Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation of the relationship of selected dietary factors and colon cancer Wolczuk, Patricia Dianne 1984

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A1 W64.pdf [ 30.05MB ]
Metadata
JSON: 831-1.0096638.json
JSON-LD: 831-1.0096638-ld.json
RDF/XML (Pretty): 831-1.0096638-rdf.xml
RDF/JSON: 831-1.0096638-rdf.json
Turtle: 831-1.0096638-turtle.txt
N-Triples: 831-1.0096638-rdf-ntriples.txt
Original Record: 831-1.0096638-source.json
Full Text
831-1.0096638-fulltext.txt
Citation
831-1.0096638.ris

Full Text

AN INVESTIGATION OF THE RELATIONSHIP OF SELECTED DIETARY FACTORS AND COLON CANCER by PATRICIA DIANNE WOLCZUK B.H.Ec, The University of Manitoba, 1967 M.Sc, The University of Manitoba, 1970 A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES HUMAN NUTRITION AND EPIDEMIOLOGY We accept this dissertation as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January, 1984 ©Patricia Dianne Wolczuk }jtsf In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department or by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department o f I n t e r d i s c i p l i n a r y S t u d i e s The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 84.04.26 DE-6 (3/81) ABSTRACT Cancer of the colon is a major cause of death in Canada and in Brit i s h Columbia. It is a disease of older persons which has a considerable latency period for development. To a large extent, i t may be preventable. The majority of past research has concentrated on current food habits in an attempt to find a single dietary cause of this disease. This case control study was designed to c l a r i f y the relationship over time of selected dietary factors in the etiology of colon cancer, total of 111 cases and 92 controls of similar age and sex were given an extensive dietary interview which probed food consumption and related factors in each of three major l i f e stages: current, middle and teen years. Approximately f i f t y percent of the current incident case population were ineligible to interview; of these thirty per cent were deceased or too i l l . Nevertheless, the case sample was considered to be representative of the population as the distributions of ages and cancer sites were similar to the case population. Controls were friends of cases or physician identified healthy individuals. The biodemographic variables of significance were urban residence more than seventy five per cent of lifespan, managerial/professional occupation, a previous history of intestinal polyps, and a history of i i i having been overweight. The general food habits found related to colon cancer were the current practice of skipping breakfast and the handling of leftover vegetables. None of the eight major food groups representing current consumption of vitamin A r i c h vegetables, vitamin C r i c h foods, cruciferous vegetables, fats, meat, vegetable source protein, f i b e r r i c h foods and beef fat were related to colon cancer. These same food scores when assessed for food consumption ten to f i f t e e n years pre-diagnosis showed a strong protective effect for the increased consumption of vitamin C r i c h foods, cruciferous vegetables and vegetable source proteins. An increased consumption of fibrous vegetables showed a similar protective effect; cheese consumption enhanced r i s k . Data from the teen years reinforced the protective r o l e of vitamin C r i c h foods. Meat and fat consumption were not found to be related to colon cancer. It i s speculated that these dietary factors could be involved i n an in t e r n a l ecosystem influencing the concentration of fe c a l b i l e acids which are promoters of carcinogenesis. iv TABLE OF CONTENTS CHAPTER PAGE 1. INTRODUCTION 1.1 BACKGROUND TO THE PROBLEM 1 1.2 STATEMENT OF THE PROBLEM 3 1.3 HYPOTHESES 4 1.4 DEFINITIONS 7 1.5 ASSUMPTIONS 9 1.6 ORGANIZATION OF THE STUDY 9 2. REVIEW OF LITERATURE 2.1 PATHOLOGY 10 2.2 DESCRIPTIVE EPIDEMIOLOGY 17 2.3 DIETARY STUDIES . 25 2.4 CARCINOGENESIS . 52 2.5 METHODOLOGICAL ISSUES 58 3. METHODOLOGY 3.1 POPULATION/SAMPLE 70 3.2 INSTRUMENTS ' 77 3.3 DATA COLLECTION 79 3.4 PERSONNEL 81 3.5 DATA ANALYSIS 82 4. RESULTS AND DISCUSSION 4.1 RESPONSE RATE 84 4.2 DESCRIPTION OF SAMPLE 89 V CHAPTER PAGE 4.3 HYPOTHESES TESTING 4.3.1 Biodemographic Variables 92 4.3.2 General Food Habits 104 4.3.3 Current Food Consumption 109 4.3.4 Previous Food Consumption 116 4.3.5 Variable Interactions 133 5. SUMMARY AND IMPLICATIONS 5.1 SUMMARY OF THE STUDY 136 5.2 LIMITATIONS 142 5.3 IMPLICATIONS 145 5.4 FUTURE RESEARCH * 147 6. BIBLIOGRAPHY 148 7. APPENDICES 164 LIST OF TABLES TABLE PAGE 1. Percentage Distribution of Histological Types 13 of Cancers in the Large Intestine, Excluding Rectum. 2. Influence of Stage of Disease on Survival of Patients 16 with Large Bowel Cancer Treated via Surgical Resection. 3. Five Year Survival of ICDA Site 153. The Crude Rate of 16 Survival for British Columbia Cancer Patients Diagnosed 1969-1973 4. Incidence of Colon Cancer (ICCD # 153) for each Registry 18 (Age Standardized Rates: World Population per 100,000). 5. Average Cl i n i c a l Incidence per.i00,000 (A l l Ages 1969-72) 19 Colon Cancer. 6. Age Standardized Mortality Rates by Province, 1970-72. 20 Large Intestine (Canadian Population). 7. Colon Cancer Incidence in Canada by Age Grouping, 1980. 22 8. Summary of Dietary Studies as Presented in Literature 46 Review 9. Age Distribution of the New Cases of Malignant Neoplasm 73 of the Large Intestine in B.C. in 1977. 10. Geographic Distribution of Live Notifications of Colon 74 Cancer in British Columbia in 1977 by School D i s t r i c t s . 11. Response Rate Results: Pre-Contact Cases. 86 12. Response Rate Results: Post-Contact Cases. 87 13. Response Rate Results: Controls 88 14. Distribution of Cancer Sites in Total Cases Identified 91 and Sample on which Interviews were Completed. 15. Summary of Results: Biodemographic Variables. 93 16. Relative Risk Estimates: Urban Residence. 94 17. Relative Risk Estimates: Mother's Country of Origin. 96 v i i TABLE PAGE 18. Relative Risk Estimates: Major Occupation. 97 19. Relative Risk Estimates: Intestinal Polyps. 99 20. Relative Risk Estimates: Obesity History. 100 21. Summary of Results: Exposure to Hazardous Chemicals. 102 22. Relative Risk Estimates: Cooking Oils and Fumes. 103 23. Summary of Results: General Food Habits 106 24. Relative Risk Estimates: Meals Skipped. 107 25. Relative Risk Estimates: Handling Leftover Vegetables. 108 26. Summary of Results: Current Food Frequencies. 110 27. Relative Risk Estimates: Fish and Poulty (Current). 112 28. Relative Risk Estimates: Wine (Current). 113 29. Summary of Results: Current food Scores. 115 30. Summary of Results: Middle Years Food Frequencies. 117 31. Relative Risk Estimates: Fibrous Vegetables 118 (Middle Years) 32. Relative Risk Estimates: Cheese (Middle Years) 120 33. Summary of Results: Middle Years Food Scores. 122 34. Relative Risk Estimates: Vitamin C Rich Foods 123 (Middle Years). 35. Relative Risk Estimates: Cruciferous Vegetables 125 (Middle Years). 36. Relative Risk Estimates: Vegetable Source Proteins 127 (Middle Years). 37. Summary of Results: Teen Years Food Frequencies. 128 38. Relative Risk Estimates: Cheese (Teen Years) 130 39. Summary of Results: Teen Years Food Scores. 131 V l l l TABLE PAGE 40. R e l a t i v e R i s k Estimates: V i t a m i n C Rich Foods 132 (Teen Y e a r s ) . 41. Summary of R e s u l t s : Age Sex C l a s s i f i c a t i o n 134 ix LIST OF FIGURES FIGURE PAGE 1. The P e r c e n t i l e D i s t r i b u t i o n of C o l o r e c t a l 11 Cancers i n Various Segments of the Large Bowel. 2. Schematic Representation of Dukes C l a s s i f i c a t i o n 14 of Carcinoma of the Large Bowel 3. The Progressive Development of Cancer 52 4. The Possible Contribution of N u t r i t i o n a l 56 Factors to the Development of Colon Cancer. X LIST OF APPENDICES APPENDIX PAGE 1 The F i n a l I n t e r v i e w Schedule 164 2 Food Photographs Used f o r P o r t i o n Size Determinations 187 3. Contact L e t t e r s 203 4. Raw Data 208 x i ACKNOWLEDGEMENT To Tim for his support and understanding. 1 CHAPTER I INTRODUCTION 1.1 BACKGROUND OF'THE PROBLEM Cancer of the colon i s a major cause of death in Canada and in Bri t i s h Columbia. It i s the second leading cause of death from cancer. Disease rates in men and women are similar. The crude survival rate of this disease at two years post diagnosis i s only f i f t y per cent. This disease costs millions of dollars, not only for Its treatment and morbidity and mortality, but also in terms of decreased quality of l i f e and loss of productivity. To a large extent i t may be preventable. Colon cancer i s more predominant i n more economically developed countries. For example, the age standardized mortality rate for colon p cancer i s higher i n the highly urbanized United States than i n less developed countries. In Connecticut, the rate i s 26.0 per 100,000 compared to less than 1.0 in Nigeria ( I . C C D . Reports, 1970). In B r i t i s h Columbia, the age standardized mortality rate i s 15.0 per 100,000, one of the higher rates i n the world. What factors account for the differences between nations? As i n a good detective novel, the clues and potential culprits are many. Using national food consumption data, a strong positive correlation has been demonstrated between fat consumption and colon cancer (Wynder, 1975). Fat degradation products have been found to be tumorigenic i n animals. A negative association has been found with dietary fiber (Burkitt, 1971). Dietary protein has been suspected, as vegetarians have a lower incidence of colon cancer. Nitrates i n foods can be chemically transformed into 2 potent carcinogens which cause colon cancer i n animals. Despite numerous studies, research i s s t i l l contradictory and the supportive data from case-control studies i s limited. Two of the principal factors which may be responsible for the i n a b i l i t y to identify the causative nature of the disease are i t s potential multi-factorial nature and a significant latent period in disease development. U n t i l recently, researchers have concentrated on mutually exclusive, single factor hypotheses such as the "fat theory" or the "fiber theory". Almost a l l of the studies have concentrated on recent food habits. The disease most l i k e l y i s i n i t i a t e d a number of years prior to development of symptoms. Thus, research which identifies only current food habits may be irrelevant as i t may r e f l e c t dietary changes made as a result 0 of the disease. Further research i s needed to delineate a model that represents the multi-factorial etiology of the disease including both risk promoting and negative risk or protective factors and to describe the biodemographic characteristics of particular groups which may be at high risk early in the disease process. 3 1.2 STATEMENT OF THE PROBLEM This study is designed to c l a r i f y the relationship of selected dietary factors in the etiology of colon cancer. Data was collected in a case-control study to address the folowing specific questions: (a) Are there any significant differences between the cases and controls in terms of the selected biodemographic variables which may be related to dietary practices such as occupation, education, marital status, parental country of origin, family history of cancer, personal disease history, stress, smoking and obesity? (b) Are there any significant differences between the cases and controls in terms of selected food habits and practices such as meal consumption patterns, vitamin supplementation, or trimming fat from meat? (c) Which of the selected dietary risk enhancing and protective factors are associated with colon cancer? (d) Are these risk factors consistently present in the diet at three stages of l i f e : currently, in the middle years and in the teenage years? 4 1.3 HYPOTHESES Biodemographic Factors Based on a review of literature, i t was hypothesized that there are differences i n biodemographic variables between those with and without colon cancer. Thus the following f i r s t set of null hypotheses were tested: That there w i l l be no significant differences between individuals with colon cancer and those of similar age and sex who do not have any known cancer of the colon with regard to: 1. principal area of residence 2. urban vs. rural residence 3. level of education 4. association with a church or f a i t h 5. father's country of origin 6. mother's country of origin 7. marital status 8. number of years married 9. major occupation 10. parental history of cancer 11. s i b l i n g s 1 history of cancer 12. personal disease history 13. smoking history 14. obesity history 15. stress i n the middle years 16. exposure to hazardous chemicals 5 General Food Habits It was hypothesized that differences would exist between cases and controls i n terms of selected food habits and practices. Therefore, the second set of hypotheses tested were: That there w i l l be no significant difference between individuals with colon cancer and those of similar age and sex who do not have any known cancer of the colon with regard to the practices of: 1. skipping meals especially breakfast 2. consuming only 3 meals a day 3. taking large doses of vitamins C and E 4. trimming fat from meat before eating 5. the addition of sugar to tea, coffee and breakfast cereal 6. the regular use of jams, j e l l i e s and honey 7. the. handling of leftover vegetables. Current Food'Consumption It was postulated that there w i l l be relationships present between selected risk enhancing and protective factors. Therefore, this third set of n u l l hypotheses were tested. 1. That there are no significant differences between individuals with colon cancer and those of similar age and sex who do not have any known cancer of the colon with regard to frequency of consumption of deep yellow vegetables, selected cruciferous vegetables, leafy green 6 vegetables, fibrous vegetables, peanuts, dried beans, peas and l e n t i l s , f i s h and poultry, beef, cured meats, cheese, eggs, citrus f r u i t s , whole grain and refined bread and cereals, tea, coffee, soda pop, liquor, smoked, g r i l l e d or pickled foods. 2. That there are no significant differences between individuals with colon cancer and those of similar age and sex who do not have any known cancer of the colon with regard to scores calculated to represent overall consumption of vitamin A rich vegetables, vitamin C rich vegetables, cruciferous vegetables, fat, meat source protein, vegetable source protein and fiber. Previous Food Consumption It was hypothesized that those relationships outlined i n the third set of hypotheses would exist in previous l i f e stages. Thus the above third set of hypotheses were tested i n two additional time frames: the middle years, that is ten to f i f t e e n years pre-diagnosis of disease, and in the teenage years, to determine i f significant differences occurred in these previous l i f e stages. 7 1.4 DEFINITIONS The following terns are defined b r i e f l y to aid understanding. More detailed explanations are provided in the text. Colon Cancer: a malignant neoplasm located in the colon (large bowel); excludes rectal cancer. Cases: persons with colon cancer as identified from the Cancer Registry of Bri t i s h Columbia. Cancer Registry of British Columbia: a registry which receives and records cancer cases reported by physicians throughout Br i t i s h Columbia. Controls: persons of similar age (+ 5 years) and the same sex as the cases, who have no known bowel disease. Food scores: scores calculated by summing the values given for frequency of consumption of foods selected to represent various categories of food. Principal areas of residence: regions identified as having the greatest total number years of residence. Urban residence: residence in a community with a population of more than 1,000 persons. Level of education: the highest educational qualification obtained categorized as: none; grades one to eight; grade eight to high school graduation; less than two years post secondary; two to four years post secondary; more than four years post secondary. Parental history of cancer: whether mother and/or father ever had any form of cancer. 8 Sibling history of cancer: whether brothers and/or sisters ever had any form of cancer. Stress in the middle years self reported assesssment of l i f e stress ten to fifteen years pre-diagnosis, cl a s s i f i e d as: none; job stress, family stress; family illness or bereavement; or multiple of stresses. Smoking history: c l a s s i f i c a t i o n of tobacco smoking practices, including cigarettes, pipes and cigars; c l a s s i f i e d as: non-smoker (never smoked); short-term smoker (less than two years); established smoker (two to five years); long-term smoker (more than five years); current smoking practice and amount smoked were also determined. Obesity history: self reported history of being at any time more than twenty pounds over what is considered an ideal weight; classified as: none (never more than twenty pounds overweight); short-term obesity (more than twenty pounds overweight for less than two years); established obesity (more than twenty pounds overweight for two to five years); long-term obesity (more than twenty pounds overweight for more than five years). Personal disease history: whether individual had had any of the following diseases; heart disease, intestinal polyps, ulcerative c o l i t i s , Crohn's disease, appendicitis, hemorrhoids, varicose veins, d i v e r t i c u l i t i s , peptic ulcer. 9 1.5 ASSUMPTIONS The underlying assumptions for this study are: 1. Cases are representative of persons with colon cancer residing i n B r i t i s h Columbia. 2. Controls are representative of persons of similar age and sex without any known cancer of the colon residing i n B r i t i s h Columbia. 3. Information provided i n the interviews was provided in good f a i t h with both cases and controls exhibiting similar a b i l i t y to re c a l l previous food consumption. 1.6 ORGANIZATION OF THE STUDY This study begins with some background to the problem and identifies the need to c l a r i f y the relationship of selected dietary factors in the etiology of colon cancer. The problem and hypothesis are stated. The review of the literature and description of the methodology follow in Chapters 2 and 3 respectively. Chapter 4 contains the results and discussion of those findings. The f i n a l chapter summarizes the study, i t s limitations, and significance. References and appendices of related material conclude this dissertation. 10 CHAPTER 2' REVIEW OF LITERATURE 2.1 PATHOLOGY The colon is a large muscular hollow organ. Its functions are to provide an outside link in the food-environment chain for the disposition of waste and to extract water and electrolytes from the cecum passing into i t from the ileum (Spiro, 1977). It is about 1.5 meters in length. It leads to the rectum. The distribution of carcinomas of the colon and rectum is illustrated in Figure 1 (Grage, 1977). Much confusion results as many studies do not differentiate carcinomas of the colon from cancer of the rectum. This study is concerned with carcinomas of the colon only, that is cancers in the International Classification System, in category number 153. This excludes the rectum and rectosigmoid junction. An established cancer of the colon usually appears as a sharply demarcated polypoid mass with central ulceration and overhanging edges (Grage, 1977). The lesions are described as polypoid, nodular, ulcerating or coll o i d . As the disease progresses, the entire circumference of the bowel may be involved. In some instances, the encircled intestinal lumen may be reduced in size to only a narrow distorted passage or even almost obliterated by this "napkin ring" configuration (Wood, 1967). The tumor may penetrate the f u l l thickness of the bowel wall and spread to the adjacent structures such as the bladder, vagina, uterus, ovaries, and regional lymph nodes. Microscopically, most (98.2%) of the cancer of the colon are 11 FIGURE 1: The Percentile Distribution of Colorectal Cancers in Various Segments of the Large Bowel. Transverse colon 7% Colon 40% Rectum Source: Grage, 1977 (p. 292) (Horton & H i l l ) 12 adenocarcinomas with distributions as i l l u s t r a t e d in Table 1 (Cowdry, 1968). The distribution in B r i t i s h Columbia i s similar. The e p i t h e l i a l barrier l i n i n g the cecum is considerably different from that of the small intestine, i t has a r e l a t i v e l y smooth surface and no absorptive v i l l i . Some of the mucoid goblet c e l l s and columnar absorptive c e l l s located i n the the walls regularly divide by mitosis throughout l i f e . They replace worn out surface absorptive and goblet c e l l s . As they synthesize DNA, they may be susceptible to carcinogens. The mucous produced by these c e l l s passes toward the surface and may serve as a barrier against carcinogens (Cowdry, 1968). The effectiveness of this barrier i s attested by the fact that few sarcomas appear i n the large intestine. Cowdry contends that i f , instead of d i r e c t l y contacting, or otherwise affecting the mucous gland c e l l s , the carcinogens were endogenous, arriving via the blood stream and entering the tissue f l u i d surrounding the subepithelial mesodermal c e l l s , sarcomas would probably be the h i s t o l o g i c a l types that predominated. Colonic mucous gland c e l l s decrease in numbers in older persons thus decreasing the amount of available lubricant rendering the surfaces more susceptible to trauma and perhaps to carcinogens. Cancers of the bowel are generally c l a s s i f i e d using the Dukes System. A Dukes Stage A tumor i s confined to the bowel wall; the tumor penetrates into, but nou through the muscular coat. The Dukes Stage B tumor has progressed through the muscularis to the serosa. The Dukes Stage C tumor has spread throughout the serosa and/or to regional lymph nodes. Approximately one third of patients with colorectal cancer have lymphatic metastases when f i r s t diagnosed (Grage, 1977). The most common site of I 13 TABLE 1 Percentage Distribution of Histological Types Cancers of the Large Intestine, excluding rectum (3,945 Cases, Connecticut 1935-51). Histological Type Per Cent 1. Adenocarcinomas 98.2 2. Malignant tumors of nerve 0.7 tissues and associated structure 3. E p i t h e l i a l carcinoma 0.3 4. A l l other malignant Rumors 0.8 Total 100.0 Source: Cowdry, 1968 14 FIGURE 2: Schematic Representation of Dukes C l a s s i f i c a t i o n of Carcinoma of the Large Bowel. Source: Grage, 1977. 15 metastatic spread is the l i v e r , with lung involvement at terminal stages of the disease. Survival i s about six to nine months in those with hepatic metastases. Patients and physicians often ignore the early symptoms of colon cancers, as these are often vague and intermittent. Thus the disease is usually well advanced on presentation of c l i n i c a l symptoms. C l i n i c a l symptoms vary with the anatomic location, size and extent of the tumor and the presence of complications such as obstruction and hemorrhage. Symptoms of lesions on the right side of the colon include dull pain, fatigue and weakness (due to anemia), weight loss, anorexia, and indigestion. Lesions on the l e f t side of the colon are characterized by symptoms such as changes in bowel habits, abdominal pain, abdominal distension, and rectal bleeding or mucous. Colon cancer i s generally treated by surgical resection. Operative mortality is generally low (5-15%) for this procedure. It may be supplemented by radiation or chemotherapy. As seen in Table 2, the five year survival rate for resected patients is high (70%) i f tumors are confined to the bowel wall (Dukes Stage A). Prognosis i s poor i f there i s regional lymph node involvement. Five year survival rates for B r i t i s h Columbians with colon cancer, a l l stages combined, are presented in Table 3. The two year percent survival is f i f t y percent. This decreases to a five year crude survival rate of only thirty-seven percent. 16 TABLE 2 Influence of Stage of Disease on Survival of Patients with Large Bowel Cancer Treated via Surgical Resection. Stage Number Per Cent Survival 5 years from when f i r s t seen Dukes A 161 70 Dukes B 89 63 Dukes C 127_ 42 Total Resected 377 59 A l l patients seen 689 35 Source: Grage, 1977 TABLE 3 Five Year Survival for ICDA Site 153. The Crude Rate of Survival for B r i t i s h Columbia Cancer Patients Diagnosed 1969-1973. of Survival Per Cent 1 62% 2 50% 3 44% 4 40% 5 37% Source: V i t a l S t a t i s t i c s . Ministry of Health. Unpublished data. 17 2.2 DESCRIPTIVE EPIDEMIOLOGY International comparisons of incidence such as those in the UICC Report (1970) provide significant clues to the etiology of colon cancer. Several features of the geographic and ethnic distribution of cancer of the colon are illustrated in Table 4 (UICC Report, 1970). The highest incidence rates were found mainly in Western Europe and the English-speaking world. Caucasians appear to be more susceptible to colon cancers than blacks. For example, rates for those of European origin in South Africa are higher than for coloured or Bantu categories. Incidence for blacks in California are lower than for whites. Rates for European-descended New Zealanders are higher than Maori New Zealanders. However, rates for blacks in California greatly exceed those for blacks in Africa. This information would suggest that place of residence is as significant a variable as race. Urban populations generally have higher risks for colon cancer than rural areas, as indicated by data from Norway. This would suggest that populations in industrialized nations are at higher risk; however, rates in industrialized areas of Japan are lower, much lower than for Japanese living in Hawaii. Rates vary widely within nations, as indicated by the American data. Table 5 represents Canadian average annual incidence rates. Rates vary considerably from province to province with the average incidence rates in B r i t i s h Columbia being among the highest in Canada. This is related to the higher proportion of older persons in British Columbia. The age-18 TABLE 4 Incidence of Colon Cancer (ICCD # 153) for Each Registry (Age Standardized Rates: World Population per 100,000) Registry- Males Females Nigeria 1.2 0.9 South A f r i c a : White 13.8 16.9 South Africa: Coloured 4.4 7.7 South Afri c a : Bantu 2.3 4.6 Columbia 3.6 4.0 U • S • A • y California: White 24.0 22.9 U.S.A., California: Black 14.1 19.0 U.S.A., Connecticut 26.7 19 .0 New Zealand: European 24,7 25.0 New Zealand: Maori 9.2 13.9 Norway: urban f 15.2 13.4 Norway: rural 10.0 10.0 Denmark - 15.5 17.5 Finland 6.8 7.3 Hawaii: Hawaiian 20.2 12.1 Hawaii: Caucasian 19.2 27.5 Hawaii: Chinese 35.9 23.5 Hawaii: Japanese 20 .7 15.3 Japan 4.1 4.0 Israel: Jews born i n Africa/Asia 4.9 5 .2 Israel: Jews born Europe/America 12.2 14.2 Source: ICCD Report. Cancer incidence i n Five Continents, 1970 19 TABLE 5 Average Annual Incidence per 100,000: A l l Ages 1969-72 Colon Cancer. Province Males Females Br i t i s h Columbia 28.6 32.1 Alberta 17.5 19.3 Saskatchewan , 23.8 24.8 Manitoba 25.6 28.0 Quebec 14.5 19.1 Maritime Provinces 21.0 29.3. Newfoundland 20.9 21.6 Source: Cancer Incidence i n Five Continents Volume III (Waterhouse 1976). 20 TABLE 6 Age-standardized Mortality Rates by Province, 1970-72, Large Intestine. (Canadian Population). Province Ages 25 - 74 Male Female British Columbia 14.9 15.4 Alberta 15.1 15.6 Saskatchewan 15.1 15.8 Manitoba 18.9 18.3 Ontario 21.5 20.7 Quebec 22.4 21.0 New Brunswick 15.3 22.1 Nova Scotia 25.3 27.9 Prince Edward Island 15.4 26.2 Newfoundland 26.2 24.4 Source: B.C. Cancer Registry, personal communication. 21 standardized mortality rates (Table 6) il l u s t r a t e a lower standardized mortality rate for Western Canada than for Eastern Canada. In countries with very low rates, the incidence is slightly higher in females. Colon cancer rates are slightly higher in nulliparous women than in parous women. The sex ratio in Canada approximates unity. Within British Columbia, the overall mortality from cancer at a l l sites in Chinese and Japanese males did not d i f f e r significantly from that of white males (Gallagher, undated). Chinese females had a decreased risk for colon cancer (R.R. = 0.4) . Colon cancer risk for native Indian males is reduced in B.C. (R.R. = 0.5) although risks for females are similar to those for white females. Colon cancer is a disease of the older years. It is seldom seen in children and young adults. As illustrated in Table 7, the incidence increases sharply after forty years of age reaching a peak in the seventy-five to seventy-nine age group. The age of diagnosis is usually sixty to seventy years. Grace (1980) reviewed four conditions considered to be pre-disposing for colon cancer: adenomatous polyps, villous adenoma, familial polyposis, and ulcerative c o l i t i s . In a population at risk for colon cancer, about 10 to 15 per cent have adenomatous polyps. The incidence increased with age; by the seventh and eight decade, adenomatous polyps may be found in the colon of almost f i f t y per cent of the population. Athough most polyps never do become malignant the fact that animal experimentally induced carcinogenesis progresses through a stage of adenomatous polyps prior to adenocarcinoma reinforces i t s role in the etiology of colon cancer. 22 TABLE 7 Colon Cancer Incidence in Canada by Age Grouping, 1980. (Age-Standardized Rates per 100,1 Age Group Canada British Columbia Male Female Male Female 0 - 2 4 0.1 0.3 0.2 -25 - 34 1.4 1.7 0.9 0.9 35 - 44 6.6 8.0 6.2 3.2 45 - 54 21.0 28.6 22.3 25.6 55 - 64 63.7 62y0 59.3 69.2 65 - 74 149.0 139.8 144.1 152.9 75 - 84 258.0 235.4 295.7 244.6 85 and over 346.1 288.9 468.1 397.7 Overall 22.3 25.8 23.3 26.3 Source: B.C. Cancer Registry, personal communication. 23 Vil l o u s adenomas appeared to be more direc t l y related to rectal cancer. Familial polyposis i s an inherited t r a i t . Approximately sixty per cent of persons with f a m i l i a l polyposis developed carcinoma, primarily of the rectum, i n the twenty-three year follow up study conducted by Grace. Persons with ulcerative c o l i t i s are more vulnerable to colon cancer, with the r i s k increasing with the duration of the disease, the extent of involvenment and age of the patient. However, these hereditary disorders account for only a small portion of the disease incidence. The probability of finding colon carcinoma in immediate relatives of patients i s two to three times that in the normal population. One study (Livett, 1974) has shown a five f o l d increase of risk i n siblings, however, these data may refle c t f a m i l i a l dietary patterns rather than genetic predisposition. Migrant studies have helped separate genetic and environmental factors. The incidence of colon cancer in European migrants (e.g. Poles) to the United States increases rapidly i n their l i f e time as they adopt American l i f e s t y l e . Similarly, for Asiatic migrants i t was found that the incidence of colon cancer i s higher among Japanese l i v i n g in the United States than those l i v i n g in Japan (UICC Report, 1970). The longer the Japanese resided i n the United States, the closer their incidence rates approached the American white rate. By the second generation after migration, the rate was similar to United States residents (Haenszel, 1961). Blacks l i v i n g i n Alameda County, Cali f o r n i a , have a much greater incidence of colon cancer than blacks i n Nigeria (from which many American blacks originated). Thus, in general, migrant populations have incidence 24 rates of colon cancer similar to the place i n which they reside rather than their place of b i r t h . This strongly implicates some environmental factor or factors which are causative, although migrant populations are not necessarily representative of the population of their native country. There i s a paucity of data on occupational exposures and increased risk of colon cancer. A study i n Ireland (Elmes and Simpson, 1971) showed an excess of colon cancer i n insulation workers. Selikbff (1973) showed a similar increase in asbestos insulation workers in the United States. Asbestos particles caught in the nasal or oral passages and swallowed could reach the large bowel and may have a carcinogenic effect l i k e that of inhaled asbestos for the lung cancer in smokers. This i s unlikely to be a factor in this study group as exposure to asbestos i s limited. Extensive studies of indus t r i a l populations have shown no group working with dyes or rubber to have an above average risk of colon cancer ( M i l , 1975). 25 2.3 DIETARY STUDIES The geographical differences and evidence from the migrant studies have led investigators to examine food as a possible source of factors affecting the incidence of colon cancer. Generally, these studies may be cl a s s i f i e d as correlational studies, case-control studies and laboratory studies. This segment of the review is organized to explain each of these groups of literature and is confined to dietary studies which may be related to colon cancer. The principal hypotheses relate to a causal role for a high fat and/or high meat diet and a protective role for dietary fiber, with i n i t i a t o r , promotor or protective roles postulated for a number of other food constituents. CORRELATION STUDIES Correlational studies based on population food consumption data provided the f i r s t general evidence of the involvement of diet in colon cancer. In 1967 Wynder and Shigematsu, using national food consumption data, found a positive correlation between dietary fat and colon cancer. Gregor (1969) correlated cancer mortality with food consumption data in twenty-eight countries. He found a high correlation (r = + 0.81) between intestinal (colon and rectum) cancer and animal protein consumption. No correlation was determined for total protein or for cereal consumption. At about the same time, Burkitt (1969) asserted that the incidence of colon cancer is inversely related to the amount of dietary fiber consumed. 26 He observed that the age-adjusted mortality rate for colon cancer was very low in the Bantu of South Africa compared to the white South African population, 2.4 and 21.0 per 100,000 respectively. Unfortunately, the level of fiber consumption attributed to the Bantu is a controversial issue. The original authors from whom Burkitt obtained data on Bantu fiber consumption have since published contradictory reports ( H i l l , 1976). Burkitt (1971) further contended that a low residue diet high in refined carbohydrate foods (white flour, sugar) is a major factor in the etiology of colon cancer and that decreased dietary fiber decreases intestinal transit time and stool bulk and alters stool consistency. He noted that a high carbohydrate diet alters fecal bacteria. He hypothesized that carcinogens are formed in the gut by abnormal bacterial flora produced by a refined carbohydrate diet and that these carcinogens are given prolonged exposure to the intestinal mucosa due to the low residue content of the diet. Population studies have been conducted to develop support for the "fiber" hypothesis. The data from a study of Chilean farm workers showed they have a very low age-adjusted death rate from colon cancer and a high fiber, low saturated fat diet (Zoldivar, 1976). A study comparing age-standardized rates of colon cancer in Denmark and Finland showed a dramatic difference (Jensen, 1974). The rates were 25.4 per 100,000 in Denmark and 10.4 per 100,000 in Finland. Both the Finns and the Danes have a high fat intake derived mostly from dairy products (polyunsaturated to saturated fat ratio = 0.15). The Finns consume a relatively higher amount of fiber. Stool bulk is greater 27 diluting the bile acid content of the stool. Thus, this data supports the protective role of fiber in the presence of other possible risk factors. Howell (1975, 1976) examining international per capita food disappearance, identified strong positive correlations between colon cancer mortality rates and total calories from meat (r = + 0.88), meat fat (r = +0.86), meat protein (r — +0.86), and total fat (r = + 0.90). Strong negative correlations were found for calories from vegetable and vegetable proteins. Factor analysis showed a further negative association between colon cancer mortality and the consumption of pulses (peas, beans), rice and rice with milk. The type of meat with the highest positive correlations was "cattle meat". Poultry, pork and fish consumption were not strongly related to colon cancer rates. Armstrong and Doll (1975) also identified a positive correlation between meat intake and colon cancer and between total fat intake and colon cancer using a population food consumption comparison. Studies on Seventh Day Adventists, about half of whom are vegetarians, support the implication of animal protein in the causation of colon cancer. Rates of colon cancer are lower in Seventh Day Adventists than in the general United States population, especially for those who do not consume meat. Serum cholesterol levels and total fat intakes are also reduced in vegetarians, complicating the interpretation of the extent to which meat consumption is related to colon cancer incidence (Finegold, 1975). However, the validity of the animal protein hypothesis as a single factor causation is seriously questioned by information on colon cancer rates in the Mormon population. Colon cancer mortality rates are low in 28 the Mormon population compared to rates for the United States, in general, and to rates for the Utah population where many Mormons reside (Lyon 1976; Lyon & Sorenson, 1973). The Mormon church advocates abstention from the use of tobacco, coffee, tea and alcohol. Mormons do consume beef at levels comparable to the general United States population (Finegold, 1977). In 1975, Wynder contended that total fat rather than meat was the causative dietary factor relating to colon cancer. He pointed out that colon cancer rates were very low in Japan where fat consumption is also very low compared to the Western world. A correlation coefficient of r = +0.62 was derived using per capita disappearance data for fats and o i l s . The same data raises questions as to the possible effects of other dietary factors. For example, Canada has one of the highest age-adjusted r mortality rates for colon cancer and yet the per capita fat and o i l consumption is considerably lower than that in New Zealand, the Netherlands, Belgium or Denmark where the rates are lower. As with meat consumption, the interpretation of these data on fat consumption as related i to cancer of the colon is complicated by the lack of knowledge of fat sources, fat quantity, and other factors such as the fiber content of the diet. Cholesterol is an additional complicating factor in these analyses. H i l l (1975a) noted that further examination of data on food disappearance and colon mortality showed that the partial correlation for cholesterol remained high when fat was controlled, however, the reverse was not true; the partial correlation for fat was not significant when cholesterol was controlled. Thus, although population studies provided the f i r s t general evidence 29 of a relationship between diet and colon cancer, the evidence is inconclusive. Unfortunately, these population comparisons may be relatively inaccurate as the food "consumption" figures are based on t national food disappearance data from Ministries of Agriculture. These figures are an "average" per capita calculation. They do not document actual consumption. They do not account for differences in food intake of different socio-economic or ethnic groups nor do they account for home grown foods, waste, or variations in food preparation. However, they have suggested avenues for more thorough investigations. CASE-CONTROL STUDIES Most of the diet and cancer research in the sixties and seventies focused on defending or refuting either one or the other of two principal hypotheses: the "high fat/high meat" theory and the "fiber" theory. The case-control study was an important vehicle for these investigations. In 1966 Higginson conducted a case-control study with 340 cases which investigated meal patterns, frequency of use of selected foods and food preparation. No consistent differences were found with regard to the type of fat used but cases did consume more fried foods including fried meat and fried foods at breakfast. No differences were determined for consumption of cheese, milk, dairy products or the use of raw vegetables. In 1967, Wynder and Shigematsu conducted an extensive case-control retrospective study of 791 persons with cancer of the large bowel and age 30 matched controls. They established that there were differences in the etiology of cancer of the colon and cancer of the rectum. Cancer of the rectum is more predominant in males than in females. In recent years, the incidence of rectal cancer is decreasing; that of colon cancer is stable. More executive/professional workers develop colon cancer; more labourers develop rectal cancer. Rectal cancer cases are more li k e l y to be alcohol drinkers than are colon cancer cases. Colon cancer patients had a higher socioeconomic status. No association could be established between slightly overweight and developing colon cancer. There was a slight association between cigar smoking and cancer of the sigmoid colon in men. They found no relationship to other medical conditions except ulcerative c o l i t i s . There was some familial basis for risk in younger individuals with colon cancer. Jewish persons were at higher risk of developing colon cancer. No consistent differences were found between cases and controls for any of the dietary factors investigated including consumption of eggs, milk, butter, cheese, meat, yellow and green vegetables, f r u i t , juices, cereals, potatoes, smoked or salted foods, and fats. No description of the dietary methodology was provided in the research report. Haenszel et al (1973) conducted a case-control study of 179 Japanese bowel cancer patients and 357 controls. They determined excess risks for persons who regularly ate Western style meals only. Meat and legumes were specifically cited as two food groups associated with a high risk of colon cancer. String beans and pea pods were identified as having relative risks of 2.5 and 2.1 respectively. Use of beef more than twelve times a month 31 versus less than eight times a month produced a calculated relative risk of 2.2. Coffee, peanut butter, papaya, orange, mushrooms, raw fish, pickled cucumbers and eggplant were suggested to be low risk or protective in nature. The authors suggested that beef as a major risk factor could be due to the fact that beef adds to the total protein intake, to the total fat intake or to the total saturated fat intake. In addition, they postulated that beef metabolism may produce a carcinogen resulting from the degradation of the amino acid, tryptophan, or from the particular long chain fatty acid composition of beef. The plant sterol composition of legumes was identified as an area requiring future research. Results of this case-control study are d i f f i c u l t to compare to other research mentioned here as this study included rectal cancer in the data analysis. The urinary metabolites of tryptophan have been associated with bladder cancer. There is an indication that the fecal concentration of tryptophan is higher in cases of large bowel cancer than in controls and that fecal tryptophan concentration increases with increased dietary protein ( H i l l , 1975a). Should beef metabolism produce carcinogens, then bowel transit time might be significant in determining carcinogen contact with bowel walls. In 1978, MacLennan compared dietary and bowel habits of two groups of Scandinavians, one having an incidence rate of colon cancer fourfold that of the other area. The "high incidence" group consumed more white bread, meat, beer and less potatoes and milk. Bowel habits were similar; although stool weights were higher in the low incidence group, transit times were 32 not significantly different. A recent cohort study over a 21 year period among 21,295 white California Adventists indicated that large bowel cancer was unrelated to meat use but was positively associated with coffee intake (Phillips & Snowdon, 1983). A complication of the interpretation of the "meat" theory data may be the presence of such diet-related factors as polycyclic aromatic hydrocarbons. Polycyclic aromatic hydrocarbons (PAH) were among the f i r s t chemical compounds found to be powerful carcinogens in animals. The major source of PAH in the environment is the result of inefficient combustion. They contaminate marine sources of food as well as vegetables and plants (Dipple, 1983). Although air pollution and tobacco smoke are the most common sources of these compounds, they are found in foods such as smoked fish , ham, barbequed beef, reused o i l s and coffee. G r i l l i n g or charcoal broiling produces polycyclic hydrocarbons. Well done hamburgers contain 0.14 parts per million of these mutagens; beef extract cubes contain 0.1 parts per million (Commoner et a l . , 1978). The higher the fat content, of both meat and fish, the greater the amount of PAH formed (Dipple, 1983). Development of polycyclic aromatic hydrocarbons has been proposed as an explanation for the relationship between colon cancer and beef. Leafy plants, such as spinach, contain more PAH than do plants with smaller surface areas. There is no evidence to indicate intake of polycyclic aromatic hydrocarbons is excessive in colon cancer patients. In 1978, Graham et al reported on a case-control study of 470 colon 33 cancer patients and 783 controls. They found no increased risk of cancer of the colon regardless of the frequency of consumption of beef or other meats ingested. An increased frequency of consumption of vegetables, (cabbage, brussel sprouts, broccoli) was associated with a decreased risk of cancer. Cruciferous vegetables (broccoli, brussel sprouts, cabbage, cauliflower, turnips) may be protective against colon cancer. Modan (1975) also found negative associations between regular consumption of cruciferous vegetables and colon cancer. Graham et al (1978) further found that a decreasing frequency of cabbage consumption was associated with an increased risk of cancer of the colon. There was a threefold increase in risk associated with rarely or never eating cruciferous vegetables (Graham, 1983). It is postulated that the indoles present in cruciferous vegetables inhibit carcinogensis by polycyclic aromatic hydrocarbons (PAH) by increasing arylhydrocarbon hydroxylase (AHH) activity (Wattenberg and Loub, 1978). Indoles in the diets of mice have inhibited benzo[ajpyrene induced tumors (Zaridze, 1981). This evidence supports a protective role for dietary indoles and cruciferous vegetables. In addition, Haenszel, Locke and Segi (1980), conducting a larger case-control study of 588 patients and 1176 controls, did not reproduce the previous significant association of beef, string beans or starches and colorectal cancer. Negative associations with cruciferous vegetables were found. This study also yielded no consistently significant differences in relative risk estimates of colon cancer for occupation, residence (urban 34 vs. rural), education, parity or tobacco use. A complication, of course, is the combining of colon and rectal cases in the sample. The authors attributed the lack of association to the d i f f i c u l t y in detecting case-control food consumption differences in areas with homogeneous diet practices. Rozen et al (1981) comparing diets of Tel Aviv residents and those of residents of a communal farm where the incidence of colon cancer was lower found no difference for calories, fat, protein and cholesterol. Significant differences were found for fiber, carotene and ascorbic acid, the intakes of which were higher on the communal farm. Data on animal protein consumption is further complicated by the close interrelationship of animal protein, animal fat and cholesterol and the possible complicating effects of other diet factors such as gross protein intake, ratio of vegetable: animal protein sources, consumption of complex carbohydrates and legumes. Also i f data is collected only on frequency of consumption, specific nutrient intakes cannot be calculated. A study supporting the fiber theory was conducted by Modan (1975). In this case-control study of 198 patients with colon cancer, a significantly lower frequency of consumption of fiber was found in the cancer patients than in the controls. The data was consistent for 61 of the 73 items on the fiber food l i s t . Patients were interviewed after admission to hospital. A recent case-control study in American blacks showed that colon cancer patients consumed foods with less than 0.5% fiber more frequently than did age (within 5 years) - sex matched controls (Dales 1978). The 35 colon cancer patients more frequently consumed foods containing greater than 5% saturated fat than did c o n t r o l s . The combination of these two factors was a s t a t i s t i c a l l y s i g n i f i c a n t f a c t o r . In t h i s same study, t o t a l beef consumption showed a non-significant negative r e l a t i o n s h i p to colon cancer i n contrast to Japanese-Hawaiian (Haenszel, 1973) and Seventh Day Adventist data (Finegold, 1977). Fi s h and fowl showed minimal differences; s t r i n g beans showed a weak p o s i t i v e association, again i n contrast to the Japanese-Hawaiian information. This is the only study which has linked two p o t e n t i a l dietary fators and i s consistent with a m u l t i f a c t o r a l hypotheses for the causation and promotion of cancer. This data would also support a modified B u r k i t t ' s hypothesis; i . e . , that d i e t a r y f i b e r i s a protective factor i n the presence of a high f a t di e t . Unfortunately, the Dales study (1978) also covered only d i e t a r y practices immediately preceding i l l n e s s . As colon cancer develops over a period of time, i t i s not possible to determine from the study i f the d i e t a r y differences observed were an a r t i f a c t of the disease, i . e . , dietary changes r e s u l t i n g from the disease before overt i l l n e s s i s apparent, or i f these factors were causative, and i f causative, what the necessary duration of exposure i s or which factor or combination of f a c t o r s . Reddy et a l (1983) compared nutrient intakes i n three population areas: a high-incidence area in New York, a high-incidence area in Sweden and an intermediate incidence area i n Sweden. Mean protein and fat intakes were comparable i n a l l areas. Fiber intake and s t o o l output were higher i n 36 the intermediate incidence area; fecal b i l e acid concentration was lower. These results would suggest that a high fiber intake may be protective even in a population with a high fat intake. Fiber may dilute and/or bind tumor promotors. A Canadian study (Jain et a l , 1980) of 348 cases of colon cancer and 194 cases of rectal cancer compared to 542 population controls and 535 hospital controls found an increased risk of bowel cancer for those with an elevated intake of saturated fat, oleic acid, cholesterol, total protein and total calories. Risk also was increased for those with a history of bowel polyps. No association was noted in this study for crude fiber or vitamin C and colon cancer. Unfortunately, the fat data included both colon and rectal cases in the analysis. The analysis from a large prospective study in Hawaii did not support an association between fat intake and colon cancer (Kolonel, 1983), although an association was found for saturated fat and breast and prostate cancer (Kolonel et a l , 1981). The relationship of fat and cholesterol to colon cancer is unclear. The Framingham Study identified an inverse relationship between serum cholesterol and colon cancer incidence in men (Williams et al, 1981). This inverse association was consistent after adjustment for age, alcohol consumption, cigarette smoking, education and relative weight. Miller et al , (1981) found a similar inverse relationship in 133 pairs of age sex matched cases and controls. Association between lower serum cholesterol levels persisted between cases with advanced tumors and controls but not between cases with early tumors and controls. This inverse association was 37 confirmed in a prospective study of more than 8000 Hawaiian-Japanese men (Stemmerman et a l , 1981). Serum cholesterol levels below 180 mg/dl predicted high rates of colon cancer. In general, patients who developed cancer showed a drop in serum cholesterol ten years prior to diagnosis. The lowest serum cholesterol values were found in those who died from colon cancer. Right colon tumors showed the strongest inverse association with serum cholesterol. A role for cholesterol in cellular repair has been hypothesized (Kaunitz, 1972). Vitamin A has a role in the maintenance of epithelial tissues. In a vitamin A deficiency, epithelial tissue is altered to a squamous keratinizing form with a decrease in mucous secreting cells (Kummet, 1983). Newburne (1977) demonstrated that vitamin A deficiency enhances the effect of dimethylhydrazine (DMH)-induced tumors. Similarly, 1,3 cis-retinoic acid, a form of vitamin A, inhibits DMH-induced tumor formation in rats. There is some evidence retinoids, the proformed vitamin A, may have a similar role in man. In a prospective study of the smoking habits of more than 8000 men over a five-year period, a low index of consumption of vitamin A was associated with a high incidence of lung (squamous cell) tumors (Bjelke, 1975). Retinoids are considered to have an antipromotor role./S -carotenes, on the other hand, may block the init i a t i o n effects of singlet oxygen, a potentially carcinogenic free radical (Peto, 1981). Thus vitamin A or its analogues may have a protective role in man for tumors of a different histology than colon cancer (i.e. adenocarcinomas). However, no metabolic studies have linked consumption of vitamin A rich foods to adenocarcinomas as found in colon 38 cancer. Alcohol is the most powerful dietary agent which has been identified as inducing cancer in humans (Higginson, 1983). The ethanol i t s e l f appears to be the principal active agent rather than the many components of alcoholic beverages. As identified previously, beer consumption has been linked to rectal cancer in case control studies (Wynder and Shigematsu, 1967). Rectal cancer has also been linked to wine consumption in Japan (Kono and Ikeda, 1979). The consumption of alcohol, primarily red wine, has been associated with an increased risk of stomach cancer in France (Hoey et a l , 1981). Tuyns (1983), studying alcohol consumption in France, noted a six fold increase in risk of esophageal cancer for heavy drinkers. The consumption of vitamin C rich foods mediated the risk of esophageal cancer such that the relative risk of esophageal cancer was more than thirteenfold for women who were heavy consumers of alcohol and non-consumers of vitamin C (primarily citrus fruits and juices). The relative risk of colon cancer was not significantly (R.R. = 1.16) elevated for heavy drinkers in the case control study. LABORATORY STUDIES Metabolic studies have linked fiber and fat to changes in fecal steriod concentrations and fecal bacteria. The "fiber" theory supporters contend that a high fiber diet decreases fecal transit time thus reducing the opportunity for carcinogens 39 to form and for carcinogens to act on the mucosal wall. The fiber in the diet alters the gut microflora. Cleave (1956) had suggested that the excess consumption of refined carbohydrates induces bacterial changes in the gut. By-products of degradation of excess bile salts are converted to carcinogens. This hypothesis, too, is complicated by conflicting information in the literature. For example, Wynder and Shigematsu (1967) had showed no relation between colorectal cancer and bowel transit time. In 1975, H i l l investigated the effects of various diets on fecal steroid concentration and gut bacterial flora. Decreasing fat from 40 per cent to 15 per cent of the diet in a low fiber diet resulted in decreased fecal concentration of bile acids and neutral steroids. In a high fiber diet, the decrease of dietary fat content also resulted in a decrease in fecal neutral steroids. Placing subjects on a no-meat diet dramatically decreased bile acids and neutral steroid excretion. Cholesterol levels were controlled. The addition of bran (16 to lOOg) to the diet consistently reduced fecal steroids to approximately sixty per cent of controls. These results would suggest support for the protective effects of fiber in the presence of a high fat, high meat diet. Unfortunately, the number of subjects was limited (4 to 25), the duration of the experiments was short and the experimenters did not use a "cross over" design. The same research group later showed that higher socio-economic groups had a higher incidence of colon cancer in Hong Kong (Crowthy, 1976). Their analysis showed a much higher incidence of strains of Clostridium AO paraputrificum, the organism most active in steroid nuclear dehydrogenation, in the feces of persons living in the high bowel cancer incidence areas. Unfortunately, there could be a confounding factor(s) affecting these results as the bacteria were isolated from persons in high socio-economic groups, not from cancer cases. Finegold et a l , (1977) performed an extensive comparison of the fecal flora of Seventh Day Adventists and non-Adventists. Fusobacterium and C.perfringens were very low and lacto-bacillus counts were very high in Adventists compared with Caucasians living on a conventional diet. The authors stressed that there were relatively few s t a t i s t i c a l l y significant differences in fecal flora. No differences were noted between Adventists who were vegetarians and Adventist non-vegetarians. A complication of the "fiber" literature is the general lack of differentiation of the various components of dietary fiber. In an excellent review, Huang ef: al (1978) differentiate four major components of dietary fiber: cellulose, hemicellullose, pectin, and lignin. Cellulose is composed of straight chain polymers of D-glucpse units with B -1,4 linkages and strong intermolecular bonds. Hemicelluloses are divided into two groups. The A-hemicelluloses contain mostly xylose units and the B-hemicelluloses contain uremic acids, hexose and pentose sugars. Pectin is a mixture of polymers of galacturonic acid, arabinose and galatose. Lignin is the non-carbohydrate fraction of dietary fiber. The Huang study further identifies that each of these fiber components has differing effects on fecal bulk and transit time. The addition of cellulose has l i t t l e effect on transit time. Pectin and the . 41 hemicelluloses increase fecal weight due to their hydrophilic capacity. Lignin, which is hydrophobic, is constipating. Bingham (1979) found that B r i t i s h diets had a fourfold range of fiber intake, from 8 to 32 grams per day, with a mean daily intake of 19.9 grams. The sources were vegetables (41%), cereals (30%) and frui t or mixed sources (28%). The mean intakes of cellulose (4.7 g/day) and lignin (1.4 g/day) were minor compared to the non-cellulosic polysaccharides. Hexoses, pentoses and uremic acid contributed to a mean of 7.4 g, 3.3g and 3.0 g per day respectively. The pentose fraction, f e l t to have the most effect on fecal weight and.transit time was derived primarily from bran, wholewheat cereals, peas and green vegetables. The lignin fraction is the most efficient binder of bile acids. a Lignins are derived from the "skeletal" components of plants i.e. stalks, pulp or strings. Wheat straw, sugar cane, sugar beet pulp and the residues of celery, corn, lettuce, potatoes and string beans adsorbed large amounts of b i l e acids in in vitro studies. The addition of lignin to the diet has been found effective in reducing serum cholesterol presumably by increasing bile acid excretion. Hemicellulose adsorbs only small amounts of bile acid but increases fecal bile acid excretion. The mechanism for this has not yet been cl a r i f i e d . Pectin also increases fecal b i l e acid excretion and has been shown to decrease serum cholesterol. The daily ingestion of 15 grams of citrus pectin resulted in a 13% reduction of total serum cholesterol (Kay and Truswell, 1977) although Durrington (1976) suggested the pectin effect is short term and largely due to a reduction of low density lipoprotein 42 cholesterol in the serum. Cellulose, which increases fecal mass, has l i t t l e effect on serum l i p i d s . Bran which is principally hemicellulose (67% of i t s total dietary fiber) reduces the concentration of bile acids and neutral steroids through a dilution effect rather than binding of these components. Bran also increases transit time in the elderly. Serum cholesterol was not significantly altered by bran. The various components of fiber also have different effects in bacterial flora, bile acid degradation, and steroid metabolism. In rat experiments Klukfeld et al (1983) demonstrated that increased dietary b i l e salts resulted in increased c e l l turnover. Tumor frequency also was increased. Increased cholesterol in these diets resulted in an increase in the number of invasive tumors. In general, the stool of persons consuming a Western diet low in fiber contains more Bacteroides, which are very active in the degradation of b i l e acids. It also contains a higher concentration of steroids which are more degraded than is found in that of persons consuming African diets. Thus, the effects of dietary fiber are very complex. The effects may be due to increasing transit time thus reducing duration of exposure to fecal carcinogens, due to binding b i l e acids thus reducing the formation of potential carcinogens, due; to a solvent action that dilutes potential carcinogens or due to i t s impact on intestinal flora which decreases the degradation of bile salts (Kritchevsky, 1983). In each instance, the various components of dietary fiber appear responsible for differing effects. Metabolic studies with closely controlled identification of fiber 43 components and their sources (age and type of plant) are necessary to c l a r i f y the mechanism of effect of dietary fiber, i f indeed, i t is an important factor in the etiology of colon cancer. The protective effect of foods high in fiber may be that they are rich in a number of nutrients such as vitamin C. It has been suggested that large doses of Vitamin C diminish the number of colonic polyps (Bruce, 1982). Thus the protective value associated with vegetables may be attributable to their vitamin content. Bruce (1982) found that the concentration of the fecal mutagens of persons living on a Western diet was diminished i f those diets were suplemented with ascorbic acid (400mg/day)/^tocopherol (400 mg/day), or wheat bran or by reducing the fat and protein content of their diets. Ascorbic acid (Vitamin C) is a potent antioxidant, colon cancer is a disease of the older years. Considering that cancer may be initiated due to the accumulation of free radicals (Harman, 1972), ascorbic acid may act as an antioxidant, reducing the number of free radials. Further, the protective effect of ascorbic acid may be related ot the ingestion or formation of nitrosamines. Readily formed from common amines and n i t r i t e s , nitrosamines are ubiquitous in the environment (Lijinsky and Epstein, 1970). They develop in the s o i l and under conditions of food storage, e.g. holding of vegetables at room temperature (Jukes, 1977). The reaction varies with temperature and pH and is catalyzed by bacteria. Nitrites and nitrates are found in water and food, naturally and as an added preservative (Walker, 1975). Beets, carrots and spinach are a l l high in nitrates (Crosby, 1976). Adverse post harvest storage conditions 44 can substantially increase n i t r i t e content of vegetables, especially spinach. Although cured meats are the most significant dietary source (White, 1975), nitrosamines are also found in beer, some cheeses, nonfat dry milk and fish (Scanlan, 1983). Fine (1977) demonstrated the in vivo formation of nitrosamines after the ingestion of a meal of a bacon, spinach and tomato sandwich and a beer. Nitrosamines are present in the serum of man. Although there is some evidence suggesting an association between nitrosamines in salted fish and stomach cancer in Japan (Haenszel et a l , 1972) there has not yet been any direct causation, even in industrial settings, by these compounds of human cancer. Ascorbic acid is a potent antioxidant, i t inhibits the in vivo nitrosation of secondary and tertiary amines to form nitrosamines in the stomach (Fine, 1977). This may be a significant protective reaction i f ascorbic acid rich foods are consumed at the same time as n i t r i t e rich foods. This observation may or may not be relevant, as nitrosamines have not been associated with colon cancer. 45 SUMMARY OF DIETARY STUDIES Various dietary studies are summarized in the following pages for ease of comparison. Comparison of mortality and incidence rates for colon cancer between different countries suggests strongly that countries with the highest rates are those where the consumption of animal fat and total protein is higher, and that of total fiber less than in countries with lower rates. However, the results of case-control studies performed to date provide l i t t l e consistency. While some studies do suggest that patients with colon cancer give a history of consumption of lower quantities of certain vegetables than non-cases, these results have not been found in other studies. The situation in regard to fiber intake and protein intake is similarly unclear. One possible reason for the discrepancy between these two types of analyses is in the type of dietary survey conducted in the case-control studies. In each of the case-control studies, the dietary information taken relates to current or very recent diet derived from a relatively detailed history of dietary patterns in recent years. However, i t is likely that the time interval between the c l i n i c a l diagnosis of the tumor and the actions of the carcinogens and cocarcinogens which i n i t i a t e the tumor is in the order of decades. Further data is needed on potentially protective factors (e.g. vitamin C or vitamin A rich foods, the role of f r u i t and vegetable consumption) and on the effects of food preparation methods. Table 8 Summary of D i e t a r y Studies as Presented i n L i t e r a t u r e Review. Author/Year of P u b l i c a t i o n Type of Study Numbers S i g n i f i c a n t Findings  Wynder & Rhigematsu, Comparison of n a t i o n a l High c o r r e l a t i o n between d i e t a r y 1967 food consumption data - fat intake and colon cancer. Gregor, 1969 Comparison of n a t i o n a l food consumption data 28 Countries High c o r r e l a t i o n between i n t e s t i n a l cancer and animal p r o t e i n consump-tio n . No c o r r e l a t i o n for t o t a l p r o t e i n or cereal consumption. B u r k i t t , 1969 Comparison of n a t i o n a l - Incidence of colon cancer i n v e r s e l y food consumption data * related to consumption of d i e t a r y f i b e r Z o l d i v a r , 1976 Comparison of food C h i l e Low age-adjusted death rate f or disappearance data colon Ca i f high f i b e r , low saturated fat d i e t . Armstrong & D o l l , 1975 Comparison of n a t i o n a l food consumption data Decreased d i e t a r y f i b e r l i n k e d with increased colon cancer incidence. Meat intake c o r r e l a t e d with colon cancer. Howell, 1975 Comparison of per c a p i t a food disappearance Colon cancer m o r t a l i t y c o r r e l a t e d p o s i t i v e l y with t o t a l c a l o r i e s from meat, meat f a t , p r o t e i n and t o t a l f a t ; negative c o r r e l a t i o n with vegetables and vegetable protein consumption. Table 8 (cont'd) Summary of Dietary Studies as Presented in Literature Review. Type of Study Numbers Significant Findings Author/Year of Publication Finegold, 1975 Population observation U.S.A. Rates of colon cancer lower in 7th Day Adventists, however, also noted that serum cholesterols and fat intakes are lower in vegetarians. Lyon, 1976 Population observation U.S.A. Rates of colon cancer lower in Mormons who do not use tea, coffee, alcohol or tobacco. Finegold, 1977 Population observation U.S.A. Mormon consumption of beef comparable to general U.S. population Wynder, 1975 Comparison of per capita food disappearance Fats and o i l s consumption correlated positively with incidence of colon cancer. Higginson, 1966 Concentrated Study 340 Cases Colon cancer cases consumed more fried foods. No differences cheese, milk, dairy products, or raw vegetables. Wynder & Shigematsu, 1967 Case control study 791 Persons Colon cancer cases had higher soci-economic status, familial risk; Jewish persons at r i s k . No differences for eggs, milk, butter cheese, meat, yellow and green vegetables, f r u i t , juices, cereals, potatoes, smoked foods, salted foods, f ats. Table 8 (cont'd) Summary of Dietary Studies as Presented in Literature Review. Author/Year of P u b l i c a t i o n Type of, Study Numbers S i g n i f i c a n t Findings Haenzel, 1973 Case control study 174 Cases 357 controls Meat (beef) and legumes (pea pods string beans) were risk enhancing. Coffee, peanut butter, papaya, oranges, mushrooms, raw fi s h , pickled cucumbers, eggplant were protective for colon cancer. H i l l , 1975a Case Control Study 4 to 25 subjects Fecal concentration of tryptophan higher in colon Ca; incresses with dietary protein. MacLennan, 1978 High incidence vs. low incidence comparison Scandinavia High incidence associated with more white bread, meat, beer and less potatotes and milk. Phillips & Snowdon 1983 Cohort 21,295 Adventisits No association with meat but coffee associated with large bowel cancer. ... ,_. y Graham, 1978 Case Control Study 470 Cases 783 Controls No increased risk associated with beef or meat consumption. Decreased r i s k associated with increased consumption of vegetables (cabbage, brussel sprouts, b r o c c o l i ) . Table 8 (cont'd) Summary of Dietary Studies as Presented in Literature Review. Author/Year of Publication Type of Study Numbers Significant Findings Haenszel, Locke & Segi, 1980 Case control study 588 Cases 1176 controls Did not reproduce previous findings of association with beef, string beans, starches. Negative association found for cruciferous vegetables. Note: Analysis combined colon and rectal cancers. Rozen, 1981 High incidence vs. low incidence comparison - Fiber, carotene and ascorbic acid intakes higher in low incidence area. H i l l , 1975a Comparison of food disappearance data Partial correlation for fat not significant when cholesterol controlled; remained high for cholesterol intake and colon cancer. Modan, 1975 Case control study 198 age matched pairs Cases had a lower fiber consumption than controls. Jensen, 1974 Comparison of food consumption data Denmark and Finland Much higher colon cancer rates in Denmark. Both countries with high fat intakes. Finns consume more fiber, less meat, less fried meat. Reddy, 1983 High incidence vs intermediate incidence comparison 2 areas Scandinavia and in New York Protein and fat intakes similar. Higher fiber intake associated with lower incidence. Table 8 (cont'd) Summary of Dietary Studies as Presented in Literature Review. Author/Year of Publication Type of Study Numbers Significant Findings Jain, 1980 Case-control study 543 cases, 1,077 controls High risk associated with elevated intakes of fat, cholesterol, total protein. No association crude fiber or vitamin C. Kolonel, 1983 Prospective study 50,000 No association with fat intake. Dales, 1978 Case control study 99 age-sex matched pairs Cases consumed high fat and low fiber foods more frequently than controls. No significant relationship for beef consumption. Williams, 1981 Prospective Study 3192 Inverse corelation between serum cholesterol and colon cancer consis-tent after adjustment for age, alcohol consumption, smoking, education and relative weight. Mill e r , 1981 Case control study 133 age, sex matched pairs Inverse relationship of advanced colon tumors and serum cholesterol, Stemmerman, 1981 Prospective study 8,000 men Serum cholesterol levels below 180 mg/dl predicted higher rates of colon cancer 5 to 9.9 years after low serum cholesterol measurement. Table 8 (cont'd) Summary of Dietary Studies as Presented in Literature Review. Author/Year of Publication Type of Study Numbers Significant Findings ^ Cleave, 1956 Population Observation - High fiber diet increases fecal transit time. Wynder & Shigimatsu 1967 Case control study 791 persons No relation between colorectal cancer and bowel transit time. Tuyns, 1983 Case control study 1,246 patients Vitamin C rich food mediated the effect of alcohol or esophageal cancer. H i l l , 1975 Metabolic study 4 to 25 subjects Decreasing dietary fat in low fiber diet or no meat diet decreased fecal concentration of Addition of bran to diets reduced fecal steroids. i j -Crowthy, 1976 \ Metabolic Study Hong Kong No difference in fecal f l o r a noted between vegetarian and non-vegetarian Adventists. Kay & Truswell, 1977 Metabolic study Citrus pectin reduced serum cholesterol. Bruce, 1982 Metabolic study - Diet supplementation with ascorbic acid or wheat bran diminished fecal mutagens. 2.4 CARCINOGENESIS The ultimate end-point of a cancer-inducing exposure to a carcinogen is the appearance of a localized collection of cells which are undergoing inappropriate proliferation, are expanding locally and invading the surrounding tissue, and are releasing new cells that may generate new collections of proliferating cells in distart sites (Farber, 1984). Most cancers are heterogeneous c e l l populations with diverse c e l l structures, genetic makeup and biochemical properties. Cancer is not acquired as a single step but rather as the result of a stepwise progression of complex interactions, the major steps of which are identified in Figure 3. In the activation phase, an external stimulus f targets an internal organ. Initiation occurs when the carcinogen causes an irreversible genetic alteration in the target c e l l . This process appears to require a round of c e l l proliferation and regenerating tissue is particularly susceptible to in i t i a t i o n . Initiation seems to be a rare event during which a mutation or some type of genetic alteration occurs. The reactive derivates of carcinogens interact with DNA and other cellular constituents. Maskens (1981) identified a sequential process for malignant transformation in the large intestines of rats dosed with 1,2 dimethylhydrazine. The f i r s t change was caused by the carcinogen which has demonstrated alkylating activity in nucleic acids of the intestinal epithelium. Under conditions of i n i t i a t i o n , discrete new focal collections of a l l cells arise. These focal collections, called nodules, papillomas or 53 FIGURE 3 The Progressive Development of Cancer External Stimulus Internal Agent Activation Cellular Transformation Initiation Tumor Growth Promo t ion [Tumor Proliferation] Disability and Death Metastases Source: Sabine, 1979 54 polyps, are sites in which cancer can arise. Colon epithelial proliferation has been shown to be promoted by bile acids. The i n i t i a t i o n phase is followed by a promotional phase during which the focal collection of cells proliferate. Only a few of the initiated c e l l s proliferate. In colon cancer, the majority of the polyps often "disappear" and only a minority persist (Faber, 1984). Once the persistent nodule appears, no further external manipulations are required for cancer development, although there may be a further step between the persistent nodule and the appearance of a probable malignant neoplastic population. Of current general interest .is the role of the oncogene. Oncogenes may work in concert with each other or with viruses or other chemicals. They may play a role in the carcinogenic process or may be an expression of the advanced progression of the disease. There may be a considerable latency period between i n i t i a t i o n and tumor growth. Maskens (1981) determined that in rat colon cancer this process took an average of ninety three days. In humans, this process may take fifteen to twenty years or more. The mechanism of promotion or inhibition of this proliferation is yet to be explained. Advanced tumor growth follows, then perhaps local invasion and metastasis. If unchecked, the disease proceeds to produce d i s a b i l i t y and eventual death. The agent, or i n i t i a t o r , may be precarcinogen, which must be transformed to a carcinogenic element, or a carcinogen, which is primarily responsible for the cellular events that i n i t i a t e the cancerous transformation. The environment may contribute promotors either as 55 cocarcinogens which are not carcinogenic per se but combine with some other element to contribute directly to the process, or as procarcinogens which provide significant stimulation to the action of the carcinogenic element. Nutrition may function at any phase in this process: by contributing carcinogens, by making the host more susceptible to the agent or by protecting the host from an environmental agent or by promoting the growth of tumors. Indeed, nutritional factors are most lik e l y to be operative at several phases of the process, that i s , they make a multi-factorial contribution to the disease. Based on the review of literature, the possible nutritional contribution to the development of colon carcinogenesis is summarized in Figure 4. Most probably a carcinogen or precarcinogen is introduced through the diet either as a natural element of food or as a product of cooking, food handling or storage, or as a food additive. Candidates for the culprit include nitrates, n i t r i t e s , polycyclic aromatic hydrocarbons and a host of unknown p o s s i b i l i t i e s . Nutritional elements then may function to promote production of carcinogens in vivo (fat, meat) or to enhance the growth or spread of the tumor (bile acids, fat). Nutritional elements may also function to protect the host by reducing contact time for the potential carcinogens (fiber) by diluting carcinogen concentration (fiber) by maintaining tissue health (vitamin A and i t s analogues), by preventing in vivo carcinogen formation (ascorbic acid, j£> -carotenes), or by deactivating potential carcinogens (indoles), or affecting the immunocompetenee of the host. Chandra (1981) indicated obesity may diminish immunological competence. Vitamin C and Vitamin A may enhance i t (Tomita, 1983). 56 FIGURE 4 The Possi b l e Contribution of Di e t Related Factors to the Development of Colon Cancer A. Contributing carcinogens or precarcinogens - natural elements of plants products of cooking - products of food handling and storage food additives B. Substrate for production of carcinogens i n vivo ac t i o n of b a c t e r i a on steroid s i n gut - p o s s i b l e formation of nitrosamine i n stomach C. Modifying the e f f e c t of carcinogens - e f f e c t on length of contact with substrate ( f i b e r ) - e f f e c t on concentration: d i l u t i o n and adsorption ( f i b e r ) a l t e r a t i o n of b a c t e r i a l f l o r a ( f a t , meat) - increasing s e c r e t i o n of b i l e acids ( f a t ) which may function promotors D. Pr o t e c t i n g host - enhancement of t i s s u e health ( r e t i n o i d s ) i n h i b i t i o n of nitrosamine formation (Vitamin C) dea c t i v a t i o n of p o l y c y c l i c aromatic hydrocarbons (indoles) deactivation of free r a d i c a l s (Vitamin C,^-carotenes) - a f f e c t i n g immunocompetence (overweight, vitamin C) Based on: D o l l , 1979 57 The complexities of this process may serve to explain some of the apparent anomalies observed in the descriptive epidemiology. If beef fat is a promoter, the low incidence of cancer among Mormons may be due to their lack of exposure to in i t i a t o r s . The low incidence among Seventh Day Adventists may be due to lack of exposure to initiat o r s and promoters. The high fat consuming Finns may have limited contact with the initiators due to varying food consumption, handling or storage practices. The Bantu may have a low rate of colon cancer due to the low contact with carcinogens somewhat due to fiber consumption but also possibly due to food preparation and storage practices. The possibility of a considerable "latency" period between i n i t i a t i o n and promotion may also assist in explaining inconsistencies in the descriptive studies between current food consumption and colon cancer incidence or mortality. More attention must be given to the investigation of food habits over a considerable period of time i f relevant factors are to be identified in dietary epidemological studies. 58 2.5 METHODOLOGICAL ISSUES One of the principal complications in the investigation of this progressive development of disease suggests the need to investigate food habits which existed some years prior to the onset of symptoms. In a prospective study, Stemmerman (1981) found that an inverse association of serum cholesterol levels and colon cancer occurred up to ten years prior to diagnosis. This relationship was not significant in subjects diagnosed 10 years or more after the i n i t i a l prospective examination. The decreased serum cholesterol level may be an organ-specific response to undiagnosed cancer at an early stage of development. The two major approaches to the investigation of such a problem are the cohort (prospective) study and the m case-control (retrospective) study. In a cohort study, the groups of persons to be studied are defined in terms of characteristics manifest prior to the appearance of the disease under investigation. The study groups are then observed over a period of time to determine the frequency of disease among them. The investigator must wait for the disease to occur. This type of study is generally chosen i f the cohorts have undergone some unusual exposure of which the effects are to be evaluated (MacMahon & Pugh, 1970). The cohort study has the advantage of clear data on exposure as the investigators are able to measure prevalence and incidence directly. In the cohort study, the relative risk of getting the disease i f a factor is present or not is assessed. Prospective studies are very costly and obviously very time consuming. Outcome, i.e. diagnosis, can sometimes be unclear. Dropouts or 59 l o s t c ontacts may be great. Such s t u d i e s are v a l i d only i f large p o p u l a t i o n s can be observed over time. The cohort study i s not the method of choice i f the exposure to the p o t e n t i a l r i s k f a c t o r i s common. The c a s e - c o n t r o l study i s an i n q u i r y i n which groups of i n d i v i d u a l s w i t h the dis e a s e (cases) and those without the disease ( c o n t r o l s ) are compared w i t h respect to e x i s t i n g or past c h a r a c t e r i s t i c s judged to be r e l e v a n t to the e t i o l o g y of the disease (MacMahon and Pugh, 1970). I n the c a s e - c o n t r o l study the p r o b a b i l i t y of having a f a c t o r given the disease being present, i s assessed versus the p r o b a b i l i t y of having that f a c t o r given the disease i s not present. A major advantage i s that the c a s e - c o n t r o l study can be done r e l a t i v e l y r a p i d l y and i n e x p e n s i v e l y compared to the p r o s p e c t i v e study. The c a s e - c o n t r o l study i s w e l l s u i t e d to the study of r a r e d i s e a s e s . The gr e a t e s t s t r e n g t h of the c a s e - c o n t r o l study i s that i t allows the e v a l u a t i o n of s e v e r a l d i f f e r e n t e t i o l o g i c a l f a c t o r s both as independent and i n t e r a c t i n g causes. One l i m i t a t i o n of the c a s e - c o n t r o l study i s that i t i s u n s a t i s f a c t o r i l y used f o r the study of r a r e exposures (Dorn, 1959). A second l i m i t a t i o n i s that the c a s e - c o n t r o l s t u d i e s a l l o w f o r the e s t i m a t i o n of r e l a t i v e r a t e s but not absolute r a t e s . I f inc i d e n c e or prevalence surveys are inc l u d e d i n a c a s e - c o n t r o l study, r i s k f a c t o r s p e c i f i c absolute r a t e s may be i n f e r r e d . A l s o , the c a s e - c o n t r o l study i s dependent on memory of the su b j e c t s r e g a r d i n g t h e i r exposure to r i s k enhancing or p r o t e c t i v e v a r i a b l e s . Depending on the l e v e l of d e t a i l i n v o l v e d , t h i s may s e r i o u s l y j e o p a r d i z e 60 the r e l i a b i l i t y of the study data. The most serious limitation of case-control studies is their susceptibility to bias, especially selection bias with regard to the selection of cases and controls. Cases selected must represent the disease population to ensure the generalizability of the results. Case definition must be clear, with the cases being a group of individuals who have the most homogeneous disease condition possible. Concise definition of cases aids in the interpretation of the study results, as specific variables can be related to specific disease entities. Thus, i t is important in studies of large bowel cancer to differentiate colon cancer and rectal cancer, to include only colon cancer cases and not rectal cancer, as the etiology of these diseases may * be very different. Cases are generally chosen from the population of persons having the disease seen at a particular c l i n i c or group of c l i n i c s , or from the population of persons with the disease found in a geographic area. Although the former is more common and less expensive, the latter is preferable. This procedure is more time consuming, as considerable effort must be expended to locate the affected individuals. It does, however, reduce the bias arising from factors which may have influenced affected individuals to a particular c l i n i c . Also i t allows for greater generalizability of results. . MacMahon and Pugh (1970) suggest that i t is preferable that cases be limited to those who were newly diagnosed within a specific period. The inclusion of a l l prevalent cases would greatly increase the number of 61 available cases, however, the long-term survivors may represent a different population than the incident cases. In this instance, where colon cancer two and five year survival rates are low, recent incident cases would be the feasible cases most representative of the disease population. Enhancing representativeness reduces selection bias. Control selection is also c r i t i c a l to the study design. The objective in selection of controls is to ensure that both cases and controls would have had equal opportunities to sustain the exposure of interest (Cole, 1978). Controls are chosen such that data can be collected in a similar manner and setting to that for cases and such that they represent a similar population to that of the population from which the cases arose i.e. who do not have the disease. Controls may be matched to enhance general similarity, however, care must be taken not to overmatch and perhaps miss the effect of an important variable. Controls are generally chosen from the same frame as the cases, i.e. the same geographical area or the same c l i n i c . Relatives of cases, either spouses or siblings, or associates of cases may also serve as controls. In a food habits study, choice of spouses as controls would very likely result in overmatching. Overmatching w i l l bias the measure of effect towards the null value. Siblings may similarly confound as controls i f the disease has a component of genetic susceptibility. To avoid overmatching, cases and control should not be matched with regard to variables that may be intermediate in the casual pathway between the factor under study and the disease (Miettinen, 1970). Nor should factors be matched that are related to the suspected cause but not-the 62 disease. For example, in a study of birth weight and nutritional intake, i f participants were matched for socioeconomic status, overmatching may occur as socioeconomic status impacts nutritional intake. Such variables should be taken into account in the analysis. A major consideration in this study is the size of the selected sample. According to Walter (1977), a retrospective study with a sample size of 100 conducted on a population with a factor prevalent in f i f t y per cent of the population and selected acceptable Type 1 C?^  and Type I I (^ ) errors of 0.05 w i l l be able to detect a relative risk of 2.84 or greater. A sample size of 200 in similar circumstances w i l l detect a relative risk of 2.07 and greater. The selection of sample size in this study w i l l balance funding pr a c t i c a l i t i e s and the concern to include only recent incident cases with the potential for detecting smaller relative risks. The homogeneity of diets in certain cultures may limit findings. As illustrated by Walter (1977), a sample size of 100 (<V/3 = 0.05) w i l l detect a relative risk of 2.84 i f f i f t y per cent of the population is exposed to a given factor. If seventy five per cent of the population is exposed, a sample size of 100 w i l l only detect a risk greater than 4.49. If ninety per cent of the population were exposed, a sample of 200 would be necessary to detect a risk of 6.50. Most studies show there is sufficient individual variation in diets or food preparation preferences for differences to be noted (Mettlin and Graham, 1979). The desir a b i l i t y of a large sample size must be balanced with the practicalities of available time and funding. Another significant but as yet controversial issue is the number of control groups used (Cole, 1978). There are two general perspectives on 63 this issue. One is that a case-control study should have one control group, the one best suited to the need of that particular study. In this instance, a second control group would be added only i f the f i r s t had some known or suspected deficiency that would be offset by the second control group. The second perspective is that every case control study should have at least two control groups. If the results are similar, one group would validate the results of the other (Gail, 1976). Further research is needed in this area as differing results may not invalidate one another but rather help explain one another, pointing out some aspect of l i f e - s t y l e of one control group versus the other that may contribute to the disease process. There is a very small increase in s t a t i s t i c a l power as the ratio of cases to control increases beyond four (Walters, 1977). Size of control groups should be equal i f access to data is similar for both cases and controls. If the number of cases is severely limited, the number of controls may be increased. Another area of methodological interest is the determination of information on exposure, its v a l i d i t y and r e l i a b i l i t y . Validity refers to the extent to which data collected reflect the "true" situation, that is the extent to which subjects in a case-control study are correctly classified as to the presence or absence, or level, of an exposure of interest (McMahon and Pugh, 1970) . Validating dietary information is a d i f f i c u l t process, as most methods selected to validate information affect usual food practices. For example, an observer weighing food may cause a binge eater to forego a dozen donuts. For this reason, Lee, Kolonel and Hankin (1983) suggest that 64 v a l i d i t y in dietary studies relates to the a b i l i t y of a given method to be interchangeable with a method of greater acceptance or one that has greater face v a l i d i t y . For example, a food frequency method is validated compared to a quantitative dietary method. They caution that the v a l i d i t y of a dietary method depends on the specific situation. Bazzarre and Yuhas (1983) further expanded this concept of v a l i d i t y related to situation to caution that although a food record may assess actual nutrient intake on a specific day or days more accurately, the diet history may be a more valid assessment of usual food intake in a given time frame. More rigorous evaluation with larger populations are required to define the best alternatives for collecting dietary data for use to answer a variety of epidemiological questions. R e l i a b i l i t y or reproducibility is equally d i f f i c u l t to measure, primarily because of individual variations in.food consumption on a daily basis. R e l i a b i l i t y can be enhanced through the use of cross-checks in the questions and food frequency l i s t s and in test/retest situations. Evidence of comparability is enhanced by presentation of data on response rates and duration of interviews. Interviewers may be asked to classify respondents with regard to r e l i a b i l i t y of responses; similarities of the d i s t r i b i t i o n of these classifications should be evident. In addition, "dummy" variables not considered to be relevant to the etiology of the disease may be included, analyzed and compared (MacMahon and Pugh, 1970). Dietary information in retrospective studies is collected via two general modes: examination of population consumption data and assessment of 65 individual patterns of consumption. As expressed previously, examination of population data is of questionable value other than to predict general trends. Foods have multiple nutritional qualities; vegetables, for example, are often high in fiber, vitamin A, vitamin C, and trace elements which complicates analysis of data for causation. Nutrient values cannot be inferred accurately without detailed information on food preparation and handling practices. In the case-control study, individual dietary assessment must be undertaken. As opposed to prospective studies which can undertake inventory or direct measurement of food consumption, retrospective data collection is limited to dietary r e c a l l , generally either a 24 hour r e c a l l , a diet history or a food frequency record (Bassarre and Myers, 1979). In the 24 hour r e c a l l , individuals are asked to re c a l l actual food consumption for the past day. Both qualitative and quantitative data are recorded. This method is considered to be reliable and accurate for the assessment of actual food intake (Graham, 1980), however, i t gives a limited representation of a typical food consumption pattern for an individual. It is useful in very large population studies to reflect the composite consumption pattern of population segments. Dietary diaries, three or seven days, and the twenty-four hour rec a l l are unsuitable for use in retrospective case-control studies (Mann, 1962). These techniques rely too heavily on the subject having the a b i l i t y to remember an inordinate amount of detail regarding his or her food habits. Further once data is collected there is no opportunity to evaluate the adequacy of information provided e.g. i f bread is without spread, was i t 66 truly eaten in this fashion or was the spread omitted from mention? Thirdly, methods like the twenty four hour recall are very much restricted in terms of representativeness of time frame due to the great variation in food habits on a day-to-day basis. Thus, the diet history method was chosen rather than the dietary r e c a l l . The dietary history is the method by which subjects are requested to recall a general pattern of food consumption based on regular frequency of consumption of selected food items. In the more sophisticated Burke method, repeated cross checks of the food frequency l i s t are employed to enhance va l i d i t y (Burke, 1947). Dietary histories have been validated by Hankin et al (1975), and Jain et al (1980). In the Hankin study (1975), the v a l i d i t y of the dietary recall questionnaire was assessed by comparing f the intakes identified for a seven day diary of food frequencies and amounts of thirty-three items with a subsequent re c a l l of the same items for the previous day and the previous seven days. Photographs were used to assist persons to estimate portion sizes. Overall, agreement as to whether an item was eaten or not eaten was only marginally less for the seven days than for the one day. Agreement was higher for items habitually eaten or associated with special events. Quantitative estimates enhanced the valid i t y of the data. The study concluded that a seven day quantitative r e c a l l of particular food items is a reasonable choice of method for studies of large groups. Jain et al (1980) compared mean estimates of thirteen nutrients calculated from detailed diet history questionnaires and from food records kept for a 30 day period. Positive correlations were found for a l l ) 67 nutrients and for food groups.. Most importantly for a case-control group, the questionaire was repeated six months later. Analysis determined that high correlation were obtained for most nutrients. Cases showed lower correlations than controls probably due to change in diet. This study concluded that the dietary history was a valid and reliable method of estimating dietary patterns. Using the food frequency method, subjects are asked to recall the frequency of consumption of various groups of food as representative of a certain period of time. The extent of the food l i s t is variable, commonly encompassing twenty to thirty items or groups of foods. The frequencies, recorded as number of times a food is consumed per unit time (day or week), may be analyzed directly or translated into a food score which is used in analysis. In investigations of disease specific morbidity, this is an efficient method as data can be collected only on those dietary factors suspected in the disease etiology (Bazzarre and Myers, 1979). As the information collected i s of a more standardized format, the interview can be less time consuming and more uniform with less bias. The interviewer training requirement is less extensive; indeed an "interview" per se may not be necessary. The questionnaire could be self-administered. Balough e t a l (1968), in preparation for a prospective study, assessed the validity of a short dietary questionnaire (food frequency l i s t ) compared with a detailed diet history modified from Burke and a weighted record of food intake. Results were correlated for total fat, animal protein, saturated fatty acids, polyunsaturated fatty acids and total calories. The weighed record and short questionnaire showed 68 correlation coefficients ranging from a high of 0.94 for total fat to a low of 0.69 for polyunsaturated fat. The authors concluded that the short dietary questionnaire (food frequency l i s t ) is l i k e l y to provide good descriptive data and also to be useful to detect associations between dietary habits and disease status. A l l of these re c a l l methods are subject to certain limitations. A l l request the subject to recall events which took place sometime earlier, thus a l l are susceptible to memory. Indeed, the a b i l i t y of subjects to reca l l dietary practices over time may be a most serious limitation of retrospective dietary studies. One recent methodological study addressed this issue. Byers et al (1983) conducted a study to assess the r e l i a b i l i t y of subjectability to rec a l l food consumption from the distant past. Subjects who completed dietary interviews in 195 7 and 1965 were re-interviewed in 1982 to assess their usual frequency of consumption of selected food items. Mean frequency of intake per month indicated general agreement between original, recalled and current intakes. For example, the respective values of bread were 29.4 ±0.21, 27.8 ±0.45, and 26.3 +0.61; for carrots the values were 6.0 ± 0.39, 5.9 ± 0.38 and 6.3 + 0.38 for the original, recalled and current intakes respectively. There was a stronger correlation for a calculated vitamin A index than for individual food items. Females had higher levels of agreement between recalled diets and originally recorded diets than did males. There were no apparent differences in the accuracy of rec a l l by age of subject. Memory is not' the only limitation of dietary recall methods. A l l of these methods presume the subject were aware of the content of their diet 69 and the manner in which i t was prepared. This may not be the case for those who frequent restaurants or are not involved in food preparation. Pretesting can reduce the number of food items which yield results of doubtful value. Salt intake is a particularly d i f f i c u l t item to assess. In addition, frequency judgments may be inaccurate. Individuals may "forget" poor food choices and/or exaggerate intake of foods thought to be nutritionally good. Internal crosschecks can be made on some foods or types of foods to assess this. Persons who are colon cancer cases may be more susceptible to gastrointestinal complaints. Then, the disease under investigation may influence current food practices. Cross-checks can be made within the disease history. Subjects should be asked to identify food pattern changes w resulting from disease. This may be a particularly important factor in the investigation of colon cancer, as most research is done on current food habits which may be altered as a result of the disease. In fact, patient's dietary changes may precede the appearance of c l i n i c a l symptoms. In conclusion, further investigation is necessary to c l a r i f y some of these methodological issues and to provide more data on the relationship of diet and colon cancer. New studies should investigate previous time frames, replicate previous findings and determine i f a dose-response relationship exists. These findings must be congruent with in vitro and animal studies. Metabolic studies are required to assist in the ver i f i c a t i o n of the involvement of separate dietary components in the etiology of the disease. The goal of a l l of this research would be to learn enough about the role of diet in this disease process to be able to control or limit the disease. 70 CHAPTER 3 METHODOLOGY A case-control study was conducted by interviewing 111 colon cancer cases and 92 controls of similar age (+ 5 years) and sex. The extensive dietary interview probed past as well as current food habits and related factors, to examine the significance of these factors in the eitology of the disease in each of three major l i f e stages. 3.1 POPULATION/SAMPLE The geographical area of interest was the province of British Columbia. The total number of live cases having cancer of the colon in 1974 was 915 males and 1166 females. The disease is rarely seen in persons under age 35, indicating a long-term process is involved in the disease. (The mean age of death for cancer of the large bowel is 69.5 for males and 70.5 for females). The most productive group for this investigation of dietary information was identified as those who have recently been diagnosed as having colon cancer and who are in the age-group 30-75. Only recently diagnosed cases were included, as food habits may change dramatically as the disease progresses. Beyond age 75, other disease entities may be complicating factors and the r e l i a b i l i t y of dietary information especially that, pertaining to previous food habits, might be questionable due to 71 problems with r e c a l l . The number of new cases diagnosed as having colon cancer in 1978 was 780, 501 of whom were between the ages of 30 and 75 years (Table 9). In this investigation, the sample consisted of a l l men and women aged 30-75, who had been diagnosed as having colon cancer in a nine-month period from September, 1981 to May, 1982. The sample was obtained from the case records on f i l e with the B r i t i s h Columbia Cancer Registry. The delay between the date of diagnosis and the date of interview was kept short (less than six months) to ensure that the sample would not be biased in that i t would represent longer term survivors only. Rectal cancer cases were excluded from this study as the factors involved in this disease d i f f e r markedly; sex ratios, age-adjusted incidence rates and geographical distribution of the two diseases are different. For practical reasons, the sample was limited to those persons living in the Lower Mainland and Southern Vancouver Island. More than eighty per cent of the live notifications of colon cancer in British Columbia are located in this area (Table 10). It was anticipated that approximately forty-five percent of the population of cases could be expected to be lost. About fifteen per cent would be deceased or too i l l to respond to the interview. Ten percent were expected to refuse to participate; ten per cent would be disqualified due to language barriers and/or mental state; ten per cent would have moved or otherwise been unable to be contacted. The assumptions were based on other similar studies (Williams, 1977). The data were examined to determine the effect of this non-response 72 bias. To determine i f the sample was representative of the case population, the comparability of age distribution and cancer sites in the sample and case population were assessed. 73 TABLE 9 Age Distribution of the New Cases of Malignant Neoplams of the Large Intestine in B.C. in 1978. AGE MALES FEMALES Group NO. Cum. Total No. Cum. Tot 20-29 1 1 0 0 30-34 1 2 1 1 35-39 3 5 6 7 40-44 5 10 2 9 45-49 17 27 17 26 50-54 15 * 42 22 48 55-59 30 72 38 86 60-64 39 111 43 129 65-69 50 161 69 198 70-74 65 226 54 252 75-79 52 278 64 316 80-84 38 316 43 359 85+ 48 364 59 416 Source: Cancer Control Agency of B.C. - Cancer Registry 74 TABLE 10 Geographic Distribution of Live Notifications of Colon Cancer in British Columbia in 1977 by School D i s t r i c t s . SCHOOL DISTRICT NUMBER SCHOOL DISTRICT NUMBER OF CASES 7 Nelson 5 11 T r a i l 10 15 Penticton 1 22 Vernon 1 24 Central Okanagan 33 27 Kamloops 13 27 Cariboo 4 31 Merritt 1 32 Hope 2* 33 Chilliwack 7* 34 Abbotsford 8* 35 Langley 9* 36 Surrey 30* 37 Delta 1* 38 Richmond 11* 39 Vancouver 170* 40 New Westminster 49* 41 Burnaby 120* 42 ,Maple Ridge 11* 43 Coquitlam 5* 46 Sunshine Coast 1 47 Powell River 7 48 Howe Sound 1 52 Prince Rupert 5 57 Prince George 15 61 Greater Victoria 100* 62 Sooke 4* 63 Saanich 2* 64 Gulf Islands 4* 65 Cowichan 1* 67 Naniamo 19* 69 Qualicum 3* 70 Alberni 3 71 Courtenay 3 72 Campbell River 2 75 Mission 3 80 Kitimat 1 88 Terrace 3 89 Shuswap 4 Total 693 * Indicates those cases located in the Lower Mainland and Southern Vancouver Island (80.2%). Source: Cancer Control Agency of B.C. - Cancer Registry 75 Controls were selected from the same geographical frame, that i s , the Lower Mainland and Southern Vancouver Island. In the design chosen, controls were to be friends of cases and of the same sex and similar age (+ 5 years). Friends were chosen to be more appropriate controls than other hospitalized persons to reduce potential confounding due to other disease conditions. Friends were selected in preference to relatives to reduce the potential of confounding should there be a component of genetic susceptibility to the disease. It was thought that the selection of relatives might also result in overmatching for dietary variables. Friends would be more l i k e l y to have a greater v a r i a b i l i t y in food consumption than immediate family. However, friends are less likely to be different in food consumption compared to cases than persons in the general population. This similarity may cause significant variables to be missed in the analysis. The selection of friends as controls proved to yield limited results. Persons in this age category appeared to be somewhat isolated; less than half could or were willing to suggest the names of friends to participate in the study. The three most frequently cited reasons were not having any friends, not having any friends thought to be willing to participate, and a wish for privacy regarding the disease. As the subjects were contacted through family physicians, a number of these physicians were approached to provide the names of patients to serve as controls. Physicians were given a random selection of age-sex categories in which to provide potential names e.g. male born 1919, female born 1925. Restrictions for inclusion were that the patient would be considered generally healthy and would have no known nor suspected cancer 76 of any kind. The author was able to observe in two instances that nurses randomly pulled charts to find suitable candidates which were then cleared for contact with the physician. The objective of the study was to obtain information on at least 100 colon cancer cases and 100 controls of similar age (+ 5 years) and sex. As the sex ratio for the disease is near unity, an approximately equivalent number of men and women were to be included in the study. 77 3.2 INSTRUMENTS The l i t e r a t u r e i d e n t i f i e d a number of d i e t a r y fators p o t e n t i a l l y involved i n the causation and/or prevention of colon cancer. Food items were selected f o r i n c l u s i o n i n t h i s study to represent a v a r i e t y of p r i n c i p a l hypotheses f o r d i e t a r y involvement sources of p o t e n t i a l carcinogens, p o t e n t i a l promoters and pro t e c t i v e f a c t o r s . This study was concerned p r i n c i p a l l y with the e f f e c t s of d i e t over time thus, an i n i t i a l interview schedule was designed to c o l l e c t data on selected biodemographical v a r i a b l e s and on di e t a r y p r a c t i c e s i n three time frames: current p r e - i l l n e s s period, the middle years of l i f e and the teenage years. The schedule included a r e s i d e n t i a l h i s t o r y , a h i s t o r y of occupations and exposures to p o t e n t i a l l y r i s k y substances i n both home and work environments, as w e l l as questions on food habits and p r a c t i c e s and an extensive food frequency l i s t . The food frequency l i s t was chosen as the method for c o l l e c t i o n of d i e t a r y data as t h i s was considered to be the most appropriate method for the i n v e s t i g a t i o n of usual intakes over the three s i g n i f i c a n t time periods. Photograhic i l l u s t r a t i o n s were used to a s s i s t the p a r t i c i p a n t s i n estimating p o r t i o n s i z e s . The instrument was pretested with twenty-five persons aged 63-68 who were members of an a c t i v e community center i n V i c t o r i a . Results from the pretest i n d i c a t e d that the proposed schedule was too long and time consuming. When the interview schedule was redesigned (Appendix 1), the food frequency l i s t was shortened and the food portion l i s t was reduced. 78 P a r t i c i p a n t suggestions l e d to the replacement of the b l a c k and white photographs w i t h c o l o r r e productions (Appendix 2) f o r the study i t s e l f . The p o r t i o n s i z e e s t i m a t i o n s were used to a s s i s t p a r t i c i p a n t s to concentrate r a t h e r than to attempt to q u a n t i f y i n t a k e and were not used i n the data a n a l y s i s . D e t a i l e d questions on food p r a c t i c e s r e l a t e d to a number of food items were c o l l e c t e d i n a n t i c i p a t i o n of in-depth i n v e s t i g a t i o n should these food items prove s i g n i f i c a n t i n the a n a l y s i s . For example, questions were asked r e l a t e d to type of bread, type of f a t used as a spread, type of f a t used i n f r y i n g , and meat preferences. 79 3.3 DATA COLLECTION The l i s t of potential cases was received by the investigator who contacted the cited physician to obtain permission, cooperation and referral prior to subject contact. Many cited physicians were not family physicians of the cases; secondary and tertiary referrals were necessary to obtain permission to contact the cases. The subjects were f i r s t contacted by letter to request cooperation (Appendix 3). A telephone c a l l was made to establish an interview time of mutual convenience. A personal interview was conducted in the subject's home by a trained interviewer. To reduce interview bias, interviewers were "blind", i.e. they did not know i f interviewing a case or a control un t i l the end of the interview. At that time interviewers opened an envelope that indicated whether or not to request referral of friends' names as potential controls. If participants indicated any discussion of illness prior to the collection of the dietary information, they were requested to withhold this information un t i l later in the interview. In only three instances was the "blindness" of the interviewer removed by the case revealing they had had cancer. In two instances, interviews on potential controls were completed before i t was determined in the medical history that these individuals had had cancer unknown to the friends who had referred them for the study. Controls were contacted f i r s t by letter, then by telephone to arrange an interview. Interviews were conducted in subjects' homes in a manner 80 s i m i l a r to that f o r the cases. I f p a r t i c i p a n t s had made any recent changes i n food h a b i t s , d i e t a r y i n f o r m a t i o n was c o l l e c t e d r e l a t i n g to food p r a c t i c e s approximately one year previous. The same procedure was used f o r both cases and c o n t r o l s . 81 3.4 PERSONNEL A total of eight interviewers were involved in the data collection, two on Vancouver Island and six in the Lower Mainland. A l l were university or college graduates. Two were graduates in nutrition and dietetics; one from biology; two from marketing background; two had graduated in social work and psychology areas; one was a law student. A l l but one of the graduate students in nutrition had experience in conducting study interviews. Each was given a minimum of two and one half hours of i n i t i a l training plus a followup session after an interview of a non-subject. The briefing included a short review of basic nutrition terminology and an extensive and comprehensive review of the total interview schedule including any potential areas of d i f f i c u l t y . The interview schedule was detailed and standardization was emphasized. Interviewers were cautioned to record any information of which they were unsure (e.g. the brand of a margarine) rather than risk interpretation (e.g. judging i f i t were saturated or unsaturated). Particular attention was paid to rehearsing interviewers in establishing a time frame ten to fifteen years previously. Fortunately, this time frame, 1967-72, was readily identified by most participants as the year 1967 was Canada's Centenary. None of the interviewers reported any d i f f i c u l t i e s with the interview schedule. Interviewers were recompensed for the interview and travel expenses. 82 3.5 DATA ANALYSIS On receipt of the schedules from the interviewers, each was checked for completeness. Incomplete questionnaires were eliminated from the analysis. Only two case and two control interviews were eliminated from the study due to incompleteness of dietary information in the previous time frames. Data was coded on the questionnaire and transferred to coding sheets. Data was keypunched onto tape by a commercial firm which verified the punches. Data was rechecked by the investigator for accuracy in preparation for analysis. Comparisons of age and cancer site distributions were made to assses the v a l i d i t y of the sample to represent the case population. Further case control comparisons were made of the duration of the interview and of the interviwers' assessment of response r e l i a b i l i t y . Frequency distributions were calculated using the FREQUENCIES of the S t a t i s t i c a l Package for the Social Sciences (SPSS) as ut i l i z e d by the University of British Columbia (Lai, 1983). As a result, data was categorized and calculations of frequencies repeated on the categorical data. Food scores were generated for groups of foods by tallying frequencies of the selected groups of items cited in the food frequency l i s t . Frequency data including means and standard deviations were generated for the data as well as selected cross tabulations using the SPSS program CROSSTABS. 83 Relative risk estimates (Mantel-Haenszel) were calculated for selected dietary factors and selected biodemographic variables using RELRISK (S p i n e l l i , 1983). Given that a significant pattern emerges, these case-control comparisons of relative risk were to be adjusted for the effects of various combinations of multiple covariables including related environmental factors, e.g. frequency of beef consumption, use. of fried foods, and smoking practices. As the dependent variable was dichotomous the popular SPSSrANOVA program was not suitable for this analysis. The data yielded a series of independent factors which had a relative risk for colon cancer which were reviewed in light of current knowlege of the etiology of the disease. 84 CHAPTER 4 RESULTS AND DISCUSSION 4.1 RESPONSE RATE A total of 356 names were identified by the Central Registry as colon cancer cases which could potentially participate in the study. On elimination of duplicates, the l i s t yielded 323 cases. Before interviews could be arranged, 67 persons died leaving a total of 256 potential live cases (Table 11). In order to contact these live cases, 217 physicians were contacted. Many were secondary or tertiary contacts. A number of cases were ineligible because the physician refused contact citing the patients's poor health or incapability or because the physician could not be reached or did not know the patient. 4.3% of the patients had moved out of the area. Patient refusals and non-responses were both less than ten percent. Two patients died between f i r s t contact and the scheduled interview date. A total of 115 interviews were conducted yielding 111 usable interview schedules (Tables 11, 12). The total usable interviews (111) represent 34.4% of the actual total cases listed (323). The number of potential controls identified was 149, 38.3% from friends, 61.7% from physicians. Approximately 12.1% refused participation and a further 13.4% did not respond to contact indicating a slightly higher rate of disinterest among potential controls than cases. Of the total number of potential controls who either did not respond to contact or were unavailable, there were 8 out of 20 (40%) non-responders 85 and 5 out of 12 (42%) unavailable from the friends-source controls. Slightly more of those controls who refused to participate (18) were from the friends-source (10 or 56%) than were from the physician-source group. Interviews were completed on 97 controls, 92 of which were usable. Two of the clients had had cancer of which their friends had been unaware (Table 13). The response rate is similar for controls and for post contact cases. 86 TABLE 11 Response Rate Results: Pre-Contact Cases Total Cases Listed September, 1981 - May, 1982 Duplicates Actual Total Cases Deceased Before Contact Total Live Cases Number of Physicians Permission Refused by Physician (too i l l , incapable) , Physician Did Not Respond to Contact Unable to Locate Family Physician (moved, locum) Physician Lost Contact with Patient Patient unknown to physician Patient Moved out of Area TOTAL E l i g i b l e to Interview Number 356 33 323 67 256 217 39 13 9 13 4 14 164 . Percentages 20.7% of Actual Total Cases % of Live Cases 12.0% 4.0% 2.8% 4.0% 1.2% 4.3% 50.7% 87 TABLE 12 Response Rate Results: Post Contact Cases Number Percentages of El i g i b l e Cases Total cases presumed eli g i b l e 164 for interview Patient did not respond to contact 14 8.5% Patient refused 15 9.1% Patient too i l l 13 7.9% Patient unavailable (e.g. vacation) 5 3.0% Patient died before interview 2 1.2% TOTAL interviews completed 115 70.1% a Clients outside age range (too old) 2 1.2% Interviews incomplete . 2 1.2% TOTAL cases used as sample 111 67.7% 88 TABLE 13 Response Rate Results: Controls Total number potential controls identified Friends Via physicians Did not respond to contact Unavailable (vacation, etc) Refused participation Too i l l TOTAL control interviewed. Interviews incomplete Controls had cancer Client outside age range (too old) TOTAL controls sample Number 149 57 92 20 12 18 2 97 2 2 1 92 Percentages 38.3% 61.7% 13.4% 8.1% 12.1% 1.3% 65.1% 1.3% 1.3% 0.7% 61.7% 89 4.2 DESCRIPTION OF THE SAMPLE The sample was assessed to determine the extent of response bias based on the distribution of cancer sites, sex, age, and place of residence. Cancer Sites As illustrated in Table 14, the pattern of distribution of cancer sites was similar in both the cases on which interviews were completed and the total sample of cases identified from the Registry within the study time period. The most frequent site for both groups was the sigmoid colon (153.3) acounting for 30.4% of the total sample and 34.7% of the completed interviews. Sex There were 62 women (55.9%) and 49 men with colon cancer in the "usable-interviewed" case group. Forty eight women and forty-four men constituted the control group. The interviewed group reflected the sex composition of the total number of potential cases which was 58 percent female. Slightly more women were willing to participate in the study than men. Age The percentage distribution for the total potential of cases in the following age categories 40-49, 50-59, 60-69, 70-75 were 7.7, 16.3, 47.7 and 28.2 percent respectively. The distribution for cases for which interviews were completed was 3.6, 11.7, 44.1 and 40.5 per cent for the same categories, indicating this study had a slightly older population than 90 disease and may die earlier. In this longer surviving population, the risk factor exposure may have been less intensive. Differences may be missed. The young may be less l i k e l y to be interested in participation in such a study. Place of Residence in B.C. Places of residence in Bri t i s h Columbia were divided into three general categories: the Vancouver area, the Valley, and the Victoria area. The percentage distribution for the total potential cases in each of these areas was 55.7, 26.6 and 17.6 respectively. The f i n a l interviewed sample closely reflected the distribution with 55.9, 27.0 and 17.1 per cent of the interviews completed in the three areas. Thus, the sample of cases in the study is considered to be representative of the total number of cases identified based on distribution of cancer sites, sex, and place of residence in Bri t i s h Columbia. FRIENDS VS PHYSICIANS SOURCE CONTROLS In addition, the data analysis related to the principal hypotheses was undertaken comparing the friends-source and physician-source controls. A sampling of the results of this analysis appears in Appendix 4 in the Raw Data. None of the variables were significantly different for friends-source versus physician-source controls. 91 TABLE 14 Distribution of Cancer Sites in Total Cases Identified and Sample on which Interviews were Completed Cancer Sites by Sample Distribution ICD Classification Total Cases Identified Completed Interviews Number % Number % 153.0 Hepatic Flexure 3 0.9 2 1.7 153.1 Transverse Colon 25 7.7 9 7.8 153.2 Descending Colon 24 7.7 9 7.8 153.3 Sigmoid Colon 98 30.4 40 34.7 153.4 Cecum 53 16.4 16 13.9 153.5 Appendix m 1 0.3 1 0.8 153,6 Ascending Colon 56 17.3 19 16.5 153.7 Splenic Flexure 15 4.6 3 2.6 153.9 Colon, NOS 48 14.9 16 13.9 TOTAL 323 100 115 100 92 4.3 HYPOTHESES TESTING 4.3.1 BIODEMOGRAPHIC VARIABLES The biodemographic variables on which data was collected included: principal area of residence (RESMAJ), urban residence (URBAN % ) , level of education (EDUC), faith (FAITH), parental country of origin (RESMOM, RESDAD), marital status (MARSTAT), number of years married (YRSMAR), major occupation (MAJOCC), familial history of cancer (MOMCA, DADCA, BROCA, SISCA), personal disease history for heart disease (CVD), intestinal polyps (POLYPS), ulcerative c o l i t i s (UCOL), Crohn's disease (CROHN), appendicitis (APPEN), hemorrhoids (HEM), varicose veins (W), d i v e r t i c u l i t i s (DIVERT), peptic ulcer (ULCER), smoking history (SMOKE), obesity history (OVRWT), stress in the middle years (STRESS) and exposure to hazardous chemicals. The results are summarized in Tables 15 to 22. Those variables which were significant are highlighed below. The summary tables e.g. Table 15 include a l l values for the variables; the highlight tables e.g. Table 18 include only the data for the relevant categories under discussion, thus the p-values may d i f f e r slightly between the two sets of tables. The raw data is included in Appendix 4. Of the 24 biodemographic variables studied, elements of 5 were found to be significant at the p=(0.5 and are highlighted in the following discussion. Urban Residence. A larger percentage of cases (70.3%) than controls (56.5%) had resided in urban setting for more than seventy-five percent of their l i f e 93 Table 15 Summary of Results: Biodemographic Variables Variable Chi Square p-Value Principal area of residence 2.77412 0.4278 Urban residence 7.92509 0.0476* Level of education 4.15345 0.5275 Faith 5.26850 0.1532 Mother's country of origin 9.91380 0.0419* Father's country of origin 5.12873 0.2743 Marital status 1.38666 0.9258 Major occupation 7.36120 0.0612* Father's history of cancer 6.83916 0.5541 Mother's history of cancer 10.55084 0.3078 Brother's history of cancer 8.71562 0.3669 Sister's history of cancer 6.88652 0.6489 Heart disease 1.12485 0.2889 Intestinal polyps 14.77452 0.0001* Ulcerative c o l i t i s 0.70780 0.4002 Crohn's Disease - -Appendicitis , 0.07606 0.7827 Hemorrhoids 0.49885 0.4800 Varicose Veins 0.08433 0.7715 Di v e r t i c u l i t i s 0.00895 0.9246 Peptic Ulcer 2.42742 0.1192 Smoking History 2.30714 0.5112 Obesity History 7.70459 0.0525* Stress in the middle years 8.29013 0.0815 *Variable highlighted in discussion. 94 Table 16 Relative Risk'Estimates: Urban Residence Urban Residence Cases Controls Relative p-Value Risk 00-75% 33 40 1.818* 0.0600 76-100% 78_ 52 111 92 * *95% Confidence Intervals = 1.020, 3.241 95 span to date. The relative risk of a large proportion of urban living was 1.818 (p=0.06). This variable may be confounded by stress, occupation and/or food habits associated with urban l i v i n g . Exposure to potential carcinogens such as nitrous oxide and polycyclic aromatic hydrocarbons is greater in an urban setting. Mother's Country of Origin The country of origin of mothers of cases most frequently identified was Britain (41.4%); the percentage of Br i t i s h mothers for controls was 23.9%. This relationship was of interest as the colon cancer rates in Britain, particularly Scotland, are some of the highest in the world. The relative risk estimate of 2.556 was not significant (p=0.1170). The trend s t a t i s t i c was distorted by the inclusion of the "other" category which simply indicates a greater diversity in the countries of origin of the mothers of controls than of cases (Table 17). Major Occupation More controls (26.1%) were employed in the resource industries such as forestry and agriculture than cases (15.3%). Sales and c l e r i c a l jobs were held by a similar portion of the two groups. More cases (26.1%) than controls (13.0%) were employed in managerial or professional occupations (Table 18). The relative risk calculated for managerial versus resource industry occupation was 3.412 (p=0.0150). Managerial occupations are associated with higher socioeconomic status which has also been associated with colon cancer. This association may be confounded by overweight, stress, food habits, or urban li v i n g . 96 Table 17 R e l a t i v e R i s k Estimates: Mother's Country of O r i g i n Mother's Country Cases C o n t r o l s R e l a t i v e p-Value of O r i g i n R i s k Western Canada 9 11 Other Canada 22 12 2.241 0.2633 B r i t a i n , 4 6 22 2.566 0.1170 Other 27_ i l 0.943 0.8851 104 80 O v e r a l l Summary Chi Square = 9.626 D.F. = 3 p = 0.0220 97 Table 18 Relative Risk Estimates: Major Occupation Major Occupation Resource Industries Clerical Service Managerial/Professional Cases 17 21 . 26 29_ 93 Controls Relative Risk p-Value 24 20 23 12_ 79 1.482 1.596 3.412* 0.5091 0.3788 0.0150 *95% Confidence Intervals 1.378, 8.445 Global Chi Square = 7.36120 DF = 3 p = 0.0612, 98 Intestinal Polyps Consistent with the literature, a previous history of intestinal polyps was related to colon cancer as illustrated in Table 19. This precursor lesion enhanced risk more than fivefold for the development of colon cancer. Obesity History A further variable of interest was obesity history which was significant (p-0.0525). Participants were asked i f they had ever been more than twenty pounds overweight and, i f so, for what length of time. The data a were then cl a s s i f i e d as: never, short-term (less than 2 years), established (2-5 years) and long-term (more than 5 years). More controls (64%) than cases (45.9%) had never been overweight. More cases (42.3%) than controls (31.4%) had established or long-term obesity. The relative risk for ever being more than twenty pounds overweight versus for never being more than twenty pounds overweight was 2.096 which was significant at p=0.0162 (Table 20). Obesity may enhance risk via several routes: through associated mental stress, through physical stress on the gut of larger quantities of food, or through physiological stress on the body. Obesity has been shown to have a significant impact on immune response, particularly an impairment of cell-mediated immune response in vivo and a reduced intracellular bacteriacidal activity by polymorphonuclear leukocytes (Chandra, 1981). 99 Table 19 Relat i v e Risk Estimates: I n t e s t i n a l Polyps I n t e s t i n a l Polyps Cases Controls Relative p-Value Risk No 73 85 Yes . 30 _6 5.822* 0.0001 103 91 *95% Confidence I n t e r v a l s = 2.465, 13.748 100 Table 20 Relative Risk Estimates: Obesity History Obesity History Cases Controls Relative p-Value Risk Never Overweight 51 57 Ever Overweight 60 32 2.096* 0.0162 111 89 *95% Confidence Intervals = 1.186, 3.704. 101 Exposure to Selected Chemicals The most frequently cited contact with chemicals was with those used in agriculture (AG). However, the exposure was not a significant factor in colon cancer. No significant relationship was found for chemicals used in working with animal products (AN); paints and varnishes (PT); plastics and rubber (PLAS); dyes and bleaches (DYE); cutting or machine o i l s (OIL); petroleum products (PET); coal tar products (TAR); asbestos (ASB); dust from wood products, grain or textiles (DUST); fumes from organic solvents (SOLV); chemicals used in hospital or medical laboratory (HOSP); or other (OTH) as illustrated in Table 21. A significant association was identified for exposure to cooking o i l s and fumes from cooking (p=0.0439). This is most likely an artifact of the data as 18.0% of cases and only 6.5% of controls identified the use of cooking o i l in the home environment as an exposure. No association was noted for occupational exposures to cooking o i l s or fumes. The Chi square for trend was not significant (Table 22). In summary, the biodemographic variables which were of significance were urban residence for more than 75% of lifespan to date, managerial/professional occupations, a previous history of intestinal  polyps and a history of having ever been overweight. A l l may be related to dietary practices. These four variables are consistent with findings in the literature, confirming the general v a l i d i t y of this study. 102 Table 21 Summary of Results: Exposure to Hazardous Chemicals Chemicals Chi-Square d.f. p-Value Agricultural Chemicals 6.00486 3 0.1114 Chemicals Used in Working with Animal Products 0.62898 3 0.8898 Paints, and Varnishes 1.47301 3 0.6885 Plastics and Rubber 1.47301 2 0.6885 Dyes and Bleaches 0.52742 2 0.7682 Cutting or Machine O i l 3.15637 3 0.3681 Petroleum Products 3.86207 3 0.2767 Cooking Oils and Fumes , 8.10577 3 0.0439* Coal Tar Products 1.06812 2 0.5862 Asbestos 0.74841 3 0.8618 Dust from Wood Products, Grain/Textiles 1.04637 3 0.0790 Fumes from Organic Solvents 2.05128 3 0.5618 Chemicals Used in Hospitals or Medical Laboratories 1.86190 3 0.6016 Other 1 .07567 3 0.7830 *Variable highlighted in discussion. 103 Table 22 Relative Risk Estimates: Cooking O i l and Fumes Exposure Cases Controls Relative p-Value Risk None 83 78 Hobby, In the Home 20 6 3.133* 0.0282 Occupationally 8 _7 1.074 0.8905 111 91 *95% Confidence Intervals, = 1.236, 7.939 Chi Square for trend = 1.275 DF = 1 p = 0.2589 104 4.3.2. GENERAL FOOD HABITS Daily total meal consumption (DAYX3) was similar between cases and controls, with 36.9% of cases and 34.8% of controls indicating a regular consumption of three meals a day. However, 17.1% of cases regularly skipped breakfast, compared to only 4.4% of controls. The pattern of skipping meals in current food habits had a significant Chi-square at p = 0.0278 (Table 23). The relative risk for skipping breakfast was fourfold that for eating three meals a day. No risk was attached to the skipping of other meals (Table 24). The data of Breslow (1980) and others strongly suggest that the * regular eating of breakfast an important health habit to acquire. It would be tempting to formulate a hypothesis to suggest that a person is at risk when subjected to periods of abstinence of food, possibly due to a hormonal imbalance (e.g. insulin) which might be consistent with the overweight risk. However, the data on previous food habits indicate that this relationship was not consistent in middle (MLSK-M) and teen years (MLSK-T). It may simply reflect late rising due to illness in the case group. Subjects were asked to indicate i f they had ever taken or were currently taking large doses of vitamin C and/or E, (VITMINS). Large doses were defined as 1000 mg of vitamin C and 400 mg of vitamin E or greater. No relationship was found between the taking of large doses of vitamins C and/or E and colon cancer. A large proportion of both cases and controls had never taken large doses of either vitamin. 105 Participants were also asked i f they trimmed the fat from their meat before eating i t (TRFAT). No significant relationship was found in fat trimming practices in the current, middle or teen years between groups. The addition of sugar to tea, coffee and breakfast cereal (SUGAR) and the use of jams, j e l l i e s or honey (SWEETS) were not related to colon cancer. Participants were asked how they handled leftover vegetables. More cases than controls (44.1% vs. 34.4%), indicated they allowed their leftover vegetables to cool before refrigeration. Controls (41.1%) more frequently refrigerated their leftover immediately than did cases (24.3%). Immediate refrigeration had a relative risk ratio of 0.462, significant at p=0.0356 indicating that this practice had less risk attached to i t than allowing vegetables to cool before refrigeration (Table 25). Although the relative risk is increased, the biological significance of this practice is questionable. Most common vegetables such as beans, carrots, peas, and spinach contain levels of nitrate (Crosby, 1976). In some plants, such as beets and spinach, n i t r i t e levels have been shown to increase with improper storage prior to cooking. This has not been demonstrated after cooking. 106 Table 23 Summary of Results: Food Habits Food Habits Chi-Square p-Value Consuming Three Meals a Day 2.32439 Large Doses of Vitamins 4.50242 Trimming Fat 0.04505 Skipping Meals - Current 15.72098 Skipping Meals - Middle 5.98229 Skipping Meals - Teen 8.98804 Addition of Sugar 0.10239 Use of Jams and J e l l i e s 2.03348 Handling of Leftover Vegetables 8.75115 0.3128 0.3423 0.8319 0.0278* 0.4252 0.2535 0.7490 0.1539 0.0328* *Variable highlighted in discussion. 107 Table 24 Relative Risk Estimates: Meals Skipped Exposure Eat 3 x Day Skip Breakfast Skip Other Meals Totals Cases Controls 72 67 19 4 » .20 _19 111 90 Relative p-Value Risk 4.420* 0.0116 0.980 0.9014 *95% Confidence Intervals = 1.533, 12.749 108 Table 25 Relative Risk Estimates: Handling Leftover Vegetables Exposure Cases Controls Relative p-Value Risk Cool and then refrigerate 49 31 Refrigerate immediately 27. 3T_ 0.462* 0.0356 Totals 76 68 *95% Confidence Intervals = 0.237, 0.900. 109 4.3.3. CURRENT FOOD CONSUMPTION CURRENT FOOD FREQUENCIES Food frequency data was collected for nineteen foods or food groups: deep yellow vegetables (YV); broccoli, Brussel's sprouts (BR); cauliflower, cabbage (CA); leafy green vegetables (LG); fibrous vegetables such as corn, celery, string beans (VR); peanuts, peanut butter (PNT); dried beans, peas, le n t i l s (DRB); fish or poultry (FPL); beef (BEEF); cured meats e.g. ham, salami, bacon (CRD); cheese (CHES); eggs (EGGS); citrus fruits or juices (CIT); tomatoes including juice and/or sauce (TOM); pasta (PAST); ice cream and cream (ICECR); salad dressings and mayonaise (MAYO); whole grain breads m and cereals (WG); refined breads and cereals (RF); milk (MILK); coffee (COFFEE); tea (TEA); soda pop (POP); beer (BEER); wine (WINE); liquor (LIQ) and liquers (LQERS). The frequency of selected food preparation procedures was also t a l l i e d , s p e c i f i c a l l y for a number of fried foods (FRFD), smoked foods (SMK), charcoal broiled foods (CHAR) and pickled foods (PICK). As illustrated in Table 26 only two of the thirty-one current food frequency factors showed any significant relationship to colon cancer: consumption of fish and poultry and consumption of wine. Fish and Poultry The data indicated a twofold increase in cancer of the colon for moderate versus low consumption of fish and poultry but no such increased 110 Table 26 Summary of Results: Current Food Frequencies Food Chi Square p-Value Food Chi-Square p-Value Yellow Vegetables 7.30239 0.1991 Mayonnaise 4.18281 0.5234 Broccoli 1.44036 0.8372 Whole Grain 2.40160 0.7912 Cauliflower Cabbage 9.53911 0.0894 Refined Cereals 4.68478 0.4555 Leafy Greens 7.32315 0.1977 Fried Foods 7.43041 0.1095 Fibrous Vegetables 3.05241 0.6919 Charcoal Broiled Foods 1.32637 0.2495 Peanuts 6.61934 0.2505 Smoked Foods - -t Dried Beans 3.35588 0.3399 Pickled Foods - -t Fish, Poultry 9.03018 0.0109* Milk 4.14280 0.5290 Beef 4.00622 0.5485 Coffee 7.46956 0.1880 Cured Meats 2.09636 0.8357 Tea 3.53830 0.6176 Cheese 14.41805 0.6358 Pop 4.10035 0.5351 Eggs 1.36616 0.9280 Beer 6.10484 0.2969 Citrus Fruits 6.20554 0.2867 Wine 11.25355 0.0466* Tomatoes 7.32524 0.1976 Liquor 4.32910 0.5031 Ice Cream 1.74198 0.8836 Liquers 7.54765 0.1096 Pasta 4.04850 0.3995 *Variables Highlighted in discussion, tlnsufficient frequency of consumption for calculation. I l l r i s k for most days or daily consumption (Table 27). This is a puzzling finding as there has been only one indication in the literature that fish or poultry consumption may be risk enhancing. Graham et al (1978) did find a slightly elevated risk of colon cancer in males who were frequent eaters of fish. In this study, the trend s t a t i s t i c was not significant. Fish.and poultry are lower in saturated fats which are associated with increased risk. Fish may be a source of PAH. This finding may also reflect a concern colon cancer patients may have regarding beef consumption and a substitution of fish and poultry in their current diets. Wine The relative risk for consumption of wine more than once a week versus never consuming wine was more than twofold (p=0.0578). Although lesser consumption showed no significant risk, the trend s t a t i s t i c was significant at p=0.0546 indicating a greater risk for increasing consumption (Table 28). Although wine consumption has not been linked to colon cancer, ethanol has been linked to esophageal and rectal cancers (Tuyns, 1983); Kono and Ikeda, 1979). FOOD SCORES The food frequencies were categorized into eight major food groups to yield food group scores as follows: vitamin A rich vegetables (AVG), vitamin C rich fruits and vegetables (CVG), cruciferous vegetables (CRVG), fats (FAT), meat source protein (MEAT), vegetable source protein (VGPR), fiber rich foods (FIBR) and sources of beef fat (BFAT). 112 TABLE 27 Relative Risk Estimates: Fish and Poultry (Current) Exposure Cases Controls Relative p-Value Risk Less Than Once a Week Two-Three Times a Week Most Days/Daily Total AO 62 9 111 51 30 11 92 2.635* 0.0024 1.043** 0.8707 *95% Confidence Intervals = 1.451, 4.785 **95% Confidence Intervals = 0.392, 2.773 Chi-Square for trend 2.499 D.F. = 1 p = 0.1139 113 TABLE 28 Relative Risk Estimates: Wine (Current) Exposure Never Once a week or less More than once a week Total Cases 46 40 25 111 Controls Relative p-Value Risk 48 33 _11 92 1.265* 0.5528 2.372** 0.0578 *95% Confidence Intervals =1.057, 5.323 **95% Confidence Intervals = 0.806, 4.362 Chi-Square for trend 3.693 D.F. =1 p = 0.0546 114 As illustrated in Table 29, none of the Chi square values showed any significant differences between cases and controls for any of the eight food scored calculated. There was no trend identified within these data that would indicate that current beef, fat or fiber consumption could be linked to colon cancer. Indeed, the data indicates that current food consumption i s unrelated to the risk of colon cancer. 115 Table 29 Summary of Results: Current Food Scores. Food Scores Vitamin A Rich Foods Vitamin C Rich Foods Cruciferous Vegetables Fats Meat Source Protein Vegetable Source Protein Fiber Rich Foods Beef Fat Chi Square p-Value 1.16037 0.4425 0.68510 0.49460 1.13857 0.74778 1.57874 2.75411 0.5598 0.9781 0.7100 0.7809 0.5659 0.6881 0.4541 0.2523 116 4.3.4 PREVIOUS FOOD CONSUMPTION FOOD FREQUENCIES: MIDDLE YEARS Food frequency data was collected in a similar fashion for food intake ten to fifteen years prior to diagnosis of the disease identified as the "middle years". As illustrated in Table 30, two items of the thirty one food frequencies were identified as being significant at p=<0.05: fibrous vegetables and cheese. Fibrous Vegetables This food category referred specifically to the consumption of corn, celery and string beans, the foods rich in plant structural fiber or lignin. Consumption of fibrous vegetables in the middle years was found to reduce risk of colon cancer considerably (Table 31). Those who had consumed fibrous vegetables two to four times a month had a relative risk of only 0.258 (p=0.036) compared to those who regularly had consumed fibrous vegetables once a month or less. Greater consumption (more than once a week) showed a similarly reduced relative risk (R.R.=0.389) although this value was not significant (p=0.1871). For c l a r i t y , data were further summarized to show threefold relative risk (R.R.=3.208) for those who had consumed fibrous vegetables only once a month. These fibrous vegetables are rich in lignin, the least digestible structural component of plant fiber, which was been shown to adsorb large amounts of unconjugated bile acids in v i t r o . The protective effect of the lignin rich vegetables may be due to their a b i l i t y to enhance bi l e acid excretion. This requires confirmation through metabolic studies in man. Table 30 Summary of Results: Middle Years Food Frequencies Food Chi Square p-Value Food < Chi Square p-Value Yellow Vegetable 7.13477 0.2108 Mayonnaise 6.37912 0.2711 B r o c c o l i 2.42978 0.6573 Whole Grain 4.85055 0.4344 Caul iflower Cabbage 10.56622 0.0607 Cereal 2.33873 0.4344 Leafy Greens 9.08448 0.1057 F r i e d Foods 4.13343 0.3882 Fibrous Vegetables 17.64468 0.0034* Charcoal B r o i l e d Foods 0.60268 0.4376 Peanuts 4.02057 0.5465 Smoked Foods - - t Dried Beans 6.94430 0.2248 P i c k l e d Foods - - T F i s h , P o u l t r y 10.89282 0.0535 Milk 1.41706 0.09224 Beef 8.73601 0.1201 Coffee 4.59250 0.4676 Cured Meats 7.67913 0.1748 Tea 3.83296 0.5737 Cheese 14.04587 0.0153* Pop 6.73061 0.2415 Eggs 3.00346 0.6995 Beer 8.39918 0.1356 C i t r u s F r u i t s 7.85108 0.1646 Wine 2.15485 0.8273 Tomatoes 4.76559 0.4452 Liquor 1.82867 0.8273 Pasta 6.40193 0.0936 Ice Cream 4.91990 0.4257 Liquers 2.64312 0.6192 *Variables highlighted i n discussion, t l n s u f f i c i e n t frequency of consumption f or c a l c u l a t ion. 118 TABLE 31 Relative Risk Estimate: Fibrous Vegetables (Middle Years) Exposure Once a month or less 2 - 4 times a month More than once a week Total Cases Controls Relative p-Value Risk 14 47 _49 96 4 52 i i 88 0.258* 0.0356 0.389 0.1871 *95% Confidence Intervals =0.084, 0.796 119 The enhanced bi l e acid excretion is of particular interest when coupled with information about the in vivo formation of nitrosamines. In a rat experiment, Narisawa (1974) demonstrated that intrarectal doses of lithocholic acid or taurodeoxycholic acid (bile acids) increased the frequency of tumors induced by a single dose of N-methyl-N-^-nitrosoguanidine. The bile acids did not induce tumors when used alone, but rather acted as promoters of carcinogenesis. Cheese Consumption of cheese more than once a week had a relative risk of 1.737 (p=0.0834) compared to consumption of cheese once a week or less. Although the overall chi-square was significant. The Chi square for trend was not significant (Table 32). Cheese does contain a significant amount of tyramine which is a potent vasopressor. Average serving of cheese (50 gm) would contain about 25 mg of tyramine which is normally metabolized through the action of monamine oxidases. Inhibition of monamine oxidases results in increased hypertension. Vasopressors mediate water reabsorption (Lee, 1982) causing constipation; this may prolong carcinogen exposure or increase carcinogen concentration. Some cheeses contain nitrosamines (Scanlan, 1983). This variable requires closer investigation to c l a r i f y any dose-response relationship. A recent in vitro study (Rooma, 1983) did show that low fat milk products had a nitrite-lowering effect; they blocked the formation of dimethylnitrosamine. Milk products with a fat content of 10-20% had no 120 TABLE 32 Relative Risk Estimate: Cheese Consumption (Middle Years) Exposure Once a week or less More than once a week Total Cases Controls Relative p-Value Risk 35 76 107 40 5_0 86 1.737* 0.0834 *95% Confidence Intervals = 0.976, 3.092 Chi-square for trend = 2.300 DF=1 p=0.1294 121 effect or even increased the nitrosation. Cream with a fat content of 35% (in the range of many cheeses) stimulated the formation of dimethylnitrosamine twenty-fold compared to a control sample with no added milk products. FOOD SCORES: MIDDLE YEARS As with current consumption, food scores were calculated for the middle years. Three of the eight food scores were significant at p=<0.05. The same trichotomous c l a s s i f i c a t i o n (high, medium, low intake) was used in the middle and teen years. When these same food scores were compiled for the middle years, a strong negative relationship with colon cancer was shown for vitamin C rich foods, for cruciferous vegetables and for vegetables source protein scores (Table 33). Chi Square for these food scores were 20.45, 7.47 and 5.99, respectively. The c r i t i c a l value of Chi-square at p = 0.05 is 5.99 indicating a l l three items were significant. Using the same trichotomous clas s i f i c a t i o n as with current food scores, relative risk estimates were then calculated for the significant variables. The food score data for Vitamin C rich foods indicated that the relative risk of colon cancer for persons with a moderate intake of these foods was six times greater than that of persons with a high intake of r— Vitamin C. Persons with a low intake of vitamin C were at a twenty fold risk of colon cancer compared to those with a high intake of vitamin C rich foods. The trend s t a t i s t i c was also significant confirming this dose-response relationship (Table 34). 122 TABLE 33 Summary of Results: Middle Years Food Scores Food Scores Chi Square p-Value Vitamin A Rich Foods 1.49387 Vitamin C Rich Foods 20.44980 Cruciferous Vegetables 7.46929 Fats " 0.21629 Meat Source Protein 1.10704 Vegetable Source Protein 5.99453 Fiber Rich Foods 0.71249 Beef Fat 2.32356 0.4738 0.0000* 0.0239* 0.8975 0.5749 0.0499* 0.7003 0.3129 *Variable highlighted in discussion. 123 TABLE 34 Relative Risk Estimate: Vitamin C Rich Foods (Middle Years) Exposure Cases Controls Relative p-Value Risk HI 2 12 MED 71 69 6.174* 0.0206 LOW 38 11 20.727** 0.0001 *95% Confidence Intervals = 1.561, 24.425 **95% Confidence Intervals = 5.208, 82.493. Chi-square for trend = 18.871 DF=1 p=0.0000 124 The protective effect of vitamin C rich foods could be attributed to several factors. For example, the vitamin C i t s e l f could act as an inhibitor of carcinogen formation or i t could act to prevent the formation of nitrosamines, which may be cancer i n i t i a t o r s . The vitamin C may also have a positive effect on the immune system. These effects would have been supported by the variable on consumption of large doses of vitamin C. This variable was not significant (although the number of persons consuming large doses of vitamins were very limited). Another potential route of protective action of the vitamin C rich foods is through the supplying of citrus pectin which binds fecal bile acids reducing the formation of carcinogens (Kay and Truswell, 1972). Cruciferous Vegetables Analysis of the relative risk data for cruciferous vegetables (CRVG) yielded a Mantel-Haenzsel relative risk of 2.182 which was significant at p=0.0105. Thus, the risk of colon cancer for persons with a low intake of cruciferous vegetables was twice that for persons with a high intake of cruciferous vegetables (Table 35). It is interesting to note that the consumption of cauliflower and cabbage (CA) had a p value of 0.0894 in the current food frequency l i s t and a value of 0.0607 in the middle years food frequency l i s t suggesting that this variable borders on significant. 125 TABLE 35 ? Relative Risk Estimate: Cruciferous Vegetables (Middle Years) Exposure Cases Controls Relative p-Value Risk HI 74 44 LOW _37_ 48 2.182* 0.0105 111 92 *95% Confidence Intervals = 1.239, 3.842 126 The mechanism of action for this effect may relate to the indoles found in cruciferous vegetables which inhibit carcinogenesis by polycyclic hydrocarbons. Cruciferous vegetables may have a protective effect by virtue of their structural composition. These vegetables also contain lignin, a fecal bile acid excretion enhancer. These findings are consistent with those of Graham (1980), Haenszel, Locke and Segi (1980) and Modan, Barell and Lubin (1975), a l l of whom identified a decreased risk of colon cancer associated with an increased consumption of cruciferous vegetables, especially cabbage. Vegetable Source Proteins The vegetable source protein score represents intake of dried beans, peas, l e n t i l s , peanuts and whole grain cereals. Consumption in the middle years was elevated in the control group versus the cases. The relative risk of low consumption of vegetable source proteins versus high consumption was fourfold (Table 36). The protective effect of these foods may be due to their fiber components. FOOD FREQUENCIES: TEEN YEARS 1 The nineteen food frequency calculations were repeated for the teen years yielding only one significant variable: cheese (Table 37). None of the other consumption variables or food preparation indicators showed any 127 TABLE 36 Relative Risk Estimates: Vegetable Source Protein (Middle Years) Exposure LO MED HI Cases 3 34 J74 111 Controls Relative p-Value Risk 10 22 60_ 92 5.152* 0.0334 4 . 1 1 1 * * 0.0551 *95% Confidence Intervals **95% Confidence Intervals 1.376, 19.209 1.172, 14.418 Table 37 Summary of Results: Teen Years Food Frequencies Food Chi Square p-Value Food Chi Square p-Value Yellow Vegetables 4.45946 0.4853 Mayonnaise 2.90047 0.7153 Broccoli 1.17933 0.8815 Whole Grain 9.68992 0.0845 Cauliflower Cabbage 4.99083 0.4170 Refined Cereals 4.63524 0.3268 Leafy Greens 1.37827 0.9267 Fried Foods 2.99891 0.5580 Fibrous Vegetables 4.56561 0.4712 Charcoal Broiled Foods — - t Peanuts 4.01745 0.5469 Smoked Foods - - t Dried Beans 5.46087 0.3623 Pickled Foods 2.62500 0.2691 Fish, Poultry 8.81402 0.1167 Milk 4.89536 0.4288 Beef 6.64634 0.2483 Coffee 4.64956 0.4601 Cured Meats 2.72742 0.7419 Tea 3.92864 0.5597 Cheese 12.75682 0.0258* Pop 5.43431 0.3652 Eggs 4.85233 0.4342 Beer 7.70339 0.1734 Citrus Fruits 4.05438 1 0.5416 Wine 2.22222 0.5276 Tomatoes 4.87177 0.4317 Liquor 3.75000 0.5276 Ice Cream 8.10811 0.1504 Liquers - -Pasta 3.07158 0.3807 *Variables highlighted in discussion, tlnsufficient frequency of consumption for calculation. 129 significant relation to colon cancer. Cheese The cheese consumption frequency data showed an increase risk for consumption of cheese versus never consuming cheese, however, the dose-response was reversed i.^e. greater consumption showed lesser risk. This would suggest cheese may be unreliable risk indicator or that the effect of cheese may be different at different ages. The trend s t a t i s t i c was not significant (Table 38). FOOD SCORES: TEEN YEARS Data for food scores in the teen years (Table 39) indicated a Chi-square value of 6.75 (p=0.0342) for vitamin C rich foods. None of the other seven food scores showed a significant relationship to colon cancer. Vitamin C Rich Foods As illustrated in Table 40, the relative risk of colon cancer is greater for those with moderate consumption than for high consumption of vitamin C; the risk is fivefold (p=0.0486). The relative risk for low versus high consumption is not significant, nor is the trend s t a t i s t i c . In conclusion, food consumption in the teen years does not yield conclusive indicators of risk associated with colon cancer although the information on vitamin C rich foods supports a protective role identified in the middle years food consumption patterns. 130 TABLE 38 Relative Risk Estimates: Cheese (Teen Years) Exposure Cases Controls Relative p-Value Risk Never consumed 18 27 A few times a month 27 11 3.682* 0.0095 One - Three times a week 56 41 2.049** 0.0749 Most days/daily _10_ 12_ 1.250 0.8727 Total 111 92 f *95% Confidence Intervals = 1.484, 9.133 **95% Confidence Intervals = 1.000, 4.196 Chi-Square for trend 0.544 D.F.= 1 p = 0.4606 131 Table 39 Summary of Results: Teen Years Food Scores. Food Score Chi-Square p-Value Vitamin A Rich Foods 0.25658 0.8796 Vitamin C rich Foods 6.75184 . 0.0342* Cruciferous Vegetables 2.86471 0.2387 Fats 0.85749 0.6513 Meat Source Protein 1.35541 0.5078 * Vegetable Source Protein 1.81221 0.4041 Fiber Rich Foods 1.08704 0.5807 Beef Fat 3.52949 0.1712 •Variable highlighted in the discussion. 132 TABLE 40 Relative Risk Estimates: Vitamin C Rich Foods (Teen Years) Exposure Cases Controls Relative p-Value Risk HI 2 7 MED 61 37 5.770* 0.0486 LO . _48 47 3.574** 0.2044 Total 111 92 *95% Confidence Intervals Chi-Square for trend 0.184 = 1.316, 25.298 D.F.= 1 p = 0.6678 133 4.3.5 VARIABLE INTERACTIONS AGE-SEX CLASSIFICATION The principal variables found to be significant were analyzed by sex for two age categories age 65 years or greater and age less than 65 years. This analysis gave further c l a s s i f i c a t i o n regarding the relationship of these variables to colon cancer. The results are summarized in Table 41. Of the fourteen variables assessed, the relationships of two of the biodemographical variables and three of the food consumption variables were influenced by age and sex. The effects of the remainder of the variables were not found to be related to age and sex. Urban Residence The percentage of urban versus rural residence was significant for men but not for women. In older men, this variable was significant at p=0.0438. This would suggest that exposure to urban residence was a greater risk for men of retirement age or greater. Polyps The variable, polyps, also was significantly related to colon cancer for males but not for females at p=<0.05. This would suggest that the familiar tendency for polyps was more of a risk factor for men than women. This has not been confirmed in the literature. TABLE 41 Summary of Results: Age-Sex Categorization FACTOR. FEMALES MALES > 65 years < 65 years y 65 years < 65 years Chi Square P-Value Chi Square P-Value Chi Square P-Value Chi Square P-Value Mother's Country of Origin 1.42083 0.4914 - — — _ 3.00000 0.2231 Urban Residence 1.77045 0.6214 1.04352 0.5935 8.10716 0.0438* 6.44526 0.0918 Major Occupation 6.08679 0.1075 4.59475 0.2040 6.11765 0.1060 2.78539 0.4259 Polyps 1.37592 0.2408 1.58242 0.2804 5.44366 0.0196* 5.24138 0.0221* Obesity History 0.11176 0.9904 7.55655 0.0561 6.05491 0.1090 1.90000 0.3867 Middle Years: Vitamin C Rich Foods 0.38378 0.8254 3.78404 0.1508 6.32124 0.0424* 7.12364 0.0284* Cruciferous Vegetables 5.98020 0.0503 0.04730 0.9766 0.55954 0.7560 0.84860 0.6542 Vegetable Source Protein 1.70794 0.4257 1.10774 0.5747 1.23992 0.5380 0.93675 0.6260 Fibrous Vegetables 11.71019 0.0390* 4.50100 0.4798 4.39414 0.2941 7 .51179 0.1853 Cheese 5.76726 0.3295 1.63836 0.8019 6.93286 0.2257 7.42124 0.1911 Teen Years: Vitamin C Rich Foods 0.66469 0.7244 0.56296 0.7547 0.57738 0.7492 1.43684 0.4875 Cheese 6.82780 0.2338 6.66389 0.1589 5.70599 0.3359 4.29176 0.5082 Current Consumption: Fish and Poultry 12.37962 0.0299* 4.67262 0.4571 0.86206 0.9299 15.38447 0.0040* Wine 6.87500 0.2301 0.97934 0.9129 5.52886 0.2372 5.18636 0.2687 *Variable highlighted in discussion. 135 Vitamin C Rich Foods The low consumption of vitamin C rich foods in the middle years was found to be significant for males in both age groups. This relationship was not found to be significant for females in either age group. The protective effect of vitamin C and the risk of urban liv i n g and polyps may be related. Fibrous Vegetables The protective effect of the consumption of fibrous vegetables in the middle years was significant (p=0.0390) for females in the older age group. A similar protective effect was'not found for males. Fish and Poultry No consistent age-sex trend was determined for the current consumption of fish and poultry. This factor was significant for older women and for younger males. This analysis of the data by age and sex categories would suggest a need for further investigation. Although the incidence of colon cancer is similar for males and for females, the mechanism of action of carcinogenisis may d i f f e r at different ages and vary between the sexes. 136 CHAPTER 5 SUMMARY AND IMPLICATIONS 5.1 SUMMARY OF THE STUDY This case-control study was undertaken to c l a r i f y the role of dietary factors in the etiology of colonic cancer. Data was collected to answer the question of which of the selected dietary factors were risk enhancing and which were protective, related to this disease. Further, these dietary factors were examined over time to determine i f the effects of these factors were consistent over three time frames: current, middle and teen years. Some biodemographic variables were examined to determine i f any were related to the disease. Procedure The case-control format was a useful procedure for this investigation. Using trained dietary interviewers, data was collected on a total of 212 participants, yielding a total of 203 usable interviews. As anticipated, a large proportion (49.3%) of the identified case population was ineligible for interview. The principal reasons were death before contact and permission refused by physician. This may be considered to limit the generalizabilty of the results, as the fina l sample of cases may represent longer survivors or a subset of cooperative individuals. However, the fi n a l case sample was comparable to the total case sample in terms of cancer sites. Patterns of refusal were similar for cases and for 137 controls. The fi n a l sample for analysis consisted of 111 cases and 92 controls of similar age and sex. Controls were either friends of cases or referrals from case physicians, the latter recruiting method being more successful and would have been a more efficient choice at onset. A larger control group may have been useful to reinforce findings. Instruments The interview schedule had three main components: food frequency l i s t s , questions on food habits, and biodemographical information including questions on exposures to various chemicals. The food frequency l i s t s and food habits data were repeated for three time frames: current food habits and food habits in the middle and teen years. Shortening the schedule to concentrate on the food frequency l i s t s would have made the interview schedule less time consuming. Data on nineteen food frequency groupings, eight food scores, information on seven principal food habits and on fifteen biodemographical factors plus exposure to fourteen groups of chemicals was analyzed. Results The results were organized in response to four questions and four sets of hypotheses. 138 A. In response to the f i r s t question as to whether selected biodemographical variables were related to colon cancer, the following results were obtained: 1. A high proportion of urban residence was a risk factor for colon cancer especially for males under 65 years. 2. The country of origin of the mother of cases was found to be related colon cancer, but not significantly. 3. A trend was shown linking obesity with increased risk for colon cancer. 4. Birth place, major residence, education, faith, marital status, occupation, and smoking were not related to colon cancer. 5. Parental and sibling history of cancer was not related to colon cancer. 6. Adenomatous polyps were strongly risk enhancing for colon cancer especially for males. 7. Other disease entities (cardiovascular disease, ulcerative c o l i t i s , appendicitis, hemorrhoids, varicose veins, d i v e r t i c u l i t i s , peptic ulcer) were unrelated to colon cancer. B. In response to the second question, regarding the relationship of general food habits to colon cancer, the following results were obtained 1. A current food habit of skipping breakfast was found more often in cases than in controls. 2. More cases than controls allowed leftover vegetables to cool before refrigeration. 3. No relationship was found for any of the other food preparation or consumption habits. «i> 139 C. In response to the third question regarding which of the current dietary factors were risk enhancing and which were protective, the following results were obtained. 1. Only two of the current food frequency items, consumption of fish and poultry and consumption of wine, were related to colon cancer. No significant trend for fish and poultry consumption could be identified. The relative risk value associated with wine consumption was significant. This current practice may be an adjustment to the disease. 2. None of the compiled food scores (AVEG, CVEG, CRVEG, FAT, MEAT, VGPR, FIBR, or BFAT) showed any significant relationship between current food scores and colon cancer. D. In response to the question regarding whether these risk relationships were similar over the middle years and teen years, the following results were obtained: 1. Increasing consumption of fibrous vegetables showed a strong protective effect for colon cancer in the middle years. 2. Cheeese consumption was identified as a risk enhancing variable in the middle years. A consumption trend could not be confirmed in the teen years. 3. Food scores calculated for the middle years showed a very strong protective effect for vitamin C rich foods. This protective effect was also seen in the teen years. 4. Protective effects were also noted for consumption of cruciferous vegetables and vegetable source proteins in the middle years. 5. None of the other frequency variables, food scores or food preparation 140 indicators showed any significant relationship to colon cancer. Thus i t may be concluded from these data that food consumption in the middle years is a better predictor of risk/protective factors related to colon cancer than is current food consumption information. In summary, this study constructs a picture of a susceptible host who has been overweight ("frisk x 2), has resided in an urban setting more than seventy-five per cent of lifespan (trisk x 1.8) and is most l i k e l y employed in a managerial or professional occupation (trisk x 3.4) The ageing host has most lik e l y developed intestinal polyps (frisk x 5). Bile acid concentration is enhanced by overweight. (Bile acids have been shown to be promoters of nitrosamine induced cancer in animals). Colon epithelial tissue proliferation is promoted by b i l e acids thus the host e p i t h e l i a l tissue is regenerating rapidly and is vulnerable to carcinogens. Bacterial action on fecal b i l e acid yields mutagens. The host's intake of fibrous vegetables is low ('trisk x 3.2); fibrous vegetables increase fecal bile acid excretion. Cheese intake is higher (trisk x 1.7) which may increase transit time and increase contact with fecal carcinogens. The intake of vitamin C rich foods of the host is low (trisk 6-20x) which may diminish protection of the host through: (a) the lack of antioxidant effect (b) diminished efficiency of the immune system (c) failure to block the formation of food or environmental source nitrosamines or (d) lack of removal of b i l e acids by citrus pectin. Intake of vegetable source protein is low (trisk 4x). The 141 susceptible host may have been exposed to cooking o i l s or fumes whi increased exposure to polycyclic aromatic hydrocarbons derived from g r i l l i n g of beef products. Intake of protective indole containing cruciferous vegetables is low ("frisk 2.1 x). 142 5.2 LIMITATIONS There are several limitations relating to the generalizability and interpretation of the results of this study. Three principal limitations are non-response, sample size, and the large number of univariate comparisons. Almost f i f t y per cent of the total potential sample of cases identified by the Provincial Cancer Registry were ineligible for interview. Twenty per cent were deceased and twelve per cent too i l l to interview. Although this is similar to other studies of this nature, there is a concern as to whether the sample is sufficiently representative of the case population such that the results can be generalized to the population at large. As the distribution of cancer sites were similar for the total case population and the interviewed case sample, the interviewed case sample was considered to be representative of the case population. A second major consideration i s the total sample size. As discussed, depending on the proportion of distribution of a given exposure in the population, the size of the sample limits the magnitude of the relative risk which is identifiable as s t a t i s t i c a l l y significant. With a small sample size, some parameters which are actually significant to the etiology of the disease may appear to be non significant in the data. Parameters of borderline significance which merit further investigation are ulcerative c o l i t i s , peptic ulcer, stress in the middle years, consumption of cauliflower and cabbage, and the consumption of pasta and of whole grain. Many univariate comparisons were made in this study. S t a t i s t i c a l l y , 143 a limited number of variables may be found to be significant simply due to chance. Care was taken to examine trends over time, dose response, and related food factors to confirm significance. The cases were a slightly older population than the total sample which may indicate a less severe exposure to disease causing or promoting factors. These factors may have been missed being significant in this study. However, care was taken to include only recent (within six months or diagnosis) incident cases in the study which should have minimized this effect. The homogeneity of the diet may also have masked significant risk factors in this study. The selection of friends as controls may have caused significant variables to be missed as diets of friends and cases might have been similar. A further limitation may be related to the v a r i a b i l i t y of friends as controls. As many cases were unable to identify friends to serve as controls, an alternative source of controls was pursued. Controls were identified from the records of physicians of cases. A l l physician identified controls were healthy patients with no known or suspected cancer of any kind. This client population may represent a particular subpopulation who have some unknown characteristic which makes them willing to participate in studies but then, too, so do the cases. Analysis indicated there were no significant differences for the variables assessed between the friends-source and physician-source controls. Thus, the controls were identified from essentially the same frame as the cases. An additional limitation relates to the long-term retrospective recall required to complete the interview schedule. Data was collected in 144 such a manner as to rely on memory of events which occurred some time previously. The schedule was designed and the interviewers trained to e l i c i t general patterns of information from set periods of time which assists memory. The Byers (1983) study was able to demonstrate agreement between originally eaten and recalled food items after a twenty to twenty-five year time span. Stronger correlations were obtained for food scored or indexes as were used in this study. The collection and analysis of data based only on frequency of intake made i t impossible to quantify nutrient consumption. Quantification would have been desirable to enhance the assessment of actual food intakes. As the assessment of usual food intake was the purpose of this study and the time frame was lengthy, quantification of nutrient intakes may not have enhanced the results. Cases may have been more susceptible to gastrointestinal complaints affecting current food habits, however, the data on disease history indicated no significant differences in disease patterns between cases and controls with the exception of the precursor disease adenomatous polyps. Finally, there is the danger of a more obscure limitation, the danger of "misclassification". Exposure categories could be inappropriate masking a significant relative risk (Gladen, 1979) although generally several categorizations were assessed. Further though, the controls may be misclassified. Control subjects were carefully checked to determine presence of cancer, however the controls have yet to complete lifespan. They may yet develop cancer of the colon. 145 5.3 IMPLICATIONS Re c a l l i n g the epidemiological model of host, agent and environment, t h i s study does provide some c l a r i f i c a t i o n of the p i c t u r e of a susceptible host exposed to harmful agents i n both a p r o t e c t i v e and r i s k enhancing environment. Colon cancer a f f e c t s an aging host. The host i s p a r t i c u l a r l y s u s c e p t i b l e to the disease i f adenomatous polyps are present. Urban residents are at higher r i s k of developing the disease. Obesity may contribute to host s u s c e p t i b i l i t y or r e f l e c t d i e t a r y experiences (exposure to carcinogens). No c l e a r i n d i c a t i o n of causative agent was apparent from t h i s study. One p o s s i b l e explanation may r e l a t e to the fact that the aging tissues i n the gut are u n i v e r s a l l y s usceptible to ubiquitous carcinogens and key pr a c t i c e s e i t h e r enhance the quantity of carcinogens present or protect the host from the carcinogenic agents. For example, the consumption of fibrous vegetables may remove a substratum f o r carcinogens from the gut, decreasing the carcinogenic exposure of host t i s s u e . The study suggests a pro t e c t i v e r o l e f o r some f i b e r components of the d i e t against the carcinogenesis-promoting e f f e c t s of b i l e acids. The study d i d demonstrate that c e r t a i n d i e t a r y p r a c t i c e s were p r o t e c t i v e to the host. S p e c i f i c a l l y , an increased consumption of vitamin C r i c h foods reduced colon cancer r i s k at least s i x f o l d . This p r a c t i c e i s consistent with theory that the antioxidant function of vitamin C could reduce a v a i l a b l e free r a d i c a l s i n ageing t i s s u e which may act as 146 carcinogens. Further, antioxidants such as vitamin C interrupt the chain of conversion from nitrates to nitrosamines and play a role in the immune system. Citrus pectin may assist in the removal of bile acids thus diminishing contact with the carcinogen promoters. Consumption of cruciferous vegetables was also a protective practice. The mechanism of protection may be the presence of indoles which protect against polycyclic aromatic hydrocarbons (PAH). This study provided no evidence to support a theory involving a dietary source of PAH, however, study controls did consume more vegetables source proteins then did cases. Thus intakes of vitamin C rich foods, cruciferous vegetables, vegetable source protein and fibrous vegetables should be closely monitored in persons with adenomatous polyps. Of considerable interest in this study is the use of previous time frames for the collection of retrospective dietary data. Data from this study showed that food patterns in current years and in the teen years were not as indicative of risk for colon cancer as were patterns ten years prior to the diagnosis of a tumor. This fact, linked with the data on serum cholesterol values which identified that this indicator changes markedly at least five years prior to onset of symptoms, suggest that further investigation of food patterns in the middle years may be useful in the determination and confirmation of the roles of dietary factors in the etiology of this disease. 147 5.4 FUTURE RESEARCH Several f a c t o r s which were i d e n t i f i e d as being r e l a t e d and which had not appeared i n previous l i t e r a t u r e , r equire further explanation. In p a r t i c u l a r , the consumption of cheese, exposure to cooking o i l s , and the p r a c t i v e of not immediately r e f r i g e r a t i n g l e f t o v e r s vegetables are i n question. Some risk-enhancing factor may be present i n cheese which coupled with a lower f i b e r intake, enhances r i s k and may a s s i s t i n explaining d i f f e r e n c e s i n colon cancer rates between Denmark and Finland. Research should be d i r e c t e d toward food habits p r i o r to the onset of disease symptoms to avoid confounding due to d i e t a r y changes r e s u l t i n g from the disease and/or pr a c t i c e s which are not relevant to the i n i t i a l phase of the disease. Using a simple d i e t a r y t o o l which assessed food scores, a prospective study could be undertaken on persons with adenamatous polyps to c l a r i f y which d i e t a r y f a c t o r s protect against or enhance r i s k of colon cancer. This study supports the need for studies such as the c l i n i c a l t r i a l of Bruce (1982) in Toronto where persons with polyps are being treated with ascorbic acid and vitamin E. L i g n i n s and c i t r u s p e c t i n should also be assessed i n such t r i a l s . Further, case c o n t r o l i n v e s t i g a t i o n s of previous food p r a c t i c e s of persons with polyps may e l u c i d a t e d i e t a r y p r a c t i c e s which are r e l a t e d to development of t h i s precursor disease. 148 6. BIBLIOGRAPHY Acheson, K.J., I.T. Campbell, O.G. Edholm, D.S. Miller, and M.J. Stock, 1980. The Measurement of Food Energy Intake in Man. American Journal of C l i n i c a l Nutrition 33:1147-1154. Adelson, S.F., 1960. Some Problems i n Collecting Dietary Data from Individuals. Journal American Dietetic Association 26:453-461. Alcantara, E.N., and E.W. Speckmann, 1976. Diet, Nutrition and Cancer. American Journal of C l i n i c a l Nutrition 29: 1035-1047. Armstrong, B., and R. Doll, 1975. Environmental Factors and Cancer Incidence and Mortality in Different Countries for Special Reference to Dietary Practices. International Journal of Cancer 15: 617-631. Balogh, M., J.H. Medalie, H. Smith, and J.J. Green, 1968. The Development of a Dietary Questionnaire for an Ischemic Heart Disease Study. Israeli Journal of Medical Science 140(2):195-203. Baserga, R., 1980. Biochemical, Molecular, Pharmacologic Approaches to Large Bowel Cancer. Cancer 45: 1168-1171. Bazzarre, T.L., and M.P. Meyers, 1979. The Collection of Food Intake Data in Cancer Epidemiology Studies. Nutrition and Cancer l(4):22-45. Bazzarre, T.L. and J.A. Yuhas, 1983. Comparative Evaluation of Methods for Collecting Food Intake Data for Cancer Epidemiology Studies. Nutrition and Cancer 5:201-214. Becker, F.F. ed., 1975. Cancer: A Comprehensive Treatise. Part 1 Etiology: Chemical and Physical Carcinogensis. Plenum Press, New York. Bennet, A.E., and K. Ritchie, 1975. Questionnaires in Medicine: A Guide to Their Design and Use. Oxford University Press, London. Berg, J.W., 1975. Diet in Fraumeni, J.F.jr. Persons at High Risk of Cancer. Academic Press, New York. Berg, J.W., and M.A. Howell, 1974. The Geographic Pathology of Bowel Cancer. Cancer 34(3): 808-814. Berg, J.W., M.A. Howell, and S.J. Silverman, 1973. Dietary Hypotheses and Diet-Related Research in the Etiology of Colon Cancer. Health Services Reports 88 (20): 915-924. Berlinger, N.T., and R.A. Good, 1980. Suppressor Cells i n Healthy Relations of Patients with Heriditary Colon Cancer. Cancer 45:1112-1116. 149 Bingham, S., J.H. Cummings, and N.I. McNeil, 1979. Intakes and Sources of Dietary Fibre in Br i t i s h Population. American Journal of C l i n i c a l Nutrition 32:1313-1319. Birkner, H.J., and F. Keen, 1974. In Vitro Adsorption of Bile Salts to Food Residues, Salicylazosulfapyridine and Hemicellulose. Gastroentrology 67:237-241. Bjelke, E., 1975. Dietary Vitamin A and Human Lung Cancer. International Journal of Cancer 15:561-565. Breslow, L., and J.E. Endstrom, 1980. Persistence of Health Habits and Their Relationship to Mortality. Preventive Medicine 9:469-483. Bresnick, E., 1980. Colon Carcinogensis: An Overview. Cancer 45:1047-1051. Broitman, S.A., 1981. Cholesterol Excretion and Colon Cancer 41;3738-3740. Bruce, W.R., 1982. Relation of Nutrition and Colon Cancer Causation as Assessed by Fecal Mutagen and Colonic Polyp Intervention Studies. Abstract. Proceedings of the 13th International Cancer Congress. September 1982. UICC. Bruce, W.R., G.M. Eyssen, A. Ciarapi, P.W. Dion, and N. Boyd, 1981. Strategies for Dietary Intervention Studies in Colon Cancer. Cancer 47:1121-1125 . Burdette, W.J., 1971. Identification of Antecedents of Colorectal Cancer. Cancer 28(a):51-70. Burdette, W.J., 1974. Colorectal Carcinogenesis. Cancer 34(3):872-877. Bureau of Epidemiology, Health Protection Branch, 1977. Cancer Patterns in Canada, 1931-74. Health and Welfare Canada, Ottawa. Burke, B.S., 1947. The Dietary History as a Tool in Research. Journal American Dietetic Association, 23:1041-1046. Burkitt, D.P., 1969. Related Diseases-Related Cause; Lancet 2:1229-1231. Burkitt, D.P., 1971a. Epidemiology of Cancer of the Colon and Rectum. Cancer 28(1):3-13. Burkitt, D.P., 1971b. Epidemiology of Dietary Factors in Large Bowel Cancer. Cancer 28:3-8. Byers, T.E., R.I. Rosenthal, J.R. Marshall, T.F. Rzepka, K.M. Cummings, and S. Graham, 1983. Dietary History From the Distant Past: A Methodological Study. Nutrition and Cancer 5(2):69-77. 150 Cady, B., A.V.Persson, D.O. Monson, and D.L. Maunz, 1974. Changing Patterns of Colorectal Carcinoma. Cancer 33(2) -.422-426. Challis, B.C., and CD. Bartlett, 1975. Possible Cocarcinogenic Effects of Coffee Constituents. Nature. 254:532-533. Chandra, R.K., 1981. Immune Response in Overnutrition. Cancer Research 41: 3795-3796. Cleave, T.L., 1956 in Burkitt, D.P., 1971. Some Neglected Leads to Cancer Cousative. Journal National of Cancer Institute 47(5):913-919. Cole, P., 1978. The Ascedence of the Case-Control Study. Unpublished Monograph. Commoner, B., A.J. Vithayathil, P. Dolara, S. Nair, P. Madyastha, and G.C. Cuca, 1978. Formation of Mutagens in Beef and Beef Extract During Cooking. Science 201(8):913-916. Cornfield, J ., and W. Haenszel, 1960. Some Aspects of Retrospective Studies. Journal of Chronic of Diseases 11:523-534. Correa, P., 1975. Comments on the Epidemiology of Large Bowel Cancer. Cancer Research 35:3395-3397. Correa, P., 1981. Epidemiological Corelations between Diet and Cancer Frequency. Cancer Research 41:3685-3690. Cowdry, E.V., 1968. Etiology and Prevention of Cancer in Man. Appleton, Century, Crofts, New York. Crosby, N.T. 1976. Nitrosamines in Foodstuffs. Residue Reviews 64:77-135. Crowthy, J.S. et al 1976. Fecal Steroids and Bacteria and Large Bowel Cancer in Hong Kong by Socio-economic Group. British Journal of Cancer 34:191-198. Cummings, J.H., 1978. Dietary Factors in the Etiology of Gastrointestional Cancer. Journal of Human Nutrition 32:455-465. Dales, L.G., G.D. Friedman, H.K. Ury, S. Crossman, and S.R. Williams, 1978. A Case-Control Study of Relationships of Diet and Other Traits to Colorectal Cancer in American Blacks. American Journal of Epidemiology 109(2):132-144. Davies, L., and M.D. Holdsworth, 1979. A Technique for Asssesing Nutritional 'At Risk' Factors in Residential Homes for the Elderly. Journal of Human Nutrition 33:165-169. Day, N.E., D.P. Byar, and S.B. Green, 1980. Overadjustment in Case-Control Studies. American Journal of Epidemiology 112(5):696-706. 151 Derby, C , and P. Drouin, 1976. Diet and Functional Disorders of the Colon. Progress in Food, Nutrition and Science 2:1-10. Deschner, E.E., 1974. Experimentally Induced Cancer of the Colon. Cancer 34:824-828. Dipple, A. 1983. Formation, Metabolism and Mechanism of Action of Polycyclic Aromatic Hydrocarbons. Cancer Research 43 (Supple):24225-24255. Doll, R., 1979. Nutrition and Cancer: A Review. Nutrition and Cancer l(3):35-45. Dorn, H.F., 1959. Some Problems Arising in Prospective and Retrospective Studies of the Etiology of Disease. New England Journal of Medicine 261:571-579. Drasar, B.S., and D. Irving, 1973. Environmental Factors and Cancer of the Colon and Breast. British Journal of Cancer 27:167-172. Durrington, P.N., A.P. Manning, CH. Bolton and M. Hartog, 1976. Effect of Pectin on Serum Lipids and Lipoproteins, Whole Gut Transit Time, and Stool Weight. Lancet. 2:394-398. Enig, M.G., R.J. Mann, and M. Keeney, 1978. Dietary Fat and Cancer Trends - A Critique. Federation Proceedings 37(9):2215-2220. Enstrom, J.E., 1981. Reassessment of the Role of Dietary Fat i n Cancer Etiology. Cancer Research 41: 3722-3723. Falterman, K.W., C.5. H i l l , J.C. Markey, J.W. Fox, and I. Cohn, 1974. Cancer of the Colon, Rectum and Anus: A Review of 2313 Cases. Cancer 34: 951-959. Farber, E., 1984. Chemical Carcinogenesis: A Current Biological Perspective. Carcinogenesis 5 ( l ) : l - 5 . Feinstein, A.R. and R.I. Horowitz, 1982. Double Standards, Scientific Methods and Epidemiologic Research. New England Journal of Medicine 307(26):1611-1617. Fenoglis, CM. and N. Lane, 1974. The Anatomical Precursor of Colorectal Carcinoma. Cancer 34(3):819-823. Fieldhouse, P., 1979. An Interview Model for Use in Dietetic Training. Journal of Human Nutrition 33:206-210. Fine, D.H., D. Ross, D.L. Roundbehler, A. Silvergleid, and L. Song, 1977. Formation In Vivo of Volatile N-Nitrosamines in Man After Ingestion of Cooked Bacon and Spinach. Nature 265:753-754. Finegold, S.M., D.J. Flora., H.R. Athenbery, and V.L. Sutter, 1975. Fecal Bacteriology of Colonic Polyp Patients and Control Patients. Cancer Research 35:3407-3417. 152 Field-house, P., 1979. An Interview Model for Use in Dietetic Training. Journal of Human Nutrition 33:206-210. Fine, D.H., D. Ross, D.L. Roundbehler, A. Silvergleid, and L. Song, 1977. Formation In Vivo of Volatile N-Nitrosamines in Man After Ingestion of Cooked Bacon and Spinach. Nature 265:753-754. Finegold, S.M., D.J. Flora., H.R. Athenbery, and V.L. Sutter, 1975. Fecal Bacteriology of Colonic Polyp Patients and Control Patients. Cancer Research 35:3407-3417. Finegold, S.M., D.L. Sutter, P.T. Sigihara, H.A. Elden, S.M. Lehmann, and R.L. Ph i l l i p s , 1977. Fecal Microbial Flora in Seventh Day Adventist Popultions and Control Subjects. American Journal of C l i n i c a l Nutrition 30 :1781-1792. Fleiss, Joseph L., 1973. St a t i s t i c a l Methods for Rates and Proportions. John Wiley and Sons, Toronto. Fleiss, Joseph L., 1974. Inference About Population Attributable Risks From Cross-Sectional Studies. American Journal of Epidemiology 110(2):103-104. Gail, M., R. Williams, D.F. Byar, and C. Brown, 1976. How Many Controls? Journal of Chronic Diseases 29:723-731. Gallagher, R.P., and J.M. Elwood, Undated. Cancer Mortality Among Chinese, Japanese, and Indians in B r i t i s h Columbia, 1964-1973. British Columbia Health Surveillance Registry. Vancouver. B.C. Prepublication Draft. Gladen, B., and W.J. Rogan, 1979. Misclassification and the Design of Environmental Studies. American Journal of Epidemiology 109(5):607-616. Gladney, V.M., 1972. Food Practices of Some Black Americans in Los Angeles County: Including a Method For Evaluating the Diet. Los Angeles County Health Services. Gori, G.B., 1977. Diet and Cancer. Journal American of Dietetics Association 71:375-379. Gori, G.B., 1978. Role of Diet and Nutrition in Cancer Cause, Prevention and Treatment. Bulletin of Cancer 65(2):115-126. Gori, G.B., 1978. Diet and Nutrition in Cancer Causation. Nutrition and Cancer 1(1) :5 -7. Grace, T.B., 1980. Gastrointestinal Tract Cancer. Chapter 10 in Cl i n i c a l Oncology. Academic Press, New York. 153 Grage, T.B., R.M. Ferguson, and R.L. Simmons, 1977. Gastrointestinal Tract Cancer, Chapter 10 in C l i n i c a l Oncology. W.B. Saunders Company, Toronto. Graham, S., 1980. Diet and Cancer. American Journal of Epidemiology 112(2):247-252. Graham, S., 1983. Results of Case-Control Studies of Diet and Cancer in Buffalo, N.Y. Cancer Research 43 (Suppl):2409S-2413S. Graham, S., H. Dayal, M. Swanson, A. Mittleman, and G. Wilkinson, 1978. Diet and the Epidemiology of Cancer of the Colon and Rectum. Journal of National Cancer Institute 61:709-714. Graham, S., and C. Mettlin, 1979. Diet and Colon Cancer. American Journal of Epidemiology 109(1);l-20. Greenwald, P., R.F. Korns, P.C. Nasca and P.E. Wolfgang, 1975. Cancer in United States. Journal of Cancer Research 35: 3507-3512. Gregor, 0., R. Toman, and F. Prusova, 1969. Gastrointestinal Cancer and Nutrition. Gut 10:1031-1034. Guthrie, H.A., and J.C. Sheer, 1981. Validity of a Dietary Score for . Assessing Nutrient Adequacy. Journal American of Dietetics Association 78:240244. Habs, M., and D. Schmahl, 1980. Diet and Cancer. Journal of Cancer Research C l i n i c a l Oncology 96:1-10. Haenszel, W., 1961. Cancer Mortality Among the Foreign Born in the United States. Journal National of Cancer Institute 26(1):37-132. Haenszel, W., and P. Correa, 1971. Cancer of the Colon and Rectum and Adematous Polyps. Cancer 28(1):14-24. Haenszel, W., and M. Kurihara, 1968. Studies of Japanese Migrants. I Mortality from Cancer and Other Diseases Among Japanese in the United States. Journal National Cancer Institute. 49:969-988. Haenszel, W., F.B. Locke, and M. Segi, 1980. A Case-Control Study of Large Bowel Cancer in Japan. Journal National of Cancer Institute 64:17-22. Haenszel, W., J.W. Ross, M. Segi, M. Kurihara, and D.B. Locke, 1973. Large Bowel Cancer in Hawaiian Japanese. Journal National of Cancer Institute 51:1765-1779. Hakulinen, T., 1981. A Mantel-Haenszel St a t i s t i c for Testing the Association Between a Polychotomous Exposure and a Rare Outcome. American Journal of Epidemiology 113(2):192-197. 154 Hal l , R.L. 1979. Naturally Occuring Toxicants and Food Additives: Our Preception and Management of Risks. Nutrition and Cancer 1(2):27-36. Hankin, J.H., V. Rawlings, and A. Nomura, 1978. Assessment of a Short Dietary Method for a Prospective Study on Cancer. American Journal of Cl i n i c a l Nutrition 31:355-359. Hankin, J.H., G.C. Rhoads, and G.A. Glober, 1975. A Dietary Method for an Epidemiological Study of Gastrointestinal Cancer. American Journal of C l i n i c a l Nutrition 28:1055-1061. Harman, D., 1972. Free-Radical Theory of Aging: Dietary Implication. American Journal of C l i n i c a l Nutrition 25:839-843. Health Protection Branch, 1977. Current Patterns in Canada, 1931-1974. Bureau of Epidemiology. Health & Welfare, Canada. Hentges, D.J., B.R. Maier, G.C. Burton, M.A. Flynn, and R.K. Tsutakawa, 1977. Effect of High Diet on the Fecal Bacterial Flora of Humans. Cancer Research 37:568-571. Higginson, J., 1966. Etiological Factors in Gastrointestinal Cancer in Man. Journal National of Cancer Institute 37:527-545. Higginson, J. 1983. Summary: Nutrition and Cancer. Cancer Research 43 (Suppl):2515S-2518S). Hilker, D.M., 1981. Carcinogens Naturally Occurring in Goods. Nutrition and Cancer 2(4)217-223. H i l l , M.J., 1974. Bacteria and the Etiology of Colonic Cancer. Cancer 34(3):815-818. H i l l , M.J., 1975a. The Etiology of Colon Cancer. CRC C r i t i c a l Reviews in Toxicology, October pp 31-82. H i l l , M.J., 1975b. Metabolic Epidemiology of Dietary Factors in Large Bowel Cancer. Cancer Research 35:3398-3402. H i l l , M.J., 1979. Bacterial Metabolism and Colon Cancer. Nutrition and Cancer 1(2):46-50. Hirayama, T., 1979. Diet and Cancer. Nutrition and Cancer 1(3).-67-81. Hirayama, T., J.A.H. Waterhouse, and J.F. Fraumeni, J r . , 1980. Cancer Risks by Site. UICC Technical Report Series Volume 41. International Union Against Cancer, Geneva. Hoey, J.C., Montvernary, and R. Lambert, 1981. Wine and Tobacco: Risk Factor for Gastric Cancer in France. American Journal of Epidemiology 113:668-674. 155 Horton, J., and G.J. H i l l , 1977. C l i n i c a l Oncology. W.B. Saunders Company, Toronto. Howell, M.A., 1975. Diet as an Etiological Factor in the Development of Cancer of the Colon and Rectum. Journal of Chronic Diseases 28:67-80. Howe, G.R., 1981. Some Methodological Issues in Epidemiology Studies of Fat and Cancer. Cancer Research 41:3731-3732. Huang, C.T.L., G.S. Gopaladrishna and B.L. Nichols, 1978. Fiber, Intestinal Steroids and Colon Cancer. American Journal of C l i n i c a l Nutrition, 31:516-526. International Union Against Cancer, 1970. Cancer Incidence in Five Continents. International Union Against Cancer, Geneva. •International Union Against Cancer, 1975. Colo-Rectal Cancer. UICC Technical Report Series, Volume 19, Geneva. Jain, M. G.M. Cook, F.G. Davis, M.G. Grace, G.R. Howe and A.B. Miller, 1980. A Case Control Study of Diet and Colorectal Cancer. International Journal Cancer 26:757-768. « Jain, M., G.R. Howe, K.C. Johnson and A.B. Mill e r , 1980. Evaluation of a Diet History Questionnaire for Epidemiologic Studies. American Journal of Epidemiology 111(2):212-219. James, W.P.T., S.A. Bingham, and T.J. Cole, 1981. Epidemiological Assesssment of Dietary Intake. Nutrition and Cancer 2(4):203-212. Jensen, O.M., 1982. Cancer of the Large Bowel: A 3-Fold Variation Between Denmark and Finland. Nutrition and Cancer. 4(1);20-22. Jensen, O.M., R. MacLennan, and J. Wahrendorf, 1982. Diet, Bowel Function, Fecal Characteristics, and Large Bowel Cancer in Denmark and Finland. Nutrition and Cancer 4(1):5-19. Jensen, O.M., J. Mosbech, M. Salaspuro, and T. Jhamaki, 1974. A Comparative Study of the Diagnostic Basis for Cancer of the Colon and Cancer of the Rectum in Denmark and Finland. International Journal of Epidemiology 3:183-186. Johnson, F.C, 1983. Cancer Promoters and Initiators, and the Free Radical Theory of Carcinogenesis. Nutrition and Cancer 4(3):169-170. Joossens, J.V., and J. Geboers, 1981. Nutrition and Gastric Cancer Nutrition and Cancer 2(4):250-261. 156 Jukes, T.H., 1976. Nitrates and Nitrites as Components of the Normal Environment. Food and Nutrition News Pt-1. 47(5):l-4; Pt.2 48(1):1-4. Jukes, T.H., 1977. Current Concepts in Nutrition. New England Journal of Medicine 297(8):427-430. Kaunitz, H. 1977. Repair Function of Cholesterol Versus the Lipid Theory of Arteriosclerosis. Chemisty and Industry, 17 September:761-763. Kay, R.M., and A.S. Truswell, 1977. Effect of Citrus Pectin on Blood Lipids and Fecal Steroid Excretion in Man. American Journal of Cli n i c a l Nutrition 30:171-175. Key, R.M., Z.I., Sabry, and A. Csima, 1980. Multivariate Analysis of Diet and Serum Lipids in Normal Men. American Journal of C l i n c i a l Nutrition 33:2566-2572. Keeney, M., 1981. Comments on the Effect of Dietary Transfatty Acids in Humans. Cancer Research 41:3743-3744. Klurfeld, D.M., E. Aglow, S.A. Tepper, and D. Kritchevsky, 1983. Modification of Dimethylhydrazine-Induced Carcinogenesis in Rats by Dietary Cholesteral. Nutrition and Cancer 5(l):21-23. Kolonel, L.N., 1980. Cancer Patterns in Four Ethnic Groups in Hawaii. Journal National Cancer Institute 65:1127-1139. Kolonel, L.N., H.H. Hankin, J. Lee, S.Y. Chu. A.M.Y. Nomura, and M.W. Hinds, 1981. Nutrient Intakes in Relation to Cancer Incidence in Hawaii. B r i t i s h Journal of Cancer 44:332-339. Kolonel, L.N., A.M.Y. Nomura, M.W. Hinds, T. Hirohata, H.H. Hankin and H. Lee, 1983. Role of Diet in Cancer Incidence in Hawaii. Cancer Research 43(Suppl): 2397s-2402s). Kono, S., and M. Ikeda, 1979. Correlation Between Cancer Mortality and Alcoholic Beverages in Japan. British Journal of Cancer 40:449-455. Kritchevstay, D., 1983. Fiber, Steroids, And Cancer. Cancer Research 43(Suppl):2491s-2495s) Kummet, T., T.E. Moon, and F.L. Meyskers, 1983. Vitamin A: Evidence for Its Preventive Role in Human Cancer. Nutrition and Cancer 5(2):96-106. Lai, C. 1983. SPSS: User Manual. University, of British Columbia, Vancouver. B.C. 157 Lee, J., L.N. Kolonel, and H.H. Hankin, 1983. On Establishing the Interchangeability of Different Dietary Intake Assessment Methods Used in Studies of Diet and Cancer. Nutrition and Cancer 5:215-218. Lee, J.A.Hi, 1976. Recent Trends of Large Bowel Cancer in Japan Compared to United States and England and Wales. International Journal of Epidemiology 5(2): 187-194. Lee, J.B., 1982. Prostaglandins. Elsevier North Holland, Inc., New York. Lee, L.J., M.C. Archer, and W.R. Bruce, 1981. Absence of Volatile Nitrosamines in Human Feces. Cancer Research 41:3992-3994. Leeds, A.R., and M.A. Gassell, 1976. Hypocholesteremic Effect of Pectin. Lancet 2:637-639. L e f f a l l , L.D. Jr. and E.B. Chung, 1974. Surgical Management of Colorectal Polyps. Cancer 34:940-947. Lijinsky, W., and S.S. Epstein, 1970. Nitrosamines as Environmental Carcinogens. Nature 225:21-22. Lipkin, M., S. Scherf, L. Schechter and D. Braun, Jr., 1980. Memorial Hospital Registry of Population Groups at High Risk for Cancer of the Large Intestine: Age of Onset of Neoplasms. Preventive Medicine 9:335-345. Lovett, E., 1974. Familial Factors in the Etiology of Carcinoma of the Large Bowel. Proc. Roy. See Med. 67:751. Lowenfels, A.B., and M.E. Anderson, 1977. Diet and Cancer. Cancer Research 39:1809-1814. Lyon, J.L. 1976. Cancer Incidence in Mormons and Non-Mormons in Utah, 1966-1970. New England Journal of Medicine 294: 129-133. Lyon, J.L. and A.W. Sorenson, 1978. Colon Cancer in a Low-Risk Population. American Journal of C l i n i c a l Nutrition 31:227s-230s. MacCay, P.B., M.M. King, and J.V. Pitha, 1981. Evidence that Effectiveness of Antioxidants as Inhibitors of 7, 12-dimethylbinz(a)antbracen induced Mammary Tumors is a Function of Dietary Fat Composition. Cancer Research 41:3745-3748. MacGregor, I.L., 1974. Carcinoma of the Colon and Stomach. Journal of American Medical Association 227(8):911-915. MacLennan R., 1979. Diet and Human Carcinogenesis with Special Reference to Colon Cancer. Nutrition and Cancer l(2):42-45. 158 MacLennan, R., O.M. Jensen, H. Mosbech, and H. Vuori, 1978. Diet, Transit Time, Stool Weight, and Colon Cancer in Two Scandanavian Populations. American Journal of C l i n i c a l Nutrition 31:S239-S242. MacMahon, B., and T.F. Pugh, 1970. Epidemiology: Principles and Methods. L i t t l e Brown and Company, Boston. Mann, G.V., G. Pearson, T. Gordon, and R.T. Dawber, 1962. Diet and Cardiovascular Disease in the Framingham Study. I Measurement of Dietary Intake. American Journal of C l i n i c a l Nutrition 11:200-225. Mantel, N., and W. Haenszel, 1959. S t a t i s t i c a l Aspects of Data From Retrospective Studies of Diseases. Journal of National Cancer Institute 22:719-748. Marshall, J., R. Priore, B Jaughey, T. Rzephka and S. Graham, 1980. Spouse-Subject Interviews and the R e l i a b i l i t y of Diet Studies. American Journal of Epidemiology 112(5):675-683. Maskens, A.P., 1981. Confirmation of the Two Step Nature of Chemical Carcinogenesis in the Rat Colon Adenocarcinoma Model. Cancer Research 41:1240:-1245. Mass, N., and B. Modan, 1969. Epidemiological Aspects of Neoplastic Disorder in I s r a e l i Migrant Population IV. Cancer of the Colon and Rectum. Journal of National Cancer Institute 42:529-536. McMicheals, A.J., J.D. Potter, and B.S. Hetzel, 1979. Time Trends in Colo-Rectal Cancer Mortality in Relation to Food and Alcohol Consumption: United States, United Kingdom, Australia, and New Zealand. International Journal of Epidemiology 8(4):295-303. Mettlin, C.J., and S. Graham, 1979. Methodological Issues in Etiologic Studies of Diet and Colon Cancer. Nutrition and Cancer l(4):46-55. Mettlin, C , S. Graham, R. Priore, I. Marshall, and M. Swanson, 1981. Diet and Cancer of the Esophagus. Nutrition and Cancer 2(3):143-147. Miettinen, 0., 1976. Estimability and Estimation in Cancer Referrent Studies. American Journal of Epidemiology 103(2):226-235. Mi l l e r , J.A., and E.C. Miller, 1976. Carcinogens Occurring Naturally in Goods. Federation Proceedings. 35:1316:1321. Mi l l e r , S.R., P.I. Tartter, A.E. Papastestas, G. Slater, and A.H. Aufsis, 1981. Serum Cholesterol and Human Colon Cancer. Journal National of Cancer Institute 67(2)297-300. 159 Mirvish, S.S., 1970. Kinetics of Dimethylamine Nitrosation in Relation to Nitrosamine Carcinogenesis. Journal of National Cancer Institute 44:633-639. Modan, B. 1980. Role of Migrant Studies in Understanding the Etiology of Cancer. American Journal of Epidemiology 112(2) 289-295. Modan, B., V. Barell, F. Lubin, and M. Modan, 1975. Dietary Factors and Cancer in Israel. Cancer Research 35: 3503-3506. Morgan, R.W., M. Jain, A.B. Mill e r , N.W. Choi, V. Matthews, L. Munon, I.D. Buich, J. Flather, G.R. Howe, and A. Kelly, 1978. A Comparison of Dietary Methods in Epidemiological Studies. American Journal of Epidemiology 107:488-498. Moore, W.E.C., and L.V. Holdeman, 1975. Discussion of Current Bacteriological Investigations of the Relationships Between Intestinal Flora, Diet and Colon Cancer. Cancer Research 35 :3418-342. Morrison, A.S., 1979. Sequential Pathogenic Components of Rates. American Journal of Epidemiology 109(6)709-718. Morson, B.C., 1974. Evaluation,of Cancer of the Colon and Rectum. Cancer 35:845-849. Narisawa, T., N.E. Magadia, J.W. Wiesburger, and E.L. Wynder. 1974. Promoting Effect of Bile Acids on Colon Carcinogenesis After Intrarectal Installation of N-Methyl-N^-nitro-N-nitrosoguaridine in Rats. Journal of the National Cancer Institute 53:1093-1097. National Research Council, 1982. Diet, Nutrition and Cancer. National Acadamy Press. Washington, D.C. Newburne, P.M. and V. Suphakarn. 1977. Preventive Role of Vitamin A in Colon Carcinogenesis in Rats. Cancer 40: 2553-2556 Newburne, P.M. and G.B. Thurman, 1981. Lipids and the Immune System: Report and Recommendation. Cancer Research 41:3803-3804. Nie, N.H., CH. Hull, J.G. Jenkins, K. Steinbrenner, and D.H. Bent, 1975. SPSS: S t a t i s t i c a l Package for the Social Sciences, Second Edition. McGraw-Hill Book Company, New York. Nigro, N.D., 1981. Animal Studies Implicating Fat and Fecal Steroids in Intestinal Cancer. Cancer Research 41:3760-3770. 160 Nomura, A., J.H. Hankin, and G.G. Rhoads, 1976. The Reproductibility of Dietary Intake Data in a Prospective Study of the Gastrointestinal Cancer. American Journal of C l i n i c a l Nutrition 29:1432-1436. Pasternak, B.S., and R.E. Shore, 1981. Sample Sizes for Group Sequential Cohort and Case - Control Study Designs. American Journal of Epidemiology 113(2):182-191. Peto, R.R., Doll, J.D. Buckley, and M.B. Sporn, 1981. Can Dietary A-Carotene Materially Reduce Human Cancer Rates? Nature 290:201-208. Ph i l l i p s , R., 1975. Role of Life-Style and Dietary Habits in Risk of Cancer Among Seventh-Day Adventists. Cancer Research 35:3513-3522. Ph i l l i p s , R.L., J.W. Bieson, and T. Lotz, 1980. Influence of Selection Versus Lifestyles on Risk of Fatal Cancer and Cardiovascular Disease among Seventh-Day Adventists. American Journal of Epidemiology 112(2):296-314. P h i l l i p s , R.L. and D.A. Snowdon, 1983. Association of Meat and Coffee Use with Cancers of the Large Bowel, Breast and Prostate Among Seventh-Day Adventists. Preliminary Results. Cance Research 43 (Suppl): 2403s-2408s. Pitot, H.C., 1981. Fundamentals of Oncology. Second Edition. Marcel Dikker, Inc., New York. Rasanen, L., and P. Pietiner, 1982. A Short Questionnaire Method for Evaluation of Diets. Preventive Medicine 11:669-676. Rawson, R.W., 1980. The Role of Nutrition in the Etiology and Prevention of Cancer. Nutrition and Cancer 2(1):17-21. Reddy, B.S., G. Ekelund, M. Bohe, A., Engle, and L. Domellof, 1983. Metabolic Epidemiology of Colon Cancer: Dietary Patten and Fecal Stool Concentrations of Three Poplations. Nutrition and Cancer 5(1) :34-40. Reddy, B.S., 1981. Dietary Fat and i t s Relationship to Large Bowel Cancer. Cancer Research 41:3700-3705. Reddy, B.S., D. Hanson, S. Mangat, L. Matthews, M. Sbaschnig, C. Sharma, and B. Simi, 1980. Effect of High-Fat, High-Beef Diet and of Mode of Cooking of Beef in the Diet on Fecal" Bacterial Enzymes and Fecal Bile Acids and Neutral Steroids. Journal of Nutrition 110:1880-1887. Reddy, B.S., A.S. Mastromarino, and E.L. Wynder, 1975. Further Leads in Metabolic Epidemiology of Large Bowel Cancer. Cancer Research 35:3400-3403. 161 Reddy, B.S., and E.L. Wynder, 1973. Large-Bowel Carcinogenesis: Fecal Constitutents of Populations with Diverse Incidence Rates of Colon Cancer. Journal National Cancer Institute 50:1437-1442. Review, 1976. The Validity of 24-hour Dietary Recalls. Nutrition Review 34:310-311. Roediger, W.E., 1980. The Colonic Epithelium in Ulcerative C o l i t i s : An Energy-Deficiency Disease? Lancet (Oct. 4):712-715. Rooma, M. and J. Uibu, 1983. Influence of Various Milk Products on the Concentration of N i t r i t e and the Formation of Nitrosodimethylamine in Vitro. Nutrition and Cancer 4(3):171-175. Rozen, P., S.M. Hellerstein, and C. Horwitz, 1981. The Low Incidence of Colorectal Cancer in a "High Risk" Population. Cancer 48:2692-2695. Sabine, J.R., 1979. Susceptibility to Cancer and the Influence of Nutrition. Nutrition and Cancer l(3):52-57. Sani, B.P., S.M. Condon, R.W. Brockman, L.H. Willand, and A.J. Schutt, 1980. Retinoic Acid-Binding Protein in Experimental and Human Colon Tumors. Cancer 45:1199-1296. * Scanlaw, R.A., 1983. Formation and Occurrence of Nitrosamines in Food. Cancer Research 43(Suppl):2435s-2440s). Sp i n e l l i , J., 1983. RELRISK: Program on Apple Computer. Cancer Control Agency of B.C. Spiro, H., 1977. C l i n i c a l Gastrointerology. Second Edition. MacMillan Publishing Co., Toronto. Staszewski, J., and W. Haenszel, 1965. Cancer Mortality Among the Polish-Born in the United States. Journal National of Cancer Institute 35:291-297. Staszewski, J., M.G. McCall, and N.S. Stenhouse, 1971. Cancer Mortality in 1962-66 Among Polish Migrants ot Australia. British Journal Cancer 25:599-610. Stemmerman, G.N., A.M.Y. Nomura, L.K. Heilburn, E.S. Polloack, and A. - Kagen, 1981. Serum Cholesterol and Colon Cancer Incidence in Hawaiian Japanese Men. Journal National Cancer Institute 67:1179-1182. Tomita, Y., 1983. Immunological Role of Vitamin A and Its Related Substances in Prevention of Cancer. Nutrition and Cancer 5(3-4):187-194. 162 Trulson, M.F., and M.B. McCann, 1959. Comparison of Dietary Survey Methods. Journal American of Dietition Association 35:672-676. Tulinius, H., 1981. Choice of Methods for Gathering Nutrition Infomation. Nutrition and Cancer 2(4):200-202 . Tuyns, A.J.,1983. Protective Effect of Citrus Fruit on Esophageal Cancer. Nutrition and Cancer 5 (3-4):195-200. Vital e , J.J., and S.A. Broitman, 1981. Lipids and Immune Function. Cancer Research 41:3706-3710. Wald, N. , M. Idle, J. Boreham, and A. Bailey, 1980. Low Serum Vitamin A and Subsequent Risk of Cancer. Lancet (Oct.' 18):813-815. Walker, A.R.P., and D.P. Burkitt, 1976. Colonic Cancer-Hypotheses of Causation, Dietary Prophylaxis and Future Research. Digestive Diseases 21 (10):910-917. j Walker, M.A., and M.M. H i l l , 1975. Homemaker' Food and Nutrition Knowledge, Practices and Opinions. Home Economics Research Report No. 39. United States Department of Agriculture, Washington. D.C. Walker, R., 1975. Naturally Ocpurring Nitrate/Nitrite in Food. Journal Science of Food Agriculture 26:1735-1742. Walter, S.D., 1977. Determination of Significant Relative Risks and Optimal Sampling Procedures in Prospective and Retrospective Comparative Studies of Various Sizes. American Journal of Epidemiology 105:387-397. Waterhouse, J., C. Muir, P. Correa and J. Powell, 1976. Cancer Incidence in Five Continents. Volume III. International Agency for Research on Cancer. World Health Organization. Switzerland. Weisburger, J.H., H. Marquardt, H.F. Mower, N. Hirota, H. Mori, and G. Williams, 1980. Inhibition of Carcinogenesis: Vitamin C and the Prevention of Gastric Cancer. Preventive Medicine 9:352-361. White, J.W., 1975. Relative Significance of Dietary Sources of Nitrate and N i t r i t e . Journal of Agriculture, Food Chemistry 23(5):386-891. Williams, R.R., P.D. Sorlie, M. Feinleib, P.M. McNamara, W.B. Kannel, and T.R. Dawber, 1981. Cancer Incidence by Levels of Cholesterol. Journal American Medical Association 245(3):247-252. Williams, R.R., N.L. Stegens, and J.W. Horn, 1977. Patient Interview Study from the National Cancer Survey: Overview of Problems and Potential of These Data. Journal National Cancer Institute 58(3):519-524. 163 Wolf, G., 1982. Is Dietary/S- Carotene an Anticancer Agent? Nutrition Reviews40(9):257-260. Wolff, W.I., and H. Shinya, 1974. Earlier Diagnosis of Cancer of the Colon Through Colonic Endoscopy. Cancer 34:912-931. Wood, D.A., 1967. Tumors of the Intestines. Sections VI in: Atlas of Tumor Pathology. Armed Forces Institute of Pathology. Washington, D.C.E. Wynder, E.L., 1975. The Epidemiology of Large Bowel Cancer. Cancer Research 35:3388-3394. Wynder, E.L., 1979. Dietary Habits and Cancer Epidemiology. Cancer 43:3388-3394. Wynder, E.L., L. Hyams, and T. Shigematsu, 1967. Correlation of International Cancer Death Rates. Cancer 20(1):113-126. Wynder, E.L., and K. Mabuchi, 1972. Etiological and Preventive Aspects of Human Cancer. Preventive Medicine 1:300-334. Wynder, E.L., and B.S. Reddy, 1974. Metabolic Epidemiology of Colorectal Cancer. Cancer 43( 3) : 801,-806 . Wynder, E.L., and T. Shigematsu, 1967. Environmental Factors of Cancer of the Colon and Rectum. Cancer 29(9):1520-1561. Young, CM., 1959. The Interview It s e l f . Journal American Dietetics Association 35:677-681. Zaldivar, R., 1977. Epidemiology of Gastric and Colo-Rectal Cancer in the United States and Chile with Particular Reference to the Role of Dietary and Nutritional Variables, Nitrate F e r t i l i z e r Pollution, and N-Nitroso Compounds. 361 Bait. Hyg.I. Abt. Orig. B. 164:193-217. Zaldivar, R., and W.H. Witterstrand. 1976. High Fibre, Low Saturated Fat Diets and the Etiology of Colorectal Carcinomata in a Low Risk Population Z. Krebsforsch 87:41-45. Zaridze, D.G. 1981. Diet and Cancer of the Large Bowel Nutrition and Cancer 21 (4):241-249. Diet and Health Study, Preliminary Report. Alberta. 164 i APPENDIX 1: THE FINAL INTERVIEW SCHEDULE 165 NUTRITION STUDY INTERVIEW SCHEDULE HELLO, MY NAME IS I AM CONDUCTING AN INTERVIEW FOR A STUDY ON LIFESTYLE AND FOOD HABITS. OTHER INTERVIEWERS AND I ARE TALKING WITH PEOPLE WHO HAVE CERTAIN ILLNESSES AND WITH OTHERS WHO DO NOT. YOU RECEIVED A LETTER ABOUT THE STUDY WHICH ARRANGED FOR AN INTERVIEW WITH YOU TODAY. MY INTERVIEW TAKES BETWEEN ONE AND TWO HOURS. ALL INFORMATION IS CONFIDENTIAL AND USED FOR SURVEY PURPOSES ONLY. YOU MAY STOP THE INTERVIEW AT ANY TIME IF YOU ARE TIRED OR DO NOT WISH TO CONTINUE. WOULD YOU BE WILLING TO PROCEED? C i r c l e : Yes No Subject no. Name: Address: Date of I n i t i a l Contact: (phone) Number of Interviewer c a l l b a c k s ( p e r s o n a l ) : C i r c l e 1 2 3 4 Date of Interview: Time of Interview: s t a r t f i n i s h dur at i o n minut es I n t e r v i e w e r : 166 Subject no. P e r s o n a l Data: FIRSTLY, I WOULD LIKE TO ASK YOU SOME QUESTIONS ABOUT YOUR PERSONAL HISTORY, FOR EXAMPLE, WHERE YOU WERE BORN, GREW UP, IF YOU WERE MARRIED, ETC, Record sex: 1 = female 2 = male • 2. WHAT WAS YOUR BIRTH DATE? year month day y r . mo. day 3. WHERE (IN WHICH CITY, COUNTRY) WERE YOU BORN? o o o c o country HOW LONG DID YOU LIVE THERE? YRS. y r s 5. AFTER THAT, PLEASE TELL ME ALL THE PLACES YOU HAVE LIVED (FOR MORE THAN ONE YEAR) AND FOR HOW LONG? Record only i f f o r more than one year. Note i f r u r a l or urban. Urban = more than 1000 persons. Note years from and t o . City/town P r o v i n c e / S t a t e / Years From Country Prov, Country and To T-I CM C r H Xi U = 1 OS II a o o o o o o c o o o o o c o o o o o o o o o o o o o c o o o o c o o c o c o o o o 167 Subject no, WHAT WAS THE HIGHEST EDUCATIONAL QUALIFICATION THAT YOU ATTAINED ( i . e . YOUR HIGHEST GRADE IN SCHOOL OR ANY FURTHER EDUCATION SUCH AS UNIVERSITY, COLLEGE OR TRADE SCHOOL)? 00 = none 01 = grades 1 - 8 02 = grades 8 - high s c h o o l g r a d u a t i o n 03 = <2 years post secondary 04 = 2 - 4 years post secondary 05 = > 4 years post secondary 99 = unknown, no response yr, 7. ARE YOU ASSOCIATED WITH ANY CHURCH OR FAITH? IF YES, WHAT WAS THAT? 00 = no a s s o c i a t i o n 05 = B u d d i s t , S h i n t o , H i n d i 01 = P r o t e s t a n t 06 = 7th Day Ad v e n t i s t 02 = C a t h o l i c 07 = Mormon 03 = Jewish 08 = Other 04 = Muslim 99 = unknown, no response 8. FROM WHAT COUNTRY IS YOUR FATHER? Probe f o r . p r o v i n c e , e t h n i c i t y , country of o r i g i n . oo ooo f a t h e r FROM WHAT COUNTRY IS YOUR MOTHER? oo ooo mother 10. WHAT IS YOUR CURRENT MARITAL STATUS? 1 = s i n g l e 5 = l i v i n g t o g e t h e r < 1 yr. 2 = married 6 = l i v i n g together >1 yr, 3 = d i v o r c e d 7 = separated 4 = widowed 9 = unknown, no response • m. s t a t u s 11. IF YOU ARE MARRIED OR HAVE BEEN MARRIED, AT WHAT AGE WERE YOU MARRIED AND FOR HOW LONG? age l e n g t h ( y r s . ) ( y r s . ) Age: How long 168 Subject no. STARTING WITH YOUR PRESENT (OR LAST) OCCUPATION COULD YOU TELL ME ABOUT EVERY JOB YOU HAVE HELD (FOR SIX MONTHS OR LONGER) DURING YOUR LIFE. FOR EVERY JOB, I WOULD LIKE TO KNOW £ WHAT YEAR YOU BEGAN AND HOW LONG YOU £ HELD THE JOB. 3 C i—( Job Industry Year D u r a t i o n £ ( y r s . ) >-3 o o o o c o coo ceo coo coo coo ooo c e o o c o c o o o o o o o o o o o c o o o c o 169 Subject no. 13. I WILL NOW READ A LIST OF CHEMICALS AND OTHER MATERIALS. I WOULD LIKE YOU TO TELL ME IF YOU HAVE EVER WORKED WITH OR BEEN EXPOSED TO THESE, WHETHER ON THE JOB OR THROUGH WORK AT HOME OR HOBBIES. For each "yes" answer, ask about SOURCE of exposure and INTENSITY of exposure. I f r e g u l a r or g r e a t e r exposure ( i . e . 1 or 2 ) , ask about AGE of f i r s t exposure and t o t a l TIME of exposure. Substance: A. B. C. E. G. H. I. J . K. M. Chemicals used i n a g r i c u l t u r e ( f e r t i l i z e r s , lime, p e s t i c i d e s ) Animal products ( h i d e s , f u r , l e a t h e r ) Chemicals used i n manufacture of p l a s t i c s , rubber or s i m i l a r products Chemicals used i n manufacture of p a i n t s , p a i n t remover, s o l v e n t s , v a r n i s h e s Chemicals used as d y e s t u f f s , bleaches or f o r tanni n g l e a t h e r C u t t i n g o i l , c o o l i n g or l i g h t machine o i l Petroleum products i n c l u d i n g g a s o l i n e , engine o i l Fumes from cooking o i l or f a t Coal t a r products, p i t c h , c o a l dust Asbestos as i n i n s u l a t i o n , brake l i n i n g s , b u i l d i n g m a t e r i a l s , mining or m i l l i n g of asbestos Dust from wood, wood products, g r a i n , " t e x t i l e s , f i b r e g l a s s Fumes from o r g a n i c s o l v e n t s eg. benzene, acetone, s o l v e n t s , or i n o r g a n i c chemicals Other, s p e c i f y CD U en 3 J 3 W -M 0 C a o X E CD i—i •p aS >> w -P i H 0 • H 3 a> w =H *—• « o a o o u CD a 3 4-> a> s 0 C •H w 00 t—I <C Exposure: 0 = no 1 = yes (ask s o u r c e , i n t e n s i t y ) Source: 1 = d i r e c t o c c u p a t i o n a l (worked with i t ) 2 = o c c u p a t i o n a l environment 3 = hobby 4 = home environment 9 = unknown, no response I n t e n s i t y : 1 = f o r the gr e a t e r p a r t of each day (g r e a t e r than 20 hours per week) 2 = r e g u l a r contact (5-19 hours per wk) 3 = o c c a s i o n a l contact (1-4 hours/wk) 4 = only r a r e l y (1 hour or l e s s per week) or s e a s o n a l l y Age: Record i n years. 1 ( r e c o r d age, time) Time of exposure: Record i n months; e.g. 9 months = 009 170 Subject no. Current Food H a b i t s : NOW I WOULD LIKE TO ASK YOU SOME QUESTIONS ABOUT YOUR CURRENT AND FORMER FOOD HABITS. FIRST YOUR CURRENT HABITS. 14. WITHIN THE LAST YEAR,HAVE THERE BEEN ANY CHANGES IN YOUR FOOD HABITS (eg. MEAL PATTERNS, FOOD REDUCTIONS, ETC.)? 0 = no 1 = yes IF YES, DESCRIBE THE CHANGES: changes PRIOR TO THESE CHANGES, IF ANY, OR BEFORE ONE YEAR AGO: Record time p e r i o d chosen -15. WHICH MEALS AND SNACKS DID YOU EAT REGULARLY ON WORKDAYS OR WEEKDAYS? Enter meal name by time. Code: 0 = no; 1 = yes, i f meal eaten i n time p e r i o d . Time a. 0500 - 0859 a. b. 0900 - 1059 b. c. 1100 - 1359 c. d. 1400 - 1659 d. e. 1700 - 1959 e. f . 2000 - 0459 f . 16. WHICH. OF THESE WAS YOUR LARGEST MEAL? . l a r g e s t 17. HOW LONG DOES/DID IT USUALLY TAKE YOU TO EAT YOUR LARGEST MEAL? (minutes) minutes 171 Subject no, 18. DESCRIBE A TYPICAL EXAMPLE OF YOUR LARGEST DAILY MEAL. Note q u a n t i t y and food c a t e g o r i e s (eg. j u i c e § cup). 19. ON WORKDAYS/WEEKDAYS DO YOU REGULARLY SKIP MEALS? Code: 0 = no 1 = yes If yes, WHICH MEALS? HOW OFTEN? Enter frequency code by a p p r o p r i a t e meal time. Meal Times: 0 = never 1 = <once a month 2 = a few times a month 3 = once a week 4 = 2 - 3 times a week 5 = most days 6 = d a i l y 0500 0900 1100 1400 1700 2000 0859 1059 1359 1659 1959 0459 Reminder: Current food h a b i t s before any recent changes/about one year ago. miss? | [ frequency code 20. ON WEEKENDS/ DAYS OFF MEALS? DO YOU REGULARLY SKIP Code: 0 = no 1 = yes If yes, WHICH MEALS? HOW OFTEN? Enter frequency code as above (#19) by a p p r o p r i a t e meal time. Meal Times: 0500 0900 1100 1400 1700 2000 0859 1059 1359 1659 1959 0459 • frequency 172 Subject no 21. I HAVE SOME ILLUSTRATIONS OF .VARIOUS SIZES OF PORTIONS OF VARIOUS FOODS. COULD YOU PLEASE INDICATE WHICH WOULD BE A TYPICAL AMOUNT OF EACH FOOD THAT YOU WOULD CONSUME AT A SINGLE MEAL. Show i l l u s t r a t i o n s . E n ter p o r t i o n s i z e s : 1, 2 or 3. I f l a r g e r s i z e s i n d i c a t e d , enter m u l t i p l e s , i . e . #3=6, Foods: a. Chicken b. Peas c. C o l d Cuts d. P a s t a e. Butter/Margarine f . J u i c e g. M i l k a. b. c. d. e. f. g-P o r t i o n 22. HOW OFTEN DO YOU EAT THE FOLLOWING FOODS? HOW ARE THEY USUALLY PREPARED? Give i n t e r v i e w e e g o l d code ca r d . Record frequency and p r e p a r a t i o n codes. How Frequency Prepared Frequency: Food L i s t : 0 = n e v e r a. deep yellow v e g e t a b l e s , eg. 1 = <once a month yams, c a r r o t s , w i n t e r squash 2 = a few times a b'. b r o c c o l i , b r u s s e l sprouts month (<4) c. c a u l i f l o w e r , cabbage 3 = once a week d. l e a f y greens: l e t t u c e , spinach 4 = 2-3 times a week e. corn , c e l e r y , s t r i n g beans 5 = most days f. peanuts, peanut b u t t e r 6 = d a i l y g- d r i e d b e a n s / p e a s / l e n t i l s h. f i s h or p o u l t r y , eg. chic k e n , How prepared: turkey, e t c . i . b eef: steak, r o a s t , hamburger 0 = as purchased J • cured meats: ham, s a l a m i , 1 = baked weiners, luncheon meat, 2 = steamed sausage, bacon 3 = b o i l e d k. cheese 4 = f r i e d 1. eggs 5 = smoked m. c i t r u s f r u i t or j u i c e 6 c h a r c o a l g r i l l e d n. tomatoes, i n c l . j u i c e / s a u c e 7 = p i c k l e d o. p a s t a : s p a g h e t t i . n o o d l e s , e t c . 8 = o t h e r , r e c o r d P- i c e cream, cream 9 ss unknown, no q- s a l a d dressing.mayonnaise response r . whole g r a i n b r e a d / c e r e a l s s. r e f i n e d b r e a d s / c e r e a l s , eg. white bread,corn f l a k e s , e t c . 1st 2nd 173 Subject no. 23. HAVE YOU EVER BEEN A VEGETARIAN? Code: 0 = no 1 = yes If yes, FOR HOW LONG? years. Years No/Yes 24. ARE YOU CURRENTLY A VEGETARIAN? Code: 0 = no 1 = yes I f yes, proceed to #29. 25. IF YOU ARE NOT A VEGETARIAN, WHICH TYPE OF MEAT DO YOU CONSUME MOST FREQUENTLY? 26. DO YOU TRIM THE FAT FROM YOUR MEAT BEFORE PREPARING IT? Code: 0 = no 8 = sometimes, s p e c i f y 27. DO YOU TRIM THE FAT FROM YOUR MEAT BEFORE EATING IT? Code: 0 = no 1 = yes 9 = unknown, no response 28. DO YOU EVER EAT MEAT AT MEALS WITHOUT VEGETABLES OR FRUIT? AT THE SAME MEAL (NOT SNACK)? Give examples, i f u n c l e a r , eg. bacon and eggs and no j u i c e ; sandwiches/ milk/no j u i c e / n o s a l a d ; meat or cheese and pasta/no v e g e t a b l e s . Code: 0 = no 1 = yes 9 = no response, unknown If yes, HOW OFTEN ?_; See frequency codes Question #22. 29. WHAT KIND OF BREAD DO YOU EAT MOST OFTEN? Types: 0 = do not eat bread 1 = rye 2 = pumpernickel 3 = white, i n c l . French, I t a l i a n 4 = cracked wheat 5 = whole wheat 6 = ot h e r , s p e c i f y .  9 = unknown, no response 30. WHAT DO YOU USE MOST OFTEN AS A SPREAD ON YOUR BREAD? Probe f o r brand of margarine. I f unsure of type, r e c o r d brand under ' o t h e r 1 . 1 = 9 = yes unknown, no response which not trimmed Types: 0 = none 1 = b u t t e r 2 = margarine 3 4 5 9 po l y u n s a t u r a t e d margarine l a r d o t h e r , s p e c i f y : unknown, no response 174 Subject no. 31. WHEN YOU EAT FRIED FOODS, WHAT KIND OF FAT ARE THEY MOST FREQUENTLY COOKED IN? Record the three most frequent c h o i c e s . 1st 2nd 3rd Types: 0 = do not eat f r i e d foods 1 = b u t t e r 7 = beef f a t 2 = margarine 8 = o l i v e o i l 3 = l a r d 9 = peanut o i l 4' = bacon f a t 10 = other vegetable o i l 5 = s h o r t e n i n g 11 = oth e r , s p e c i f y 6 = p o u l t r y d r i p p i n g s 99 = unknown, no response 32. DO YOU USE HOME CANNED OR BOTTLED FOODS? Code: 0 = no 1 = yes 9 = unknown, no response I f yes, WHAT FOODS DO YOU EAT THAT ARE PREPARED IN THIS MANNER? 33. WHAT DO YOU USUALLY DO WITH LEFTOVER COOKED VEGETABLES? Probe to determine i f d i s c a r d immediately, c o o l and r e f r i g e r a t e , r e f r i g e r a t e immediately, use at a l a t e r date, d i s c a r d at a l a t e r date. E s p e c i a l l y i n t e r e s t e d i n c a r r o t s , beets, spinach i f eaten. 34. WHAT TYPE OF CHEESE DO YOU EAT MOST OFTEN? Record three most frequent c h o i c e s . 35. IF YOU HAVE CHEESE WITH MOULD (NOT INTENDED AS IN BLUE OR ROQUEFORT) , DO YOU: 1 = DISCARD THE CHEESE 2 = REMOVE THE MOULD AND EAT THE CHEESE 3 = EAT THE CHEESE AND THE MOULD 0 = do not eat cheese 9 - unknown, no response 175 Subject no. 36. HOW OFTEN DO YOU DRINK EACH OF THE FOLLOWING BEVERAGES? Record frequency code. Frequency: 0 = never MILK 1 = <once a month 2 = a few times a month COFFEE 3 = once a week 4 = 2-3 times a week TEA 5 = most days 6 d a i l y SODA POP DIET POP 37. HOW FREQUENTLY DO YOU CONSUME THESE ALCOHOLIC BEVERAGES? Record frequency code as i n #36 above, BEER WINE 38. HOW MUCH SUGAR DO YOU ADD TO: Record number of teaspoons? 89 = do not consume 99 = unknown, no response 39, 40, * Request person draw a  r e p r e s e n t a t i o n of a teaspoon. DO YOU USE JAM, JELLY, MARMALADE, HONEY? Code: 0 = no 1 = yes If yes, ESTIMATE YOUR TOTAL DAILY USE OF THESE IN NUMBER OF TEASPOONS t s p Record # teaspoons LIQUOR LIQUERS TEA COFFEE BREAKFAST CEREAL No/Yes Tsp. Code: 99 = unknown, no response S i z e DO YOU TAKE OR HAVE YOU EVER TAKEN ON A REGULAR BASIS A LARGE DOSE OF VITAMIN C ( 1000 mg./day), VITAMIN E( 400 IU/day)^ OR BOTH? Regular = at l e a s t once a week. Code: 1 = Vi t a m i n C 2 = Vitamin E 3 = both 0 = no 8 = not sure of dosage 9 = unknown, no response I f yes, WHEN, WHAT TYPE ( r e c o r d brand i f a v a i l a b l e ) WHAT AMOUNTS, AND FOR HOW LONG? 176 Subject no. Food H a b i t s i n the Middle Years COULD WE NOW SWITCH TO THINKING ABOUT YOUR MIDDLE YEARS, SPECIFICALLY 10 TO 15 YEARS AGO. YOU WERE LIVING IN (I n t e r v i e w e r to f i l l i n from p r e v i o u s data to e s t a b l i s h s e t t i n g ) THAT WAS BETWEEN 1967 AND 1972. S e l e c t a p e r i o d i n which the s u b j e c t was f a i r l y g e o g r a p h i c a l l s t a b l e w i t h i n the time frame. Record what time was chosen and why e.g. Spring,1969 s e t t l e d i n t o new job. WAS THIS A VERY STRESSFUL TIME IN YOUR LIFE? WERE THERE UNUSUAL EVENTS TAKING PLACE? I f yes, DESCRIBE THEM FOR ME PLEASE. (Record major events, eg. moves, marriage, bereavement, d i s m i s s a l from job, d i v o r c e ) . I WOULD LIKE TO ASK YOU SOME ADDITIONAL QUESTIONS ABOUT YOUR FOOD HABITS AT THAT TIME. 41. WHEN DID YOU EAT YOUR LARGEST OF THE DAY? Enter a p p r o p r i a t e meal time code (1-6): 1 = 0500 - 0859 - 5 = 1700 - 1959 2 = 0900 - 1059 6 = 2000 - 0459 3 = 1100 - 1359 4 = 1400 = 1659 9 = unknown 42. DESCRIBE A TYPICAL EXAMPLE OF YOUR LARGEST DAILY MEAL. Note q u a n t i t y and food c a t e g o r i e s , eg. j u i c e J cup: 43. AT THAT TIME, WHICH MEALS DID YOU REGULARLY SKIP, IF ANY? If p o s i t i v e response, ask HOW OFTEN? Record frequency code by a p p r o p r i a t e meal time. Frequency: Meal 0 = never Times: 0500 - 0859 1 = <once a month 0900 - 1059 2 = a few times a month 1100 - 1359 3 = once a week 1400 - 1659 4 2-3 times a week 1700 - 1959 5 = most days 2000 - 0459 6 = d a i l y 177 Subject no. 44. AT THAT TIME, HOW OFTEN DID YOU EAT THE FOLLOWING FOODS? IN WHAT WAY WERE THEY PREPARED? Record frequency and p r e p a r a t i o n codes. Frequency How Prepared Frequency: Food L i s t : 0 = never a. deep ye l l o w v e g e t a b l e s , eg. 1 = <once a month yams, c a r r o t s , winter squash 2 = a few times a b'. b r o c c o l i , b r u s s e l s p r o u t s month (<4) c. c a u l i f l o w e r , cabbage 3 = once a week d. l e a f y greens: l e t t u c e , s p i n a c h 4 = 2-3 times a week e. c o r n , c e l e r y , s t r i n g beans 5 = most days f. peanuts, peanut b u t t e r 6 = d a i l y fif. d r i e d b e a n s / p e a s / l e n t i l s h. f i s h or p o u l t r y , eg. c h i c k e n , How prepared: t u r k e y , e t c . i . b eef: steak, r o a s t , hamburger 0 = as purchased J • cured meats: ham, s a l a m i , 1 ss baked weiners, luncheon meat, 2 = steamed sausage, bacon 3 b o i l e d k. cheese 4 = f r i e d 1. eggs 5 - smoked m. c i t r u s f r u i t o r j u i c e 6 = c h a r c o a l g r i l l e d n. tomatoes, I n c l . j u i c e / s a u c e 7 = p i c k l e d o. p a s t a : s p a g h e t t i , n o o d l e s , e t c . 8 = o t h e r , r e c o r d P. i c e cream, cream 9 ss unknown, no q- s a l a d dressing,mayonnaise response r. whole g r a i n b r e a d / c e r e a l s s. r e f i n e d b r e a d s / c e r e a l s , eg. white bread,corn f l a k e s , e t c * 1st 2nd / WAS THIS REPRESENTATIVE OF YOUR EATING HABITS AT THIS TIME? I f not, WHY NOT? 178 Subject no. 45. DID YOU TRIM THE FAT FROM YOUR MEAT BEFORE EATING IT? Code: 0 = no 1 = yes 8 = d i d not eat meat 9 = unknown, no response 46. WHAT KIND OF BREAD DID YOU EAT MOST OFTEN? Types: 0 = d i d not eat bread 1 = rye 2 — pumpernickel 3 = white, i n c l u d i n g Freeh, I t a l i a n 4 = cracked wheat 5 = whole wheat 6 = ot h e r , s p e c i f y 9 = unknown, no response 47. WHAT SPREAD DID YOU USE ON YOUR BREAD MOST FREQUENTLY? Record type code. Types: 0 = none 1 = b u t t e r 2 = margarine 3 = p o l y u n s a t u r a t e d margarine 4 = l a r d 5 = o t h e r , s p e c i f y 9 = unknown 48. DID YOU EVER EAT MEAT AT MEALS WITHOUT VEGETABLES OR FRUITS AT THE SAME MEAL? Record frequency code. Codes: 0 = never 1 = <once a month 2 = a few times a month 3 = once a week 4 = 2-3 times a week 5 = most days 6 = d a i l y 49. DID YOU EAT HOME CANNED OR BOTTLED FOODS? Record code: 0 = no 1 = yes 9 = non response I f yes, WHAT TYPES OF FOODS WERE PREPARED IN THIS MANNER? 179 Subject no. 50. WHAT TYPES OF CHEESE DID YOU EAT MOST OFTEN? Record three most frequent c h o i c e s . 51. IF YOU HAD CHEESE WITH MOULD (NOT INTENDED AS IN BLUE OR ROQUEFORT), DID YOU : 1 = DISCARD THE CHEESE 2 = REMOVE THE MOULD AND EAT THE CHEESE 3 = EAT THE CHEESE AND THE MOULD 0 = do not eat cheese 9 = unknown, no response 52. HOW OFTEN DID YOU DRINK THE FOLLOWING BEVERAGES? Frequency: 0 = never 1 = once a month 2 = a few times a month 3 = once a week 4 = 2-3 times a week 5 = most days 6 = d a i l y 53. HOW FREQUENTLY DID YOU CONSUME THESE ALCOHOLIC BEVERAGES? Record frequency code as above Q#52. BEER WINE LIQUOR • MILK TEA COFFEE SODA POP DIET POP LIQUERS 180 Subject no. Food Hab i t s i n the Teens COULD YOU NOW TRY TO REMEMBER BACK TO WHEN YOU WERE A TEENAGER (AGE 13 - 19 APPROX.). I AM INTERESTED IN YOUR LAST YEARS AT HOME. THAT WOULD HAVE BEEN IN 19 TO 19 ( I n t e r v i e w e r to f i l l i n from p r e v i o u s i n f o r m a t i o n ) AND YOU WERE LIVING IN (Chose a r e l a t i v e l y s t a b l e p e r i o d r e p r e s e n t i n g the l a s t years at home). 54. WHEN DID YOU EAT YOUR LARGEST MEAL OF THE DAY? Record meal time code. Time codes: 1 = 0500 - 0859 2 0900 - 1059 3 1100 - 1359 4. = 1400 - 1659 5 = 1700 = 1959 6 = 2000 = 0459 55. DESCRIBE A TYPICAL EXAMPLE OF YOUR LARGEST DAILY MEAL? Note the q u a n t i t y and food category, eg. j u i c e i cup. 56. WHICH MEALS DID YOU REGULARLY SKIP, IF ANY? HOW OFTEN? Enter frequency code by a p p r o p r i a t e meal time. Frequency: Meal Times: 0 = never 0500 - 0859 1 = <once a month 0900 - 1059 2 a few times a month 1100 - 1359 3 = once a week 1400 - 1659 4 = 2-3 times a week 1700 - 1959 5 = most days 2000 - 0459 6 = d a i l y 181 Subject no. 57. AT THAT TIME, HOW OFTEN DID YOU EAT THE FOLLOWING FOODS? IN WHAT WAY WERE THEY PREPARED? Record frequency and p r e p a r a t i o n codes. How Frequency Prepared Frequency: Food L i s t : 1st 2nd 0 never a. 1 = <once a month 2 a few times a b\ month (<4) c. 3 = once a week d. 4 = 2-3 times a week e. 5 most days f . 6 = d a i l y g-h. How prepared: 0 = as purchased X « j • 1 baked 2 = steamed 3 = b o i l e d k. 4 = f r i e d 1. 5 = smoked m. 6 = c h a r c o a l g r i l l e d n. 7 = p i c k l e d o. 8 = othe r , r e c o r d P-9 = unknown, no q. response r . s. deep ye l l o w v e g e t a b l e s , eg. yams, c a r r o t s , w i n t e r squash b r o c c o l i , b r u s s e l sprouts c a u l i f l o w e r , cabbage l e a f y greens: l e t t u c e , spinach corn, c e l e r y , s t r i n g beans peanuts, peanut b u t t e r d r i e d b e a n s / p e a s / l e n t i l s f i s h or p o u l t r y , eg. ch i c k e n , turkey, e t c . beef: steak, r o a s t , hamburger cured meats: ham, s a l a m i , weiners, luncheon meat, sausage, bacon cheese eggs -c i t r u s f r u i t or j u i c e tomatoes, i n c l . j u i c e / s a u c e p a s t a : s p a g h e t t i , n o o d l e s , e t c . i c e cream, cream s a l a d dressing,mayonnaise whole g r a i n b r e a d / c e r e a l s r e f i n e d b r e a d s / c e r e a l s , eg. white bread,corn f l a k e s , e t c . WAS THIS REPRESENTATIVE OF YOUR EATING HABITS AT THIS TIME? I f not, WHY NOT? 182 Subject no. 58. AS A TEENAGER, DID YOU TRIM THE FAT FROM YOUR MEAT BEFORE EATING IT? Code: 0 = no 1 = yes 8 = d i d not eat meat 9 = unknown, no response 59. WHAT KIND OF BREAD DID YOU EAT MOST OFTEN? Types: 0 d i d not eat bread 1 = rye 2 = pumpernickel 3 = white, i n c l u d i n g French, I t a l i a n 4 = cracked wheat 5 = whole wheat 6 = ot h e r , s p e c i f y 9 = unknown, no response 60. WHAT SPREAD DID YOU USE ON YOUR BREAD MOST FREQUENTLY Record type code. Types: 0 = none 1 = b u t t e r 2 = margarine 3 = p o l y u n s a t u r a t e d margarine 4 = l a r d 5 = ot h e r , s p e c i f y 9 = unknown 61. DID YOU EVER EAT MEAT AT MEALS WITHOUT VEGETABLES OR FRUIT AT THE SAME MEAL? Record frequency code. Codes: 0 = never 1 = once a month 2 = a few times a month 3 = once a week 4 = 2-3 times a week 5 = most days 6 = d a i l y 62. DID YOU EAT HOME CANNED OR BOTTLED FOODS? Record code: 0 = no 1 = yes 9 = no response I f yes, WHAT TYPES OF FOOD WERE PREPARED IN THIS MANNER? 183 Subject no 63. WHAT TYPES OF CHEESE DID YOU EAT MOST OFTEN? Record three most frequent c h o i c e s . 64. IF YOU HAD CHEESE WITH MOULD (NOT INTENDED) DID YOU: 1 = DISCARD THE CHEESE 2 = REMOVE THE MOULD AND EAT THE CHEESE 3 = EAT THE CHEESE AND THE MOULD 0 = d i d not eat cheese 9 = unknown, no response 65. HOW OFTEN DID YOU CONSUME THE FOLLOWING BEVERAGES? Record frequency code. Frequency: 0 = never 1 =< once a month 2 = a few times a month 3 = once a week 4 = 2-3 times a week 5 = most days 6 = d a i l y MILK TEA COFFEE SODA POP DIET POP 66. HOW FREQUENTLY DID YOU CONSUME THESE ALCOHOLIC BEVERAGES? Record frequency code as above i n #65, BEER WINE LIQUOR LIQUERS Me d i c a l Information MY LAST SET OF QUESTIONS RELATE TO YOUR HEALTH, IN GENERAL. 67. AT ANY TIME IN YOUR LTFE WERE YOU OVERWEIGHT? * I f yes, FOR HOW LONG AND AT WHAT AGE? * Overweight = more than 20 l b s . over recommended 184 Subject no, 68. WHAT WAS YOUR WEIGHT ABOUT A YEAR AGO? Record i n pounds. 69. WHAT IS YOUR HEIGHT? Record i n in c h e s . 70. HAVE YOU EVER HAD ANY OF THE FOLLOWING CONDITIONS? I f yes, AT WHAT AGE WAS THIS DETECTED? Code: 0 = no HEART DISEASE 1 = yes 9 = no response, INTESTINAL POLYPS unknown CANCER AT WHAT SITE? ULCERATIVE COLITIS CROHN'S DISEASE APPENDICITIS HEMORRHOIDS VARICOSE VEINS DIVERTICULITIS PEPTIC ULCER wt ht. CHD IP CA yes/no age UC CD A H VV D PU If yes, PLEASE PROVIDE BRIEF DETAILS OF THESE DIAGNOSES AND ANY MODIFICATIONS YOU HAVE MADE IN YOUR FOOD HABITS BECAUSE OF THESE ILLNESSES. 185 Subject no, 71. WAS YOUR MOTHER OR FATHER EVER DIAGNOSED AS HAVING CANCER? Code: 0 = no No/Yes 1 = yes 9 = unknown, no response If yes, WHICH PARENT, WHAT SITE(S)? 72. HAVE ANY OF YOUR BROTHERS OR SISTERS (IF ANY), BEEN DIAGNOSED AS HAVING CANCER? Code: 0 = no 1 = yes 8 = no b r o t h e r s / s i s t e r s 9 = unknown, no response If yes, BROTHER OR SISTER? WHAT SITE(S)? 73. HAVE YOU EVER SMOKED? Code: 0 = no No/Yes 1 = yes 9 = unknown What If yes, WHAT? Code: 1 = c i g a r e t t e s 2 = c i g a r s 3 = pip e s FOR HOW LONG? Record number of years to one Long decimal. STARTING AT WHAT AGE? Record year. Age ARE YOU STILL SMOKING? Code: 0 = no S t i l l 1 = yes 9 = unknown, no response HOW MUCH? OOO I n t e r v i e w e r : See coded l i s t to determine i f f u r t h e r i n f o r m a t i o n r e q u i r e d . I f so, proceed. If not, thank the s u b j e c t f o r time and co - o p e r a t i o n . P l e a s e assure him/her of the v a lue of h i s / h e r a s s i s t a n c e . Record your impression of the s u b j e c t ' s a b i l i t y to remember past food h a b i t s and ease w i t h the i n t e r v i e w . Was a spouse or i n t e r p r e t e r present? 186 Subject no. Question to be asked i f coded l i s t i n d i c a t e s more i n f o r m a t i o n r e q u i r e d . TO FURTHER OUR STUDY, WE WOULD LIKE TO INTERVIEW MORE PEOPLE LIKE YOURSELF. WOULD YOU PLEASE PROVIDE THE NAMES OF A FEW OF YOUR FRIENDS WHO MAY BE WILLING TO ASSIST US IN OUR RESEARCH AS YOU HAVE DONE. WE WOULD LIKE THE NAMES OF PEOPLE OF SIMILAR AGE (± 5 YEARS), THE SAME SEX, AND, I F POSSIBLE, A SIMILAR BACKGROUND, I.E. LOCATION LIVED. OF COURSE, THEIR PARTICIPATION IN THE STUDY AND THE INFORMATION THEY GIVE US WILL BE KEPT CONFIDENTIAL. IF REQUESTED, WE WOULD HAVE TO IDENTIFY YOU AS THE SOURCE OF THEIR NAMES, AS ANOTHER PARTICIPANT IN THE RESEARCH STUDY. ONLY ONE OR TWO WILL BE CHOSEN. 1. 2. 3. 4. 5. Record f i v e , i f p o s s i b l e . Arrange f o r phone c a l l back i f c l i e n t wishes to check with f r i e n d s before g i v i n g names. Thank the s u b j e c t f o r time and c o - o p e r a t i o n . Data from R e g i s t r y Report to be Entered L a t e r : ,74. SITE OF TUMOR 75. CLASSIFICATION OF TUMOR 76. DATE OF DIAGNOSIS ocooo oocoo (XXX) 187 APPENDIX 2: FOOD PHOTOGRAPHS USED FOR PORTION SIZE DETERMINATIONS (black and white copies). 192 194 197 202 203 APPENDIX 3: CONTACT LETTERS 206 T H E U N I V E R S I T Y O F BRITISH C O L U M B I A Faculty of Medicine Department of Health Care and Epidemiology Mather Building 5804 Fairview Crescent Vancouver, B.C. V6T 1W5 Dear Dr. I am c o n d u c t i n g a st u d y t o dete r m i n e the r e l a t i o n s h i p o f s e l e c t e d e n v i r o n m e n t a l f a c t o r s e s p e c i a l l y d i e t a r y h a b i t s , t o c o l o n c a n c e r . I would l i k e t o r e q u e s t your p e r s m i s s i o n t o c o n t a c t the p a t i e n t s on the a t t a c h e d l i s t f o r whom you are i d e n t i f i e d as the t r e a t i n g p h y s i c i a n . The s t u d y has been approved by the Committee f o r A p p r o v a l o f S t u d i e s i n v o l v i n g Human S u b j e c t s a t the U n i v e r s i t y o f B r i t i s h Columbia. My t h e s i s chairman i s Dr. C . J . Mackenzie. My t h e s i s a d v i s o r y committee i n c l u d e s Dr. H. S t i c h (Cancer R e s e a r c h ) , Dr. G. H i s l o p (Cancer C o n t r o l ) , D r s. M. Lee and H. K u h n l e i n (Human N u t r i t i o n ) and Dr. W. Powrie (Food S c i e n c e ) . This s t u d y w i l l i n v o l v e a one t o two hour i n t e r v i e w on l i f e s t y l e h a b i t s . As you can see from the a t t a c h e d s h e e t , I would l i k e t o know the p a t i e n t ' s a d d r e s s and t e l e p h o n e number from your r e c o r d s . I w i l l be c a l l i n g y o u r o f f i c e t h e week o f A p r i l 15 t o o b t a i n t h i s i n f o r m a t i o n and t o i d e n t i f y i f t h e r e a r e any p a t i e n t s who s h o u l d n o t be c o n t a c t e d and t o answer any q u e s t i o n s r e g a r d i n g the study which you may have. Thank you f o r your c o o p e r a t i o n . Yours t r u l y , Pat Wolczuk, M.Sc. 2 0 7 T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A Faculty of Medicine Department of Health Care and Epidemiology Mather Building 5804 Fairview Crescent Vancouver. B.C. V6T 1W5 DOCTOR'S NAME: LOCATION-: P a t i e n t ' s Name P a t i e n t ' s A d d r e s s P a t i e n t ' s Phone No. ' P a t i e n t 1 s B i r t h Date Thank vou. P l e a s e f o r w a r d t h i s s h e e t i n the a t t a c h e d envelope 208 APPENDIX 4 RAW DATA 209 Summary of Selected Results: Comparison of Friends-Source versus Physician-Source Controls Variable Chi Square p-Value Urban Residence 2.45821 0.;4829 Level of education 1.85155 0.7630 0.82901 0.8425 Mother's Country of Origin 2.96498 0.5637 Major Occupation of origin 3.07082 0.6891 Heart Disease Polypsoccupation Smoking History Obesity History Current Foods: Vitamin A Rich Foods Fatrt disease Fibertinal polyps Vitamin C Rich Foods 0.00150 1.40807 4.11451 2.36527 0.9691 0.2354-0.2494 0.5001 0.81244 0.89070 12.74542 2.20797 0.6662 0.6406 0.2532 0.3315 Middle Years: Vitamin A Rich Foods Vitamin C Rich Foods Cruciferous Vegetables Fiber Rich Food 1.69744 0.4280 2.95866 0.2278 1.46024 0.4819 0.89357 0.6397 Teen Years: Vitamin A Rich Foods Vitamin C Rich Foods Fiber 0.97772 0.6133 0.17613 0.9157 0.34430 6.8419 210 C R O S S T A B U L A T I O N OF • • • BY GROUP - PAGE 1 OF 1 GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 I 2 1 I 91 I 73 I SS .5 I 44.5 I 82 .0 I 80.2 I 45 .0 I 36. 1 2 I e I 6 I 60 .0 1 40.0 I a . i J 6.6 1 4 .5 I 3.0 3 ] 7 I 4 I 63 .6 I 36.4 I 6 .3 I 4.4 I 3 S I 2.0 4 1 4 I 8 I 33 . 3 I 66. 7 I 3 .6 I 8.8 I 2 .0 I 4.0 CC-UMN 111 81 TOTAL 55 .0 45.0 ROW TOTAL 164 81 .2 Major Residence 202 100.0 1 OUT OF 8 ( 12.5X> OF THE VALID CELLS HAVE EXPECTED CELL FREQUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 4.955 RAW CHI SQUARE - 2.77412 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.4278 NUMBER OF MISSING OBSERVATIONS • C R 0 S S T A A T I O N 8» GROUP COUNT ROW PCT COL PCT I TOT PCT I 1 I 2 1 I 4 I 9 I 30.8 I 69.2 I 3.6 I 9.8 1 2.0 I 4.4 2 I 12 I 7 I 63.2 I 36.8 I 10.8 I 7.6 I 5.9 I 3.4 3 I 17 I 24 I 4 1.5 I 58.5 1 15.3 I 26. 1 I 8.4 I 11.8 4 I 78 I 52 1 60.0 1 40.0 I 70.3 I 56.5 I 38.4 I 25.6 COLUMN TOTAL RAW CHI SQUARE • 111 54.7 92 45.3 7.S2S09 WITH ROW TOTAL 13 6.4 Urban Residence 130 64.0 203 100.0 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.0476 211 C R 0 S S T A I 0 N GROUP GROUP COUNT 1 ROW PCT leases COL PCT I TOT PCT I 1 grl-7 poste«c{2 poatsec2-4 post sec) 4 COLUMN TOTAL 1 tOO.O 0.9 O.S 23 44.2 20.9 11.4 59 O 0.0 0.0 0.0 29 55. B 31.5 14.4 45 ROW TOTAL 5S.7 I 43 .3 53.£ I 48 .9 29.3 I 23 .3 11 I 6 S4.7 I 35 .3 10.0 I 6 .5 5.4 I 3 .0 9 6 60.0 I 40 .0 8.2 I 6. .5 4.5 I 3. .0 7 I 6 53.8 I 46.3 6.4 I 6.5 3.5 I 3.0 no 92 54.5 45.5 E d u c a t i o n 52 25.7 104 51.5 17 8.4 IS 7.4 13 6.4 202 100.0 2 OUT OF 12 ( 16.7X> OF THE VALID CELLS HAVE EXPECTED CELL FREQUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.455 RAW CHI SQUARE • 4.15345 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.5275 NUMBER OF MISSING OBSERVATIONS C R 0 S S T A U L A T I 0 N BV GROUP PAGE 1 OF 1 GROUP COUNT I ROW PCT leases COL PCT I TOT PCT bath Jewish COLUMN TOTAL ROW TOTAL 1 I 2 39 I 23 62.9 I 37. 1 36.4 I 30.7 31.4 I 12.6 54 I 43 55.7 I 44.3 50.5 I 57.3 29.7 I 33.6 13 j 5 73.2 I 37.8 12. 1 I 6.7 7. 1 I 2.7 1 I 4 20.0 I SO.O 0.9 I 5.3 0.5 1 2.2 107 75 58.8 41.2 F a i t h 18 9.9 182 100.0 2 OUT OF 8 ( 25.07.) OF THE VALID CELLS HAVE EXPECTED CELL FREQUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREQUENCY • 2.O60 RAW CHI SQUARE • 5.26850 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.1532 NUMBER OF MISSING OBSERVATIONS • 31 212 c i o s s n A T BY I O N GROUP COUNT ROW PCT COt. PCT TOT PCT 1 leases 1 I 1 COLUMN TOTAL 9 I 11 4S.0 1 5S.0 8.1 I 13. 1 4.6 1 5.6 22 I 12 84. 7 I 35.3 19.8 I 14.3 11.3 1 6.3 46 1 22 67.6 I 32.4 41.4 I 26.2 23.6 I 11.3 27 I 35 43.S I 56.5 24.3 I 41.7 13.8 I 17.9 T I 4 63.6 I 36.4 6.3 I 4.8 3.6 I 2. 1 111 56.8 84 43. 1 PAGE 1 OF ROW TOTAL 20 10.3 34 17.4 68 34.9 62 31.8 Mother's Country of O r i g i n 195 100.0 1 OUT OF 10 ( 10.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 4.738 RAW CHI SQUARE • 9.91380 WITH 4 DEGREES OF FREEDOM. SIGNIFICANCE • 0.0419 NUMOER OF MISSING OBSERVATIONS • C R O S S T A A T BY COUNT I ROW PCT leases COL PCT I COLUMN TOTAL 3 I 5 37.5 I 62.5 2.7 I 5.7 1.5 I 2.5 25 I 15 62.5 I 37.5 22.5 I 17.2 12.6 I 7.6 49 I 30 62.0 I 38.0 44 . 1 I 34.5 24.7 1 15.2 30 I 34 46.9 I 53. 1 27.0 I 39. 1 15.2 I 17.2 4 I 3 57. 1 I 42.9 3.6 I 3.4 2.0 I I.S 1 1 1 87 56. I 43.9 PAGE 1 OF ROW TOTAL 40 20.2 79 39.9 7 3.5 F a t h e r ' s Country of O r i g i n 188 100.0 4 OUT OF lO ( 40.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREQUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREQUENCY • 3.076 RAW CHI SQUARE • 5.12873 WITH 4 OEGREES OF FREEDOM. SIGNIFICANCE • 0.2743 NUMBER OF MISSING OBSERVATIONS -213 C R O S S T A B U L A I 0 N GROUP COUNT ROW PCT COL PCT TOT PCT eonp2 leases I I I I t I I 1 I I I I I I I 1 I 1 ROW TOTAL COLUMN TOTAL 1 I 3 7 I 3 70.0 I 30.0 6.3 1 3.3 3.4 I 1.5 68 I 60 53. 1 1 46.9 61.3 I 65.2 33.5 I 29.6 7 I 5 58.3 1 41.7 . 6.3 I 5.4 3.4 I 2.5 33 I 19 53.7 I 46.3 19.8 I 20.7 10.8 9.4 2 1 3 5O.0 I 50.0 1.8 I 2.2 1 .0 1 1 .0 5 1 3 62.5 I 37.5 4.5 I 3.3 3.5 I 1.5 111 92 54.7 45.3 10 4.9 128 63.1 12 5.9 41 20.2 4 2.0 8 3.9 M a r i t a l Status 203 100.0 5 OUT OF 12 ( 41.7%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 1.813 RAW CHI SQUARE • 1.38666 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.9358 C R O S S T A B U L A T I 0 N GROUP PAGE 1 OF 1 COUNT I ROW PCT I COL PCT I TOT PCT I controls 1 I 17 I 24 I 4 1 5 I 58 5 I 18.3 I 30 4 I 9.9 I 14 .0 2 I 26 I 23 I 53. 1 I 46.9 I 28 .0 I 29. 1 I 15. 1 I 13 4 3 1 2 1 I 20 I 51.2 I 48.8 COLUMN TOTAL RAW CHI SQUARE • 22.6 12.2 2S.3 11.6 29 I 12 70.7 1  29.3 31.2 ! 1 15.2 16.9 I 1 7.0 93 79 . 54 . 1 45.9 7.36120 WITH ROW TOTAL Major Occupation 172 lOO.O 3 OEGREES OF FREEDOM. SIGNIFICANCE • 0.0612 NUMBER OF MISSING OBSERVATIONS 2 J. 4 C R O S S T A B U L COUNT ROW PCT leases control a COL PCT TOT PCT 1 1 2 0 I BS I 7S 1 S3. a I 46 . 1 I BO 2 I 82.6 1 43 8 1 37 .4 1 s I 3 1 73.0 1 25.0 I S 4 I 2.2 1 3 0 I 1.0 2 2 I 1 1 SS 7 I 33.3 I 1 8 I 11 ! t 0 I 0.3 4 1 I 3 1 3S 0 I 75.0 I 0 8 I 3.3 I 0 5 I 1.3 S 3 I 1 I EG 7' 1 33.3 I 1 8 I 1.1 t 1 0 I O.S 6 8 t 2 1 7S 0 I 25.0 I S 4 I 2.2 t 3 0 I 1.0 7 3 I 3 I 40 0 I 60.0 I 1 8 I 3.3 1 1 0 I 1.3 • 1 I 0 J lOO 0 I O.O I O 8 1 O.O I 0 5 I O.O a 2 I 4 I 33 3 I 66.7 1 ' 1 8 I 4.3 1 1 0 I 2.0 COLUMN 111 82 TOTAL 54 .7 45.3 ROW TOTAL 165 81.3 8 3.8 1 1.3 3 1.3 6 3.8 3 2.3 I O.S 6 3.0 203 100.0 GROUP COUNT I ROW PCT leases controls ROW COL PCT I TOTAL TOT PCT I I I 2 1 I I 1 O > '83 I 76 1 158 I 53.2 I 47.8 I 78.3 I 74.8 I 82.6 I 1 40.8 I 37.4 I -I I I I I 3 1 3 1 6 I 50.0 I 50.0 I 3.0 I 2.7 I 3.3 I 1 1.5 1 t.S I -I J 2 2 1 3 1 0 1 3 1 100.0 I 0.0 I 1.3 I 3.7 I 0.0 I I t.S 1 0.0 I -I I I 3 1 S I 3 1 7 1 71.4 I 28.6 I 3.4 I 4.5 I 3.2 I I 2.3 I 1.0 I -I I 1 4 1 3 1 S I 8 I 37.5 I 62.S I 3.9 I 2.7 I 5.4 I I 1.S 1 2.5 I -I — I — I 5 1 2 1 0 1 2 1 100.0 I 0.0 I 1.0 I 1.8 I 0.0 t I 1.0 I O.O I -I .--I I 6 1 6 1 3 1 8 I 75.0 I 25.0 I 3.9 I 5.4 I 2.2 I --- I— 2^-1—- ' .9-1 7 1 I I 0 1 1 -I 100.0 I 0.0 I 0.5 I O.S I O.O I I 0.5 1 0.0 I -I I ---I • I 2 1 0 1 2 1 100.0 I 0.0 I I.O I 1.8 I 0.0 I I 1.0 I 0.0 I -I T , , 9 1 3 1 4 1 7 ' I 42.9 I 57.1 I 3.4 I 2.7 I 4.3 I I 1.5 1 2.0 I - I I I COLUMN 111 82 203 TOTAL 54.7 45.3 100.O F a t h e r ' s H i s t o r y of Cancer Mother's H i s t o r y of Cancer 16 OUT OF 18 < 88.37.) Of THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.453 RAW CHI SOUARE - 6.83916 WITH 8 OEGREES OF FREEOOM. SIGNIFICANCE • 0.5541 18 OUT OF 20 ( 9O.07.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY - 0.453 RAW CHI SOUARE - 10.SSO84 WITH 9 OEGREES OF FREEOOM. SIGNIFICANCE - 0.3078 215 c » o > i t < « ii i n i o « O F SISCA GROUP COUNT I ROW PCT Icssas COL PCT 1 TOT PCT I -I 1 ! I I I -I 2 I I I I -I 4 I I I I -I 8 I I -I 6 I I I I -1 T I 1 I I -1 a i ROW TOTAL [ 1 I 2 83 I 71 S3. 9 I 46. 1 74.8 i 77.2 40. S l 33.0 8 I 2 80.0 I 20.0 7.2 l 2.2 3.a I t.O [ 1 1 1 50.0 I 50.0 0.9 I 1.1 0.5 1 O.S 1 1 4 20.0 I ao.o 0.9 I 4.3 0.3 I 2.0 3 I 3 50.0 I 50.0 2.7 I 3.3 1.5 I 1.3 3 I 2 60.0 I 40.0 2.7 I 2.2 1.3 I 1.0 4 I 3 44.4 1 55.6 3.6 1 3.4 2.0 t 2.3 2 I 3 40.0 I 6O.0 1.8 I 3.3 1.0 I 1.3 6 I i as .7 : 14.3 3. .4 i 1. 1 1 3. O i O.S ] COLUMN TOTAL 111 54.7 10 4.9 2 1.0 3 2.5 6 3.0 3 2.5 3 2.5 7 3.4 92 45.3 303 100.0 CROUP COUNT I ROW PCT Icaiai COL PCT I TOT PCT I O I I -I 3 1 5 I I 55.6 I I 4.5 1 1 2.5 I I.a 1.0 I I -I-I 0.9 O.S 0.0 0.0 2.7 I 1.5 I COLUMN TOTAL 111 54.7 92 45.3 ROW TOTAL 1 l 3 87 I 80 52. t I 47.9 78.4 I 87.0 42.a I 39.4 4 l 2 66.7 I 33.3 3.6 I 2.2 2.0 I 1.0 3 I 0 100.0 t 0.0 i .a l 0.0 1.0 I 0.0 44.4 I 4.3 I 2.0 I I I 3 1 0 100.0 I 0.0 2.7 1 0.0 1.3 I o.o 1 I 0 100.0 I 0.0 0.9 1 0.0 O.S I o.o I -1 I 1 I I -I I 1 66.7 I 33.3 I 1.1 I 0.5 I -I O I I 100.0 I 0.0 I I 5O.0 I 50.0 I 3.3 I 1.5 I j I I 3 I I 60.0 I 40.0 I I 2.7 I 2.2 1 I t.S I .0 I 1 6 3.0 2 1.0 » 4.4 3 1.5 1 0.5 3 t.S < 3.0 3 3.3 303 100.0 15 OUT OF 18 ( 83.3X1 OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTED CELL FREOUENCY • 0.906 RAW CHI SQUARE " 8.71562 WITH 8 DEGREES OF FREEOOM. SIGNIFICANCE • 0.3669 SISCA 18 OUT OF 30 ( 90.0X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.4S3 RAW CHI SQUARE • 6.88653 WITH 9 DEGREES OF FREEOOM. SIGNIFICANCE • 0.648s B r o t h e r ' s H i s t o r y of Cancer S i s t e r ' s H i s t o r y of Cancer 216 C R 0 S S T A B U 1 A T at [ 0 N GBOUP PACE I OF CROUP COUNT 1 ROW PCT lease* COL PCT t TOT PCT 1 1 CVD 1 0 1 I I I -I 1 I I I I - I COLUMN TOTAL CORRECTED CHI SOUARE • RAW CHI SOUARE • ' ROW TOTAL oo 1 58 1 139 67 .S 1 43.4 1 68.5 72. 1 1 84. 1 I 39.4 I 39. 1 1 31 1 33 1 64 4a.4 ] 51 .8 1 31. S 37.a I 35.8 1 IB.3 1 16.3 I 111 83 303 64.7 45.3 100.0 H i s t o r y of Heart Disease 1.13485 WITH 1 DECREE OF FREEDOM. 1.46973 WITH 1 OEGREE OF FREEOOM. SIGNIFICANCE • SIGNIFICANCE • O.3B89 0.3354 C R O S S T A U L A T 1 0 N er GROUP COUNT ROW PCT COL'PCT leasas 1 ROW TOTAL POL VPS TOT PCT I 1 I 3 I 0 | 73 1 85 1 158 1 46.3 1 53. a 1 81.4 I 70.9 1 S3. 4 I 1 37.6 I 43.8 1 1 1 30 I C I 36 1 83.3 I 16.7 I 18.6 1 29. 1 I C.« I 1 15.5 I 3. 1 t I- -I COLUMN 103 91 194 TOTAL 53. 1 46.9 IOO.O H i s t o r y of Polyps CORRECTED CHI RAW CHI SOUARE SQUARE 14.77432 WITH 16.33125 WITH NUMBER OF MISSING OBSERVATIONS DEGREE OF FREEOOM. DEGREE OF FREEDOM. SIGNIFICANCE • 0.OOO1 SIGNIFICANCE • O.OOOt C R O S S T A B U L A I O N GROUP PACE 1 OF COUNT ROW PCT COL PCT TOT PCT leases I I 1 I 103 I SO I S3.4 I 46.6 I 94.3 I 87.8 I St.2 I 44.8 -I ] I S I I 73.0 1 I 5.3 I I 3.0 I -I 1 COLUMN TOTAL I -1 I I I I 1 3 I 25.0 I 2.2 I 1.0 I 1 109 S4.2 92 45.8 ROW TOTAL 183 96.0 H i s t o r y of U l c e r a t i v e C o l i t i s 301 IOO.O 3 OUT OF 4 ( 30.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCV LESS THAN S O MINIMUM EXPECTEO CELL FREOUENCY • 3.663 CORRECTED CHI SOUARE • 0.70780 WITH 1 OEGREI OF FREEDOM. SIGNIFICANCE • O 4002 RAW CHI SOUARE " 1.44819 WITH 1 DEGREE OF FREEOOM. SIGNIFICANCE - 0.2288 NUMBER OF MISSIMG OBSERVATIONS - 3 217 C R O S S T A B U L A T > 0 N CROUP PACE 1 OF COUNT I ROW PCT Icaaa CDL PCT I TOT PCT I 110 84.7 88. t 84.2 1 so.o o.a O.B COLUMN TOTAL 111 54.7 45.3 88.8 44.8 I -I 1 1 I I -I 1 I 50.0 I I.I I o.e i - i 81 45.3 ROW TOTAL 301 89.0 a 1.0 301 IOO.O H i s t o r y of Crohn's Disease 3 OUT OF 4 ( 50. OX) OF THE VALID CELLS HAVE EXPECTEO CELL- FREQUENCY LESS THAN 8 O MINIMUM EXPECTED CELL FREOUENCY - 0.906 CORRECTED CHI SOUARE - O.O WITH 1 OECREE OF FREEDOM. SIGNIFICANCE - 1 . OOOO RAW CHI SOUARE - 0.01785 WITH 1 OEGREE OF FREEDOM. SIGNIFICANCE • 0 8937 C R O S S T A B U L A I 0 N GROUP PAGE 1 OF GROUP ' COUNT I ROW PCT Icaaaa control! ROW COL PCT I TOTAL TOT PCT I I I 3 1 1 J 1 0 1 7 1 1 S3 I 133 I 53.4 I 46.S I 65.8 I 64.5 I 67.4 I I 35.1 I 30.T I _1 1 1 I I 39 I 30 I 69 I 56.5 I 43.5 I 34.3 I 35.5 I 33.6 1 I 19.1 I 14.9 I -1 1 — I COLUMN 110 93 303 TOTAL 54.S 45.5 IOO.O H i s t o r y of A p p e n d i c i t i s CORRECTED CH! SOUARE RAW CHI SOUARE 0.07606 WITH O.18041 WITH OEGREE OF FREEDOM. OEGREE OF FREEOOM. SIGNIFICANCE SIGNIFICANCE 0.7837 0.6710 NUMBER OF MISSING OBSERVATIONS " C R O S S T A B U L A T I 0 N GROUP PAGE 1 OF GROUP COUNT I ROW PCT Icaaaa control a ROW COL PCT 1 TOTAL TOT PCT I I I 3 1 1 1 1 0 1 61 I 56 I 117 I 53.1 I 47.9 I 57.6 I 55.0 1 GO.9 I I 30.0 I 37.6 I -1 1 1 I I SO I 36 1 86 I 58.1 I 41.8 I 43.4 I '45.0 I 39.1 I I 34.6 I 17.7 I -I 1 1 COLUMN 111 93 303 TOTAL 54.7 45.3 IOO.O H i s t o r y of Hemorrhoids CORRECTED CHI SOUARE -RAW CHI SQUARE • O.49885 WITH 1 OECREE OF FREEDOM. SIGNIFICANCE • O.48O0 0.72073 WITH 1 DEGREE OF FREEDOM. SIGNIFICANCE • 0.3959 218 C R 0 S S T A B U L A T 1 O N BY CROUP CROUP COUNT I ROW PCT Icaitl controls ROW COL PCT I TOTAL TOT PCT I 1 1 1 1 — - I - -I --I o i as I 72 I 1(1 1 B9.2 I 44.7 I 7».7 I ao.a i 78.3 I I 44.1 I 39.< I -I I 1 I I 21 I 20 I 41 I SI.2 I 48.8 I 30.3 1 19.1 I 31.7 I I 10.4 I 9.9 I -I -I " — I COLUMN 110 93 303 TOTAL 54.S 45.S 100.0 H i s t o r y of V a r i c o s e Veins CORRECTED CHI RAW CHI SOUARE • SQUARE • O.08433 WITH 1 DECREE Or FREEOOM. O.31718 WITH 1 DECREE OF FREEDOM. SIGNIFICANCE SIGNIFICANCE 0.7715 0.64 12 hLIMBER OF MISSING OBSERVATIONS • C R O S S T A B U L A T ION GROUP COUNT ROW PCT COL PCT Tcss«B I ROW TOTAL TOT PCT I 1 I 3 I O I 97 I a s I 180 I 53.9 1 4 6 . 1 I 9 0 . 0 1 8 9 . a I 9 0 . 2 I I 4 8 .S I 4 1 . 5 I - 1 -1 I 11 I 9 I - 2 0 I 5 5 . 0 I 4 5 . 0 I 1 0 . 0 I 10.3 1 a . a I I S.S I 4.S I -COLUMN i o a 93 2 0 0 TOTAL 5 4 . 0 4 6 . 0 1 0 0 . 0 H i s t o r y of D i v e r t i c u l i t i s CORRECTED CHI SOUARE • RAW CHI SOUARE - O.O WITH 0.00895 WITH NUMBER OF MISSING OBSERVATIONS • DEGREE OF FREEOOM. DEGREE OF FREEDOM. SIGNIFICANCE SIGNIFICANCE 1 .0000 0.9346 C R O S S T A 8 U L A T BT I O N GROUP PAGE I OF 1 COUNT I ROW PCT Icssas COL PCT I ROW TOTAL TOT PCT I 1 I 2 I • I - - I O I 9 8 t 7 3 t 171 1 3 7 . 3 I 4 2 . 7 I a s . i I 8 9 . 1 1 8 0 . 2 I 1 4 8 . a 1 3 6 . 3 t 1 I 12 •1 18 I 3 0 I 4 0 . 0 -I G O O I 1 4 . 9 I I O . 9 1 1 9 . a I I C O I 9 . 0 I -C O L U M N n o 9 1 301 T O T A L 5 4 . 7 4 6 . 3 1 0 0 . 0 H i s t o r y of P e p t i c U l c e r CORRECTED CHI RAW CHI SOUARE SQUARE 3.42742 WITH 3.08653 WITH OEGREE OF FREEOOM. DEGREE OF FREEOOM. SIGNIFICANCE • O.1193 SIGNIFICANCE • 0.0789 NUMBER OF MISSING OBSERVATIONS -219 CROUP COUNT I BOW PCT I O K I COL PCT I T01 PCT I 1 C R O S S T A B ROW TOTAL U L A T I 0 N BT CROUP PAGE 1 OF o 1 . 31 I 33 I 63 never 1 48.2 l so.a 1 31.7 1 3B.4 I 35.6 1 I IS.6 I 16. 1 1 1 1 S I a I 7 1 ignt I 71.4 I 3a. 6 I 3.6 1 4.6 I a.2 I 1 3.B I 1.0 I 2 I 3 I 4 1 7 estaol lahed I 43.1 I B7. 1 I 3.S 1 3.a I 4.4 1 I 1.S I 3.0 I 3 1 70 I S3 1 133 long tera : 57.4 I 42.6 1 61.3 i '64.3 I 57.8 t i 35.3 I 36. 1 I COLUMN 109 90 199 TOTAL 64. a 4S.3 IOO.O Smoking H i s t o r y 4 OUT OF 8 ( SO.OX) OF THE V A L I D C E L L S HAVE EXPECTED C E L L FREOUENCT L E S S THAN 5.0. MINIMUM E X P E C T E D C E L L FREOUENCT - 3.166 RAW CHI SQUARE . 2.30714 WITH 3 DEGREES OF FREEOOM. S I G N I F I C A N C E • 0.5113 NUMBER OF M I S S I N G O B S E R V A T I O N S C R 0 S S T A 8 U L A T I 0 N BT GROUP PAGE I OF COUNT ROW PCT COL P CT TOT PCT anort Ura establIsned long tera> COLUMN TOTAL leases 1 I I •I 1 I I I I I I I I I I I 1 I I I I I I 1-lO 66.7 9.0 S.O 37 61.7 33.3 18.5 111 ss.s s 33.3 S.6 3.5 3 3 3 8 . 3 3 5 . 8 1 1 . 6 8 9 44.5 RAW C H I SOUARE ' 7 . 7 0 4 5 9 WITH MJMBER OF M I S S I N G O B S E R V A T I O N S • ROW TOTAL -I-SI 1 8 7 4 7 . 2 1 5 2 . 8 4 5 . 9 I 6 4 . 0 2 5 . 6 I -I-28.S 13 I 4 7 6 . S I 23.S 11.7 I 4.6 6.S I 3.0 lOB 54.0 17 8.5 15 T.S 60 30.0 O b e s i t y H i s t o r y 20O IOO.O 3 DEGREES OF FREEOOM. S I G N I F I C A N C E • O.OS25 C R 0 S S T A A T I O N BT GROUP PAGE 1 OF COUNT ROW PCT COL P CT leases I ROW TOTAL TOT PCT I t 1 a I 0 1 5 2 | 4 3 I 9 5 1 5 4 . 7 1 4 6 . 3 1 4 8 . 0 I 4 9 . 1 1 4 6 . 7 I I 2 6 . 3 1 2 1 . 7 I 1 1 2 5 I 1 3 I 3 8 I 6 5 . 8 I 3 4 . a I 18.3 I 3 3 . 6 I 14. 1 1 I 12.8 I 6.6 I a I 7 I 6 I 12 I 5 8 . 3 I 4 1 . 7 I 6. 1 I 6.S I 5.4 I 1 3.6 I a . 6 I a I 17 1 17 I 3 4 I 5 0 . 0 I so.o 1 17.2 I 1 6 . 0 I 1 8 . 5 I I 8.6 I 8.6 I - r 6 I 5 1 14 i 19 1 3 6 . 3 1 7 3 . 7 i 9.6 .1 4.7 I i s . a i I 3.5 I T. 1 i COLUMN 106 9 3 198 TOTAL S 3 . 5 4 6 . 6 IOO.O S t r e s s i n the Middle Years RAW C H I SOUARE 8.29013 WITH 4 DEGREES OF FREEDOM. S I G N I F I C A N C E NUMBER OF M I S S I N G O B S E R V A T I O N S - 5 220 ) C R O S S ! * U I A I ION BT GROUP COItNl ROW rel COL PCT 1 TOTAL TOT PCT 1 1 1 2 1 0 1 IOO 1 76 1 176 1 56.8 1 43.2 I 86.7 I SO. 1 1 82.6 I I 49.3 1 37.4 I 1 1 4 1  a ! 13 1 33.3 1 1 66.7 I 5.9 I 3.6 I 1 8.7 | 1 2.0 1 1 3.9 I •I-2 I 3 1 3 1 6 1 50.0 1 1 SO.O I 3.0 1 2.7 1 1 3.3 I 1 1.3 1 1 1.3 1 3 1 2 1 I 3 I 4 I 50.0 1 50.0 I 2.0 I 1.8 1 2.2 1 1 1 .0 1 1 .O 1 4 1 2 I 3 1 5 1 40.0 1 60.0 1 2.5 I t.B I 3.3 1 1 1.0 1 1.5 I - -I COLUMN 1 1 1 92 203 TOT AL 54.7 45 . 3 100.0 Animal Products 6 OUT OF lO ( SO Oil OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCY • 1.813 RAW CHI SOUARE • 3.05449 WITH 4 OEGREES OF FREEOOM. SIGNIFICANCE • 0.54B7 C R O S S T A B U L A T I 0 N GROUP PAGE 1 OF COUNT ROW PCT COL PCT I TOTAL TOT PCT I 1 I 3 I -1 0 I 32 1 33 I IOS I 49.5 1 50.5 I 51.7 1 46.8 1 57.6 I I 25.6 I 26. 1 I -1 -1 1 I B 1 1 1 I 19 1 42. 1 I 57 .9 I 9.4 I 7.2 1 12.0 I 1 3.9 1 5.4 I 2 1 8 I Q I 17 I 47 1 1 S3 .9 1 8 4 I 7.2 1 9 8 1 • 3.9 I 4 . 4 1 3 I 5 I 2 I 7 1 71.4 I 28. 6 I 3.4 I 4.5 I 2.2 I 1 2.5 1 1 .0 4 I 38 I 17 I ss 1 69. 1 I 30.9 1 27. 1 1 34.2 1 18.5 I 18.7 1 8.4 COLUMN 111 92 303 TOTAL 54.7 45.3 IOO.O OF lO < 20.OX) OF THE VALIO CELLS A g r i c u l t u r a l Chemicals MINIMUM EXPECTEO CELL FREOUENCY • RAW CHI SOUARE - 8.13890 WITH 4 OEGREES OF FREEOOM. SIGNIFICANCE • O.OB66 221 C R O S S T A B U L A T 1 0 N GROUP PAGE 1 OF 1 COUNT ROW PCT COL PCT TOT PCT lenses I I I COLUMN TOIAL 0.9 0.3 111 54 .7 I. I 0.5 92 5.3 ROW TOTAL 1 54 .5 1 45 .3 91 .9 1 93 . 4 50 .3 • 4 1 .9 3 1 2 £0 .0 1 40 .0 2 .7 I 2 . 2 1 .5 1 1 .0 5 1 3 63 .5 1 37 .5 4 .3 I 3 .3 3. .5 I 1 , .50 1 1 O. 0 1 ICO. O O. o I 1. 1 O. 0 1 o. 5 1 1 1 SO. 0 I 50. 0 187 93. I 5 2.3 P a i n t s and V a r n i s h e s 203 1O0:O 8 OUT OF 10 ( BO.OX) OF THE VALIO CELLS HAVE EXPEC1E0 CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY - 0.453 RAW CHI SOUARE • 1.480O9 WITH 4 DEGREES OF FREEDOM. SIGNIFICANCE • O 8302 C R O S S T A B U L A T ION GROUP PAGE I OF C0UN1 I ROW PCT leases controls COL PCT I TOTAL TOT PCT I 1 I 3 O I 77 I 72 I 149 I SI . 7 1 48.3 1 73.8 I 69 . 4 1 79. 1 I I 38 . 1 1 35.6 I | I 6 I 6 1 13 I SO .0 1 50.0 I 5.9 I 5 . 4 1 6.6 1 I 3 O 1 3.0 I 3 I 7 1 5 I 13 I 58 . 3 I 4 1.7 1 5.9 1 6 .3 I 5.5 I I 3 .5 1 2.5 I 4 I 21 1 8 I 29 I 73 .4 1 37 .6 I 14.4 I IB. 9 I 8.8 I 1 IO. 4 I 4.0 I COLUMN 11 1 1 91 ' 202 TOTAL 55. 0 45.0 100.0 P l a s t i c s RAW CHI SOUARE 4.39156 WITH 3 DEGREES OF FREEOOM. SIGNIFICANCE NUMBER Of HISSING OBSERVATIONS 222 GROUP COUNT I ROW PCT lease* COt PCT I TOT PCT I I 0 I 1 I -I 1 I I I I -1 2 I COLUMN TOTAL C R O S S T A 8 U L A T ROW TOTAL S3. 8 1 46.2 90. 1 1 93.5 49.3 1 42.4 5 I 3 71.4 I 38 .6 4.S I 2.2 2.S t 1.0 2 1 2 50.0 1 50.0 1.8 I 2.2 1.0 I 1.0 4 1 3 £6.7 1 33. 3 3.6 1 3.2 2.0 1 1.0 1 1 1 92 54.7 45.3 188 916 D y e s t u f f and Bleaches 4 2.0 6 3.0 303 100.0 6 OUI OF 8 I 75.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY > 1 813 PAW CHI SOUARE • 1.23867 WITH 3 DEGREES OF FREEOOM. SIGNIFICANCE • O 7437 . . . . . . . . . . . . . . . . C R O S S T A B U L A T ION OF EX01L BY GROUP" GROUP COUNT I ROW PCT leases COL PCT TOT PCT I I 1 I 3 I TOTAL O I 90 I 73 I 163 1 5S.7 1 44 . a I BO. 3 1 91. I 1 79.3 1 1 44.3 1 36.0 I 1 I 13 ; 10 1 22 I 54.5 1 45.5 1 10.8 I 10.9 I 10.9 1 I 5.9 I 4.9 1 3 I 8 1 7 1 15 I S3. 3 1 46.7 I 7 . 4 I 7.2 I 7.6 I I 3.9 I 3 4 1 3 I 1 I 0 I 1 I 100.0 I 0.0 I 0.5 I 0.9 I O.O I I O.S I O.O I 4 I 0 I 2 I 3 1 o.o 1 lOO.O 1 1.0 1 o.o I 2.2 I 1 0.0 1 1.0 I -COLUMN 111 92 203 TOIAL 54.7 45.3 lOO.O C u t t i n g O i l or Machine O i l 4 OUT OF lO ( 40.071) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5 O MINIMUM EXPECTED CELL FREOUENCY . 0.453 RAW CHI SOUARE • 3.27182 WITH 4 DEGREES OF FREEOOM. SIGNIFICANCE • 0.5134 223 C R O S S T A B U L A T l o N BY GDOUP PAGE I OF I GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I I COLUMN TOTAL control 1 s 82 1 73 52. 9 1 47 . 1 73.9 1 79.3 AO. a 1 36.0 t o 1 t o 50.0 I 50.0 9.0 I 10.9 A .9 1 4.9 13 1 6 6S.7 1 33.3 10.8 1 6.5 5.9 1 3.0 0 [ 1 0.0 I 100.0 0.0 I 1.1 0.0 I 0.5 7 [ 3 77.8 I 22.2 6.3 I 2.2 3.4 1 1.0 111 92 54.7 45.3 ROW TOIAL 155 76.4 20 9.9 18 8.9 1 0.5 9 4.4 Petroleum Products 203 IOO.O 4 OUT OF 10 ( 40.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.453 RAW CHI SOUARE " 4.56200 WITH 4 OEGREES OF FREEDOM. SIGNIFICANCE • 0.33S3 C R 0 S S T A 8 U L A T I 0 N GROUP PAGE 1 OF COUNT ROW PCT COL PCT TOT PCT leases I I 1 controls 83 51.6 74 .8 40.9 5 41.7 4.5 2.5 58 7 3 7.6 3.4 3 IOO.O 3.7 1.5 O.O 0.0 0.0 o O.O O.O O.O ROW TOTAL 161 79.3 12 5.9 Cooking O i l or Fat 20 I 6 I I 36 76.9 I 23. 1 I 12.9 18.0 1 6.5 I 9.9 1 3.0 I 11 1 92 303 54.7 45.3 100.0 COLUMN TOTAL 4 OUT OF lO (.40.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0 MINIMUM EXPECTEO CELL FREOUENCY • 0.453 HAW CHI SOUARE • 10.33933 WITH 4 OEGREES OF FREEOOM. SIGNIFICANCE • 0.0351 224 C P 0 S S T 1 I 0 N CROUP P»GE I OF COUNT I ROW PCT Ic COL PCT I TOT PCT I 1-O I COLUMN TOTAL 80.0 3.6 3.0 18 64.3 16.4 8.9 1 10 54.5 30.0 1. 1 0.5 10 3S.7 10.9 5.0 93 45.5 ROW TOTAL 78 47.9 77 .3 1 84 .8 43. , 1 I 39 .6 3 I 3 SO 0 1 SO 0 3. 7 1 3. 3 1 . 5 1 1 . .5163 80.7 S 3.S 38 13.9 Coal Tar Products 302 t o o O 4 OUt OF 8 ( SO.07.) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 3.377 RAW CHI SOUARE • 3.80464 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.433T NUMBER OF MISSING OBSERVATIONS • C R O S S T A B U L A T BY GROUP PAGE 1 OF GROUP COUNT I ROW PCT leases COL PCT I TOT PCT 1 0 I I I I 1 I I ROW TOTAL COLUMN TOIAL 1 I 3 96 I 83 53.6 I 46.4 87.3 I 90.3 47.5 I 4 1.1 7 I 3 70.0 I 30.0 6 4 I 3.3 3.5 1 1.5 3 I 3 . SO O I SCO 2.7 I 3 3 1 .5 I 1 . 5 1 1 50.0 1 1 50.0 0.9 1 1 . 1 0.5 I OS 3 I 3 60.0 1 40.0 3.7 I 3.3 1.5 I 1.0 1 i o 92 54.5 45. 5 179 88.6 10 SO 6 3.0 2 1.0 5 3.5 Asbes to s 302 lOO.O 7 OUT OF IO ( 70.07.1 OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5. MINIMUM EXPECTEO CELL FREOUENCY - 0.911 RAW CHI SOUARE - 1.14929 WITH 4 OEGREES OF FREEDOM. SIGNIFICANCE • 0.8864 NUMBER Or MISSING OBSERVATIONS • 1 225 c u s s u • T B r I 0 N GROUP PAGE 1 OF 1 ROW PCT leases controls ROW COL PCT 1 w TOTAL TOT PCT 1 1 1 2' I 0 I 70 I 6 1 134 1 32. 7 i 47.8 1 66.3 I 63.6 1 69.6 1 I 34.7 1 31.7 1 1 I 16 1 9 I 25 1 64 .0 I 36 .0 I 12.4 1 14 .5 I 9.8 I I 7.9 I 4.5 I 2 I 16 I 13 I 78 1 57. 1 I 47.9 1 13.9 1 14 .5 I 13.0 1 1 7.9 I 5.9 I 3 I 3 1 3 | 5 I 40.0 1 60.0 1 3.5 1 1 . B 1 3.3 I I 1.0 I 1.5 1 4 I 6 I 4 1 10 I 60.0 1 40.0 I 5.0 I S.5 I 4.3 I I 3.0 1 3.0 I I- 1 COLUMN 1 10 93 203 TOTAL 54.5 43.5 IOO.O Dust From Wood or Gr a i n Products 3 OUT OF 10 ( 30.0X1 OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5. MINIMUM EXPECTED CELL FREOUENCY • 2.277 RAW CHI SOUARE • 1.81050 WITH 4 DEGREES OF FREEOOM. SIGNIFICANCE • 0.7706 NUMBER OF MISSING OBSERVATIONS C R O S S T A B U L A T COUNT ROW PCT COL PCT TOT PCT leases I 1 1 controls I 0 N GROUP PAGE 1 OF ROW TOTAL 0 I 85 I 78 I 163 1 '52. 1 I 47 .9 I 80. 3 I 76 6 I 84 .8 I 1 4 1.9 I 38 . 4 1 1 I 12 I 8 I 30 I 60.0 I 40 O 1 .9.9 I 10.8 1 8 . 7 I I 5 . 9 I 3 .9 I 2 1 10 I 5 I 15 1 66 . 7 1 33 .3 I 7 4 1 9 0 I 5 . 4 1 I 4.9 I 3 .5 I -I- -I 3 I 1 I 1 1 2 I 50.0 I SO .0 I 1 .O 1 0.9 I 1 . 1 I I O.S ' 0 .s I 4 I 3 1 0 1 3 1 IOO.O 1 0 0 1 1.5 1 2 7 1 0. 0 I t.S 1 0. 0 1 COLUMN 1 1 1 93 203 TOTAL 54.7 45. 3 IOO.O Fumes From Organic S o l v e n t s 4 OUT OF 10 < 40.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 3.0. MINIMUM EXPECTED CELL FREOUENCY • 0.906 RAW CHI SOUARE " 4.03421 WITH 4 DEGREES OF FREEDOM SIGNIFICANCE • 0.4027 226 C R O S S T A B U L A T I O N Of BY GROUP • • • • • . • • . • . • • • • • • • • • • • » • • • . . . . p 4 0 c , o r GROUP COUNT I ROW PCT leases controls ROW COL PCT 1 TOTAL TOT PCT I I I I I EXHOSP 1 I I 0 I 101 1 SS I IS6 I 54.3 I 45.7 I 91.6 _ , . ., . T T • "1 i 91.0 i 92.4 i Chemicals i n H o s p i t a l s I 49.8 I 41.9 I -I I I I I S I 3 1 9 I SG.7 I 33.3 I 4.4 I 5.4 I 3.3 I I 3.0 I 1.S I -I I I 2 1 2 1 I I 3 I SS.7 I 33.3 1 IS I 1.8 I 1.1 I I I.O I O.S I -1 I I 3 1 0 1 I I I I 0.0 I 100.0 I 0.5 I 0.0 I 1.1 I I 0.0 I O.S 1 -I I I 4 1 2 1 2 1 4 I 50.0 1 5O.0 I 2.0 I 1.8 I 2.2 I 1 1.0 I 1.0 I -I 1 --I COLUMN 11 I 92 203 TOTAL . 54.7 45.3 IOO.O 8 OUT OF 10 ( 80.OX) Or THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.453 RAW CHI SOUARE • 1.94842 WITH 4 OEGREES OF FREEOOM. SIGNIFICANCE • 0.7452 C R O S S T A B U L A T I O N GROUP PAGE 1 OF COUNT ROW PCT COL PCT leases I controls ROW TOTAL TOT PCT 1 1 1 3 1 O 1 94 1 83 I 176 I 53 4 1 46.6 1 86.7 1 84 7 1 89. 1 1 I 46. 3 I 40.4 I 1 1 7 I 4 I 1 1 I 63 6 1 36 . 4 1 5 4 I 6.3 I 4 . 3 1 • 3.4. I 2.0 1 2 I 3 I 3 I 6 I 50.0 1 50.0 1 3 0 I 2.7 I 3.3 I I 1.5 I IS 3 1 1 I . 0 I 1 I IOO.O 1 O.O I O.S I 0.9 1 O.O I I O.S 1 O.O 1 4 I 6 1 3 I 9 I 66.7 1 33.3 1 .4.4 I S.4 1 3.3 I 1 3.0 1 1.5 1 -I- •I COLUMN 111 92 303 TOTAL 54.7 45.3 IOO.O Other Chemicals 7 OUT OF 10 ( 7O.0X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.453 RAW CHI SOUARE • 1.87446 WITH 4 OEGREES OF FREEOOM. SIGNIFICANCE • 0.7588 227 C R O S S 1 A B U L A T I O H BY CROUP CROUP COUNT I ROW PCT ICM.I COL PCT I TOT PCT I t COLUMN TOTAL 17 e a . o IS.3 *.« 33.0 (.0 4.0 41 t 31 SS.9 1 43. 1 1 36.9 1 .34.1 ; 20.B t 1S.S 1 S3 51. 3 47.7 76.5 111 55.5 I SO I 49.5 I 56.3 I 35.0 -I I B9 44.5 PACC 1 OF ROW TOTAL 35 13.3 73 36.0 103 31.3 Consuming Three Meals a Day 300 t o o . o RAW CHI SOUARE - 3.37439 WITH NUMBER OF MISSING OBSERVATIONS • 3 DECREES OF FREEDOM. SIGNIFICANCE • 0.3139 C R 0 S S T A A T I 0 N BY GROUP PAGE 1 OF GROUP COUNT I ROW PCT leases COL PCT 1 TOT PCT I I 71 31.9 64.0 35.0 66 4S.3 71.7 33.3 13 I 3 75.0 I 35 .0 13.3 I 3. .4 7.4 I 3 .3 ( 1 9 47. 1 I S3. .9 7.3 I 9. .1 3.9 1 4. .4 ROW TOTAL 137 67.8 30 9.9 17 9.4 Consuming Large Q u a n t i t i e s of Vitamins -I 15 I i o 1 33 SO.O I 40.0 I 13.3 13.3 1 10. 9 I 7.4 1 4.S 1 3 I 3 I 4 SO.O I s o . o I 3.0 t.e ! 3.3 I 1 o I 1.0 I i n 93 303 54. 7 43.3 ICO 0 COLUMN TOTAL 3 OUT OF IO ( 20.0<> OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5. MINIMUM EXPECTEO CELL FREOUENCY • 1.913 RAW CHI SOUARE - 4.50242 WITH 4 DEGREES OF FREEOOM. SIGNIFICANCE • 0.3433 C R O S S T A A T I 0 N BY GROUP PAGE 1 OF GROUP COUNT I ROW PCT leases controls ROW COL PCT I TOTAL TOT PCT 1 I I 2 1 1 1 1 O I J l I 34 I 35 t 56.4 1 43.C I 29.3 I 30.4 1 27.9 I I 16.3 I 12.9 I - | 1 1 I I 71 I C3 I 133 I 33.4 I 46.6 I 70.7 I 69.6 I 73.1 I I 37.8 I 33.0 I -I I — 1 COLUMN 103 86 188 TOTAL 34.3 45.7 lOO.O Trimming Fat From Meat (Current) CORRECTED CHI SOUARE • RAW CHI SOUARE • 0.04505 WITH 0.13924 WITH DEGREE OF FREEOOM. DEGREE OF FREEDOM. SIGNIFICANCE SIGNIFICANCE 0.8319 O.7090 NUMBER OF MISSING OBSERVATIONS - 15 228 C R O S S T A B U L A T I 0 N CROUP COUNT ROW PCT COL PCT TOT PCT Icaaaa I I 1 (CONTINUED) COLUMN TOTAL I 72 I 01.1 I 64.B 1 ss. a I 19 82.C 17. 1 a . s B 33.7 4.3 3.3 3 66.7 i.a 1.0 111 35.3 I 33.1 1.1 0.5 1 1 3 33.3 I 66.7 O.S I 3.2 0.6 1 1.0 C 1 4 60.0 I 40.0 6.4 1 4.4 l.O I 2.0 6 I 1 as.7 1 14.3 5.4 I 1 . 1 3.0 I 0.5 SO 4.8 PACE I OF ROW TOTAL CROUP COUNT 1 ROW PCT Icaaaa COL PCT 1 TOT PCT I 1 1 control* 67 48.2 74.4 33.1 4 I 17.4 I 4.4 I 1.0 I — I a 64.1 1O.0 4.3 138 69.2 23 11.4 14 7.0 1 1.6 J 1.6 lO 5.0 7 3.6 201 100.0 COLUMN TOTAL I o I 0.0 I O.O I ' O.O -I l i t 33.3 3 I I 2 I 100.0 I 2.2 I 1.0 I I SO 44.8 ROW TOTAL 3 1.0 201 IOO.O S OUT OF 16 ( 56.3X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 3.0. MINIMUM EXPECTED CELL FREOUENCY - 0.896 SAW rm SOUARE - 13.72098 WITH 7 DEGREES OF FREEOOM. SIGNIFICANCE - 0.O276 NUMBER OF MISSING OBSERVATIONS • 2 S k i p p i n g Meals (Current) C R O S S T A B U L A T I O N BY GROUP AGE 1 OF 1 COUNT ROW PCT COL PCT TOT PCT 1 ease's I controls 9 I I I I -I I I 1 I I I I I I I -I-I I I I - I -I I I 1 - I -I 1 I I - I -I I I I - I -[ 1 I 3 1 73 I 6S [ 51.4 I 48.6 65.8 I 75.0 36.0 I 34.0 21 I to 67.7 I 33.3 IS.a I 10.S 10.3 I 4.S 8 I 5 61.6 I 38 .5 7.2 I S.4 3.8 I 2.5 6 1 3 O.O 1 IOO.O O.O I 2.2 O.O I l.O 1 1 1 SO.O 1 SO.O O.S I 1.1 O.S 1 O.S COLUMN TOTAL 2 SO.O i.a t . o I 2 I SO.O I 3.2 I l.O 6 66.7 S.4 3.0 3 33.3 3.3 1.5 111 64.7 S3 43.3 SOW TOTAL 142 70.0 31 IS.3 13 6.4 2 1.0 2 1.0 S k i p p i n g Meals (Middle Years) 8 4.4 203 IOO.O 8 OUT OF 14 ( 37.11) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 3.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.SO6 RAW CHI SOUARE • S.B8228 WITH 6 OEGREES OF FREEDOM. SIGNIFICANCE • O 4352 229 C RO i S 1 I I UL I I D N CROUP PAGE I OF CROUP COUNT I ROW PCT k i w i COL PCT ! TOT PCT I 1 ROW TOTAL COLUMN TOTAL 1 13 I 76 [ 51.3 1 46.a I 74.5 1 o4 a 1 40.C I sa. c 13 I a 05.7 1 14.3 10.S 1 3.3 5.S I 1.0 4 t 5 44.4 I 55.6 3.C I 6.4 3.0 1 3.S 1 I 3 33.3 I 66.7 O.S I 3.2 O.B ! 1.0 1 1 0 100.0 1 0.0 0.9 1 0.0 O.S I 0.0 S t 3 62.6 I 37.3 4.S I 3.3 3.S 1 1.6 3 1 1 75.0 > 35.0 3.7 1 1.1 1.5 I 0.5 MO 54.5 92 43.5 160 79.2 14 6.9 8 4.6 3 1.3 1 O.S a 4.0 303 lOO.O IMLSK COUNT ROW PCT COL PCT TOT PCT CROUP I ICASaS I I 1 1 contnolo ROW TOTAL COLUMN TOTAL I 3 I 66.7 I i.a I 1.0 - I 110 64.6 3 1.8 93 45.6 302 100.0 12 OUT OF 16 ( 75.OX)_0F THE VALIO CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 6.0. MINIMUM EXPECTED CELL FREOUENCY - 0.455 «A* CHI SOUARE • S.98804 WITH 7 DEGREES OF FREEOOM. SIGNIFICANCE • 0.2535 NUMBER OF MISSING OBSERVATIONS • 1 S k i p p i n g Meals (Teen Years) (CONTINUED) C R O S S T A B U L A T ION GROUP GROUP COUNT 1 ROW PCT leases COL PCT 1 TOT PCT I 1 COLUMN TOTAL CORRECTED CHI RAW CHI 47 56.6 42.3 33.2 64 53.3 57.7 31.5 111 S4.7 43.3 ROW TOTAL 3 I 1 36 t 43.4 t 39. 1 I 17.7 I 96 I 46.7 I 60.9 I 37.6 I 1 92 93 40.9 130 59. 1 203 1O0.0 Adding Sugar to Beverages/Cereals SOUARE SOUARE O.10239 WITH 1 DECREE OF FREEOOM. O.21472 WITH 1 DECREE OF FREEDOM. SIGNIFICANCE • SIGNIFICANCE • 0.T490 0.6431 230 C R O S S T A B U L A I 0 N GROUP PACE 1 OF GROUP COUNT 1 BOW PCT Icaasa controla BOW COL PCT t TOTAL TOT PCT I I I 2 1 I I I 0 1 21 I 28 I SI I AS.I I 54.8 I 2S.1 I 20.7 I 30.4 I I 11.3 I 13.8 I -I I I I I 88 I 64 I 1S2 I S7.9 I 42.1 I 74.9 I 78.3 I 68.6 I I 43.3 I 11.1 1 -I I I COLUMN 111 92 201 TOTAL 54.7 45.3 100.0 Regular Consumption of J a m s / J e l l i e s CORRECTED CHI SOUARE RAW CHI SOUARE 3.03348 WITH 1 DEGREE OF FREEOOM. 3.52345 WITH 1 OEGREE OF FREEOOM. SIGNIFICANCE SIGNIFICANCE O.1338 O.1122 C R 0 S S T A GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 COLUMN TOTAL > I I I I -I-I I I I I -I-I I I -I-I I I I - I -26 56.5 23.4 13.8 48 61.1 44. I 34.4 27 42.2 24.3 13.4 81.8 8. I 4.S 1 11 S5.2 20 I 43.5 I 22.2 I lO.O I 1 31 I 38.6 I 34.4 I 15.4 I 1 37 I 57.8 41. 1 18.4 I I I 1 3 I 18.2 3.3 l.O SO 44.8 B U L A T I O N BT CROUP PAGE I OF ROW TOTAL 80 39.8 64 31.8 301 IOO.O Handling of L e f t o v e r Vegetables 1 OUT OF 8 I 12.5X1 OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCT • 4.S2S RAW CHI SOUARE • 8.73115 WITH 3 OEGREES OF FREEOOM. SIGNIFICANCE • 0.O338 NUMBER OF MISSING OBSERVATIONS • 3 231 run freq FILE NONAME (CREATION DATE • 09/25/13) 09/25/83 C R O S S T A B U L A T BT I 0 N GROUP GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 ROW TOTAL Onea a wo dally S day COLUMN TOTAL 6.S 3.6 t3 32.2 11.2 6.3 22 AT.B 20.6 It. 3 I I I I -I 1 I I I -I I I I I -I-1 1 I I -I 1 tS I BO.O I 13.0 I a.3 I I c I 46.2 1 3.6 1 3.1 I 107 SS.7 a 33.3 9.4 4.2 I -I I I I I •I 11 I 47.8 I 12.9 I 3.7 I I 24 52 .2 28.2 12.3 44 38.7 41.1 22.* I I I I 1 31 I I I 41.3 I I 36.5 I I 16.1 I •1 -I I 4 I I 20.0 I 1 4.7 1 I 2.11 •1 -I 1 7 I I 33.8 I 1 a.3 1 I 3.6 I I 1 as 44.1 RAW CHI SOUARE • 7.30238 WITH NUM8ER OF MISSING OBSERVATIONS • 15 7.8 23 12.0 46 24.0 73 39. I 30 10.4 13 6.a 192 100.0 Yellow Vegetables (Current) 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.1991 C R 0 S S T A A T BY I 0 N GROUP PAGE 1 OF COUNT RCi PCT COL PCT TOT PCT lcaaas 1 once a MO few a MO once a wfc s day COLUMN TOTAL 85 34.3 ROW TOTAL 7 I 9 43.8 I 56.3 T.4 1 11.3 4.0 I 3. 1 18 I 17 53.8 I 47.3 20.0 1 21.3 10.8 I 9.7 32 I 38 S3.3 I 46.7 33.7 I 35.0 IB. 3 I 16.0 33 I 34 97.8 1 43. 1 34.7 1 30.0 18.9 I 13.7 4 { 2 66.7 I 33.3 4.2 I 2.3 2.3 I 1. 1 80 43.7 36 20.6 60 34.3 57 32.6 B r o c c o l i / B r u s s e l ' s S p r o u t s ( C u r r e n t ) 175 IOO.O 3 OUT OF lO ( 20.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 3.0. MINIMUM EXPECTEO CELL FREOUENCY • 2.743 RAW CHI SOUARE - 1.44036 WITH 4 DEGREES OF FREEDOM. SIGNIFICANCE - 0.8372 NUMBER OF MISSING OBSERVATIONS * 38 232 T A 8 U l A T I 0 N sr GROUP PAGE I OF I COUNT I ROW PCT I c i i . l COL PCT 1 ROW TOTAL ones a MO amity TOT PCT I 1 I 2 1 1 t 10 I • t 19 > MO 1 92.8 1 47.4 I 10.7 1 lO.l I 11.1 I 1 5.8 I 8. 1 I -!• 3 1 32 1 38 1 SO •o 1 53.3 I 4S.7 I 33.7 I 33.0 I 34.8 I 1 18.0 I 15.7 | -I- -1 1 I 38 1 33 I GO •« 1 S3.3 1 38.7 I 33.7 ! 39.2 1 37.2 I 1 21.3 I 12.4 I -I- -1 4 1 13 I 22 I 35 1 37. 1 I S2.8 t 19.7 I 13.4 1 27.2 I I 7.3 1 12.4 1 -I-5 I 3 I O I 3 day I IOO.O I 0.0 I 1.7 1 3. 1 I 0.0 I 1 1.7 1 O.O 1 -1-C I 1 1 0 I 1 I IOO.O I 0.0 1 O.C I l.O I 0.0 I I O.S 1 0.0 I -I-COLUMN 97 81 178 TOTAL 54.8 45.5 IOO.O Cauliflower/Cabbage (Current) 4 OUT OF 12 ( 33.3*) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 3.0. MINIMUM EXPECTEO CELL FREOUENCT • 0.455 RAW CHI SOUARE - 9.53911 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.O894 NUMBER OF MISSING OBSERVATIONS • 25 C R O S S T A B U L A T I O N BT GROUP COUNT ROW PCT COL PCT TOT PCT I Icasaa 1 I 1 ROW TOTAL Illy 3 day COLUMN TOTAL 3 33.3 3.0 1.8 S SS.T C . 8 3.3 36 69.0 36.0 ' 19. I 19 43.9 18.0 S.6 23 34.8 33.0 12.3 IOO 53.2 I •I I I 1 I I I I I I 1 I I I I I I 25 1 1 41.0 1 I 28.4 I I 13.1 I •I 1 1 7 I 10 [ 41.3 I 58.6 7.0 I 11.4 3.7 • 8.3 13 1 3 73.2 I 37.6 13.0 I 5.7 6.8 ; 2.7 23 66. 1 26. 1 13.2 I I I 1 1 19 I 45.2 I 21.6 I lO. 1 I 1 96 46.6 8 4.6 17 1 8.0 18 9.6 61 32.4 Leafy Green Vegetables (Current) 42 22.3 188 IOO.O 3 OUT OF 12 ( 16.7X1 OF THE VAL10 CELLS HAVE EXPECT EO CELL FREOUENCT LESS THAN 3.0. MINIMUM EXPECTEO CELL FREOUENCT • 4.213 RAW CHI SOUARE • 7.32315 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.1877 NUMBER OF MISSING OBSERVATIONS • 15 233 • • • • • • • c • o i i i i • u i i i I o n O F CFVR. 8V CROUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . p40t , of , CROUP COUNT I ROW PCT lcasas controls ROW COL PCT I TOTAL TOT PCT I t ! 2 1 1 I ( 1 5 1 11 One* a AO I 54 .5 I 45.S 1 6.1 1 3.C I 9 .6 I I 3 . t *I 3 .1 1 -I- -1 a 1 IB 1 19 I 34 f •« m mo I 44. t 1 55.9 I 17.3 I 14.0 I 21.3 ! I 7.7 I 9.7 I 3 I 30 I 38 I 58 once • wk I St.T 1 48.3 I 29. S I 3S.O I 31.5 I 1 15.3 1 14.3 1 -I- -I 4 I 41 1 28 I 69 wk 2-2 1 SS.4 I 40.1 I 35.3 I 38.3 I 31.S I I 20. B t 14.3 I -I- -I- -I S 1 9 1 < X 15 »OSt day I 60.0 I 40.0 1 7.7 I a.4 I 6.7 1 I 4.6 I 3. 1 I -I- -I- -I < I « I 3 t 9 Sally I 66.7 1 33.3 I 4 .1 I 5.6 I 3.4 t I 3.1 I 1.5 I -I- -I- -I COLUMN 107 89 196 TOTAL 54. C 45.4 100.0 Fibr'ous V egetables (Current) 3 OUT OF 13 ( 25.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCT • 4.087 RAW CHI SOUARE • 3.05241 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE - 0.6919 NUMBER OF MISSING OBSERVATIONS • 7 C R O S S T A B U L A I 0 N CROUP PAGE 1 OF CFPNT onci a MO COUNT ROW PCT COL PCT TOT PCT I 1 I I I I I -I I I I I -I I Icaaas IROW TOTAL ones a wk S Oay COLUMN TOTAL 1 I 3 f 20 I 14 I 34 S9.8 I 41.3 I 36.0 35.6 I 36.4 I 15.3 I 10.7 I -1 13 1 S 7 18 68.4 I 31.1 1 14.5 16.7 t 11.3 I 9.9 t 4.6 I 14 1 11 I 35 56.0 I 44.0 I 19. 1 17.9 I 20.1 I 10.7 I 9.4 I -1 13 I to 1 33 56. S I 43.5 I 17.6 16.7 I 18.9 1 9 .1 I 7.6 I t l 94.C 14. I 1.4 7 41.3 9.0 ft.I 78 69. S 2 IS.4 3.8 1.5 to 68.9 18.9 7.6 53 40.S Peanuts/Peanut B u t t e r (Current) 13 a . a 17 13.0 131 too.o RAW CHI SOUARE • 6.61934 WITH 9 DECREES OF FREEOOM. SIGNIFICANCE - 0.35OS NUMBER OF MISSING OBSERVATIONS • 73 234 C R 0 S S T A 9 U L A T I 0 N C r D R » B* OOOUP PAOE I OF I GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 ROW TOTAL 1 1 31 I 21 I 52 • • no I 59.6 1 40.4 I 53. 1 I CO.B 1 44.7 1 1 31.6 I 21.4 I 2 I S I 11 I 20 X 1 I 45.0 I 55.0 I 20.4 1 17.6 1 23.4 I 1 B.3 I 11.2 I 3 1 a I 13 I 21 1 • wk I 38. 1 I 61.8 I 21.4 I 15.7 I 27.7 1 I 8.2 I 13.3 I 4 1 3 I 2 I 5 -3 ! 60.0 1 40.0 I S.I I 5.8 I 4.3 I I 3.1 I 2.0 I COLUMN 81 47 98 TOTAL 52.0 48.0 100.0 D r i e d Beans/Peas (Current) M r N i i ^ J v S ^ r . n 2 ! W > ° F ™ E V * L I D C E U S H A V t EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 2.398 RAW CHI SOUARE • 3.3S588 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.3399 NUMBER OF MISSING OBSERVATIONS 109 C R O S S T A B U L A 1 O N GROUP PAGE 1 OF COUNT ROW PCT COL PCT TOT PCT leases I I I one* a ao one. • wk dally 5 day COLUMN TOTAL 111 54.7 ROW TOTAL [ G I 3 66.7 I 33.3 5.4 I 3.3 3.0 I I.S 13 I IS 46.4 I 53.6 11.7 I 16.3 6.4 I 7.4 21 I 33 38.9 I 61. 1 19.9 1 35.9 10.3 I 16.3 62 I 30 67.4 I 32.6 55.9 I 32.6 30.5 1 14.9 7 I 10 41.2 I 58. a 6.3 I 10.9 3.4 1 4.9 2 I 1 66.7 I 33.3 1.8 I 1.1 1.0 I 0.9 92 43.3 9 4.4 28 13.8 S4 26.6 92 4S.3 17 a.4 3 1.5 F i s h / P o u l t r y (Current) 303 100.0 4 OUT OF 12 ( 33.37.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY - 1.360 RAW CHI SOUARE - 14.14832 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE - 0.0147 23: C R O S S T A B U L CROUP COUNT I ROW PCT Icaees' COL PCT I TOT PCT I I one* a MO one* a wfc a 37.9 3.1 t.S I S I S3.S 1 B.« I l.S I ao i so . c I 18.7 i to.a i I 18 t 43.7 I 13.0 t B.t I 0 N CROUP PAGE t or ROW TOTAL I I I 1 1 I I 13 I 38.4 I 14.4 I C.S I 1 18 1 54.3 I a i . i I >.S I * 4.1 33 16.8 35 17.8 Beef (Current) -I-4 I Bl I 4a I S3 wk a-3 1 54.8 I 4S.a I 47.2 I 47.7 I 46.7 2 I 35.8 I 21.3 I -1- -I 5 I IS l 8 I 23 •K>«t day I 65.a l 34.8 S 11.7 I 14.0 I 8.8 I I 7.6 I 4.1 I -I-6 1 a I a 1 9 dally I 40.0 I 6O.0 I 2.S I t . f I a.a I 1 l.O I l.S I -1 -COLUMN 107 SO 1S7 TOTAL 54. a 43.7 100.0 4 OUT or ia ( 33.3X) or THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL EREOUENCY - 3.384 RAW CHI SOUARE • 4.00622 WITH 5 DEGREES Of EHEEDOM. SICN1MCAHCE - 0.5485 NUMBER Or MISSING OBSERVATIONS • * C R O S S T A B U L A T I O N GROUP GROUP COUNT 1 ROW PCT Icassa COL PCT I TOT PCT I 1 one* a MO f t . once a wk ally I I I I -1 1 I I I -I I I I I -I -I S I day 1 I I - I -c I I I 1 COLUMN TOTAL 20 55.6 20.2 1 1.7 I 16 I 44.4 I 22.3 I 8.4 26 55.3 26.3 15.3 I 21 I 44 .7 I 28.3 I 12.3 S 83. a 9.1 a.s a 6O.0 3.0 1.8 89 37.8 ROW TOTAL 38 I 18 ! 60. S 1 39. I 28.3 I 3S.0 16.4 1 10.9 17 1 14 54.8 I 43.a 17.3 1 18.4 9.9 I a.a 1 1 I 16.7 I 1.4 1 I 0.6 1 -I 1 1 3 1 1 40.0 I I a.a i I i.a i - l 1 72 42.1 36 21 . 1 47 27.S 46 26.8 31 18. I Cured Meats (Current) 8 2.8 171 100.0 4 OUT Of 12 ( 33.3X) OF THE VALID CELLS HAVE EXPECTED CELL fREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 3.105 RAW CHI SQUARE - 3.09636 WITH 9 OEGREES OF FREEDOM. SIGNIFICANCE • 0.8337 NUMBER OF MISSING OBSERVATIONS • 32 236 c i a s s i i i u i i T I O N BT CROUP PAGE 1 Of I one* a MO CROUP COUNT I ROW PCT Icaaas COL PCT I TOT PCT I I I 1 D I 62.6 I 4.7 i a . » i < I co.o 5.7 3. I ROW TOTAL ones A wfc dally I 1 -1 3 I I t I -I-4 I 1 I 1 -I-5 I day I I 3 37.5 3.3 1.B 13 41.9 13.3 S.C 43 59.7 40. C 31.9 31 SO.O 19.9 10.7 COLUMN TOTAL I -I t 19 I 54. S I 17.0 I 9.3 -I IOC 54.1 4 40.0 4.4 3.0 I I I I I I 19 I I 59.1 I I 30.0 I I 9.3 1 I 1 I 39 I I 40.3 I 33.3 1 14.9 1 I I I 1 21 I 1 SO.O I 23.3 I 10.7 I I I I I IS I I 45.5 I I 1C.7 1 I 7.7 I I 1 SO 45.9 9 4.1 10 5.1 31 15.9 72 36.7 42 21.4 33 16.9 196 lOO.O Cheese (Current) 3 OUT OF 12 { 25.OS) Of THE VALID CELLS HAVE EXPECTEO CELL fREOUENCT LESS THAN S.O. MINIMUM EXPECT EO CELL fREOUENCT • 3.673 RAW CHI SOUARE • 3.41BOS WITH 5 OECREES Of FREEOOM. SIGNIFICANCE • 0.6356 NUMBER Of MISSING OBSERVATIONS • C R O S S T A B U COUNT ROW PCT COL PCT GROUP I Icaaaa I one* a ao PCT I 1 « 3 1 t 2 I 3 I 40.0 f 6O.0 I 1.9 I 3.4 I 1 .O I 1.5 2 I 7 I C I 53.8 I 46.3 I 6.6 I 6.7 I 3.6 I 3. 1 3 1 19 I 14 ! 57.6 1 42.4 I 17.S 1 15.7 I 9.7 I 7.2 s day S7 55.3 53.9 29.2 46 44.7 SI.7 23.C dally 45.0 S.S 4.6 SS.O 12.4 S.S 13 S7. I 11.3 C.2 S 43.9 lO. 1 4.9 COLUMN TOTAL IOC 54.4 99 4S.6 L A T I 0 N BY CROUP PAGE 1 OF ROW TOTAL 5 3.6 13 6.7 33 16.9 103 93.9 20 10.3 21 10.8 Eggs (Current) 195 100.0 3 OUT Of 13 ( 16.7X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCY - 3.282 RAW CHI SOUARE • 1.36616 WITH 5 DECREES OF FREEDOM. SIGNIFICANCE - 0.9280 NUMBER OF MISSING OBSERVATIONS • 9 237 C R 0 S S T A 8 U L A T I O M 8T CROUP CROUP COUNT I ROW PCT !<:•••• COL PCT I TOT PCT I 1 1-dalty 1 one* « mo one* m wk I -I-3 I I I I -I-4 I I I I -I-5 I day I I COLUMN TOTAL I 33.1 1.1 O.C t IS.7 1. I O.S a 37. S 3.3 1.8 11 S5.0 11.a S.4 14 SO.O IS. I a.3-I -I I S3 I S9.4 I S7.T I 36. a -I 93 54.4 a GS.7 3.6 1.2 3 83.3 i 6.4 I 3.9 1 1 I S I 63.5 1 6.4 I a.9 I 9 I 45.0 I 11.5 I S.3 t • t -I I 14 I SO.O I • 17.9 I 8.2 I I I I •I I 43 I 1 40.6 I I 53.1 I I 35.1 I I 1 78 45.6 PACE 1 OF 1 ROW TOTAL 3 1 .a s 3.3 a 4.7 30 II.7 3a 16.4 IOC 62.0 171 100.0 C i t r u s F r u i t s / J u i c e (Current) 6 OUT OF 12 ( 50.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCY • 1.368 RAW CHI SOUARE - C.20554 WITH . 5 DECREES OF FREEOOM. SIGNIFICANCE • 0.2867 NUMBER OF MISSING OBSERVATIONS • C R O S S T A B U L A T I 0 N BY GROUP GROUP COUNT 1 ROW PCT I C I t A I COL PCT I TOT PCT I 1 I l l y 9 d a y COLUMN TOTAL 3 27.3 3.0 1.6 89.2 B.O 4.8 I I I I -I I I I I -I I 33 I 63.9 I 33.0 I 13.4 -I 1 31 I 5*.4 1 31.0 I IS.7 -I I 18 1 31.4 I 19.0 I S.7 -I I IS I 44.4 I 16.0 I a . s -I 100 53.a 8 73.7 9.3 4.3 30.8 4.7 2.2 I 13 I 36. 1 I IS. 1 I T.O 1 34 I 43.6 I 27.9 I 12.9 1 17 I 48.S I 18.a I 8.1 ao 55.6 33.3 10.a a s 46.2 PACE 1 OF I ROW TOTAL 11 3.9 13 T.O 36 19.4 55 39.6 Tomatoes/Tomato J u i c e (Current) 36 19.4 196 100.0 RAW CHI SOUARE 7.32324 WITH S DEGREES OF FREEDOM. SIGNIFICANCE • 0.1976 NUMBER OF MISSING OBSERVATIONS - 17 238 C I 0 S 1 1 A B U L A T I O N BT GROUP GBJOUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT ! t one a a mo 1 I I I I -I a i i i i - i -a i i i i t - i -4 I I I I - I -9 I Oay I I I - I -COLUMN TOTAL 2f 4B. I 30. 6 16.6 35 97.4 41.3 33.3 IS 91.4 21.3 11.9 9 83.3 9.8 3.2 1 IOO.O 1.2 O.S 89 94. 1 1 1S.T 1.4 O.S I I I I •I O I 0.0 I O.O I O.O I -I 72 49.8 PAQE 1 OF ROW TOTAL 2 f 1 28 I 91 .8 I 38-. S I 17.8 I I 38 I 42.S 3S. i 16.8 ! I I — I 17 I 48.6 I 23.6 I 10.8 I 54 34.4 41 38.8 35 22.3 6 3.8 1 0.« 157 IOO.O Pasta (Current) 4 OUT OF to ( 40.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 9.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.43B RAW CHI SOUARE • 4.0485O WITH 4 OEGREES OF FREEOOM. SIGNIFICANCE • 0.3995 NUMBER OF MISSING OBSERVATIONS C R O S S T A B U L A T I 0 N BY GROUP PAGE 1 OF COUNT ROW PCT COL PCT TOT PCT I Icaaaa I I 1 once a wfc bally 9 oay COLUMN TOTAL 20 64.9 21.7 13.3 I I I I -I I 17 I 68.0 I IB. 5 I 11.2 -I I 15 I 57.7 6.3 S.S a 33.0 13.3 5.3 11 18.3 7.2 24 60.0 26. 1 15.8 16 40.0 26.7 10.8 7.6 4.S 41.7 8.3 3.3 82 SO.5 SO 39.9 SOW TOTAL 1 I 1 11 I 39.9 I 18.3 7.2 1 9 I 8 so o 1 SO.O 1 8, .8 I ts.o ». .8 I 5.8 31 20.4 25 16.4 26 17. 1 Ice Cream (Current) 40 26.3 18 11.8 1S2 IOO.O 1 OUT OF 12 ( 8.31) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 9.0. MINIMUM EXPECTED CELL FREOUENCY • 4.737 RAW CHI SOUARE • 1.74188 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.8836 NUMBER OF MISSING OBSERVATIONS • SI 239 C 10 I S T I III L A T I 0 N BT CROUP CROUP COUNT I ROW PCT l » u l COL PCT I TOT PCT I f ROW TOTAL one* a wit I B 41.7 S.4 2.9 I 7 I 56.3 I ' (.« I 4.1 I 11 I 71.3 I 13.0 I C.4 -I I > I 43.0 I B.B ! 9.2 I I 3B.7 I B.O I 3.3 •I I 11 I 5S.0 ! 13.1 I 4.4 -I-4 1 33 1 38 wV 2 -3 I 34. 1 I 49 .8 I 35.» I 33.0 1 IB.2 I 16.3 -1- -I-9 I 19 I 14 moat day 1 S7.6 1 42.4 I 20.7 1 17.9 I 11.0 1 8.1 -I-« I 19 1 16 dally I 4B.4 I 91.6 I 16.3 I 20.0 I S.7 1 8 .1 -I-COLUMN 82 80 TOTAL 93.9 46.8 RAW CHI SOUARE • 4.18381 WITH 12 7.0 IS 8.7 20 11.6 61 33. B 33 IB.2 31 IB. O Salad Dressings/Mayonnaise (Current) 172 1O0.O 5 DECREES OF FREEDOM. SIGNIFICANCE NUMBER OF MISSING OBSERVATIONS • C R 0 5 S T A B U L A CFWG GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 contraI a Brtca 8 wk U a i l y 1 I I I I -I 2 I I I I -I 3 I I I I -I 4 I I I 1 -I-9 I day 1 I I -I-6 I 1 I I -I -COLUMN TOTAL 4 66.7 4.2 3.3 T I 0 N V CROUP PAGE I OF 1 ROW TOTAL I 3 I •I 1 I 2 I I 33.3 I I 3.4 I I 1.11 I 1 I 3 I Whole Grains (Current) 1 40.0 1 60.0 I 3.8 3.1 I 3.6 I 1.1 I 1.7 I 9 I 4 I a 99.6 I 44.4 I 9. 1 9.3 I 4.8 I 3.8 I 2.2 I S 1 .13 I 23 40.8 1 39. 1 I 12.4 9.9 I 19.7 I 9. 1 I 7.3 1 7 I 8 I 12 98.3 I 41.7 I 6.7 7.4 I CO I 3.8 I 2.8 I -I 68 I 96 I 124 94.8 I 49.3 I 68.7 71 .6 I 67.S 1 38.2 1 31.9 I -1 89 S3 178 83.4 46.6 IOO.O 6 OUT OF 12 ( SO.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 3.0 MINIMUM EXPECTED CELL FREOUENCY - 2.331 RAW CHI SOUARE • 2.40160 WITH S DEGREES OF FREEDOM. SIGNIFICANCE • 0.7912 NUMBER OF MISSING OBSERVATIONS • 25 240 COUNT BOW P C T COL P C T TOT P C T c a o u P I l e n t ! I C R O S S T A B U L A SOW TOTAL T 1 0 H BY GROUP PAOC 1 OF one* AI no once a wtt I I B I 55.* l * . ;> I 4.0 - I I « I a s . T I T . a I 4.a - I l t o I 71.4 I 13.0 l a . o - I -I 12 I 48.0 I 19.< I 9.C I 44.4 9.3 3.1 1 14.3 2.1 o . a a a . c a .3 3.2 I I I I 13 I 53.0 I 27.1 I 10.4 -I-8 1 7 I 3 no«t day I 70.0 I 30.0 I 9.1 I 9.3 1 9.C I 2.4 - I -C 1 37 1 23 dally I Cl.T I 39.3 1 48. 1 1 47.9 I 29.C I 18.4 - I -COLUHN 77 48 TOTAL S I . S 38.4 t 7.2 7 s . a R e f i n e d C e r e a l s (Current) 28 20.0 10 8.0 SO 48.0 128 10O.0 4 OUT OF 12 ( 33.3%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCT • 2.£88 RAW CHI 5QUARE • 4.£9478 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE - 0.4555 NUMBER OF MISSING OBSERVATIONS. CROUP COUNT 1 ROW PCT lcasaa COL P C T I TOT PCT I 1 C R 0 S S T A ROW TOTAL onct a no few K no bally I I I I I - I I I I I -1 I I I - I -COLUMN TOTAL 33 £7.3 35. t 19.3 32.7 30.5 9.3 I I - 1 -5 I day I 1 I - I -6 I 3S.4 4.3 2.3 7 S3.S a . o 4.1 a S O . O 3. 1 t . a a SO.O a . s 1.2 i I 1OO.0 I i . i I O . S - I •4 54.7 O O . O o.o o.o 79 45.3 A 7 I 0 N a r GROUP PAGE 1 OF 1 30 I 34 I 54 ' 55. S I 44.4 1 31.4 I 31.9 I 30.a I 17.4 J 14.0 I 34 1 39 I 53 45.3 I 54.7 I 30.8 35. S I 37 .3 I 14.0 I 16.9 I F r i e d Foods (Current) 48 38.5 I 0.6 173 1OO.0 S OUT OF 12 ( 41.77.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCT • 0.453 RAW CHI SOUARE • 7.4304I W I T H 5 DEGREES OF FREEOOM. SIGNIFICANCE • O. I90S NUMBER OF MISSING OBSERVATIONS ' 31 241 C I d S S T 1 I U L I GROUP COUNT I BOW PCT Icaaea COL PCT I TOT PCT I 1 I one* a mo 18 e > .a S4.T 60.0 ^ 1 35.0 3.3 3.3 « 30.1 73.7 36.7 3 73.0 37.1 10.0 COLUMN TOTAL 10 61.1 11 16.7 : o N CB0UP PACE 1 Of 1 ROW TOTAL 36 66.7 4 11.3 C h a r c o a l B r o i l e d Foods (Current) 30 100.0 3 OUT Of 4 ( 50.0%) OF THE VALID CELLS HAVE EXPECTED CELL fREOUENCT LESS THAN 3.0. MINIMUM EXPECTEO CELL fREOUENCT • 1.467 CORRECTED CHI SOUARE • 1.33637 WITH 1 DEGREE OF FREEDOM. SIGNIFICANCE • 0.3*95 RAW CHI SOUARE • 2.33050 WITH 1 DEGREE OF FREEDOM. SIGNIFICANCE - O.OS73 NUMBER OF MISSING OBSERVATIONS • 173 C R 0 S S 7 A 8 U L A T I 0 N . BY GROUP PAOE 1 OF 1 GROUP COUNT I ROW PCT Icaaas COL PCT I TOT PCT I 1 1 ones a no 1 50.0 100.0 SO.O 50.0 100.0 50.0 COLUMN TOTAL 1 50.0 1 SO.O a ROW TOTAL I -I 1 2 I 100.0 I I -I 2 100.0 P i c k l e d Foods (Current) STATISTICS CANNOT BE COMPUTED WHEN THE NUMBER OF NON-EMPTY ROWS OR COLUMNS IS ONI. NUMBER OF MISSING OBSERVATIONS - 201 C R O S S T A B U L COUNT ROW PCT COL PCT TOT PCT I Icaaaa I I 1 COLUMN TOTAL 3 33.3 3.4 I.B 6 66.7 8.2 3.7 3 37.5 3.4 1.9 S 63. S 6.8 3. 1 3 60.0 1.4 1.9 9 69.2 IO. 1 5.6 7 63.6 7.9 4.3 64 I 92 5S.2 I 44.8 71.9 I 71.2 39. S I .33.1 89 54.8 73 49. 1 A T I 0 N BY GROUP PAGE 1 OF ROW TOTAL I -I I I I I -I I I I I 1 3 I 40.0 I 2.7 I 1.2 I 1 4 I 30.8 I 9.9 I 2.9 I 1 4 I 36.4 9.9 3.9 8 5 . 6 5 3. 1 13 8.0 11 6.8 116 71.6 M i l k (Current) 162 100.0 7 OUT OF 13 ( 58.3X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.O. MINIMUM EXPECTED CELL FREOUENCY • 2.253 RAW CHI SOUARE • 4.14280 WITH 5 DEGREES DF FREEDOM. SIGNIFICANCE - 0.5390 NUMBER OF MISSING OBSERVATIONS • 242 C 10 1 ! I I I H I T ] OK BY GROUP COUNT HOW PCT COL PCT TOT PCT I C B B B B t I 1 BOW TOTAL 1 I a 33.3 1 68.7 ' 1.0 I a.« 0.6 I i.a a 1 8 aa.c I 71.4 a. 1 I «.S i.a I a.9 3 I 1 78.0 1 23.0 3. 1 1 1.3 1.7 I 0.6 COLUMN TOTAL 7 I 9 43.8 1 56.3 7.3 I 11.7 4.0 1 3.3 4 I 0 lOO.O I 0.0 4.3 1 0.0 a. 3 1 0.0 78 I 60 36.9 I 43.3 93.3 1 77.9 49.7 'I 34.7 96 77 33.3 44.3 7 4.0 C o f f e e (Current) IS 9.3 4 3.3 139 90.3 173 100.0 9 OUT Of 13 ( 66.7%) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTED CELL FREOUENCY • 1.335 RAW CHI SOUARE • 7.46956 WITH 5 DEGREES Of FREEDOM. SIGNIFICANCE • 0.1980 NUMBER OF MISSING OBSERVATIONS - 30 C R O S S T A B U L A T I O N BY GROUP PAGE 1 OF 1 GROUP I COUNT ROW PCT IcaaaB COL PCT I TOT PCT I 1 3 SO.O 3.9 1.7 5 43.3 4.9 3.9 59.3 S.9 3.9 control B I a I I 1 I 3 I I 40.0 I. I 3.6 1 I 1.1 I I 1 I S I I S4.S I I 7.8 I I 3.3 I I 1 I 5 I I 41.7 I I 6.5 1 3.8 1 7S .0 1 35.0 11 .7 I s. a S .7 I a.a S I a 71 .4 1 38.6 4 .9 I a.s 3 .9 I 1.1 71 53.0 69.9 39.4 39 43.0 73.3 33.3 COLUMN TOTAL 103 57.2 77 42.8 ROW TOTAL 5 3.8 Tea (Current) i i - i 4 I I I I I I 1 I I I I I I I I 13 S.7 16 8.9 7 3.9 129 71 .7 190 1O0.0 5 OUT OF 12 ( 41.77.) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 2.139 RAW CHI SOUARE • 3.53830 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE - 0.617S .NUMBER OF MISSING OBSERVATIONS • C 1 0 ! S I I 243 B U I 1 0 N CROUP PAGE 1 OF COUNT I ROW PCT I c e . COL PCT TOT PCT I 12 I 65.T I 2S.0 1 15.8 I i a I 47. 1 I 16.7 to.4 COLUMN TOTAL 6 33.3 30.7 7.a 8 52.8 31 .0 11.7 8 61.5 16.7 10.4 8 81.8 is.a 11.7 3 50.0 6.3 3.8 a 66.7 16.7 10.4 I 5 I 38.5 I 17.2 I 6.5 • I -1 3 1 18.2 1 6.9 t 2.S •I 1 3 I SO.O I 10.3 I 3.8 I I 4 I 33.3 I 13.8 I 5.2 48 62.3 38 37.7 ROW TOTAL IS 23.4 17 22. 1 13 16.8 11 14.3 6 7.8 12 15.6 Soda Pop (Current) 77 100.0 5 OUT OF 12 ( 41.7X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 2.260 RAW CHI SOUARE • 4. 1O03S WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE - 0.5351 NUM8ER OF MISSING OBSERVATIONS • 126 C R O S S T A B U L A T I 0 N GROUP PAGE 1 OF 1 GROUP COUNT I ROW PCT Iceeee COL PCT I TOT PCT I 1 I to I 8 I SS .6 I 44 .4 [ 25 .6 I 28 .6 [ 15 .2 I 12 . 1 7 I 3 70 .O I SO .0 17 .8 I 11 . 1 lO .6 I 4 .5 4 I 7 36 .4 I 63. .6 10 .3 I 25. .8 C. . 1 I lO. .6 - I -6 I 6 SO. O I SO. 0 IS. .4 I 22. 2 9. 1 I S. 1 5 1 1 83. 3 I 16. 7 12. a I 3. 7 7. 6 I 1. S 7 77 .8 3 22.2 7.4 l.O ROW TOTAL 18 27.3 10 15.2 Beer (Current) 12 18.2 6 9. 1 COLUMN TOTAL 27 40.8 66 100.0 6 OUT OF 12 ( SO.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MI Ml MUM EXPECTEO CELL FREOUENCY • 2.455 RAW CHI SOUARE * 6.10484 WITH 5 DECREES OF FREEOOM. SIGNIFICANCE ' 0.2962 NUMBER OF MISSING OBSERVATIONS • 137 2 4 4 C R 0 S S T A 9 U L A T 1 0 N BT CROUP PAGE 1 Of COUNT I ROW PCT !<:•••• COL PCT I TOT PCT 1 1 I I 1 1 -I I I I 1 I -I I I I I -I I I I I I -I I I I I -I-I I 1 I -I-I 1 1 3 15 t • 65.3 I 34. a 23. 1 I 18.2 13.• I 7.3 12 I 14 46.2 1 53.8 19.S I 31.8 II.O I 12.8 ROW TOTAL 23 31.1 36 33.8 Wine (Current) 13 X 11. I 34 54.3 I 45.8 1 33.0 3O.0 1 35.0 I 11.» I 10.1 I 14 I 6 I 30 70.0 1 30.0 1 18.3 31.5 I 13.6 I 13.8 I 5.5 I O I 3 I 3 O.O I 10O.0 t 3.8 0.0 I 6.8 I O.O 1 3.8 I 11 I 2 1 13 84.6 I 15.4 1 11.8 16.8 I 4.5 I 10.1 I 1.8 I 65 44 109 58.6 40.4 lOO.O COLUMN TOTAL 3 OUT Of 13 ( 16.7X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5 O MINIMUM EXPECTED CELL FREOUENCT • 1.311 RAW CHI SOUARE • 11.25355 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.0466 NUMBER OF MISSING OBSERVATIONS - 94 C R O S S T A B U L A T I O N BY GROUP PAGE 1 OF GROUP COUNT I ROW PCT IcasMB C°'. PCT I TOT PCT I 1 I I - I -I I I I -I-I I I -I-I 1 I I -I-I I I -I I I COLUMN TOTAL 11 55.0 19.3 13.0 5 63.5 8.8 5.4 lO 5S.C 17. S 10.9 13 80.0 21.1 13.0 44.4 7.0 4 . 3 57 62.0 3 37.5 3.3 35 38.0 ROW TOTAL 9 45.0 25.7 9.8 1 -I I I I I -I I I I I 1 8 I 44.4 I 33.9 I 8.7 I 1 3 I 20.0 I S.S I 3 . 3 I 1 5 I SS.S 14.3 S.4 15 I 7 68.2 I 31. .8 26.3 ' I 20. O 16.3 I 7. . 6 1 20 21.7 8 8.7 18 19.6 IS 16.3 9 9.9 22 23.9 L i q u o r ( C u r r e n t ) S3 lOO.O 3 OUT OF MINIMUM EXPECTED CELL FREOUENCY • •AW CHI SOUARE • 4.33910 WITH ( 35.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN S.O. 3.043 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.S03I MJ«BER OF MISSING OBSERVATIONS • 111 245 c t o I S t I I l i t I T I 0 N BT GO OUT" GROUP COUNT 1 fiOW PCT lc»s«» COL PCT I TOT PCT I I IB T1.4 33.6 38.9 COLUMN TOTAL 1 1 -I I T I IOO.O I 29.0 I IT .8 -I 1 2 I 33.3 I 7. I I 9.1 -I I 3 I 75.0 I 10.7 I T.7 -I I 1 I IOO.O I 3.8 I 2.C -I 28 71.8 1 t 1 I 4 1 SC.7 I 36.4 I 10.3 I 1 1 1 35.0 8. I 2.6 I I t I 0 I 0.0 I 0.0 J 0.0 I 1 11 28.2 PACT. i or ROW TOTAL 2 I I 8 I 28.6 I 54.8 I 15.4 I 1 O 1 O.O 0.0 0.0 31 53.6 T 17.9 6 15.4 4 10.3 1 2.6 39 IOO.O L i q u e r s (Current) 7 OUT Of 10 f 70.OX) o r THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 9.0. MINIMUM EXPECTED CELL FREOUENCY • 0.282 RAW CHI SOUARE - 7.54765 WITH 4 DECREES OF FREEOOM. SIGNIFICANCE - 0.1096 NUMBER OF MISSING OBSERVATIONS • 164 * C R O S S T A B U L A T I O N _ V G C BY GROUP PAGE I Of A VGC lo GROUP COUNT I ROW PCT Ic»m»s COL PCT I TOT PCT 1 1 I IS 26 58.3 29.2 13.8 30 41.7 22.2 10.0 COLUMN TOTAL 111 95.2 44 90 8 RAW CHI SOUARE • 1.16037 WITH NUMBER OF MISSING OBSERVATIONS -ROW TOTAL I 47.1 I 53.B I 14.4 I 30.0 I S.O I 9.0 -I 1 1 67 I 62 I 96.3 I 43.7 I 60.4 I 57.8 I 33.3 I 35.8 -I-I I I I -1-16.8 118 59.3 23.8 V i t a m i n A Ric h Vegetables (Current) 201 100.0 2 OEGREES OF FREEDOM. SIGNIFICANCE • 0.5598 C R O S S T A B U L A I 0 N CROUP CVGC lO GROUP COUNT I ROW PCT I C A H I COL PCT I TOT PCT I 1 I I I I -1 2 I COLUMN TOTAL [ 1 I 2 17 1 13 56.7 I 43.3 13.3 I 14.1 8.4 1 6.4 69 I 97 54.8 I 45.2 62.2 1 62.6 34.2 I 28.2 23 I 21 54.3 I 43.7 22.9 1 23.1 13.4 I 10.4 III 91 93.0 4S.0 PAGE 1 OF ROW TOTAL 30 14.9 136 63.4 46 32.8 V i t a m i n C R i c h V e g e t a b l e s (Current) 302 IOO.O RAW CHI SOUARE • 0.0443S WITH 3 OEGREES OF FREEDOM. SIGNIFICANCE • 0.6781 NUMBER OF MISSING OBSERVATIONS • I 246 GROUP COUNT J ROH PCT leases COL PCT I TOT PCT I 1 C R O S S T A B U L A T ROW TOTAL I 0 N GROUP AGE 1 OF t i I 41 1 33 I 76 lo I 53.B I 46.1 1 37.4 1 36.9 I 38.0 I > 30.3 I 17.3 I i I 67 I 56 1 133 • m I 54.5 I 43.3 1 60.6 1 60.4 I 60.8 I 1 33.0 I 37.6 I 3 I 3 I 1 I 4 hi I 75.0 1 33.0 I 2.0 I 3.7 1 1. 1 I I 1.5 I 0.5 I - I -COLUMN 111 83 203 TOTAL 34.7 45.3 10O.0 C r u c i f e r o u s Vegetables (Current) 3 OUT OF 6 ( 33.3X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY - 1.813 RAW CHI SOUARE - 0.68510 WITH 2 DEGREES OF FREEDOM. SIGNIFICANCE • 0.7100 C R O S S T A B U L A I 0 N GROUP PAGE 1 OF 1 GROUP COUNT t ROW PCT Icam COL PCT I 10T PCT !•• 1 FATC lo COLUMN TOTAL 17 53. 1 15.3 8.4 87 55.8 78.4 43.8 7 46.7 6.3 3.4 111 54 .7 69 44.3 75.0 34.0 8 53.3 8.7 3.9 93 43.3 ROW TOTAL 3 I 1 15 I 46.9 16.3 7.4 33 15.6 156 76.8 Fat (Current) 303 IOO.O RAW CHI SOUARE • O.49460 WITH 2 DEGREES OF FREEOOM. SIGNIFICANCE • 0.7809 C R 0 S 5 T A I O N GROUP MEATC lO GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 I 14 I 46.7 I 13.6 I 6.9 - I I 90 I 36.6 I 81. 1 I 44.3 - I I 7 I 50.0 I 6.3 I 3.4 - I Ill 54.7 COLUMN TOTAL I 16 I 53.3 I 17.4 I 7.9 69 43.4 73.0 34.0 7 SO.O 7.6 3.4 93 45.3 ROW TOTAL 30 4.8 159 78.3 Meat (Current) 303 100.0 RAW CHI SOUARE I.13837 WITH 2 OEGREES OF FREEDOM. SIGNIFICANCE • 0.5659 247 C R O S S T A B U L VCPRC l a GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 I I I I -I-I I I I -I-I I 1 I -I-43 S3.3 37. 8 30.8 83 57.3 5S.8 31.3 S 48.3 5.4 3.0 I 7 t S3.8 I 7.7 I 3.5 COLUMN TOTAL 111 55.0 81 45.0 A T BY I 0 N GROUP PAOE 1 Of SAW CHI SOUARE • 0.74778 WITH NUMBER OF MISSING OBSERVATIONS • ROW TOTAL 3 ! - I 37 I 46.8 I 40.7 I 18.3 I -I 47 I 43.7 51.6 33.3 78 38. I 110 54.5 13 6.4 Vegetable P r o t e i n s (Current) 203 100.0 2 DEGREES OF FREEDOM. SIGNIFICANCE - 0.6B8I C R O S S T A B U L A T I 0 N BY GROUP PAGE I OF COUNT ROW PCT COL PCT TOT PCT FIBRC lo leases I I 1 COLUMN TOTAL I 25 I 62.5 I 22.5 I 12.3 -I-t I I I -I-I I I 8 60.0 8.1 4.4 111 54 .7 I IS I 37.5 1 16.3 I 7.4 77 52.0 69.4 37.9 71 49.0 77.2 35.0 6 40.0 6.5 3.0 92 45.3 ROW TOTAL 149 72.9 15 7.4 303 lOO.O F i b e r R i c h Foods (Current) RAW CHI SOUARE • 1.57874 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.4541 C R 0 S S T A A T I 0 N OF 8Y GROUP PAGE 1 OF COUNT ROW PCT COL PCT TOT PCT leases I COLUMN TOTAL ROW TOTAL 1 I 2 13 I 5 73.3 I 27.9 11.7 I 5.4 6.4 I 2.5 69 I 59 54.3 I 45.7 63.3 I 63.0 34:0 I 29.6 29 I 29 SO.O I SO.O 26. 1 I 31.9 14.3 I 14.3 111 92 S4.7 45.3 Beef Fat (Current) 127 62.6 58 28.6 203 lOO.O RAW CHI SOUARE - 2.7S411 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.2533 248 [ > D S S 1 I I 0 N CROUP COUNT ROW PCT COL PCT TOT PCT Icaaaa I onca a wk dally I I I I I -I 3 I I I I -I 3 I I I I -I' 4 I I I I - i s I day I I I -I « I I I I -I-1 1 3 s I 2 Tt.4 I 28.6 4.7 1 2.3 2.6 I 1 .0 7 I 8 43.6 I 56.3 6.9 1 10.5 3.6 1 4.7 33 1 35 47.9 I 52. 1 21 .S I 29. 1 11.9 1 13.0 59 I 42 56.4 I 41.6 55. 1 I 48.8 30.6 I 21.8 9 1 2 SI.8 I 18.2 8.4 I 2.3 4.7 I 1.0 4 I 6 40.0 1 60.0 3.7 I 7.0 2.1 I 3. 1 COLUMN TOTAL 107 55.4 PAGE 1 OF ROW TOTAL 7 3.6 16 1.3 46 34.8 101 52.3 11 5.7 10 5.2 Y e l l o w V e g e t a b l e s ( M i d d l e Y e a r s ) 193 100.0 4 OUT OF 13 ( 33.3*) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCT • 3.119 RAW CHI SOUARE • 7.13447 WITH 5 OEGREES OF FREEOOM. SIGNIFICANCE • 0.2108 NUMBER OF MISSING OBSERVATIONS • C R 0 S S T A ION GROUP PAGE I OF COUNT ROW PCT COL PCT TOT PCT lcasaa I faw x mo onca a wk 5 day COLUMN TOTAL 2 IOO.O 2.1 1.2 96 98.8 ROW TOTAL ! 1 I 2 I -1 19 I 10 I 25 I 60.0 I 40.0 I 15.2 [ 15.6 I 14.7 I ! 9. 1 I 6. 1 I 20 I 14 I 34 58.8 I 41.2 I 20.7 20.8 I 20.6 I 12.2 I 8.5 I 33 I 28 I 60 53.3 I 46.7 I 36.6 33.3 I 41.2 I 19.5 I 17. 1 I 27 I 16 I 43 62.9 I 37.2 I 36.2 28. 1 I 23.5 I 16.5 I 9.8 I o 0.0 O.O 0.0 68 41.8 B r o c c o l i / B r u s s e l ' s S p r o u t s ( M i d d l e Y e a r s ) 2 1.2 164 100.0 2 OUT OF 10 ( 20.0*) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.829 RAW CHI SOUARE • 2.42978 WITH 4 OEGREES OF FREEDOM. SIGNIFICANCE • 0.6573 NUMBER Or MISSING OBSERVATIONS • 39 249 GROUP 1 COUNT ROW PCT leases COL PCT I 101 PCT I 1 one* a ao one* * wk B day • lly COLUMN TOTAL 11 73.J 10.7 S.S I 1 I I I I 30 I 43.S I IS.4 I IO.S I--I 48 63.2 46.6 25.8 I 27 I S7.4 t 32.S I 14.S I 26 I 36.8 I 33.7 I 1S.1 32 21.4 11.6 2 lOO.O 1.8 1.1 O O.O O.O O.O 103 SS.4 I 23 I SI . I ! 27.7 1 12.4 •I -I O I 0.0 I 0.0 I 0.0 •I I 1 t t o o . o I 1.2 I O.S I 83 44.6 C R O S S 7 A 8 U L A ROW TOTAL i a - i I 4 I 36.7 I 4.8 1 3.3 7 I 0 N 1 GROUP Pace 1 OF I IB 8.1 47 2S.3 76 40.8 45 24.2 a i . i i O.S Cauliflower/Cabbage (Middle Years) 186 100.0 4 OUT OF 12 ( 33.3%) OF THE V4LI0 CELLS HAVE EXPECT EO CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCT • 0.446 RAW CHI SOUARE - 10.S6E23 WITH S OEGREES OF FREEDOM. SIGNIFICANCE • 0.0607 NUMBER OF MISSING OBSERVATIONS • 17 C R O S S T A B U L A T ION OF • • • • • • • • • • • • • • • • • BT GROUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ptat , Qf GROUP COUNT I ROW PCT leases controls ROW COL PCT I TOTAL TOT PCT I I I 3 I -I I 4 1 3 1 7 I 57.1 I 43.S I 3.8 I 4.0 I 3.6 I i 3.3 J i.l ! Leafy Greens - i 1 1 1 1 II I I  1  1 4.0 1.3 I 1 6 -I-3 I 13 30.0 I 80.0 3.0 I 14.3 1.6 I 6.S 17 I 11 60.7 I 39.3 16.8 I 13. I S.3 I 5.9 (Middle Years) I 1 I I 9 I I I I I -I-I I I I 9 1 I I 9 I _| 1 1 4 1 38 I 3S I S3 wk 3-3 I SO.3 I 39.7 I 34.1 I 37.S I 39.8 I I 30.5 I 13.5 1 -| 1 1 9 1 30 I 14 1 34 •oat o>ay I 59.8 I 41.3 I 18.4 I 19.9 I IS.7 1 1 IO.S I 7.S I -! ; 1 1 C I IS t IS I 38 dally I SO.O I SO.O I 30.6 I 18.8 1 33.S I I 10.3 I 10.3 I -| 1 1 COLUMN lOI 84 185 TOTAL 54.6 45.4 100.0 3 OUT OF 13 ( 16.7%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCT • 3.1/6 RAW CHI SOUARE - S.OS448 WITH S OEGREES OF FREEDOM. SIGNIFICANCE • O.IOST NUMBER OF MISSING OBSERVATIONS • 18 250 C R O S S T A S U L A T I O N ar GROUP PAGE I OF COUNT I ROW PCT Icaaae COL PCT I ROW TOTAL d a l l y MFVR TOT PCT I 1 t 2 1 1 10 I 0 onca a BO I 100.0 1 0.0" I a.4 I 0.0 I a.2 I 0.0 3 1 10 I 22 few x « I 31.3 1 68.8 I 8.4 I 25.0 I 8.2 I 11.3 -I-3 I 37 I 30 onca a wk 1 55. 2 1 44.8 I 34.8 I 34.1 I IS. 1 I 15.5 4 I 36 I 31 wk 2-3 I 83.7 I 46.3 I 34.0 I 33.3 I 18.6 I 16.0 -1-S I 8 I 3 noat day t 72.7 1 37.3 I 7.3 ! 3.4 I 4.1 I 1.3 6 71.4 4.7 3.6 COLUMN TOTAL 106 54.6 2 28.6 2.3 1.0 88 45.4 10 5.2 33 16.3 67 34.8 67 34.3 11 8.7 7 3.6 F i b r o u s V e g e t a b l e s ( M i d d l e Y e a r s ) 184 100.0 4 OUT OF 13 ( 33. 3X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 3.175 RAW CHI SOUARE • 17.64468 WITH 5 OEGREES OF FREEOOM. SIGNIFICANCE • O.0O34 NUMBER 6F MISSING OBSERVATIONS - 8 C R O S S T A B U L A T I 0 N Y GROUP PAGE I OF I COUNT ROW PCT COL PCT Icaaas I ROW TOTAL TOT PCT I 1 I 2 I I 15 I 8 onca a MO I 65.2 I 34 .8 I 20.3 I 15 .7 I 12.0 I 6 .4 -1-2 I 13 I 9 f aw x » o I 59. 1 1 40. 9 I 17.6 I 17. 6 I 10.4 I 7. 3 -I-onca a wk da I l y 3 I I I I -I 4 I I I I -I 5 I day I I I -I-6 I I I I -I-14 I 7 66.7 I 33.3 18.9 I 13.7 11.2 I 3.6 16 I 8 66.7 I 33.3 21.6 I 13.7 12.8 1 6.4 10 I 12 45.3 I 34.3 13.3 I 23. S 8.0 I 9.6 6 46.3 8.1 4.8 I 7 I 53.8 I 13.7 1 5.6 33 IS. 4 31 16.8 34 19.2 22 17.6 13 10.4 P e a n u t s / P e a n u t B u t t e r ( M i d d l e Y e a r s ) COLUMN TOTAL 74 39.2 51 40.8 125 IOO.O RAW CHI SOUARE 4.O20S7 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.5465 NUMBER OF MISSING OBSERVATIONS • 78 251 C R O S S T A ION GROUP PAGE 1 OF I GROUP COUNT I ROW PCT l e a s e s COL PCT I TOT PCT I 1 one* a wk 1*ost dal ly t I 1 1 I -I 2 I I I I -I 3 I I 1 I -I 4 I I I 1 -I-5 I day I I 1 -I-e i 1 I COLUMN TOTAL 26 61.9 44.8 13.0 lOO.O 1.7 0.8 1 lOO.O 1.7 0.9 58 51.3 16 38. 1 29. I 14.3 I I 0.0 I O.O I 0.0 ROW TOTAL 14 I 31 40.0 I 60.0 34. 1 I 38.2 12.4 I 18.6 13 I 16 42.9 I 67. 1 30.7 I 29. 1 10.6 I 14.2 4 1 3 66.7 I 33.3 6.9 I 3.6 3.8 I 1.8 0 I t I • I O I 0.0 I 0.0 I 0.0 I I 35 48.7 42 37.2 35 31.0 29 34.9 6 6.3 1 O.S D r i e d Beans/Peas (Middle Years) 113 100.0 6 OUT OF 13 ( SO.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5 O MINIMUM EXPECT EO CELL FREOUENCY - 0.487 RAW CHI SOUARE - 6.94430 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE " 0.3248 NUMBER OF MISSING OBSERVATIONS • SO C R 0 S S T A 6 U L A PAGE 1 OF COUNT ROW PCT COL PCT TOT PCT leases I ROW TOTAL once e Mo dally I 1 -I 4 I I 1 I -I 5 I day I I I -I S I I I t COLUMN TOTAL 1 I 2 3 I 2 6O.0 I 40.0 3.7 I 2.2 1.5 I 1.0 11 I io 32.4 I 47.6 9.9 I 10.9 5.4 I 4.9 37 I 45 45. 1 I 54.9 33.3 I 48.9 19.2 I 22.2 53 I 29 65.5 I 34.5 49.9 I 31.5 27. 1 I 14.3 3 I 3 62.5 I 37.6 4.3 I 3.3 2.3 I 1.5 0 I 3 0.0 I 100.0 0.0 I 3.3 0.0 I 1.5 111 82 34.7 43.3 5 2.5 F i s h / P o u l t r y (Middle Years) 8 3.9 203 100.0 . 6 OUT OF 12 ( SO.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY - 1.360 RAW CHI SOUARE - IO.89385 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • O.OS35 252 C> 0 S S t 1 1 UL I T I 0 N V GROUP GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT onca a MO brice a wk illy 1 I 1 I I -I 2 I I I I -I 3 1 I I I -t 4 I I 1 I -1 6 I day I 1 I -I C I I 1 1 -I 1 t X 2 [ 1 I 0 100.0 I 0.0 0.9 I 0.0 0.5 1 0.0 4 I 4 50.0 I 50.0 3.6 I 4.4 3.0 I 2.0 10 1 14 41.7 I SB .3 9. 1 1 15.4 5.0 I 7 .0 E7 I 3B 63.B I 36.2 60.9 I 41.8 33.3 I IB .9 24 I 30 44.4 I 55.6 21.9 I 33.0 11.9 I 14.9 4 I S 44.4 I 5S.6 3.6 1 5.5 2.0 I 3.5 COLUMN TOTAL 110 54.7 81 46.3 PAGE 1 OF 1 ROW TOTAL I 0.6 6 4.0 24 11.9 105 52.2 54 26.8 ' 9 4.6 Beef (Middle Years) 201 100.0 6 OUT OF 12 < SO.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCT • 0.453 RAW CHI SOUARE • 8.73601 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.1201 NUMBER OF MISSING OBSERVATIONS • 2 C R O S S T A S U L A T I O N BT GROUP PAGE. 1 OF COUNT ROW PCT COL PCT TOT PCT Icaaea I I 1 onca a MO onca a wk dally 9 day COLUMN TOTAL IS 71.4 14.6 8.0 6 28.6 7.1 3.3 11 I 17 39. 3 I 60.7 10.7 I 20.3 5.9 I 8. 1 33 I 32 S9.3 1 40.7 31.1 I 26.2 17. 1 I 1 1.8 28 I 30 48.3 I SI .7 27.2 I 3S.7 19.0 1 16.0 13 1 » 65.0 I 35.0 13.6 I 8.3 7.0 I 3.7 4 1 3 66.7 1 33.3 3.8 I 2.4 103 55. 1 84 44.8 ROW TOTAL 21 11.2 38 15.0 54 28. S 58 31.0 30 10.7 6 3.3 Cured Meats (Middle Years) 187 IOO.O 2 OUT OF 12 ( 16.7X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY - 2.695 RAW CHI SOUARE • 7.67913 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.1748 NUM8ER OF MISSING OBSERVATIONS • 16 253 . . . . . . . . . . . . . . . . C I D i S T I 1 UL I I | 0 N "'CMES B» COOOP PAGE I OF I COUNT I BOW PCT I C I H I COL PCT TOT PCT onca a mo once a wfc • I l y 1 1 I 1 I - I -2 I I I -I 3 I I . I 1 -I 4 I 1 I 1 - i s I day I 1 I -I-C ) I I I - I -COLUHN TOTAL ( lOO.O S.T 3.1 S 31.3 4.7 2.S II s o . a 12.a 5.7 62.S 42.8 23.4 21 63.6 19.a IO.S 37 37.8 31.4 14.1 13 36.4 14.0 6.3 lO 43.8 S.4 5.3 13 56.5 19.1 6.a 106 55.2 86 44 . a ROW TOTAL 2 I 1 O I I 0.0 I o . o I 0.0 IS I 23 45.2 I 54.8 17.S I 26.7 ] 9.9 I 12.0 ] 1 1 I I I I I I 1 I I I I I -I I I I -I I I I I -I 6 3.1 16 a.3 42 21.9 73 37.8 33 17.3 23 13.0 Cheese (Middle Years) 193 100.0 2 OUT OF 12 { 16.7%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5 O MINIMUM EXPECTEO CELL FREOUENCT • 2.699 RAW CHI SOUARE . 14.0458T WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE - 0.0153 F4JMBER OF MISSING OBSERVATIONS • 11 C R 0 S S T A 8 U I A T I 0 N BY GROUP PAGE 1 OF COUNT I ROW PCT Icaaea COL PCT I ROW TOTAL fa* x mo A l l y PCT I 1 1 3 I 1 I 1 I 3 I 3 I 33.3 I S6.7 1 1.S I O.S I 3.2 • I O.S I 1.0 1 -1-2 I 6 I 3 I 9 I 66.7 I 33.3 I 4.6 I S.S I 3.3 I I 3.0 1 1.5 I 3 t 20 I 13 I 32 I 62.5 I 37.5 I 16.3 I 18.7 I 13.3 1 I 10.2 I 6.1 I 4 I 44 I 37 1 91 I S4.3 I 45.7 I 41. 1 I 41.1 I 41. 1 I t 32.3 I 18.8 I 8 I 18 I 32 I 41 day 1 46.3 I 83.7 I 20.8 I 17.8 1 24.4 I I • .a > 11.3 X 6 1 17 I 14 1 31 I 54.8 I 45.3 1 19.7 1 15.B 1 1S.S I I S.S I 7.1 t COLUMN TOTAL lOT 54.3 BO 45.7 Eggs (Middle Years) 197 100.0 4 OUT OF 12 ( 33.3%) Of THE VALID CELLS HAVE EXPECTEO CELL fREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL fREOUENCT • I.3TI RAW CHI SOUARE - 3.O0346 WITH 5 DECREES OF FREEDOM. SIGNIFICANCE - 0.6995 NUMBER OF MISSING OBSERVATIONS 254 C R O S S T A B U L A T I O N BT GROUP GROUP I COUNT ROW PCT Icaaaa COL PCT 1 TOT PCT I ROW TOTAL onca a MO f l H X N onca a wk Nik 2-3 1 I 2 4 I 1 SO.O I 20.0 4.3 1 1.3 2.4 I O.S S I 8 5O.0 I SO.O S.3 I 6.7 3.0 I 3.0 4 I 10 38.6 ! 71.4 4.3 I 13.3 2.4 I S.9 IS I IS SO.O I s o . o 18.0 I 30.0 8.8 I a . s 5 I 18 1 17 Moat day I S3.a I 47.2 1 30.2 I 22.7 1 11.2 I 10. 1 6 I 47 I 27 dally I 63. S I 36.5 I SO.O I 36.0 I 37.8 I 16.0 1- -I-COLUMN 84 75 TOTAL SS.6 44 .4 5 3.0 10 9.8 14 8.3 30 17.8 36 31.3 74 43.8 C i t r u s F r u i t / J u i c e ( M i d d l e Y e a r s ) 168 IOO.O 3 OUT Of 12 ( 25.OX) Of THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCT • 2.219 RAW CHI SOUARE • 7.85108 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.1646 NUMBER OF MISSING OBSERVATIONS • . . . . . . . . . . . . . . . . C R O S S T A B U L A T I O N Of MFTOM BY GROUP PAGE I OF GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 onca a MO onca a wk dally 1 I I 1 I -I 2 I I I I -I 3 I I I I -I 4 I I I I -I-9 I day I I I -I-« I I I I -I-COLUKN TOTAL 3 100.0 3.0 1.6 I O I 0.0 I O . O I O . O 6 I 5 54.5 I 45. S 6.1 I 6.0 3.3 I 2.7 16 I 13 55.2 I 44.8 16.2 I 1S.S B.7 I 7.1 45 I 33 57.7 I 43.3 45.5 I 39.3 24.6 I 18.0 16 43.2 16.2 8.7 21 56.8 25.0 11.S 13 52.0 13. 1 7.1 I 12 I 48.0 I 14.3 I 6.6 .99 54. 1 B4 45.9 ROW TOTAL 3 1.6 T o m a t o e s / J u i c e ( M i d d l e Y e a r s ) 37 20.2 25 13.7 183 100.0 2 OUT OF 12 ( 16.7%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN S.O MINIMUM EXPECTED CELL FREOUENCY - 1.377 RAW CHI SOUARE - 4.76559 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.4453 NUMBER Of MISSING OBSERVATIONS • 20 255 C R 0 S S T A GROUP COUNT I ' ROW PCT Icaaaa COL PCT I TOT PCT I I one* * ma 14 40.0 IT. 1 S.I 36 63.3 43.9 33.4 3 1 35 1 27 onca A wk I 48 . 1 I SI .8 1 30 .5 1 37.5 I 16 .2 I 17.B 4 I 7 I 3 wk 3-1 I 70 .0 I 30.0 t 8 .5 I 4.3 I 4 .8 I 1.8 -I-COLUMN 82 72 TOTAL S3 .2 46.8 U L A T I 0 N BT GROUP PAGE 1 OF I RC'i TOTAL 3 I I 21 I 60.0 1 28.2 I 13.6 1 I. 31 > 36.8 28.2 13.6 33 32.7 37 37.0 32 33.8 10 6.S Pasta (Middle Years) 134 IOO.O 1 OUT OF 8 ( 12.3X1 OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 3.0. MINIMUM EXPECTED CELL FREQUENCY • 4.673 RAW CHI SOUARE • 6.40193 WITH 3 OEGREES OF FREEDOM. SIGNIFICANCE • 0.0936 NUMBER OF MISSING OBSERVATIONS • 48 C R 0 S S T A U L A T I O N BY GROUP GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 onca a mo onca a wk illy 9 Oay COLUMN TOTAL 12 1 4 73.0 I 23.0 12. 1 I . 5.S 7.0 I 3.3 14 1 13 S3.8 I 46.3 14 . 1 I 16.4 8. 1 I 7.0 IS I 20 47.4 I 33.6 18.3 I 37.4 10.3 I 11.6 34 I 19 64.2 I 35.8 34.3 I 26.0 18.8 I 11.0 7 46.7 9.6 4.1 13 54.2 13. 1 7.6 11 43.6 15.1 6.4 99 37.6 73 42.4 PAGE 1 OF 1 RAW CHI SOUARE - 4.91990 WITH NUMBER OF MISSING OBSERVATIONS • ROW TOTAL 16 9.3 26 IS. 1 38 23. 1 S3 30.8 15 8.7 24 14.0 Ice Cream (Middle Years) 172 100.0 5 OEGREES OF FREEOOM. SIGNIFICANCE • 0.42S7 256 C R O S S T A B U L A T ION GROUP COUNT ROW PCT COL PCT ICAARS I ROW TOTAL TOT PCT I 1 I 2 I 1 I 14 I 21 ! as one* a no I 40.0 I 60.0 I 22.7 I 17. 1 1 29.2 1 I 9. 1 I 13.6 I a . I ' 36 1 21 I 57 faw x mo I 63.2 I 36.8 I 37.0 I 43.9 I 29.2 I I 23.4 I 13.6 I I- -I a I 25 I 27 I S3 onca a wk I 48. 1 I 51.9 I 33.8 I 30.5 I 37.8 I I 16.2 I 17.5 1 I-4 I 7 I 3 I 10 wk 2-3 I 70.0 I 30.0 I 6.S I 8.5 I 4.2 I 1 4.5 I 1.8 I -I-COLUMN 82 73 154 TOTAL 53.2 46.8 100.0 Pasta (Middle Years) 1 OUT OF 8 < 12.3%) OF THE VALID CELLS HAVE EXPECTED CELL FREQUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREQUENCY - 4.67S RAW CHI SQUARE • 6.40183 WITH 3 OEGREES OF FREEOOM. SIGNIFICANCE - 0.0936 NUMBER OF MISSING OBSERVATIONS ' 4 8 C R O S S T A B U L A T BY ION GROUP GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 onca a lao onca a wk dal ly 1 . I " I t I -I 3 I I I I -I 3 I I I I -I 4 I I I I -I 9 I day I I I -I 8 I I I I -I-13 I 4 75.0 I 25.0-12.1 I 5.5 7.0 I 2.3 14 I 12 53.8 I 46.3 14. 1 I 16.4 8. 1 I 7.0 18 I 20 47.4 I 52.6 18.2 I 27.4 10.5 I 11.6 34 I 19 64.2 I 35.8 34.3 1 26.0 19.8 I 11.0 8 I 7 93.3 I 46.7 8. 1 I 9.6 4.7 I 4.1 13 84.3 13. 1 7.6 11 4S.8 IS. 1 6.4 PAGE 1 OF ROW TOTAL 16 9.3 26 15. 1 38 22. 1 53 30.8 19 8.7 34 14.0 Ice Cream (Middle Years) COLUMN TOTAL 99 97.6 73 42.4 173 100.0 RAW CHI SOUARE - 4.91990 WITH 5 DECREES OF FREEDOM. SIGNIFICANCE ' 0.4257 NUMBER OF MISSING OBSERVATIONS, • 31 257 COUNT ROW PCT COL PCT Icaass I C R 0 S S T A 8 U L A ROW TOTAL few it i-11 l y T 1 O N BY GROUP PAGE 1 OF TOT PCT I 1 1 3 1 I a I 3 I 5 a ao 1 40.0 1 60.0 I 3.9 I a. t I 3.9 I 1 t .a 1 1.7 I - I - - I - -1 a I 3 I 7 I 10 it BO I 30.0 t 70.0 I s. a 1 3.3 1 a. i I 1 1.7 1 4.1 1 - I - -I 3 I 13 I 14 I 37 • wk 1 4«. 1 I s i . a I 18.7 1 13.7 1 18.3 X I 7.6 I a. i 1 -1- -I 4 1 41 1 33 I 64 •3 I 64 . 1 I 35.8 I 37.3 I 43.3 I 39.9 I I 33.S I 13.4 I 5 1 33 I 16 I 38 day I 59.0 I 41.0 I 33.7 t 34.3 : 20.3 X t 13.4 i 9.3 I I-6 I 13 l 14 I 27 I 43. 1 i 31.9 I 13.7 I 13.7 i 19.3 I t 7.6 ! 9.1 I COLUMN 85 77 172 TOTAL S5.3 44.8 lOO.O S a l a d D r e s s i n g s / M a y o n n a i s e ( M i d d l e Y e a r s ) 3 OUT OF 12 ( 33.07;) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 3 O MINIMUM EXPECTED" CELL FREOUENCY • 3.238 RAW CHI SOUARE - 6.37913 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.3711 NUMBER OF MISSING OBSERVATIONS - 31 W h o l e G r a i n s ( M i d d l e Y e a r s ) . . . . . . . . . . . . . . . . . . C R O S S T A B U L A T I O N OF . . . . . . . . . . . . . . . . . MFWG BY GROUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PAGE 1 OF GROUP COUNT I ROW PCT Icasas controls ROW COL PCT I TOTAL TOT PCT I I I 3 1 HFWQ 1 1 1 I I 3 1 4 1 6 onca a MO I 33.3 I 66.7 I 4.1 I 3.6 I 3.7 I I 1.4 I 3.7 I _I 1 j 3 1 3 1 3 1 6 fax XMO I SO.01 SO.01 4.1 I 3.9 I 4.3 I I 3.0 I 2.0 I ,j ] 1 3 1 S I 3 1 9 onca a wk I 66.7 I 33.3 I 6.1 I 7.9 I 4.3 I I 4.1 I 2.0 I -I 1 1 4 1 9 1 13 I 22 wk 2-3 I 40.9 1 S9.1 I 13.0 1 11.7 I 19.S I I S.I I 8.8 I .j 1 1 9 1 3 1 9 1 8 •oat day 1 33.3 I 66.7 I 6.1 1 3.9 I 8.S I I 2.0 I 4.1 I .1 1 1 S I 94 I 41 I 93 dally I 36.8 I 43.3 I 64. S I 70.1 I 58.6 I I 36.7 I 37.8 I -I 1 1 COLUMN 77 70 147 TOTAL 53.4 47.6 100.0 8 OUT OF 13 ( 66.7X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY - 2 . B5T RAW CHI SOUARE - 4.85055 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.4344 NUMBER OF MISSING OBSERVATIONS • 36 258 C P. 0 S S T A I I 0 N BY CROUP PACE I OF 1 CROUP COUNT I ROW PCT Icaaaa COL PCT I lOT PCT I 1 ROW TOTAL onct • wk daily 4 66.7 4.7 S.7 • i i - i a l I I I -I 4 I 1 I I -I 5 I day t I I -I C I 1 I I -I a SO.O a.e 7.0 a t 1 a l aa. a i a.a i 1.4 I a 50.0 4.a 3.0 9 I 3 75. .0 I a s . o 10. .6 I 4.B : s. . 1 I 2.0 ] COLUMN TOTAL IB I 12 87 . 1 1 42.8 18.8 1 18.4 10.8 I 8. 1 C I a 66.7 1 33.3 7.0 I 4.8 4.1 I 2.0 48 I 38 55.2 I 44.8 55.8 I £2.8 33.4 t 26.4 88 62 58. 1 41.8 4.1 C 4.1 12 8.1 28 18.8 8 6.1 87 sa.B R e f i n e d C e r e a l s ( M i d d l e Y e a r s ) 148 IOO.O 5 DUT OF 12 ( 41. 7%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY - 2.514 RAtj CHI SOUARE - 2.33873 WITH 5 DECREES OF FREEDOM. SIGNIFICANCE • 0.8OO6 PAJMBER OF MISSING OBSERVATIONS • COUNT ROW PCT COL PCT lcasas I C R 0 S S T A ROW TOTAL U L A T I 0 N BT GROUP AGE I OF 1 TOT PCT I t I 3 I 1 I 31 I 18 I 40 a no 1 S3. 5 I 47.5 I 31.6 I 30.6 I 22.8 I I 11.4 I 10.3 I 2 I 39 I 29 I 58 a MO I 50.0 I SO.O I 31.4 I 38.4 I 34.9 I 1 15.7 I 15.7 I a I 37 I 34 I 51 a wk I 53.8 1 47. 1 I 37.6 I 36.5 I 28.9 I I 14.6 I 13.0 I -I-4 1 34 1 10 I 34 •3 I 70.6 I 39.4 1 18.4 1 23.S 1 13.0 I I 13.0 1 9.4 I -I- -I S I 1 I 1 I 3 day I SO.O I SO.O I 1. 1 I l.O 1 i.a 1 I O.S I 0.9 I -COLUMN 103 83 185 TOTAL SS. 1 44.8 100.0 F r i e d F o o d s ( M i d d l e Y e a r s ) 3 OUT OF lO ( 30.OX) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0 MINIMUM EXPECTEO CELL FREOUENCT - 0.897 RAW CHI SOUARE - 4. 13343 WITH 4 DEGREES OF FREEDOM. SIGNIFICANCE • 0.3S83 NUMBER OF MISSING OBSERVATIONS • 18 259 . . . . . . . . . . . . . . . . . . C R O S S T A B U L A T ION OF . . . MFSMK. BY GROUP GROUP COUNT 1 ROV PCT leases controls ROW COL PCT 1 TOTAL TOT PCT I I I 2 1 MF SMK 1 1 1 I I 0 1 I I I ones A so .1 0.0 I 100.0 I BO.O _ i o.o i 100.0 i Smoked Foods I O.O I BO.O I -I I I 3 1 1 1 0 1 1 one* A wk i IOO.O i o.o i so.o (Middle Years) I IOO.O I 0.0 I I SO.O I 0.0 I -I — I I COLUMN 1 1 3 TOTAL SO.O 80.0 1O0.0 4 OUT OF 4 (1OO.0X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.500 FISHER'S EXACT TEST • O.SOOOO (1-TA1LE0) t.OOOOO (3-TAILEO) NUMBER OF MISSING OBSERVATIONS • 301 C R O S S T A B U L A T I O N BY GROUP PAGE 1 OF MFCHAR Ones a GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 controls I 3 I I 13 o I 61.9 I 86.7 I 4B.1 -I i i a I 33.3 I 13.3 I 7.4 -I COLUMN IS TOTAL 55.6 8 38. 1 66.7 39.S 66.7 33.3 14.8 ROW TOTAL 13 44.4 31 77.8 6 33.3 C h a r c o a l B r o i l e d Foods (Middle Years) 37 IOO.O M,^.i.2"J °l 4 ' 5 O 0*> 0 F T H E "*<•"> CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCY • 3.66T CORRECTED CHI SOUARE - 0.60268 WITH I DEGREE OF FREEDOM. SIGNIFICANCE - O 4376 RAW CHI SOUARE . 1.S4386 WITH 1 DEGREE OF FREEDOM. SIGNIFICANCE • O 3143 NUMBER OF MISSING-OBSERVATIONS 178 C R O S S T A B U L A T I O N BY GROUP P i c k l e d Foods GROUP COUNT I ROW PCT Icaaaa control a ROW COL PCT I TOTAL TOT PCT I I I 3 1 1 1 1 I I I I 0 1 1 mo 1 IOO.O I O.O I 50.0 I IOO.O I O.O I I SO.O I 0.0 I -1 1 1 - , M 4 i tl I i *°° (Middle Years) I O.O I SO.O I -I I I COLUMN 1 1 3 TOTAL SO.O 50.0 100.0 4 OUT OF 4 (1OO.0X) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY - 0.500 FISHER'S EXACT TEST - O.5O0O0 (l-TAILEO) 1.OOOOO (3-TAILEO) PAGE 1 OF NUMBER OF MISSING OBSERVATIONS • 301 260 C • O S S U I U L I I I O N O F • • • » • • • • » • • • • • • • • BY OROUP • • • • • • • • • • • • • • • • • • a * * * * * * * * * * * PAGE I OF COUNT ROW PCT COL PCT TOT PCT « Icaaw* 1 COLUMN TOTAL 94 S4.S S3.8 33.7 a s 53.9 ROW TOTAL 1 I 2 9 I 4 s o . o I 40.0 * 7.0 I 9.4 3.7 I 2.5 3 t S 37. B I 62.8 3.5 I 6,9 1.9 I 3. 1 B 1 3 62.5 I 37.8 5.9 ! 4. 1 3. 1 1 1.9 12 I 12 s o . o I 50.0 14.0 1 16.2 7.5 I 7.5 s • 54.5 7.0 3.7 8 4S.S 6.8 3.1 4S 45.3 60.8 28. 1 46 10 5.3 8 S.O 8 S.O M i l k (Middle Years) i i 6.9 160 100.0 5 OUT OF 12 ( 41.7X) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 3.0. MINIMUM EXPECTEO CELL FREOUENCY - 3.700 RAW CHI SOUARE - 1.41706 WITH 5 DECREES OF FREEDOM. SIGNIFICANCE - 0.9224 NUMBER OF MISSING OBSERVATIONS • 43 C R 0 S S T A A T I 0 N BY GROUP PAGE I OF 1 COUNT ROW PCT COL PCT IC4ISM4 I ROW TOTAL TOT PCT I 1 I 3 I 1 I 3 I 1 I • 4 I 75.0 I 3S.0 1 3.3 I 3.0 I 1.3 I I 1.7 I 0.6 I -I-2 I 4 I 3 I 6 I 66.7 1 33.3 I 3.4 I 4.0 I 3.6 I 1 2.2 I 1. 1 I -I-3 I 5 I 3 I 7 I 71.4 I 28.6 I 3.8 I S.O I 3.6 I I 2.8 I 1. 1 I -I-4 I S I 6 I 11 I 45.3 I 54.5 I E.3 t S.O 1 7.8 I 1 2.8 1 3.4 I -1-8 I 3 I 9 I 14 I 35.7 I 64.3 ! 7.9 I 3.0 I 11.7 I I 3.8 I S. 1 I S I 79 I 37 I 136 I SB. 1 I 41.9 I 76.4 I 78.2 I 74.0 I t 44.4 I 33.0 I COLUMN lOI 77 179 TOTAL 86.7 43.3 100.0 Coffee (Middle Years) 7 0U7 OF 13 < 58.3X) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5 O MINIMUM EXPECTED CELL FREOUENCT - 1.730 «•-««.""-i l"«N 3.O. RAW CHI SOUARE - 4.S92SO WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE - 0.4676 NUMBER OF MISSING OBSERVATIONS - 35 261 C IO S S T 1 1 U L I ! I ON OF . . . . . . . . . . . . . . . . . BY GROUP • • • « • . . . » « • • • • • • • • « • . . . . . . . . . . p A 0 £ , 0 F CROUP COUNT I BOW PCT Icasss controls ROW COL PCT I TOTAL TOT PCT I I I a I -I I a I a t « 1 i 5 a . ? ! ' •* Tea (Middle Years) 1 1.T I l i l I -I I I a i 4 i I I s i ao .o i ao .o i a.B I 4.2 1 1.2 I I i . a i o . s i -I 1 1 a t o i a i 2 I 0.0 I IOO.O I 1.1 I 0.0 I 2.S I I 0.0 t 1.1 I -I I I 4 ! 5 1 4 1 9 I 53.S I 44.4 I 5.1 I 3.2 I 4.9 I 1 2.B I 2.3 I - i i : 5 1 B 1 B I 10 I 50.0 I 50.0 I 3.6 1 5.2 1 6.2 I 1 2.8 1 2.8 I -1 1 1 6 1 79 I 66 I 145 I 84.B 1 45.5 1 81.9 I 62.3 ! 81.5 I I 44.6 t 37.3 I -I 1 — J COLUMN 96 81 177 TOTAL 54.2 45.8 IOO.O 8 OUT OF 12 ( 75.0%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 3 0. MINIMUM EXPECTEO CELL FREOUENCY - 0.913 RAW CHI SOUARE - 3.83296 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • O.S73T NUMBER OF MISSING OBSERVATIONS - 36 C R O S S T A S U L A T I O N BY GROUP PAGE 1 OF COUNT ROW PCT COL PCT TOT PCT Icasaa I I 1 COLUMN TOTAL 10 76. S 17.8 13.3 9 69.3 16. 1 11.1 7 43.8 12.5 8.6 17 77.3 30.4 21.0 4 66.7 7.1 4.9 control• 9 s i . a 16. 1 11.1 56 69. 1 3 23. 1 12.0 3.7 4 30.8 16.0 4.S 8 56.3 36.0 11.1 I I I I I 2 I 1 33.3 1 I 8.0 I I 2.9 1 I 1 I 2 1 I 18.2 I I 8.0 I I 2.S I 25 30.9 ROW TOTAL I I I I I I I I I 1 I I 1 I I I 1 1 I 9 I I 22.7 I 20.0 I 6.2 13 16.0 13 16.0 16 19.8 22 27.2 Soda Pop (Middle Years) i i 13.6 81 100.0 6 OUT OF 12 < 50.0%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 3.0. MINIMUM EXPECTED CELL FREOUENCY - 1.852 RAW CHI SOUARE - 6.73061 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.241S NUMBER OF MISSING OBSERVATIONS - 122 262 C I 0 S S ! < I « l < T I 0 N BY OROUP PAGE 1 OF 1 GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 I I 14 I 70.0 I 58.S I 17. 1 -I I 2 I 28.6 1 4. 1 I 2.4 -I I 7 I 41.2 14.3 S . S 14 73.7 28.6 17.1 3 SO.O S.I 3.7 COLUMN TOTAL 1 69.2 I 18.4 I 11.0 -J 49 59.8 I S I 30.0 ! 18.2 I 7.3 3 SO.O 9.1 3.7 4 30.3 12. 1 4.9 ROW TOTAL I I I I i I I S I 71.4 I 15.2 I 5. 1 I 1 10 I SB.a i 30.3 I 12.2 ! 1 S I 26.3 13.2 6. 1 20 24.4 7 8.3 17 20.7 19 23.2 6 7.3 13 15.9 Beer (Middle Years) 33 40.2 82 100.0 4 OUT OF 12 ( 33.3%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 2.415 RAW CHI SOUARE • 8.39918 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE • 0.13S6 NUMBER OF MISSING OBSERVATIONS -C R O S S T A B U L A T I O N OF BY GROUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . p i 0 E t Qf GROUP COUNT I ROW PCT Icaaaa controls ROW COL PCT I TOTAL TOT PCT I I I 2 1 -I I 14 I 8 1 22 I 63.S I 36.4 I 26.8 [ "t { lit J Wine (Middle Years) I 17.1 I 8.8 I -I 1 1 2 1 9 1 8 1 17 I 32.8 I 47.1 I 20.7 I 47.6 I 2S.8 I I 11.0 I 9.8 I -! 1 , 3 1 6 1 4 1 10 I SO.O I 40.0 I 12.2 1 11.8 I 12.9 I I 7.3 I 4.9 I -I 1 1 4 1 16 I 9 1 24 I 66.7 I 33.3 I 29.3 I 31.4 I 25.8 I I 19.S I 9.8 I - I — I I 9 1 2 1 0 1 2 I IOO.O I 0.0 I 2.4 I 3.9 I 0.0 I 1 2.4 I 0.0 I -I 1 I S I 4 1 3 1 7 I 37.1 I 42.9 I 8.3 I 7.8 I 9.7 I I 4.9 I 3.7 I -I 1 -I COLUMN 31 31 82 TOTAL 62.2 3T.6 1O0.0 3 dOT OF 12 ( 41.7X) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0 MINIMUM EXPECTED CELL FREOUENCY - 0.7SS RAW CHI SOUARE - 2.1S485 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE - 0.8273 NUMBER OF MISSING OBSERVATIONS - 121 263 C R O S S T A S U L A T I O N BY OROUP GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I I COLUMN TOTAL 12 60.0 18.S 11.0 a 93.3 13.3 7.3 10 50.0 15.4 9.2 17 68.0 26.2 15.6 3 60.0 4.6 2.8 19 62.8 23. 1 13.8 65 59.6 6 40.0 18.3 7.3 I 7 I 46.7 I 15.9 I 6.4 -I I 10 I 50.0 I 33.7 I 9.2 8 32.0 18.2 7.3 2 40.0 4.5 1.8 I 9 I 37.9 t 20.9 I 8.3 44 ROW TOTAL 20 18.3 19 13.8 20 18.3 25 32.9 L i q u o r ( M i d d l e Y e a r s ) 5 4.6 24 22.0 109 100.0 2 OUT OF 12 ( 16.7%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCT - 2.018 RAW CHI SOUARE • 1.82867 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.8723 NUMBER OF MISSING OBSERVATIONS -C R 0 S S T A B U L A T BY I 0 N GROUP COUNT I ROW PCT Icaaaa COL PCT I ROW TOTAL TOT PCT I 1 I 2 I 1 I 8 I 3 I 12 I 75.0 I 35.0 1 37.8 I 40.9 I 30.0 I I 38. 1 I 8.4 I 2 I 5 I 2 I 7 I 71.4 I 28.6 I 21.8 I 22.7 I 20.0 I I 15.6 I 6.3 I 3 I B I 2 I 7 I 71.4 I 38.6 I 21 .9 I 22.7 1 20.0 I I 15.6 1 6.3 I 4 1 3 I 2 I 5 I 60.0 1 40.0 I 15.6 I 13.6 I 20.0 I I 9.4 1 6.3 I 5 I 0 I 1 I 1 I O.O I IOO.O I 3. 1 I 0.0 I 10.0 I I 0.0 1 3. t I COLUMN 22 10 32 TOTAL 6S.8 31.3 100.0 L i q u e r s ( M i d d l e Y e a r s ) 9 OUT OF 10 ( 90. OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 9 O MINIMUM EXPECTED CELL FREOUENCY • 0.313 RAW CHI SOUARE - 3.64312 WITH 4 DEGREES OF FREEDOM. SIGNIFICANCE - 0.6192 NUMBER OF MISSING OBSERVATIONS • 171 264 ( 1 0 S SI I I II l 1 T I Q N BT CROUP COUNT ROW PCT COL PCT TOT PCT Icaaai 1 I t AVGM .lo COLUMN TOTAL RAW CHI SQUARE • 11 AS.a a.a i n 59.0 ROW TOTAL 24 I 15 81 .5 I 38.5 21.6 I 16.5 11.9 I 7.4 76 I 63 54.7 I 45.3 68.5 I 68.2 37.6 I 31.2 1 13 I 54.3 I 14.3 I 6.4 91 45.0 39 19.3 139 69.9 24 11.9 V i t a m i n A Rich Vegetables (Middle Years) .49387 WITH 202 100.0 2 OEGREES OF FREEOOM. SIGNIFICANCE • 0.4738 NUMBER OF MISSING OBSERVATIONS • C R O S S T A B U L A ; 0 N GROUP PAGE 1 OF GROUP COUNT 1 ROW PCT Icaatl COL PCT I TOT PCT I I CVGM t o COLUMN TOTAL 38 77.6 34.3 ia.7 2 14.3 i.a i . o i n 54.7 71 SO. 7 64.0 35.0 ROW TOTAL 2 t -I 11 I 22.4 I 12.0 1 5.4 I 1 69 49.3 75.0 34.0 I 12 1 95.7 ! 13.0 I I 5.9 1 I 1 92 45.3 49 24. 1 140 69.0 14 6.9 303 100.0 V i t a m i n C R i c h Vegetables (Middle Years) RAW CHI SOUARE - 30.44 990 WITH 3 OEGREES OF FREEDOM. SIGNIFICANCE GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 C R 0 S S T A ROW TOTAL U L A T I 0 N BT GROUP PAGE 1 OF :RVCM 1 I 37 1 48 I SS lo I 43.S I 56.5 I 41.9 1 33.3 1 53.7 I 1 19.3 I 33.6 I -I-2 I 73 I 43 I 116 awd I 63.9 I 37. 1 I 37. 1 1 65.9 I 46.7 I I 36.0 I 31.3 I 3 I 1 I 1 I 3 hi I SO.O I SO.O 1 1.0 I 0.9 I 1.1 I I 0.5 I O.S I COLUMN 111 93 203 TOTAL 54.7 4S.3 100.0 3 OUT OF 6 ( 33.3X) OF THE VALID CELLS C r u c i f e r o u s Vegetables (Middle Years) MINIMUM EXPECT EO CELL FREOUENCT -RAW CHI SOUARE • 7.46926 WITH .906 2 DEGREES OF FREEDOM. SIGNIFICANCE • 0.0239 265 C R O S S T A B U L A T I 0 N BY GROUP COUNT ROW PCT COL PCT leases I ROW TOTAL FATH TOT PCT I 1 I 2 I 1 1 10 I 8 I 18 lo ! 55.6 I 44.4 I 8.8 1 s.o I 8.7 I I 4 .9 I 3.9 I •I- -I 2 I 78 1 88 I 147 m l I S3.7 I 48.3 I 72.4 I 71.2 I 73.8 I I 38.8 I 33.S I -I-hi 3 I 22 1 IS I 38 I S7.8 1 42. 1 1 18.7 I 19.8 I 17.4 t I 10.8 I 7.9 I I- -t COLUMN 111 92 303 TOTAL 54.7 45.3 100.0 Fat (Middle Years) RAW CHI SOUARE • O.21629 WITH 2 DEGREES OF FREEDOM. SIGNIFICANCE • 0.8975 . . . . . . . . . . . . . C R O S S T A S U L A T I O N OF BY GROUP • ...«.....»••••.••••••••........,,...,,,,, p(l(.E ( Q f 1 GROUP COUNT 1 ROW PCT Icasas controls ROW COL PCT I TOTAL TOT PCT I 1-1 2 1 1 I 7 I 3 I 10 lo I 70.0 I 30.0 I 4.9 I 6.3 I 3.3 I I 3.4 I 1.5 I 2 I 87 1 76 I 163 I 53.4 I 46.6 I SO. 3 I 78.4 I 82.6 I I 42.9 I 37.4 I --' 3 I 17 I 13 I 30 hi I 56.7 I 43.3 I 14.8 I tS.3 I 14. 1 I I 9.4 I 6.4 I COLUMN 111 S3 203 TOTAL 54.7 45. 3 100.0 Meat (Middle Years) 1 OUT OF 6 < 16.7X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5 O MINIMUM EXPECTED CELL FREOUENCY -4.532 RAW CHI SOUARE - 1.10704 WITH 2 DEGREES OF FREEOOM. SIGNIFICANCE - 0.5749 C R O S S T A B U L A T I 0 N GROUP PAGE 1 OF 1 GROUP COUNT I ROW PCT Icasas COL PCT I COLUMN TOTAL RAW CHI SOUARE • 111 54.7 92 45.3 ROW TOTAL PCT I 1 I 2 1 I 49 I 41 I 54.4 I 45.6 1 44. 1 1 44.6 1 24. 1 I 20.2 2 1 59 I 41 1 59.0 1 41.0 I 53.2 I 44.6 I 29. 1 I 20.2 3 I 3 I 10 I 23. 1 I 76.9 I 2.7 I 10.9 I 1.5 I 4.9 1O0 49.3 13 6.4 Vegetable P r o t e i n s (Middle Years) 5.99453 WITH 203 100.0 2 DEGREES OF FREEOOM. SIGNIFICANCE • 0.0499 266 C R O S S T A B U L A T I 0 N BT CROUP COUNT ROW PCT COL PCT TOT PCT I ICaseS t COLUMN TOTAL S3 • 3 7.2 3.S 111 S4.7 7 46.7 7.6 3.4 92 43.3 ROW TOTAL 1 1 2 27 I 27 SO.O I SO.O 24.3 1 29.3 13.3 I 13.3 76 I 58 56.7 1 43.3 6B.S 1 63.0 37.4 I 36.6 54 36.6 134 66.0 203 1CO.O F i b e r R i c h Foods (Middle Years) RAW CHI SQUARE • 0.71249 WITH 2 DEGREES OF FREEDOM. SIGNIFICANCE - 0.7O01 C R O S S T A B U L BFATM lO GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 controls COLUMN TOTAL T 70.0 6.3 3.4 3 30.0 3.3 1.3 • 63 I 61 s o . a I 49.3 56. a I 66.3 31 .0 1 30.0 41 I 28 59.4 I 40.6 36.9 I 30. 4 30.3 I 13.8 111 54.7 92 45.3 A T BT I O N GROUP PAGE 1 OF ROW TOTAL 10 1.9 134 SI . 1 69 34.0 Beef Fat (Middle Years) 303 IOO.O 1 OUT OF 6 ( 16.7X> OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 3.0. MINIMUM EXPECTED CELL FREQUENCY • 4.332 RAW CHI SOUARE • 2.32356 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.3138 2 6 7 C P O S S T A GROUP COUNT • ROW PCT Icain COL PCT 1 TOT PCT I t one* • wk S day dally COLUMN TOTAL 1 IS.7 1.0 O.S s 83.3 S.S S.S I 8 I B 61.B 1 38.S 7.S ! S.S 4.2 1 2.6 26 I 20 S6.5 I 43.S 24.8 I 23.S 13.7 I 10.5 52 54.2 49.5 27.4 45.8 51.8 33.2 11 I 7 61.t I 38 .8 IO.S I 8. .2 1 5.8 I 3. .7 I 7 ' 6 3 . r S.7 3.7' lOS 55.3 85 44.7 1 0 N GROUP PAGE 1 OF 1 ROW TOTAL 6 3.2 13 6.8 46 34.3 96 SO.5 18 9.5 11 S.S Yellow Vegetables (Teen Years) ISO 10O.O 3 OUT OF 13 ( 35.OX) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY . 3.684 RAW CHI SOUARE • 4.45946 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE - 0.4853 NUMBER OF MISSING OBSERVATIONS • 13 C R 0 S S T A A T I 0 N BY GROUP GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT 1 faw a mo I I 1 I -I I I I I -I 3 I rk I I I -I 4 I I I -I 5 I day I I I -I COLUMN TOTAL 1 I 3 8 I 9 50.0 I SO.O 16.7 I 24.2 9.9 I 9.9 t o I 5 66.7 1 33.3 20.8 I 15.3 13.3 I 6.3 20 I 12 G3.S I 37. S 41.7 I 36.4 34.7 I 14.9 9 I 7 56.3 I 43.9 19.9 I 21.2 11.1 I 9.6 1 1 1 50.0 I 50.0 3. 1 1 3.0 1.3 I 1.2 48 S9.3 33 40.7 PAGE I OF ROW TOTAL 16 19.8 32 39.5 16 19.8 3 3.5 B r o c c o l i / B r u s s e l ' s Sprouts (Teen Years) 3 OUT OF lO ( 30.OX) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCT • 0.815 RAW CHI SOUARE - 1.17933 WITH 4 DEGREES OF FREEOOM. SIGNIFICANCE • 0.8815 NUMBER OF MISSING OBSERVATIONS • 133 268 C R O S S T A B U L A T I O N OF • BY GROUP •• PAGE 1 OF I GROUP COUNT I ROW PCT lc.1.1 COL PCT 1 TOT PCT I 1 one* • mo onca a wk dally I 1 I I I -I 3 I I I 1 -I 3 I 1 I I -I 4 I I I I -I 5 I day I I I -I-e i i COLUMN TOTAL 3 50.0 3.a l.S o O.O 0.0 O.O IOS 56.8 ROW TOTAL 3 I 4 43.8 I 57.1 3.8 I 4.9 1.6 I 2.1 13 I 12 52.0 I 48.0 12.0 I 14.6 6.8 1 6.3 39 I 32 54.8 t 45. 1 36. 1 I 39.0 30.5 I 16.8 50 I 29 63.3 I 36.7 46.3 I 35.4 36.3 I 15.3 I 3 1 SO.O I 3.7 I 1.6 -I I 3 I IOO.O 1 2.4 I 1.1 82 43.2 7 3.7 35 13.2 71 37.4 78 41.6 6 3.2 2 1.1 Cauliflower/Cabbage (Teen Years) ISO 100.0 6 OUT OF 12 ( SO.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCT • 0.863 RAW CHI SOUARE • 4.99083 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.4170 NUMBER OF MISSING OBSERVATIONS • 13 C R O S S T A B U L A COUNT ROW PCT COL PCT Icaaaa I T I O N BT GROUP PAGE I OF ROW TOTAL TFLG TOT PCT I 1 I 2 I 1 I 11 { 8 I 19 onca a mo I 57.9 I 42. 1 I 12.2 I 13.3 I 11.0 I I 7.1 I 5. 1 I -I-2 I 11 I 10 I 21 faw • m o 1 52.4 I 47.6 I 13.5 I 13.3 I 13.7 I I 7.1 I 6.4 I -I-3 I 31 I 18 I 39 onca a i «k I 63.8 I 46.2 I 35.0 I 25.3 I 24.7 I I 13.S t 11.5 1 -I-4 1 30 I 25 I 55 wk 3-3 I 54.5 1 45. 8 1 35.3 1 36. 1 I 34.3 I I 19.2 I 16.0 I -I-8 I a I 11 I 19 •oat day I 42. 1 I 57.9 I 13.2 I 9.6 I 15. 1 I I 5. 1 I 7.1 I -I -6 I 2 I 1 I 3 dally I 66.7 1 33.3 I 1.9 I 2.4 I 1.4 I 1 1.3 I 0.6 I - I -COLUMN 83 73 156 TOTAL 53.2 46.8 100.0 Leafy Green Vegetables (Teen Years) 3 OUT OF 12 ( 16.7X) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 9.0. MINIMUM EXPECTEO CELL FREQUENCY • 1.404 RAW CHI SOUARE - 1.37827 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.9267 NlA'BER OF MISSING OBSERVATIONS • 47 269 C R 0 S S T A 8 U L A T er I o N GROUP PAGE 1 OF one* a mo COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 9 64 .3 9.9 4.9 onca a wk dally 8 . day I I t 1 -I-I I I I -I-I 1 I I -I I I I I I I I I I -I I I I I -I-COLUMN TOTAL 14 83.9 13.7 7.6 41 56.9 40.2 22.3 I 50.0 1.0 O.S 5 35.7 6. I 2.7 46.2 14. C 6.8 31 43. I 37.8 16.8 SO.O 1.2 O.S 102 55.4 82 44.6 ROW TOTAL I I t I I I 1 12 I 33 I 33 50.0 I s o . o 32.4 I 40.3 17.9 I 17.9 4 I 0 IOO.O I 0.0 3.9 I 0.0 2.3 I 0.0 14 7.6 36 14. 1 73 39. 1 66 35.9 4 2.2 3 1. I F i b r o u s Vegetables (Teen Years) 184 IOO.O 4 OUT OF 13 ( 33.3X) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTED CELL FREOUENCT • 0.891 RAW CHI SOUARE - 4.56S61 WITH 5 OEGREES OF FREEOOM. SIGNIFICANCE • 0.4713 NUMBER Of MISSING OBSERVATIONS • IB C R O S S T A B U L A I 0 N GROUP 'one m onca a wk GROUP COUNT ! ROW PCT leases COL PCT I TOT PCT I 1 I I I I t -1 3 I I I I -I 3 I 1 I 1 -I 4 I I 1 I -1 5 I day I I dally COLUMN TOTAL 9 I 4 66.7 I 33.3 13. I I 10.0 7.9 I 4.0 S I 4 53.6 I 44.4 8.3 I 10.0 5.0 I 4.0 13 I 4 76.5 I 33.5 31.3 I 10.0 13.9 I 4.0 17 I 13 58.6 I 41.4 37.8 I 30.0 16.8 I 11.9 11 I 7 61.1 I 38.9 18.0 I 17.3 lO.8 1 6.9 7 I 9 43.8 I 56.3 11.5 I 22.9 . 6.9 I 8.9 61 40 60.4 39.6 PAGE I OF I ROW TOTAL 12 11.9 9 9.9 17 16.8 29. 28.7 19 17.8 IE IS.8 Peanuts/Peanut B u t t e r (Teen Years) 101 100.0 2 OUT OF 12 < 16.77.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCT • 3.564 RAW CHI SOUARE - 4.01745 WITH S OEGREES OF FREEDOM. SIGNIFICANCE • 0.S469 NUMBER OF MISSING OBSERVATIONS • 103 270 COUNT I BOW PCT Icaaaa COL PCT I C R O S S T A B U L ROW TOTAL A T I 0 N BT GROUP PAGE i o r onca a WW dally TOT PCT I 1 I 3 I 1 1 11 I 8 I 20 a ao I BS.P I 45.0 I 13.3 I 13.4 I 13.0 I 1 T . 3 1 8.0 1 -1 -2 I 17 I 20 1 37 • "O 1 49.S 1 54. 1 I 34.9 I 30.7 1 28.0 I 1 I t . 3 I 13.3 1 - I -3 I 3 0 1 22 1 S3 ww 1 57.7 1 43.3 I 34.4 I 36.6 I 31.8 I 1 IB. a | 14.S I - 1 -4 I i s I IS I 31 - 3 I BI.S I 48.4 I 30.8 I IB .9 ! 31.7 I I lO.S I 8.8 I - 1 -B I 7 1 1 I a clay 1 87.5 1 12.8 I 5 . 3 I a . 9 I 1.4 I 1 4.S I 0.7 ! -:- - I - - I c 1 1 I 3 I 3 r I 33.3 I 66.7 1 3.0 I 1.3 I 2.8 1 J 0.7 1 1.3 I - I - - • I COLUMN 83 68 151 TOTAL 54.3 45.7 IOO.O D r i e d Beans /Peas (Teen Years) 4 OUT Of 12 < 33.3%) Of THE VALID CELLS HAVE EXPECTEO CELL fREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL fREOUENCT • 1.371 BAW CHI SOUARE - S. 46087 WITH 5 OEGREES Of fHEEOOM. SIGNIFICANCE - 0.3623 NUMBER or MISSING OBSERVATIONS • S3 C R 0 5 S T A B U L A T I O N BT GROUP CROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I I ROW TOTAL onca a MO oally 9 oay COLUMN TOTAL 1 6 I 8 1 40.0 I 60.0 S.E I 10. 1 t 3. 1 I 4.6 16 I 15 51 .6 I 48.4 IS.O I 16.8 8.3 I 7.7 28 I I S 45.1 I S4.7 27. 1 I 19.1 14.8 I . , 7 - B SO I 26 65.8 1 14.2 46.7 1 29.2 2S.9 I 13.3 5 1 2 71.4 I 28.6 4.7 1 2.3 2.6 I 1.0 : 1 I 3 13.3 I 66 .7 1 0.9 I 3. .3 ] 0.9 1 1. . o : 107 94.6 89 45.4 15 7.7 31 15.8 64 33.7 76 38.8 7 3.6 1 t . S F i s h / P o u l t r y (Teen Years) 196 IOO.O 4 OUT O r 13 ( 33.TX) Or THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN S.O. MINIMUM EXPECTED CELL FREOUENCT • 1.363 RAW CHI SOUARE • 8.81402 WITH 5 OEGREES OF FREEDOM SIGNIFICANCE - 0.1167 NUMBER OF MISSING OBSERVATIONS • 7 271 C t O S S 1 I I U L I I 0 N CROUP COUNT ROW PCT COL PCT TOT PCT I onca a ao onca a wk dally I I I I t a i i I i - i a i i l i - i 4 I t I I -I B I day I 1 I -I 6 I I I ! -r f 1 a a I 4 43.9 I 97. 1 3.9 1 4.9 t.S 1 3.1 4 ! S 40.0 1 SO.O a .s I 7.1 3.1 I 3.2 3S I 11 70.3 I 39.7 34.9 I 13. 1' 13.9 I 5.9 93 I 42 55.9 I 44.2 50.5 I 50.0 38.0 I 22.3 14 I 19 43.9 I 56.3 13.3 1 31.4 7.4 I 9.5 COLUMN TOTAL s 63.5 4.9 2.6 105 55.6 3 37.5 3.6 1.6 94 PACE I OF ROW TOTAL 7 3.7 10 5.3 37 19.6 95 50.3 32 16.9 9 4.2 Beef (Teen Years) 199 100.0 5 OUT OF 12 ( 41.7X) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 3.111 RAW CHI SOUARE • 6.64634 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE • 0.3483 NUMBER OF MISSING OBSERVATIONS • C R 0 S S T A B U L A T I 0 N BY GROUP PAGE I OF 1 GROUP COUNT I ROW PCT leases COL PCT I TOT PCT ROW TOTAL onca * MO onca a wk tk 2-3 dally 1 I I I I -I 2 I 1 I I -I 3 I I I I -I-4 1 I I I -I-5 I day I I I -I-S I I I I -I-[ 1 I 3 12 I 6 66.7 I 33.3 13.5 I 9.3 7.1 I 3.6 14 I 7 66.7 I 33.3 14.6 I 9.6 8.3 I 4. 1 37 I 35 51.9 1 48. 1 39. 1 I 34.3 16.0 I 14.8 34 I 38 54.0 I 46.0 35.4 I 39.7 30. 1 I 17.3 6 I 3 66.7 I 33.3 6.3 I 4. 1 3.6 I I.S 3 I 3 SO.O I s o . o 3. 1 I 4.1 1.9 I 1.8 COLUMN TOTAL 96 56.8 73 43.3 18 10.7 31 12.4 Cured Meat (Teen Years) 63 37.3 9 5.3 6 3.6 169 IOO.O 3 OUT OF 13 ( 25.07.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 2.592 RAW CHI SOUARE • 2.73743 WITH S DEGREES OF FREEDOM. SIGNIFICANCE • 0.7419 NUMBER OF MISSING OBSERVATIONS • 34 272 GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 1 onca a a>o onca a wk dally I I I I -I 2 I I 1 I -I 3 I I I I -I 4 I I I I -I 5 I day I I I -I-6 I I I I -I-COLUMN TOTAL 3 33. 1 3.2 i . a 7 77.8 7.8 4.5 83 59.S C R O S S T A B U L * 15 I 8 82.5 1 37.5 16. 1 I 14.3 9.6 I 5.8 12 I 3 85.7 I 14.3 12.9 I 3.3 T.7 I 1.3 24 I 19 55.8 I 44.3 25.8 1 30.2 15.4 I 12.2 33 I 21 60.4 I 39.6 34.4 I 33.3 20.5 I 13.5 10 76.9 15.9 6.4 2 22.2 3.2 1.3 63 40.4 PAGE 1 OF 1 ROW TOTAL 34 15.4 14 9.0 43 27.6 53 34.0 13 8.3 8 5.8 Cheese'(Teen Years) 156 100.0 1 OUT OF 12 ( 6.3%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCT • 3.635 RAW CHI SOUARE - 12.7S683 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0.0258 NUMBER OF MISSING OBSERVATIONS C R p S S T A B U L A T BY I 0 N CROUP PAGE 1 OF 1 TFCCGS GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 ROW TOTAL onca a ao dal ly 1 3 60.0 3.8 1.5 2 40.0 1.9 I .0 5 day COLUMN TOTAL 2 I I 2 I 40.0 I 2.3 I 1 .0 I 1 3 I 60.0 3.4 1.5 -I-3 I IS I 11 onca a wk I 63. 1 I 37.9 I 16.7 I 13.6 I 8.3 I 5.6 -I-4 I 49 I 39 wk 3-3 I 55.7 I 44.3 I 45.4 I 44.8 I 35. 1 I 30.0 34 I 14 63.2 I 36.8 22.2 I 16. 1 13.3 I 7.3 12 I IS 40.0 I 60.0 11.1 I 20.7 6.2 I 9.2 108 87 55.4 44.6 5 2.6 5 2.6 29 14.6 88 45. I 38 IS.5 30 15.4 Eggs (Teen Years) 195 100.0 4 OUT OF 12 ( 33.3%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 3.231 RAW CHI SOUARE • 4.SS233 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE - 0.4343 NUMBER OF MISSING OBSERVATIONS - 8 273 C R O S S T A B U L A T I 0 N B' GROUP PAGE t OF 1 GROUP COUNT I ROW PCT Icasea COL PCT I TOT PCT I 1 ROW TOTAL one* a mo once a wk dally B day COLUMN TOTAL B I a s o . o I 30.0 13.3 I 4.3 7.5 1 1.9 11 1 6 G4.7 1 35.3 18.3 I 13.a 10.3 1 5.6 S 1 7 46.3 I 53.8 10.0 I 14.9 5.6 1 6.S 18 t 16 53.8 I 47. 1 30.0 1 34.0 16.8 I 15.0 8 I 7 56.3 I 43.8 15.0 I 14.8 8.4 I 6 .B a I 9 47. 1 1 53.9 13.3 I 19. 1 7.5 I 8.4 60 47 56. 1 43.9 10 8.3 17 15.9 13 13. I 34 31 .8 16 15.0 17 i s . a C i t r u s F r u i t / J u i c e (Teen Years) 107 IOO.O 1 OUT OF 13 ( B.3X) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN S.O. MINI-MUM EXPECTEO CELL FREOUENCT • 4.393 RAW CHI SOUARE - 4.05438 WITH S OEGREES OF FREEOOM. SIGNIFICANCE • 0.S4I6 NUMBER OF MISSING OBSERVATIONS 96 • * * C R O S S T A B U L A T I O N OF . . . . . . . . . . . . . . . . . . TFTOM ..- Bv GROUP • p 4 C E 1 o r 1 GROUP COUNT I ROW PCT Icaaaa control a ROW COL PCT I TOTAL TOT PCT I I I 3 1 TFTOM 1 1 1 dnce a 1 I 7 I 8 I 15 ao I 46.7 I 53.3 I 8.8 I 7.1 I 11.1 I 1 4. 1 I 4.7 I 3 I 11 I 8 I 19 few x m o I 57.9 I 43. 1 I 11.2 I 11.3 I 11.1 I I 6.5 I 4.7 I -I- -I 3 I 37 I 18 I SS onca a I 67.3 I 32.7 I 33.4 I 37.8 I 35.0 I I 31.8 I 10.6 I 4 I 39 I 26 I 55 wk 3-3 I 53.7 I 47.3 I 33.4 I 39.6 I 36. 1 I I 17. 1 I 15.3 I -I- -I 5 I 13 I 8 I 30 noat day I 6O.0 I 40.0 I 11.9 I 13.3 I 11.1 I I 7. 1 I 4.7 I 6 I 2 I 4 I 6 dal ly I 33.3 I 66.7 I 3.5 I 3.0 I 5.6 I I 1.3 I 3.4 I -I- -I COLUMN 68 72 170 TOTAL 57.6 42.4 IOO.O Tomatoes/Juice (Teen Years) 3 OUT OF 13 ( 16.77;) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCY " 3.541 RAW CHI SOUARE • 4.87177 WITH 5 OEGREES OF FREEOOM. SIGNIFICANCE - 0.4317 NUMBER OF MISSING OBSERVATIONS • 33 274 C l t S S T I I U l l T I 0 N BY GROUP COUNT I ROW PCT leases COL PCT I ROW TOTAL T1PAST TOT PCT I 1 I 3 1 1 I 17 1 7 I 24 one* a i eO 1 70. B 1 29.2 1 18.9 I 21.3 I 11.3 I 1 12.0 1 4.» I 1 I 19 1 IS 1 34 few X KH 5 I 55.9 1 44 . 1 I 23.9 1 23.8 I 24.2 I I 13.4 I 10.8 1 3 1 30 I 30 1 60 nnce S « >k I 50.0 I SO.O I 42.3 1 37.5 1 48 4 I 1 31.1 1 31.1 I 4 I 14 I 10 1 74 wk 2-3 1 58.3 I 41.7 I 16.9 1 17.5 1 IS. 1 I 1 8.9 I 7.0 I COLUMN 80 82 142 TOTAL 58. 3 43.7 IOO.O Pasta- (Teen Years) RAW CHI SOUARE • 3.07158 WITH NUMRER Of MISSING OBSERVATIONS • 3 OEGREES OF FREEDOM. SIGNIFICANCE • 0.3807 C R O S S T A U L A T I 0 N BY GROUP PAGE 1 OF GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 a wk • d s l Ally day I COLUMN TOTAL 9 81.8 11.5 1 6.8 -I I T 1 35.0 1 8.0 I 5.3 -I 78 59.5 ROW TOTAL IT 1 8 68.0 I 33.0 21.8 I IS. 1 13.0 1 6. 1 11 I 6 57.9 I 42. 1 14. 1 1 15. 1 8.4 I 6. 1 18 I 11 63. 1 I 37.8 33. 1 I 20.8 13.7 I 8.4 16 I 11 59.3 1 40.7 20.5 I 20.B 12.2 I 6.4 2 18.2 3.8 1.5 I I I I I I I I I I I I I I I 13 I 65.0 24.5 8.8 S3 40.5 25 19. 1 19 14.S 29 22. 1 27 20.6 20 15.3 Ice Cream (Teen Years) 131 100.0 1 OUT OF 12 ( 8.37.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY - 4.450 RAW CHI SOUARE • 8.10811 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE • 0.1S04* NUMBER OF MISSING OBSERVATIONS > 72 275 C I 0 S S 1 I 1 II L I T I 0 N GROUP PAGE 1 Of 1 COUNT ROV PCT COL PCT TOT PCT Icaaes I I I onca a mo onca a wk dally 1 1 1 1 I -I 2 t 1 I I -J 3 I I 1 I -I 4 I I I I -I 5 I day 1 I I -I C I I I I -I COLUMN TOTAL 10 71.4 16.4 9.6 11 64.7 18.0 10.6 3 43.8 4.8 3.8 61 S8.7 28.6 9.3 3.8 6 39.3 14.0 S.8 S7.1 9.3 3.9 43 41.3 ROK TOTAL 16 I II 99.3 I 40.7 36.3 1 2S.6 19.4 I tO.6 14 I ' 14 SO.O 1 SO.O 33.0 1 33.6 13.S I 13.S 7 I 4 63.6 I 36.4 11.S I 9.3 6.7 I 3.8 17 16.3 37 36.0 38 26.9 11 tO.6 7 6.7 Salad Dressing/Mayonnaise (Teen Years) 104 100.0 3 OUT OF 12 ( 23.0*> OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THA MINIMUM EXPECTED CELL FREOUENCY • 2.884 RAW CHI SOUARE • 2.90047 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.7153 NUMBER OF MISSING OBSERVATIONS » C R 0 S S T A L A T I 0 N BY GROUP GROUP COUNT I ROW PCT Icaasa COL PCT I TOT PCT I 1 onca a MO onca a wk dal ly 1 I I I I -I 2 I I I I -I 3 I I I I -I 4 I I I I -1 5 I day I 1 I -I-C I I I I -I-[ 1 1 4 30.0 I 90.0 1.5 I S.9 0.7 I 3.0 3 I 0 100.0 I 0.0 3.0 I 0.0 1.5 I O.O 3 1 3 SO.O I SO.O 3.0 I 3.9 1.5 I I.S 16 I "6 73.7 I 37.3 33.9 1 9.9 11.9 I 4.4 14 I 14 50.0 I SO.O 30.9 I 30.6 10.4 I 10.4 33 I 43 43.3 I S6.9 47 .9 1 61.9 33.7 I 31. 1 COLUMN TOTAL 67 49.6 69 50.4 PAGE 1 OF ROW TOTAL 5 3.7 Whole Grains (Teen Years) 32 16.3 36 30.7 74 54.8 13S 100.0 6 OUT OF 13 ( 50.0%) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN S.O. MINIMUM EXPECTEO CELL FREOUENCY • 0.993 RAW CHI SOUARE • 9.68993 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.0845 NUMBER OF MISSING OBSERVATIONS • 68 276 C R O S S T A B U L A T I O N BY GROUP PAGE 1 OF 1 COUNT ROW PCT COL PCT TOT PCT GROUP I ICB«*A I I 1 1 once • eo once e wk ifldsx del ly I I I I - I 3 I I I I - 1 4 I I I I - I a I day I I I - I 6 I I I I - I COLUMN TOTAL 1 I 0 100.0 I 0.0 1.1 I 0.0 0.6 ! 0.0 1 I 1 S O . O I 50.0 1: t 1 l . S o:6 I O . S 7 t 8 46.7 1 S3. 3 7.4 I 13. 1 4.3 1 S . O 3 I S 33.3 I SS.7 3.3 1 8. 1 1.8 I 3.7 83 I 51 81.8 I 38. 1 87.4 I 77.3 S t .8 I 31.7 93 SS 39.0 41.0 ROW TOTAL 1 0.6 13 8.3 9 5.6 134 83.2 R e f i n e d C e r e a l s (Teen Years) 161 100.0 5 OUT OF 10 ( 50.0%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCT - 0.410 RAW CHI SOUARE • 4.63S24 WITH 4 OEGREES OF FREEDOM. SIGNIFICANCE • 0.3268 NUMBER OF MISSING OBSERVATIONS • C R O S S T A B U L A GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I I once a as once a wk - I 3 I day I I I - I COLUMN TOTAL [ 36 I 32 ! 63. 1 I 37.9 35.0 I 27.6 19.a I 12. 1 33 I 21 61.1 I 38.8 32.0 I 26.6 18. 1 I 11.5 22 I 23 48 .8 I 51 . 1 21.4 I 39. 1 12. 1 I 12.6 11 I 13 47.8 I 52.2 10.7 I 15.3 6.0 I 6.6 1 S O . O 1.0 O.B S O . O 1 . 3 O . S 103 56.6 78 43.4 T I 0 N BT GROUP PAGE 1 OF ROW TOTAL SB 31.8 54 29.7 45 24.7 23 12.6 F r i e d Foods (Teen Years) 182 100.0 2 OUT OF 10 ( 20.0%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.B68 RAW CHI SOUARE - 2.99899 WITH 4 DEGREES OF FREEDOM. SIGNIFICANCE • 0.5580 NUMBER OF MISSING OBSERVATIONS • 21 GROUP COUNT I ROW PCT Icatil controls ROW COL PCT I TOTAL TOT PCT I I I 2 1 TFSMK 1 J 1 I I 4 1 7 I II one* m mo I 36.4 I 63.6 I 100.0 I 100.0 I IOO.O I r, i j T-i j i 36.4 i 63.6 i Smoked Foods - i i i COLUMN 4 7 It TOTAL 36.4 63.6 100.0 / „, „ s (Teen Years) STATISTICS CA'INOT BE COMPUTED WHEN THE NUMBER OF NON-EMPTY ROWS OR COLUMNS IS ONE. NUMBER OF MISSING OBSERVATIONS • 192 . . . . . . . . . . . . . . . . . . C R O S S T A B U L A T I O N O F TFCHAR BT GROUP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . p 4 0 £ , Qf GROUP COUNT I ROW PCT leases ROW COL PCT I TOTAL TOT PCT I I I TFCHAR 1 1 1 I 1 I 1 once a »o i l o o . o i i c o . o C h a r c o a l B r o i l e d Foods I 100.0 I I 100.0 I COLUMN i i (Teen Years) TOTAL 1O0.0 IOO.O STATISTICS CANNOT BE COMPUTED WHEN THE NUMBER OF NON-EMPTY ROWS OR COLUMNS IS 0NE~1 NUMBER OF MISSING OBSERVATIONS • 203 / . . . . . . . . . . . . . . . . . . C R O S S T A B U L A T I O N TFPICK BY GROUP GROUP COUNT I ROW PCT leases controls ROW COL PCT I TOTAL TOT PCT I I I 3 1 TFPICK 1 1 1 I I 3 1 I I 4 once a mo I 7S.0 I 35.0 I 66.7 I 75.0 I SO.O I I SO.O I 16.7 I -| 1 1 3 1 I I O I 1 lew < mo I IOO.O 1 O.O 1 16.7 I 35.0 I O.O I I 16.7 I 0.0 I -! 1 1 4 1 0 1 I I 1 wk 3-3 I 0 0 I ICO.O I 16.7 I 0 0 I SO.O I I 0.0 I 16.7 I -I I I COLUMN 4 2 6 TOTAL 66.7 33.3 100.0 6 OUT OF 6 (ICO.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.333 RAW CHI SQUARE - 2.625O0 WITH 2 OEGREES OF FREEOOM. SIGNIFICANCE - 0.2691 PAGE 1 OF P i c k l e d Foods (Teen Years) NUMBER OF MISSING OBSERVATIONS • 197 278 c a o s s I A 8 U L A T I 0 N BY GROUP caoup COUNT I ROW PCT lcaaea COL PCT I TOT PCT 1 1 COLUMN TOTAL 1 IOO.O 1.1 0.6 7 77.8 7.4 4.0 8 61 .6 8.4 4.5 10 55.6 IO.S 5.7 3 33.3 3.3 1.7 I O I 0.0 t O.O I 0.0 3 33.3 3.5 1. 1 5 38.5 6.3 3.8 44.4 9.8 4.5 6 66.7 7.4 3.4 66 I 60 53.4 1 47.6 69.5 I 74.1 37.5 I 34. 1 95 81 54.0 46.0 PAGE 1 OF ROW TOTAL 1 0.6 9 5. 1 13 7.4 18 10.3 9 5.1 136 71.6 M i l k (Teen Years) 176 100.0 6 OUT OF 13 ( SO.OX) Of THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY ' 0.460 RAW CHI SOUARE • 4.89536 WITH 5 DEGREES OF FREEOOM. SIGNIFICANCE - 0.4388 NUM8ER OF MISSING OBSERVATIONS -C R 0 S S T A A T I 0 N BY GROUP PAGE I OF 1 GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT I 1 ROW TOTAL 100.0 1.4 0.8 66.7 5.5 3.3 3 I 1 O I O.o . o 0.0 33.3 3.B 16 5 I 1 83.3 I 16 .7 6.8 I 1. 9 4.0 I 0. .8 : 7 70.0 9.6 5.6 3 37.5 4.1 3.4 3 30.0 5.7 3.4 5 63.5 9.4 4.0 1 O.B 6 4.8 Coffee (Teen Years) i i i i i 3 I I I I -1 3 I I I I -I 4 I I I I -I 5 I I I 1 -I 6 I I I I -I COLUMN TOTAL 9 OUT OF 13 ( 7S.0X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.431 RAW CHI SOUARE • 4.649S6 WITH 5 DEGREES OF FREEDOM. SIGNIFICANCE • 0.4601 10 7.9 9 6.3 S3 I 43 I 95 SS.S I 44.3 I 75.4 73.6 I 79.3 I 43. 1 I 33.3 I 73 53 136 57.9 43. 1 100.0 NUMBER OF MISSING 06SERVATIONS • 77 279 C I O S I 1 I COUNT ROW PCT COL PCT TOT PCT CROUP I Icaaaa I I 1 1 I I I -I-I I I I -I-I I I I -I-I I I I -I-I I I I -I-I I I I -I-COLUMN TOTAL 4 IOO.O S.l B.J 1 SO.O 7.0 3. a 5 62. B 11.6 6.6 7 46.7 16.3 S.2 1 50.0 2.3 1.3 23 56. 1 S3.5 30.3 43 S6.6 O 0.0 O.O 0.0 3 SO.O a.t 3.8 3 37.5 3.8 8 53.3 34.2 10.5 1 SO.O 3.0 1.3 18 43.8 54.5 23.7 33 A T BT I 0 N CROUP PAGE I OF I ROW TOTAL 4 5.3 Tea (Teen Years) IS 18.7 3 2.6 41 53.8 76 100.0 8 OUT OF 13 ( 66.7X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTEO CELL FREOUENCY • 0.868 RAW CHI SOUARE • 3.92864 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE • 0 5597 NUMBER OF MISSING OBSERVATIONS • C R 0 S S T A A T I 0 N BY GROUP PAGE 1 OF 1 COUNT ROW PCT COL PCT TOT PCT GROUP I Icaaaa I I 1 control i I 2 ROW TOTAL I I I I -I I I I I I I I I -I-I I I 1 -I-1 1 I I -I-I I I I -I-8 I 1 88.8 I 11 . 1 19.0 I 5 .0 12.9 I 1 .6 10 I 3 76.9 I 33 . 1 23.8 I 15 .0 16. 1 I 4. .8 COLUMN TOTAL 9 7S.0 31.4 14.5 I 3 I 35.0 I 15.0 I 4.8 8 SO.O 19.0 12.9 3 6O.0 7. I 4.8 4 57. 1 9.5 6.5 I 8 I 50.0 I 40.0 I 12.8 | I 2 I 40.0 I 10.o I 3.3 I I 3 I . 43.9 I 15.0 I 4.8 43 67.7 30 32.3 13 21.0 13 19.4 16 25.8 7 11.3 Soda Pop (Teen Years) 63 100.0 7 OUT OF 12 ( 5B.3X) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREQUENCY • 1.613 RAW CHI SOUARE • 5.43431 WITH 5 OEGREES OF FREEDOM. SIGNIFICANCE - 0.3653 NUMBER OF MISSING OBSERVATIONS • 141 280 [ I OSS1 I 1 U L I I I 0 N r GROUP PAGE I OF 1 GROUP COUNT I ROW PCT Icaaaa COL PCT.I TOT PCT!I t 1 COLUMN TOTAL a IOO.O 21.4 13.1 I I I I -I I 1 1 33.3 I 7.1 I 3.3 -I I 6 I 43.7 43.S 31.6 I I -I I 1 t 33.3 I 7.1 ! 3.3 -I I 1 I IOO.O I 7.1 I 3.3 -I I 2 I 1O0.0 I 14.3 1 10.3 -J 14 73.7 ' O 0.0 0.0 0.0 2 66.7 40.0 10.S 14.3 20.0 3.3 1 2 I 66.7 I 40.0 I 10.9 -I O 0.0 0.0 0.0 I 1 1 I -I-1 o I 0.0 I 0.0 I 0.0 6 26.3 ROW TOTAL 3 19.6 a 19.1 7 36.8 3 13.8 1 8.3 Beer (Teen Years) 18 100.0 11 OUT OF 13 < 91.7%) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 3.0. MINIMUM EXPECTED CELL FREOUENCT • 0.263 RAW CHI SOUARE • 7.70339 WITH 3 DEGREES OF FREEOOM. SIGNIFICANCE - 0.1734 NlArSER OF MISSING OBSERVATIONS C R 0 S S T A U L A T I 0 N BY GROUP PAGE I OF 1 GROUP COUNT I ROW PCT Icaaaa COL PCT I TOT PCT COLUMN TOTAL 1 I 2 2 I 1 66 .7 I 33.3 33 . 3 I 33.0 20 .0 I 10.0 2 I 3 40 .0 I 60.0 33 .3 I 73.0 20. .0 I 3O.0 1 I 0 t o o .0 I 0.0 16. .7 I O.O 10. .0 I 0.0 1 I 0 100. 0 I 0.0 16. 7 I 0.0 10. 0 I 0.0 6 4 60. 0 40.0 ROW TOTAL 3 30.0 5 30.0 Wine (Teen Years) i 10.0 10 100.0 8 OUT OF 8 (lOO.OX) OF THE VALID CELLS HAVE EXPECTEO CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY • 0.400 RAW CHI SOUARE - 2.22222 WITH 3 DEGREES OF FREEDOM. SIGNIFICANCE • 0.3276 NUMBER OF MISSING OBSERVATIONS • 183 281 C I 0 S S 1 1 1 U l I I 0 N GROUP PACE 1 OF CROUP COUNT 1 ROW PCT Icasn COL PCT I TOT PCT I 1 ROW TOTAL 1 1 I I -:-2 IOO.O 2S.0 20.0 O 0.0 0.0 0.0 2 SO.O 25.0 2O.0 • ! 1 -I I I I I -I I I I 100.0 I 12.5 I 10.0 -I • 2 50.0 100.0 20.0 3 100.0 37.6 •O.O COLUMN TOTAL 80.0 O 0.0 0.0 O.O O I O.O I 0.0 I 0.0 I 1 2 20.0 2 20.0 4 40.0 3 30.0 1 10.0 10 100.0 L i q u o r (Teen Years) 8 OUT OF 8 (1OO.0X) OF THE VALIO CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY " 0.200 RAW CHI SOUARE • 3.7SOOO WITH 3 OEGREES OF FREEDOM. SIGNIFICANCE • 0.2898 NUMBER OF MISSING OBSERVATIONS C R 0 S S T A A T I 0 N BY GROUP PAGE 1 OF I GROUP COUNT I ROW PCT leases COL PCT I ROW TOTAL AVGT TOT PCT I 1 I a I 1 I 44 I 33 I 77 lo t 57. 1 I 42.9 I 38. 1 ! 39.6 I 36.3 I I 21.5 I 16.3 I 2 I 62 1 54 I 116 med I 53.4 I 46.6 I 57.4 I 55.9 I 59.3 I I SO. 7 I 26.7 I I- -I- -I a I 9 I 4 I 9 hi I 55.6 I 44.4 I 4.5 I 4.9 I 4.4 I I 2.5 1 2.0 I -I- -I COLUMN lit 91 202 TOTAL 55.0 45.0 IOO.O 3 OUT OF S ( 33.3X) OF THE VALID CELLS V i t a m i n A Rich Foods (Teen Years) MINIMUM EXPECTED CELL FREOUENCY • RAW CHI SOUARE • O.25658 WITH NUMBER OF MISSING OBSERVATIONS • 4 .054 2 DEGREES OF FREEOOM. SIGNIFICANCE C R O S S T A B U L A T I 0 N BY GROUP PAGE 1 OF GROUP COUNT I ROW PCT Icasas COL PCT I TOT PCT I 1 controls CVGT lo COLUMN TOTAL 48 SO.S 43.2 23.9 61 62.2 55.0 DO 2 2 22.2 1.8 1.0 111 55.0 49.5 51.6 23.3 I 37 I 37.5 I 40.7 1 19.3 I I I I I I 7 I I 77.8 I I 7.7 I I 3.9 I 91 45.0 ROW TOTAL 95 47.0 98 48.5 302 1OO.0 V i t a m i n C Ric h Foods (Teen Years) 2 OUT OF 6 ( 33.37.) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCT LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCT • 4.054 RAW CHI SOUARE - 6.75184 WITH 2 DEGREES OF FREEDOM. SIGNIFICANCE • 0.0342 NUMBER OF MISSING OBSERVATIONS • 1 282 GROUP COUNT I ROW PCT leases COL PCT I TOT PCT I 1 C R O S S T A B U L ROW TOTAL A T BT I 0 N GROUP PAGE t OF :RVOT 1 I GS I 63 I 130 lo 1 S3. 3 I 47.7 I ,64.7 I 61 .8 1 68. 1 I I 33.8 I 30.8 I -I-2 I 41 1 36 I 67 ned I 61.3 I 38.8 I 3 3 . 3 I 37.3 I 38.8 I I 20.4 I 13.8 I 3 I 1 1 3 I 4 hi I 25.0 I 75.0 I 2.0 I O.B 1 3 . 3 I 1 0.8 I 1.8 I COLUMN 110 91 201 TOTAL 54.7 45.3 IOO.O 5 OUT OF 6 ( 33.3*) OF THE VALID CELLS C r u c i f e r o u s Vegetables (Teen Years) MINIMUM EXPECTED CELL FREOUENCY -RAW CHI SOUARE . 2.86471 WITH NUMBER OF MISSING OBSERVATIONS • 1.811 2 OEGREES OF FREEDOM. SIGNIFICANCE • 0.238T C R O S S T A B U L A 7 I 0 N BT GROUP PAGE 1 OF GROUP COUNT 1 ROW PCT leases COL PCT I TOT PCT FATT lo COLUMN TOTAL ROW TOTAL 41 I 37 52.6 I 47.4 36 .8 I 40.7 20.3 I 18.3 66 I 49 57.4 I 42.6 59.5 I 53.8 32.7 I 24.3 4 I 5 44.4 I 5S.6 3.6 I 5.6 2.0 I 3.5 111 91 55.0 45.0 Fat (Teen Years) 115 56. a 9 4.S 202 100.0 2 OUT OF 6 ( 33.3%) OF THE VALID CELLS HAVE EXPECTED CELL FREOUENCY LESS THAN 5.0. MINIMUM EXPECTED CELL FREOUENCY - 4.0S4 RAW CHI SOUARE • 0.8S749 WITH 2 DEGREES OF FREEOOM. SIGNIFICANCE • 0.6513 NUMBER OF MISSING OBSERVATIONS • C R 0 S S T A T I 0 N BY GROUP PAGE I OF GROUP COUNT 1 ROW PCT leases COL PCT I TOT PCT MEATT lo COLUMN TOTAL RAW CHI SOUARE • 111 55.0 91 45.0 ROW TOTAL 1 I 2 15 I 16 49.4 I 51.6 13. S I 17.6 7.4 I 7.9 83 I 68 55.0 I 4S.0 74.8 I 74.7 41.1 I 33.7 13 I 7 65.0 I 35.0 11.7 I 7.7 6.4 I 3.5 31 15.3 151 74.8 Meat (Teen Years) 1.35541 WITH 202 IOO.O 2 DEGREES OF FREEDOM. SIGNIFICANCE NUMBER OF MISSING OBSERVATIONS • 283 C R D S S T A B U L A I o N CROUP VGPRT lo CROUP COUNT I BOW PCT lc...» COL PCT I TOT PCT I 1 49 60.9 44. I 14.3 BOW TOTAL 32 39.5 35.2 15.1 COLUMN TOTAL 91 40.1 1 51 I 47 I 99 1 52.0 I 49.0 1 48.5 1 45.9 I 51.9 I [ 25.2 X 23.3 j II I 12 1 23 47.9 1 52.2 I 11.4 9.9 I 13.2 I 5.4 I 5.9 I 111 91 202 55.0 45.0 IOO.O Vegetable P r o t e i n s (Teen Years) BAW CHI SOUARE • 1.81231 WITH NUMBER OF MISSING OBSERVATIONS -3 OEGREES OF FREEOOM. SIGNIFICANCE • 0.4041 I C R O S S T A B U L A T I 0 N T GROUP PAGE I OF 1 FIBRT lo GROUP COUNT I ROW PCT leases COL PCT I TOT PCT 1 - I I ROW TOTAL 1 I 1 I 3 38 I 35 60.3 I 39.7 34.3 I 37.5 18.8 I 13.4 69 I 63 53.7 1 47.3 63.3 1 69. 1 34.3 I 30.7 4 I 4 50.0 I SO.O 3.6 I 4.4 2.0 J 2.0 COLUMN TOTAL 111 SS.O 91 4S.0 63 31.2 131 64.9 a 4.0 F i b e r R i c h Foods (Teen Years) 302 100.0 2 OUT OF 6 ( 33.3%) OF THE VALIO CELLS HAVE EXPECTEO CELL FREOUENCT LESS THAN 5. MINIMUM EXPECTED CELL FREOUENCT • 3.604 RAW CHI SOUARE • 1.087O4 WITH 2 DEGREES OF FREEDOM. SIGNIFICANCE • 0.5807 NUMBER OF MISSING OBSERVATIONS • I C R O S S T A B U L A I O N GROUP PAGE 1 OF GROUP COUNT I BOW PCT I cases COL PCT I TOT PCT BOW TOTAL BFATT lo 1 COLUMN TOTAL RAW CHI SOUARE • I 1 I 2 I I 24 I 27 I 31 1 47. 1 I 32. S 1 3S.3 1 31.6 I 29.7 I I 11.9 I 13.4 I I 73 I 49 I 131 1 SO.3 I 39.7 I 59.9 1 s s . a I 32.7 I I 36. 1 I 33.9 I 1 14 I 16 I 30 I 46.7 I 33.3 I 14.9 I 12.S I 17.6 I I S.S I 7.9 I III 53.0 91 43.0 303 100.0 3.32949 WITH Beef Fat (Teen Years) 2 OEGREES OF FREEOOM. SIGNIFICANCE NUMBER OF MISSING OBSERVATIONS " 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096638/manifest

Comment

Related Items