"Medicine, Faculty of"@en . "Population and Public Health (SPPH), School of"@en . "DSpace"@en . "UBCV"@en . "Kim, Amanda L."@en . "2009-06-11T19:13:49Z"@en . "1998"@en . "Master of Science - MSc"@en . "University of British Columbia"@en . "Adolescent pregnancy and sexuality has traditionally been approached from the female\r\npoint of view. Research has extensively documented characteristics, antecedents and\r\nconsequences of adolescent pregnancy for the female, but very little is known about the male.\r\nHere, this study begins to describe the adolescent father. Comparisons were made between\r\nadolescent males who are and are not sexually active, as well as between those sexually active\r\nmales who have and have not caused a pregnancy in order to provide clues for the prevention\r\nand counseling regarding initiation of sexual intercourse or adverse sequelae of sexual activity.\r\nData from the British Columbia Adolescent Health Survey was utilized. The survey was\r\nheld in 1992 among 12 to 18 year old students in British Columbia. This survey had collected\r\ndata from 7,254 young males. In this group, 2,345 (31.9 percent) were sexually active, of whom\r\n181 (7.7 percent) had caused one or more pregnancies. An additional 100 (4.3 percent) indicated\r\nthey were unsure about having caused a pregnancy. The associations of sexual activity and\r\nfatherhood with six categories of variables were studied: individual and family demographics;\r\nschool achievement and self-esteem; physical and mental health; risk taking behaviours;\r\nalcohol, tobacco and recreational drug use; and sexuality and sexual practice.\r\nEngaging in sexual activity was independently associated with provincial region, having\r\nacademic difficulties (such as skipping school), and use of alcohol, tobacco, marijuana and other\r\nrecreational drugs. Furthermore, it was associated with risk taking behaviours such as binge\r\ndrinking and involvement in physical fights. These Canadian findings confirm reports from the\r\nUnited States. Among the sexually active young men, fatherhood was not only associated with\r\nuse of contraceptives, early age of sexual activity and history of sexually transmitted infections,\r\nbut also with physical health problems and history of sexual abuse. Religiosity was associated\r\nwith delayed sexual activity, whereas among the sexually active, religious youth were more\r\nlikely to be fathers. These finding have not been reported previously. They are significant in\r\nproviding clues for prevention and counseling, and are the basis for the recommendations made\r\nin this thesis."@en . "https://circle.library.ubc.ca/rest/handle/2429/8974?expand=metadata"@en . "2896781 bytes"@en . "application/pdf"@en . "CHARACTERISTICS OF THE ADOLESCENT FATHER by Amanda L. Kim B.A., The Johns Hopkins University, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Health Care and Epidemiology We accept this thesis as conforming to the required s^ndard THE UNIVERSITY OF BRITISH COLUMBIA October 1998 \u00C2\u00A9 Amanda L. Kim, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. ,> Department of ri^Alih P/*f^ Ef> i A/SMI alaj tf The University of British Columbia Vancouver, Canada DE-6 (2/88) A B S T R A C T Adolescent pregnancy and sexuality has traditionally been approached from the female point of view. Research has extensively documented characteristics, antecedents and consequences of adolescent pregnancy for the female, but very little is known about the male. Here, this study begins to describe the adolescent father. Comparisons were made between adolescent males who are and are not sexually active, as well as between those sexually active males who have and have not caused a pregnancy in order to provide clues for the prevention and counseling regarding initiation of sexual intercourse or adverse sequelae of sexual activity. Data from the British Columbia Adolescent Health Survey was utilized. The survey was held in 1992 among 12 to 18 year old students in British Columbia. This survey had collected data from 7,254 young males. In this group, 2,345 (31.9 percent) were sexually active, of whom 181 (7.7 percent) had caused one or more pregnancies. An additional 100 (4.3 percent) indicated they were unsure about having caused a pregnancy. The associations of sexual activity and fatherhood with six categories of variables were studied: individual and family demographics; school achievement and self-esteem; physical and mental health; risk taking behaviours; alcohol, tobacco and recreational drug use; and sexuality and sexual practice. Engaging in sexual activity was independently associated with provincial region, having academic difficulties (such as skipping school), and use of alcohol, tobacco, marijuana and other recreational drugs. Furthermore, it was associated with risk taking behaviours such as binge drinking and involvement in physical fights. These Canadian findings confirm reports from the United States. Among the sexually active young men, fatherhood was not only associated with use of contraceptives, early age of sexual activity and history of sexually transmitted infections, but also with physical health problems and history of sexual abuse. Religiosity was associated with delayed sexual activity, whereas among the sexually active, religious youth were more ii likely to be fathers. These finding have not been reported previously. They are significant in providing clues for prevention and counseling, and are the basis for the recommendations made in this thesis. iii T A B L E O F C O N T E N T S Abstract ii Table of Contents iv List of Tables v Acknowledgements , vi INTRODUCTION 1 Adolescent Pregnancy as a Public Health Concern 1 Who is the Adolescent Father - A Review of the Literature 3 The Picture in Canada and British Columbia 11 The British Columbia Adolescent Health Survey (AHS) 12 Objectives 17 METHODOLOGY 18 Statistical Analysis 20 RESULTS 22 Individual and Family Demographics 22 Physical Activities, School Achievement and Self-Esteem 27 Physical Health, Chronic Illness/Disability and Mental Health 31 Behaviours which result in Intentional and Unintentional Injuries 32 Tobacco, Alcohol and Drug Use 34 Sexuality, Contraceptive Use and Sexually Transmitted Diseases 37 Multivariate Analysis - Sexual Activity 40 Multivariate Analysis - Fatherhood 43 DISCUSSION 46 Sexual Activity 47 Fatherhood 50 Limitations 53 Future Directions and Summary 54 REFERENCES 57 iv L I S T O F T A B L E S TABLE 1.1 Individual and Family Demographics: Frequency Distribution 24 TABLE 1.2 Individual and Family Demographics: Univariate Analysis 26 TABLE 2.1 Physical Activities, School Achievement and Self-Esteem: Frequency Distribution 29 T A B L E 2.2 Physical Activities, School Achievement and Self-Esteem: Univariate Analysis 30 TABLE 3.1 Physical Health, Chronic Illness/Disability and Mental Health: Frequency Distribution 31 TABLE 3.2 Physical Health, Chronic Illness/Disability and Mental Health: Univariate Analysis 32 TABLE 4.1 Behaviours which result in Intentional and Unintentional Injuries: Frequency Distribution 33 TABLE 4.2 Behaviours which result in Intentional and Unintentional Injuries: Univariate Analysis 33 TABLE 5.1 Tobacco, Alcohol and Drug Use: Frequency Distribution 34 TABLE 5.2 Tobacco, Alcohol and Drug Use: Univariate Analysis 36 TABLE 6.1 Sexuality, Contraceptive Use and Sexually Transmitted Diseases: Frequency Distribution 38 TABLE 6.2 Sexuality, Contraceptive Use and Sexually Transmitted Diseases: Univariate Analysis 39 TABLE 7.1 Multivariate Analysis by Sexual Activity 41 TABLE 8.1 Multivariate Analysis by Fatherhood 44 v A C K N O W L E D G E M E N T S This thesis could not have come to completion without several people: Dr. Sam Sheps, my thesis advisor, who has provided much appreciated support and guidance. My committee members, Ruth Milner and Dr. Roger Tonkin, whose comments have been helpful in the progress of writing this thesis. Aileen Murphy of the McCreary Centre who assisted me with the data and provided information regarding the Adolescent Health Survey. And the students of British Columbia who participated in the survey, without whom none of this would have been possible. vi I. I N T R O D U C T I O N This thesis approaches adolescent pregnancy and sexuality from a male perspective. This introduction begins with a description of adolescent pregnancy from a public health point of view, and then discusses the male involved. Next follows the nature of this problem in Canada, and more specifically in British Columbia. The Adolescent Health Survey (AHS) of British Columbia is the study upon which this thesis is based and its purpose and development are discussed. Finally, the research questions that this thesis proposes to answer are presented. A. ADOLESCENT PREGNANCY AS A PUBLIC HEALTH CONCERN Traditionally, pregnancy has been viewed from the female perspective, and pregnancy involving adolescents is no different in this respect. The female has predominantly suffered the direct consequences of unprotected intercourse, including both pregnancy and sexually transmitted infections (16). The consequences, such as educational outcomes and economic difficulties, have been extensively documented for the young mother (16, 39, 59). Biologically, the outcomes of early maternity are different for younger adolescent females, 12 to 15 years old, versus older adolescent females, 16 to 19 years old. Studies have examined the relationships between adolescent motherhood and marital experience, subsequent fertility, educational attainment, patterns of labour force participation, job status, earnings, and poverty and welfare status (39, 59). In Western nations such as Canada and the United States, many women are choosing not to marry, since the social need to legitimize birth with marriage is not as great as it once was (40). While older unmarried mothers tend to be in a stable relationship or environment when they give birth, teenage mothers tend not to be (40). The increased proportion of female-1 headed, single-parent families also tends to lead to more of the population becoming vulnerable to poverty (81). Data from the Youth Risk Behaviour Survey administered by the Division of Adolescent and School Health of the U .S . Centers for Disease Control reported that 54.2 percent of all sampled high school students in 1990 had ever had sexual intercourse and 39.4 were currently sexually active (that is, had intercourse during the three months prior to the survey) (11). A significant trend of increasing sexual activity by grade level was also observed (11). As a behavioural trend, this is quite noteworthy since sexual activity contributes greatly to morbidity and mortality among adolescents via infections, physical and psychological effects of pregnancy and other adverse consequences (53, 74). There are differing perceptions of the problem of adolescent pregnancy; that the problem lies in the morality of premarital sexual activity, that the focus should be taking responsibility and use of effective contraceptive methods, or that the central issue is the choice of continuing the pregnancy versus termination (15, 31). The perception also exists that adolescent girls are too immature to raise children, a perception not supported by cross-cultural ethnographic evidence (45). Many theories have been proposed explaining why young girls become mothers in their teenage years. The problem behaviour theory of Jessor and Jessor posits that engaging in some problem behaviours leads down the path toward other problem behaviours such as adolescent parenthood. That is, they cluster. Problem behaviours are defined as \"behaviors that constitute transgressions of societal and/or legal norms and that tend to elicit some sort of social control response\" (53, page 80). Another theory proposes that adolescents who are depressed or 2 emotionally deprived will turn to parenthood as a means to capture a sense of intimacy and love (45). A third theory states that adolescent pregnancy offers an alternate life course to achieving adulthood, particularly for those who experience obstacles to other paths (45). Recognizing adolescent pregnancy as a public health concern has given rise to programs targeting its prevention. This prevention is usually in the form of sex education programs or family planning programs most often addressing the adolescent girls (56). In contrast, the partner in adolescent parenthood, the father, has been virtually ignored (1, 59, 68). Considering sexually active adolescent females do not get pregnant or sexually transmitted infections alone, it seems odd that males have managed to be neglected for so long. The role of the teenage father, however, is slowly emerging from this neglect, but it still has a long way to go in catching up to the amount and breadth of information regarding adolescent mothers. B. WHO IS THE ADOLESCENT FATHER - A REVIEW OF THE LITERATURE Adolescent males have recently been identified as a priority for data collection and research (31). Current research in adolescent pregnancy reflects the fact that policymakers, service providers, parents, and the teenagers themselves have traditionally regarded adolescent pregnancy as a female problem (10, 37). While adolescent mothers, and to a slightly lesser degree, their children, have been extensively studied, the males impregnating the adolescent females have managed to escape scientific scrutiny for some time. Nevertheless, many myths and stereotypes regarding these males abound. Such myths and stereotypes often characterize the male as being worldly when it comes to matters of sexuality, or exploitative of unsuspecting and helpless adolescent females 3 (69). Others depict him as needing to prove his masculinity by impregnating young girls (69). He is also seen as emotionally cold about relationships with the opposite sex, or non-participating in his relationship with the young mother and raising of his child (69). However, there is no research evidence upon which to base these characterizations. Thus with the acknowledgment of the father's role in the growth and development of his children, the fathers of children born to adolescent mothers have begun to be recognized and become the object of study as well. However, this has not been without its own share of research problems. 1. The Forgotten Father Most investigations of adolescent parenthood omit the father. When he is mentioned, it is only in his relation to the adolescent mother (e.g., 38, 43, 48, 71). For instance, a review article regarding partner support in adolescent pregnancy offered only a view of how partner support affected the young mother while no description was given of how the father benefited from participating, i f he participated at all (71). A recent article examining intergenerational transmission of adolescent parenthood looked at both male and female offspring, but only at the teenage mother. The father was nowhere mentioned with regard to his influence (38). A report submitted to the British Columbia Task Force on Teen Pregnancy and Parenting that outlined a support group for parents defined the clients specifically as teenage mothers (76). The father was mentioned as one of the discussion topics, and permitted to answer questions i f present, but his presence or participation did not seem to be of importance to the support group facilitators (76). From Ontario, a report on single adolescent mothers experiences only mentions the father of the child in relation to social contact with the mother or child, while the mother's economic, health and child care situations are extensively documented (49). 4 Even at an international level, males are not included in this public health issue. A joint statement published by the World Health Organization (WHO), the United National Fund for Population Activities, and UNICEF regarding the reproductive health of adolescents does not mention males. The term \"young people\" is used, as well as \"adolescent females\" or \"young females\", but nowhere are males specifically mentioned (82). When attempts are made to include the father, it is often inferential or indirect. In an early study of fathers, the fathers who were adolescents were not distinguished from those who were adults (69). Some studies or surveys used proxy measures such as \"going steady\" for sexual activity (27), or household structure as an indicator for commitment to fatherhood (52). In other instances, the mother was used as the source of information about the partner. These data are subject to bias, depending on the nature of the relationship between the adolescent mother and her young partner and how she wishes to portray him (20, 29, 30, 46, 48, 77). As well, such information is limited in details and reliability. Information needs to be collected and analyzed in a timely manner; what was relevant for that cohort of adolescents may not have been for following generations (9, 68). As well, the timing of when fatherhood was determined is usually retrospective (e.g., 13, 20, 27). Fatherhood is usually defined as being when the adolescent female chooses to continue the pregnancy and the male partner has chosen to continue his involvement with the mother rather than whether he has impregnated an adolescent female, regardless of the outcome of the pregnancy (9, 19, 22, 33, 34, 59). 5 2. Early Research Initial research on adolescent fathers was exploratory, using sample sizes that were quite small, and therefore did not offer a great deal of statistical power (3, 19, 22, 23, 27, 33, 34, 58, 62, 67, 78, 79, 80). Recruitment of participants in the study was often through the female partner (4, 17, 19, 22, 34, 58, 62, 65, 78, 79), or used other sampling procedures that resulted in unrepresentative samples such as convenience samples (47, 63), analysis restricted to particular racial groups (27), community (60), or a birth cohort (12, 13). In addition, the data often gathered by such studies were very limited in detail. Such data were usually restricted to age, education, or employment (46, 50). But the dearth of information and the unrepresentativeness of samples have slowly been addressed through large surveys, such as the Youth Risk Behavior Survey, the National Survey of Adolescent Males, the National Longitudinal Survey of Work Experience of Youth ( N L S Y ) , High School and Beyond (HSB), and the National Chi ld Development Study. Unfortunately, most of these surveys were conducted in the United States, thus the findings are not always generalizable to Canadian context (39). The few studies in Canada with some information relating to adolescent pregnancy and fatherhood had either a limited study objective, study population, or are now out-dated (35, 55). Other than that, Canadian information consists of vital statistics and reports from task forces and family planning agencies such as Planned Parenthood (8, 61, 76). Canada has participated in a W H O international survey of youth health that addressed many areas, such as alcohol, tobacco, and other drug use, physical activity, nutrition, and relationships with others (42). However, this ideal opportunity to also survey sexuality issues was not taken. A late 1980 report details teenage pregnancy and parenthood in British Columbia, 6 and mentioned that fathers were neglecting their role to provide financial support and other responsibilities, but little other reference was made to males (73). Provincial statistics were provided, but no further study was conducted (73). In Ontario, a descriptive study of 50 adolescent mothers and their children participating in an interdisciplinary clinic hardly mentioned fathers, except in a judgmental manner, or in the context of contact with the mother and child and as a source of income (72). Also, from other countries, outside of the United States, little information is available on adolescent pregnancy. In Santiago, Chile, a survey regarding sexual activity and contraceptive use among a sample of 15 to 24 year old males and females was conducted, but questions regarding pregnancy were only addressed to the females (36). The National Child Development Study took place in Great Britain, but the data were from a cohort of children born in 1958 (12, 13). Focus group discussions in Nigeria and Kenya looked at influences on adolescent sexuality, but sampling methods were not used (2). In Mexico City, a study looked at the influence of family on the sex lives of adolescents, and one part of that study used adolescent males (63). Unfortunately, poor sampling methods were used here as well (63). Since there are few studies and most are exploratory in nature, comparisons between studies are difficult to make. Definitions of adolescence and fatherhood appeared inconsistent. The variables of interest and how they are measured or scaled are not standardized. Moreover, methods of data collection and analysis vary between the different studies. Despite these limitations being identified in the early 1980s (37, 68), the current literature continues to be methodologically poor, though some studies have begun to address these problems (75). Much of previous research suffers from study design flaws, so in fact, the conclusions drawn cannot be 7 definitive. When future research quotes conclusions drawn from such research, the flaws are perpetuated. 3. Definition of the Adolescent Mother and Father Adolescence itself is somewhat difficult to define. In using chronological age, the boundaries of adolescence are at the lower end, around 10 to 12 years old, and at the upper end, 18 to 20, and even beyond (53). Other criteria, such as biological, psychological, social and institutional are needed for a more precise definition; many of these criteria change in relation to age, including age of menarche, entry into secondary school, and initiation of sexual activity (53). The adolescent mother is usually recognized as a female in her adolescent years, most commonly under the age of 20 and usually over the age of 12. The adolescent father, on the other hand, is not so easily defined. He could be considered analogous to the female as simply a male in his adolescent years (69). He may also be the partner of an adolescent mother, regardless of age (69). The mere act of impregnating an adolescent female could make him a father, or he could be considered a father only i f the pregnancy is carried to term. His level of participation in raising the child may be another factor to consider in the adolescent father definition, that is, whether he assists with the raising of the child, or chooses to have nothing to do with the child. For the purposes of this research, all references to adolescent fathers w i l l be those males in their adolescent years that have caused a pregnancy, regardless of the outcome of the pregnancy and his degree of involvement in raising offspring. 4. Characteristics of the Young Father Through the slowly growing body of research, the characteristics of adolescent fathers are 8 being defined. However, just as there are differences between adolescent females who choose to continue a pregnancy and raise their child and those who choose to terminate or to place for adoption (25, 83), there may also be differences between those males who stay with their partner and are involved with their children versus those who only provide financial assistance, versus those who do not acknowledge their role in the pregnancy. The father has several degrees of involvement from which he can choose. However, the characteristics that are often quoted in the literature do not make the distinction between these various degrees of involvement of the adolescent father. Previous research has found that typically adolescent fathers have some type of difficulty in school, are educationally disadvantaged in some way (12, 13, 21, 27, 29, 30, 58, 66) have low expectations for their future (27), or are limited by the expectations of others (12, 13, 27). They tend to come from low socioeconomic backgrounds (12, 13, 24, 30, 65, 66), are more often of black racial background (21, 27, 30, 65), and live in an urban environment (65). They also tend to be older than the mothers are (13, 29, 46, 50, 75). Some studies have suggested that older male partners of teen mothers are more similar developmentally to teen fathers than their peers who have adult female partners, such as having low educational attainment and an inconsistent work history (46). Age difference raises the issue of life experiences, and perhaps more importantly, issues of pressure, abuse, and power in relationships (26, 46). The behavioural characteristics of teenage fathers are not often described. Considering that the primary causes of morbidity and mortality among adolescents are behavioural in origin, this is quite an oversight (53, 74). It is reported that adolescent fathers tend to engage in risky behaviours such as illegal drug use, sexual intercourse, cigarette smoking, and alcohol use (21, 9 75). Some have participated in lawbreaking activities (13, 20, 21, 75). Males also generally report earlier sexual experience than females do (46). However, causation theories for male sexual behaviour are difficult to determine, much less prove, since paternity is not easily established (16). A 1985 study by Rivara, Sweeney and Henderson is one of the few that compare fathers with non-fathers. Age and race matched peers were used to compare demographic background, attitudes and knowledge regarding pregnancy and contraception, and family characteristics and dynamics, but the controls were not necessarily sexually active and the subjects were recruited from a black, urban, low socioeconomic population, often through partner identification or clinics in the hospital offering the teen mother program (65). Across cultures, facets of \"maidenhood\" have been researched, such as premarital sexual activity, menarche, and measures to control or encourage sexual development (45). However, anthropological studies on the male corollary, \"bachelorhood\", have been noticeably lacking (45). Most of the research has come from the United States, and it is difficult to draw conclusions from the American population and apply them to Canada. In a comparison of teenage pregnancy among industrialized countries, including several western European nations, Canada's birth rate, abortion rate and overall pregnancy rate was in the middle of two extremes: the highest rates being in the United States and the lowest in the Netherlands (39). Canada also has a different social system and dissimilar demographics than the United States (39). However, the extent of adolescent pregnancy research in Canada is quite limited. It mainly consists of reports of birth rates, abortions, live births, rural versus urban distribution and provision of services (61), case studies and clinical experiences (70), or dated studies using unrepresentative or questionable samples (35, 64). 10 Unfortunately, one of the problems of this type of research is gathering data. Few data sets have been designed to provide data on sexual activity among male adolescent (59). In order to become a father, the teenage male needs to become sexually active first. Therefore, it is important to determine characteristics not only associated with fatherhood, but also with initiation of sexual activity. Since information regarding the men involved in pregnancies are usually gathered from women, basic information may be missing or inaccurate (46). Fathers ages are not reported on a high proportion of birth certificates for infants born out of wedlock (46). Other information, regarding behaviour and characteristics other than basic demographics are not recorded. C. THE PICTURE IN CANADA AND BRITISH COLUMBIA According to a survey conducted in 1990 by the Canadian Institute of Child Health, 50 percent of both teenage males and females, aged 15 to 19, were sexually active in the past year (28). In a 1992 survey of 660 females, aged 15 to 18 in three Canadian cities, over one-quarter of those who were sexually active (41 percent) used no form of contraception (28). A Canadian report on the reproductive health of adolescents recognized the need for research in this field (32). Early sexual activity, pregnancy among adolescents, negative effects of early parenthood, public costs, and infections from sexually transmitted diseases are now recognized. However, there is little research to help deal with these problems (32). Teenage girls often have low birth weight infants due to inadequate nutrition, delayed prenatal care and cigarette smoking. This leads to higher morbidity and mortality among their infants, who will in turn require more health services (32). 11 From two nationally representative surveys of sexual activity, knowledge, attitudes and behaviours of youth in Canada, it was found that adolescents initiate intercourse at younger ages and are often without adequate protection against pregnancy or sexually transmitted diseases (41). However, that national report is based mainly on statistics and a few provincial reports and two national studies that have limited bases of information. Manitoba Health published a report on teenage pregnancy, but the statistics were based on the total population of females aged 15 to 19, not all of whom may be sexually active (51). In 1995, six out of every 100 live births in British Columbia were to teenage mothers, who are defined as women less than 20 years old (8). There were a total of 2,639 births to teen mothers, of which 2,618 were live births and 21 were stillbirths (8). Teen pregnancy rates for British Columbia from 1985 to 1993 have shown no dramatic change (8). Abortion statistics were last available for 1994, with over half of all teenage pregnancies terminated (8). However, this was only for abortions performed in hospitals; abortions conducted in clinics or medical abortions were not included in this figure. Vital statistics regarding the male partner were limited to age, which was often not reported, and marital status. Of all live births to teenage mothers in 1995, the age of the father was not recorded for almost one-third (8)! As well, seventeen percent of these young mothers had given birth previously at least once (8). D. THE BRITISH COLUMBIA ADOLESCENT HEALTH SURVEY (AHS) While it is recognized that there are a variety of health and behavioural problems among adolescents, relatively little information was available on the health status and health practices of adolescents in British Columbia. Thus, services and programs had been designed without any 12 actual information regarding characteristics, needs or problems of adolescents. This need for data regarding adolescents was recognized by the McCreary Centre Society, a non-profit organization dedicated to issues regarding the health of adolescents, and led to the development of the Adolescent Health Survey. One of the problems recognized in the research of adolescent pregnancy is the gathering of data. There are few sources that provide information on the demographics, much less the behaviours of adolescent males. 1. Survey Design The Adolescent Health Survey (AHS) was developed in 1990 because there was very little information available on the health status and health practices of adolescents in British Columbia. The survey is comprised of questions from three other tested instruments: the Youth Risk Behavior Survey (YRBS) of the Division of Adolescent and School Health of the U.S. Centers for Disease Control, the Minnesota Adolescent Health Survey ( M A H S ) , and the World Health Organization Cross-National Youth Survey. Additional items concerning specific health problems and information related to eating disorders were also developed and included (54). The questionnaire is a paper and pencil format, and self-administered in a classroom setting. It consists of 123 multiple-choice questions. Seventy-three questions regarding core risk behaviour questions were taken from the Y R B S . Thirty-two demographic and health status items were adapted from the M A H S . In addition, some questions on self-esteem were taken from the World Health Organization Cross-National Youth Survey. The Y R B S in its entirety was used in the survey, and therefore comprised the largest portion of the A H S . The Y R B S was developed by the Division of Adolescent and School Health 13 of the U.S. Centers for Disease Control and Prevention as a component of its Youth Risk Behavior Surveillance System (YRBSS) to assess the comparative prevalence of behaviours contributing most to adverse health and social outcomes (44). These behaviours often are established during youth, often extend into adulthood, and are often inter-related (44). Federal scientists met to delineate priority behaviours and develop survey items used to monitor the prevalence of these behaviours in the six categories of unintentional and intentional injuries; drug and alcohol use; sexual behaviours leading to HIV infection, other sexually transmitted disease, and unintended pregnancy; tobacco use; dietary behaviours; and physical activity (44). The aim was to be able to administer the survey in approximately 45 minutes, the average length of a class period (44). A draft of the survey was presented to state and local education agencies and survey methodologists, and modifications were made. The National Center for Health Statistics conducted laboratory and field tests of the survey. The first wave of testing included focus group discussions, the second wave involved a series of personal one-on-one interviews, the third wave involved interviewing peer educators, and the fourth wave was conducted with a group of summer students, some of whom had very low reading and comprehension abilities (14). A test-retest reliability study on the YRBS was conducted using a convenience sample from five school districts, one suburban, and two each of urban and rural (7). It has already been documented that adolescent provide reliable and valid measures regarding substance use, but it is not known whether this is the case for other sensitive information such as many of the risk behaviours on the YRBS (7). In calculating Kappa statistics for the various items of the Y R B S , the majority of Kappa values fell in the moderate or greater ranges (41-100%); however because all inconsistent responses were counted as response errors, it is likely to result in a conservative estimate for 14 reliability (7). One point of note is the responses of those in grade seven were far less consistent than of those in the upper grades; in fact, the responses of those in grades eight through twelve were not very different (7). However, when using lifetime as the point of reference, the younger and older students both had better reliable responses (7). Overall, though, the YRBS has been determined to have adequate reliability (7). The Minnesota Adolescent Health Survey (MAHS) also underwent similar development and pre-testing (5). Health educators, practitioners and social scientists collaborated to identify content domains to be covered by the M A H S , and items from other instruments were included (5). The first draft was pretested on a sample of predominantly rural youth at a statewide youth gathering (5). Reliability, validity, and feasibility were checked; revisions were made and then another pretest conducted on students at a private urban school (5). After further modifications, the final draft of the questionnaire was printed (5). The pre-testing conducted for the separate components of the survey, as well as the AHS as a whole add to the strength of the data collected and conclusions drawn from the statistical analysis. The AHS itself also contains several items which demonstrate the internal validity of the questionnaire. The survey was then field-tested in British Columbia during the 1991 school year. Domains covered by the Adolescent Health Survey included: 1. individual and family demographics 2. physical activities, school achievement and self-esteem 3. physical health, chronic illness/disability and mental health 4. behaviours which result in intentional and unintentional injuries 15 5. tobacco, drug and alcohol use 6. sexuality, sexually transmitted diseases and pregnancy 2. Sampling and Survey Administration From 1992 census estimates, there were approximately 296,000 youth between 12 and 18 years old in B.C.; 256,884 were enrolled in public and private schools. The sampling frame included all youth in schools (87%). Those who were not represented included youth not attending school (employed or unemployed), youth in care or corrections facilities, and street youth. Therefore, the sample is only representative of those in school, and caution is made against generalizing to those not included in the sampling frame (54). The sample was selected from among the districts and schools comprising the eight regions of British Columbia. For each school district, a ten percent sample of students was drawn by randomly selecting schools and classrooms representing grades 7 through 12 in public and independent schools. However, of 75 school districts, only 48 agreed to permit access to selected schools. This subsample contained 67 percent of all students (54). Students took approximately 35 to 55 minutes to complete the survey. Representatives administered the AHS from the provincial health units, under the direction of the Medical Health Officer. Students were invited to participate in the survey i f parental consent was obtained, or in the absence of objections or refusals by parents. Surveying was voluntary, anonymous and confidential. Instructions were given to the survey administrators from the health units on how to administer the survey, collect information on class enrollment, absenteeism, and parent/student refusals. The instructions given to students were standardized across all grades and schools. The data were then entered and checked for errors. Incomplete surveys made up 1 16 percent of all participants, that is, those which did not have enough questions answered or failed to give reliable answers, were omitted, as part of the non-responders (54). E. OBJECTIVES The objective of this thesis is to describe the adolescent father, as has been earlier done for the adolescent mother. Comparisons will be made between adolescent males who are and are not sexually active as determined by self-reports, as well as between those sexually active males who have and have not caused a pregnancy in order to provide clues for the prevention and counseling regarding initiation of sexual intercourse or adverse sequelae of sexual activity. 17 I I . M E T H O D O L O G Y The Adolescent Health Survey consists of 123 questions, collecting a variety of information from the student respondents. To analyze all this information is rather tedious, and some of the responses do not translate into useful information. For example, one series of questions asks about how one feels about oneself using various statements. Individually, these statements do not give any useful information, but collectively, it gives an indicator of low, medium or high self-esteem, which is much more relevant. Many of these questions were used, so that collectively, a series of them could give relevant information. Responses to a series of five questions were used to determine a psychological distress score. Questions regarding various physical activities were consolidated as well as questions regarding safety practices that could prevent unintentional injuries such as use of helmets when bicycling or riding a motorcycle. One set of questions asked about various health problems, which was reduced to whether they had health problems and health problems that limited them in any way. These were separated into problems of a physical nature, such as allergies, infections, and chronic conditions, and other problems such as attention deficit disorder and learning disabilities. Data relating to eating disorders, diet, part-time work, HIV/AIDS knowledge, general mood and suicide are not presented or analyzed since they were not considered relevant to the current study, nor were they found to be meaningful in the literature review. For several questions, response categories were very detailed, and therefore, difficult to analyze. For this reasons, response categories were collapsed. For many of the questions regarding behaviour, data were collected on the age of initiation of the behaviour, lifetime 18 engagement of the behaviour, and current engagement, that is in the past 30 days, of the behaviour. Usually, the responses were collapsed into three categories of none, one or two, three or more episodes of the behaviour. The one or two episode category was used to account for any experimentation in which adolescents may engage. Regarding education, since the survey participants were already in school, the categories of interest were high school completion or incompletion and education beyond high school to determine what educational aspirations students had. For this analysis, two questions were the main focus of dividing up the sample of students who answered the Adolescent Health Survey. One question asked students whether they had ever engaged in sexual intercourse, and the other, whether they had ever caused a pregnancy. In order to compare among sexually active versus non-sexually active males, responses to the former question were used to group the sample. For comparing among those who had caused a pregnancy and those who did not, the males who were sexually active according to the previous grouping were then subdivided according to their response to the question regarding pregnancy. Males who had not replied to either question were excluded from the analysis. The Statistical Package for Social Sciences (SPSS) was used to conduct the statistical analyses. Frequencies were first presented for the following groupings to describe each group and show trends: Group 1 non-sexually active males Group 2 non-fathers; males, sexually active, did not cause a pregnancy Group 3 fathers; males, sexually active, did cause a pregnancy They are reported as percentages of those who responded to the particular question. 19 Statistical Analysis Univariate logistic regression analysis of sexual activity compares the non-sexually active males of Group 1 with the non-fathers of Group 2 combined with the fathers of Group 3. Significance of variables is defined when 95 percent confidence intervals (CI) of the odds ratios (OR) exclude 1.00. The value to which all other values of the variable are compared to is indicated by the odds ratio equal to 1.00 without a confidence interval. For the logistic regression analysis for fatherhood, the males in Group 1 were excluded and comparison was only between Groups 2 and 3. Significance criteria were as mentioned above. Because of the large number of variables, prior to multivariate analysis, survey questions were grouped into the following categories: 1. individual and family demographics 2. physical activities, school achievement and self-esteem 3. physical health, chronic illness/disability and mental health 4. behaviours which result in intentional and unintentional injuries 5. tobacco, drug and alcohol use 6. sexuality, sexually transmitted diseases and pregnancy For each of the above 6 groups, a multivariate logistic regression model for sexual activity and for fatherhood were constructed using forward and backward stepwise selection procedures (inclusion criteria p<0.10, exclusion criteria p>0.05). Variables that were found to be significant in these multivariate models were subsequently brought together in an overall model using again the same forward and backward stepwise selection procedures and the same inclusion and 20 exclusion criteria. These final models for sexual activity and for fatherhood are presented. 21 I I I . R E S U L T S From the total of 15,549 respondents, 7,650 (49.2 percent) were males. After removing those respondents who had not indicated whether they were sexually active or had caused a pregnancy, 7,354 males remained. Of these respondents, 4,728 (64.3 percent) were not sexually active (assigned to Group 1); 2,345 (31.9 percent) were sexually active but had not caused a pregnancy (non-fathers: assigned to Group 2); and 181 (2.5 percent) were sexually active and had caused a pregnancy (fathers: assigned to Group 3). Of these 181 fathers, 71 (39.2 percent) indicated to have caused multiple pregnancies. Aside from those in Groups 1 through 3, a remaining 100 respondents (1.4 percent) were sexually active but had indicated not to know whether they had caused a pregnancy. In comparing this fourth group of males with those in Group 3, there were no substantial or statistically significant differences in the characteristics. For reasons of accuracy this fourth group has been excluded from the analyses presented. A. Individual and Family Demographics (Tables 1.1 and 1.2) Respondents originated from eight regions in British Columbia. Greater Vancouver comprised the majority of non-sexually active and sexually active respondents. With respect to the proportion of pregnancies caused, Greater Vancouver, Fraser Valley and Upper Island regions comprised the majority (Table 1.1). Relative to Greater Vancouver, a higher percentage of respondents from all other regions were more sexually active (Table 1.2). Those most likely to be sexually active were in the Fraser Valley, Northeast, and Capital Region (Table 1.2). These associations are different for fatherhood and not statistically significant, due in part to smaller 22 numbers. Mean age increased from Group 1 to Group 2 to Group 3, though the difference in the mean ages of Groups 2 and 3 was quite minimal: 16.95 and 17.02, respectively (Table 1.1). For the univariate odds ratios, ages were grouped as: 14 years or less, 15 to 16 years, and 17 years or older. Increasing age was clearly associated with increased likelihood of being sexually active, but regarding fatherhood, no distinct statistically significant associations were shown (Table 1.2). The majority of students reported to be of European ethnicity. Asians and Native Americans were the two other ethnic groups of note. An equal proportion of students .did not declare any ethnicity in all three groups (categorized as none). Becoming sexually active was associated greatly with Native American ethnicity. A decreased association was observed with / Asian ethnicity. Fatherhood was not significantly associated with ethnicity. The results of the Hispanic respondents should be interpreted with caution due to the small numbers. The percentage of males living with both parents declines from Group 1 through to Group 3. Both groups of sexually active males live with a single parent or in an alternate family structure other than with their parents, and by far a greater percentage live alone. This trend observed among the frequency distribution is further emphasized.in the univariate odds ratios for sexual activity and for fatherhood. Not living in a two-parent family structure has an increased likelihood of the male becoming sexually active and causing a pregnancy. There is a steep drop from Group 1 to Group 2 (73.3 to 58.3 percent) for the proportion having biological parents who are married. Moreover, a larger percent of males who have caused a pregnancy tend to be adopted, have one parent deceased or have parents that are not married than the other two groups. But while marital status is associated with males becoming sexually 23 active, it does not show the same significant relationship with fatherhood. Educational attainment of the mother and father is lowest among the males of Group 3, with 21.8 and 29.6 percent not finishing high school, respectively. A small association is shown relating to males becoming sexually active, but fatherhood does not appear to be influenced. As well, in all three groups, the majority of mothers and fathers were permanently employed. However, not as many mothers worked as fathers. Association of this variable with sexual activity or fatherhood was not significant. TABLE 1.1 - Individual and Family Demographics: Frequency Distribution Group 1 Group 2 Group 3 N (total = 7,254) 4,728 2,345 181 % of total 64.3 31.9 2.5 region Greater Vancouver 33.7 19.4 20.4 Capital 8.4 10.9 7.7 Fraser Valley 9.4 13.7 20.4 Interior 13.7 17.1 9.9 Kootenays 8.6 8.2 8.3 Northeast 8.2 11.5 11.6 Northwest 4.1 4.0 3.3 Upper Island 13.8 15.3 18.2 age mean 15.54 16.95 17.02 ethnicity none 16.7 16.7 16.3 European, only 30.7 37.7 32.0 Asian, only 19.9 6.3 9.6 Hispanic, only 0.5 0.8 0.0 Native American, only 1.5 3.1 2.2 other, only 6.2 5.7 7.3 multiple 24.5 29.7 32.6 24 TABLE 1.1 (continued) Group 1 Group 2 Group 3 family structure two parents 75.4 61.8 53.0 single parent 19.9 27.4 23.8 alternate structure 4.6 9.6 16.6 alone 0.1 1.1 6.6 parents' marital status married 73.3 58.3 53.9 divorced/separated 19.9 31.8 27.2 one or both deceased 2.6 2.9 7.2 never married 2.7 5.2 8.3 adopted 1.5 1.7 3.3 education, father beyond high school 43.2 40.4 33.5 did not finish high school 16.3 24.6 29.6 finished high school 17.0 19.3 18.4 do not know 23.5 15.7 18.4 education, mother beyond high school 41.2 40.8 33.0 did not finish high school 12.7 18.9 21.8 finished high school 25.0 27.6 29.1 do not know 21.1 12.7 16.2 employment, father permanent 81.3 77.9 70.0 unemployed/retired 2.9 3.2 2.8 seasonal/intermittent 7.8 10.8 12.8 househusband 1.0 0.4 1.7 do not know 7.0 7.6 12.8 employment, mother permanent 59.1 60.3 56.7 unemployed/retired 2.0 2.3 5.0 seasonal/intermittent 10.4 11.3 10.0 housewife 24.9 22.4 20.6 do not know 3.6 3.7 7.8 25 TABLE 1.2 - Individual and Family Demographics: Univariate Analysis sexual activity fatherhood OR 95% CI OR 95% CI region Greater Vancouver Capital Fraser Valley Interior Kootenays Northeast Northwest Upper Island 1.00 2.20 (1.83,2.65) 2.61 (2.20,3.10) 2.09 (1.78,2.45) 1.65 (1.36,2.01) 2.42 (2.02,2.91) 1.66 (1.28,2.16) 1.95 (1.66,2.29) 1.00 0.67 (0.36, 1.26) 1.41 (0.88,2.28) 0.55 (0.31,0.99) 0.96 (0.51, 1.79) 0.96 (0.55, 1.67) 0.78 (0.32, 1.91) 1.13 (0.69, 1.84) age 17 or older 15 to 16 14 or less 1.00 0.42 (0.37,0.47) 0.16 (0.14,0.19) 1.00 0.84 (0.59, 1.20) 1.12 (0.76,1.63) ethnicity none European, only Asian, only Hispanic, only Native American, only other, only multiple 1.00 1.21 (1.05,1.40) 0.33 (0.27,0.41) 1.46 (0.78,2.73) 1.99 (1.41,2.81) 0.94 (0.74, 1.18) 1.22 (1.05,1.41) 1.00 0.87 (0.55, 1.38) 1.54 (0.82,2.89) 0.03 (0.00,790.22) 0.73 (0.25,2.14) 1.31 (0.66,2.60) 1.13 (0.71, 1.79) family structure two parents single parent alternate structure alone 1.00 1.68 (1.50,1.89) 2.73 (2.26,3.30) 14.58 (6.15,34.55) 1.00 1.01 (0.70,1.46) 2.00 (1.30,3.09) 6.96 (3.41,14.22) parents' marital status married divorced/separated widowed never married adopted 1.00 2.00 (1.79,2.24) 1.59 (1.19,2.12) 2.54 (1.98,3.26) 1.51 (1.04,2.20) 1.00 0.93 (0.65,1.32) 2.66 (1.42,4.99) 1.73 (0.97,3.07) 2.09 (0.86,5.05) father's education beyond high school did not finish high school high school only do not know 1.00 1.66 (1.46,1.89) 1.23 (1.07,1.41) 0.73 (0.64,0.84) 1.00 1.45 (0.99,2.12) 1.15 (0.74, 1.78) 1.41 (0.91,2.20) 26 TABLE 1.2 (continued) sexual activity fatherhood OR 95% CI OR 95% CI mother's education b e y o n d h i g h school d i d not f i n i s h h i g h school h i g h school o n l y do not k n o w 1.00 1.53 (1 .33,1 .77) 1.13 (1 .00,1 .28) 0.63 ( 0 . 5 4 , 0 . 7 3 ) 1.00 1.43 (0 .94,2 .18) 1.30 (0.89, 1.92) 1.58 (0 .99,2 .51) father's e m p l o y m e n t permanent u n e m p l o y e d seasonal/intermittent househusband do not k n o w 1.00 1.16 (0 .87 ,1 .53) 1.49 (1 .26,1 .75) 0.53 ( 0 . 2 9 , 0 . 9 9 ) 1.20 (1 .00,1 .44) 1.00 0.96 (0 .38,2 .41) 1.31 ( 0 . 8 3 , 2 . 0 9 ) 4.32 (1.17, 15.88) 1.86 (1 .16,2.97) mother's e m p l o y m e n t permanent u n e m p l o y e d seasonal/intermittent housewife do not k n o w 1.00 1.22 ( 0 . 8 8 , 1 . 6 9 ) 1.06 (0.91 ,1 .25) 0.88 ( 0 . 7 8 , 0 . 9 9 ) 1.07 ( 0 . 8 3 , 1 . 3 8 ) 1.00 2.29 (1 .10,4.77) 0.94 (0 .56,1 .58) 0.98 (0 .66,1 .44) 2.26 (1 .24,4.12) B. Physical Activities, School Achievement and Self-Esteem (Tables 2.1 and 2.2) T h e majority o f the respondents reported part ic ipat ing i n p h y s i c a l education classes, sports teams, or other forms o f exercise (Table 2.1). N o t being p h y s i c a l l y active was reported by s imi lar proport ions o f respondents i n G r o u p s 1 through 3. U n i v a r i a t e analyses revealed that increasing p h y s i c a l act ivi ty had signif icant statistical association w i t h increasing l i k e l i h o o d o f sexual act iv i ty , but not w i t h fatherhood (Table 2.2). N o t surpr is ingly , i n accordance w i t h age, the majority o f non-sexual ly active males are i n the earlier grades (Table 2.1), w h i l e those w h o are sexual ly experienced or are fathers are i n the higher grades. S i m i l a r l y , odds ratios for sexual act ivi ty further shows that be ing i n an earlier grade has a protective effect. The trend for fatherhood was not found to be statistically significant. Comparing Groups 1 to 2 to 3, it is evident that grade failure increases and the frequency of skipping school increases. Failing a grade is significantly associated with being sexually active, but not with causing a pregnancy. However, skipping school, especially more than twice, is associated both with being sexually active and with causing a pregnancy. Religiosity is one variable that has a slightly unexpected trend. The non-sexually active males and those who had caused a pregnancy had approximately the same percentage declaring themselves religious (35.4 and 39.8 percent, respectively). But those who had not caused a pregnancy had a lower percentage (26.3 percent). This trend is emphasized in the univariate analysis where sexual activity is associated with low religious sentiment (OR = 1.46), but fatherhood is less associated with low religious sentiment (OR = 0.54). Self-esteem is low in a larger proportion of fathers than in the other two groups. High self-esteem is found greatest among the non-sexually active males of Group 1. Low self-esteem is associated with both becoming sexually active and fatherhood. 28 TABLE 2.1 - Physical Activities, School Achievement and Self-Esteem: Frequency Distribution Group 1 Group 2 Group 3 physical activities * very active 27.2 34.9 32.2 somewhat 55.2 47.3 48.9 not active 17.5 17.9 18.9 grade 12 10.9 28.1 28.2 11 13.5 23.1 21.5 10 16.8 17.3 14.9 9 18.4 13.7 16.0 8 17.7 11.0 16.6 7 22.8 6.8 2.8 grade failure ever 13.9 29.0 35.2 skipping school never 76.7 40.6 23.8 1 to 2 times 17.3 33.4 28.2 3 or more 6.1 26.0 48.0 religiosity little or none 35.4 26.3 39.8 self-esteem high 48.8 35.4 28.3 medium 46.9 59.1 61.3 low 4.3 5.5 10.4 * includes physical education classes, participation in team sports and exercise 29 TABLE 2.2 -Physical Activities, School Achievement and Self-Esteem: Univariate Analy sexual activity fatherhood OR 95% CI OR 95% CI physical activities * very some not active 1.00 0.67 (0.60,0.75) 0.80 (0.70,0.93) 1.00 1.12 (0.79,1.58) 1.14 (0.74,1.77) grade 12 11 10 9 8 7 1.00 0.66 (0.56,0.78) 0.40 (0.34,0.47) 0.29 (0.25,0.35) 0.25 (0.21,0.30) 0.11 (0.09,0.14) 1.00 0.93 (0.61, 1.44) 0.86 (0.53, 1.40) 1.17 (0.73,1.88) 1.50 (0.94,2.42) 0.40 (0.16, 1.03) grade failure never ever 1.00 2.60 (2.30,2.93) 1.00 1.33 (0.96, 1.83) skip school never 1 to 2 times 3 or more 1.00 3.73 (3.31,4.20) 8.91 (7.64, 10.40) 1.00 1.44 (0.95,2.19) 3.16 (2.16,4.61) religiosity some little \u00E2\u0080\u00A2 1.00 1.46 (1.31,1.62) 1.00 0.54 (0.40,0.74) self-esteem high medium low 1.00 1.77 (1.59,1.96) 1.91 (1.52,2.41) 1.00 1.29 (0.91, 1.83) 2.35 (1.33,4.17) * includes physical education classes, participation in team sports and exercise 30 C. Physical Health, Chronic Illness/Disability and Mental Health (Tables 3.1 and 3.2) More of the fathers proportionately have many physical health problems and physical health problems that limit their activities (Table 3.1). Having physical health problems is positively associated with sexual activity and fatherhood. In accordance, physical health problems that limit activity is also associated with both sexual activity and fatherhood. Emotional and psychological measures such as emotional distress tend to be reported more frequently among fathers than non-fathers. They are also more often found among sexually active than non-sexually active males. Of note is abuse, both physical and sexual, which is highest among fathers and then next among non-fathers. These variables are strongly associated with earlier onset of both sexual activity and fatherhood. TABLE 3.1 - Physical Health, Chronic Illness/Disability and Mental Health: Frequency Distribution Group 1 Group 2 Group 3 physical health problems * none 13.1 10.5 7.2 some 75.1 74.4 55.2 many 11.8 14.8 37.6 limiting physical health problems any 15.8 17.9 33.3 other health problems f any 24.6 34.6 46.4 other limiting health problems any 4.8 6.4 10.0 emotional distress severe 4.1 7.4 16.0 physically abused ever 10.8 18.2 34.5 sexually abused ever 2.2 5.6 23.3 31 TABLE 3.2 Analysis -Physical Health, Chronic Illness/Disability and Mental Health: Univariate sexual activity fatherhood OR 95% CI OR 95% CI physical health problems * none 1.00 1.00 some 1.20 (1.03,1.40) 1.12 (0.62, 2.03) many 1.72 (1.42,2.08) 3.86 (2.09, 7.13) limiting physical health problems none 1.00 1.00 any 1.24 (1.10, 1.41) 2.25 (1.62,3.12) other health problems f none 1.00 1.00 any 1.69 (1.52, 1.87) 1.63 (1.20, 2.21) other limiting health problems none 1.00 1.00 any 1.42 (1.16, 1.75) 1.60 (0.96, 2.68) emotional distress low to 1.00 1.00 severe 2.02 (1.65,2.48) 2.39 (1.56,3.67) physically abused never 1.00 1.00 ever 1.97 (1.72,2.25) 2.37 (1.71,3.29) sexually abused never 1.00 1.00 ever 3.27 (2.55, 4.20) 5.10 (3.45, 7.54) * includes allergies, infections, and chronic conditions | includes attention deficit disorder and learning disabilities D. Behaviours which result in Intentional and Unintentional Injuries (Tables 4.1 and 4.2) Unsafe behaviours include not wearing a seatbelt or safety helmet and swimming in unsupervised areas. Such behaviours were reported more commonly by sexually active respondents and young fathers. The univariate odds ratios in this respect were substantial and statistically significant. In addition, carrying weapons, involvement in physical fights, and drinking and driving are injury-causing behaviours that appeared positively associated sexually activity and fatherhood. TABLE 4.1 - Behaviours which result in Intentional and Unintentional Injuries: Frequency Distribution Group 1 Group 2 Group 3 unsafe behaviours f little 22.2 10.9 7.8 (in past 12 months) some 56.3 48.5 37.8 many 21.4 40.6 54.4 carried a weapon never 79.9 60.0 32.4 (in past 30 days) 1 to 3 times 11.5 18.2 18.6 4 or more 8.7 21.7 49.2 physical fights never 63.1 40.0 24.9 (in past 12 months) 1 to 3 times 29.5 40.0 28.8 4 or more 7.3 19.9 46.3 drinking & driving never 84.2 50.1 25.4 once 11.4 24.0 20.4 2 or more 4.4 25.9 54.1 TABLE 4.2 -Behaviours which result in Intentional and Unintentional Injuries: Univariate Analysis sexual activity fatherhood OR 95% CI OR 95% CI unsafe behaviours f few 1.00 1.00 (in past 12 months) some 1.76 (1.52,2.05) 1.09 (0.61, 1.98) many 4.04 (3.44, 4.73) 1.88 (1.06,3.35) carried a weapon none 1.00 1.00 (in past 30 days) 1 to 3 times 2.19 (1.91,2.52) 1.91 (1.22,2.97) 4 or more 3.77 (3.27, 4.34) 4.21 (2.97, 5.97) physical fights none 1.00 1.00 (in past 12 months) 1 to 3 times 2.15 (1.93,2.40) 1.16 (0.77, 1.75) 4 or more 4.85 (4.16, 5.65) 3.74 (2.55, 5.48) drinking & driving never 1.00 1.00 once 3.62 (3.17, 4.14) 1.68 (1.08,2.62) 2 or more 11.10 (9.38, 13.12) 4.13 (2.87, 5.94) f includes not wearing safety helmets or seatbelts and swimming in unsupervised areas 33 E. Tobacco, Alcohol and Drug Use (Tables 5.1 and 5.2) Cigarette, alcohol and recreational drug use exhibit similar trends of increase from Group 1 of non-sexually active males to Group 2 of non-fathers to Group 3 of fathers. Fathers begin engaging in these behaviours earlier than non-fathers who in turn begin earlier than non-sexually active males. Lifetime and current use tends to increase from Group 1 to Group 2 to Group 3. The younger the male when he first smokes, drinks or tries other illicit drugs, the greater the odds ratio that the male is sexually active or causes a pregnancy. Current use that is high (greater than 3 times) also has the same relationship. Riskier behaviours such as binge drinking show high odds ratios as well. Confidence intervals tend to be quite large, since for many of these variables, numbers were not very large. TABLE 5.1 - Tobacco, Alcohol and Drug Use: Frequency Distribution Group 1 Group 2 Group 3 ever smoked, age never 69.5 31.0 19.6 10 or less 7.4 22.9 42.4 11 to 14 18.7 36.1 29.6 15 or older 4.3 9.9 8.4 regular smoker, age never 92.2 62.6 40.1 10 or less 0.5 4.5 16.3 11 to 14 5.1 22.6 31.6 15 or older 2.2 10.4 11.9 smoking, current never 88.1 58.5 36.5 (in past 30 days) 1 to 2 times 4.5 7.4 7.2 3 to 19 3.8 10.1 11.1 20 to 30 3.6 23.9 45.3 alcohol, age of initiation never 41.4 5.6 3.4 10 or less 17.3 34.6 59.6 11 to 14 31.9 48.6 31.4 15 or older 9.4 11.2 5.7 34 TABLE 5.1 (continued) Group 1 Group 2 Group 3 alcohol, lifetime use none 44.2 7.9 4.4 1 to 2 times 16.7 9.1 6.6 3 or more 39.0 83.0 88.9 alcohol, current use none 72.2 28.8 16.6 (in past 30 days) 1 to 2 times 17.0 23.9 13.3 3 or more 10.8 47.2 70.2 binge drinking none 87.0 45.4 27.6 (in past 30 days) 1 to 2 days 9.4 27.7 22.7 3 or more 3.7 26.9 49.6 marijuana, age of initiation never 87.7 40.8 22.8 10 or less 0.9 7.9 25.0 11 to 14 7.1 32.4 35.6 15 or older 4.2 18.9 16.6 marijuana, lifetime use none 87.8 40.8 22.9 1 to 2 times 4.8 12.3 7.3 3 or more 7.4 46.9 69.9 marijuana, current use none 95.1 66.9 41.9 (in past 30 days) 1 to 2 times 2.9 12.5 17.9 3 or more 2.0 20.6 40.2 cocaine, age of initiation never 99.0 85.0 52.2 10 or less 0.1 1.5 10.6 11 to 14 0.5 5.2 19.1 15 or older 0.5 8.3 18.0 cocaine, lifetime use none 99.0 85.3 53.9 1 to 2 times 0.7 8.2 13.9 3 or more 0.2 6.5 32.2 cocaine, current use none 99.7 95.6 66.9 (in past 30 days) 1 to 2 times 0.1 2.5 11.6 3 or more <0.1 1.9 21.6 other drugs, lifetime use none 94.4 62.8 36.5 1 to 2 times 3.4 14.9 7.2 3 or more 2.2 22.2 56.3 35 TABLE 5.2 -Tobacco, Drug and Alcohol Use: Univariate Analysis sexual activity fatherhood OR 95% CI OR 95% CI ever smoked, age never 10 or less 11 to 14 15 or older 1.00 7.51 (6.44,8.75) 4.38 (3.88,4.95) 5.19 (4.25,6.35) 1.00 2.94 (1.94,4.45) 1.30 (0.84,2.02) 1.35 (0.72,2.51) regular smoker, age never 10 or less 11 to 14 15 or older 1.00 16.25 (10.39,25.40) 6.94 (5.90,8.17) 7.11 (5.62,8.99) 1.00 5.74 (3.57,9.25) 2.18 (1.52,3.14) 1.78 (1.07,2.96) smoking, current never (in past 30 days) 1 to 2 times 3 to 19 20 to 30 1.00 2.53 (2.06,3.11) 4.24 (3.46,5.18) 10.89 (9.10, 13.02) 1.00 1.55 (0.84,2.87) 1.75 (1.04,2.94) 3.04 (2.16,4.26) alcohol, age of initiation never 10 or less 11 to 14 15 or older 1.00 16.01 (13.11,19.54) 11.29 (9.33,13.67) 8.77 (6.95,11.06) 1.00 2.86 (1.23,6.65) 1.08 (0.45,2.54) 0.83 (0.30,2.34) alcohol, lifetime use none 1 to 2 times 3 or more 1.00 3.09 (2.51,3.81) 12.32 (10.50, 14.46) 1.00 1.30 (0.52,3.26) 1.92 (0.93,3.96) alcohol, current use none (in past 30 days) 1 to 2 times 3 or more 1.00 3.52 (3.08,4.03) 11.68 (10.25,13.32) 1.00 0.96 (0.56, 1.67) 2.58 (1.71,3.88) binge drinking none (in past 30 days) 1 to 2 days 3 or more 1.00 5.77 (5.04,6.62) 15.47 (12.93, 18.50) 1.00 1.35 (0.88,2.06) 3.04 (2.12,4.35) marijuana, age of initiation never 10 or less 11 to 14 15 or older 1.00 21.32 (15.38,29.56) 10.23 (8.86,11.83) 9.80 (8.19,11.73) 1.00 5.64 (3.59, 8.85) 1.96 (1.31,2.94) 1.58 (0.97,2.57) marijuana, lifetime use none 1 to 2 times 3 or more 1.00 5.54 (4.60,6.67) 14.48 (12.62, 16.61) 1.00 1.05 (0.56, 1.99) 2.66 (1.85,3.82) 36 TABLE 5.2 (continued) sexual activity fatherhood OR 95% CI OR 95% CI marijuana, current use none (in past 30 days) 1 to 2 times 3 or more 1.00 6.39 (5.20,7.86) 16.28 (12.98,20.43) 1.00 2.28 (1.48,3.51) 3.12 (2.22,4.37) cocaine, age of initiation never 10 or less 11 to 14 15 or older 1.00 29.58 (10.80,80.99) 15.18 (9.77,23.58) 22.92 (14.77,35.56) 1.00 11.62 (6.40,21.08) 5.96 (3.87,9.20) 3.51 (2.29,5.39) cocaine, lifetime use none 1 to 2 times 3 or more 1.00 15.14 (10.41,22.03) 35.87 (20.44,62.96) 1.00 2.68 (1.68,4.26) 7.91 (5.49,11.40) cocaine, current use none (in past 30 days) 1 to 2 times 3 or more 1.00 22.25 (10.29,48.09) 31.77 (13.08,77.18) 1.00 6.68 (3.93,11.37) 16.36 (10.24,26.13) other drugs, lifetime use none 1 to 2 times 3 or more 1.00 6.51 (5.36,7.90) 17.32 (13.96,21.49) 1.00 0.83 (0.45, 1.52) 4.36 (3.15,6.04) F. Sexuality, Contraceptive Use and Sexually Transmitted Diseases (Tables 6.1 and 6.2) Comparisons regarding sexuality, contraceptive use and sexually transmitted diseases were limited to the sexually active respondents (Groups 2 and 3). Relative to non-fathers, a greater proportion of fathers had an earlier onset of sexual activity. Fathers had more partners, and participated in risk behaviours, such as using alcohol and recreational drugs prior to intercourse, and not using condoms or other contraceptives. In terms of the univariate odds ratios this was reflected in statistically significant associations with respect to both initiation of 37 sexual activity, number of partners, contraceptive use, and sexually transmitted diseases. TABLE 6.1- Sexuality, Contraceptive Use, and Sexually Transmitted Diseases: Frequency Distribution Group 1 Group 2 Group 3 ever had intercourse 0.0 100.0 100.0 age of initiation 12 or less X 22.1 47.3 13 to 15 X 54.0 43.4 16 or more X 23.8 9.4 lifetime partners 1 to 2 X 59.8 19.0 more than 2 X 40.3 81.0 current partners not active X 43.2 10.1 (in past 3 months) 1 X 41.8 36.9 more than 1 X 15.0 53.1 alcohol/drug use before sex X 27.4 42.7 no condom use X 33.7 61.2 contraceptive use pill X 21.2 21.7 condom X 57.5 30.6 other/not sure X 10.9 20.5 none X 10.4 27.2 caused a pregnancy never X 100.0 0.0 once X 0.0 60.8 more than once X 0.0 39.2 sexually transmitted infection ever X 2.7 16.6 38 TABLE 6.2 - Sexuality, Contraceptive Use, and Sexually Transmitted Diseases: Univariate Analysis fatherhood OR 95% CI age of initiation 16 or more 13 to 15 12 or less 1.00 2.02 (1.18,3.45) 5.37 (3.15,9.16) lifetime partners 1 to 2 2 or more 1.00 6.33 (4.32,9.28) current partners not active (in past 3 months) 1 more than 1 1.00 3.78 (2.23,6.41) 15.16 (9.03,25.44) alcohol/drug use before sex no yes 1.00 1.97 (1.45,2.69) condom use yes no 1.00 3.11 (2.27,4.26) contraceptive use pill none condom other 1.00 2.57 (1.64,4.01) 0.52 (0.34,0.80) 1.85 (1.15,2.98) sexually transmitted infection never ever 1.00 7.19 (4.51,11.44) In summary, the frequency tables and univariate analyses have shown that risky behaviours increase from Group 1, the non-sexually active males, to Group 2, the non-fathers, to Group 3, the fathers. Risky behaviours include low school achievement and attendance, abuse, injury-causing behaviours, use of tobacco, alcohol and recreational drugs, and behaviours associated with sexual activity such as multiple partners, and contraceptive use. Other variables that are often associated with disadvantaged backgrounds, such as particular ethnic groups, 39 fragmented family structures, and parents who are not well-educated, are found more often among the males who are sexually active or have caused a pregnancy. Univariate odds ratios have highlighted those trends observed among the frequency distributions with effects that were substantial and statistically significant. G. Multivariate Analysis - Sexual Activity (Table 7.1) In the final multivariate model for variables associated with sexual activity, there are a few demographic variables remaining. The majority are related to school or risk-taking behaviours such as injury-causing behaviours and use of tobacco, alcohol and recreational drugs (Table 7.1). Two provincial regions stand out: relative to Greater Vancouver and independent of all other variables in the model, more respondents from the Capital and Fraser Valley are sexually active . Also, adjusting for all variables in the model, younger age reduces the likelihood of being sexually active. Asian ethnicity, analogous with the univariate analyses, has a protective effect with regards to sexual activity, while due to the very small number of Hispanics in the survey sample, that particular odds ratio should be interpreted with caution. Furthermore, family structure is an independent and statistically significant factor of sexual activity among young males: those not in two-parent households, being more likely to be sexually active. Grade level, independent of age and the other variables in the model, demonstrates the effect that, like age, earlier grades are less likely to be associated with sexual activity than higher grades. In addition, failing a grade and skipping school are associated with increased likelihood of being sexually active. Little or no religiosity also shares this association. 40 Being not or only moderately physically active shows a decreased association with sexual activity. Participation in risk behaviours such as weapon carrying, fighting, drinking and driving, and binge drinking, were all independently associated with increased sexually activity. A s well, use of tobacco, alcohol, marijuana, cocaine, and other recreational drugs, were each independently associated with increased sexual activity. TABLE 7.1 - Multivariate Analysis by Sexual Activity OR 95% CI provincial region Greater Vancouver 1.00 Capital 1.35 (1.01, 1.80) Fraser Valley 1.36 (1.03, 1.80) Interior 1.14 (0.89, 1.47) Kootenays 0.87 (0.64, 1.18) Northeast 1.34 (1.00, 1.79) Northwest 1.09 (0.72, 1.65) Upper Island 1.23 (0.95, 1.58) age group 17 or older 1.00 15 to 16 0.77 (0.58, 1.03) 14 or less 0.54 (0.34, 0.85) ethnicity none 1.00 European only 1.02 (0.83, 1.27) Asian only 0.62 (0.46, 0.85) Hispanic only 5.32 (1.79, 15.80) Native American only 1.49 (0.86, 2.59) other only 0.99 (0.70, 1.41) multiple 0.95 (0.76, 1.19) family structure two parent 1.00 single parent 1.21 (1.02, 1.44) alternate structure 1.77 (1.30,2.41) alone 3.27 (1.04, 10.31) 41 TABLE 7.1 (continued) OR 95% CI grade twelve 1.00 eleven 0.65 (0.50, 0.84) ten 0.38 (0.27, 0.55) nine 0.35 (0.23, 0.54) eight 0.45 (0.27, 0.74) seven 0.28 (0.16,0.48) grade failure no 1.00 yes 1.62 (1.34, 1.98) skipping school never 1.00 once or twice 1.39 (1.17, 1.65) 3 or more 2.31 (1.84,2.90) religiosity some to very 1.00 little or none 1.36 (1.16, 1.60) physically active very 1.00 somewhat 0.63 (0.53, 0.75) not active 0.41 (0.32, 0.52) weapon carrying none 1.00 1 to 3 days 1.26 (1.02, 1.55) 4 or more 1.45 (1.17, 1.81) physical fights none 1.00 1 to 3 times 1.61 (1.37, 1.90) 4 or more 2.77 (2.16,3.57) drinking and driving never 1.00 once 1.17 (0.96, 1.41) 2 or more 1.51 (1.15, 1.97) regular smoking, age of never 1.00 10 or under 3.14 (1.66, 5.94) 11 to 14 1.52 (1.20, 1.93) 15 or older 1.35 (0.97, 1.86) alcohol, age of initiation never 1.00 10 or under 2.40 (1.50,3.85) 11 to 14 2.18 (1.37,3.46) 15 or older 1.72 (1.05,2.82) 42 TABLE 7.1 (continued) OR 95% CI binge drinking none 1 or 2 days 3 or more 1.00 1.27 (1.04,1.54) 1.56 (1.19,2.06) marijuana, age of initiation never 10 or under 11 to 14 15 or older 1.00 3.19 (1.96,5.18) 1.79 (1.42,2.25) 1.58 (1.23,2.04) cocaine, age of initiation never 10 or under 11 to 14 15 or older 1.00 29.26 (3.65,234.48) 2.86 (1.49,5.50) 1.62 (0.94,2.80) other drugs, lifetime use never 1 or 2 times 3 or more 1.00 1.67 (1.29,2.17) 1.99 (1.44,2.74) H. Multivariate Analysis - Fatherhood (Table 8.1) Among the sexually active subgroup, no variables of individual and family demographics, age, grade and school performance appeared independently associated with fatherhood. Family structure appears associated with fatherhood independent of all other variables in the model (Table 8.1). In particular, those males who live alone are most likely to cause a pregnancy. Fathers were also more likely to report a limiting physical health condition. Risk behaviours such as weapon carrying, and drinking and driving, were independent associates of pregnancy, as they were for sexual activity (Table 7.1 and 8.1). Use of tobacco, alcohol, and marijuana were not associated with fatherhood. In contrast, current use of cocaine and lifetime 43 use of other illicit drugs such as hallucinogens and amphetamines were associated with fatherhood. Several variables relate to sexuality, including number of partners, contraceptive and condom use, sexually transmitted infections and age of initiation of intercourse, appeared associated with fatherhood. Sexual abuse stands out. Young males who reported sexual abuse had an odds ratio of 2.43 for fatherhood. 44 TABLE 8.1 - Multivariate Analysis by Fatherhood OR 95% CI family structure two parent 1.00 one parent 0.87 (0.55, 1.38) alternate structure 1.39 (0.79, 2.45) alone 3.37 (1.18,9.60) limiting physical health conditions none 1.00 any 1.57 (1.02,2.41) sexual abuse never 1.00 ever 2.43 (1.41,4.18) weapon carrying none 1.00 1 to 3 days 1.14 (0.66, 1.96) 4 or more 1.86 (1.17,2.96) drinking and driving never 1.00 once 1.74 (1.01,3.00) 2 or more 2.00 (1.21,3.31) cocaine, current use never 1.00 (in past 30 days) once or twice 2.61 (1.26, 5.39) 3 or more 2.67 (1.27, 5.61) other drugs, lifetime use never 1.00 1 or 2 times 0.59 (0.29, 1.22) 3 or more 1.71 (1.04,2.82) sexual intercourse, age of initiation 16 or older 1.00 13 to 15 1.45 (0.76, 2.75) 12 or under 2.54 (1.27,5.06) sexual partners, lifetime 1 or 2 1.00 3 or more 1.93 (1.13,3.29) sexual partners, current none 1.00 one 4.05 (2.15, 7.63 2 or more 7.11 (3.56, 14.21 condom use yes 1.00 no 2.06 (1.04, 4.08) contraceptive use pill 1.00 none 2.44 (1.34,4.45) condom 1.02 (0.50, 2.08) other/do not know 1.55 (0.85,2.82) sexually transmitted infection never 1.00 ever 3.34 (1.68, 6.63) 4 5 IV. DISCUSSION The primary intent of this thesis is to describe the adolescent father and to higlight risk factors to address prevention with regards to young fatherhood. Such prevention may occur at two levels: first, at the point of initiation of sexual activity, and second, in the practice of contraception and prophylaxis. Comparisons were therefore made between adolescent males who are and are not sexually active, as well as between those sexually active males who have and have not caused a pregnancy. Descriptive data was first reported to provide an overall picture. Univariate logistic regression analysis was conducted in order to show associations with sexual activity or with fatherhood. Further multivariate analysis removed confounding factors and identified the most important demographic and behavioural factors associated with sexual activity or fatherhood. The descriptive analyses have shown a general trend of increase of risky behaviours across the three groups of non-sexually active males to non-fathers to fathers. Risky behaviours are defined as behaviours that constitute transgressions of societal and/or legal norms that tend to elicit some sort of social control response (53, page 80). Jessor and Jessor describe such behaviours as problematic (53). These not only include behaviours related to tobacco, alcohol and drug use and violence, but also school performance such as failure and skipping. The univariate analyses emphasized those variables that shared a statistical relationship with either sexual activity or fatherhood. The same trend observed in the descriptive analyses of an increase of risky behaviours continued in this analysis. However, while fatherhood and sexual 46 activity both were related to measures of violent behaviours and tobacco, alcohol and drug use, fewer other measures, such as demographics, were related to fatherhood as to sexual activity. Prior to the final multivariate models that would highlight the most important variables, intermediate multivariate models were constructed for each of the six categories of variables of: 1. individual and family demographics 2. physical activities, school achievement and self-esteem 3. physical health, chronic illness/disability and mental health 4. behaviours which result in intentional and unintentional injuries 5. tobacco, alcohol and drug use 6. sexuality, contraceptive use and sexually transmitted diseases Since there are various interrelations between measures within each of these categories, the intermediate multivariate analyses reduced the large number of variables and identified those that were statistically most important. These factors, representing the most important effects of each of the six categories were then used to construct the final multivariate model which resulted in interpretable outcomes. In contrast, the construction of a multivariate model where all variables were thrown together resulted in a very different and difficult to interpret results. These results are then discussed below. A. Sexual Activity Of all males in the survey, 2,526 (34.4 percent) were sexually active. The provincial regions of the Capital and Fraser Valley had adolescent males who were more likely to be 47 sexually active than in Greater Vancouver. Apparently, sexual activity is not a phenomenon that is isolated to urban areas. The youngest age group were the least likely to be sexually active, with the highest sexual activity in the oldest group. This seems a reasonable relationship and shows the important of controlling for age in such studies. Those in earlier grades, like younger age, are less likely to be sexually active. However, because both these measures are independently related to sexual activity, another effect must relate to grade that is not accounted for by age. Students in the same grade are not necessarily of the same age, and therefore, peer groups will be different. This would lead to differing cohort experiences. Asian ethnicity showed a protective effect. The Asian population in British Columbia is fairly young, therefore whether this effect will continue with subsequent generations is not known. Hispanics constituted such a small percentage of the sample, therefore, though the odds ratios and confidence intervals are statistically significant for this ethnic group, it is not significant in the context of British Columbia. While those of Native American background showed a relationship in the univariate analysis, this does not continue into the multivariate model, indicating that other factors may place this ethnic group at a disadvantage to others in the province or that statistically, the numbers were too small to have an effect in the multivariate model. Typically, the usual explanation given for lower level of sexual activity in Greater Vancouver is attributed to the large Asian population, but it is now clear that there must be other factors at play. The question of ethnicity may have been interpreted slightly differently by respondents, leading to misclassification which would underestimate the true effect of ethnicity. 48 Young males who came from two-parent families were least likely to be sexually active. Whether this is because it is more likely that both parents can share the responsibilites of childrearing or another factor that is related to a two-parent family structure cannot be determined. Unlike family structure, marital status appeared to not be relevant as long as both parents were present. Those males who lived alone were relatively small in number, and therefore had large confidence interval, but it is still evident that these males who are unsupervised or without a guardian are most likely to be sexually active. Two academic measures, grade failure and skipping school indicate that these males may have difficulties with school, or may not value education. Because they are not in school, they may be more likely to have the time to participate in other behaviours. Having little or no religious feeling was related to being sexually active. While being of a particular religion may not be associated with becoming sexually active, feeling religious may indicate the degree of influence religion has on the decision to become sexually active. While it may seem that those who are not physically active would have more time, and therefore more likely to be sexually active, in this analysis, those who were very physically active were more likely to be sexually active. There may be underlying factors that indicate these physically active males are willing to spend more time on improving their physical appearance through exercise, in part due to a desire to attract females. Therefore, they are more likely to have the opportunity and desire to be sexually active. Violent behaviours such as weapon carrying, physical fighting and drinking and driving fall under risk taking behaviours. Those adolescent males who were to engage in such behaviours were more likely to be found among those who are sexually active. In fact, the more 49 often such behaviours were reported, the greater the odds ratio to be sexually active. As per Jessor and Jessor's problem behaviour theory, males who are more willing to take these violent risks may be more willing to take risks in other aspects of their life, such as sexual activity (53). Use of tobacco, alcohol and drugs are another group of risk taking behaviours. Measures of use of these substances were for age of initiation, lifetime and current use. Age of initiation of such behaviours were most important, according to this multivariate model. Other drug use, such as hallucinogens and amphetamines, was only measured by lifetime use, and also was included in the sexual activity model. Cocaine use was the most extreme of all drug use surveyed, the others being tobacco, alcohol and marijuana. While fewer males began smoking, drinking or using illicit drugs at younger ages, less than 10 years old, these were the males who were more likely to be sexually active. Binge drinking is a clear example of why drug use and sexual activity are often tied together. When one is intoxicated, one's judgment is impaired, and therefore, engaging in sexual intercourse may be more likely. There may also be an increased likelihood of being coerced into behaviours that normally one would not participate in. B. Fatherhood Adolescent fathers were defined as a male of adolescent years who causes a pregnancy, regardless of the outcome of the pregnancy and whether he participated in the upbringing of the child. The second set of analyses focussed on only the 2,526 sexually active males to reveal their associations with fatherhood. The variables found to be independently related were less numerous and different from those found in the previous analysis of sexual activity. 50 Family structure remained an important demographic factor. This is in large part due to the strong relationship between those who lived alone and fatherhood, indicating those adolescent males who were without any guardian were at higher risk of becoming fathers than other adolescent males. Whether males who are living alone are more likely to be sexually active and cause pregnancies, or are sexually active and cause pregnancies and then find themselves without a home cannot be distinguished from this analysis. Those males who had physical health conditions that limited their life in any way were also found in a higher proportion among fathers than non-fathers. It could be speculated that in order to overcome such limitations, be they diabetes, epilepsy, respiratory problems, allergies or other conditions, these males may, try to prove that they are like everyone else by engaging in behaviours that they believe everyone else is doing, such as sexual intercourse. As well, having a limiting physical health condition requires a great deal of vigilance to prevent any health complications, and it is possible that these males may be fed up with such vigilance and become careless in other aspects of their life, such as caring for their sexual health. There may also be a relationship between those males who tend to suffer from such physically limiting conditions and the environment from which they come. They may be socially disadvantaged, with limited opportunities and less control of their lives, and having a physical health condition may further put them at a disadvantage. Sexual abuse had a clear relationship with fatherhood. Those males who had been sexually abused may have affected psychologically and emotionally and their relationships suffer as a result. In adolescent females, it has been observed that there is a link between abuse and sexual acting out (26). Theses females also had a higher risk of social malfunctioning and 51 feelings of powerlessness and loss of control of self so that they do not take care of themselves such as through contraception (6, 26). It is possible that adolescent males could suffer similar effects. Jessor and Jessor's problem behaviour theory (53) once again applies in this situation of fatherhood. Weapon carrying, drinking and driving, other drug use and current cocaine use, and variables relating to sexual activity itself, such as age of initiation, partners and contraceptive and condom use, and sexually transmitted infections, are all related to fatherhood. These males do not engage in only one risky behaviour, but in fact several. These analyses have confirmed that being sexually active or causing a pregnancy is not a behaviour that exists alone. While Jessor and Jessor did not study the same population, their problem behaviour theory applies in the context of British Columbia to a large degree. Because such behaviours do not exist alone, the presence of some indicates the likelihood that others are present. The mechanisms and interactions of such variables is unclear, but has clear implications for intervention efforts and prevention planning and programs. Factors important for describing sexual activity and fatherhood are in accordance with the problem behaviour theory except for religiosity. In the univariate analysis, religiosity was related with delayed sexual activity, but for those who were sexually active, those who were religious were more likely to be fathers. This is possibly due to sexually active religious males not having adequate knowledge of contraceptives to use them properly. This may explain that when contraceptive use is controlled for, the association between religiosity and fatherhood is no longer statistically significant. 52 C. Limitations The size of the Adolescent Health Survey is large. With its number of participants, it has great statistical power. This is an enormous asset, given the many variables available and the potential to evaluate them simultaneously. However, studies with such great statistical power may reveal associations to be statistically significant, but not of practical relevance since the size of the effect is small. Therefore, the size of the odds ratio should be taken as a guide. For example, use of cocaine. The survey collected information both anonymously and confidentially, so names or other unique identifiers were not recorded. Since participants were aware of this, response bias may have been minimized. However, the survey is cross-sectional in design, and therefore inferences regarding causality should be considered with caution. A case-control or cohort design would be preferred for such purposes. As well, we are relying on accurate self-report for the responses to the questionnaire. It is difficult to corroborate such personal information such as sexual practices. No information was collected regarding age of partner(s), nor when the pregnancy was caused relative to the behaviours currently reported in response to the survey. Therefore, if any changes in behaviour had occurred, they will not be detected. Misclassification may have been a problem as well. Mothers are easier to identify than fathers. If the female partner did not disclose the pregnancy to the male partner, then he will not be able to answer accurately whether he had ever caused a pregnancy. In such cases, the \"father\" will in fact end up being classified as a non-father. Such misclassifications, however, would lead to an underestimation of the actual association between behavioural factors and fatherhood. 53 The questionnaire did not collect any information regarding the female partners of the adolescent fathers. As well, data on older males who have fathered children of adolescent mothers are even less available than data on adolescent males. Previous research has identified various characteristics associated with male partners of adolescent females, both adolescent and adult males, such as age, relationship history, educational and vocational stability, socioeconomic status and sexual history (18, 46, 50, 65). The populations studied were quite specific, or no sampling procedures were used (13, 46, 50, 65), therefore, no definitive criteria have been identified as to particular characteristics of male partners of adolescent mothers. D. Future Directions and Summary Most research has come from the United States, so it was not clear whether the Canadian situation would be similar. For the most part, risk behaviour clustering has been observed in this analysis. Those who engaged in one risky behaviour were more likely to also engage in others. One reason to describe the adolescent father is for prevention purposes. This analysis has shown that most of these risky behaviours cluster together. Some behaviours, such as school difficulties, may serve as early indicators if they were determined to occur in a particular sequence. It would then be easier to target particular males for prevention efforts for risky behaviours in general and to young fathers in particular. For example, students who are truant or have failed a grade, could be identified and then targeted for interventions that would assist them with their academic difficulties that may prevent other problems further on, such as early sexual activity and fatherhood. Similarly, adolescents who have suffered a history of abuse could receive regular counseling with respect to various things including sexuality and pregnancy in 54 their teenage years. Use of alcohol, tobacco and recreational drugs has been identified to be associated with sexual activity in adolescents. Places where youth engage in the use of these substances could provide a place for education and counseling regarding sexual health. By interviewing fathers, it can be determined what events or circumstances occurred that may have led them to fatherhood. While these variables have been identified in multivariate analysis as statistically significant, it is necessary to determine whether they are significant in a practical context. Qualitative research would be helpful in this situation. As well, now that variables of importance with regards to sexual activity and fatherhood have been identified, the reason for their importance needs to be determined. For example, why does family structure stand out as a demographic measure? Do these families require support socially and financially? In summary, these Canadian findings confirm reports from the U.S. Among the sexually active young men, fatherhood was not only associated with use of contraceptives, early age of sexual activity and history of sexually transmitted infections, but also with perceived poor health status and history of sexual abuse. Religiosity was associated with delay sexual activity, whereas among the sexually active, religious youth were more likely to be fathers. The phenomenon of clustering of risk behaviours is clearly observed. This study confirms results reported elsewhere, as well as identifying new factors not previously reported, and provides clues for prevention and counseling. Such clues are the basis for some of the recommendations made, such as for students having academic difficulties being identified and targeted for prevention efforts, or students who have suffered a history of abuse being given regular counseling, or using venues where adolescents gather to consume alcohol or 55 other substances as a place to provide information and counseling with regards to their sexual health. General survey data has thus proved useful here to determine relationships, but the future research should focus on determining the causal relationships involved. 56 REFERENCES 1. Amini S. B., P. M . Catalano, L. J. Dierker, and L. J. Mann. 1996. Births to teenagers: Trends and obstetric outcomes. Obstetrics and Gynecology, 87(5, Pt l):668-674. 2. Barker G. K. , and S. Rich. 1992. Influences on adolescent sexuality in Nigeria and Kenya: findings from recent focus-group discussions. Studies in Family Planning, 23(3): 199-210. 3. Barret R. L. , and B. E. Robinson. 1982. A descriptive study of teenage expectant fathers. Family Relations, 31:349-352. 4. Barth R. P., M . Claycomb, and A. Loomis. 1988. Services to adolescent fathers. Health and Social Work, 13(4):277-287. 5. Blum R., L . J. Harris, M . D. Resnick, and K. Rosenwinkel. 1989. Technical Report on the Adolescent Health Survey. Minneapolis: University of Minnesota. 6. Boyer D., and D. Fine. 1992. Sexual abuse as a factor in adolescent pregnancy and child maltreatment. Family Planning Perspectives, 24(1):4-11, 19. 7. Brener N . D., J. L . Collins, L. Kann, C. W. Warren, and B. I. Williams. 1995. Reliability of the youth risk behavior survey questionnaire. American Journal of Epidemiology, 141(6):575-580. 8. British Columbia Provincial Health Officer. 1995. Provincial Health Officer's Annual Report: Report on the Health of British Columbians. Victoria, B.C.: British Columbia Ministry of Health and Ministry Responsible for Seniors. 9. Card J. J., and L. L . Wise. 1978. Teenage mothers and teenage fathers: The impact of early childbearing on the parents' personal and professional lives. Family Planning Perspectives, 10(4): 199-205. 10. CarreraM. A. 1992. Involving adolescent males in pregnancy and STD prevention programs. Adolescent Medicine: State of the Art Reviews, 3(2):269-281. 11. Centers for Disease Control and Prevention. 1992. Sexual behavior among high school students-United States, 1990. Morbidity and Mortality Weekly Report, 40(51 & 52):885-888. 12. Dearden K. , C. Hale, and J. Alvarez. 1992. The educational antecedents of teen fatherhood. British Journal of Educational Psychology, 62:139-147. 57 13. Dearden K. A. , C. B. Hale, and T. Woolley. 1995. The antecedents of teen fatherhood: a retrospective case-control study of Great Britain youth. American Journal of Public Health, 85(4):551-554. 14. Division of Adolescent and School Health, Center for Chronic Disease Prevention and Health Promotion. 1989, revised 1990. Handbook for conducting youth risk behavior surveys. Atlanta, GA: Centers for Disease Control. 15. Dryfoos J. G. 1990. Adolescents at Risk: Prevalence and Prevention. New York: , Oxford University Press. 16. Dryfoos J. G. 1988. Putting the Boys in the Picture: A Review of Programs to Promote Sexual Responsibility among Young Males. Santa Cruz, CA: Network Publications. 17. East P. L. , and M . E. Felice. 1996. Adolescent Pregnancy and parenting: Findings from a Racially Diverse Sample. Mahwah, NJ: Lawrence Erlbaum Associates. 18. Elster A . B., and M . E. Lamb, eds. 1986. Adolescent Fatherhood. Hillsdale, NJ: Lawrence Erlbaum Associates. 19. Elster A. B., M . E. Lamb, and N . Kimmerly. 1989. Perceptions of parenthood among adolescent fathers. Pediatrics, 83(5):758-765. 20. Elster A. B., M . E. Lamb, L. Peters, J. Kahn, and J. Tavare. 1987. Judicial involvement and conduct problems of fathers of infants born to adolescent mothers. Pediatrics, 79(2):230-234. 21. Elster A . B., M . E. Lamb, and J. Tavare. 1987. Association between behavioral and school problems and fatherhood in a national sample of adolescent youths. The Journal of Pediatrics, 111 (6):932-936. 22. Elster A . B., and S. Panzarine. 1983. Teenage fathers: stresses during gestation and early parenthood. Clinical Pediatrics, 22(10):700-703. 23. Elster A. B. , and S. Panzarine. 1980. Unwed teenage fathers: Emotional and health educational needs. Journal of Adolescent Health Care, 1 (2): 116-120. 24. Flick L. H . 1986. Paths to adolescent parenthood: implications for prevention. Public Health Reports, 101(2): 132-147. 25. Foster V. , and N . A . Sprinthall. 1992. Developmental profiles of adolescents and young adults choosing abortion: stage sequences, decalage, and implications for policy. Adolescence, 27(107):655-673. 58 26. Gershenson H . P., J. S. Musick, H. S. Ruch-Ross, V . Magee, K. K . Rubino, and D. Rosenberg. 1989. The prevalence of coercive sexual experience among teenage mothers. Journal of Interpersonal Violence, 4(2):204-219. 27. Hanson S. L. , D. R. Morrison, and A. L . Ginsburg. 1989. The antecedents of teenage fatherhood. Demography, 26(4):579-596. 28. Hanvey L. , and D. Avard. 1994. The Health of Canada's Children: A Canadian Institute of Child Health Profile, 2nd ed. Ottawa, Ontario: Canadian Institute of Child Health. 29. Hardy J. B. , A . K. Duggan, K. Masnyk, and C. Pearson. 1989. Fathers of children born to young urban mothers. Family Planning Perspectives, 21 (4): 159-163, 187. 30. Hardy J. B. , and A. K . Duggan. 1988. Teenage fathers and the fathers of infants of urban, teenage mothers. American Journal of Public Health, 78(8):919-922. 31. Hayes C. D., ed. 1987. Risking the Future: Adolescent Sexuality, Pregnancy, and Childbearing, Volume I. Washington, D.C.: National Academy Press. 32. Health Services and Promotion Branch. 1990. Report on Adolescent Reproductive Health. Ottawa, Ontario: Health and Welfare Canada. 33. Hendricks L. E., and T. Montgomery. 1983. A limited population of unmarried adolescent fathers: a preliminary report of their views on fatherhood and the relationship with the mothers of their children. Adolescence, 18(69):201-210. 34. Hendricks L. E. 1982. Unmarried black adolescent fathers' attitudes toward abortion, contraception, and sexuality: a preliminary report. Journal of Adolescent Health Care, 2(3):199-203. 35. Herold E. S. 1984. Sexual Behaviour of Canadian Young People. Markham, Ontario: Fitzhenry & Whiteside. 36. Herold J. M . , M . S. Valenzuela, and L. Morris. 1992. Premarital sexual activity and contraceptive use in Santiago, Chile. Studies in Family Planning, 23(2):128-136. 37. Hofferth S. L. , and C. D. Hayes, eds. 1987. Risking the Future: Adolescent Sexuality, Pregnancy, and Childbearing, Volume II. Washington, D.C.: National Academy Press. 38. Horwitz S. M . , L . V . Klerman, H. S. Kuo, and J. F. Jekel. 1991. Intergenerational transmission of school-age parenthood. Family Planning Perspectives, 23(4): 168-172, 177. 59 39. Jones E. F., J. G. Forrest, N . Goldman, S. Henshaw, R. Lincoln, J. I. Rosoff, C. F. Westoff, and D. Wulf. 1986. Teenage Pregnancy in Industrialized Countries. New Haven, CT: Yale University Press. 40. Jones E. F., J. G. Forrest, S. Henshaw, J. Silverman, and A . Torres. 1989. Pregnancy, Contraception, and Family Planning Services in Industrialized Countries. New Haven, CT: Yale University Press. 41. King A. J. C , R. P. Beazley, W. K. Warren, C. A. Hankins, A. S. Robertson, and J. L. Radford. 1988. Canada Youth and AIDS Study. Kingston, Ontario: Queen's University. 42. King A. J. C , and B. Coles. 1992. The Health of Canada's Youth. Ottawa: Minister of National Health and Welfare. 43. Klein L . 1978. Antecedents of Teen Pregnancy. Clinical Obstetrics and Gynecology, 21(4):1151-1159. 44. ' Kolbe L. J. 1990. An epidemiological surveillance system to monitor the prevalence of youth behaviors that most affect health. Health Education, 21(6):44-48. 45. Lancaster J. B., and B. A . Hamburg, eds. 1986. School-Age Pregnancy and Parenthood: Biosocial Dimensions. Hawthorne, N Y : Aldine de Gruyter. 46. Landry D. J., and J. D. Forrest. 1995. How old are U.S. fathers? Family Planning Perspectives, 27(4): 159-161, 165. 47. Leland N . L. , and R. P. Barth. 1992. Gender differences in knowledge, intentions, and behaviors concerning pregnancy and sexually transmitted disease prevention among adolescents. Journal of Adolescent Health, 13(7):589-599. 48. Lorenzi M . E., L . V . Klerman, and J. F. Jekel. 1977. School-age parents: How permanent a relationship? Adolescence, 12(45):13-22. 49. MacKay H. , and C. Austin. 1983. Single Adolescent Mothers in Ontario. Ottawa: The Canadian Council on Social Development. 50. Males M . , and K . S. Y . Chew. 1996. The ages of fathers in California adolescent births, 1993. American Journal of Public Health, 86(4):565-568. 51. Manitoba Health. 1989. Teenage Pregnancy in Manitoba: A Statistical Report. Winnipeg: Manitoba Health. 52. Marsiglio W. 1987. Adolescent fathers in the United States: Their initial living arrangements, marital experience and educational outcomes. Family Planning Perspectives, 19(6):240-251. 60 53. Matarazzo J. D., S. M . Weiss, J. A . Herd, N . E. Miller, and S. M . Weiss, eds. 1984. Behavioral Health: A Handbook of Health Enhancement and Disease Prevention. New York: John Wiley & Sons. 54. McCreary Centre Society. 1993. Adolescent Health Survey: Province of British Columbia. Vancouver, BC: The McCreary Centre Society. 55. Meikle S, J. A . Peitchinis, and K. Pearce. 1985. Teenage Sexuality. San Diego, C A : College-Hill Press. 56. Miller B. C , J. J. Card, R. L . Paikoff, and J. L. Peterson, eds. 1992. Preventing Adolescent Pregnancy: Model Programs and Evaluations. Newbury Park, C A : SAGE Publications. 57. Mott F. L. , and R. J. Haurin. 1988. Linkages between sexual activity and alcohol and drug use among American adolescents. Family Planning Perspectives, 20(3): 128-136. 58. Nakashima 1.1., and B. W. Camp. 1984. Fathers of infants born to adolescent mothers: a study of paternal characteristics. American Journal of Diseases of Children, 138:452-454. 59. Nord C. W., K. A . Moore, D. R. Morrison, B. Brown, and D. E. Myers. 1992. Consequences of teen-age parenting. Journal of School Health, 62(7):310-318. 60. Orr D. P., M . L . Wilbrandt, C. J. Brack, S. P. Rauch, and G. M . Ingersoll. 1989. Reported sexual behaviors and self-esteem among young adolescents. American Journal of Diseases of Children, 143:86-90. 61. Orton M . J., and E. Rosenblatt. 1986. Adolescent pregnancy in Ontario: Progress in Prevention. Hamilton, Ontario: Planned Parenthood Ontario. 62. Panzarine S., and A. B. Elster. 1983. Coping in a group of expectant adolescent fathers: an exploratory study. Journal of Adolescent Health Care, 4(2): 117-120. 63. Pick S, and P. A . Palos. 1995. Impact of the family on the sex lives of adolescents. Adolescence, 30(119):667-675. 64. Redmond M . A . 1985. Attitudes of adolescent males toward adolescent pregnancy and fatherhood. Family Relations, 34:337-342. 65. Rivara F. P., P. J. Sweeney, and B. F. Henderson. 1985. A study of low socioeconomic status, black teenage fathers and their nonfather peers. Pediatrics, 75(4):648-656. 66. Robbins C , H. B. Kaplan, and S. S. Martin. 1985. Antecedents of pregnancy among unmarried adolescents. Journal of Marriage and the Family, 47:567-583. 61 67. Robinson B. E., R. L . Barret, and P. Skeen. 1983. Locus of control of unwed adolescent fathers versus adolescent nonfathers. Perceptual and Motor Skills, 56:397-398. 68. Robinson B. E., and R. L . Barret. 1982. Issues and problems related to the research on teenage fathers: a critical analysis. Journal of School Health, 52(10):596-600. 69. Robinson B. E. 1988. Teenage. Fathers. Lexington, M A : Lexington Books. 70. Ross A. 1982. Teenage Mothers, Teenage Fathers. Toronto: Personal Library, Publishers. 71. Roye C. F., and S. J. Balk. 1996. The relationship of partner support to outcomes for teenage mothers and their children: a review. Journal of Adolescent Health, 19(2):86-93. 72. Sacks D., J. G. Macdonald, B. Schlesinger, and C. Lambert. 1982. The Adolescent Mother and Her Child: A Research Study. Toronto: University of Toronto. 73. Scambler N . R. 1980. An Overview of Teenage Pregnancy and parenthood in British Columbia. Vancouver: Social Planning and Review Council of British Columbia. 74. Sells C. W., and R. W. Blum. 1996. Morbidity and mortality among U.S. adolescents: an overview of data and trends. American Journal of Public Health, 86(4):513-519. 75. Spingarn R. W., and R. H. DuRant. 1996. Male adolescents involved in pregnancy: Associated health risk and problem behaviors. Pediatrics, 98(2):262-268. 76. Timewell B. A. , and K. Cassels. 1989. B.C. Task Force on Teenage Pregnancy and Parenthood: Developing a Support Group for Young Parents. Vancouver, BC: Vancouver Y W C A . 77. Toledo-Dreves V. , L. S. Zabin, and M . R. Emerson. 1995. Durations of adolescent sexual relationships before and after conception. Journal of Adolescent Health, 17(3): 163-172. 78. Vaz R., P. Smolen, and C. Miller. 1983. Adolescent pregnancy: involvement of the male partner. Journal of Adolescent Health Care, 4(4):246-250. 79. Westney O. E., O. J. Cole, and T. L . Munford. 1988. The effects of prenatal education intervention on unwed prospective adolescent fathers. Journal of Adolescent Health Care, 9(3):214-218. 80. Williams-McCoy J. E., and F. B. Tyler. 1985. Selected psychosocial characteristics of black unwed adolescent fathers. Journal of Adolescent Health Care, 6(1): 12-16. 62 81. Wilson W . J . 1987. The Truly Disadvantaged. Chicago: The University of Chicago Press. 82. World Health Organization, United Nations Fund for Population Activities, and UNICEF. 1989. The Reproductive Health of Adolescents: A Strategy for Action. Geneva: World Health Organization. 83. Zabin L. S., and V . Sedivy. 1992. Abortion among adolescents: research findings and the current debate. Journal of School Health, 62(7):319-324. 63 "@en . "Thesis/Dissertation"@en . "1999-05"@en . "10.14288/1.0088912"@en . "eng"@en . "Health Care and Epidemiology"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Characteristics of the adolescent father"@en . "Text"@en . "http://hdl.handle.net/2429/8974"@en .