Open Collections

UBC Faculty Research and Publications

Cancer incidence and adverse pregnancy outcome in registered nurses potentially exposed to antineoplastic… Ratner, Pamela A; Spinelli, John J; Beking, Kris; Lorenzi, Maria; Chow, Yat; Teschke, Kay; Le, Nhu D; Gallagher, Richard P; Dimich-Ward, Helen Sep 16, 2010

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

Item Metadata


52383-12912_2009_Article_76.pdf [ 475.99kB ]
JSON: 52383-1.0221293.json
JSON-LD: 52383-1.0221293-ld.json
RDF/XML (Pretty): 52383-1.0221293-rdf.xml
RDF/JSON: 52383-1.0221293-rdf.json
Turtle: 52383-1.0221293-turtle.txt
N-Triples: 52383-1.0221293-rdf-ntriples.txt
Original Record: 52383-1.0221293-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessCancer incidence and adverse pregnancyoutcome in registered nurses potentially exposedto antineoplastic drugsPamela A Ratner1*, John J Spinelli2, Kris Beking3, Maria Lorenzi2, Yat Chow4, Kay Teschke5, Nhu D Le2,Richard P Gallagher2, Helen Dimich-Ward3AbstractBackground: To determine the relationships of potential occupational exposure to antineoplastic drugs withcancer incidence and adverse pregnancy outcomes in a historical prospective cohort study of female registerednurses (RNs) from British Columbia, Canada (BC).Methods: Female RNs registered with a professional regulatory body for at least one year between 1974 and 2000formed the cohort (n = 56,213). The identifier file was linked to Canadian cancer registries. An RN offspring cohortfrom 1986 was created by linkages with the BC Birth and Health Status Registries. Exposure was assessed by workhistory in oncology or cancer agencies (method 1) and by estimating weighted duration of exposure developedfrom a survey of pharmacists and nursing unit administrators of all provincial hospitals and treatment centers andthe work history of the nurses (method 2). Relative risks (RR) were calculated using Poisson regression for cancerincidence and odds ratios (OR) were calculated for congenital anomaly, stillbirth, low birth weight, and prematurityincidence, with 95% confidence intervals.Results: In comparison with other female RNs, method 1 revealed that RNs who ever worked in a cancer center orin an oncology nursing unit had an increased risk of breast cancer (RR = 1.83; 95% CI = 1.03 - 3.23, 12 cases) andtheir offspring were at risk for congenital anomalies of the eye (OR = 3.46, 95% CI = 1.08 - 11.14, 3 cases). Method2 revealed that RNs classified as having the highest weighted durations of exposure to antineoplastic drugs had anexcess risk of cancer of the rectum (RR = 1.87, 95% CI = 1.07 - 3.29, 14 cases). No statistically significant increasedrisks of leukemia, other cancers, stillbirth, low birth weight, prematurity, or other congenital anomalies in the RNs’offspring were noted.Conclusions: Female RNs having had potential exposure to antineoplastic drugs were not found to have anexcess risk of leukemia, stillbirth, or congenital anomalies in their offspring, with the exception of congenitalanomalies of the eye, based on only three cases; however, elevated risks of breast and rectal cancer wereobserved.BackgroundThe nursing profession is known to involve occupationalexposures that may have adverse health effects. Lie andKjæheim’s [1] review of 19 published epidemiologicalstudies of nurses, conducted between 1983 and 2001,concluded that RNs may be at increased risk for breastcancer and leukemia related to their work. Some studieshave shown that the offspring of nurses who continueto work during pregnancy are at increased risk of prena-tal development of congenital anomalies [2,3].More specifically, there has been concern regardingthe potential for carcinogenic and teratogenic effectsrelated to antineoplastic drugs exposure. Antineoplasticdrugs have been used in the treatment of malignant dis-eases for more than 50 years. There are almost 100 anti-neoplastic drugs currently in use, many of which are* Correspondence: pam.ratner@ubc.ca1School of Nursing, University of British Columbia, 302-6190 Agronomy Road,Vancouver, V6T 1Z3, CanadaFull list of author information is available at the end of the articleRatner et al. BMC Nursing 2010, 9:15© 2010 Ratner et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (, which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.mutagenic and either known or probable human carci-nogens [4]. Second malignancies, including leukemiaand bladder cancer, have been reported in patients whohave previously received antineoplastic drugs [4]. Nursesemployed in oncology treatment programs may be occu-pationally exposed to antineoplastic drugs through inha-lation of aerosolized drug products and by direct skincontact of residues on drug vials, contaminated intrave-nous tubing, drug spillage, and patients’ excreta (urine,feces, blood, or vomitus), or orally through hand-to-mouth contact [5]. Although the intensities of exposureare much less than those of patients receiving treatment,occupational exposures typically are of longer cumula-tive duration. Despite the introduction of safety guide-lines and protective measures, health-care workers canstill be exposed to these toxic drugs, as demonstrated bydetectable levels of biomarkers found in the urine ofnurses and other health professionals, and DNA damageor chromosomal aberrations observed in their lympho-cytes and exfoliated buccal cells [6-9]. Mutagenic effectsof antineoplastic drugs provide an explanatory mechan-ism for elevated risks of cancer and adverse reproductiveoutcomes. Antimetabolites, a commonly used class ofantineoplastic drugs, may also affect reproductive out-comes through folate antagonist activity [10,11].Several epidemiological studies have been conductedof the health risks of nurses, in general, as a result ofantineoplastic drug exposure. Increased risks of leuke-mia and breast cancer were suggested in two studies ofnurses who handled antineoplastic drugs [12,13].A 1985 study showed an increased risk of congenitalanomalies in the offspring of RNs exposed to antineo-plastic drugs [14]. Fransman et al. [15], however,reported that RNs exposed to antineoplastic drugs, rela-tive to other nurses, had difficulty conceiving, and wereat greater risk of premature delivery and low birthweight infants, although not at greater risk of sponta-neous abortion, stillbirth, or congenital anomalies.We had the opportunity to examine a very largecohort of registered nurses (RNs) and their offspring inrelation to this occupational exposure. The objectives ofour study were to determine whether a cohort of femaleRNs employed since 1974 by oncology/cancer agenciesor by any nursing departments determined to have pos-sible or probable exposure to antineoplastic drugs, inthe province of BC, had increased risks of cancer (breastcancer or leukemia, in particular) or adverse pregnancyoutcomes, including stillbirth, congenital anomalies, lowbirth weight, and prematurity.MethodsStudy PopulationThe cohort, described previously [16], consisted of56,213 females registered with their professionalregulatory body for at least one year between 1974 and2000 in the province of BC. The cohort file was firstlinked with the Canadian Cancer Registry. Observationswere censored for RNs lost to follow-up (i.e., no longerregistered and known to have left Canada before theend of the study period). Otherwise, they were assumedto be cancer-free at the end of 1999, unless there was acancer diagnosis reported in the cancer registry. Theexposure period was lagged 10 years prior to the inci-dent cancer cases to allow for a minimum latency per-iod of 10 years. The cancer cases were categorized bysite according to the World Health Organization’s 9thedition of the International Classification of Disease(ICD-9).The cohort file was linked to the live and stillbirthrecords of the BC Vital Statistics Agency to create anRN offspring cohort of 22,491 live and 115 stillbirths.Records of birth weight and gestational age at birthenabled the identification of offspring with low birthweight and those born prematurely. Stillbirth wasdefined as the delivery of a dead fetus following 20 ormore weeks of gestation. Low birth weight included off-spring weighing less than 2500 grams and prematurebirth included offspring born at a gestational age of lessthan 37 weeks. The offspring cohort was linked to theBC Health Status Registry (HSR) to obtain informationabout the presence of congenital anomalies diagnosedany year (up to 20 years) after birth and, if present, theirtype, as categorized according to the ICD-9 classificationsystem. The HSR was established in 1952 with a man-date to ascertain, record, and classify “handicapping”conditions, congenital anomalies, and genetic defects inthe population. Because recording practices were lessconsistent in the earlier periods of surveillance, BC VitalStatistics limited access to HSR data to the interval,1986 to 2000, to improve upon the consistency in dataquality. Cases were limited to live, singleton offspring toprevent potential confounding from the adverse effectsassociated with stillbirth and multiple-birth pregnancies.The study was approved by the University of BritishColumbia’s Clinical Research Ethics Board.Exposure AssessmentFor all cohort members, the work history file derivedfrom the annual registration renewal records containedthe following information for each year the RNs wereregistered to practice in the province: name of employer,type of employer, employment position, level of nursingeducation attained, primary area of responsibility, full orpart-time status, and number of hours worked.Exposure to antineoplastic drugs was assessed by twomethods. The first method (method 1) calculated thenumber of years the RN had been employed in the fieldof oncology or by the provincial cancer agency, asRatner et al. BMC Nursing 2010, 9:15 2 of 11reported in the annual registration renewal records. Forthese RNs, the BC Cancer Agency (the organization thatcoordinates all provincial cancer centers) was recordedas the place of employment for the entire study period,or the nurses reported that they had been employed inoncology by another hospital. Oncology was added tothe registry as a field of employment in 1996. The BCCancer Agency was listed as a place of employment forthe full study period. The majority of nurses in oncologywere employed outside of the BC Cancer Agency (71.0%for the years 1996-2000), typically in a general hospitalsetting. Before 1996, nurses employed in oncology out-side of the BC Cancer Agency could not be identified.For the calculation of total years exposed, full-timeemployment was assigned a weight of 1.0 and part-timeemployment was assigned a weight of 0.5.A second method of ascertaining exposure to antineo-plastic drugs (method 2) was based on telephone inter-views with pharmacists and with senior nurses innursing departments from all 94 general hospitals (357departments) and 19 diagnostic and treatment centersin BC. Pharmacists were asked about the periods oftime antineoplastic drugs were used at the facility andabout the frequency of use of up to 74 individual anti-neoplastic drugs. Senior nurses were asked whether anynurses in their department had administered or mixedantineoplastic drugs or cared for patients who receivedthe drugs. Positive responses were followed by furtherquestions about the probability of exposure (i.e., noexposure, possible exposure, or probable exposure) bynursing position (i.e., supervisor/coordinator, clinicalnurse specialist, head nurse/unit manager, charge nurse,staff nurse) and year. For all positions and years withpossible and probable exposures, these senior nurseswere asked whether specific personal protective equip-ment was used, and whether waste disposal, spill, andpatient care guidelines were in place and followed.Finally, if nurses in the department were involved in themixing of antineoplastic drugs, the location of the pro-cedure was queried (i.e., bio-safety cabinet, laminar flowhood, desk/nursing station, or medication room). Expo-sure to antineoplastic drugs was then classified as no,unlikely, possible, or probable using an algorithm cre-ated by the study hygiene team and completed by theoccupational hygienist (see Figure 1). The algorithmused the available published evidence about factorsinfluencing exposures, and was applied to the followingdepartmental survey data elements for each nursingposition and year: probability that the job involved anti-neoplastic drug administration; the number of times perweek the drugs were administered; whether the mixingof antineoplastic drugs was performed; whether glovesand long-sleeved gowns were used; and whether specialhandling procedures were followed for waste disposal,patient handling, and spill clean-up.A weighted exposure duration was calculated for eachnurse’s working history during the cohort period. Toacknowledge the skewed distributions of exposures ofnurses to antineoplastic agents [17], and to ensure thatonly those with reasonable possibilities of exposure wereattributed exposure-time, a multiplicative weighting(0.00, 0.0625, 0.25, and 1.00) was assigned to the expo-sure probabilities from the algorithm (none, unlikely,possible, and probable exposure, respectively). In addi-tion, part-time employment was assigned a weight of 0.5and full-time employment a weight of 1.0. Weightedexposure durations were cumulated over the person’sfull job history, as follows:P Hjjnj=∑ ×1where j is the specific job in a specific department fora specific employer for a specific year, Pj is the weightfor the assigned probability of exposure for that job/department/employer/year and Hj is the weight for full-time and part-time employment. A weighted duration ofexposure of 15 days is equivalent to full-time work forone year (240 days) at a job considered unlikely to beexposed; a weighted duration of exposure of 60 days isequivalent to full-time work for one year at a job con-sidered possibly exposed.Exposure assessment for the birth outcomes analysiswas determined by the above format for two measures:(a) estimated exposure during the first trimester of preg-nancy and (b) cumulative exposure over the ten yearspreceding the date of birth.Statistical AnalysisPoisson regression analysis was performed on the cancerincidence of the RN cohort, using the R statistical soft-ware package [18]. Adjustment was made for calendaryear and the age of the RN, and a lag of 10 years wasapplied. Based on the distribution of cumulative workyears, the years of work in oncology for method 1 werecategorized into 3 groups: (a) none, (b) < 5 years, and(c) ≥ 5 years. Weighted duration of exposure to antineo-plastic drugs for method 2 was also categorized into 3groups: (a) < 15 days, (b) ≥ 15 to 60 days, and (c) > 60+days.Logistic regression analysis was performed on thereproductive outcomes of the offspring cohort, usingSPSS 16 (SPSS Inc., Chicago, IL). Adjustments weremade for the age of the mother at birth, year of birth,and the sex of the child. Because of the relatively smallRatner et al. BMC Nursing 2010, 9:15 3 of 11number of births with adverse reproductive outcomes,the years of work in oncology or a cancer center(method 1) were categorized into 2 groups ("none” or1+ years ("ever”)), and exposure to antineoplastic drugs(method 2) was collapsed into two groups (< 15 daysand ≥ 15 days).For all analyses, two-sided testing with a 5% signifi-cance level was used, corresponding to a 95% confidenceinterval.ResultsOf the 56,213 female RNs who had a work historybetween 1974 and 2000, 905 (1.6%) RNs were identifiedas ever having been exposed to antineoplastic drugsaccording to method 1, based on their employment inoncology nursing units or at a cancer center, and 7,635(13.6%) of the 56,213 RNs were assigned at least 15 daysweighted duration of exposure to antineoplastic drugsaccording to method 2, based on information acquiredthrough the agency-based survey.During 1986-2000, 12,741 RNs gave birth to 22,491live and singleton offspring, which defined the offspringcohort (see Table 1). Sixty nine RNs were identified ashaving worked in oncology for the time it was availableas a field code (1996-2000), of which 20 (29.0%) wereFigure 1 Algorithm of probability of antineoplastic drug exposure. Algorithm used for individual assessment of probability of exposure,based on survey questions administered to head nurses of each department included in the study. Based on responses displayed in the boxesto each of the listed, ordered questions, nurses employed by a given department were allocated in a deterministic manner to either havingunlikely, possible, or probable exposure to antineoplastic drugs.Table 1 Descriptive characteristics of the RN offspringcohort, 1986-2000N %Total births 23,222Singletons 22,613 97.4%Stillbirths1 147 0.6%Twins 581 2.5%Triplets 28 0.1%Live singletons 22,491Only child 9,163 40.7%2 children 8,454 37.6%≥ 3 children 4,874 21.7%Males 10,961 48.7%Females 11,530 51.3%RN mothers 12,741Mother’s age at birth< 30 years 7,943 35.3%30 - 34 years 9,371 41.7%≥ 35 years 5,177 23.0%1Value includes twins and triplets, and cases with missing or incompletecongenital anomaly diagnosis data.Ratner et al. BMC Nursing 2010, 9:15 4 of 11also identified as having worked in a cancer center dur-ing pregnancy. A total of 141 were identified as havingworked in a cancer center during pregnancy.As shown in Table 2, having ever worked in oncologyor a cancer center (method 1) was associated with a sig-nificantly elevated relative risk of breast cancer (RR =1.83, 95% CI = 1.03 - 3.23, 12 cases). The risks for allcancers (RR = 1.28, 95% CI = 0.83 - 1.79, 21 cases), orfor cancers specifically of the uterus (RR = 2.58, 95%CI = 0.96 - 6.94, 4 cases), were elevated, but were notstatistically significant. Cases of other cancers had fre-quencies below 3 and no cases of leukemia wereobserved among those having worked in oncology or acancer center.Table 3 shows the relative risk of cancer by probabilityof exposure to antineoplastic drugs (method 2). Forthose in the top two exposure categories, the risk of rec-tal cancer was significantly elevated, based on 14 cases(RR = 1.87, 95% CI = 1.07 - 3.29), and the risk of breastcancer was elevated, but was not statistically significant(RR = 1.12, 95% CI = 0.89 - 1.39, 87 cases). The risksfor all cancers (RR = 1.08, 95% CI = 0.93 - 1.25, 194cases), and for each other type of cancer, were not sta-tistically significantly elevated. Three cases of leukemiawere observed among those in the top exposurecategories.Of 22,491 live births of the offspring cohort, 1,391(6.2%) were diagnosed with congenital anomalies. Onehundred and seventy (0.8%) mothers (n = 12,741) wereidentified as having worked in oncology nursing units orin a cancer center between 1986 and 2000 duringpregnancy. A total of 2,650 (11.8%) of the mothers inthe cohort were assigned as having had at least 15 daysweighted duration of exposure to antineoplastic drugs inthe 10 years preceding the pregnancy.Among the offspring of mothers who had ever workedin oncology nursing units or a cancer center duringtheir pregnancy (method 1), the risk of all congenitalanomalies or congenital anomalies of the circulatory ormusculoskeletal systems were elevated, but not statisti-cally significantly; the risk of anomalies of the eye wasstatistically significantly elevated at 3.46 (95% CI = 1.08- 11.14), but based on only 3 cases (see Table 4). Asshown in Table 5, there were no statistically significantelevated risks of congenital anomalies among the off-spring of mothers with exposure to antineoplastic drugsduring the first trimester of pregnancy or over the 10years preceding pregnancy (method 2). However, therisk of cleft palate or lip, for the 10-year exposure per-iod, was notably high, at 1.84 (95% CI = 0.75 - 4.49, 6cases). Stillbirths were an infrequent outcome for theoffspring cohort, with only 115 cases. There was noincreased risk of stillbirth related to antineoplastic drugexposure during the first trimester of pregnancy (OR =0.67, 95% CI = 0.21 - 2.13, 3 cases) or during the 10years preceding pregnancy (OR = 0.63, 95% CI = 0.31 -1.30, 8 cases). There were no cases among those whohad worked in oncology nursing units or for a cancercenter during pregnancy.There were 746 cases of low birth weight and 1,133cases of prematurity among the offspring cohort.Among those employed in oncology or a cancer centerTable 2 Relative risk of selected cancer incidence of female registered nurses according to cumulative years worked inoncology or a cancer center (method 1), 1996-2000 (N = 56,213)1,2,3,4Site (ICD-9 code) Employment Duration Observed RR CI (95%) p-value5All cancers except non-melanoma skin never 3078 1.00 0.135< 5 years 14 1.10 0.65-1.87> 5 years 7 1.89 0.90-3.97ever 21 1.28 0.83-1.79Breast (174) never 1274 1.00 0.379< 5 years 9 1.75 0.91-3.38> 5 years 3 2.10 0.67-6.53ever 12 1.83 1.03-3.23Uterus (179-182) never 302 1.00 0.068< 5 years 3 2.50 0.80-7.82> 5 years 1 2.85 0.40-20.38ever 4 2.58 0.96-6.941Based on Poisson regression using a 10-year lag.2Cancers with an incidence of less than 3 cases were excluded.3N for ever worked in oncology or cancer center = 905.4Adjusted for age and calendar year.5Significance of the trend, derived from chi-square analysis.Ratner et al. BMC Nursing 2010, 9:15 5 of 11Table 3 Relative risk of selected cancer incidence for female registered nurses according to weighted duration ofexposure to antineoplastic agents (method 2), 1974-2000 (N = 56,213)1,2,3,4Site (ICD-9 code) Weighted duration of exposure5 Observed RR CI (95%) p-value6All cancers except < 15 days 2905 1.00 0.452non-melanoma skin 15 to < 60 days 150 1.11 0.94-1.30≥ 60 days 44 1.00 0.74-1.35top 2 categories combined 194 1.08 0.93-1.25Colon (153,159) < 15 days 225 1.00 0.71015 to < 60 days 11 1.16 0.63-2.13≥ 60 days < 3top 2 categories combined 13 0.99 0.56-1.74Rectum (154) < 15 days 116 1.00 0.03715 to < 60 days 10 1.85 0.96-3.56≥ 60 days 4 1.93 0.71-5.25top 2 categories combined 14 1.87 1.07-3.29Lung (162) < 15 days 202 1.00 0.86715 to < 60 days 11 1.24 0.67-2.28≥ 60 days 3 0.83 0.27-2.61top 2 categories combined 14 1.22 0.65-1.94Melanoma (172) < 15 days 191 1.00 0.40015 to < 60 days 12 1.28 0.71-2.30≥ 60 days 3 1.28 0.41-4.02top 2 categories combined 15 1.28 0.75-2.18Breast (174) < 15 days 1199 1.00 0.26715 to < 60 days 64 1.08 0.84-1.39≥ 60 days 23 1.23 0.82-1.87top 2 categories combined 87 1.12 0.89-1.39Uterus (179-182) < 15 days 290 1.00 0.69615 to < 60 days 9 0.72 0.37-1.40≥ 60 days 7 1.66 0.78-3.53top 2 categories combined 16 0.95 0.57-1.59Ovary (183) < 15 days 144 1.00 0.20915 to < 60 days 6 0.90 0.40-2.06≥ 60 days < 3top 2 categories combined 6 0.68 0.30-1.55Bladder - including in-situ (188) < 15 days 78 1.00 0.39415 to < 60 days 6 2.00 0.86-4.65≥ 60 days < 3top 2 categories combined 7 1.67 0.76-3.67Brain (191-192) < 15 days 43 1.00 0.29215 to < 60 days 3 1.76 0.54-5.80≥ 60 days < 3 1.88 0.26-13.87top 2 categories combined 4 1.79 0.63-5.10Ratner et al. BMC Nursing 2010, 9:15 6 of 11during pregnancy, the risks of low birth weight (OR =1.41, 95% CI = 0.44 - 4.54, 3 cases) and prematurity(OR = 1.88, 95% CI = 0.80 - 4.41, 6 cases) were bothelevated, but were not statistically significant. Therewere no increased risks of low birth weight or prematur-ity, respectively, related to antineoplastic drugs exposureduring the first trimester of pregnancy (OR = 0.97, 95%CI = 0.43 - 2.20, 6 cases; OR = 1.05, 95% CI = 0.56 -2.00, 10 cases) or during the 10 years preceding preg-nancy (OR = 0.67, 95% CI = 0.32 - 1.42, 7 cases;OR = 0.62, 95% CI = 0.33 - 1.17, 10 cases).DiscussionThis cohort of RNs who worked in BC between 1974and 2000 showed a slightly increased but non-significantrisk of all cancers if they had potential exposure to anti-neoplastic drugs. The risk of breast cancer was signifi-cantly elevated among those who had ever worked inoncology nursing units or for a cancer center, whichshowed a possible trend of increasing incidence withincreasing years of work. However, this increase was ele-vated but not statistically significant in RNs classified asTable 3 Relative risk of selected cancer incidence for female registered nurses according to weighted duration ofexposure to antineoplastic agents (method 2), 1974-2000 (N = 56,213)1,2,3,4 (Continued)Thyroid (193) < 15 days 82 1.00 0.40815 to < 60 days 4 0.93 0.34-2.56≥ 60 days< 3top 2 categories combined 4 0.74 0.27-2.05Ill-defined (195) < 15 days 81 1.00 0.18115 to < 60 days 5 1.33 0.54-3.31≥ 60 days 3 2.09 0.66-6.65top 2 categories combined 8 1.54 0.74-3.21Lymphatic & Hematopoietic < 15 days 219 1.00 0.482(200-208) 15 to < 60 days 9 0.93 0.47-1.82≥ 60 days < 3top 2 categories combined 11 0.84 0.46-1.55Non-Hodgkin Lymphoma < 15 days 121 1.00 0.643(200, 202) 15 to < 60 days 6 1.06 0.47-2.43≥ 60 days < 3 0.49 0.07-3.52top 2 categories combined 7 0.91 0.42-1.96Leukemia (204-208) < 15 days 59 1.00 0.93015 to < 60 days < 3≥ 60 days < 3top 2 categories combined 3 0.89 0.27 - 2.881 Based on Poisson regression using a 10-year lag.2 Cancer with an incidence of less than 3 were excluded.3 N for top 2 categories of antineoplastic drug exposure = 7,635.4 Adjusted for age and calendar year.5 A weighted duration of exposure of 15 days is equivalent to full-time work for one year (240 days) at a job considered unlikely to be exposed; a weightedduration of exposure of 60 days is equivalent to full-time work for one year at a job considered possibly exposed.6 Significance of the trend, derived from chi-square analysis.Table 4 Risk of congenital anomalies among theoffspring of RNs employed in oncology nursing units ora cancer center during pregnancy (method 1), 1996-2000(N = 22,491)1,2Congenital anomaly category(ICD9 code)Employed Cases OR3 CI (95%)All congenital anomalies (740-759)never 1024 1.00ever 17 1.42 0.86-2.36Eye (743) never 69 1.00ever 3 3.46 1.08-11.14Circulatory system (747) none 94 1.00ever 3 2.68 0.84-8.59Musculoskeletal system never 335 1.00(754-756) ever 5 1.210.49-2.961Congenital anomaly types with an incidence of less than 3 cases wereexcluded.2N for ever employed in oncology or cancer center = 190.3Adjusted for sex, year of birth, and age of mother.Ratner et al. BMC Nursing 2010, 9:15 7 of 11Table 5 Risk of congenital anomalies among the offspring of RNs according to weighted duration of exposure toantineoplastic drugs (method 2), 1986-2000 (N = 22,491)1,2Congenital anomaly category(ICD9 code)Weighted duration ofexposure3Exposure during first trimester ofpregnancyExposure during 10 years precedingpregnancycases OR4 CI (95%) cases OR4 CI (95%)All congenital anomalies < 15 days 1,328 1.00 1229 1.00(740-759) ≥ 15 days 63 0.93 0.72-1.21 162 0.98 0.82-1.16Nervous system < 15 days 51 1.00 47 1.00(740-742) ≥ 15 days < 3 6 0.96 0.41-2.24Eye < 15 days 89 1.00 81 1.00(743) ≥ 15 days 6 1.31 0.57-3.00 14 1.26 0.71-2.22Ear, face, neck < 15 days 85 1.00 82 1.00(744) ≥ 15 days 4 0.96 0.35-2.64 7 0.64 0.30-1.39Heart < 15 days 137 1.00 127 1.00(745-746) ≥ 15 days 10 1.44 0.75-2.74 20 1.15 0.72-1.84Other circulatory system < 15 days 127 1.00 115 1.00(747) ≥ 15 days 8 1.26 0.61-2.58 20 1.28 0.79-2.06Cleft palate/lip < 15 days 29 1.00 25 1.00(749) ≥ 15 days < 3 6 1.84 0.75-4.49Upper alimentary tract < 15 days 83 1.00 81 1.00(750) ≥ 15 days 4 0.93 0.34-2.55 6 0.55 0.24-1.25Other digestive system < 15 days 35 1.00 32(751) ≥ 15 days < 3 5 1.17 0.46-3.01Genital organs < 15 days 213 1.00 192 1.00(752) ≥ 15 days 6 0.55 0.24-1.25 27 1.06 0.70-1.59Urinary system < 15 days 101 1.00 93 1.00(753) ≥ 15 days 5 0.97 0.39-2.38 13 1.05 0.59-1.88Musculoskeletal system < 15 days 443 1.00 419 1.00(754-756) ≥ 15 days 18 0.80 0.49-1.28 42 0.74 0.54-1.02Integument < 15 days 96 1.00 88 1.00(228, 757) ≥ 15 days 5 1.01 0.41-2.49 13 1.08 0.60-1.94Chromosomal anomalies < 15 days 60 1.00 56 1.00(758) ≥ 15 days < 3 4 0.54 0.20-1.49Multiple anomalies < 15 days 15 1.00 13 1.00(759.7-759.8) ≥ 15 days < 3 3 1.76 0.50-6.191 Congenital anomalies with an incidence of less than 3 cases were excluded.2 N for ≥ 15 days of antineoplastic drug exposure during first trimester of pregnancy = 1,062; during 10 years preceding pregnancy = 2,650.3 A weighted duration of exposure of 15 days is equivalent to full-time work for one year (240 days) at a job considered unlikely to be exposed; a weightedduration of exposure of 60 days is equivalent to full-time work for one year at a job considered possibly exposed.4 Adjusted for sex, year of birth, and age of mother.Ratner et al. BMC Nursing 2010, 9:15 8 of 11potentially exposed to antineoplastic drugs, based on asurvey-based exposure assessment protocol. Anincreased risk of cancer of the rectum was observed inRNs determined to have had a probability of exposureto antineoplastic drugs. No increased risk of any othercancer type, including leukemia, was observed usingeither exposure assessment method.It has been well established that nurses in oncologywho handle antineoplastic drugs have increased frequen-cies of biomarkers of exposure, including chromosomalaberrations and sister chromatid exchanges [19-22].However, to date, very few relevant epidemiological stu-dies have estimated RNs’ cancer risk related to exposureto antineoplastic drugs. A linkage study of Danishnurses who worked in oncology departments, preparingand administering antineoplastic drugs, reported a sta-tistically significantly increased relative risk (RR = 10.65)for one site, leukemia, based on 2 cases [12]. A nestedcase-control study of 59 breast cancer cases from anIcelandic cohort study of female nurses found that therisk estimates were highest (although not statisticallysignificant and based on 7 cases) among those who hadever handled cytotoxic drugs (OR = 1.65, 95% CI = 0.53- 5.17), after adjustment for year of birth, breast cancerin a first-degree relative, marital status, and nulliparity[13]. Our finding of a significantly increased risk ofbreast cancer among RNs who had ever worked inoncology or a cancer center provides some additionalevidence of this association and supports the relevanceof further investigating occupational risk factors fornurses. In future research about nursing and cancerincidence, consideration should also be given to the pos-sible influence of mammography screening rates. It ispossible that oncology nurses, who directly care forwomen with breast cancer, are predisposed to mammo-graphy screening participation and thus are subject toan over-detection bias. It is recognized that some breastcancers, if undetected and thus not treated, would notprogress and that some women are “over-diagnosed”[23,24].The mutagenic effects of antineoplastic drugs mayincrease the risk of congenital anomalies and stillbirthamong exposed nurses. Of three case-control studies,one showed an odds ratio of 4.7 (95% CI = 1.2 - 18.1)for congenital anomalies among those exposed to anti-neoplastic drugs at least once a week [14], one showed arelative risk of 1.7 (95% CI = 1.0 - 2.8) for spontaneousabortions among occupationally exposed nurses [25],and a third showed an odds ratio of 2.3 (95% CI = 1.2 -4.4) for fetal loss among nurses exposed during theirfirst trimester of pregnancy [26]. A cross-sectional studyobserved 8 congenital anomalies among offspring of 152physicians and nurses who administered antineoplasticdrugs during pregnancy (4.05 expected; p = 0.05) [27].More recent studies have shown few reproductive risksfrom antineoplastic drug exposure. A questionnaire-based study found log-linear odds ratios of 1.20 (95% CI= 0.98 - 1.47) for stillbirth and 0.97 (95% CI = 0.86 -1.09) for congenital anomalies among the offspring ofnurses exposed to antineoplastic drugs [12]. Two case-control studies found odds ratios of 1.02 (95% CI = 0.47- 2.06) and 2.2 (95% CI = 0.7 - 7.2) for congenitalanomalies among the offspring of nurses who handledantineoplastic drugs, compared with nurses withoutexposure [12,28]. A recent meta-analysis of four studiesof nurses’ exposure to antineoplastic drugs estimated anodds ratio for congenital anomaly incidence of 1.64(95% CI = 0.91 - 2.94) [7].In our study, the risk of congenital anomalies of theeye was significantly increased in the offspring of RNswho had ever worked in oncology nursing units or forcancer centers during pregnancy (method 1), althoughthis estimate was based on only 3 cases. The risks forother congenital anomalies among this employmentgroup were also notably elevated, but not statisticallysignificantly and were limited by the low frequency ofcases. The comparatively higher risk of all congenitalanomalies in the employment group may suggest anoncology/cancer agency-specific effect that is perhapsdiluted in the broader department-based exposureassessments. The risks of stillbirth, low birth weight,and prematurity were not significantly increased amongRNs potentially exposed to antineoplastic drugs,assessed through either their employment in oncologyor estimated weighted duration of exposure.According to the BC Cancer Agency Benefit DrugList, there was an increase in the number of differentantineoplastic drugs used, from 66 in 1986 to 88 in2000. Standardized safety practices for handling antineo-plastic drugs and other hazardous materials were estab-lished in 1985. However, a recent review of methodsused to monitor occupational exposure to antineoplasticdrugs concluded that, despite the introduction of safetyguidelines and protective measures, health-care workerscan still be exposed [8]. Nonetheless, the adoption ofthese practices may partly account for the predomi-nantly null or non-significant risks observed in this andother recent studies [15,12,28].Whereas many previous studies were questionnaire-based and were, therefore, susceptible to exposure recalland selection biases, our record linkage used compre-hensive recruitment of an employment group based onregistry-derived data and allowed for complete ascer-tainment of outcomes according to standardized medi-cal reporting systems. Furthermore, the historicalprospective design of the study meant that exposurevariables were assessed independently from healthoutcomes.Ratner et al. BMC Nursing 2010, 9:15 9 of 11Despite our large sample size, limitations of our studyarise from limited statistical power associated with thesmall numbers of cases and the low prevalence of expo-sure, possibly leading to missed associations, as well asthe testing of multiple hypotheses, such that a statisti-cally significant association may have occurred bychance alone.As is common in historical cohort studies, we had noinformation about potential confounding factors relatedto lifestyle; however, all comparisons were within thenursing profession, a narrow, well educated, socioeco-nomic stratum. Nurses have been documented to havehealthier habits than the population as a whole, butthere have been temporal patterns, for example strongdeclines in smoking rates among US nurses in the last30 years (33% smokers in 1976 vs. 8.4% in 2002/2003[29]). There also were temporal patterns in antineoplas-tic drug use in the British Columbia healthcare agenciesduring the study period (e.g., hand mixing of the drugswas completed in 45% of the facilities in the 1970 s and1980 s, but only 8% in the 1990 s and later). If smokingand antineoplastic drug exposures were correlated andboth related to the outcomes of interest, there could beuncontrolled confounding in the study results. In arecent study of 1,147 live births among nurses, control-ling for parity, smoking, alcohol, coffee, multivitamin,and folic acid intake did not materially change the effectestimates for congenital anomalies [15].We were not able to adjust for other chemical expo-sures in the RNs’ workplaces or other environments.Unfortunately, there is little if any data available aboutthe presence and usage of such chemicals. We are leftto assume that the RNs exposed to antineoplastic drugswere similar to the unexposed RNs with respect to theiropportunity to having been exposed to other toxins.The exposure assessment methods used in the studywere crude and may have resulted in misclassification ofexposure. For example, in method 1, most nurses work-ing in oncology departments were outside of the cancercenters and were not identifiable in the registry, prior to1996. This likely led to underestimation of the numberof exposed RNs but would have been highly specific.The survey (method 2) identified relevant departmentsin general hospitals and treatment centers that adminis-tered antineoplastic drugs and ascribed individual nursestherein with a probability of exposure. Within thesedepartments, some of the nurses ascribed a probabilityof exposure may not have been exposed and, withinother departments, some of the nurses may have beenmissed because of poor information recall. This likelyresulted in misclassification, which would have biasedthe effect estimate toward the null. Nonetheless, themethod was novel in its attempt to more preciselycharacterize exposure estimates of a cohort of nursesthrough the use of relevant questions about employmenthistory that were simple enough to assume sufficientrecall for an approximation of exposure; that is, the clas-sification was broadly based.The premise underlying method 1 was that the oncol-ogy departments, as cancer treatment facilities by defini-tion, would have consistently used antineoplastic drugs,whereas other departments (in general hospitals andother agencies) may not have used them, or when theydid, perhaps they did so with less regularity. Thismethod more definitively assigned exposure through theuse of reliable employment records, limited the exposuregroup, and introduced some uncertainty about theunexposed group (it may have contained nurses fromnursing departments with possible exposure).Consequently, these limitations of the assessments ofexposure may have affected their accuracy and couldpossibly have resulted in an underestimation of the risksassociated with exposure to antineoplastic drugs or havedistorted the shape of the dose-response relationship.For instance, the relative risk found for breast cancerusing method 2, as opposed to method 1, was elevated,but not statistically significant.ConclusionsIn summary, RNs potentially exposed to antineoplasticdrugs through their employment had an elevated risk ofbreast and rectal cancer. There was no demonstratedrisk of other types of cancer or of adverse reproductiveoutcomes, apart from an elevated risk for anomalies ofthe eye based on three cases. Further investigations,including more detailed measurements of exposures anddosages received, may be helpful in evaluating themechanisms by which occupational exposures to anti-neoplastic drugs may act as risk factors for cancer andadverse reproductive outcomes.AcknowledgementsWe appreciate the assistance provided by Olha Demchenko, Joanna Chung,and Zenaida Abanto, as well as the linkages facilitated by Statistics Canada,the BC Cancer Registry, and BC Vital Statistics. We are especially grateful tothe College of Registered Nurses of British Columbia for their consistentsupport of the study. The study was funded by the Workers’ CompensationBoard of British Columbia (WorkSafeBC) and the Canadian Institutes ofHealth Research. Dr. Ratner was supported by the Michael Smith Foundationfor Health Research.Author details1School of Nursing, University of British Columbia, 302-6190 Agronomy Road,Vancouver, V6T 1Z3, Canada. 2BC Cancer Agency Research Centre, 675 West10th Avenue, Vancouver, V5Z 1L3, Canada. 3Department of Medicine,University of British Columbia, 890 West 10th Avenue, Vancouver, BC, V5Z1M9, Canada. 4School of Environmental Health, University of BritishColumbia, 2206 East Mall, Vancouver, V6T 1Z3, Canada. 5School ofPopulation and Public Health, University of British Columbia, 5804 FairviewAvenue, Vancouver, V6T 1Z3, Canada.Ratner et al. BMC Nursing 2010, 9:15 10 of 11Authors’ contributionsHD conceived of the study proposal and participated in the draftconstruction; KB performed background literature searches; PR, JS, RG, KB,and NL participated in the draft construction; ML, YC, and KB conducted thedata analyses. All authors read and approved of the final manuscript.Competing interestsThe authors declare that they have no competing interests.Received: 22 July 2009 Accepted: 16 September 2010Published: 16 September 2010References1. Lie JA, Kjaerheim K: Cancer risk among female nurses: a literature review.Eur J Cancer Prev 2003, 12:517-526.2. Bjerkedal T: Occupation and outcome of pregnancy: a population-basedstudy in Norway. Prog Clin Biol Res 1985, 163B:265-268.3. Matte TD, Mulinare J, Erickson JD: Case-control study of congenitaldefects and parental employment in health care. Am J Indust Med 1993,24:11-23.4. Connor TH, McDiarmid MA: Preventing occupational exposures toantineoplastic drugs in health care settings. CA Cancer J Clin 2006,56:354-365.5. Labuhn K, Valanis B, Schoeny R, Loveday K, Vollmer WM: Nurses’ andpharmacists’ exposure to antineoplastic drugs: findings from industrialhygiene scans and urine mutagenicity tests. Cancer Nurs 1998, 21:79-89.6. Cavallo D, Ursini CL, Perniconi B, Francesco AD, Giglio M, Rubino FM,Marinaccio A, Iavicoli S: Evaluation of genotoxic effects induced byexposure to antineoplastic drugs in lymphocytes and exfoliated buccalcells of oncology nurses and pharmacy employees. Mutat Res 2005,587:45-51.7. Dranitsaris G, Johnston M, Poirier S, Schueller T, Milliken D, Green E,Zanke B: Are health care providers who work with cancer drugs at anincreased risk for toxic events? A systematic review and meta-analysis ofthe literature. J Oncol Pharm Pract 2005, 11:69-78.8. Sessink PJ, Bos RP: Drugs hazardous to healthcare workers. Evaluation ofmethods for monitoring occupational exposure to cytostatic drugs. DrugSaf 1999, 20:347-359.9. Ündeğer Ü, Başaran N, Kars A, Güç D: Assessment of DNA damage innurses handling antineoplastic drugs by the alkaline COMET assay.Mutat Res - Genetic Toxicology and Environmental Mutagenesis 1999,439:277-285.10. Kaye SB: New antimetabolites in cancer chemotherapy and their clinicalimpact. Br J Cancer 1998, 78(Suppl 3):1-7.11. Locksmith GJ, Duff P: Preventing neural tube defects: the importance ofpericonceptional folic acid supplements. Obstet Gynecol 1998,91:1027-1034.12. Skov T, Maarup B, Olsen J, Rorth M, Winthereik H, Lynge E: Leukaemia andreproductive outcome among nurses handling antineoplastic drugs. Br JIndust Med 1992, 49:855-861.13. Gunnarsdottir HK, Aspelund T, Karlsson T, Rafnsson V: Occupational riskfactors for breast cancer among nurses. Int J Occup Environ Health 1997,3:254-258.14. Hemminki K, Kyyrönen P, Lindbohm ML: Spontaneous abortions andmalformations in the offspring of nurses exposed to anaesthetic gases,cytostatic drugs, and other potential hazards in hospitals, based onregistered information of outcome. J Epidemiol Community Health 1985,39:141-147.15. Fransman W, Roeleveld N, Peelen S, de Kort W, Kromhout H, Heederik D:Nurses with dermal exposure to antineoplastic drugs: reproductiveoutcomes. Epidemiology 2007, 18:112-119.16. Dimich-Ward H, Lorenzi M, Teschke K, Spinelli J, Ratner PA, Le N, Chow Y,Shu D, Gallagher R: Mortality and cancer incidence in a cohort ofregistered nurses from British Columbia, Canada. Am J Indust Med 2007,50:892-900.17. Fransman W, Peelen S, Hilhorst S, Roeleveld N, Heederik D, Kromhout H: Apooled analysis to study trends in exposure to antineoplastic drugsamong nurses. Ann Occup Hyg 2007, 51:231-239.18. Crawley MJ: The R Book New York: John Wiley 2007.19. Anwar WA, Salama SI, el Serafy MM, Hemida SA, Hafez AS: Chromosomalaberrations and micronucleus frequency in nurses occupationallyexposed to cytotoxic drugs. Mutagenesis 1994, 9:315-317.20. Milkovic-Kraus S, Kraus O, Krsnjavi H, Kubelka D: Chromosomalabnormalities among nurses occupationally exposed to antineoplasticdrugs. Am J Indust Med 1991, 19:771-774.21. Milkovic-Kraus S, Kraus O, Krsnjavi H, Kubelka D: Environmental effects onchromosomes in oncology and radiology department personnel. PrevMed 1992, 21:498-502.22. Norppa H, Sorsa M, Vainio H, Gröhn P, Heinonen E, Holsti L, Nordman E:Increased sister chromatid exchange frequencies in lymphocytes ofnurses handling cytostatic drugs. Scand J Work Environ Health 1980,6:299-301.23. Duffy SW, Lynge E, Jonsson H, Ayyaz S, Olsen AH: Complexities in theestimation of overdiagnosis in breast cancer screening. Br J Cancer 2008,99:1176-1178.24. Waller M, Moss S, Watson J, Møller H: The effect of mammographicscreening and hormone replacement therapy use on breast cancerincidence in England and Wales. Cancer Epidemiol Biomarkers Prev 2007,16:2257-2261.25. Stucker I, Caillard JF, Collin R, Gout M, Poyen D, Hemon D: Risk ofspontaneous abortion among nurses handling antineoplastic drugs.Scand J Work Environ Health 1990, 16:102-107.26. Selevan SG, Lindbohm ML, Hornung RW, Hemminki K: A study ofoccupational exposure to antineoplastic drugs and fetal loss in nurses. NEngl J Med 1985, 313:1173-1178.27. McDonald AD, McDonald JC, Armstrong B, Cherry NM, Côté R, Lavoie J,Nolin AD, Robert D: Congenital defects and work in pregnancy. Br J IndMed 1988, 45:581-588.28. Saurel-Cubizolles MJ, Hays M, Estryn-Behar M: Work in operating roomsand pregnancy outcome among nurses. Int Arch Occup Environ Health1994, 66:235-241.29. Sarna L, Bialous SA, Jun HJ, Wewers ME, Cooley ME, Feskanich D: Smokingtrends in the Nurses’ Health Study (1976-2003). Nurs Res 2008, 57:374-82.Pre-publication historyThe pre-publication history for this paper can be accessed here: this article as: Ratner et al.: Cancer incidence and adversepregnancy outcome in registered nurses potentially exposed toantineoplastic drugs. BMC Nursing 2010 9:15.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at et al. BMC Nursing 2010, 9:15 11 of 11


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items