SEARCH

SEARCH BY CITATION

Keywords:

  • colorectal neoplasms;
  • cohort study;
  • oral contraceptive use;
  • hormone replacement therapy;
  • reproductive factors

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Evidence from epidemiologic studies suggests a possible role of exogenous and endogenous hormones in colorectal carcinogenesis in women. However, with respect to exogenous hormones, in contrast to hormone replacement therapy, few cohort studies have examined oral contraceptive use in relation to colorectal cancer risk. We used data from a large cohort study of Canadian women enrolled in a randomized controlled trial of breast cancer screening to assess the association of oral contraceptive use, hormone replacement therapy and reproductive factors with risk of colorectal cancer, overall and by subsite within the colorectum. Cancer incidence and mortality were ascertained by linkage to national databases. Among 89,835 women aged 40–59 at enrollment and followed for an average of 16.4 years, we identified 1,142 incident colorectal cancer cases. Proportional hazards models were used to estimate the associations between the exposures of interest and risk of colorectal cancer. Ever use of oral contraceptives at baseline was associated with a modest reduction in the risk of colorectal cancer (hazard ratio 0.83, 95% confidence interval 0.73–0.94), with similar effects for different subsites within the colorectum. No trend was seen in the hazard ratios with increasing duration of oral contraceptive use. No associations were seen with use of hormone replacement therapy (ever use or duration of use) or reproductive factors. Our results are suggestive of an inverse association between oral contraceptive use and colorectal carcinogenesis. However, given the lack of a dose–response relationship and the potential for confounding, studies with more complete assessment of exogenous hormone use throughout the life course are needed to clarify this association. © 2007 Wiley-Liss, Inc.

In virtually every region of the world, incidence rates of colorectal cancer are higher in men compared to women for a given age.1 Early etiological studies indicated a correlation between colorectal cancer and breast cancer rates and elevated colon cancer rates in nuns.2 These observations led McMichael and Potter3 to postulate that endogenous and/or exogenous steroid hormone exposure may protect against colon cancer.

Experimental studies indicate that estrogens may play a protective role in colorectal carcinogenesis either indirectly by reducing secondary bile acids and insulin-like growth factor-I or directly by regulating cell growth in the colonic epithelium and inhibiting cell proliferation of colorectal tumors by binding to the estrogen receptor.4, 5 In cultured colon cancer cells, methylation-associated loss of expression of the estrogen receptor gene results in deregulated growth.6, 7 Issa et al.7 reported that in individuals without colonic tumors, methylation of the estrogen receptor increases with age, and that in individuals with colonic tumors, estrogen receptor methylation is almost universally present, suggesting that the estrogen receptor may act as a tumor suppressor. In addition, estradiol has been shown to inhibit chemically-induced murine colon carcinogenesis,8 and in rats treated with dimethylhydrazine, the colon tumor yield is higher in males compared to females.9

A large number of epidemiologic studies have examined use of exogenous hormones and reproductive factors in relation to colorectal cancer risk in women, providing some evidence of a possible reduced risk in users of oral contraceptives (OCs) and hormone replacement therapy (HRT).10, 11, 12 However, relatively few cohort studies have examined OC use,13, 14, 15, 16, 17, 18 and several of those did not have adequate numbers of cases to examine subsites within the colorectum. Furthermore, few studies have examined both HRT and OC use in the same study population. We report here on the association of hormone use and reproductive factors in a large cohort of Canadian women followed for 16 years.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Study population

Our investigation was conducted in the Canadian National Breast Screening Study, a randomized controlled trial of screening for breast cancer, which has been described in detail elsewhere.19, 20 In brief, 89,835 women aged 40–59 were recruited from the general Canadian population between 1980 and 1985. On enrollment into the study, information was obtained from participants on demographic characteristics, smoking habits, reproductive history and exogenous hormone use, using a self-administered lifestyle questionnaire. Regarding hormone use, participants were asked: “Have you ever used OCs either for birth control or menstrual irregularities?” and “Have you ever taken estrogen, with or without progesterone, for menopause or change of life symptoms?” Women who answered yes to either question were further asked (i) the total length of time (in years or months) the preparations were taken and (ii) if a woman had stopped, the year in which she stopped. Starting in 1982, a self-administered food frequency questionnaire (FFQ) was distributed to all new attendees at all screening centers and to women returning to the screening centers for rescreening.21 The FFQ elicited information on usual portion size and consumption of 86 food items and included photographs of portion sizes to assist respondents in quantifying intake. A total of 49,654 completed dietary questionnaires were returned.

Ascertainment of colorectal cancer and deaths

Incident cases of colorectal cancer and deaths from all causes were ascertained, respectively, by means of computerized record linkages to the Canadian Cancer Database and to the National Mortality Database through 1998–2000, depending on the province. During an average of 16.4 years (786,588 person-years) of follow-up of the cohort of 89,835 women, we identified 1,142 incident colorectal cancer cases.

Statistical analysis

Cox proportional hazards models (using age as the time scale) were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for the association of OC use, use of HRT and reproductive factors, with risk of colorectal cancer overall, and with risk of cancers of the colon, proximal colon, distal colon and rectum. We carried out the analysis in 2 stages. First, we used the entire cohort. In these analyses, covariates included age (time to event variable); body mass index (kg/m2) (quintiles); menopausal status (pre-, peri-, postmenopausal), pack-years of smoking (never smoked + 5 levels for smokers); and education (3 levels). Addition of indicator variables for screening center1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 and randomization group in the original clinical trial (intervention/usual care) did not affect the risk estimates, and therefore these variables were omitted. The second stage of the analysis was restricted to the dietary cohort. These analyses included the above covariates as well as dietary intakes of total calories, red meat, fat, fiber, folic acid and calcium (all continuous); alcohol intake (continuous); and physical activity (none or moderate, vigorous, missing). To test for trends in risk with increasing levels of the exposures of interest, we assigned the categorical variables (parity) their ordinal number or the category median (age at first live birth, duration of OC use, duration of HRT) and then fitted the assigned value of each factor as a continuous variable in the risk models. We then evaluated the statistical significance of the corresponding coefficient using the Wald test.22 Analyses were repeated excluding cases of colorectal cancer diagnosed within the first 2 years of follow-up. We further classified women simultaneously by OC use (ever/never) and HRT use (ever/never). All statistical significance tests were 2-sided. All analyses were performed using SAS version 9 (SAS Institute Cary, NC).

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Cases tended to be older than non-cases, were more likely to be postmenopausal, and were less likely to have used OCs (Table I). Cases and non-cases were similar in terms of body mass index, other menstrual and reproductive variables, smoking, HRT use, alcohol consumption, physical activity and dietary variables.

Table I. Baseline Characteristics of the Study Population by Outcome in the Entire Cohort
FactorIncident colorectal cancer cases (n = 1,142)Non-cases (n = 88,655)
  • BMI, body mass index; HRT, hormone replacement therapy.

  • 1

    Among parous women.

  • 2

    HRT results among postmenopausal women only.

  • 3

    For physical activity and dietary variables, n cases and non-cases were 642 and 48,995, respectively.

Mean age at baseline (years)51.0 (5.4)48.5 (5.6)
Mean BMI (kg/m2)25.3 (4.4)25.1 (4.7)
Mean age at first live birth124.4 (4.5)24.2 (4.8)
Parity (%)
Nulliparous14.614.9
 1–230.835.8
 3–442.938.6
 5+11.710.7
Postmenopausal (%)53.037.2
Cigarette smoking
 Never50.253.3
 Former29.227.3
 Current20.619.3
Oral contraceptive use (% ever)48.059.5
HRT use (% ever)246.547.8
Alcohol consumption (grams/day)3
 Never27.826.3
 1–<319.722.4
 3–<1023.325.1
 10+29.226.2
Vigorous physical activity
 None95.792.7
 Any4.37.3
Mean calorie intake (kcal/day)2,1062,111
Mean fat intake (grams/day)101101
Mean fiber intake (grams/day)2121
Mean folate intake (mg/day)312312
Mean calcium intake (mg/day)839854
Mean red meat intake (grams/day)2930

In multivariate models adjusting for potential confounding variables, ever use of OCs was associated with reduced risk of colorectal cancer (HR 0.83, 95% CI 0.73–0.94). Among current users the HR was 0.76, 95% CI 0.50–1.17 (based on only 24 current users among the cases), whereas among exusers the HR was 0.83, 95% CI 0.73–0.95. Similar associations for ever use of OCs and duration of use were seen for colon and rectal cancer and for the proximal and distal colon (Table II). However, there was no clear evidence of a trend in risk with increasing duration of OC use (Table II) or with years since first, or last, use (data not shown). HRT use (ever/never; current and former use; and duration of use) showed no association with risk of colorectal cancer overall or by subsite. When cases of colorectal cancer diagnosed within the first 2 years of follow-up were excluded from the analysis, the results were unchanged.

Table II. Multivariate-Adjusted Hazard Ratios (HR) and 95% Confidence Intervals (95% CI) for the Association of Oral Contraceptive Use and Hormone Replacement Use with Colorectal Cancer, Overall and by Anatomic Subsite
FactorMultivariate HR1 (95% CI)
Total colorectal cancer (n = 1,1422)Colon cancer (n = 790)Proximal colon (n = 394)Distal colon (n = 299)Rectum (n = 366)
  • 1

    Hazard ratios are adjusted for: age (time to event variable); body mass index (kg/m2) (continuous); menopausal status (pre-, peri-, postmenopausal); pack-years of smoking (never + five levels) and education (3 levels), as well as for the remaining variables in the table.

  • 2

    Fourteen women had a diagnosis of both colon and rectal cancer and were included in the analyses for both the colon and the rectum.

  • 3

    Number of total colorectal cancer cases.

  • 4

    p for trend omitting never users.

  • 5

    Postmenopausal women only.

Oral contraceptives
 Never (n = 585)31.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 Ever (n = 557)0.83 (0.73–0.94)0.81 (0.70–0.94)0.83 (0.67–1.03)0.78 (0.61–1.00)0.85 (0.67–1.03)
Duration of OC use
 Never (n = 585)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 1–11 month (n = 113)0.86 (0.70–1.06)0.84 (0.65–1.07)0.75 (0.52–1.09)0.92 (0.62–1.35)0.87 (0.60–1.27)
 12–35 month (n = 135)0.89 (0.73–1.09)0.88 (0.69–1.12)0.82 (0.58–1.17)0.80 (0.54–1.19)0.88 (0.61–1.26)
 36–71 month (n = 174)0.75 (0.63–0.90)0.75 (0.60–0.93)0.84 (0.62–1.13)0.71 (0.50–1.02)0.75 (0.54–1.04)
 ≥72 month (n = 135)0.84 (0.69–1.03)0.80 (0.63–1.02)0.91 (0.66–1.25)0.75 (0.50–1.11)0.93 (0.66–1.31)
  p for trend0.0050.0080.300.030.23
  p for trend40.590.560.270.340.95
HRT use5
 Never (n = 315)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 Ever (n = 278)1.00 (0.87–1.15)1.01 (0.86–1.20)0.95 (0.75–1.19)1.15 (0.87–1.51)0.97 (0.75–1.26)
Duration of HRT
 Never (n = 315)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 1–12 month (n = 75)0.98 (0.79–1.21)1.02 (0.80–1.31)1.13 (0.81–1.56)1.06 (0.70–1.60)0.94 (0.63–1.39)
 13–59 month (n = 82)1.03 (0.84–1.28)1.05 (0.82–1.35)0.87 (0.60–1.26)1.08 (0.71–1.65)0.95 (0.64–1.43)
 60–119 month (n = 42)0.95 (0.70–1.30)0.79 (0.53–1.17)0.51 (0.26–0.99)1.15 (0.65–2.05)1.32 (0.81–2.15)
 ≥120 month (n = 46)0.86 (0.64–1.15)0.89 (0.63–1.24)0.91 (0.57–1.43)0.90 (0.51–1.61)0.75 (0.42–1.32)
  p for trend0.500.470.170.750.68
  p for trend40.470.390.160.640.71

Compared to women who had never used either OC or HRT, women who were ever users of OC but who had never used HRT were at reduced risk of colorectal cancer (HR 0.86, 95% CI 0.74–1.00), whereas women who had ever used HRT but never used OC showed no association (HR 1.05, 95% CI 0.88–1.26). Women who had ever used both OC and HRT had an HR of 0.80 (95% CI 0.66–0.98).

No associations were seen for age at menarche, age at first live birth, or parity with risk of colorectal cancer overall or by subsite, although parity was associated with a 70% statistically nonsignificant increase in the risk of proximal colon cancer (Table III). The analyses described earlier were repeated in the cohort members for whom dietary information was available, which permitted additional adjustment for physical activity and intake of calories, fat, red meat, fiber, folate, calcium and alcohol (data not shown). Similar reductions in risk to those observed in the total cohort inassociation with OC use were seen for colorectal cancer (HR= 0.79, 95% CI 0.67–0.93), colon cancer (HR = 0.80, 95% CI 0.65–0.97) and rectal cancer (HR = 0.76, 95% CI 0.56–1.02). As in the total cohort, there was no clear trend with duration of OC use. Neither HRT use nor reproductive factors showed evidence of any association with all colorectal cancer or with cancer at specific subsites in the dietary cohort.

Table III. Multivariate-Adjusted Hazard Ratios (HR) and 95% Confidence Intervals (95% CI) for the Association of Reproductive Factors with Colorectal Cancer, Overall and by Anatomic Subsite
FactorMultivariate HR1 (95% CI)
Total colorectal cancer (n = 1,1422)Colon cancer (n = 790)Proximal colon (n = 394)Distal colon (n = 299)Rectum (n = 366)
  • 1

    Hazard ratios are adjusted for: age (time to event variable); body mass index (kg/m2) (continuous); menopausal status (pre-, peri-, postmenopausal); pack-years of smoking (never + five levels); OC use (ever/never), HRT use (ever/never) and education (3 levels), as well as for the remaining variables in the table.

  • 2

    Fourteen women had a diagnosis of both colon and rectal cancer and were included in the analyses for both the colon and the rectum.

  • 3

    Number of total colorectal cancer cases.

Age at menarche
 <12 years (n = 183)31.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 12 years (n = 270)1.15 (0.95–1.41)1.22 (0.96–1.55)1.19 (0.85–1.68)1.43 (0.97–2.12)1.01 (0.71–1.43)
 13 years (n = 342)1.16 (0.96–1.40)1.20 (0.96–1.52)1.12 (0.80–1.56)1.38 (0.94–2.02)1.03 (0.74–1.44)
 ≥14 years (n = 347)1.12 (0.92–1.36)1.17 (0.93–1.48)1.29 (0.94–1.79)1.05 (0.71–1.56)0.97 (0.69–1.35)
  p for trend0.400.340.180.730.84
Age at first live birth
 Nulliparous (n = 162)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 <23 years (n = 338)1.04 (0.85–1.27)1.03 (0.81–1.31)1.38 (0.96–1.98)0.79 (0.55–1.14)1.04 (0.73–1.48)
 23–25 years (n = 297)1.10 (0.90–1.34)1.16 (0.92–1.48)1.57 (1.10–2.24)0.90 (0.63–1.31)0.93 (0.65–1.34)
 26–29 years (n = 231)1.10 (0.89–1.35)1.10 (0.86–1.42)1.27 (0.87–1.86)0.83 (0.56–1.23)1.05 (0.72–1.51)
 ≥30 years (n = 109)0.94 (0.73–1.20)0.97 (0.72–1.31)1.15 (0.74–1.79)0.94 (0.58–1.43)0.94 (0.61–1.46)
  p for trend0.940.740.690.820.81
Parity
 Nulliparous (n = 162)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)1.00 (reference)
 1–2 (n = 372)1.17 (0.80–1.73)1.10 (0.70–1.75)1.69 (0.87–3.38)0.71 (0.34–1.45)1.17 (0.59–2.32)
 3–4 (n = 448)1.21 (0.84–1.74)1.17 (0.76–1.80)1.71 (0.91–3.20)0.84 (0.43–1.64)1.11 (0.59–2.12)
 5+ (n = 159)1.18 (0.81–1.71)1.20 (0.77–1.86)1.79 (0.94–3.39)0.79 (0.40–1.59)1.05 (0.54–2.05)
  p for trend0.500.280.200.730.71

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In the time since McMichael and Potter3 proposed that exposure to endogenous and exogenous steroid hormones might confer protection against colon cancer, numerous studies have been carried out examining the role of reproductive and hormonal factors in the etiology of the disease. A number of meta-analyses and reviews of the association of either OC or HRT use with colorectal cancer have appeared.10, 11, 12, 23, 24, 25 However, relatively few cohort studies have examined OC use,13, 14, 15, 16, 17, 18 and most had relatively small numbers of cases (501, 203, 170, 146, 206 and 267 cases, respectively), providing limited power to examine subsites within the colorectum. The large number of cases of colorectal cancer ascertained in the current study enabled us both to examine the effect of different exposures for subsites within the colorectum while adjusting for a number of known risk factors for colorectal cancer, and to examine the joint effect of OC and HRT use.

Our finding of reduced risk of colorectal cancer among users of OC (HR 0.83, 95% CI 0.73–0.94) is in agreement with the findings of some13, 15, 16, 18 but not all studies,14, 17, 26 which have examined this association. In the meta-analysis by Fernandez et al.,10 the pooled estimate from 4 cohort studies was 0.84 (95% CI 0.72–0.97), and the pooled estimate from cohort and case–control studies combined was 0.82 (95% CI 0.74–0.92), without apparent heterogeneity. Longer duration of use was not associated with a greater reduction in risk, but there was a suggestion that more recent use was associated with reduced risk (RR = 0.46, 95% CI 0.30–0.71). In a recent report based on analysis of a cohort study, Lin et al.18 obtained a relative risk of 0.67 (95% CI 0.50–0.89) for ever use of OCs, but there was no evidence of a trend with duration of use. In the present study, there were very few current users of OCs at baseline, so that almost all OC use was in the past. There were no clear trends with duration of use, years since first use, or years since last use. It is possible that the lack of a dose–response relationship with OC use is due to poor recall of actual duration of OC use.

In our data, HRT use showed no association with colorectal cancer risk, whereas meta-analyses of HRT and colon or colorectal cancer indicate a modest reduction in risk on the order of 15–20% for both colon and rectal cancer.11, 12 A number of studies reporting an association of HRT use with colorectal cancer risk did not adjust for possible confounding effects of prior OC use.14, 26, 27, 28, 29, 30 In our data, there was no association with HRT use, whether or not OC use (ever/never) was included as a covariate in the model.

The lack of association between reproductive or menstrual factors and risk of colorectal cancer in the present study is in accord with the results of most of the large number of studies that have addressed this question over the past 25 years.30, 31, 32

Among the strengths of our study are the large number of cases, the completeness of follow-up and the ability to control for a number of risk factors for colorectal cancer. In addition, reporting of ever use of OC at baseline in this cohort should be reliable, since the average age of women at entry was 49, by which time most of those who had ever used OCs had stopped using them. Limitations of our study include the fact that exposure information was obtained only at baseline and that changes in exposure over the 16-year follow-up period might have resulted in misclassification of exposure, reducing statistical power to detect an association. This is particularly true for HRT use (both ever use and duration). Another limitation, which applies to most previous studies, is the lack of information on dosage of exogenous hormone preparations and the type of HRT.

In conclusion, our results are suggestive of a modest inverse association between OC use and colorectal cancer risk but not of an association of HRT use with this disease. Further studies are needed to clarify the effects of specific hormone preparations, the timing of postmenopausal hormone use and the combined effects of OC and HRT.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Statistics Canada, the provincial and territorial Registrars of Vital Statistics, and the Cancer Registry directors for their assistance in making the cancer incidence and mortality data available.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics. CA Cancer J Clin 2005; 55: 74108.
  • 2
    Fraumeni JF,Jr, Lloyd JW, Smith EM, Wagoner JK. Cancer mortality among nuns: role of marital status in the etiology of neoplastic disease in women. J Natl Cancer Inst 1969; 42: 45568.
  • 3
    McMichael AJ, Potter JD. Reproduction, endogenous and exogenous sex hormones, and colon cancer: a review and hypothesis. J Natl Cancer Inst 1980; 65: 12017.
  • 4
    Pasi AJ, Robert JM. Chemoprevention of colorectal cancer. New Engl J Med 2000; 342: 19608.
  • 5
    Martineti V, Picariello L, Tognarini I, Sala SC, Gozzini A, Azzari C, Mavilia C, Tanini A, Falchetti A, Fiorelli G, Tonelli F, Brandi ML. ERβ is a potent inhibitor of cell proliferation in the HCT8 human colon cancer cell line through regulation of cell cycle components. Endocr Relat Cancer 2005; 12: 45569.
  • 6
    Foley EF, Jazaeri AA, Shupnik MA, Jazaeri O, Rice LW. Selective loss of estrogen receptor β in malignant human colon. Cancer Res 2000; 60: 2458.
  • 7
    Issa JP, Ottaviano YL, Celano P, Hamilton SR, Davidson NE, Baylin SB. Methylation of the oestrogen receptor CpG island links ageing and neoplasia in human colon. Nat Genet 1994; 7: 536540.
  • 8
    Smirnoff P, Liel Y, Gnainsky J, Shany S, Schwartz B. The protective effect of estrogen against chemically induced murine colon carcinogenesis is associated with decreased CpG island methylation and increased mRNA and protein expression of the colonic vitamin D receptor. Oncol Res 1999; 11: 25564.
  • 9
    Odagiri E, Jibiki K, Kato Y, Nakamura S, Oda S-I, Demura R, Demura H. Steroid receptors in dimethylhydrazine-induced colon carcinogenesis. Cancer 1985; 56: 262734.
  • 10
    Fernandez E, La Vecchia C, Balducci A, Chatenoud L, Franceschi S, Negri E. Oral contraceptives and colorectal cancer risk: a meta-analysis. Brit J Cancer 2001; 84: 7227.
  • 11
    Hébert-Croteau N. A meta-analysis of hormone replacement therapy and colon cancer in women. Cancer Epidemiol Biomarkers Prev 1998; 7: 6539.
  • 12
    Grodstein F, Newcomb PA, Stampfer RJ. Postmenopausal hormone therapy and the risk of colorectal cancer: a review and meta-analysis. Am J Med 1999; 106: 57482.
  • 13
    Martinez ME, Grodstein F, Giovannucci E, Colditz GA, Speizer FE, Hennekens C, Rosner B, Willett WC, Stampfer MJ. A prospective study of reproductive factors, oral contraceptive use, and risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev 1997; 6: 15.
  • 14
    Troisi R, Schairer C, Chow WH, Schatzkin A, Brinton LA, Fraumeni JF,Jr. Reproductive factors, oral contraceptive use, and risk of colorectal cancer. Epidemiology 1997; 8: 759.
  • 15
    Beral V, Hermon C, Kay C, Hannaford P, Darby S, Reeves G. Mortality associated with oral contraceptive use: 25 year follow up of cohort of 46,000 women from Royal College of General Practitioners' oral contraception study. Br Med J 1999; 318: 96100.
  • 16
    Hannaford P, Elliott A. Use of exogenous hormones by women and colorectal cancer: evidence from the Royal College of General Practitioner's Oral Contraception Study. Contraception 2005; 71: 958.
  • 17
    Vessey M, Painter R. Oral contraceptive use and cancer: findings in a large cohort study, 1968–2004. Br J Cancer 2006; 95: 3859.
  • 18
    Lin J, Zhang SM, Cook NR, Manson JE, Buring JE, Lee I-M. Oral contraceptives, reproductive factors, and risk of colorectal cancer among women in a prospective cohort study. Am J Epidemiol 2007; 165: 794801.
  • 19
    Terry PD, Jain M, Miller AB, Howe GR, Rohan TE. Dietary carotenoid intake and colorectal cancer risk. Nutr Cancer 2002; 42: 16772.
  • 20
    Miller AB, Baines CJ, To T, Wall C. Canadian National Breast Screening Study. I. Breast cancer detection and death rates among women aged 40–49 years. II. Breast cancer detection and death rates among women aged 50–59 years. Can Med Assoc J 1992; 147: 145988.
  • 21
    Jain M, Harrison L, Howe GR, Miller AB. Evaluation of a self-administered dietary questionnaire for use in a cohort study. Am J Clin Nutr 1982; 36: 9315.
  • 22
    Rothman KJ, Greenland S. Modern epidemiology. Philadelphia, PA: Lippincott-Raven, 1998. 319.
  • 23
    Calle EE. Hormone replacement therapy and colorectal cancer: interpreting the evidence. Cancer Causes Control 1997; 8: 1279.
  • 24
    Franceschi S, La Vecchia C. Oral contraceptives and colorectal tumors. Contraception 1998; 58: 33543.
  • 25
    Barrett-Connor E, Grady D, Stefanick ML. The rise and fall of menopausal hormone therapy. Annu Rev Public Health 2005; 26: 11540.
  • 26
    Chute CG. Willett WC, Colditz GA, Stampfer MJ, Rosner B, Speizer FE. A prospective study of reproductive history and exogenous estrogens on the risk of colorectal cancer in women. Epidemiology 1991; 2: 2017.
  • 27
    Furner SE, Davis FG, Nelson RL, Haenszel W. A case-control study of large bowel cancer and hormone exposure in women. Cancer Res 1989; 49: 493640.
  • 28
    Gerhardsson de Verdier M, London S. Reproductive factors, exogenous female hormones, and colorectal cancer by subsite. Cancer Causes Control 1992; 3: 35560.
  • 29
    Jacobs EJ, White E, Weiss NS. Exogenous hormones, reproductive history, and colon cancer (Seattle, Washington, USA). Cancer Causes Control 1994; 5: 35966.
  • 30
    Kampman E, Potter JD, Slattery ML, Caan BJ, Edwards S. Hormone replacement therapy, reproductive history, and colon cancer: a multi-center case-control study. Cancer Causes Control 1997; 8: 14658.
  • 31
    Marcus PM, Newcomb PA, Young T, Storer BE. The association of reproductive and menstrual characteristics and colon and rectal cancer risk in Wisconsin women. Ann Epidemiol 1995; 5: 3039.
  • 32
    Potter JD, Hunter D. Colorectal cancer. In: AdamiH-O, HunterD, TrichopoulosD, eds. Textbook of cancer epidemiology. Oxford: Oxford University Press, 2002; 188211.