NHS Evidence Health Information Resources formerly National Library for Health

On 12 September 2007, seven newspapers (1-7) reported that the contraceptive pill protects women against cancer. These were generally accurate reports of a well conducted large retrospective study. The results are likely to be reliable for a population similar to that studied, but may not be generalisable to current oral contraceptive (OC) pill users.

On 12 September 2007, seven newspapers (1-7) reported that the contraceptive pill protects women against cancer. Six reported that taking for over eight years however could increase cancer risk.
The reports were based on a well conducted retrospective study (8) and an associated editorial (9) published in the BMJ which analysed cancer rates for a large number of women recruited to a study initiated in the 1960s. Overall, using the main dataset, OC use reduced the risk of any cancer by 12%; significant reductions were seen in the risk of large bowel/rectum (28%), uterine (42%), and ovarian (46%) cancers. After eight years of OC use, women had an increased risk of all cancers (22%), invasive cervical cancer (173%), and central nervous system/pituitary cancer (451%), and a decrease in ovarian cancer (62%). The reduction in ovarian cancer persisted for up to 15 years after stopping OC use. The results are likely to be reliable and applicable to a population similar to that studied rather than current OC users.
The newspapers generally accurately reported the results of the research. Three of the newspapers (1-3) reported the number of women who would need to take the OC pill to avoid one cancer, figures which were not presented in the published report.

Evaluation of the evidence base for the risk of women developing cancer as a result of long-term use of oral contraceptives

Where does the evidence come from?

The study was conducted by researchers from the University of Aberdeen, and the Centre for Rural Health, Inverness, and led by Professor Hannaford. The research was funded by the Royal College of General Practitioners, Medical Research Council, Imperial Cancer Research Fund, British Heart Foundation, Schering AG, Schering Health Care, Wyeth Ayerst International, Ortho Cilag and Searle.

What were the authors' objectives?

To investigate the associated risk of different types of cancer with long-term oral contraceptive use.

What was the nature of the evidence?

The study was based upon data from The Royal College of General Practitioners' oral contraception (OC) study which recruited approximately 47,000 women during 1968-1969. All the women were either married or in a stable relationship with approximately half who used OCs (ever users) and half who had never used OCs (never users). From the resulting database, the authors produced two sets of data. It appears that the main set of data contained information on cancer rates for the 76% of women who were still being followed by the study in the mid 1970s when the recording of the incidence of cancers and deaths began (with the exception of women who were classified as never users who were under the age of 38 in 1996, and were lost to the study prior to this date), as well as women who were lost to follow-up until that point of loss. The second data set was restricted to women who were followed-up by their general practitioners until either developing cancer or dropping out of the study. Women recruited as never users who subsequently started using OCs were included in the ever users group from the date of starting OC.

Cancer incidence (standardised by age, parity, smoking, social class and hormone replacement therapy) and relative risk were presented for the following cancer sites: large bowel/rectum; gallbladder/liver; lung; melanoma; breast; invasive cervix; uterine body; ovary; central nervous system/pituitary; main gynaecological (combined); site unknown, and other cancers. Overall standardised cancer incidence rates and relative risk were also calculated for different age groups of women (<30, 30-39, 40-49, 50-59, >/=60), different durations of OC use (<48 months, 49-96 months, >97 months) and the number of months since last use of OC. (<60; 61-120; 121-180; 181-240; >/=241).

What were the findings?

Overall, using the main dataset, ever users had a significant reduction (12%) in the risk of any cancer compared to never users; significant reductions were seen in the rate of cancers of the large bowel/rectum (28%), uterine body (42%), ovaries (46%), and cancers classified as unknown site (36%) or 'other' (12%). There was no difference in the rate of breast cancer.

Women who used OC for over eight years had significant increases in the risk of all cancers (22%), invasive cervical cancer (173%), and cancer of the central nervous system/pituitary (451%), and a significant decrease in the incidence of ovarian cancer (62%) when compared to never users; these differences were not evident in women who took OC for shorter periods. There were no other significant differences between ever users and never users for any category of cancer when classified according to duration of OC use.

The significant reduction in the incidence of ovarian cancer persisted for up to 15 years after discontinuing OC use, after which time, there was no difference in the incidence between previous users and never users. Results for other categories of cancer were less consistent with time after discontinuation of OC use.

What were the authors' conclusions?

The authors concluded that OC use was not associated with an overall increase in the risk of cancer, but further research is required to quantify the balance of cancer risks and benefits in different parts of the world, including effects on mortality, as cancer risk will vary depending upon the levels and duration of OC use.

How reliable are the conclusions?

The research was a well conducted retrospective review of information from a large number of women held on a database constructed during a long-term observational study, although it was unclear as to the total number of women included in this study. There were some limitations of the study. Firstly, the data related to women taking a variety of different OC pills, including progesterone only and pills containing progesterone combined with <50, 50 and >50 micrograms of oestrogen; 75% of the women were taking the combined OC pill with 50 micrograms of oestrogen. It was therefore unclear whether particular preparations of OC pills were responsible for the responses observed. Secondly, given that the women were originally recruited in the 1960's, the preparations used at this time may vary considerably to those prescribed today, and the cancer risk relating to currently used preparations may be different. Thirdly, the cohort of women were primarily women under the age of 30 years at recruitment (64%) and white, and there were some disparities between the ever user and never user women; ever users tended to be younger, less likely to be smokers, although these factors were adjusted for in the analyses. Fourthly, 37% of women were no longer being followed-up, and this is a high dropout rate. Finally, making a large number of comparisons from such a dataset is likely to result in significant differences between groups for some of these comparisons by chance.

Although there seems to be some inconsistencies of the differences between ever and never users across the analyses, the results are based on a large number of women, and the results are likely to be reliable and applicable to a population similar to that studied. However, given the limitations of the study mentioned, and the differences in the preparations used in current practice, application of these results to current OC users should be made with caution.

Systematic reviews

Information staff at CRD searched for systematic reviews relevant to this topic. Systematic reviews are valuable sources of evidence as they locate, appraise and synthesize all available evidence on a particular topic.

There were no related systematic reviews identified on the Cochrane Database of Systematic Reviews (CDSR), however there were five on the Database of Abstracts of Reviews of Effects (DARE) (10-14).

References and resources

1. The pill: new evidence shows it helps protect against cancer. The Guardian, 12 September 2007, p1.

2. Taking the Pill cuts the risk of cancer. The Times, 12 September 2007, p1,4.

3. The Pill 'protects against cancer'. The Independent, 12 September 2007, p16.

4. The Pill 'reduces the risk of cancer in later life'. Daily Mail, 12 September 2007, p40.

5. Taking Pill cuts cancer risk. Daily Express, 12 September 2007, p11.

6. The Pill reduces risk of womedn getting cancer. Daily Mirror, 12 September 2007, p8.

7. Long-term Pill use link to cancer. Daily Telegraph, 12 September 2007, p4.

8. Hannaford PC, Selvaraj S, Elliott AM, Angus V, Iversen L, Lee AJ. Cancer risk among users of oral contraceptives: cohort data from the Royal College of General Practitioner's oral contraception study. BMJ September 2007; doi:10.1136/bmj.39289.649410.55.

9. Meirik O, Farley TMM. Risk of cancer and the oral contraceptive pill. BMJ September 2007; doi:10.1136/bmj.39336.503067.BE.

10. Costa H L, Doyle P. Influence of oral contraceptives in the development of post-molar trophoblastic neoplasia: a systematic review. Gynecologic Oncology 2006;100(3):579-585. [DARE Abstract]

11. Fernandez E, La Vecchia C, Balducci A, Chatenoud L, Franceschi S, Negri E. Oral contraceptives and colorectal cancer risk: a meta-analysis. British Journal of Cancer 2001;84(5):722-727. [DARE Abstract]

12. Pfahlberg A, Hassan K, Wille L, Lausen B, Gefeller O. Systematic review of case-control studies: oral contraceptives show no effect on melanoma risk. Public Health Reviews 1997;25(3-4):309-315. [DARE Abstract]

13. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713-1727. [DARE Abstract]

14. Smith J S, Green J, Berrington de Gonzalez A, Appleby P, Peto J, Plummer M, Franceschi S, Beral V. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159-1167. [DARE Provisional Abstract]