The use of combined oral contraceptives, progestogen-only pills, or hormonal intrauterine devices neither protects against nor contributes to the risk of colorectal cancer (CRC) in premenopausal women, according to a study.
“Although some subgroup estimates were imprecise, no support was found for an association between hormonal contraceptive use and CRC,” the researchers said.
A total of 1,956,948 women were followed for a median of 12.5 years. Of these, 1,878 had a CRC diagnosis. [BMJ 2026;394:e100065]
The incidence rate ratio (IRR) of CRC among current and recent users vs never users of hormonal contraception was 0.94 (95 percent confidence interval [CI], 0.83‒1.06). For use of <5 years, the IRR was 0.97 (95 percent CI, 0.85‒1.11), while that for >10 years was 0.86 (95 percent CI, 0.62‒1.18).
However, the IRR in previous users was comparable to that of never users, with some indication for a reduced IRR at 10 or more years following cessation (IRR, 0.81, 95 percent CI, 0.66‒0.99). No consistent differences were observed in the risk between newer and older combined oral contraceptives.
Finally, the most used progestogen-only pills had an IRR of 1.09 (95 percent CI, 0.74‒1.61), and the most used progestogen-only nonoral product (ie, levonorgestrel releasing intrauterine device) showed an IRR of 0.96 (95 percent CI, 0.81‒1.15).
“These findings indicate that hormonal contraceptives do not influence CRC risk in premenopausal women,” the researchers said. “Hormonal contraceptives should neither be presented as giving protective nor contributory effects to the rising incidence of CRC among young women.
Study design
These results are also consistent with those of the case-control study by Nichols and colleagues, the cohort studies form the Nurses’ Health Study, and Rosenblatt and colleagues. [Eur J Cancer 2004;40:590-593; Cancer Epidemiol Biomarkers Prev 2005;14:1212-128; Cancer Epidemiol Biomarkers Prev 2015;24:1214-1221]
“Our findings do not align with a [2022] review and meta-analysis, which also added studies of postmenopausal women,” the researchers said. “Because previous studies assessed older formulations at an earlier time, differences in formulation, dose, and calendar period could explain inconsistencies in study findings over time.” [Ann Med Surg (Lond) 2022;83:104254]
The current nationwide cohort study used national registers in Denmark and included women aged 15‒49 years who had at least 5 years of residence and no history of cancer, hysterectomy, oophorectomy, or sterilization. The researchers calculated the adjusted IRRs for a first-time diagnosis of CRC by type and duration of hormonal contraceptive use.
“Our study does not assess a potential protective effect from past hormonal contraception use occurring decades after cessation in postmenopausal women, a topic for future studies,” the researchers said.
“Risk estimates for less commonly used products, such as combined products with drospirenone (six events of use in our study), were imprecise. Studies across countries may help provide more accurate estimates for these less frequently used products,” they added.