Presentation
Contraception, Hormones, and Cancer: Exploring the Link Between Birth Control and Cancer Risk
SessionPoster Session 2
DescriptionIntroduction:
In 2021, breast cancer became the most commonly diagnosed cancer in women worldwide, with 2.26 million new cases and 685,000 deaths, surpassing lung cancer as the leading cancer in women. It remains a major cause of mortality, responsible for the death of about one in five women diagnosed. Ovarian cancer, though less common, is among the most lethal gynecologic malignancies, with 313,959 new cases and 207,252 deaths reported globally in 2020, and U.S. estimates for 2025 projecting 20,890 new cases and 12,730 deaths. The global burden continues to rise, with ovarian cancer cases nearly doubling since 1990, and breast cancer projected to increase by 38% in incidence and 68% in mortality. These patterns underscore the urgency of clarifying modifiable risk factors.
The widespread use of hormonal and non-hormonal contraceptives has raised questions about long-term safety in relation to hormone-sensitive cancers. While contraceptives are effective in preventing unintended pregnancies, evidence on their cancer associations is mixed. A review of 51 studies reported that oral contraceptives protect against ovarian and endometrial cancers without increasing breast cancer risk in healthy women, though BRCA1/2 carriers showed elevated breast cancer risk. Complementing this, a large cohort study found a modest overall increase in breast cancer risk, equivalent to one additional case per 7,690 hormonal contraceptive users. Despite extensive research, findings remain inconsistent, particularly for breast cancer, and most studies focus primarily on hormonal methods. Limited comparisons with non-hormonal options, such as intrauterine devices and tubal ligation, highlight the need for broader analyses to better understand how different contraceptive methods influence breast and ovarian cancer risk.
This study aims to evaluate the potential link between commonly used contraceptive methods and the risk of developing breast/ovarian cancer, with particular attention to hormonal methods (OC: oral contraceptives; IUDs: intrauterine devices) compared to non-hormonal approaches (TL: tubal ligation; NI: non-invasive methods). We hypothesize that contraceptive methods have different effects on breast and ovarian cancer risk, with hormonal methods potentially increasing breast cancer risk while offering protection against ovarian cancer, whereas non-hormonal methods are expected to show neutral or minimal effects.
Approach:
This study utilizes data from the All of Us Research Program, a large-scale national health initiative developed by the National Institutes of Health. The program provides a representative U.S. participant pool, making it a robust resource for population-level cancer research. The All of Us Research Program’s Registered Tier Dataset v7 was used to build the cohort, extract features, and conduct similarity analysis of contraceptive methods and cancer risk.
Female patients were stratified by contraceptive history, with cases defined as breast or ovarian cancers diagnosed ≥6 months after use. Exposures were grouped into four categories: oral contraceptives (OC, hormonal), intrauterine devices (IUDs, hormonal), tubal ligation (TL, non-hormonal), and non-invasive (NI, non-hormonal) methods.
To evaluate the association between contraceptive use and cancer incidence, a Jaccard similarity matrix was applied, which quantifies the overlap between women exposed to a given contraceptive method and those diagnosed with breast or ovarian cancer. The index is expressed as the ratio of the intersection to the union of the two groups.
Findings:
Dataset Characteristics: The cohort included 888 cases with 811 OC users, 67 IUD users, 20 with TL, and 12 using NI methods. Among them, 763 had breast cancer and 127 had ovarian cancer. Most participants were aged more than 45 years old and are predominantly White, with smaller African American, Asian, and mixed-race groups.
Similarity Results: The Jaccard index ranges from 0 (minimal similarity) to 1 (strong similarity), with values >0.70 indicating strong similarity. Oral contraceptives (OC, hormonal) showed the highest similarity with cancer incidence overall (J = 0.945), driven largely by breast cancer (J = 0.812), while the association with ovarian cancer was weak (J = 0.147). Intrauterine devices (IUDs, hormonal) demonstrated minimal similarity (J = 0.075 overall; J = 0.074 breast; J = 0.049 ovarian). Tubal ligation (TL, non-hormonal) and non-invasive (NI, non-hormonal) methods yielded very low similarity indices (J ≤ 0.024 across outcomes), reflecting negligible overlap with cancer incidence. Collectively, these findings suggest oral contraceptives are most strongly linked to breast cancer risk, whereas IUDs, TL, and NI methods show little or no association with cancer outcomes.
Comparison With Other Studies: The associations observed in this study are largely consistent with prior epidemiological evidence. The strong similarity between oral contraceptive use and breast cancer (J = 0.812) aligns with findings from previous findings, including a meta-analysis of over 50,000 cases across 54 studies that reported a relative risk of 1.24 for current users, and a large Danish cohort showing a 20–30% increase in breast cancer risk varying by formulation and duration. By contrast, our analysis did not identify a protective association between oral contraceptives and ovarian cancer (J = 0.147), in contrast to landmark evidence from the Collaborative Group on Epidemiological Studies of Ovarian Cancer, which reported a 29% risk reduction per five years of use and long-term protective effects. The weaker association in our study may be due to the relatively small number of ovarian cancer cases, limiting statistical power
The minimal similarity observed for intrauterine devices (IUDs; J = 0.075 overall) suggests little measurable influence on cancer risk, consistent with prior evidence showing mixed or modest effects depending on device type. The current dataset used in this study did not distinguish between hormonal and non-hormonal IUDs, which may explain the low indices, as copper IUDs have been linked to reduced ovarian cancer risk while levonorgestrel-releasing devices have shown neutral or slightly increased breast cancer risk. Similarly, the very low similarity values for tubal ligation (TL) and non-invasive (NI) methods (J ≤ 0.024) indicate negligible overlap with cancer incidence, which is biologically plausible given their non-hormonal mechanisms. The absence of measurable associations in our analysis may in part reflect the limited size of the TL subgroup; however, further investigation in a larger, well-powered cohort is needed to clarify the potential relationship.
Overall, the present findings reinforce the complexity of contraception-hormone-cancer relationships. Oral contraceptives appear most strongly associated, showing a role of modestly increasing breast cancer risk. Non-hormonal methods such as tubal ligation and non-invasive methods appear to exert little to no measurable effect on breast or ovarian cancer risk in this dataset.
Takeaways From Human Factors Perspectives:
From a human factors’ perspective, the link between contraception and cancer risk extends beyond clinical outcomes to how women perceive, choose, and adhere to contraceptive methods. Decision-making is shaped by risk perception, trust in medical advice, cultural attitudes, and health literacy. Hormonal methods such as oral contraceptives and hormonal IUDs raise unique challenges: while effective, oral contraceptives require strict daily adherence, and hormonal IUDs may raise concerns about systemic side effects, both of which influence continuation rates. Non-hormonal methods, including copper IUDs, tubal ligation, and barrier-based or fertility-awareness strategies, reduce cognitive and behavioral burden by eliminating daily compliance but may be seen as less flexible, particularly when irreversibility or user-dependent consistency are factors. Communicating nuanced risks, for example, oral contraceptives modestly increasing breast cancer risk while (possibly) decreasing ovarian cancer risk, requires clear, culturally sensitive approaches that account for these differences in usability and perception. Addressing disparities in access and representation within contraceptive care is essential, as women may face additional barriers both in obtaining services and in accessing accurate cancer risk information.
In 2021, breast cancer became the most commonly diagnosed cancer in women worldwide, with 2.26 million new cases and 685,000 deaths, surpassing lung cancer as the leading cancer in women. It remains a major cause of mortality, responsible for the death of about one in five women diagnosed. Ovarian cancer, though less common, is among the most lethal gynecologic malignancies, with 313,959 new cases and 207,252 deaths reported globally in 2020, and U.S. estimates for 2025 projecting 20,890 new cases and 12,730 deaths. The global burden continues to rise, with ovarian cancer cases nearly doubling since 1990, and breast cancer projected to increase by 38% in incidence and 68% in mortality. These patterns underscore the urgency of clarifying modifiable risk factors.
The widespread use of hormonal and non-hormonal contraceptives has raised questions about long-term safety in relation to hormone-sensitive cancers. While contraceptives are effective in preventing unintended pregnancies, evidence on their cancer associations is mixed. A review of 51 studies reported that oral contraceptives protect against ovarian and endometrial cancers without increasing breast cancer risk in healthy women, though BRCA1/2 carriers showed elevated breast cancer risk. Complementing this, a large cohort study found a modest overall increase in breast cancer risk, equivalent to one additional case per 7,690 hormonal contraceptive users. Despite extensive research, findings remain inconsistent, particularly for breast cancer, and most studies focus primarily on hormonal methods. Limited comparisons with non-hormonal options, such as intrauterine devices and tubal ligation, highlight the need for broader analyses to better understand how different contraceptive methods influence breast and ovarian cancer risk.
This study aims to evaluate the potential link between commonly used contraceptive methods and the risk of developing breast/ovarian cancer, with particular attention to hormonal methods (OC: oral contraceptives; IUDs: intrauterine devices) compared to non-hormonal approaches (TL: tubal ligation; NI: non-invasive methods). We hypothesize that contraceptive methods have different effects on breast and ovarian cancer risk, with hormonal methods potentially increasing breast cancer risk while offering protection against ovarian cancer, whereas non-hormonal methods are expected to show neutral or minimal effects.
Approach:
This study utilizes data from the All of Us Research Program, a large-scale national health initiative developed by the National Institutes of Health. The program provides a representative U.S. participant pool, making it a robust resource for population-level cancer research. The All of Us Research Program’s Registered Tier Dataset v7 was used to build the cohort, extract features, and conduct similarity analysis of contraceptive methods and cancer risk.
Female patients were stratified by contraceptive history, with cases defined as breast or ovarian cancers diagnosed ≥6 months after use. Exposures were grouped into four categories: oral contraceptives (OC, hormonal), intrauterine devices (IUDs, hormonal), tubal ligation (TL, non-hormonal), and non-invasive (NI, non-hormonal) methods.
To evaluate the association between contraceptive use and cancer incidence, a Jaccard similarity matrix was applied, which quantifies the overlap between women exposed to a given contraceptive method and those diagnosed with breast or ovarian cancer. The index is expressed as the ratio of the intersection to the union of the two groups.
Findings:
Dataset Characteristics: The cohort included 888 cases with 811 OC users, 67 IUD users, 20 with TL, and 12 using NI methods. Among them, 763 had breast cancer and 127 had ovarian cancer. Most participants were aged more than 45 years old and are predominantly White, with smaller African American, Asian, and mixed-race groups.
Similarity Results: The Jaccard index ranges from 0 (minimal similarity) to 1 (strong similarity), with values >0.70 indicating strong similarity. Oral contraceptives (OC, hormonal) showed the highest similarity with cancer incidence overall (J = 0.945), driven largely by breast cancer (J = 0.812), while the association with ovarian cancer was weak (J = 0.147). Intrauterine devices (IUDs, hormonal) demonstrated minimal similarity (J = 0.075 overall; J = 0.074 breast; J = 0.049 ovarian). Tubal ligation (TL, non-hormonal) and non-invasive (NI, non-hormonal) methods yielded very low similarity indices (J ≤ 0.024 across outcomes), reflecting negligible overlap with cancer incidence. Collectively, these findings suggest oral contraceptives are most strongly linked to breast cancer risk, whereas IUDs, TL, and NI methods show little or no association with cancer outcomes.
Comparison With Other Studies: The associations observed in this study are largely consistent with prior epidemiological evidence. The strong similarity between oral contraceptive use and breast cancer (J = 0.812) aligns with findings from previous findings, including a meta-analysis of over 50,000 cases across 54 studies that reported a relative risk of 1.24 for current users, and a large Danish cohort showing a 20–30% increase in breast cancer risk varying by formulation and duration. By contrast, our analysis did not identify a protective association between oral contraceptives and ovarian cancer (J = 0.147), in contrast to landmark evidence from the Collaborative Group on Epidemiological Studies of Ovarian Cancer, which reported a 29% risk reduction per five years of use and long-term protective effects. The weaker association in our study may be due to the relatively small number of ovarian cancer cases, limiting statistical power
The minimal similarity observed for intrauterine devices (IUDs; J = 0.075 overall) suggests little measurable influence on cancer risk, consistent with prior evidence showing mixed or modest effects depending on device type. The current dataset used in this study did not distinguish between hormonal and non-hormonal IUDs, which may explain the low indices, as copper IUDs have been linked to reduced ovarian cancer risk while levonorgestrel-releasing devices have shown neutral or slightly increased breast cancer risk. Similarly, the very low similarity values for tubal ligation (TL) and non-invasive (NI) methods (J ≤ 0.024) indicate negligible overlap with cancer incidence, which is biologically plausible given their non-hormonal mechanisms. The absence of measurable associations in our analysis may in part reflect the limited size of the TL subgroup; however, further investigation in a larger, well-powered cohort is needed to clarify the potential relationship.
Overall, the present findings reinforce the complexity of contraception-hormone-cancer relationships. Oral contraceptives appear most strongly associated, showing a role of modestly increasing breast cancer risk. Non-hormonal methods such as tubal ligation and non-invasive methods appear to exert little to no measurable effect on breast or ovarian cancer risk in this dataset.
Takeaways From Human Factors Perspectives:
From a human factors’ perspective, the link between contraception and cancer risk extends beyond clinical outcomes to how women perceive, choose, and adhere to contraceptive methods. Decision-making is shaped by risk perception, trust in medical advice, cultural attitudes, and health literacy. Hormonal methods such as oral contraceptives and hormonal IUDs raise unique challenges: while effective, oral contraceptives require strict daily adherence, and hormonal IUDs may raise concerns about systemic side effects, both of which influence continuation rates. Non-hormonal methods, including copper IUDs, tubal ligation, and barrier-based or fertility-awareness strategies, reduce cognitive and behavioral burden by eliminating daily compliance but may be seen as less flexible, particularly when irreversibility or user-dependent consistency are factors. Communicating nuanced risks, for example, oral contraceptives modestly increasing breast cancer risk while (possibly) decreasing ovarian cancer risk, requires clear, culturally sensitive approaches that account for these differences in usability and perception. Addressing disparities in access and representation within contraceptive care is essential, as women may face additional barriers both in obtaining services and in accessing accurate cancer risk information.
Event Type
Poster Presentation
TimeTuesday, March 244:45pm - 6:15pm EDT
LocationRhinelander Gallery
Patient Safety Research and Initiatives