This massive Swedish cohort study serves as a critical update for anyone managing contraception. While the absolute risk remains low—approximately 1 extra case per 7,752 users per year—the relative increase in breast cancer risk is substantial, driven largely by specific hormonal components. 

The overall Hazard Ratio of 1.24 suggests a 24% increase in relative risk of diagnosis for ever-users of hormonal contraceptives. However, the data strongly suggest that this risk is not uniform. Formulations containing desogestrel (both oral and implant) showed higher relative risks than those with levonorgestrel or drospirenone.

Clinical Bottom Line

This large, nationwide cohort study provides compelling evidence that the risk of breast cancer associated with hormonal contraceptives (HCs) varies significantly by formulation. While the overall association between HC use and breast cancer risk remains modest (Hazard Ratio 1.24), the study highlights a specific, higher risk profile for progestin-only formulations containing desogestrel (oral pills and etonogestrel implants) compared to levonorgestrel-containing options (combined pills and IUS).

Clinicians should note that while the absolute risk increase is small (approximately 13 additional cases per 100,000 users per year), the choice of progestin appears to be a modifiable risk factor, with levonorgestrel-based products and combined drospirenone demonstrating a safer profile in this dataset. However, as an observational study, these findings establish association, not causation, and potential confounding by indication or unmeasured lifestyle factors cannot be fully ruled out.

Results in Context

Main Results

• Overall Risk: Ever use of any hormonal contraceptive was associated with a statistically significant increased risk of breast cancer compared to never use.

• Hazard Ratio (HR): 1.24 (95% CI, 1.20-1.28).

• Absolute Risk: This corresponds to 1 additional case per 7,752 users per year.

Formulation-Specific Risks:

• Progestin-Only vs. Combined: Progestin-only contraceptives showed a higher risk (HR 1.21; 95% CI, 1.17-1.25) compared to combined formulations (HR 1.12; 95% CI, 1.07-1.17).

• Desogestrel (Progestin-Only): Oral desogestrel-only pills were associated with a higher risk (HR 1.18; 95% CI, 1.13-1.23).

• Etonogestrel Implant: This implant (containing the active metabolite of desogestrel) was associated with a higher risk (HR 1.22; 95% CI, 1.11-1.35).

• Levonorgestrel: Combined levonorgestrel pills (HR 1.09; 95% CI, 1.03-1.15) and the 52 mg Levonorgestrel IUS (HR 1.13; 95% CI, 1.09-1.18) showed lower relative risks compared to desogestrel products.

• Drospirenone: Combined oral drospirenone was not associated with a statistically significant increase in breast cancer risk (HR 1.04; 95% CI, 0.96-1.12).

Definitions

• Hazard Ratio (HR): A measure of how often a particular event (in this case, breast cancer) happens in one group compared to another over time. An HR of 1.24 means the risk of diagnosis at any given time was 24% higher in the contraceptive user group compared to the non-user group.

• Person-Years: A measurement combining the number of persons in the study and the amount of time they were followed. This study included over 21 million person-years.

Participants

The study included 2,095,130 adolescent girls and women aged 13 to 49 years residing in Sweden.

Assertive Critical Appraisal

Limitations & Bias (STROBE & RECORD Frameworks)

• Confounding Variables (Missing Data): A significant limitation of this study is the handling of key confounders. Data on Body Mass Index (BMI), smoking status, and age at first birth were only available for women with a history of pregnancy (via the Medical Birth Register), representing only 64% of the cohort. For nulliparous women (who may use contraceptives differently), these variables were missing. While the authors conducted sensitivity analyses treating these as time-fixed covariates, residual confounding remains a valid concern. Unmeasured confounders such as alcohol consumption, physical activity, and family history were not available in the registers.

• Detection Bias: There is a risk of detection bias if women seeking contraception have more frequent interactions with the healthcare system, leading to higher rates of breast cancer diagnosis (surveillance bias). The authors argue against this by noting that the risk persisted in women over 40 (when screening begins), suggesting the finding is not solely due to differential screening access.

• Exposure Misclassification: The study defines exposure based on dispensed prescriptions (redeemed), not actual consumption. This “intention-to-treat” style approach in observational data typically biases results toward the null (underestimating risk) if women bought the pills but didn’t take them.

Reporting Quality Assessment (STROBE & RECORD)

• Strengths: The study adheres well to STROBE guidelines. The authors utilized a “counting process” format for the Cox regression, which appropriately handles time-varying exposures (switching between contraceptives) and avoids immortal time bias (incorrectly attributing untreated time to the treatment group).

• Data Source Transparency (RECORD): As a study utilizing Real-World Data (RWD) from Swedish registers, the reporting is strong. The linkage between the Prescribed Drug Register and the Cancer Register is clearly described. However, the authors note that detailed contraceptive history prior to July 2005 was unavailable, potentially misclassifying some “early users” as “new users,” which could dilute the observed effects of duration.

Applicability

The findings are highly applicable to Western populations with similar contraceptive prescribing patterns. The study’s strength lies in its analysis of progestin-only products, which are widely used in Sweden (more so than in many other countries). This makes the data particularly valuable for clinicians in regions where progestin-only pills (POPs) and long-acting reversible contraceptives (LARCs) are becoming the standard of care.

Study Details

• Research Objective: To estimate the difference in breast cancer risk associated with different hormonal contraceptive formulations, specifically comparing combined versus progestin-only methods and different progestin types (e.g., desogestrel vs. levonorgestrel).

• Study Design: Nationwide, population-based prospective cohort study. Key elements included linked national registers, time-dependent Cox regression analysis, and adjustment for multiple covariates including education and parity.

• Setting: Sweden.

• Dates: Follow-up from January 1, 2006, to December 31, 2019.

• Population: All girls and women aged 13-49.

• Exclusions: Prior history of breast, ovarian, cervical, or uterine cancer; bilateral oophorectomy; or infertility treatment.

Bibliographic Data

• Title: Hormonal Contraceptive Formulations and Breast Cancer Risk in Adolescents and Premenopausal Women

• Authors: Hadizadeh F, Koteci A, Karlsson T, Ek WE, Johansson Å.

• Journal: JAMA Oncology

• Year: 2025

• DOI: 10.1001/jamaoncol.2025.4480

Disclaimer: This AI-generated analysis is for informational and research purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of a qualified health provider with any questions you may have regarding a medical condition.