Coverage Gaps and Contraceptive Use Among Medicare Enrollees With Disabilities

Abstract

Importance: Medicare is the primary health insurance payer for 1.5 million reproductive-aged women with disabilities, yet it is the only major form of US health insurance that is not required to cover contraceptives for pregnancy prevention.

Objective: To evaluate whether Medicare's contraceptive coverage gaps were associated with reduced use of contraceptives by enrollees with disabilities.

Design, setting, and participants: In this national, cross-sectional study, traditional Medicare (TM), Medicare Advantage (MA), and Medicaid claims from female enrollees aged 20 to 49 years receiving Social Security Disability Insurance or Supplemental Security Income from January 1, 2016, to December 31, 2020, were linked. The propensity score-weighted probability of contraceptive use by public insurance type was estimated, then the association between gaining contraceptive coverage through a transition from Medicare to dual Medicare-Medicaid enrollment and contraceptive use was evaluated using a staggered-entry difference-in-differences design. Data were analyzed from December 3, 2024, to April 5, 2025.

Exposures: Public insurance enrollment in TM, MA, dual TM-Medicaid, dual MA-Medicaid, or Medicaid.

Main outcomes and measures: Monthly use of permanent contraceptives, long-acting reversible contraceptives (intrauterine device and implant), and short-acting contraceptives (injectable and oral contraceptives, patch, and ring).

Results: A total of 51 501 303 monthly observations from 1 606 129 women were included in the analysis. Mean (SD) age was 35.93 (8.58) years; 1.8% of monthly observations were from Asian women, 30.7% from Black women, 13.0% from Hispanic women, 52.6% from White women, and 1.9% from multiracial women or women identifying as another race and ethnicity not reported on previously. Those enrolled in TM and MA were more often older and non-Hispanic White compared with those dual enrolled or enrolled in Medicaid. The estimated monthly probability of use of any contraceptive method was lowest among TM (4.9%; 95% CI, 4.9%-4.9%) and MA (6.6%; 95% CI, 6.5%-6.6%) enrollees, followed by Medicaid (11.0%; 95% CI, 11.0%-11.0%), dual MA-Medicaid (11.3%; 95% CI, 11.3%-11.4%), and dual TM-Medicaid (13.1%; 95% CI, 13.0%-13.1%) enrollees. Gaining contraceptive coverage through dual enrollment was associated with an increase of 3.9 (95% CI, 3.5-4.3) percentage points (35%) in use of any contraceptive method, with the largest increase in use of short-acting methods at 2.6 (95% CI, 2.3-3.0) percentage points (45%).

Conclusions and relevance: In this cross-sectional study of contraceptive use in the Medicare program, gaining contraceptive coverage through dual Medicare-Medicaid enrollment was associated with increased contraceptive use among disabled Medicare enrollees, suggesting that Medicare's coverage gaps pose a financial barrier to desired contraceptive use. Given these findings, Medicare should be required to cover all US Food and Drug Administration-approved contraceptive methods without cost-sharing. Doing so would align Medicare's coverage requirements with those of Medicaid, private insurance plans, and TRICARE.

Figure 1.. Contraceptive Use by Public Insurance Type

The estimated probabilities of contraceptive use are shown among women with disabilities enrolled in traditional Medicare (TM), Medicare Advantage (MA), dual TM-Medicaid, dual MA-Medicaid, and Medicaid. The bars represent 95% CIs. IUD indicates intrauterine device.

Figure 2.. Association of Gaining Contraceptive Coverage Through a Transition From Medicare to Dual Medicare-Medicaid Enrollment With Contraceptive Use

The percentage point changes in the estimated probabilities of any contraceptive use (A), use of a permanent method (B), use of a long-acting method (C), and use of a short-acting method (D) are shown before and after a transition from Medicare alone to dual Medicare-Medicaid enrollment. Values are from propensity score–weighted staggered difference-in-differences models with estimators from Callaway and Sant’Anna. The bars represent 95% CIs. Blue dots represent pretransition associations and orange dots represent posttransition associations.