Hormonal Contraceptive Use and Association with Thromboembolism in Women with Sickle Cell Disease

Introduction Women with sickle cell disease (SCD) require broad access to highly effective contraceptive methods due to high rates of unplanned pregnancy and adverse maternal and fetal outcomes. Combined (estrogen-containing) hormonal contraception carries a known risk of thromboembolism (TE) in the general population, and women with SCD have higher baseline risk of TE, thus use of these agents in women with SCD may further increase that risk. The paucity of safety data on hormonal contraceptive use in women with SCD limits the ability of providers and patients to make informed decisions.

Objectives This study aims to use population-level administrative claims data of publicly insured women with SCD in the United States (US) to assess patterns of hormonal contraception use and evaluate risk of TE during the first year of use. We hypothesized that TE rates are higher in women prescribed combined hormonal contraceptives (CHC) as compared to women prescribed progestin-only contraceptives (POC).

Methods We obtained data using the Centers for Medicare and Medicaid Services Analytic eXtract (CMS MAX) from 2006-2018. Study population included females of reproductive age (12-44 years) with >3 ICD-9 or ICD-10 codes consistent with SCD. Women had to have ≥12 months of continuous enrollment with at least 6 months of data as 'washout period' prior to enrollment period. A new prescription for hormonal contraception was identified using national drug codes in pharmaceutical claims or ICD-9 or 10 procedure codes. CHC included the combined oral contraceptive pill (OCP), transdermal patch, and vaginal ring. POC included the progestin-only pill, depot medroxyprogesterone acetate (DMPA) injection, intrauterine device (IUD) and subdermal implant. TE was defined as the presence of ≥2 ICD-9 or ICD-10 codes for venous TE (deep vein thrombosis (DVT) or pulmonary embolism (PE)) or arterial TE (stroke or myocardial infarction (MI)). Arterial TE was also required to have an associated inpatient admission. Women were included in the analysis for 12 months after contraception initiation, or censored in the event of either contraceptive switch, pregnancy or abortion, or TE occurrence. Data were summarized descriptively and TE incidence rates were calculated as events per 1,000 person-years.

Results We identified a total of 41,012 reproductive-aged women with SCD enrolled in CMS MAX, of whom 22.6% (n=9,261) had received ≥1 new prescription for hormonal contraception. Among women with new contraceptive use, 58.5% (n=5,420) initiated CHC and 41.5% (n=3,841) initiated POC. Median age at the start of contraception was 22.3 years (range: 12.0-43.9) and median age at the start of CHC and POC were 22.1 and 22.7 years respectively. Among CHC users, 84.6% (n=4,584) of women were prescribed the combined OCP, 7.7% (n=419) the transdermal patch, and 7.7% (n=417) the vaginal ring. For POC users, DMPA was prescribed to 72.3% (n=2,777), progestin-only pills to 21.2% (n=816), IUD to 4.2% (n=163) and subdermal implant to 2.2% (n=85) of women.

A total of 141 (1.5%) new contraceptive users with SCD were identified to have a TE during the follow-up period. The median time from contraception initiation to TE diagnosis was 183 days (IQR: 82-244). The most common TE event was PE (46.1%) followed by DVT (41.8%), stroke (6.4%), and MI (2.1%); 3.5% of women had multiple TE events. TE rate was 16.4 events per 1,000 person-years in CHC users, and the TE event rate was 17.0 events per 1,000 person-years in POC users (p=0.84). Rates of TE did not differ between the two most commonly used methods in each group, the combined OCP (15.5 events per 1,000 person-years ) and DMPA (17.7 events per 1,000 person-years) (p=0.48).

Conclusion Our work provides population-level data regarding contraceptive practices for women with SCD in the US. We found that CHC was more frequently prescribed than POC, and identified low rates of long-acting reversible contraception use. The most frequently prescribed agents were the combined OCP and the DMPA injection. TE rates after initiation of the combined OCP and DMPA did not differ. Although limited by the retrospective study design and use of administrative claims data, this study suggests a similar safety profile for CHC and POC in women with SCD with respect to risk of TE.

o'Brien:Pfizer-BMF alliance: Other: Dr O'Brien's institution receives salary support for her role as study chair in a Children's Oncology Group study funded by the BMS-Pfizer alliance..