OBJECTIVES: To evaluate risk of intrauterine device (IUD) expulsion by postpartum timing of IUD insertion.
METHODS: APEX IUD was a retrospective cohort study conducted within four health care systems with electronic health records—three Kaiser Permanente sites (Northern California, Southern California, Washington) and Regenstrief Institute (Indiana). The study included data from 326,658 women (aged ≤50 years) with an IUD insertion. Date of delivery, IUD insertion, IUD expulsion, and potential confounders were identified using structured and unstructured data. Expulsion was defined as complete (IUD in the vagina, not visible with imaging, or patient-reported) or partial (any portion of the IUD in the cervix). We calculated postpartum timing of IUD insertion and categorized it into discrete time periods: (1) 0-3 days, (2) 4 days to ≤6 weeks, (3) >6 to ≤14 weeks, (4) >14 to ≤52 weeks, and (5) >52 weeks or with no evidence of delivery. We estimated incidence rates and adjusted hazard ratios (aHRs), using propensity scores to adjust for confounding.
RESULTS: Across all sites, IUD expulsion incidence [95% confidence interval (CI)] per 1,000 person-years of follow-up were: (1) 46.5 [40.5-53.2], (2) 10.9 [9.8-12.1], (3) 9.3 [8.7-9.9], (4) 14.4 [13.2-15.6], (5) 14.9 [14.6-15.3]. The aHRs [95% CI] comparing groups 1 through 4 with group 5 were: (1) 5.34 [4.47-6.39], (2) 1.22 [1.05-1.41], (3) 1.06 [0.95-1.18], and (4) 1.43 [1.29-1.60].
CONCLUSIONS: Risk of IUD expulsion was low overall, but highest in those with IUD insertions 0-3 days postpartum. Differential rates warrant consideration in the benefit-risk evaluation and should be an important part of counseling patients.
