Treatment patterns for early pregnancy failure in Michigan.

AIMS We describe current treatment patterns for early pregnancy failure (EPF) among women enrolled in two Michigan health plans. METHODS We conducted a retrospective review of EPF treatment among Michigan Medicaid enrollees between January 1, 2001, and December 31, 2004, and enrollees of a university-affiliated health plan between January 1, 2001, and December 31, 2005. Episodes were identified by the presence of a diagnostic code for EPF. Surgical treatment was distinguished from nonsurgical management using procedure codes. Facility charges, procedure, and place of service codes were used to determine whether a procedure was done in an office as opposed to an operating room. Cases without a claim for surgical uterine evacuation were examined for a misoprostol pharmacy claim and, if present, were classified as medical management. Cases without a procedure or pharmacy claim were classified as expectant management. RESULTS Respectively, we identified 21,311 and 1,493 episodes of EPF in the Medicaid and university-affiliated health plan databases, respectively. Women enrolled in Medicaid were more likely to be treated with surgery than were enrollees of the university-affiliated health plan (35.3 vs. 18.0%, respectively, p < 0.000). Among Medicaid enrollees, only 0.5% of surgical evacuations occurred in the office, but office procedures were common among enrollees of the university-affiliated health plan (30.5%, p < 0.000). The proportion of cases managed with misoprostol was <1% in both groups. Caucasian race and age were both associated with having a surgical uterine evacuation (p < 0.001). CONCLUSIONS EPF is primarily being treated with expectant management or surgical evacuation in an operating room and may not reflect evidence-based practices or patient preferences.