JOURNAL OF WOMEN’S HEALTH Volume 17, Number 4, 2008 © Mary Ann Liebert, Inc. DOI: 10.1089/jwh.2007.0407 A Retrospective Managed Care Claims Data Analysis of Medication Adherence to Vaginal Estrogen Therapy: Implications for Clinical Practice LEE P. SHULMAN, M.D., 1 DAVID J. PORTMAN, M.D., 2 WON CHAN LEE, Ph.D., 3 SANJEEV BALU, Ph.D., M.B.A., 4 ASHISH V. JOSHI, M.S., Ph.D., 5 DAVID COBDEN, M.Sc., M.P.H., 5 QIN WANG, M.A., 3 and CHRIS L. PASHOS, Ph.D. 4 ABSTRACT Objectives: This study sought to assess refill-based treatment duration and adherence to vagi- nal estrogen therapy (VET) in clinical practice and to compare treatment duration in women prescribed vaginal tablets (VT) or vaginal creams (VC) in clinical trials with that in clinical practice. Methods: Adults initiating VET between January and June 2004 were identified in 57 com- mercial health plans in the United States and followed for up to 10 months after treatment initiation. Treatment duration (Kaplan-Meier analysis) and adherence (medication possession ratio [MPR]) were examined for VC and VT cohorts and compared with a weighted average of treatment duration calculated for seven clinical trials identified in the literature. Results: Of 13,074 women (mean age 54  9.1 years) identified, patients on VT had signif- icantly longer treatment duration compared with patients on VC (149.1  101.1 days vs. 91.6  30.0 days, p  0.01). Among 5,599 patients receiving multiple prescriptions, higher treatment durations were observed for both VT-treated and VC-treated patients (198.5  82.4 days vs. 177.1  86.7 days, p  0.01) compared with 165 days observed in clinical trials. Medication ad- herence (MPR  80%) was significantly higher among VT users compared with VC users (74%  27% vs. 52%  32%], p  0.01); OR, 3.24, 95% CI 2.84-3.70). Conclusions: Duration of VET among patients receiving multiple prescriptions was longer in real-world clinical practice than in clinical trials. Newly treated patients initiating VT were more likely to continue treatment for longer periods and exhibited greater medication ad- herence than did those on VC. 569 1 Division of Reproductive Genetics, Department of Obstetrics and Gynecology, Feinberg School of Medicine, North- western University, Chicago, Illinois. 2 The Columbus Center for Women’s Health Research, Columbus, Ohio. 3 HERQuLES, Abt Associates Inc., Bethesda, Maryland. 4 HERQuLES, Abt Associates Inc., Lexington, Massachusetts. 5 Novo Nordisk Inc., Princeton, New Jersey. This study was funded by a grant from Novo Nordisk Inc., Princeton, New Jersey. This work was presented at the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 9th Annual European Congress, Copenhagen, Denmark, October 11–14, 2006, and at the North American Menopause So- ciety 17th Annual Meeting, Nashville, Tennessee, October 28–31, 2006.