ABSTRACT
BACKGROUND: Many women have discontinued hormone replacement therapy (HRT) in view of recent findings. The goal of this study was to determine if HRT discontinuation is associated with changes in health-related quality of life (HRQOL) in elderly women. METHODS: We studied women enrolled in Pennsylvania's Pharmaceutical Assistance Contract for the Elderly (PACE) program, linking prescription claims with data from a longitudinal mail survey. HRQOL measures included the number of days out of the last 30 that physical health was not good and analogous measures for mental health, pain, and interference with activities, as well as a composite "healthy days" measure developed by CDC. Longitudinal analyses focused on 2,357 women who completed surveys in both 2002 and 2003, and who used HRT at baseline (mean age = 75.5, range = 65-102). Propensity scores were used to match HRT continuers and discontinuers according to HRT type, demographics, and baseline HRQOL. Analysis of covariance was used to compare HRQOL change in continuers and discontinuers. RESULTS: Between 2002 and 2003, 43% of HRT users discontinued therapy. Analysis of covariance to examine HRQOL change revealed complex interactions with age. Discontinuers aged 65-74 reported greater increases in days in which mental health was not good (p < .05), fewer "healthy days" (p < .05), more days in which health interfered with activities (p < .01), and more days with pain (p < .01). Among women aged 75-84, HRT discontinuers reported more days in which physical health was not good (p < .01); no other significant effects were observed in this group. Relative to HRT continuers, discontinuers aged 85 and older experienced apparent HRQOL improvements following cessation, with fewer days in which physical health was not good (p < .01), fewer days of poor mental health (p < .05), and more "healthy days" (p < .01). CONCLUSIONS: These results suggest that there are substantial age differences in response to HRT discontinuation. While women aged 65-74 experienced apparent declines in HRQOL following HRT cessation, women aged 85 and older experienced relative improvements. The HRQOL declines observed among younger women underscore the importance of communication between clinicians and patients throughout the discontinuation process. These results also demonstrate the value of HRQOL surveillance as a component of health program administration.
ABSTRACT
BACKGROUND: Recent Food and Drug Administration-mandated and company-initiated withdrawals of drug products from the marketplace have had an impact on utilization in related drug classes. OBJECTIVE: To investigate the impact of withdrawal of the prokinetic agent cisapride (Propulsid) on utilization of other gastrointestinal (GI) agents. STUDY DESIGN: A longitudinal, retrospective study using electronic prescription data from a state-funded geriatric prescription benefit program. PATIENTS AND METHODS: Prescription claims for 2644 patients using cisapride between January 10, 2000, and October 1, 2000, were analyzed with respect to points in time at which (1) prospective drug utilization review edits were implemented denying reimbursement of cisapride because of drug interactions, (2) the manufacturer announced its intent to cease production, and (3) the agent was withdrawn from the market. Prevalence of use, claims volume, and expenditures were compared for cisapride, proton pump inhibitors, histamine-2 receptor antagonists, and the prokinetic agent metoclopramide during these periods. RESULTS: Use of cisapride decreased precipitately even before implementation of a "medical exception only" reimbursement policy. After the change in policy, metoclopramide use increased, although this increase was not proportional to cisapride's decline. Although total GI expenditures declined within the cisapride cohort, this change had little impact on overall program GI expenditures. CONCLUSIONS: The loss of cisapride did not significantly affect program-wide costs for GI drugs. However, the withdrawal of cisapride appears to have resulted in increased use of metoclopramide, a medication with a more serious adverse effect profile than cisapride. Further study is needed to evaluate the long-term clinical impact of such therapy changes.
