Health utilities associated with hemoglobin levels and blood loss in postmenopausal women: the Women's Health Initiative.

OBJECTIVES: The purpose of our study was to use health-related quality of life data from the Women's Health Initiative to calculate health-related utility weights and examine differences in these health utility weights across different hemoglobin (Hgb) levels. These utility weights could then be used in future cost-effectiveness studies. METHODS: Health utility weights were measured by the Short Form-6D (SF-6D), a health utility index derived from the Short Form Medical Outcomes questionnaire. Adjusted least square means were calculated for each level of Hgb at baseline and in longitudinal regression analysis the relationship between change in Hgb and change in the SF-6D was examined. Both baseline and longitudinal analyses were performed for all postmenopausal women and separately for those with self-reported heart failure, cancer, and osteoarthritis. RESULTS: Women with Hgb in the anemic range had lower health utility weights than those with higher Hgb levels. Longitudinally, a loss of of 2 g/dl Hgb or more was associated with a statistically significant and clinically meaningfully decline in SF-6D in all participants and also in the group of participants with cancer and osteoarthritis, but not in those with heart failure. CONCLUSIONS: Lower levels of Hgb and a loss of Hgb are associated with a statistically significant and clinically meaningful decrement in health utility in all postmenopausal women we studied and also in those with chronic conditions.

The impact of publicly funded managed care on adolescent substance abuse treatment outcomes.

This study compares the 12-month changes in substance use following admission to substance abuse treatment in Massachusetts between adolescents enrolled in Medicaid managed care and other publicly funded adolescents. Two hundred and fifty-five adolescents were interviewed as they entered substance abuse treatment and at 6 and 12 month follow-ups. Medicaid enrollment data were used to determine the managed care enrollment status. One hundred forty two (56%) adolescents were in the managed care group and 113 (44%) comprise the comparison group. Substance use outcomes include a count of negative consequences of substance use, days of alcohol use, days of cannabis use, and days of any substance use in the previous 30 days. Repeated measures analysis of covariance (ANCOVA) was used to assess change with time of measurement and managed care status as main effects and the interaction of time and managed care included to measure differences between the groups over time. Although several changes across time were detected for all four outcomes, we found no evidence of an impact of managed care for any of the outcomes. The results of our study do not support the fears that behavioral managed care, by imposing limits on services provided, would substantially reduce the effectiveness of substance abuse treatment for adolescents. At the same time, the results do not support those who believe that the continuity of care and improved resource utilization claimed for managed care would improve outcomes.

Variation in cost of informal caregiving and formal-service use for people with Alzheimer's disease.

This study used a geographically diverse sample to estimate the total cost of informal care and formal services for community-residing Alzheimer's disease (AD) care recipients. Baseline data were used for 1200 family caregivers from the Resources for Enhancing Alzheimer's Caregiver Health (REACH) study, a multisite intervention trial. The replacement-wage-rate approach estimated informal cost. Formal services were assigned a cost based on secondary sources. Annual cost per care recipient amounted to 23,436 dollars for informal care and 8064 dollars for formal services. Variation in informal cost was almost entirely due to instrumental activities of daily living (IADLs) assistance. Cross-site differences in cost persisted after controlling for caregiver and care-recipient characteristics. Geographic variation may suggest regional preferences or ethnic/cultural values. Further study is needed to determine whether this reflects differences in access or availability or how including a control group for care recipients with nondementia diagnoses might have affected these findings.