As part of a program to evaluate the health care provided to its beneficiaries, HCFA requires Medicare managed care plans to participate in the Medicare Health Outcomes Survey (HOS). One thousand randomly-sampled beneficiaries per plan are selected for the HOS each year and are followed for two years, using the SF-36 Health Survey and additional case-mix variables to assess physical and mental health. Baseline data from the first (1998) cohort (n=167,069 from 268 plans) and the second (1999) cohort (n=194,615 from 283 plans) are reported here.
The overall response rate was 60.0% in 1998 and 66.6% in 1999. SF-36 physical and mental summary scores could be calculated for 95.6% of respondents, and scale scores could be calculated for 97.4% to 99.6% of the eight SF-36 scales (1998 data). Tests of scaling assumptions consistently were passed. Internal consistency reliability (Cronbach's alpha) ranged from 0.83 to 0.93 (median=0.88) across scales. Scales generally demonstrated hypothesized relationships with physical and mental health factors. Clinically meaningful groups that could be ordered in terms of health status scores were defined; as in previous studies, SF-36 scores for these groups conformed to hypotheses. Some variation in psychometric results among subgroups (e.g., age, education) was noted.
From these results, we conclude that the HOS is valid for use in assessing health status and health outcomes in the Medicare population. We also comment on the large differences observed across plans at baseline and the implications of these differences for the analysis and interpretation of health outcomes.
Learning Objectives: At the conclusion of this session, the participant will be able to discuss the reliability and validity of a questionnaire used to measure health status in a major HCFA initiative to evalute the quality of care provided to Medicare managed care beneficiaries
Keywords: Medicare, Managed Care
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.