Diabetes and disenrollment in a health maintenance organization setting - A 4-year longitudinal study with a matched cohort

OBJECTIVE - The increasing enrollment of Medicare beneficiaries in health m aintenance organizations (HMOs) in recent years has caused concern about wh ether HMOs and their providers have created an unfavorable environment for members who are chronically ill. This study was designed to examine whether there are any differences in disenrollment rates among enrollees with diab etes and enrollees without diabetes. RESEARCH DESIGN AND METHODS- This was a 4-year longitudinal follow-up study with a matched cohort. Medicare beneficiaries (aged greater than or equal to 65 years) with diabetes identified through pharmacy records in 1994 were matched with a comparison group according to age, sex, comorbidities, and type of provider groups in an HMO in California. RESULTS - The overall distribution of the characteristics of members in the diabetic and matched nondiabetic group is almost identical. The matched-pa ir chi(2) tests indicated that there were no statistical differences in dis enrollment rates between diabetic and nondiabetic members during all three follow-up periods (P = 0.16-0.85). CONCLUSIONS - We found that the HMO members with diabetes did not disenroll from the HMO at a higher rate than those without diabetes. The findings sh ould alleviate some of the concern that HMOs and their contracted providers have created an unattractive environment for members who have chronic dise ases such as diabetes.