April 2, 1997

Extending Health Maintenance Organization Insurance to the UninsuredA Controlled Measure of Health Care Utilization

Author Affiliations

From the Departments of Pediatrics (Dr Bograd), Clinical Profiles (Ms Ritzwoller), Preventive Medicine (Dr Calonge), and Community Medicine (Mss Shields and Hanrahan), Kaiser Permanente of Colorado, Denver.

JAMA. 1997;277(13):1067-1072. doi:10.1001/jama.1997.03540370057037

Objective.  —To investigate the utilization of health care services of previously uninsured low-income patients after becoming insured by a health maintenance organization (HMO).

Design.  —Retrospective study of utilization in a previously uninsured study group compared with an age- and sex-matched randomly selected control group of commercial HMO enrollees.

Setting.  —Group model HMO.

Patients.  —A study group of 346 previously uninsured low-income patients and 382 controls.

Measures.  —Outpatient visits for primary and specialty care, outpatient pharmacy, laboratory, and radiology use, and inpatient admissions and hospital days over a 2-year period. Self-reported health status measures were obtained to control for differences in health status.

Principal Findings.  —There were no differences between the study and control groups in hospital admissions, hospital days, and measures of outpatient laboratory, pharmacy, and radiology use. The odds of having an outpatient visit per patient per month was 30% higher for the study group. Approximately half the increase in the odds ratio for outpatient visits was related to the worse self-perceived health status of the study group. While both groups utilized more services in the early phase of their enrollment, the intensity of this start-up effect was similar for both groups.

Conclusions.  —Compared with a commercial group of the same age and sex, the patterns of utilization were similar and the financial costs of care were only moderately more for a previously uninsured group provided with comprehensive HMO insurance. With the growth of managed care, these data should be beneficial in the development of health care programs for the growing number of uninsured Americans.