Medical care utilization among older HMO members with and without hypertension

To compare hospital utilization, health status, and sociodemographic characteristics of older persons with and without hypertension and to examine the nature of the association of hypertension with medical care utilization.

A retrospective, nonexperimental study.

A large health maintenance organization (HMO), located in the Portland, Oregon and Vancouver, Washington areas, that provides comprehensive, prepaid benefits to its members.

Approximately 4200 older (> or = 65 years) members enrolled in the Social HMO Demonstration Project.

The dependent variables were (1) whether a member was hospitalized (0 = no; 1 = yes) and (2) the number of hospitalizations during the 12-month period before the return of the Health Status Form, a mail questionnaire. The predictor variables were high blood pressure and other medical conditions, health status, and sociodemographic characteristics. Patients classified as hypertensive were those who self-reported high blood pressure and who received two or more dispensings for antihypertensive medications.

Older hypertensives were more likely than older nonhypertensives to have been hospitalized (odds ratio (OR) = 1.14 (95% CI 1.01, 1.27)), self-report their health as only good or fair (P < .001), self-report IADL functional limitations (P < .001), and to be younger than nonhypertensives (P < .001). Utilization differences were similar after controlling for patients' age and gender. Factors independently associated with an increased likelihood of hospitalization were male gender (OR = 1.14 (1.00,1.31)), age (OR = 1.35 (1.00,1.81)), poor (OR = 1.32 (1.02,1.72)) or fair (OR = 1.20 (1.01,1.44)) self-rated health status, functional limitations in activities of daily living (OR = 1.74 (1.48,2.03), and self-reported heart trouble (OR = 1.50 (1.34,1.68)), stroke (OR = 1.37 (1.15,1.64)), or cancer (OR = 2.89 (1.72,4.84)). Factors independently associated with a decreased likelihood of hospitalization were excellent self-reported health (OR = 0.65 (0.47,0.90)), no physical impairments (OR = 0.59 (0.48,0.72)), and no circulation problems (OR = 0.85 (0.74,0.98)).

Older hypertensive patients had poorer self-reported health, more functional limitations, used more hospital services, and were not as old as nonhypertensive older patients. To the extent that general health is improved by treatment of hypertension, appropriate treatment of high blood pressure may not only improve health, it may also reduce medical care utilization. Our findings offer further support for hypertension prevention, detection, and treatment programs designed to reduce or minimize hypertension's later medical complications and to reduce future utilization.