HMO membership and patient age and the use of specialty care for hospitalized patients with acute stroke - The Minnesota stroke survey

BACKGROUND. The number of older patients enrolling in health maintenance or ganizations (HMOs) is increasing. Concerns have been raised that older pati ents may be targeted by HMOs for more stringent cost-containment mechanisms , including reduced access to expensive specialty care. OBJECTIVES. We investigated the relationship between membership in an HMO a nd the decision to consult with a neurologist or admit to a neurology ward far patients hospitalized with acute stroke. We then compared 1-year mortal ity of patients who received neurology care to the 1-year mortality of thos e who did not receive neurology care. DESIGN. Retrospective medical record review. SUBJECTS, A sample of hospitalized acute stroke patients (age range, 30-79 years) who were discharged from Minneapolis-St. Paul metropolitan hospitals with a diagnosis code of acute cerebrovascular disease from 1991 to 1993. MEASURES. Trained nurses abstracted the medical records. Stroke events (n = 2,320) were validated using clinical criteria and neuroimaging reports. Mo rtality data were obtained from the Minnesota Death Index. RESULTS. Among patients enrolled in HMOs, 30% of validated stroke patients did not receive neurology care in comparison with 19% of patients not enrol led in HMOs. After adjusting for patient mix and hospital characteristics, the odds of receiving neurology care were half as great for patients enroll ed in HMOs as compared with patients not enrolled in HMOs (odds ratio [OR] = 0.52, 95% confidence int ental [CI] 0.36-0.74). The association of member ship in HMOs with lower use of neurology care was concentrated in older pat ients. Within each age group, the odds ratios and 95% CI of receiving neuro logy care for patients enrolled in HMOs versus patients not enrolled in HMO s were: < 55 years (1.06, 0.42-2.67), 55 to 64 years (0.54, 0.34-0.87), 65 to 74 years (0.51, 0.36-0.71), and >75 years (0.40, 0.24-0.68). Using Cox r egression, 30-day mortality did not differ between patients who received ne urology care and those who did not. Among 30-day survivors, the mortality h azards ratio (HR) during the next II months for patients who received neuro logy care was 71% of the hazard for patients who aid not receive neurology care (HR = 0.71, 95% CI = 0.55-0.91). CONCLUSIONS. These data suggest that membership in an HMO was associated wi th reduced access to neurology care for older patients with acute stroke an d that patients who received neurology care had a lower risk of death durin g the year after their stroke. It remains to be determined if these differe nces in outcome are caused by true differences in stroke management or by u nmeasured characteristics.