Relationship between procedures and health insurance for critically ill patients with Pneumocystis carinii pneumonia

The objective of the present study was to assess the association between type of health insurance coverage and use of diagnostic tests and therapies among patients with AIDS-related Pneumocystis carinii pneumonia (PCP). Fifty- six private, public, and community hospitals in Chicago, Los Angeles, and Miami were selected for the study, and the charts of 890 patients with empirically treated or cytologically confirmed PCP, hospitalized during 1987 to 1990 were retrospectively reviewed. Patients were classified by insurance status: self-pay (n = 56), Medicaid (n = 254), or private insurance, including health maintenance organizations and Medicare (n = 580). Primary outcomes were the use and timing of bronchoscopy, the type and timing of PCP therapy, and in-hospital mortality. The results indicate that Medicaid patients were less likely than privately insured patients to undergo bronchoscopy (relative odds = 0.61; 95% CI = 0.40, 0.93; p = 0.02) or to have their diagnosis of PCP confirmed (relative odds = 0.51; 95% CI = 0.33, 0.77), after adjusting for patient, severity of illness, and hospital characteristics. Medicaid patients were approximately three-fourths more likely than privately insured patients (relative odds = 1.73; 95% CI = 1.01,2.96; p = 0.04) to die in-hospital, after adjusting for patient, severity of illness, and hospital characteristics. However, with further adjustment for confirmation of PCP, Medicaid patients no longer had a significantly higher likelihood of dying in-hospital. We conclude that Medicaid patients are less likely to receive diagnostic bronchoscopy than privately insured or self-insured patients, more likely to be empirically treated for PCP, and more likely to die in-hospital. Higher mortality among Medicaid patients may result from failure to diagnose and treat alternative pathogens, a consequence of worse access to invasive diagnostic tests.