Predictors of resource utilization for hospitalized patients with Pneumocystis carinii pneumonia (PCP): a summary of effects from the multi-city study of quality of PCP care.

To determine whether patient and hospital characteristics were significantly associated with variations in Pneumocystis carinii (PCP) care and outcomes, we analyzed the use of diagnostic tests, intensive care units (ICUs), anti-PCP medications for persons hospitalized with human immunodeficiency virus (HIV)-related PCP, and hospital discharge status. We conducted retrospective chart reviews of a cohort of 2,174 patients with PCP hospitalized in 1987-1990. Outcomes included process of care for PCP and in-hospital mortality rates. Persons with PCP who were more severely ill at admission were more likely to have early medical care, to receive care in an intensive care unit, and to die in hospital. After we adjusted for differences in this severity of illness, we noted that Medicaid patients, injection drug users (IDUs), and patients treated at VA or county hospitals were significantly less likely than others to have diagnostic bronchoscopies and that persons covered by Medicaid, with a previous diagnosis of acquired immunodeficiency syndrome (AIDS), who did not receive prior zidovudine (AZT) or who received care in a VA hospital had the highest chances of in-hospital death. Insurance and risk group characteristics, severity of illness, and hospital characteristics appear to be the most important determinants of the intensity and timing of medical care and outcomes among patients hospitalized with PCP.