The most frequent infectious cause of death in the United States is pneumonia. The Health Care Financing Administration (HCFA) is interested in assessing the quality of care in Medicare patients admitted to the hospital because of pneumonia and in determining what performance measures are associated with improved survival. Meehan and colleagues reviewed the medical records of 14,069 elderly patients hospitalized with pneumonia to evaluate these issues.

A standardized data collection system was used to collect information about four quality of care indicators in patients hospitalized because of pneumonia: (1) the interval between hospital arrival and administration of antibiotics, (2) whether blood cultures were collected before antibiotic therapy was instituted, (3) whether blood cultures were obtained within 24 hours of hospital arrival and (4) whether oxygenation status was assessed within 24 hours of hospital arrival.

Cases of pneumonia were identifed by HCFA from the Medicare National Claims History File. Patients were excluded from the study if they were under 65 years of age, had human immunodeficiency virus (HIV) infection, had recently been exposed to chemotherapy or immunosuppressive therapy or had been transferred from another acute care facility. Severity of illness variables, process of care variables (such as the rapidity of initial administration of antibiotic therapy, the time of blood culture collection and oxygenation assessment) and outcomes of care were evaluated.

Approximately 25 percent of the patients were transferred to the hospital from skilled or intermediate nursing care facilities. Most of the study subjects (58.2 percent) had at least one other illness. Antibiotic therapy was initiated within eight hours of admission in 75.5 percent of the patients.

The risk of death at 30 days increased as the interval between hospital admission and initiation of antibiotic therapy increased beyond eight hours. Similarly, the risk of death at 30 days was significantly lower in those patients in whom blood cultures were obtained within 24 hours of admission. Whether blood cultures were collected before or after antibiotic administration had no significant effect on 30-day mortality.

The authors conclude that the findings of this retrospective review encourage speedy antibiotic administration and blood culture collection in elderly patients with pneumonia. Wide variation was found in the interval between hospital admission and implementation of these two measures, suggesting an area for improvement in the hospital care of elderly patients with pneumonia.