Provider practice patterns in nursing home-acquired pneumonia.

Objective: To describe provider practice patterns in the diagnosis and treatment of Nursing Home-Acquired Pneumonia (NHAP) and to document associations with cure, mortality, and transfer.

Design: A retrospective cohort.

Setting: Six nursing homes in Seattle, Washington.

Participants: A total of 94 patients, 65 years of age or older (mean 83 +/- SD 9), in whom the diagnosis of pneumonia was documented in the nursing home medical record between July 1, 1994, and June 6, 1995.

Measurements: Multivariate logistic regression was used to assess the relationship between descriptive, diagnostic, or therapeutic measures and three outcomes, cure, 30-day mortality, and hospital transfer.

Results: Ninety-four episodes of pneumonia were identified. Allowing for more than one outcome per patient, there were 71 (75.5%) cures, 16 (17%) deaths, and nine (9.6%) transfers. Eighty-five percent of patients identified as having NHAP by their providers had chest X-rays (CXRs), and 69% had physical examinations. Sputum examination was ordered in 5%, blood cultures in 6%, and white blood cell counts in 33% of patients. In multivariate analysis, patients with functional decline were more likely to die (Odds Ratio (OR) 36.5 (95% CI 6.1, 220)). Cognitive decline was a risk factor for mortality (OR 6.8 l (CI 1.8, 26)) and transfer (OR 7.5 (CI 1.2, 46)). Those patients receiving only oral antibiotics (OR 3.2 (CI 1.1, 9.7)) were more likely to be cured. Length of therapy >1 week was also associated with cure (OR 2.9 (CI 1.0, 8.6)). Providers with Certificate of Added Qualifications (CAQ) in Geriatric Medicine were more likely to achieve cure (OR 3.1 (CI 1.0, 9.0)).

Conclusions: Most patients with NHAP had diagnostic CXRs and physical examinations. In multivariate analysis, death was more likely to occur in patients with cognitive or functional decline. Cure was associated with the use of oral antibiotics alone and with care by providers with CAQ in Geriatric Medicine.