Epidemiology, antibiotic therapy, and clinical outcomes in health care-associated pneumonia: a UK cohort study

Abstract

Background: The recently introduced concept of health care-associated pneumonia (HCAP), referring to patients with frequent healthcare contacts and at higher risk of contracting resistant pathogens, is controversial.

Methods: This prospective observational study recorded the clinical features, microbiology, and outcomes in a UK cohort of hospitalized patients with pneumonia. The primary outcome was 30-day mortality. Logistic regression was used to adjust for confounders when determining the impact of HCAP on clinical outcomes.

Results: A total of 20.5% of patients met the HCAP criteria. HCAP patients were older than patients with community-acquired pneumonia (CAP) (median 76 y, IQR 65-83 vs 65 y, IQR 48-77; P < .0001) and more frequently had major comorbidities (62.1% vs 45.2%; P < .0001). Patients with HCAP had higher initial severity compared to CAP patients (Pneumonia Severity Index, mean 3.7 [SD 1.1] vs mean 3.1 [SD 1.3]; P < .0001) but also worse functional status using the Eastern Cooperative Oncology Group scale (mean 2.4 [SD 1.44] vs mean 1.4 [SD 1.13]; P < .0001) and more frequently had treatment restrictions such as do not resuscitate orders (59.9% vs 29.8%; P < .0001). Consequently mortality was increased (odds ratio [OR] 2.15 [1.44-3.22]; P = .002) in HCAP patients on univariate analysis. Multivariate analysis suggested this relationship was primarily due to confounders rather than a higher frequency of treatment failure due to resistant organisms (adjusted OR .97 [.61-1.55]; P = .9). The frequencies of Pseudomonas aeruginosa, methicillin-resistant Staphylococcus aureus, and Gram-negative Enterobacteriaceae were low in both cohorts.

Conclusions: HCAP is common in the United Kingdom and is associated with a high mortality. This increased mortality was primarily related to underlying patient-related factors rather than the presence of antibiotic-resistant pathogens. This study did not establish a clear indication to change prescribing practices in a UK cohort.