Epidemiology and Pathogen Characteristics of Pneumonia Types in the US Hospital Setting

Background: Understanding the epidemiology of pneumonia types in hospitals (PNA) is imperative for appropriate management. Based on the setting, PNA is classified into community acquired (CAP) or nosocomial (NP); NP is further classified into healthcare-associated (HCAP), hospital-acquired (HAP), and ventilator-associated (VAP) pneumonia. Few studies have elucidated the epidemiology of all PNA types using a single data-source, and largely restricted to culture-positive cases.

Methods: A retrospective analysis using data from a large US hospital-based database (Premier) was conducted. Eligible patients were those admitted as inpatient between January 1, 2008 and March 31, 2015 with a primary or secondary diagnosis of pneumonia (ICD-9 CM: 481.x-483.x, 485.x- 486.x, 997.31). Patients were then grouped into PNA types based on a pre-defined algorithm. Distributions of PNA types along with associated pathogens were the outcomes of interest.

Results: A total of 2,810,674 patients were deemed eligible, of whom 79.5%, 10%, 6.4%, 1.3%, and 2.6% patients were categorized as CAP, HCAP, HAP, VAP, and unclassified, respectively. Among CAP patients, 18.7% were intubated (iCAP). Among a subset of eligible patients with microbiology data (18% of PNA), positive bacterial cultures were available for 57% of CAP, 77% of iCAP, 65% of NP, 63% of HCAP, 62% of HAP, and 92% of VAP patients. Among VAP, S. aureus (40%) and P.aeruginosa (24%) were the most common gram-positive and negative pathogens, respectively, followed by E. coli and Klebsiella sp. (17% each). These pathogens were more predominant in VAP than other PNA types. iCAP patients showed higher proportion of S.aureus (27 vs. 14%) and P.aeruginosa (12 vs. 5%) than non-intubated CAP patients.

Conclusion: In our study, CAP identified based on antibiotic use within 2 days of admission, could potentially be an over-estimate as real-world use of antibiotics would be influenced by other factors (e.g. infections, surgeries). To our knowledge, this is the first study characterizing the serious iCAP population, who may require a different treatment management than traditional CAP patients.