Predictors for length of hospital stay in patients with community-acquired Pneumonia: Results from a Swiss Multicenter study
1 Department of Internal Medicine, Division of Endocrinology, Diabetes and Metabolism, University Hospital Basel, Basel, Switzerland
2 Medical University Clinic, Kantonsspital Liestal, Liestal, Switzerland
3 Medical University Clinic, Kantonsspital Aarau, Aarau, Switzerland
4 Harvard School of Public Health, Boston, USA
Citation and License
BMC Pulmonary Medicine 2012, 12:21 doi:10.1186/1471-2466-12-21Published: 20 May 2012
Length of hospital stay (LOS) in patients with community-acquired pneumonia (CAP) is variable and directly related to medical costs. Accurate estimation of LOS on admission and during follow-up may result in earlier and more efficient discharge strategies.
This is a prospective multicenter study including patients in emergency departments of 6 tertiary care hospitals in Switzerland between October 2006 and March 2008. Medical history, clinical data at presentation and health care insurance class were collected. We calculated univariate and multivariate cox regression models to assess the association of different characteristics with LOS. In a split sample analysis, we created two LOS prediction rules, first including only admission data, and second including also additional inpatient information.
The mean LOS in the 875 included CAP patients was 9.8 days (95%CI 9.3-10.4). Older age, respiratory rate >20 pm, nursing home residence, chronic pulmonary disease, diabetes, multilobar CAP and the pneumonia severity index class were independently associated with longer LOS in the admission prediction model. When also considering follow-up information, low albumin levels, ICU transfer and development of CAP-associated complications were additional independent risk factors for prolonged LOS. Both weighted clinical prediction rules based on these factors showed a high separation of patients in Kaplan Meier Curves (p logrank <0.001 and <0.001) and a good calibration when comparing predicted and observed results.
Within this study we identified different baseline and follow-up characteristics to be strong and independent predictors for LOS. If validated in future studies, these factors may help to optimize discharge strategies and thus shorten LOS in CAP patients.