Methods: Consecutive patients admitted to teaching and nonteaching hospitalist services in a tertiary care hospital with a primary diagnosis of CHF or pneumonia were assessed for in-hospital mortality and length of hospital stay (LOS). Categorical variables were expressed as counts (percentages) and compared using Chi-square test. Linear regression models were used to examine the risk adjusted difference in LOS in both Pneumonia and CHF patients using Medical-Severity – Diagnosis Related Groups weights/Case Mix Index (MS-DRG Weight/CMI) as measures of illness severity and clinical complexity and further adjusting for home use of Angiotensin-Converting Enzyme (ACE)-Inhibitors and Angiotensin II Receptor Blockers (ARBs) for CHF patients. Logistic regression models were used to examine the difference in in-hospital death. P-value ≤ 0.05 was considered statistically significant.
Results: A final cohort of 384 CHF and 310 pneumonia cases were identified from September to December 2014. CHF patients managed by the non-teaching team had less in-hospital death (1.46% vs. 5.45%, P-value=0.0364. Adjusted Odds Ratio for in-hospital mortality for CHF patients managed in non-teaching vs. teaching services was: 0.196, 95% CI: 0.050-0.770, P = 0.0196 after adjusting for severity of illness; 0.248, 95% CI: 0.068-0.899, P =0.0338 after adjusting for home use of ACE-inhibitors; and 0.192, 95% CI: 0.071-0.929, P =0.0382 after adjusting for home use of ARBs). There was no difference in mortality for pneumonia patients admitted to teaching and non-teaching services. Patients with pneumonia admitted to nonteaching service had longer LOS (6.00 days (95% CI: 5.46-6.59) vs. 4.94 days (95% CI: 4.18-5.82, P = 0.0450). There were no statistically significance differences between services with respect to LOS and home use of ACE-I and ARBs for patients admitted for CHF.
Conclusions: Patients with CHF admitted to nonteaching hospitalists services had lower adjusted in-hospital mortality compared to similar cohorts admitted to teaching services. Patients with pneumonia admitted to teaching hospitalists services had similar inpatient morality but lower LOS compared to those admitted to nonteaching services.