Abstract
BACKGROUND: The Accreditation Council for Graduate Medical Education reforms in 2003 instituted an 80-hour weekly limit for resident physicians. Critics argue that these restrictions have increased handoffs among residents and the potential for a decline in patient safety. "Never events" hospital-acquired conditions (HACs) are a set of preventable events used as a quality metric in hospital safety analyses. OBJECTIVE: This analysis evaluated post-work hour reform effects on HAC incidence for US hospital inpatients, using the National Inpatient Sample. METHODS: Data were collected from 2000-2002 (pre-2003) and 2004-2006 (post-2003) time periods. HAC incidence in academic and non-academic centers was evaluated in multivariate analysis assessing for likelihood of HAC occurrence, prolonged length of stay (pLOS), and increased total charges. RESULTS: The data encompassed approximately 111 million pre-2003 and 117 million post-2003 admissions. Patients were 10% more likely to incur a HAC in the post-2003 versus pre-2003 era (odds ratio [OR] = 1.10; 95% confidence interval [CI] 1.06-1.14; P < .01). Teaching hospitals exhibited an 18% (OR = 1.18; 95% CI 1.11-1.27; P < .01) increase in HAC likelihood, with no change in nonteaching settings (OR = 1.03; 95% CI 1.00-1.06; P > .05). Patients with ≥ 1 HAC were associated with a 60% likelihood of elevated charges (OR = 1.60; 95% CI 1.50-1.72; P < .01) and 65% likelihood of pLOS (OR = 1.65; 95% CI 1.60-1.70; P < .01). CONCLUSIONS: Post-2003 era patients were associated with 10% increased likelihood of HAC, with effects noted primarily at teaching hospitals.