Surgical Safety and Hospital Volume Across a Wide Range of Interventions

Bibliographic Data

ID9097993
AuthorsYves Eggli (0000-0002-2979-7936, University of Lausanne, corresponding author), Patricia Halfon (Institute of Social and Preventive Medicine), Danielle Meylan (University Hospital of Lausanne, corresponding author), Patrick Taffé (0000-0002-2049-6651, Institute of Social and Preventive Medicine)
Year2010
Volume48
Issue11
Pages962-971
Publication date2010-11-01
Open AccessNo
TypeARTICLE
VenueMedical Care (JOURNAL)
Journal identifiersISSN: 0025-7079 • E-ISSN: 1537-1948
PublisherOvid Technologies (Wolters Kluwer Health) (PUBLISHER)
DOI10.1097/mlr.0b013e3181eaf9f6
PMID20829722
OpenAlexW2335425975
LanguageEN
References cited40

OBJECTIVES: For certain major operations, inpatient mortality risk is lower in high-volume hospitals than those in low-volume hospitals. Extending the analysis to a broader range of interventions and outcomes is necessary before adopting policies based on minimum volume thresholds. METHODS: Using the United States 2004 Nationwide Inpatient Sample, we assessed the effect of intervention-specific and overall hospital volume on surgical complications, potentially avoidable reoperations, and deaths across 1.4 million interventions in 353 hospitals. Outcome variations across hospitals were analyzed through a 3-level hierarchical logistic regression model (patients, surgical interventions, and hospitals), which took into account interventions on multiple organs, 144 intervention categories, and structural hospital characteristics. Discriminative performance and calibration were good. RESULTS: Hospitals with more experience in a given intervention had similar reoperation rates but lower mortality and complication rates: odds ratio per volume deciles 0.93 and 0.97. However, the benefit was limited to heart surgery and a small number of other operations. Risks were higher for hospitals that performed more interventions overall: odds ratio per 1000 for each event was approximately 1.02. Even after adjustment for specific volume, mortality varied substantially across both high- and low-volume hospitals. CONCLUSION: Although the link between specific volume and certain inpatient outcomes suggests that specialization might help improve surgical safety, the variable magnitude of this link and the heterogeneity of hospital effect do not support the systematic use of volume-based referrals. It may be more efficient to monitor risk-adjusted postoperative outcomes and to investigate facilities with worse than expected outcomes

Decile · Intervention (counseling) · Logistic regression · Odds · Odds ratio · Psychological intervention · Statistics · Aortic aneurysm repair treatments · Cardiac, Anesthesia and Surgical Outcomes · Emergency Medicine · Internal Medicine · Medicine · Nursing · Surgical Simulation and Training

Tools

Open DOISci-Hub