Error bars indicate 95% CIs.
Anderson JE, Chang DC. Does the Effect of Surgical Volume on Outcomes Diminish Over Time?. JAMA Surg. 2014;149(4):398–400. doi:10.1001/jamasurg.2013.4654
Many studies have found that high-volume hospitals have improved surgical outcomes.1,2 However, as surgical techniques become more commonly used, knowledge becomes disseminated over time. We hypothesize that with the diffusion of knowledge of surgical techniques, high-volume hospitals have a diminished comparative advantage compared with low-volume hospitals over time.
A retrospective review of the Nationwide Inpatient Sample from 1998 to 2010 was performed. The Nationwide Inpatient Sample is the largest all-payer inpatient database in the United States, containing data from an approximately 20% stratified sample of all US community hospitals.3
All patients with one of the following procedures identified by International Classification of Diseases, Ninth Revision procedure codes were included: aortic valve replacement (AVR), abdominal aortic aneurysm (AAA) repair, pancreatic resection, and esophagectomy. These surgical procedures were selected because of the presence of the Leapfrog Group quality standards.4 Patients younger than 18 years of age or patients with more than 1 procedure of interest during the same hospitalization were excluded.
Hospital volume per year was calculated by adding the number of patients who underwent each procedure. Hospitals were considered to be “high-volume” hospitals with regard to each procedure if they met the following caseloads: 120 or more AVR cases, 50 or more AAA repair cases, 11 or more pancreatic resection cases, and 13 or more esophagectomy cases.5 This was calculated for each year. Thus, hospitals may have been considered high-volume hospitals one year but low-volume hospitals another year, depending on the caseload.
Using logistic regression and adjusting for age, race, female sex, Charlson comorbidity index, and teaching hospital status, we found odds ratios of inpatient mortality by comparing patients at high-volume hospitals with those at low-volume hospitals, by procedure. Separate analyses were performed for each 2-year interval in the study period to observe changing odds ratios over time. The Charlson comorbidity index is a measure of comorbidities based on the presence or absence of a number of diagnoses for the patient and is combined in a weighted formula.6
Statistical analysis was performed using the Stata 64-bit special edition, version 11.2 (StataCorp). A P value of less than .05 was considered to be statistically significant.
A total of 361 686 patients were included (Table). Most patients underwent an AAA repair (44.1%) or an AVR (44.0%), whereas a minority of patients underwent an esophagectomy (6.1%) or a pancreatic resection (5.9%). A majority of patients (62.6%) received care at a teaching hospital. Around half (47.3%) of all patients received care at a high-volume hospital (designated as high volume for that particular procedure). Patients who underwent a pancreatic resection were more likely to receive care at both a teaching hospital (80.4%) and a high-volume hospital (62.9%). The inpatient mortality rate was 6.2%, with the highest mortality rate among patients who underwent an esophagectomy (7.9%).
Using adjusted logistic regressions, we found that, over time, patients had decreased odds ratios of inpatient mortality when they received care at a high-volume hospital compared patients who received care at a low-volume hospital (Figure). There were several years in which there was no difference in mortality, including 2002-2003 for esophagectomy and pancreatic resection and 2010 for AAA repair, AVR, and pancreatic resection. Trends were similar for unadjusted analysis (data not shown).
Our study finds that the effect of volume on surgical outcomes does not diminish over time. Instead, patients who receive care at high-volume hospitals, even when taking into account age, comorbidities, and teaching hospital status, generally have decreased odds of inpatient mortality compared with patients who receive care at low-volume hospitals. There were several years in which there was no difference in mortality, but these were the exception, and there was no clear trend. In conclusion, high-volume hospitals have maintained lower mortality rates compared with low-volume hospitals for patients who underwent an AAA repair, an AVR, an esophagectomy, or a pancreatic resection during the period from 1998 to 2010.
Corresponding Author: Jamie E. Anderson, MPH, Center for Surgical Systems and Public Health, Department of Surgery, University of California, San Diego, 200 W Arbor Dr, 8400, San Diego, CA 92103 (firstname.lastname@example.org).
Published Online: February 5, 2014. doi:10.1001/jamasurg.2013.4654.
Author Contributions: Ms Anderson and Dr Chang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: All authors.
Acquisition of data: Chang.
Analysis and interpretation of data: All authors.
Drafting of the manuscript: Anderson.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: All authors.
Administrative, technical, and material support: All authors.
Study supervision: Chang.
Conflict of Interest Disclosures: None reported.