Seasonal variation in surgical outcomes as measured by the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP)

Abstract

Conclusion: There are seasonal variations in surgical outcomes with respect to postoperative morbidity and mortality. Summary: The complete experience of a surgical procedure is complex and involves interactions between both systems and individuals. It is assumed that increased surgeon experience can help produce favorable outcomes. In July and August, there are relatively new trainees in most academic medical centers. At this time of the year, the trainees are less familiar with their roles and responsibilities. It is postulated there may be more adverse outcomes in surgical procedures performed early in the academic year vs later. The authors propose that the National Surgical Quality Improvement Program (NSQIP) is a database that is sufficiently large to allow a multicenter, risk-adjustment analysis of surgical morbidity and mortality on a month-to-month basis. In this study, the authors use 30-day morbidity and mortality rates derived from NSQIP-participating hospitals and compared two periods of care: an early group from July 1 to August 30 and a late group from April 15 to June 15. The postoperative morbidity rate was 18% higher in the early (n = 9941) vs late (n = 10, 310) groups (odds ratio, 1.18; 95% confidence interval [CI], 1.07-1.29, P = .0005; c-index, 0.794). Mortality was 41% higher in the early group compared with the late group (odds ratio, 1.41; 95% CI, 1.11-1.80; P = .005; c-index 0.938). Comment: The study really cannot reliably implicate inexperience of resident trainees as contributing to the so-called July effect. No hospitals without training programs were included as controls. In addition, other studies have suggested a “July effect” does not exist (Surgery 2001;130:346-53 and J Gen Intern Med 2003;18:639-45). The authors also noted an uptake in morbidity and mortality in December. At this point, resident inexperience is not likely important. Also, many attending surgeons take a vacation in July and August; therefore, slightly more urgent cases may be performed in those months than at other times of the year. It may be that less attending supervision in peak vacation months or more urgent surgery in the summer months and December are contributing more to seasonal variation in surgical morbidity and mortality than the inexperience of resident physicians.