Multicentre observational study of performance variation in provision and outcome of emergency appendicectomy

Abstract

Identification of variation in practice is a key step towards standardization of service and determination of reliable quality markers. This study aimed to investigate variation in provision and outcome of emergency appendicectomy. A multicentre, trainee-led, protocol-driven, prospective observational cohort study was performed during May and June 2012. The main outcome of interest was the normal histopathology rate; secondary outcomes were laparoscopy and 30-day adverse event rates. Analysis included funnel plots and binary logistic regression models to identify patient- and hospital-related predictors of outcome. A total of 3326 patients from 95 centres were included. An initial laparoscopic approach was performed in 66.3 per cent of patients (range in centres performing more than 25 appendicectomies over the study period: 8.7-100 per cent). A histologically normal appendix was removed in 20.6 per cent of patients (range in centres performing more than 25 procedures: 3.3-36.8 per cent). Funnel plot analysis revealed that 22 centres fell below three standard deviations of the mean for laparoscopy rates. Higher centre volume, consultant presence in theatre and daytime surgery were independently associated with an increased use of laparoscopy, which in turn was associated with a reduction in 30-day morbidity (adjusted for disease severity). Daytime surgery further reduced normal appendicectomy rates. Increasing volume came at the cost of higher negative rates, and low negative rates came at the cost of higher perforation rates. This study reveals the extremely wide variation in practice patterns and outcomes among hospitals. Organizational factors leading to this variation have been identified and should be addressed to improve performance.