‘Taking a calculated risk’- Is American College of Surgeon’s National Surgical Quality Improvement Program universal risk calculator a valuable tool to improve quality of surgical services in Indian surgical scenario?

Abstract

Background: American College of Surgeon’s National Surgical Quality Improvement Program (ACS NSQIP) risk calculator is a valuable tool exercised in the hospitals across the USA for predicting risk rates for 8 post-operative complications till the 30th post-operative day, taking into account 21 risk factors. Its applicability in the surgical practice in India has not been popular. This study explores, if this model is accurate in predicting risk of post-operative complications in surgical patients in India. Methods: Retrospective study done on 368 patients who underwent elective and emergency surgical procedures in a medical college hospital. Demographic data, contemplated surgery, co-morbid factors entered into the risk calculator and predicted personal risk calculated and compared to average risk. Post-operative progress reports retrieved. Complications till 30th post op day compared with the predicted risk estimates. Brier score, sensitivity and specificity, PPV/NPV, z and p values calculated for the probabilistic, predictive values of the model for validation.Results: Emergency procedures done in 143 and elective in 225. 67 developed major, 35 minor, 11 fatal complications. The average Brier score was 0.00324. Mean sensitivity was 96.08 %, specificity 98.94%. Mean PPV 82.6%, NPV 99.6%. The predictions were inaccurate for serious, “any” complications, discharge to nursing and rehabilitative facility, the colectomy ileus and length of stay. The percentage of error ranged from 2.22 to 25.Conclusions: Overall, the ACS NSQIP surgical risk calculator very fairly predicted the complications till the 30th post-operative day which naturally matched with actual complications that were seen.