The costs of complications after emergency gastrointestinal surgery in Kenya

Background: The financial burden of surgery is substantial worldwide. Postoperative complications increase costs in high-resource settings, but this is not well studied in other settings. Our objective was to review the financial impact of postoperative complications. Method: Patients undergoing emergency gastrointestinal operations at a center in Kenya were reviewed between January 2017 and June 2019. In a cost analysis, we ascertained the outcome of total hospital costs, adjusted for inflation, and converted to international dollars using purchasing power parities. Costs were analyzed for their association with a postoperative complication, defined using standardized criteria. We calculated the Africa Surgical Outcomes Study surgical risk scores and clustered for discharge diagnosis in a mixed-effects generalized linear model accounting for confounding factors related to costs and complications. Results: A total of 361 individuals had cost data available. The cohort had 251 men (69.5%) and 110 women (30.5%) with a median age of 41 years (interquartile range: 29–57 years). A total of 122 (33.8%) patients experienced a postoperative complication with an overall all-cause mortality rate of 10.5%. The median total cost of hospitalization was 1,949 (interquartile range: 1,516–2,788) international dollar purchasing power parities. When controlling for patient factors and diagnoses, patients who did not develop complications had costs of 2,119 (95% confidence interval 1,898–2,340) compared to costs of 3,747 (95% confidence interval 3,327–4,167) for patients who developed a postoperative complication, leading to a 77% increase of 1,628 international dollar purchasing power parities for patients with complications. Conclusion: Our findings demonstrated a substantial financial burden generated by postoperative complications in patients undergoing emergency gastrointestinal operations. Reducing complications could allow cost savings, an important consideration in variable-resource settings.