Study populations and casemix: influence on analysis of postoperative outcomes.

BACKGROUND: The importance of patient casemix as a determinant of surgical outcome is now being recognized. The present study was undertaken in order to compare the presentation and outcomes in colorectal patients managed surgically by the same surgeon, in the same way, in different settings.
METHODS: Colorectal outcome data from the University Department of Surgery in Wellington and the King Faisal Specialist Hospital in Riyadh were analysed in order to determine casemix differences between the two hospitals. Data relating to the type of surgery, the surgeon, the patient's disease, the operation performed and the postoperative complications were compared. Specific colorectal clinical indicators were compared for two commonly performed operations for rectal cancer: anterior resection and abdomino-perineal resection of the rectum.
RESULTS: Wellington patients were slightly older and there were more females. Emergency surgery was more frequent in Wellington. Left hemicolectomy, sigmoid colectomy, abscess drainage and pilonidal surgery were more common in Wellington whereas abdomino-perineal resection and anterior resection of the rectum, stoma closure, fistula surgery, seton insertion, restorative proctocolectomy and ileostomy were undertaken more frequently in Riyadh. More complex anal fistulas were managed in Riyadh. Condylomata accuminata, pilonidal abscess, anorectal abscess, rectal prolapse and diverticular disease were rarely seen in Riyadh. There were more postoperative pulmonary and cardiac complications in Wellington. Patients having anterior resection of the rectum were younger in Riyadh and there were proportionally more females. There were some obvious numerical outcome differences in postoperative atelectasis, wound infection, anastomotic leak and deep vein thrombosis rates but none of these reached statistical significance except atelectasis. In Riyadh the usual male-to-female ratio of patients undergoing abdomino-perineal resection was reversed but, again, none of the numerical outcome differences observed reached statistical significance except postoperative atelectasis and intraabdominal abscess.
CONCLUSION: Although not statistically significant, the results of the present study suggest that when the same surgeon operates using the same technique in different communities, the outcomes may be different. Care should thus be taken when comparing different populations with different casemixes before definitive conclusions are made in comparative studies.