HOME
Return to Table of Contents
116 A COST-EFFECTIVENESS MODEL TO EVALUATE THE ROLE OF REPEAT STANDARD ENDOSCOPY, IN ADDITION TO CAPSULE ENDOSCOPY OR PUSH ENTEROSCOPY, IN THE INVESTIGATION OF OBSCURE GASTROINTESTINAL BLEEDING P Tartaro, JC Gregor, TP Ponich PURPOSE: To compare the yield and cost-effectiveness of repeating standard endoscopy (gastroscopy and/or colonoscopy) versus proceeding directly to push enteroscopy (PE) or capsule endoscopy (CE) in detecting a bleeding source anywhere in the gastrointestinal (GI) tract in patients with obscure gastrointestinal bleeding (OGB).
Division of Gastroenterology, London Health Sciences Centre, University of Western Ontario, London, Ontario
METHODS: Using TreeAge Pro 2004 software, two decision trees were created, one incorporating PE and the other CE. Various permutations of repeat standard endoscopy were then incorporated into each of the models. It was assumed that a positive CE would be followed by a gastroscopy, PE or colonoscopy if the lesion was felt to be reachable by this means and hadn't already been performed in the model. Probability data was derived from published articles with data on the yield of PE, yield of CE, and rates of lesions that were missed by standard endoscopy. Costs ($CDN) were from a third party payer perspective and calculated from Ontario hospital administration data and Ontario physician (OHIP) fees. One-way and two-way sensitivity analyses were performed on all variables.
RESULTS: In the decision tree incorporating CE, a strategy of repeating gastroscopy prior to proceeding to CE had the highest yield and lowest cost-effectiveness ratio (CER), with base case values of 74% and $1,969 per bleeding source detected, respectively. One-way sensitivity analysis was unremarkable. However, two-way sensitivity analysis revealed that proceeding directly to CE had the highest yield with the lowest cost-effectiveness ratio if CE could detect nearly 100% of upper GI lesions plus at least one of the following: low yield of initial gastroscopy, high cost of EGD or low cost of CE. In the decision tree incorporating PE (assuming CE would be performed if the PE was negative), a strategy of proceeding directly to PE (i.e. without repeat standard endoscopy) had the highest yield and lowest CER with base case values of 67% and $2,137 per bleeding source detected, respectively. This was insensitive to all variables in the one-way and two-way sensitivity analyses.
CONCLUSIONS: If CE is chosen as the investigation of the small bowel, then a strategy of initially repeating the gastroscopy (without colonoscopy) would be most cost-effective. However, if PE is chosen as the investigation of the small bowel (on the premise that a CE would be performed if the PE is negative), then repeat standard endoscopy would not be cost-effective.