Purpose: The purpose of this study was to model the effectiveness and cost-effectiveness of multi-society guideline recommendations for post-polypectomy surveillance colonoscopy. Methods: A Markov model was developed to compare the effectiveness and cost-effectiveness of several surveillance strategies. The published literature was used to estimate polyp and cancer transition rates, risks, benefits, and costs. The target population was men and women over the age of 50 with a new diagnosis of colonic adenomas. The model was calibrated to one- and three-year data from the National Polyp Study and lifetime data from the SEER registry. Results: In the base-case analysis, colonoscopy every 3 years for all patients (the 3/3 strategy) was the most effective approach, with an 83% reduction in colorectal cancer (CRC) mortality compared to no surveillance. The incremental cost-effectiveness ratio (ICER) of this strategy was $162000 per life-year saved (LYS) compared to the 3-year/5-year strategy recommended by current guidelines (the 3/5 strategy). The 3/5 strategy resulted in a 76% reduction in CRC death compared to no surveillance. The ICER of this strategy was $57000 per LYS compared to a 3-year/10-year strategy (the 3/10 strategy). The 3/10 strategy was the least effective but also the least costly of the three approaches, with an ICER of $4500 per LYS compared to no surveillance. Sensitivity analysis revealed 2 underlying factors that had important effects on the ICER: (1) the advanced adenoma miss rate; and, (2) the rate of malignant transformation of an advanced adenoma. Specifically, increasing the advanced adenoma miss rate from 4% (base) to 12% decreased the ICER of the 3/3 strategy to $79000 per LYS and decreased the ICER of the 3/5 strategy to $28000 per LYS. Decreasing the advanced adenoma transformation rate from 5% (base) to 1% increased the ICERs of both the 3/3 and the 3/5 strategies to over $100000 per LYS. Conclusions: A 3/5 or 3/10 strategy for surveillance colonoscopy, as recommended by current guidelines, is cost-effective by traditional standards while a 3/3 strategy is not cost-effective. However, a 3/3 strategy may be reasonable in settings where the miss rate for advanced adenomas can be expected to be higher than that reported in the literature. Future research should attempt to better define both the miss rate and the rate of malignant transformation for advanced adenomas.
Read full abstract