Endoscopist-directed propofol administration versus anesthesiologist assistance for colorectal cancer screening: a cost-effectiveness analysis.

ABSTRACT

BACKGROUND: Propofol for colonoscopy is largely administered by anesthesiologists or anesthesiology nurses in the United States (US) and Europe. Endoscopist-directed administration of propofol (EDP) by nonanesthesiologists has recently been proposed, with potential savings of anesthetist reimbursement costs. We aimed to assess potential EDP-related benefit in a screening setting. METHODS: In a Markov model the total number of screening and follow-up colonoscopies in a cohort of 100 000 US subjects were estimated. Anesthetist-assisted colonoscopy was compared with an EDP strategy. Model outputs were projected onto the 50 - 80-year-old US population, assuming 27 % as the current uptake for colonoscopy screening. Anesthetist costs were estimated using the mean reimbursement for the corresponding Medicare code (≥ 65-year-olds) and from commercial insurance information (50 - 64-year-olds). The proportion of colonoscopies with anesthesiologist assistance was estimated from the Medicare database. Mean nurse salary was used to estimate the cost of a 2-week EDP training. The absolute number of US endoscopists was estimated by inflating by 33 % the number of board-certified gastroenterologists. No EDP mortality was assumed in the reference scenario, and 0.0008 % mortality in the sensitivity analysis. US census data were adopted. Analogous inputs were used for France to assess EDP-related benefit in a European country. RESULTS: EDP training for 17 166 nurses (one for each US endoscopist) showed a cost of $ 47 million. Cost estimates for anesthesiologist assistance for colonoscopy were $ 95 (Medicare) and $ 450 (non-Medicare commercial insurance), with 34.8 % of colonoscopies requiring anesthesiologist assistance. US implementation of an EDP policy showed a 10-year saving of $ 3.2 billion (Monte Carlo analysis 5 - 95 % percentiles $ 2.7 - $ 11.9 billion). In the sensitivity analysis, assuming 50 % of colonoscopies were anesthetist-assisted showed an EDP benefit of $ 4.6 billion. Assuming a 0.0008 % mortality rate, the incremental cost - effectiveness of anesthetist-assisted colonoscopy versus an EDP policy was $ 1.5 million per life-year gained, supporting EDP as the optimal choice. A 31-fold increase of EDP-related mortality or a 17-fold cost reduction for anesthetist-assisted colonoscopy was required for EDP to become not cost-effective in this scenario. Implementation of an EDP policy in France, within a guaiac-fecal occult blood test (g-FOBT) screening program, was estimated to save € 0.8 billion in 10 years. CONCLUSIONS: The absolute economic benefit of EDP implementation in a screening setting is probably substantial with 10-year savings of $3.2 billion in the US and €0.8 billion in France. The impact of an eventual EDP-related mortality on EDP cost - effectiveness seems marginal. The huge economic and medical resources entailed by anesthetist-assisted colonoscopy could be more efficiently invested in other clinical fields.