COST-EFFECTIVENESS OF A HYPOTHETICAL INTERVENTION TO IMPROVE PHYSICIAN COMPLIANCE WITH TRAUMA TRIAGE GUIDELINES

Purpose: We used cost-effectiveness analysis to identify the best target for an intervention to improve physician compliance with guidelines that recommend the regionalization of moderately to severely injured trauma patients.

Method: Signal detection theory identifies two targets for interventions to improve decision making: decisional threshold (attitudes towards transferring patients to trauma centers) and perceptual sensitivity (ability to identify patients who meet triage guidelines). Changing decisional thresholds may be easier but increases the number of patients with minor injuries transferred to trauma centers. Changing perceptual sensitivity, which requires modifying heuristics, may be harder but more efficient. We constructed a decision model to compare the outcomes of patients taken initially to a non-trauma center with no intervention (e.g., given current compliance rates) to outcomes after a hypothetical intervention that either changed physicians’ decision threshold or their perceptual sensitivity. We used the societal perspective and assumed that the cost of the hypothetical intervention would be no greater than current spending on Advanced Trauma Life Support certification programs (~$40/patient). We assumed that an intervention to change decisional thresholds would increase compliance by 40%, and the intervention to change perceptual sensitivity would be half as effective. We drew model inputs from the literature. We performed a series of one-way analyses for all variables and examined the most influential variables in a multi-variable sensitivity analysis using a Monte Carlo simulation.  

Result: The incremental cost-effectiveness ratio (ICER) of an intervention to change perceptual sensitivity compared with no intervention was $64,449/quality-adjusted life year (QALY) saved. The ICER of an intervention to change decisional threshold compared with an intervention to change perceptual sensitivity was $108,668/QALY-saved. The intervention to change perceptual sensitivity remained more cost-effective as long as it was 2/5 as effective as the intervention to change the decisional threshold. The most significant drivers of the ICER were the cost of hospitalizing patients with moderate-severe injuries and the relative risk of dying after moderate-severe injuries at non-trauma centers compared with trauma centers. Probabilistic sensitivity analyses suggested that at a willingness-to-pay threshold of $100,000/QALY-saved the intervention to change perceptual sensitivity was the most cost-effective 48% of the time. 

Conclusion: We found that an intervention to change physicians’ perceptual sensitivity was likely to be the most cost-effective way of increasing compliance with trauma triage guidelines.