Technical Appendix 1

Technical Appendix 1

The warranty allowance model division between flat fee and proportional rate

Editor’s Note: This online data supplement contains supplemental material that was not included with the published article by Francois de Brantes and colleagues, “Should Health Care Come With A Warranty?” Health Affairs 28, no. 4 (2009): w678–w687 (published online 16 June 2009; 10.1377/hlthaff.28.4.w678)

Because ECRs are severity-adjusted based on patient and other factors, there are wide variations in the total episode costs per patient. Let’s assume that the inpatient portion of the AMI ECR can vary from as low as $16,670 to as high as $71,180. Additionally, let’s assume that the average cost of potentially avoidable complications in this example is $15,000. Halving that number to create the warranty allowance pool yields an average of $7,500 per episode. Given the size of this number relative to the lower end of the severity-adjusted ECRs, if equally distributed on all cases, this amount could potentially create a significant incentive for cherry picking less severe patients. Conversely, while PROMETHEUS recognizes that there is a higher likelihood of having PACs when patients are more severe, thus necessitating a higher allowance, allocating the allowance on a purely proportional basis could lead providers to upcode the severity of all patients to maximize the allowance on each patient. As a result, and in order to mitigate these two potential effects, PROMETHEUS decided to split the allocation of the per ECR warranty using a flat and proportional amount as illustrated in the following figure. While it’s unclear that this split will effectively avoid cherry-picking or upcoding, we analyzed the relative size of the per-ECR warranty when the flat fee portion is increased from 0% to 100%. More in full text below…