For those who couldn't make it, we spent the lion's share of our time looking at critical differences in how we are reporting our performance in patients with severe sepsis. Jim Kaplan did an ooutstanding analysis of how ICD codes have changed over the past few years and how those changes may affect our reported results; as usual, his presentation was clear, logical, and insightful. Perhaps more importantly, he showed how our benchmarked performances are based on APR-DRG's (a proprietary invention of 3M that stands for All Patient Refined Diagnosis Related Groups; http://www.qualityindicators.ahrq.gov/\downloads\listserv\Understanding%20the%203M%20APR-DRG.pdf); this tool does not identify the same patients in the discharge database that we have been identifying for the past several years in our tracking of severe sepsis patients. Jim showed how those differences arise, i.e. how assigning patients APR-DRG 720 differs in nature from identifying them by use of ICD 9s 995.91, 995.92, and 785.5. The dilemma is twofold: we need to evaluate whether we have identified different sets of patients in successive years by use of ICD codes, and we need to attempt to understand how much credence to place in our benchmarking data. Obviously, benchmarks are imposed upon us from the outside, in that we cannot (at least as of now we don't have the clout to) change what data are used by various organizations for the purpose of benchmarking. However, it will take some analysis on our part to determine what is of more use to the patients we serve. More on that in upcoming posts.
We attempted to record the entire session, but it remains to be seen whether we were entirely successful. If so, we will have it posted on the web site, www.mwcritcare.org, for review by all.
