Dear Hybrid Hospital Wide Mortality and Readmissions stewards and CMS,

Please consider listening to participating hospitals and staff that can provide valuable insight to these measures and the problems with what you, CMS, are producing in terms of reporting. The value of CMS reporting on these measures is extremely diminished by removing data and even further so by replacing data with mean values for your own purposes but at the same time making things less clear and much more difficult to make actionable on the part of all hospitals.

Our hospital has SQL logic built for each of the hybrid measures to do internal analysis and know exactly how we perform. Our internal analysis is drastically different than what appears in the HWM and HWR HSR. The HSR as distributed is miscalculating exclusions in place of data that was successfully submitted, suggesting that our successful submission of each metric CCDE is lower than it really is, underreporting our HWM metrics by about 42% and our HWR metrics by about 14%. For instance, Exclusions 8 and 9, for failed linkage of claims data and 50% or more of CCDEs missing are the only exclusions that should count against us and combined that was only 6%, not 22% as our results state.

It also makes sense to report an unabridged version separately from the altered versions to show what was successfully submitted aside from the failed logic that has been applied by CMS’s contractor. After many years of doing this, it is obvious that more expertise lies on the hospital end than those that are producing these measures and associated reports.
If you can be so humble hear me out for a better approach. Start with vendors. Force adherence to specific data structuring for which all hospitals will already be set up to report on like metrics. Don’t ask five thousand hospitals to all create their own mapping and logic to hopefully come to the same data structure and completeness. Create auto pulls via sftp weekly with a 6-month lag from discharge to allow coding and any final billing procedures to flesh out. Take all the data you want.

These measures don’t do anything to improve care. Some may even suggest that there is negligence in coercing hospital policies to collect and therefore perform invasive procedures that are not clinically beneficial and could result in harm. CMS could better support their premise by collecting all lab and vital signs that are available and then looking for trends. Don’t create mandates on percent collected, all for risk adjustment, this is entirely unethical.

We, participating hospitals, should be seeking further reimbursement as the effort put forth to collect and report on this data is substantially more than the effort you allocated for it in the final rule. All of this to result in grossly incorrect and negligent reports that, if published, will result in reputational harm to reporting hospitals and potential for litigation.

Please reach out if you would like to work together on improving this all the way around.
Thank you,

Andrew Heiler MBA,RN
Michigan Medicine
Clinical Reporting Manager- Staff Specialist