Challenges with eCQMs in General

The AAMC is supportive of CMS’ efforts to improve the quality of care by developing measures on dimensions of patient harm or adverse patient safety events, but notes that CMS has previously recognized and responded to the challenges regarding the feasibility of electronically-submitted measures and has reduced the number of eCQMs hospitals must report for FY 2019 and 2020 payment. There is considerable burden required to map the necessary data elements from the EHR to the appropriate Quality Reporting Data Architecture (QRDA) format, and some vendors are not properly equipped to collect and transmit such data through the CMS portal.

Mandatory eCQM reporting depends on hospitals using the correct version of specifications, which is generally in the control of the EHR vendors, not the hospitals. The AAMC urges CMS to continue outreach to EHR vendors, hospital quality staff, and other affected stakeholders to identify underlying structural problems and barriers to successful reporting of these measures. With this in mind, the Association continues to have concerns that hospitals and vendors may not be adequately prepared to fully report eCQMs, and asks CMS to focus resources on sufficiently addressing current concerns with eCQM reporting rather than on developing additional eCQMs for inclusion in hospital reporting programs for the future. Focusing on the inclusion of a small number of measures in the eCQM program that are meaningful and not overly burdensome will provide hospitals with additional time and bandwidth to address the considerable challenges of electronic data reporting.

Finally, the AAMC advises that completed testing of these eCQMs under development should demonstrate reliability and validity in the acute care setting and these measures should be submitted to National Quality (NQF) for review and endorsement. CMS should vet these new eCQMs across a selection of vendors and hospitals prior to considering the measures for addition to a CMS quality reporting program for implementation.

Measure Comments: Hypoglycemia

The Association agrees that it is important to develop measures that focus on reducing the most common adverse drug events and that hospitals should implement protocols to manage hypoglycemia for critically ill patients. That said, this measure as currently developed does not appear to be useful in assessing and improving the care for patients with severe hypoglycemia due to the administration of an anti-hyperglycemic medication 24 hours prior to the harm event because of the timing and manner of the glucose measurement. The majority of tests for monitoring blood glucose levels are done with a glucometer at the bedside (point-of-care testing) because it is cost effective and expedient and not from a drawn blood sample that goes to the lab. In the event that a drawn blood sample goes to the lab and comes back as less than 40 mg/dL there is no immediate course to provide the sugar and 5-minute glucose measure follow-up because the lab test was often done hours before. By measuring only lab tests, and not including point-of-care testing, the measure removes the majority of blood glucose testing from measure.

Additionally, by measuring quality performance (via removal from the numerator) as a rise from less than 40 mg/dL to greater than 80 mg/dL in 5 minutes, the quality exclusion is essentially only eliminating the false positive test results. That is, the measurement of less than 40 mg/dL was likely not correct, because it is physiologically unlikely that even with administration of glucose that the level will rise that quickly. Was the intention of the measure developer to remove false positives when removing cases with a follow-up of another glucose test within 5 minutes with a result greater than 80 mg/dL?

It should also be noted that administering an anti-hyperglycemic is the standard of care to lower the glucose for a patient that is hyperglycemic and that a measure to incentivize management of hypoglycemia should not have the indirect potential to cause second guessing of that standard of care. Instead, it should incent care workflows to ensure that there is appropriate glucose monitoring after the administration of the anti-hyperglycemic.