We have been working to build 2025 EH eCQMs off a population health platform that receives data from multiple sources. The EHR that provides the most data across our client base doesn't consistently provide an indicator to us that a location is an ICU.  We've been in conversations with them, and we found that they offer an accomodation code that registration changes to ICU when the patient becomes an ICU patient, regardless of the physical unit.  The advantage of using that is that we would capture patients who aren't in a physical ICU (most likely due to census overflow- this happened frequently during COVID but occurs fairly frequently in regular circumstances) but who are expected to be treated as ICU patients. It would also make it easier to exclude patients who are physically in the ICU but have been transferred to step-down (also likely due to census overflow-no beds available in lower level of care).  We think we may be able to use this for facility.location but it depends on how literally the verbiage is interpreted.

The question is about the intent of the measure. Is this measure intended to ensure that patients identified as critical, ICU patients are treated within the standard, regardless of physical location, or is it simply to audit a specific location and allows omission of patients who would be expected to receive the higher level of care even though they're not physically in an ICU unit? If the intent is to make sure the patient receives the right level of care in a timely fashion regardless of whether a bed is available in an actual ICU location, the accomodation code they provide will provide the most accurate start/stop times. If it's not as concerned that every critical patient receives the right level of care, only the ones physically in an ICU, then we are still left with no clear way to identify an ICU from this vendor.