Developing electronic health record-based measures of the 4Ms to support implementation and evidence generation for...

To support implementation of the 4Ms framework and more rigorous evidence of 4Ms impact, we translated Institute for Healthcare Improvement's (IHI's) recommended 4Ms routine care practices into electronic health record-based, encounter-level adherence measures and then implemented measures at a large academic medical center.

We started with the 19 care practices in IHI's 4Ms implementation guide and developed encounter-level adherence measures using structured EHR data. We also developed overall 4Ms-level and M-level composite measures. Next, we operationalized measures at UCSF Health-an academic medical center that has implemented the 4Ms using the IHI guide. We identified UCSF Health patients who should have received 4Ms care during their inpatient admission (19,335 individuals 65 years and older with an admission between January 1, 2019 and December 31, 2021), then implemented the individual measures and composite measures (all at the encounter level) using Epic EHR data. We focused on 4Ms inpatient care processes, but similar approaches can be followed for ambulatory, post-acute, and other settings.

We developed 18 EHR-based measures that captured all IHI care practices, 16 of which could be implemented using UCSF Health EHR data. For example, the EHR-based measure for the Medication care practice "deprescribe high risk medications" was measured using EHR data as "Patient had no previously existing prescriptions for high-risk medications OR patient had ≥1 previously existing prescriptions for high-risk medications deprescribed during the encounter," and 29.5% of UCSF Health encounters met this measure. For composite measures, on average, UCSF Health encounters had 61.1% adherence to the 4Ms (SD = 14.4%), with the lowest average adherence to What Matters (50.9%; SD = 44.3%) and the highest for Mentation (68.4%; SD = 13.4%).

It is feasible to construct encounter-level measures of 4Ms adherence using EHR data and derive insights to guide ongoing implementation efforts. Future efforts should refine measures based on assessments of reliability and validity.