There is no diagnosis associated with the denominator or the numerator. The atribute for underweight/overweight is snomed code which is a finding. If a BMI is above or below the threshold as stated n the measure, the logic requires the provider to assign an underweight or overweight condition (finding) which would be tied in the EHR to the snomed codes.

The issue can be described as the implementation of CMS 69 with the reason requirement added to the follow-up orders for low or high BMI within the most current version. Concerns have been raised regarding the feasibility of implementing this linkage between BMI finding and subsequent intervention (see Jira issue CQM-748 for specific details).   The current measure specifications require a linkage between the finding and the intervention or medication follow up. This is captured using a "reason" attribute for each follow up action with valid reasons defined by the value sets "Overweight" and "Underweight" which are populated with SNOMED-CT concepts. These value sets are a proxy for below normal and above normal BMI findings which cannot be directly used as an attribute in this context. The earliest version of this eCQM in 2013 did not require an association between the BMI finding and follow up action. Concerns were raised at the time that we may be capturing interventions which were not specific to an abnormal BMI finding. Adding the "reason" attribute to the follow up items attempts to bring the measure more in line with its intent.   While no specific proposal for an alternate approach has been developed we welcome the opportunity for feedback and discussion with all interested stakeholders to improve the feasibility of this measure. Description/Background: A few vendors and implementers have found it difficult to implement the reason attributes found in the numerator logic of CMS69 Preventive Care and Screening: Body Mass Index (BMI) Screening and Follow-Up Plan. Workflow and feasibility assessment performed in 2016 using three provider practices who have implemented the eCQM found that overall, the measure was feasible. The most difficult data element to capture was documentation of a follow-up plan for patients whose BMI values fell outside the normal parameters, consistent with this issue ( CQM-748 ). However, one of our three testing sites’ EHRs had a drop-down menu that clinicians used to capture follow-up plans for patients with a BMI value outside normal parameters. The other two testing sites documented the majority of follow-up in patient’s care plans, which were infrequently captured in structured fields, thus not captured correctly to meet performance for this measure. In a recent discussion with the Expert Work Group the burden of extra "clicks" needed to document the reason for a follow up was acknowledged, however no consensus could be reached on a solution which would reduce the burden of documentation while preserving the measure's validity. As mentioned previously during the posting for CRP, we propose removing the attribute (documenting a reason for a specific follow-up plan) from the data elements whose value sets specifically relates to a BMI value. At this time we cannot recommend removing the "reason" attribute from the referral coding because of the potential of inaccurate assessment of the numerator criteria. Voting Options: Our proposed voting options to include for this issue: Option 1: In the numerator logic, remove the attribute (reason: Overweight) from the following data elements: Intervention, Order: Above Normal Follow-up Medication, Order: Above normal medications Also remove attribute (reason: Underweight) from the following: Intervention, Order: Below normal follow-up Medication, Order: Below normal medications Option 2: No change to numerator logic (retain the attributes: reason: Overweight and reason: Underweight) Please provide your vote by leaving a comment indicating "Yes" or "No." • Vote ‘Yes’ if you agree the proposed changes in option 1. • Vote ‘No’ if you would like no change to the numerator logic

Stakeholders voted overwhelmingly (10/11 - 91%) to remove the attribute (documenting a reason for a specific follow-up plan) from the data elements whose value sets specifically relate to a BMI value. This involves a minor change to the numerator logic. We would simply delete the attributes from the relevant data elements as shown below.   Numerator = AND: ...   OR:  AND: Union of: "Intervention, Order: Above Normal Follow-up (reason: Overweight) "  "Intervention, Order: Referrals where weight assessment may occur (reason: Overweight)"  "Medication, Order: Above Normal Medications (reason: Overweight) "  <= 6 month(s) starts before end of "Occurrence A of Encounter, Performed: BMI Encounter Code Set"   AND: "Physical Exam, Performed: BMI LOINC Value" satisfies all: Most Recent: (result) <= 6 month(s) starts before end of "Occurrence A of Encounter, Performed: BMI Encounter Code Set"  (result >= 25 kg/m2)   OR:  AND: Union of: "Intervention, Order: Below Normal Follow up (reason: Underweight) " "Intervention, Order: Referrals where weight assessment may occur (reason: Underweight)" "Medication, Order: Below Normal Medications (reason: Underweight) " <= 6 month(s) starts before end of "Occurrence A of Encounter, Performed: BMI Encounter Code Set"   AND: "Physical Exam, Performed: BMI LOINC Value" satisfies all: Most Recent: (result) <= 6 month(s) starts before end of "Occurrence A of Encounter, Performed: BMI Encounter Code Set"  (result < 18.5 kg/m2)   We will retain the attribute related to the “Intervention, Order” data elements as the value set which defines it contains more general concepts and can capture referrals for almost any reason, not just those related to BMI findings. In future updates (beyond 2018) we hope to offer further solutions to ease implementation burden and improve the measure’s alignment with clinical evidence.