In response to https://www.ncqa.org/about-ncqa/contact-us/public-comments/hedis-public-comment-2/, the Massachusetts Health Data Consortium (MHDC) and its Data Governance Collaborative (DGC) have submitted the following comment.

NOTE: These comments were submitted in sections per the NCQA process but are presented collectively in this document

About MHDC

Founded in 1978, MHDC, a not-for-profit corporation, convenes the Massachusetts’s health information community in advancing multi-stakeholder health data collaborations. MHDC’s members include payers, providers, industry associations, state and federal agencies, technology and services companies, and consumers. The Consortium is the oldest organization of its kind in the country.

MHDC provides a variety of services to its members including educational and networking opportunities, analytics services on both the administrative and clinical side (Spotlight), and data governance and standardization efforts for both clinical and administrative data (the Data Governance Collaborative/DGC and the New England Healthcare Exchange Network, respectively).

About DGC

The DGC is a collaboration between payer and provider organizations convened to discuss, design, and implement data sharing and interoperability among payers, providers, patients/members, and other interested parties who need health data. It is a one stop interoperability resource. The DGC primarily focuses on three areas:

1. Collaboration: Development of common understanding of and specifications for data standards, exchange mechanisms, and what it means to participate in the modern health IT ecosystem
2. Education: helping members understand their regulatory obligations, the data and exchange standards they're expected to use, and modern technology and related processes
3. Innovation: Identification and development of projects and services needed to make modern health data practices and exchange a reality

Particularly pertinent to this comment is the DGC history working on quality measures. The first major project of the DGC involved creating a (pre-FHIR) flat file data exchange using Secure FTP to send data between payers and providers. This specification – called the MHDC Quality Measures specification - was developed collaboratively in conjunction with payers and providers in Massachusetts, was first released on December 31, 2019, and is still in use.

General Comments

This section comments on the general approach taken by NCQA in their posted proposal and were provided under the Other Feedback option.

Inconsistency of Follow Up Trigger Date

We note that there is an inconsistency in the methodology NCQA uses to start the window for follow up care when looking at abnormal BI-RADS results following a mammogram vs high risk results from a cervical cancer screening.

In the case of the breast cancer measure, the trigger date to start the 30 day follow up window appears to be the mammogram itself. In the case of the cervical cancer measure, the trigger date to start the 90 day follow up window appears to be the date the screening results are reported.

We believe there should be consistency across measures regarding approach to calculating applicable follow up windows. Starting the clock when results are reported is likely a more reasonable approach, especially given that the measure with the shorter window is also the one that currently starts from the date of screening rather than the date results are reported.

Response to Specific Questions

This section will list specific topics for which the proposal solicits input and our feedback on them. This includes new measures, modified measures, and overarching changes proposed by NCQA available for specific comments.

Gender-Inclusive Measurement for HEDIS

Status: Support with Modification

MHDC and participants in its Data Governance Collaborative approve of the impetus to make HEDIS measures more gender inclusive. However, while this was posed as an effort cutting across multiple measures, the details of the proposal appear to be specific to the Chlamydia Screening in Women measure. We believe these changes should be made across all potentially gender-specific measures.

Further, we also note that these changes are primarily changes of language and do not provide specifics regarding how to apply them to a population at large. While specifying the inclusion of patients/members recommended for screening is a step in the right direction, we believe the measure should supply specific instructions on exactly how to consistently identify this population so the measure calculations are comparing apples to apples when looking at the results across different organizations. This is also in keeping with other specifics of measures wherein NCQA provides general language such as “dementia medications” in high-level discussions within a measure but follows on with a list of the specific medications that qualify the patient/member for the population being discussed.

These specifics are important because they may also vary by measure type. One could imagine measures applying to the entirety of a population assigned a specific sex at birth, the entirety of a population with specific biological features (and thus excluding people who have received certain types of gender affirming care), excluding individuals taking certain hormones common in gender affirming care that might alter their risk factors, or other fact combinations that make inclusion or exclusion in the measure more or less appropriate. Similarly, some measures may be centered on gender identity irrespective of biological status, especially if tracking health equity is a component of the measure.

Finally, we propose measure names should also use gender inclusive terms. For example, Chlamydia Screening in Women should be renamed to something like Chlamydia Screening in Qualified Individuals.

Acute Hospitalization Following Outpatient Surgery Measure

Status: Support with Modification

Measure Title

Given the inclusion of colonoscopies, we feel using procedures instead of surgeries in the title captures the measure content more accurately.

Dual Eligible Status

If the population size is too low to support stratification by race/ethnicity, it likely is also too small to stratify by dual eligible status, but DGC participants are not opposed to seeing if it yields statistically useful data.

Endoscopies

We found it interesting that NCQA proposed the inclusion of colonoscopies and not endoscopies as procedures worth tracking. Providers participating in our conversation felt that the two were analogous in terms of risks of complications requiring hospitalization, and that many (but not all) of the risks were the same for both procedures. We ask NCQA to consider adding another rate for endoscopies, and perhaps also one for combined endoscopy/colonoscopy procedures as they would likely be excluded from the individual procedures counts given the increased risks of the combination.

Procedures Occurring 12/17-12/31

Procedures occurring between December 17 and December 31 in any year are always excluded from this measure. We suggest using a procedure date of December 17 of the year prior through December 16th of the measure year to calculate this measure so as not to exclude relevant procedures. Further, as the excluded procedures are at both a busy and stressful time of the year, they may be more likely to have higher rates of issues so omitting that specific period even to ensure that the timing fully fits into the measure year (if that is the rationale) seems problematic. [Note: we had dissent from one payer participant who believes this would be an additional burden]

Procedures the Day Before Admission

Omitting hospitalizations that are not scheduled but occur the day after an outpatient procedure seems problematic. Unplanned hospitalizations occurring the day after an outpatient procedure likely have some correlation to that procedure. Certainly, correlation is just as likely if not more so than other admissions that occur 2-15 days post-procedure.

Irritable Bowel Disease

Providers participating in our discussion found it odd irritable bowel disease is an exclusion for colonoscopies. They did not think it a likely factor to contribute to hospitalization and could not think of any other reason to exclude such patients.

Death

DGC participants found it unusual that patients who died during the measure year were not excluded from the measure.

Blood Pressure Control for Patients with Hypertension Measure

Status: Support with Modification

Use of Clinical Data

Given that this is an electronic measure, we were struck by the proposed measure specifically requiring diagnosis data come from claims rather than allowing it to be acquired via other means such as clinical data exchange. One of the major pros of electronic measures is the ability to automate some of the data collection via FHIR APIs and other mechanisms; by discounting this data and requiring that diagnoses come solely from claims NCQA is shutting out a potentially rich data source as well as precluding the use of more modern data exchange mechanisms that the industry is quickly moving toward adopting.

Medication information may also be available via this mechanism and, if it includes indication that the medication was actually dispensed and not just prescribed, would be sufficient to satisfy the medication requirements of the measure if its use is permitted for that purpose.

We further note the NCQA FAQ on ECDS Measures (https://www.ncqa.org/hedis/the-future-of-hedis/ecds-frequently-asked-questions/) indicates that any ECDS data can be used to identify any element of a measure’s specification including identifying the eligible population. Requiring the use of claims or pharmacy data for this denominator is not in alignment with this statement.

Incomplete Blood Pressure Readings

According to the numerator definitions, the patient is non-compliant if their last recorded blood pressure reading is incomplete. We strongly recommend that previous blood pressure readings within the measure period should be used instead if they exist and also meet the other measure requirements (such as timing related to diagnosis events).

Formal Diagnosis vs Repeated High Values

Patients with complex medical conditions as well as an existing hypertension diagnosis may not have a formal diagnosis of hypertension attached to all or even any of their medical encounters within the pertinent time period if those encounters are for a purpose other than blood pressure management. This may be true even if they exhibit blood pressure values above the normal range at some of those times. It may make sense to consider identifying patients with K blood pressure readings above 140/90 taken on different days during the measure period as patients who should be part of the measure population where K is some value higher than 2, perhaps 4 or 6.

Documented BI-RADS Assessment after Mammogram Measure

Status: Support with Modification

Mammograms Occurring 12/18-12/31

Procedures occurring between December 18 and December 31 in any year are always excluded from this measure. We suggest using a procedure date of December 18 of the year prior through December 17^th of the measure year to calculate this measure so as not to exclude relevant procedures. Further, as the excluded procedures are at both a busy time and stressful time of the year, they may be more likely to have higher rates of delays so omitting that specific period even to ensure that the timing fully fits into the measure year (if that is the rationale) seems problematic. [Note: we had a dissent from one payer participant who believes extending the measure period this way would be an additional burden]

Gender Diverse Populations

We realize this measure only applies to patients who get a mammogram and is not directly measuring whether the appropriate people were scheduled for screenings, but since the issue of transgender individuals and gender diverse populations was raised several times within the supporting documentation and also as a cross-cutting issue of concern for the MY2025 changes, we would like to see it more directly addressed within the measure in some way.

Some preliminary thoughts include a stratification for cisgender women vs other patients screened (if there is sufficient data available) or perhaps a count of patients in the population who are not cisgender women as a starting point to at least ensure someone is thinking about the need to consider them in the screening processes.

Follow Up after Abnormal Breast Cancer Assessment Measure

Status: Support with Modification

Type of Follow Up

We believe it may be useful to track the type of follow up provided for abnormal screening results. It may be useful to provide counts of patients receiving a biopsy as a follow up vs another mammogram vs an ultrasound. If this is not deemed helpful or is too onerous, we suggest that tracking ultrasound vs an additional mammogram as the follow up for inconclusive results would be a good first step, especially as it might lead to future tracking of whether one type of follow up or the other leads to better outcomes.

Type of High-Risk Assessment

We wonder if it is also useful to track the frequency of follow up based on the specific type of abnormal screening result rather than collectively for all abnormal results together. We suggest considering stratification by category, tracking the values 4, 4A, 4B, 4C, and 5 individually.

Gender Diverse Populations

We realize this measure only applies to patients who get adverse results from a screening mammogram and is not directly measuring whether the appropriate people were scheduled for screenings, but since the issue of transgender individuals and gender diverse populations was raised several times within the supporting documentation and also as a cross-cutting issue of concern for the MY2025 changes, we would like to see it more directly addressed within the measure in some way.

Some preliminary thoughts include a stratification for cisgender women vs other patients (if there is sufficient data available) or perhaps a count of patients in the population who are not cisgender women as a starting point to at least ensure someone is thinking about the need to consider them in the screening processes. The ultimate goal, however, should be the ability to track whether follow up rates differ for cisgender women vs others who get abnormal screening results.

Cervical Cancer Screening Follow Up Measure

Status: Support with Modification

Age Discrepancy

The introductory materials and measure description indicate the measure is meant to apply to patients 21-64 years of age but the initial population defined for the measure lists patients 22-64 years of age. The applicable age range should be consistent throughout the entire measure documentation.

Gender Diverse Populations

We realize this measure only applies to patients who get adverse results from cervical cancer screening and is not directly measuring whether the appropriate people were scheduled for screenings, but since the issue of transgender individuals and gender diverse populations was raised several times within the supporting documentation and also as a cross-cutting issue of concern for the MY2025 changes, we would like to see it more directly addressed within the measure in some way.

Some preliminary thoughts include a stratification for cisgender women vs other patients (if there is sufficient data available) or perhaps a count of patients in the population who are not cisgender women as a starting point to at least ensure someone is thinking about the need to consider them in the screening processes. The ultimate goal, however, should be the ability to track whether follow up rates differ for cisgender women vs others who get abnormal screening results.

Potentially Harmful Drug Disease Interactions Measure

Status: Support with Modification

Age of Initial Population

The discussion of changes at the start of the proposal file does not call out any changes to the age of the population for this measure, but the red lining of the measure text adjusts the minimum age from 65 to 67. If this is an intentional change it should be called out in any future documentation or other information provided about changes to the measure.

Follow Up After ED Visit for Mental Illness Measure

Follow Up After Hospitalization for Mental Illness Measure

Status: Support with Modification

Note: Our comments are identical across the two measures

Occupational Therapy as Follow Up for Mental Illness

It is unclear to us why occupational therapy would qualify as direct follow up care for acute mental health incidents.

Avenues for Mental Health Services

Over time, it has become common to access mental health services outside of the traditional healthcare system. Mental health services are offered at schools (as noted in the proposed measure), but also at community centers, homeless shelters, churches, via mobile and web applications, and more. If follow up from these and other non-traditional sources of care are not counted in this measure it will not be an accurate representation of the follow up rate achieved.

Tracking Mental Health Follow Up Services

As the avenues for receiving mental health services have expanded, the ability to track them using traditional healthcare mechanisms decreases. While some of these services may be covered by payers, the majority are not (often including some of the mechanisms already noted in the proposal such as peer support services). Mechanisms for tracking receipt of these services are needed, preferably mechanisms that are consistent, automated, and auditable in keeping with the goals of electronic data collection. It is not clear at this time what those mechanisms could be. An interim step might be accepting manual documentation such as receipts for payment or allowing for some type of attestation process. However, these require direct involvement of the patient and uses their time for tasks that they get no direct benefit from and thus are unlikely to prioritize.

Adult Immunization Status Measure

Status: Support with Modification

Handling of Allergies

It seems odd that people with allergies to the Hepatitis B vaccine would be counted in the numerator of the measure rather than be treated as exclusions in the denominator.

Use of Childhood Vaccination Status

It seems unlikely that childhood vaccine records would be directly available in a verifiable way for adult patients, particularly as they get older. In most cases, patients change their entire cadre of providers when they transition to adulthood. Individuals often move to attend college or for other reasons around this transition age. The older a patient gets, the less likely direct access to childhood immunization records remain available. Statewide immunization databases may be available for recent vaccinations, but many adults would have received their childhood vaccinations before these services were stood up and the records for others may be located in public health records outside of the jurisdiction and accessibility of the patient’s current providers or payers. Some indication of the proof deemed acceptable for childhood immunization records would be appropriate, but attestation of some sort may be the only option available.

Age ranges

Participants in our Data Governance Collaborative believe tracking in multiple age bands makes more sense than incorporating all ages into one cohort. That said, we do not have strong feelings on what those bands should be. We have no objection to the suggested 19-30 and 31-59 age bands, but we would also support other age distributions such as two equal bands of 19-39 and 40-59.

Acute Hospitalization Measure

Status: Support

We approve of extending this measure to cover Medicaid patients age 16-64.