Leveraging HEDIS® Data

HEDIS® hones in on 5 domains of care and measures whether the healthcare services rendered are actually improving conditions. The domains include effectiveness of care, access or availability of care, experience of care, utilization and resource use, and information about the health plan.

In the domain of access or availability of care, advancing health equity and improving quality of care for underserved populations is an up and coming topic that the NCQA is focusing on. There will be a requirement, tentatively set for 2024, for health plans to have race and ethnicity data on 80% of a health plan’s population. Health plans should be auditing their member information now to identify incomplete information sets and develop a plan for incorporating the right data into race and ethnicity stratifications. This data can be sourced directly or indirectly. Direct data comes directly from the member. For example, from a survey, enrollment information, or even requested from provider’s EMRs. Indirect data is the practice of using data for a purpose other than the reason it was originally collected, such as census data.

Another change we can expect from the NCQA is the shift to acquire more electronic sources for data collection. They’ve started to develop specific measures, Electronic Clinical Data System sets, ECDS. It’s not required today for the Medicare Advantage population, but this should be on every health plan’s radar because it will be a NCQA requirement in the future.

HEDIS data is valuable beyond HEDIS. Forward-thinking plans can leverage the output of HEDIS data to support CAHPS® and HOS. The data can be pushed downstream to other areas of the business to provide member-level insights like member compliance for care management teams or member communications. This is where risk adjustment—closing gaps, health management, communications, and member experience intersect. This should be each health plan’s future goal—to have a holistic model where all business units are on the same page, fueled by the same data and working together towards the same goal.  

Host: This is Episode 9, Leveraging HEDIS® Data. Our guest today is Alyse Schwartz, Director of HEDIS Analytics. Alyse has a Masters in public health with a focus on epidemiology. She’s an experienced researcher and has been in the quality improvement space for over 10 years. Welcome Alyse.  

Alyse: Thanks; I’m excited to be here.  

Host: HEDIS is such a nuanced area of quality, and not many people fully understand it, even in the health plan space. So, to start, will you just give us a little background information on HEDIS?

Alyse: Sure. HEDIS stands for Healthcare Effectiveness and Data Information Set. It helps determine health plans' performance in terms of quality of care for their population. As you know, HEDIS is governed by NCQA and includes over 90 measures, but the measures could change every year. HEDIS is also used by Medicaid and commercial plans, but the uses can vary.  

Host: So, can you give me an idea about how HEDIS contributes to a plan’s overall Star rating?  

Alyse: A subset of the HEDIS measure set is included in the CMS Star Rating program.  The number of HEDIS measures included can vary from year to year.  Its purpose is to support consumers in making informed decisions about which health plan to choose. HEDIS hones in on 5 domains of care and measures whether the healthcare services rendered are actually improving conditions. The domains are effectiveness of care, access or availability of care, experience of care, utilization and resource use, and information about the health plan. For example, for effectiveness of care, they’ll look at preventative measures like breast cancer or colorectal screenings, both of which are included in the CMS Star Rating. For utilization measures, they may look at mental health utilization, inpatient utilization, frequency of elective procedures, and readmission rates.  

Host: Are there any particular measures that you’re watching closely right now?

Alyse: Yeah, I’m keeping an eye on kidney health evaluation for patients with diabetes, KED is the acronym. The reason why we’re keeping an eye on that one is that, in the most recent release of the HEDIS specifications from NCQA, they retired the sub-measure medical attention for nephropathy. It is part of the comprehensive diabetes care measure. They’re making a bunch of changes to the comprehensive diabetes care measure by breaking it up into a few different new measures, and they’re retiring the medical attention for the nephropathy sub-measure. What that means is that it’s no longer available to be part of the Star Ratings in measurement year 2022. KED is a new measure from the measurement year 2020. What our thought is…that CMS may potentially pivot to this measure.

Another up-and-coming topic that the NCQA is focused on… is advancing health equity and improving the quality of care for underserved populations that aren’t receiving the right quality of care. They’re trying to highlight those disparities in membership populations. To do this, we have to incorporate the right data to be able to calculate the race and ethnicity stratifications. One of the projects we’ve been working on is auditing our member information to see what race and ethnicity information we already have on our members and then coming up with ways to get more complete information sets. And then asking, how can we use this data when calculating these rates? NCQA is really pushing this initiative forward in their timeline, and there will be a requirement for a certain percentage of race and ethnicity data to be available for a certain percentage of your health plan’s population. There are different ways to classify this data for these sources, too. They’ve identified direct data and indirect data.  

Direct data would be something that comes directly from the member. So it could be from a survey or enrollment information. Potentially, it’s something you could request from providers through the EMR. This would all be considered direct. The goal for NCQA is that all plans have 80% direct info. Indirect data is using data for a purpose other than the reason why it was originally collected, like census data. We can also use this data to make inferences about the member. There are vendors you can use that will attribute a race and ethnicity based on various information that you do know about the member. For example, using information about where the member lives.

There’s no requirement for data completeness yet; it’s tentative for the year 2024. However, starting this year, 2022, they’re updating their output to be able to handle the stratification of these measures. There are going to be 5 measures that will have output for stratification by race and ethnicity. 3 of those are Medicare Advantage, and 2 are commercial only. So we’re gathering that data now and working with our HEDIS auditors to confirm our plans.  

Host: It sounds like an incredible amount of data is collected through these measures. What are the sources, you know? Where is all of this data coming from?

Alyse: Great question. Yeah, it’s a lot of IT work; it’s very data-heavy work. The data is coming from so many sources. Claims data, medical, pharmacy, and vision claims,  abstracted medical record chart data. There’s an annual medical review for hybrid measures. Random chart samples are reviewed, and relevant information is abstracted. For example, there’s a measure controlling blood pressure. As you can imagine, not every provider is sending over a claim with a CPT 2 code that tells you what the blood pressure is. So what they’re looking at is that they want to know that a member with a diagnosis of hypertension has their blood pressure under control. So, within a larger health plan, there might be thousands of hypertensive members. NCQA allows health plans to take a random sample and go through medical record review and chart abstraction to get those blood pressures. Then, it gets submitted to NCQA and CMS. There are many measures that require more detailed information, results, and data from labs. Another example is there’s a diabetes care measure that gets looked at. A1C’s need to be in control. Again, unless you’re getting lab results from the lab vendor or CPT 2 codes of what the A1C value was, it doesn’t really help you with reporting on that specific measure.

One of the other things that has to be looked at is that the provider mapping logic is accurate. There are certain measures where you are only compliant for that measure if the service was completed by a provider of a specific specialty. Again, the HEDIS vendors have to have this data in their standard file format. Some health plans are moving in the direction of receiving EMR data in the FHIR format for interoperability. But then, currently, it likely still needs to be converted to the standard file format for submission to the HEDIS vendor.  

HEDIS also accounts for provider information, enrollment data, and core member data to help us understand who’s in your population, what their health burdens are, and what we need to focus on.

A big component of the data that’s not a requirement, but we look at it, is EMR feeds from provider organizations. You know, NCQA is trying to acquire more electronic sources for data collection. They’ve started to develop specific measures, such as Electronic Clinical Data System sets, ECDS. As I said, it’s not required today for the Medicare Advantage population, but this should be on every health plan’s radar because it will be an NCQA requirement in the future. And some HEDIS vendors aren’t ready. I recommend working towards the ability to submit the ECDS measures as soon as you can because this is the future.  

Host: Again, with so much data, this has to be valuable beyond HEDIS. Are health plans leveraging this data to support other areas of the business?  

Alyse: Yeah, in addition to submitting to NCQA and CMS, the health plan can use HEDIS results internally, they can share it with providers for incentive programs to help improve member care. Forward-thinking plans could leverage the output of HEDIS data to support CAHPS® and HOS….with the aim of improving member experience. That’s an opportunity that can be built into a member experience model. If you have a care management program, you should make sure they are plugged in somehow. HEDIS data can go downstream to support care management and refer historically non-compliant members to a care manager. It’s really about quality of care, and this data can provide insights at the member level. You know, point to members who need support. Another part of HEDIS initiatives includes strategies for member compliance.  

For example, if we’ve noticed that a member hasn’t completed their colorectal cancer screening and we’ve sent out the communications reminders, and they still haven’t done it. Maybe they have some inhibitions that are preventing them from getting the screening. We want to close that gap to ensure the members’ health, so we have to find a way to break through. So, we can mail out a kit directly to the member, and now that the member has the kit at home, can do the test, and send it back to whatever vendor partner processes the labs. It’s these well-thought-out solutions for member compliance that can close that gap and support communication efforts. Some other creative member incentive tactics I’ve seen are offering gift cards or rewards for completing screenings.  

This is where risk adjustment—closing gaps, health management, communications, and member experience intersect. This should be each health plan’s future state—to have this holistic model where all business units are on the same page, working together towards the same goal.  

Host: Alyse, you’ve given us a lot to think about in regards to HEDIS and how dynamic this part of quality is. I really appreciate you being here today.  

Alyse: Yeah, I enjoyed it.

Host: Thank you to our listeners for joining us.