Best Practices

Learning Network participants identified a variety of best practices to consider when leveraging race/ethnicity data for quality improvement purposes at the health plan level. Some of the best practices are named below and fall within three major areas of focus: data collection and management, linking race/ethnicity to quality performance data and using race/ethnicity data to inform specific quality improvement initiatives. For more information on these and other best practices, please refer to the Learning Network summary report below. Additionally, NCQA held a public webinar on best practices in July 2023; the recording and slides can be accessed for free here.

Download the Summary Report

Improving Race/Ethnicity Data Collection and Management:

  1. Build the capacity to continuously conduct an inventory of all race/ethnicity data sources at your organization’s disposal. Such a process should enable your staff to compare strengths and weaknesses of various sources, and it should allow for centralized data mapping to take place to accommodate different use cases.
  2. Develop processes to prioritize different data sources based on various attributes such as access, completeness and accuracy. This may involve the creation of logic documents to facilitate prioritization of sources based on specified criteria set by your organization.
  3. Determine how to best engage different functional units of your organization to support collection and storage of different data.
  4. Focus resources on getting known, member-reported data through driving member traffic to certain sources such as member portals.
  5. Allow members to choose more than one field when possible. Creating this option can help reflect the unique needs and experiences of members with multiracial backgrounds.
  6. For organizations that rely on external data, determine if it is possible to receive data through alternative channels such as supplemental state sources or additional race/ethnicity feeds. This may require conversations or negotiations with external entities about the need to receive improved race/ethnicity information.

Linking Race/Ethnicity with Quality Performance Data:

  1. Create and refine existing processes to match and link race/ethnicity with quality measurement data. This should involve training all staff who interact with the data to fully understand the motivation behind data linkage as well as developing consistent rules for mapping to desired race/ethnicity categories from multiple sources.
  2. Consider how your organization might engage additional functional units, in particular the IT division, in linking race/ethnicity with quality metrics and analyzing identified gaps. This may require the involvement of a range of business units, some of which may need training on understanding how to link such data.
  3. Develop a set of clearly-defined objectives behind analyzing race/ethnicity data in the wider context of quality metrics. Staff must have a clear understanding of why they are working with the data and specifically for what use case(s), as the data could potentially be used for a number of purposes.

Leveraging Stratified Data for Quality Improvement:

  1. Consider how to utilize and report on data when sample sizes are small. Where possible, leverage category flexibility (e.g., combine smallest categories) to creatively evaluate performance and to make the data more actionable.
  2. Develop processes that allow the flexibility of reporting race/ethnicity in multiple ways for different audiences. This might mean having the ability to report race/ethnicity values reflecting higher-level categories (e.g., OMB standard) and more granular categories.
  3. Consider opportunities to meaningfully collaborate with external partners in the community such as accountable health partners, county organizations, community health entities and member committees. Soliciting input from external voices can ensure that the most critical community needs are focused upon.
  4. Leverage opportunities to build and maintain trust with members as well as all stakeholders who may view results. Ensure that staff have resources to appropriately present the importance of the data and how results inform quality improvement efforts.
  • Building on direct data and having one place to store all the data at an enterprise level so that everybody's contributing to the same database with their data that's collected.

    Client Services
  • I would say success would be the enterprise level accountability for direct data, the ability to ingest it into our HEDIS tool and then our ability to analyze data and start to close gaps that we identify.

    HEDIS Operations
  • It's very hard for a health plan to move forward without engaging in direct data collection, not through an intermediary.

    Senior Vice President
    Performance Measurement and Improvement
  • We have to be consistent across the organization in terms of how we’re going to report on race and ethnicity. And we have to be consistent on what the sources are. So that’s probably one of the biggest challenges I see within our organization.

    Data Analytics
  • You have to have IT on board and you have to have the funding to be able to get the infrastructure in place because that’s the biggest challenge.

    Data Analytics
  • We were very intentional about choosing the membership of the team to make sure that we had representation from really all of the business units that are necessary to motivate change within the organization to eliminate inequities.

    Quality Officer
  • You can’t look for disparities where data is missing.

    Data Analytics
  • You need to know how the data gets put together before it gets to you.

    Information Technology Developer
  • One of the big lessons learned was to question the data before it gets to your doorstep.

    Information Technology Developer
  • It does seem that there are sources that will use different categorizations. So yes, sometimes we get the OMB values, sometimes we get more of the CMS values. It's just kind of all over the road.

    Principal Health Care Analyst