Memorandum: Recommendations Related to Data Collection Requirements in Section

Below are the collective recommendations related to implementing this provision from organizations representing communities and populations who will be most affected by the effective implementation of this new requirement.

• Ensure community input and engagement in the design, planning, implementation and dissemination of data. HHS should ensure that community stakeholders are included in the decision-making process and testing of the Secretary?s data collection, reporting and analysis protocol under section 4302. We recommend that HHS also create an expert advisory committee, similar to the U.S. Census Advisory Committees, to ensure that all federal surveys have a process of including non-governmental partners from all Office of Management and Budget (OMB) racial communities to participate in the design, planning, implementation and dissemination of survey data. The inclusion of non-governmental partners will help address the unique cultural, linguistic, and social barriers that deter certain populations from survey participation. The Secretary should build partnerships with local community organizations and trusted groups within these hard-to-count communities to develop data collection protocols that are culturally respectful, and ensures confidentiality for the survey participants.
• Utilize community-based outreach strategies to ensure robust participation of respondents from targeted communities, including racial, ethnic and limited English proficient populations. HHS should form community partnerships to ensure sufficient geographic and population-specific representation in any data collected, reported or analyzed pursuant to section 4302. National survey projects such as the U.S. Census have benefitted from high participation rates due to the investments in community outreach that were built into the research design.
• Train staff in collecting demographic data, including explaining why this data is being collected. The Health Research and Educational Trust (HRET) developed a toolkit for collecting race, ethnicity and language data after testing different rationales for collecting this data.i Similar training toolkits should be developed and made available for the other demographic categories.
• Adopt clear privacy and nondiscrimination protections. For this requirement to be impactful, individuals must feel comfortable disclosing personal information that can help to improve the care they receive and foster a broader understanding of health care disparities. We support the language in section 4302(e)(1)(A)(i) regarding the application of the Health Insurance Portability and Accountability Act (HIPAA) of 1996, and encourage HHS to ensure that the privacy protections applied to demographic data should be not only as broad as HIPAA, but as stringent. Patients should be made aware of their privacy protections and rights ? including those granted under applicable state laws as well as HIPAA ? and have a clear understanding of why this information is being collected and who will have access to what forms of information.
• Safeguard that patient/enrollee reporting of demographic data be voluntary. While health care systems and providers should be required to ask for data of patients/enrollees, the responses to data collection requests are (and should be) voluntary for  patients/enrollees to report and should be self-reported to ensure accuracy.
• Support analyses based on multiple demographic variables. While this section lists several specific demographic variables for data collection to better understand and address health disparities,ii it should be emphasized that these variables are neither mutually exclusive nor unrelated. As individuals, each person has a sex, race, ethnicity, primary language, and disability status, and all these demographic identities interact in relevant ways for understanding and addressing health disparities. At the community and population level, these variables, both individually and in combination, can be explanatory for both risk and protective factors. Racial and ethnic minority women receive poorer quality care than racial and ethnic minority men, who receive poorer care than white men.iii Spanish-speaking Hispanics experience poorer quality care than English-speaking Hispanics, who experience poorer care than non-Hispanic whites.iv Compared to women without disabilities, women with disabilities are more likely not to have regular mammograms or Pap tests.v Racial and ethnic minorities with disabilities experience greater disparities in diagnoses and utilization of assistive technology.vi When additional variables such as socioeconomic status, sexual orientation, and gender identity also are considered, there is evidence that these multiple variables may have a cumulative or compounding impact on health disparities.vii While health disparities research and analyses often focuses on variations based on a single demographic variable, in our increasingly multicultural society, it is vital that HHS?s future data collection support these types of analyses based on multiple demographic variables. This requires standardized categories and definitions for all these demographic variables as well as inclusion of all these variables in as many data collection opportunities as possible. Finally, given the smaller population sizes of some of these variables, it is critical that oversampling and other small population data collection methods continue to be used by HHS, especially when seeking data based on multiple demographic variables.