By Oliver Kim
May marks the annual Asian American and Pacific Islander Heritage Month, which recognizes the history and contributions of this diverse population in the United States. Accounting for that diversity though is one of the challenges facing the Asian American-Pacific Islander (AAPI) community: for example, the Library of Congress commemorative website recognizes that AAPI is a “rather broad term” that can include
all of the Asian continent and the Pacific islands of Melanesia (New Guinea, New Caledonia, Vanuatu, Fiji and the Solomon Islands), Micronesia (Marianas, Guam, Wake Island, Palau, Marshall Islands, Kiribati, Nauru and the Federated States of Micronesia) and Polynesia (New Zealand, Hawaiian Islands, Rotuma, Midway Islands, Samoa, American Samoa, Tonga, Tuvalu, Cook Islands, French Polynesia and Easter Island).
Understanding that diversity has huge policy and political implications, particularly in health policy. For example, the incidence of colorectal cancer is similar between whites and AAPIs, but “rates of both CRC screening and survival are especially low in many Asian subpopulations.” In other words, if you looked at AAPIs as a whole, you would not notice that difference but if you were able to slice the data, you would find significant variation: “For example, Japanese Americans have been found to undergo CRC screening at rates akin to those of non-Latino Whites, while the lowest screening rates have been found among Korean and Vietnamese Americans.” That is why data policy is a top issue for leading AAPI organizations such as the Asian & Pacific Islander American Health Forum (in full disclosure, I sit on the board of this organization).
A series of federal laws—including Section 185 of the Medicare Improvements for Patients and Providers Act, Section 3002 of the HITECH Act, and Section 4302 of the Affordable Care Act—included greater reporting requirements for Medicare, Medicaid, and the Children’s Health Insurance Program as well as electronic medical records purchased through federal Medicare and Medicaid incentives. Such richer data sets would “represent a powerful new set of tools to move us closer to our vision of a nation free of disparities in health and health care.”
For example, Section 185 required the Centers for Medicare and Medicaid Services (CMS) to improve the collection of “data on disparities on the basis for race, ethnicity, and gender” across all parts of Medicare and report to Congress on its success by January 2010 and every four years afterwards. That would mean not only traditional Medicare but also the private plans that operate in Medicare Advantage and Medicare’s prescription drug program, giving a broader look into private insurers’ data and potentially creating uniformity with their non-Medicare data collection. Given Medicare’s role in health policy, the assumption was that this requirement would lead to greater uniformity in health data collection. A 2011 Medicaid report and a 2011 minority health activities report—well after the January 2010 deadline— reference “a series of reports to Congress” that will update Congress on progress implementing Section 185, but it is not clear whether CMS fulfilled its obligation to report Congress on its approach.
Similarly, one of HITECH’s purposes was to reduce health disparities by ensuring that EMRs at a minimum could collect information on race, ethnicity, gender, and primary language. Collecting this data was supposed to show that eligible providers were meaningfully using health IT and thus entitled to Medicare or Medicaid reimbursement. This information in turn would be used as part of quality improvement efforts. Yet critics have noted that federal EMR requirements could have built a much strong infrastructure in order to tackle disparities, but the federal government either missed or even rejected opportunities to advance more robust data collection. It’s also not clear why there isn’t a closer connection between Medicare and Medicaid’s data collection requirements with federal meaningful-use incentives.
These missed opportunities though can have larger implications. First, as big data helps shape delivery system reform, an outstanding question is whether these reforms can reduce disparities, particularly if the data is not there to aim the reforms in the right direction. Second, if the federal government does not connect reducing disparities in its policy decisions for data collection or health IT utilization, then it also becomes less of a priority for vendors developing the next software or app. Finally, having data sets that are too white may put some innovations like precision medicine out of reach for some minorities, exacerbating disparities as well as feelings of mistrust in the healthcare system.
Imagine if there was a richer data set that could more accurately not only the breadth and diversity of the AAPI community but our nation as a whole. Unfortunately data policy is often not the bright and shiny object that grabs the attention of legislators, one possible reason for the seeming gaps here in implementation. While it is unlikely that the current administration will take advantage of these tools in order to reduce health disparities, it is well worth advocates’ time to continue to push for a data set that looks more like America.
Thank you to Katie Rubinger for research assistance.