It is sobering to publish Sell’s call for every federal survey to include questions about sexual orientation and gender identity precisely when the Trump Administration excludes such questions from the proposed 2017 National Survey of Older Americans Act Participants (NSOAAP) conducted by the Administration for Community Living (ACL) in the Department of Health and Human Services.
In particular, NSOAAP question DE1a, introduced in 2014, asked “Which of the following best represents how you think of yourself? Lesbian or Gay; Straight, that is, not lesbian or gay; Bisexual; Something else; Refused; Don’t know.” The ACL has invoked the Paperwork Reduction Act to justify excluding this question.
DE1a was allegedly being pilot tested but “unfortunately, because extremely few people identified themselves as LGBT [lesbian, gay, bisexual, or transgender], there were not enough respondents for the data to be statistically reliable or reportable” (see the Appendix, available as a supplement to the online version of this article at http://www.ajph.org). For this same reason, the 2014 and 2015 data “are not available for viewing or downloading at this time” (bit.ly/2ntXyXl).
This explanation is surprising. The act has tremendous public health implications for the LGBT community, particularly those in the older adult subpopulation. It has caught the broader research and community interest groups off guard because the ACL did not actively seek their input before publishing the proposed 2017 NSOAAP with the deletion in the Federal Register for public comment. On March 13, 2017, written or electronic comments were invited by the ACL to be submitted by May 12, 2017, which was a very short notice (bit.ly/2oONoAD).
Even more puzzling is the apparent lack of perception that complex survey methods allow us to study small communities by oversampling them. Oversampling is used routinely in survey research to get statistically meaningful information about discrete subpopulations in everything from election polls to consumer marketing surveys. We would expect the ACL to explain why greater attempts have not been made over the years to oversample the LGBT community to produce reliable estimates. The Williams Institute estimates that the number of LGBT adults older than 50 years in the United States is currently 2.4 million, and may be more than 5 million by 2030 (bit.ly/2pgHPXY), as the catastrophic impact of the HIV/AIDS epidemic in the 1980s fades away. Actually, national surveys need to consider collecting additional data on sexual identity: in the February issue of AJPH, reports estimating that the number of transgender persons in the United States is more than a million, discussed ways of identifying them in censuses and surveys.
We may learn more about the motivations behind the NSOAAP proposed modification as the Human Rights Campaign filed a Freedom of Information Act request to understand how the decision was made. AJPH will also explore these questions in an upcoming issue. But as of now the “lack of statistical significance” justification given by the ACL is so weak on its face that we are left to wonder whether other political reasons are interfering with accepted research methodology. Is it an attempt to marginalize the LGBT community, to not offend anti-LGBT survey takers, to justify reduced health measures targeting the LGBT community, or some other reason? This is especially alarming when similar rollbacks of data collection on sexual orientation and gender identity are occurring for the Annual Program Performance Report for the Centers for Independent Living and for the 2020 Decennial Census and the American Community Survey. As Sell indicates, not collecting public health surveillance data about a community is equivalent to ignoring it and pushing it back into the dark. Who would want that?
Because of the importance of health surveys to developing health policy, the government has the burden of proving that removing the questions about sexual orientation is scientifically and ethically defensible. As of now, the ACL action under the Trump Administration appears as an attempt to set back the improved public health justice achieved in the last decade.