The U.S. Bureau of the Census is considering a change to how it tracks disability, which would lead to significant drops in estimates of how many Americans are disabled. The change may result in cuts to funding and resources for disabled communities, experts say, including the millions of Americans with Long Covid.
If the change is implemented, the census would switch from asking Americans about their disabled status with a series of yes or no questions to using scaled questions. To determine who is disabled, the agency would count only respondents who say they have significant difficulty completing basic tasks. With this shift, bureau researchers estimate that the share of Americans with a disability, based on their data, would drop from 13.9% to 8.1%.
“The national estimates of people with disabilities will drop by 40%,” said Jean Hall, director of the Institute for Health and Disability Policy Studies at Kansas University. She also co-authored a letter, along with other researchers who study disability and have lived experience of disabilities, asking the Census Bureau not to go through with the change. “The number of people with disabilities in our country is actually growing, and this change will make it look like there’s many fewer people,” Hall said.
Advocates for people with Long Covid and other post-infectious chronic diseases have similarly sounded the alarm about this proposed change, citing the importance of census data for research and government programs. “I can’t overemphasize how much the census is used in, not just federal policymaking and budgeting, but state-level budgeting,” said Angela Vazquez, former president of the Long Covid advocacy group Body Politic. Time and resources go into ensuring other marginalized communities are “properly counted,” she said, adding that the same should be true for disabled people.
While disability researchers and advocates are calling on the Census Bureau to cancel the proposed change, they also say the current questions, used to estimate disability in the U.S. for the last 15 years, are far from perfect. Both sets of questions fail to include all people with chronic illnesses like Long Covid, people with intellectual and developmental disabilities, people with mental illnesses, and many others, said Bonnie Swenor, director of Johns Hopkins University’s Disability Health Research Center and another author on the letter.
The Census Bureau needs to “find better ways to include the disability community” across its programs, Swenor said. The letter she co-authored calls for the agency to improve engagement with disabled people and scientists studying disability — echoing similar calls for better patient engagement in biomedical research. Poor data collection, such as the change currently under consideration, would hide the challenges that people with disabilities face, she said.
In response to questions from The Sick Times, a spokesperson from the Census Bureau said that the agency is currently seeking comments from the public to “inform a final decision” about the proposed change. The spokesperson also noted that the bureau “regularly engages with disability researchers and data users” through events such as conference presentations and a disability research committee that coordinates across federal agencies.
A small question change with major implications
The Bureau of the Census is the federal agency responsible for tracking the U.S. population. It runs the Decennial Census every 10 years, which seeks to count every U.S. resident, along with more frequent programs that use data from smaller groups of Americans to estimate national patterns.
Since 1830, the decennial census has included questions about disability. In 2008, the agency started using its current question framework, called the ACS Questions after the annual American Community Survey. The framework includes a series of six yes or no questions asking whether the respondent has “serious difficulty” seeing, hearing, concentrating or making decisions, walking or climbing stairs, dressing or bathing, and doing errands alone. Respondents who answer “yes” to any question are counted as disabled.
Now the bureau has proposed switching to a different set of six questions, potentially starting with the American Community Survey in 2025. The question topics would stay mostly consistent, with a new question about communicating replacing the question about errands. But rather than answering yes or no, respondents would answer on a scale: no difficulty, some difficulty, a lot of difficulty, cannot do at all. Per the bureau’s proposal, only respondents who say they have “a lot of difficulty” or “cannot do at all” a given task are counted as disabled.
In defending this change, federal scientists presenting at an advisory committee meeting on November 16 explained that the new questions better align with international standards of measuring disability. A spokesperson for the Census Bureau similarly told The Sick Times that the proposal “reflects advances in the measurement of disability” and is “conceptually consistent” with a research framework used by the World Health Organization. While the current ACS Questions were developed specifically for the U.S., the new questions were developed by an international consensus group and are used by demographic agencies in many other countries.
However, disability researchers and advocates argue that switching the questions — and counting only people with more severe limitations as disabled — would have drastic implications. Census researchers compared the two measurement frameworks during a test in 2022 and reported that the proposed change would identify 8.1% of the U.S. population as disabled, down from 13.9% with the current framework.
Many people counted as disabled under the current framework would suddenly be considered “non-disabled,” Swenor said. While the scaled response options do provide more information about respondents’ abilities, the cutoff at “a lot of difficulty” would cut many people out of the disabled classification. This may be particularly true for people with chronic, post-infectious diseases like Long Covid, which often have fluctuating symptoms, said Jaime Seltzer, head of scientific and medical outreach at the ME/CFS advocacy group ME Action. “At least in ME/CFS, we have found that people underplay their troubles,” she said.
One alternative proposal is to change the cutoff and include people who report “some difficulty” in response to at least one question as disabled. According to the Census Bureau’s testing, such a change would increase the estimate of Americans with disabilities to 31.7%. To researchers asking the bureau to reconsider, however, this may still be a poor solution, as the questions’ construction would still allow policymakers to access and interpret scaled data in other ways.
“The only times, historically, that I know the U.S. government has ever wanted to increase the size of the disabled population is when they either wanted to sterilize us or institutionalize us,” said Scott Landes, a sociologist at Syracuse University and co-author on the letter. The problem with the proposed census questions, he said, is that government officials and policymakers would “get to decide who counts as disabled” — and they could make that decision with an eye toward cutting funding for disability services.
The Census Bureau’s choices around measuring disability have major ramifications for disabled people across the U.S., Landes and other experts say. Census data are the basis for federal, state, and local policies, for funding decisions, for designing programs that support people with disabilities, and more. For example, census data may be used at a local level to identify how many people need accessible housing. If the proposed change goes through, the census would “decrease the number of apparent disabled people,” therefore “essentially arguing for budget cuts to disability services and programs,” Seltzer said.
Such a change would create particular challenges for researchers and advocates trying to understand how many people have become disabled since the start of the Covid-19 pandemic, as it would make data collected before and after 2025 not comparable — and make analyses like the recent New York Times story exploring cognitive issues more difficult. “We know that people with Long Covid are falling through the cracks as far as things like qualifying for federal disability programs,” Hall said. “Without knowing who they are, and how big of a problem that is, it’s not likely to change.”
The only times, historically, that I know the U.S. government has ever wanted to increase the size of the disabled population is when they either wanted to sterilize us or institutionalize us.
Scott Landes
Advocating for better data on disability
Researchers and patient-advocates are calling on the Census Bureau not to go through with its proposed question change. But they don’t want the census to keep using the same ACS questions, either, as those questions exclude Americans with disabilities whose conditions fall outside the narrow buckets prescribed by the questions, including many with Long Covid.
At the Census Bureau’s advisory committee meeting on November 16, Susan Popkin, co-director of the Urban Institute’s Disability Equity Policy Initiative, presented findings from a survey using the census’ current questions, conducted in collaboration with the Patient-Led Research Collaborative. Among adults under age 65 who identified as having severe Long Covid symptoms, only 60% would count as disabled using the census questions, Popkin reported. The finding shows a “disconnect” between the agency’s data-collection methods and the experiences of people with Long Covid, she said.
Researchers and Long Covid advocates agree that better census questions are needed to capture people with Long Covid as well as those with different post-infectious chronic diseases and other types of disabilities. The current questions are focused on “functional limitations” to respondents’ day-to-day lives, Landes said, which only capture one dimension of disability.
Other options might include questions that are more focused on demographic identity, such as “Are you disabled, yes or no?” followed by additional questions that provide more specificity. Vazquez would like to see questions asking about accommodations that respondents use and their capacity for different activities. Seltzer would like to see cumulative measurements that add up multiple disabilities in the same respondent.
Significant work would be needed to devise new questions that truly capture the disability community in the U.S., Swenor said. The Census Bureau would need to bring together a variety of disability groups, actively seek out different perspectives, and listen to feedback, she said. In their letter, Swenor and her co-authors propose that the Census Bureau consider disabled scholars and advocates for its advisory boards and committees. They also recommend that the agency support a national task force that would improve disability data collection “across the federal government.”
People with disabilities had no input on the proposed census change before it was publicly announced, according to the researchers’ letter, mirroring challenges with patient engagement for biomedical research. But there’s still time to provide input before the Census Bureau implements a new framework that could lead to a 40% cut in disability estimates.
Here’s how you can get involved:
- Write a comment for the Federal Register by December 19.
- Write a letter directly to the Census Bureau.
- Contact your Congressional representatives about the change.
- Apply to join public committees on data collection, either at federal agencies or at the state or local level.
- Get involved with disability advocacy groups working toward better data collection.
Editor’s note, November 21, 7:30 pm EST: This article has been updated with comments from the Census Bureau.
All articles by The Sick Times are available for other outlets to republish free of charge. We request that you credit us and link back to our website.
Leave a Reply