A proposed redesign of the federal government's annual National Health Interview Survey would remove questions about disabilities, a change that advocates say will cause significant undercounting of Americans with intellectual and developmental disabilities.

The National Health Interview Survey, run by the federal government on an annual cycle, has historically undercounted people with disabilities even before the proposed changes. That baseline matters for any research organization that uses NHIS data as a sampling frame, a benchmark, or a denominator in disability health studies. Removing the relevant questions does not merely trim a module; it removes the variable itself from future public-use datasets.

Katy Neas, CEO of The Arc, an advocacy group for people with intellectual and developmental disabilities, called the redesign a "head-scratcher." Her objection goes to the core of what a federal statistical system provides: a trusted, population-level measurement layer that no single academic center or private vendor can replicate.

"If we don't have the government as a trusted source of data, asking questions about real people and what their real experiences are, we're never going to get better [health outcomes]," Neas said.

What is at stake for researchers

The National Health Interview Survey is one of the federal government's principal instruments for measuring population health. For R&D managers and principal investigators whose portfolios include disability epidemiology, health services research, or disparities work, the redesign raises three immediate questions.

First, comparability. Any longitudinal analysis that tracks disability prevalence or health outcomes across survey years will face a structural break once the questions disappear. Researchers will need to decide whether to truncate time series, switch instruments, or model the discontinuity directly — each option carries cost and methodological risk.

Second, replacement data. No alternative currently identified in the proposal's public discussion matches NHIS coverage. Advocates' assessment that the survey "historically undercounted" people with disabilities suggests the existing instrument was already imperfect. The redesign replaces a flawed measurement with, in their view, no measurement at all — a worse starting point for any study design that depends on nationally representative disability estimates.

Third, downstream funding decisions. Federal and foundation grant portfolios that allocate resources by disability prevalence depend on surveillance numbers. If the undercounting that advocates predict materializes, budget allocations built on those figures will systematically underserve the population in question. The data pipeline and the funding pipeline are not independent.

Claims and caveats

The assertion of "significant undercounting" comes from advocacy organizations, not from an independent methodological review. The Arc has an institutional stake in the visibility of the population it represents, and its assessment of the redesign's impact should be read in that light. At the same time, the group's claim rests on a documented baseline — the survey's historical undercount of people with disabilities — that is not in dispute in the source material. What remains unquantified is how large the incremental undercount from the redesign would be, since no sample sizes, sensitivity analyses, or Census Bureau or CDC methodological justifications appear in the available reporting.

The Centers for Disease Control and Prevention, which administers the survey redesign process, has not yet provided its full rationale in the material available. Researchers weighing the change should watch the formal federal register comment period and any methodological appendices that accompany the redesign, which would show whether the dropped questions were replaced by proxy measures elsewhere in the instrument.

What comes next

The proposal remains at the redesign stage, which typically means an opportunity for public comment before final implementation. Advocates including The Arc are making their case now, before the changes take effect. Whether the CDC restores the disability questions, substitutes an alternative measurement approach, or proceeds as drafted will determine whether the next NHIS cycle can still support credible disability health research at the national level.