How Cuts to Health Surveys Impact Data on Children and Families

By Robin Ghertner, MPP

How Cuts to Health Surveys Impact Data on Children and Families

By Robin Ghertner, MPP, Founding Director of Strategic Policy Intelligence

Here at the Wonk, we love data.

You’ve seen it in the Data Drops we’ve put out. It reflects something deeper about our work; separating signal from noise to lift up what matters in policy.

If you’re not in the weeds of federal data but want to know how it shapes policy and decisions, this piece is for you.

I’m going to lay out some of the context for current cuts to federal health surveys and why those cuts matter for those of us working on child and family issues.

Before I joined the Wonk to oversee our data and analytic work, I spent 15 years leading teams in several federal agencies focused on using data to improve decision making.

I’ve seen numerous changes – mostly improvements – in how the federal government builds data systems.

The Stakes

You already know that reorganizations at the Department of Health and Human Services (HHS), staffing cuts, and budget pressures are impacting child and family policy.

What you may not know is that those actions will lead to changes to survey data that inform our understanding of children’s health and wellbeing. Here are a few examples.

Moving Key Health Data Offices to the Office of the Secretary

This spring, HHS proposed reorganizations involving two of the main sources of high-quality health data.

In May HHS proposed moving the National Center for Health Statistics (NCHS) out of CDC to the Office of the Secretary – one of the biggest potential changes to health data.

NCHS is in charge of national surveys and the National Vital Statistics System. Among many uses, NCHS data inform child fatality reviews, disability statistics, family formation trends, and child welfare risk factors.

Earlier in March, HHS planned to also move the Agency for Healthcare Research and Quality (AHRQ; wonks pronounce it “ark”) to the Office of the Secretary. AHRQ supports science and improved health care practices.

It also runs the Medical Expenditure Panel Survey (MEPS), which we and others have used to understand psychotropics, mental health, and other issues child welfare populations face.

These moves could change the entire orientation of NCHS and AHRQ.

These offices are supposed to produce objective data that are measured consistently over time – meaning you can trust that a number for 2025 is directly comparable to a number for 2015.

The Office of the Secretary is political by definition – this is where political appointees make decisions about funding, regulations, and other policies.

NCHS and AHRQ would be directly subject to the whims of changing political leadership, no longer having the same autonomy to determine their own priorities.

The data they produce may lose consistency and objectivity, which in turn could affect the decisions federal and state governments make.

Staff Cuts to the Most Important Behavioral Health Survey

The National Survey on Drug Use and Health (NSDUH), run by the Substance Abuse and Mental Health Services Administration for decades, tracks mental health and substance use, treatment access, and many other things.

It gathers essential information for anticipating pressures on child welfare systems and shaping prevention efforts that drive policy.

The NSDUH is housed in SAMHSA’s Center for Behavioral Health Services and Quality.

The NSDUH lead and the entire office that runs it were let go in April’s Reduction in Force.

Without staff to lead the survey effort, it’s questionable whether it can continue without significant data problems.

Gold Standard for Maternal and Infant Health in Limbo

At the end of January, the Pregnancy Risk Assessment Monitoring System (PRAMS) was put on hold, putting at risk both current availability of data and future collections (Handler et al., 2025).

PRAMS is a main source for research on maternal and infant health, run out of CDC’s National Center for Chronic Disease Prevention and Health Promotion.

Cutting Surveys to Save Money is Not New

Survey data collection is expensive.

For years the federal government has looked to save money by being smarter about the data it collects. Surveys are often one of the targets for cost efficiencies.

Over the past 15 years, I’ve been in budget debates with leadership in multiple agencies over the future of supplemental survey questions to the Current Population Survey.

These supplements cover a range of topics, from civic engagement to child support. When agency budgets are thin, it can be hard to justify spending money on these surveys, especially when other data sources look like they can fill the same needs.

Our health and human services systems automatically collect massive amounts of data through administrative records and insurance claims.

It seems reasonable that we can save money by using these data, rather than paying for surveys.

In truth, surveys are experiencing lowering response rates (Czajka and Beyler, 2016), making it harder and more expensive to get enough people to respond to make the results meaningful.

Surveys Feed Research and Drive Policy

To a non-research crowd this might seem like semantics, but these complementary approaches do different things.

I won’t rehash what a lot of smart people have written on the pros and cons of surveys relative to other data sources (Graeff and Baur 2020).

What I think matters most for child welfare is that survey data are intentionally designed to be of high quality for research.

Other sources - case records, claims, or third-party commercial data – are built for reporting, operations, or billing, not for research.

While these other sources are extremely valuable for many research topics, they have limitations.

For example, analyzing foster care records misses everyone who is not in foster care, meaning you have no one to compare to. In health claims data, if information is not needed for billing purposes, it may be incomplete, inaccurate, or outright wrong.

Administrative or claims data can also be biased against specific populations based on how they interact with the systems (including the service providers) producing the data.

Surveys can overcome all of these issues. Yes, they aren’t perfect, but they are not obsolete.

Getting Insight on Substance Use in Child Welfare Populations Is an Example of Why We Need Surveys

Parental substance use is one of the key drivers of child welfare system involvement. The Children’s Bureau reported that 32% of kids in FY2023 entered foster care because of parental substance use.

That analysis is based on the Adoption and Foster Care Analysis and Reporting System (AFCARS), a mandatory federal data collection used to study trends in foster care. But that number is most certainly an undercount.

There are numerous problems with how AFCARS collects substance use data, related to how state systems collect the data.

These problems lead to underreporting and inconsistency across states. (Give me some time and I’ll get my act together to publish something on it).

AFCARS also only includes kids in child welfare systems. If you want to know whether substance use has gotten worse for families in child welfare systems, you need to compare them to families not in the system.

We have to rely on surveys like the NSDUH, the NCSH, and National Survey of Child and Adolescent Well-Being to reliably measure substance use trends.

This isn’t just about research. Back in the early 2010s, we would have missed the full signal that the opioid crisis was pushing more kids into foster care if we just relied on AFCARS.

What's at Stake in Child Welfare Policy

Child welfare agencies face so many challenges today that solid data help address.

Administrators, providers, and families want to know if interventions work. They want to know if certain groups are being excluded from services.

They want to know how to prioritize strapped budgets.

Pick the topic that you are working on right now, and I can point to a health survey that can inform it.

Do you want to know how mental health and treatment differ for children with experience in foster care? Turn to the National Survey of Children’s Health, which we did recently.

Do you want to know how psychotropic use differs between children on Medicaid and those not, to design interventions before children enter foster care? That’s something we looked at with the MEPS.

Do you want to know which behavioral health issues homeless youth are facing and how it may have changed? The Youth Risk Behavioral Surveillance System can give you insight.

Understanding many child welfare policy issues relies on surveys run by CDC, SAMHSA, and AHRQ.

If those agencies face cuts or reorgs that diffuse their mission, those surveys may either go away or lose their value.

It may not happen overnight.

But losing these data sources would raise questions about where decision makers on child and family issues can turn for the insights they need.

References

Handler, A. S., Johnson, K., Rankin, K. M., Velonis, A. J., James, A. R., & Kotelchuck, M. (2025). Shuttering the Pregnancy Risk Assessment Monitoring System (PRAMS): A Dangerous Attack on US Mothers and Infants. American Journal of Public Health, 115(6), 848–850. https://doi.org/10.2105/AJPH.2025.308107

Czajka, J. L., & Beyler, A. (2016). Declining Response Rates in Federal Surveys: Trends and Implications. Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services. https://aspe.hhs.gov/sites/default/files/private/pdf/255531/Decliningresponserates.pdf

Graeff, P., & Baur, N. (2020). Digital Data, Administrative Data, and Survey Compared: Updating the Classical Toolbox for Assessing Data Quality of Big Data, Exemplified by the Generation of Corruption Data. Historical Social Research / Historische Sozialforschung, 45(3), 244–269.