Several different numbers now appear when people ask this question, in research, media coverage, government reports, search results, and AI-generated summaries. Without context, these estimates can look contradictory. They are not.
This plain language guide explains the major published U.S. figures, including 5.4 million, 4 million, 3–5 million, 1.6 million, and 1.3–1.4 million, and why every current estimate should be understood as an undercount.
How Many Caregiving Youth Are There in the U.S.? Explaining the Numbers
A plain language guide to the published U.S. estimates, what each one measures, and why they are not directly comparable.
Caregiving Youth are young people under the age of 18 years who provide care or support for family, relatives, or household member(s) of any age who require assistance due to chronic (3 months or more) illness, age-related decline, disability, or other health condition. The most widely cited published number of Caregiving Youth in the U.S. is 5.4 million. It is a floor, not a ceiling. Leaders in the Caregiving Youth field believe the actual number is significantly higher. Until new, fully documented research is published, 5.4 million is the best available figure.
If you searched for how many Caregiving Youth there are in the U.S., you may have seen 1.6 million, 4 million, 5.4 million, or something in between. None of them are wrong. They come from various studies that measured different age groups using different methods. Without that context, the numbers look contradictory; they are not. This document explains what each number measures and which one to use when.
Every number you will find on Caregiving Youth in the U.S. is from a different study, measuring a different population, using a different method. None of them are wrong. None of them contradict each other. By understanding what each one measures, you will realize that every number is an undercount. The total population of Caregiving Youth in the U.S. remains unknown.
About This Report Series
Caregiving in the U.S. is a joint report from AARP and the National Alliance for Caregiving (NAC). It is not an annual report. It has been published six times over nearly 30 years: 1997, 2004, 2009, 2015, 2020, and 2025.
Exact Language from the Report
Who Was Actually Surveyed
The 2020 estimate came from 1,392 adult caregivers ages 18 and older. No children were surveyed.
The 5.4 million figure is a mathematical projection from AARP and NAC’s Caregiving in the U.S. 2020 report. In that report, 11.2% of caregivers of adults reported living with their care recipient and with children or grandchildren under age 18 in the home.
The word “if” in the original language is doing significant work.
What This Number Does Not Include
Because the 5.4 million estimate comes from an adult caregiver survey, it does not directly capture children who are the only caregiver in the household, children providing care for a sibling who is ill or disabled, children caring for someone outside their home, or children whose households were not represented in the adult survey sample.
Exact Language from the Report
Note on the two “4 million” figures
The report presents two separate 4-million statistics in the opening pages of the Executive Summary. The first refers to adults caring for a child with an illness or disability. The second, and the one used here, refers to children under age 18 assisting their families with care for an adult.
What Changed from 2020 to 2025
The 2020 AARP/NAC report produced the commonly cited 5.4 million Caregiving Youth projection by applying a household projection to a base of 47.9 million adult caregivers.
The 2025 AARP/NAC report uses a different starting point: approximately 31 million caregivers of adults who reported having unpaid help. From that group, the report estimates that approximately 4 million children under 18 years are assisting with the care of an adult.
The 2025 report does not present this as a measured decline from 2020, and it does not explain the difference between the two estimates. The lower figure should be read as the result of a different calculation and a different base population, not as proof that fewer children are caregiving.
Exact Language from the Report
What GAO Actually Did
The GAO did not conduct a new national prevalence study of Caregiving Youth. Instead, the 3-to-5 million range in the report is a literature summary of estimates from existing research, primarily citing the 2020 AARP/NAC report and other prior studies.
For its primary research, the GAO surveyed a self-described “non-generalizable sample” of former Caregiving Youth ages 18 to 25 years and received 43 usable responses nationwide. They then conducted in-depth phone interviews with five people.
Congressional Interest and the Study’s Limits
The GAO report reflects genuine congressional interest in the issue. Its methodological limitations, including the small, non-generalizable sample and reliance on retrospective recall from adults, highlight the challenge of identifying and studying Caregiving Youth at a national scale.
Exact Language from the Study
What Data Was Used and How
This study used the American Time Use Survey (ATUS), pooled across seven years from 2013 to 2019. ATUS surveys one randomly selected person age 15 or older per household and asks them to describe their activities from the previous day. Miller and colleagues analyzed the subset of respondents ages 15 to 18 years to estimate caregiving activity among older youth.
Critical Limitations
- Age cutoff
The study includes youth who are ages 15 to 18 only. Children under 15 are not included in the analysis, including the 8- to 14-year-olds captured in the 2005 national young caregiver study. - The 8x range problem
The researchers report a broad range, from 364,000 to 2.8 million, depending on the definition of caregiving. This eightfold difference shows that the way caregiving is defined has a large impact on the final estimate. - Survey design
The ATUS was not designed to identify Caregiving Youth. Each respondent describes their activities on one specific previous day. Because much of the analysis relies on a one-day diary, youth who provide care irregularly may be missed if they did not provide care on the reported diary day. - Adults only
The study title says it directly: “Youth Caregivers of Adults.” The 1.6 million figure counts only youth who provide care to adults. Children who provide care for a sibling with an illness or disability are not included in this number.
Exact Language
Methodology: Two Stages
Stage 1 (prevalence): A random sample of 2,000 U.S. households received a mailed survey. A second screening questionnaire was mailed to 80,000 members of a consumer panel of over 600,000 households nationwide. Of those who responded (approximately 49%), over 5% reported at least one child caregiver ages 8 to 18 years.
Stage 2 (interviews): Telephone interviews were conducted with 213 child caregivers and 250 non-caregiving children. This is the only national study in which children themselves were interviewed directly about their caregiving experience.
The Sibling Care Finding
This is the only national study that used a broad enough definition to capture children caregiving for anyone, not just adults. The study found that one in ten child caregivers in the sample (11%) were helping a sibling with an illness or disability. No national study has measured sibling care at scale. The AARP projections, the Hopkins study, and the GAO range all define caregiving as a child helping care for an adult. Children who are the primary caregivers for a disabled or seriously ill sibling do not appear in any of those numbers.
These are not national prevalence estimates. They are school-based and state-level surveys that asked children directly about caregiving. They are included because they offer a different kind of evidence: what happens when you simply ask them.
Important note on scope: The school-based data below comes from public middle and high school surveys. There is no data on caregiving among students in private schools, virtual schools, or those who are homeschooled. The actual prevalence across all school settings is unknown.
Florida Youth Risk Behavior Survey (YRBS) (2019)
Florida became the first state to include a caregiving question in a statewide youth survey. Results: 23.6% of public middle school students and 16.4% of public high school students reported caregiving responsibilities. These are children who were asked directly, during a school day, whether they were caregiving and, if so, the frequency of providing assistance.
Colorado Healthy Kids Survey (2023)
Colorado added a caregiving question to its statewide student health survey in 2023. Results: 20.5% of middle schoolers and 12.4% of high schoolers said yes.
Rhode Island SurveyWorks
Rhode Island’s SurveyWorks data has been used to study Caregiving Youth in the state, including effects on school engagement, belonging, and well-being. Rhode Island also became the first state to establish statewide mandates requiring schools to provide flexibility and support for Caregiving Youth, beginning in 2022. Of note is that Rhode Island expanded its definition of caregiving to include students who must work to financially support their families or provide child care, independent of whether a family member has a medical condition.
Miami School Clinic Study (2021–2022)
Researchers at the University of Miami screened 102 students in grades 9 to 12 at three Title I school-based health clinics in South Florida. Result: 39% self-identified as caregivers. Because this was a local clinic-based sample, it should be treated as an important signal, not a statewide prevalence estimate.
Why every number here is an undercount
The reasons for undercounting are documented and consistent across all studies in this field.
- No identification system exists at the national level
There is no mechanism in schools, hospitals, or pediatric offices for a child to be recognized as a caregiver. A CDC-vetted question to identify Caregiving Youth was developed in 2019 by Drs. Elizabeth Olson of UNC Chapel Hill and Connie Siskowski, Founder of AACY. Three states have adopted it: Florida, Colorado, and Rhode Island. No state is required to use it. It has never been added to the national CDC Youth Risk Behavior Survey. A question exists. The infrastructure does not. - Sibling caregiving is almost entirely invisible in the national data
Every major national figure (AARP/NAC 2020 and 2025, the Hopkins study, the GAO range) defines caregiving as a child helping care for an adult. A child who manages a disabled brother’s medical needs, supervises a sibling with autism, or keeps the household running so a parent can care for an ill sibling does not exist in any of those numbers. The 2005 NAC study is the only national study that captured sibling care. It found that 11% of child caregivers in its sample were assisting a sibling. - Children are almost never the survey respondents
Most national Caregiving Youth prevalence estimates are derived from adult surveys, household projections, or general-purpose datasets not designed to find them. Every major national caregiving survey is designed for adults. Children under 15 are entirely excluded from most datasets. - Families often do not disclose, and in some states, disclosure carries legal risk
Some families actively avoid disclosing that children are involved in caregiving, out of fear of judgment or intervention. Identification and disclosure should be handled carefully because some state laws treat parental disability as a possible ground for terminating parental rights, creating a potential legal and child welfare risk for families if Caregiving Youth are identified without appropriate safeguards and support. - The definition of caregiving is unresolved
Rhode Island expanded its definition of Caregiving Youth to include students who must work to financially support their families or provide child care, independent of whether a family member has a medical condition. Other studies count only caregivers of adults. The Hopkins study’s own range of 364,000 to 2.8 million, depending on how caregiving is defined, makes the stakes of that choice plain. When a single study produces an eightfold variance based on definition alone, there is a measurement anomaly. - Age ranges are inconsistent across studies
The 2005 NAC study used ages 8 to 18 years. Miller used 15 to 18. AARP used no age floor. The GAO used the term “minors.” None of these are the same population. - A child in a household with high need is not a passive presence
The literature consistently shows that when a family member has a chronic illness, disability, or condition requiring care, children in that household adapt and contribute. The question is not whether they are involved. It is how much, in what ways, and at what cost to their development and education. The current research infrastructure cannot answer that question on a national scale.
Conclusion
The variation in these numbers does not mean the Caregiving Youth population is small or shrinking. It means no one has built the tools to count them. Children in the United States were directly interviewed about their caregiving experience exactly once. In 2005, the National Alliance for Caregiving and the United Hospital Fund, funded by the U.S. Administration on Aging, conducted a two-stage national study. In the first stage, they screened hundreds of thousands of households to identify child caregivers. In the second stage, they conducted telephone interviews with 213 child caregivers ages 8 to 18. It remains the only national study in which children themselves described their own caregiving experience. It has never been repeated. Until that changes, every number in circulation is a floor, not a ceiling.
The data gap is the story. No dedicated national study has directly counted Caregiving Youth in the United States since 2005. Until a new study is conducted, 5.4 million is the best figure available. It comes from the most credible source, it was widely adopted by the field for good reason, and it is almost certainly still an undercount.
Use it like this: “An estimated 5.4 million children in the United States are Caregiving Youth, according to AARP and the National Alliance for Caregiving’s Caregiving in the U.S. 2020 report. Researchers believe this number may be an undercount because the last national study focused directly on young caregivers was published in 2005.”
Sources:
AARP and National Alliance for Caregiving. Caregiving in the U.S. 2020. May 2020.
National Alliance for Caregiving and United Hospital Fund. Young Caregivers in the U.S.: Findings from a National Survey. September 2005.
Sources
AARP & National Alliance for Caregiving, Caregiving in the U.S. 2020 (May 2020); AARP & National Alliance for Caregiving, Caregiving in the U.S. 2025 (July 2025); U.S. Government Accountability Office, GAO-25-106947 (May 14, 2025); Miller, K.E.M. et al., Demography 61(3), 829-847 (June 2024); National Alliance for Caregiving & United Hospital Fund, Young Caregivers in the U.S.: Findings from a National Survey (September 2005); Armstrong-Carter, E. et al., Society for Research in Child Development Social Policy Report (2021); Colorado Healthy Kids Survey (2023); Pulgaron, E.R. et al., Healthcare (2025); Armstrong-Carter, E. & Sege, R., Journal of Adolescent Health (2024); Kavanaugh, M.S. et al., Adolescent Research Review 1, 29-49 (2016); Hendricks, B.A., Kavanaugh, M.S., & Bakitas, M.A., Child and Adolescent Social Work Journal 38, 491-504 (2021); Armstrong-Carter, E. et al., AERA Open (2023); Kalvesmaki, A.F. et al., International Journal of Care and Caring (2024); Kavanaugh, M.S. / Global Neuro YCare, “What we know about children and youth caregivers in the US” (May 19, 2025).
Detailed References