Data for an AAA Area Plan Needs Assessment (2026 Guide)

An AAA area plan needs assessment should use ACS 5-year data on older adults, CMS and HRSA service locations, and local unmet-need data. Federal rules set the minimum, and your state unit on aging sets the format, the definitions and often the data list.

What the Older Americans Act requires

ACL's 2024 update to the Older Americans Act (OAA) regulations took effect March 15, 2024. Regulated entities had until October 1, 2025 to comply. The rule was published in the Federal Register on February 14, 2024 and revised 45 CFR Part 1321.

Area plan rules in 45 CFR 1321.65

Section 306(a)(1) has area plans weigh the number of older individuals in several groups. These are people with low incomes, in greatest economic or social need, at risk for institutional placement, and who are Indians.

Section 306(a)(4)(A) requires specific objectives for low-income minority older individuals, those with limited English proficiency, and rural residents. The Act defines an older individual as someone 60 years of age or older.

Greatest economic and social need (45 CFR 1321.3)

Section 1321.3 defines greatest economic need as income at or below the federal poverty level, as further defined by State and area plans. Those plans may weigh local factors, including geography and expenses.

Greatest social need is need caused by noneconomic factors. The listed factors are physical and mental disabilities, language barriers, and cultural, social or geographical isolation, including isolation due to:

• racial or ethnic status, Native American identity, or religious affiliation
• sexual orientation, gender identity or sex characteristics, or HIV status
• chronic conditions
• housing instability, food insecurity, lack of reliable clean water, lack of transportation, or utility assistance needs
• interpersonal safety concerns
• rural location
• any other status that restricts routine daily tasks or threatens independent living

State and area plans may also add other needs based on local and individual factors.

Each State plan must state how the State defines both terms, and those definitions must include the § 1321.3 populations. Read your state's definition before you pick income cutoffs.

What three states ask for

State guidance varies more than the federal rule:

• New Mexico. The SFY 2026–2030 area plan guide asks for one table per county with total and 60+ population. Its 60+ rows cover race and ethnicity, poverty, low-income minority, limited English, rural, Alzheimer's, living alone, kinship caregivers, disability, veterans and institutional risk. Its "LGBT" row has no age qualifier.
• South Carolina. The 2026 area plan program instruction requires "objectively collected and statistically valid data" but names no data source. It requires a readiness assessment for anticipated change in the older population over the next 10 years, and suggests an 85+ analysis. AAAs must hold at least three in-person public participation events and comply by July 1, 2027.
• California. The Department of Aging's planning page posts its 2024–2028 Area Plan Guidance, a Needs Assessment Guidance document and a 2025 service data dictionary. CDA also publishes key characteristics of Californians 60 and over by county and planning and service area. Its fields include lives alone, low income, minority, non-English speaking and rural.

Data elements, sources and tables

The Census tables below are ACS 5-year Detailed Tables plus Subject Table S0102, with labels checked in the Census Data API. The 2020–2024 estimates were released January 29, 2026 for areas down to the census tract for all tables. ZCTAs are published with 2020 Census boundaries.

Notes on the service files:

• The CMS home health ZIP file has one record for each ZIP code where the agency served at least one patient.
• As of November 2026, CMS's latest nursing home Provider Information file was released October 8, 2026.
• PACE enrollees must be at least 55, live in a PACE service area, and need a nursing home level of care. CMS's MA Plan Directory lists a contact for each PACE organization.
• The Eldercare Locator offers an API for third-party developers, with registration, and a raw data file of its core tables.

To view these layers together, GeoChat AI has a free tier with no credit card required.

Its layers include CMS nursing home, home health, hospice and PACE data, HRSA HPSAs, and ACS 2020–2024 estimates. Sources are rebuilt by hand after each release, so check dates on our methods page.

How to map older adults living alone far from services

1. Use census tracts as the base unit. The 5-year release covers tracts for all data tables. Label maps with town names, and add ZCTA views only when partners work in ZIP codes.
2. Pull living-alone counts. For each tract, download the B09021 rows "65 years and over" and "65 years and over: Lives alone," with margins of error. Map both the count and the share.
3. Test reliability. Published ACS margins of error use a 90 percent confidence level. Divide each MOE by 1.645 to get the standard error. Then compute the coefficient of variation, and flag tracts above your cutoff.
4. Build the service layer. Geocode senior centers, meal sites and case management offices from your contract records. Add CMS home health and nursing home addresses and Eldercare Locator sites.
5. Measure travel time. Draw drive-time areas around each site. The free openrouteservice plan allows 500 isochrone requests per day, as of November 2026. Record your threshold in the methods; see drive-time access.
6. Overlay need and distance. Flag tracts with many older adults living alone whose centers fall outside every drive-time area. Use B25045 no-vehicle households and RUCA codes as second filters.
7. Check home health reach separately. Count the agencies listing each ZIP code in the CMS ZIP file. A listed ZIP shows past service, not open capacity.
8. Test the map with residents. Bring flagged areas to public sessions and ask where people go for help. Record the answers as primary data for your § 1321.65(b)(3) conclusions on unmet need.

Common mistakes

• Labeling 65+ data as the OAA population. The Act counts people 60 and older, but Detailed Tables B09021, B11007, B18101, B18106, B18107, B16004 and B25045 split at 65. B17024 uses 55–64, 65–74 and 75+, so label those indicators by their real ages. For 60+ living alone, disability and limited English, use Subject Table S0102.
• Ignoring margins of error in small areas. Tract estimates for small groups, such as grandparents raising grandchildren, can be unreliable. The Census Bureau suggests aggregating geographic areas for larger samples.
• Treating ZIP codes as areas. ZIP Codes are point-based; ZCTAs are generalized areal representations, and not every ZIP Code has a ZCTA. HUD's ZIP–tract and ZIP–county crosswalks are updated quarterly. See ZCTA vs. ZIP code.
• Stopping at the county. County tables hide which towns are far from services. Analyze tracts, then roll results up for county templates such as New Mexico's.
• Comparing overlapping or mixed data years. The Census Bureau says not to compare overlapping 5-year datasets, and to compare 5-year estimates only with 5-year estimates. Note each CMS file's release month beside the ACS years.
• Counting registrations as staff. NPPES provider counts are registrations, not full-time clinicians; see provider-to-population ratio.
• Treating ACS disability as ADL need. ACS self-care difficulty means difficulty bathing or dressing. Under the Act, "at risk for institutional placement" means being unable to perform at least 2 activities of daily living without substantial assistance. It also requires a State determination that the person needs long-term care facility placement, so call the ACS figure a proxy.

Next steps

1. Read your state's current area plan guidance, and copy its definitions of greatest economic and social need into your methods.
2. List each § 1321.3 population with the table or local source you will use. Mark groups, such as HIV status or interpersonal safety, that only primary data can cover.
3. Download 2020–2024 ACS tables for every tract in your planning and service area, with margins of error, and set a CV cutoff.
4. Download the CMS, HRSA and Eldercare Locator files, and record each file's release date.
5. Draft the living-alone and travel-time maps before the 30-day comment period opens.
6. For county examples using ACS and HRSA data, see healthcare access in 10 California counties.
7. To explore ZIP codes and compare regions in plain language before building your own GIS, try GeoChat Atlas; the platform is new.