County data desk
Where is caregiver supply thinnest? Jefferson County, at 0.07 businesses per 1,000 older adults.
A ratio that neither source agency publishes: home care establishments divided by residents aged 65 and over, county by county, sorted so the thinnest coverage comes first.
Methodology
- Metric
- Private home care establishments per 1,000 residents aged 65 and over, sorted ascending. Fewest businesses per older adult first. This ratio is computed by this desk and is not published by either source agency.
- Numerator
- U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, 2024 Q4, county file, private ownership (own_code 5), NAICS 621610 home health care services, column qtrly_estabs. Employment is column month3_emplvl.
- Denominator
- U.S. Census Bureau, Population Estimates Program, Vintage 2024, column AGE65PLUS_TOT, rows coded YEAR 6, reference date July 1, 2024.
- Pulled
- Both files September 10, 2026. The QCEW county files were read at data.bls.gov/cew/data/api/2024/4/area/.
- Industry caveat
- NAICS 621610 mixes medical and non-medical home care into one industry code. A Medicare certified skilled nursing agency and a company that sends a caregiver to help with bathing, meals and errands are counted in the same column. This ratio measures registered business presence, not the availability of any particular service.
- Missing rows
- Five counties have no NAICS 621610 row in the 2024 Q4 file at all. QCEW publishes an establishment count even when it withholds employment for confidentiality (disclosure code N), which is visible three times in this very table, so an entirely absent row reads as nothing reported rather than as something suppressed. Those counties are shown as no row published and no ratio is computed for them. They are placed at the head of the ascending order because they represent the thinnest end of the range, not because a zero was measured.
- Licence
- Public domain, works of the United States government.
| # | County | State | Establishments | Employment | Aged 65 and over | Per 1,000 |
|---|---|---|---|---|---|---|
| 1 | Calhoun County | Illinois | no row published | withheld | 1,077 | not computed |
| 2 | Bond County | Illinois | no row published | withheld | 3,272 | not computed |
| 3 | Jersey County | Illinois | no row published | withheld | 4,663 | not computed |
| 4 | Monroe County | Illinois | no row published | withheld | 7,471 | not computed |
| 5 | Macoupin County | Illinois | no row published | withheld | 9,911 | not computed |
| 6 | Jefferson County | Missouri | 3 | 86 | 41,443 | 0.07 |
| 7 | Warren County | Missouri | 1 | withheld | 7,504 | 0.13 |
| 8 | Clinton County | Illinois | 1 | withheld | 7,320 | 0.14 |
| 9 | Lincoln County | Missouri | 2 | withheld | 10,226 | 0.20 |
| 10 | Franklin County | Missouri | 6 | 164 | 21,777 | 0.28 |
| 11 | Madison County | Illinois | 15 | 419 | 51,448 | 0.29 |
| 12 | St. Charles County | Missouri | 35 | 524 | 76,887 | 0.46 |
| 13 | St. Clair County | Illinois | 27 | 1,153 | 46,500 | 0.58 |
| 14 | St. Louis County | Missouri | 242 | 5,023 | 199,345 | 1.21 |
| 15 | St. Louis city | Missouri | 77 | 949 | 46,992 | 1.64 |
Regional totals: 409 establishments reported across the ten counties with a published row, against 535,836 residents aged 65 and over, a regional rate of 0.76 per 1,000. Employment shown as withheld is suppressed by BLS under disclosure code N; the establishment count in the same row is still published.
The shape of the gap
Two jurisdictions hold most of the industry. St. Louis County reports 242 establishments and the city of St. Louis 77, so 319 of the 409 establishments in the region sit in those two rows, against 46 percent of the region's older population. The city posts the highest rate in the table at 1.64 per 1,000, St. Louis County the second highest at 1.21.
Everywhere else the rate falls by an order of magnitude. St. Charles County, with 76,887 residents aged 65 and over and the second largest older population in the region, reports 35 establishments and a rate of 0.46. Jefferson County reports three establishments against 41,443 residents aged 65 and over, a rate of 0.07, the lowest computed figure on this page by a wide margin.
The outer counties are thinner still. Warren County reports one establishment against 7,504 residents aged 65 and over. Clinton County, Illinois, reports one against 7,320. Lincoln County reports two against 10,226. Five Illinois counties, together holding 26,394 residents aged 65 and over, have no row for this industry code at all.
The employment column tells a second story
Establishment counts describe how many businesses exist. The employment column describes how large they are, and the two do not track. St. Clair County, Illinois, reports 27 establishments and 1,153 employees in the third month of the quarter, an average near 43 per establishment. Madison County, Illinois, reports 15 establishments and 419 employees. St. Charles County reports 35 establishments and 524 employees, an average near 15.
Three counties in this table have their employment withheld under BLS disclosure code N, which the agency applies when publishing a figure could identify an individual employer. In each of those three the establishment count is still published, and that asymmetry is the reason this desk reads an entirely missing row as nothing reported rather than as a withheld figure.
What a low rate does and does not mean
A county with few establishments is not necessarily a county without providers. QCEW counts establishments where they are registered as employers, and a home care company headquartered in one county routinely sends caregivers into several. A resident of Warren County may be served from an office in St. Charles County without that ever appearing in the Warren County row.
Nothing here measures quality, licensure, capacity or whether any particular provider is accepting new clients. It measures registered business presence against resident population, and that is all. The practical step for a family is to check a provider's own published service area against their address, then call.
Both source files, their column codes and their limits are set out in full on the methodology page. The population side of the join is published separately on the county ranking page.
Back to the region at a glance · Sources, vintage and limits