Abstract
This cross-sectional study examines whether there is an association between county-level rates of food insecurity and the availability of food benefits through Medicare Advantage plans.
Introduction
Nearly 17% of Medicare enrollees experience food insecurity,1 a share projected to increase following recent cuts to the Supplemental Nutrition Assistance Program.2 Since 2020, Medicare Advantage (MA) plans—which now enroll more than half of the Medicare population—have been able to offer a Food and Produce supplemental benefit, allowing plans to provide targeted allowances to beneficiaries to purchase healthy foods.
Supplemental benefits have drawn scrutiny, given mixed and incomplete evidence on how they are used and to whom they are offered.3,4,5 We examined the characteristics of plans offering the Food and Produce benefit in 2025 and assessed whether enrollment in these plans matched county-level need for nutritional support.
Methods
For this cross-sectional study, data on county-level MA plan characteristics, including enrollment and Food and Produce benefit offerings, came from the Centers for Medicare & Medicaid Services (CMS) 2025 MA public use files. County-level estimates on the proportion of residents who were food insecure came from Feeding America’s Map the Meal Gap.6
We used 2-sample tests of proportion to assess for differences in the characteristics of plans offering the Food and Produce benefit relative to those not offering the benefit. We examined the county-level association between food insecurity and the proportion of MA beneficiaries enrolled in plans offering the Food and Produce benefit using ordinary least squares regression, sequentially adjusting for county-level plan characteristics to isolate whether plan availability or enrollment drove any observed associations. We defined significance as P < .05 with Bonferroni corrections for multiple comparisons.
This study was deemed not human participant research by the institutional review board at Weill Cornell Medicine and followed STROBE reporting guidelines for cross-sectional studies. Analyses were performed using Stata version 18.5 (StataCorp). See eMethods in Supplement 1 for additional details.
Results
The sample included 5105 MA plans across 3064 counties (median [IQR] food insecurity, 15.1% [12.8%-17.6%]); 33.8% of plans offered the Food and Produce supplemental benefit (Table). Plans that offered Food and Produce benefits, relative to those that did not, were more likely to be health maintenance organizations (difference, 22.8 percentage points [PP], 95% CI, 19.9-25.6 PP; P < .001), newer (contract effective date >2020: difference, 9.8 PP, 95% CI, 7.3-12.3 PP; P < .001), smaller (lowest quartile enrollment: difference, 7.6 PP; 95% CI, 5.0-10.2 PP; P < .001), and Special Needs Plans (SNPs; difference, 62.2 PP, 95% CI, 59.8-64.5 PP, P < .001). Among plans offering the benefit, 82.7% did not require prior authorization to access the benefit.
Table. Characteristics of Medicare Advantage Plans Stratified by Offering of Food and Produce Supplemental Benefits, 2025.
| Characteristic | Plans, No. (%)a | Difference (95% CI), percentage points | P value | ||
|---|---|---|---|---|---|
| With Food and Produce benefit | No Food and Produce benefit | Total | |||
| Total | 1727 (33.8) | 3378 (66.2) | 5105 (100.0) | NA | NA |
| Plan type | |||||
| HMO | 949 (55.0) | 1087 (32.2) | 2036 (39.9) | 22.8 (19.9 to 25.6) | <.001 |
| HMO-POS | 404 (23.4) | 762 (22.6) | 1166 (22.8) | 0.8 (−1.6 to 3.3) | .50 |
| PFFS | 1 (0.1) | 20 (0.6) | 21 (0.4) | −0.5 (−0.8 to −0.3) | .005 |
| Local PPO | 370 (21.4) | 1461 (43.3) | 1831 (35.9) | −21.8 (−24.4 to −19.3) | <.001 |
| Regional PPO | 3 (0.2) | 48 (1.4) | 51 (1.0) | −1.2 (−1.7 to −0.8) | <.001 |
| Plan parent organization | |||||
| Humana Inc | 214 (12.4) | 511 (15.1) | 725 (14.2) | −2.7 (−4.7 to −0.8) | .008 |
| UnitedHealth Group Inc | 266 (15.4) | 609 (18.0) | 875 (17.1) | −2.6 (−4.8 to −0.5) | .02 |
| CVS Health Corporation | 267 (15.5) | 428 (12.7) | 695 (13.6) | 2.8 (0.7 to 4.8) | .006 |
| Centene Corporation | 93 (5.4) | 226 (6.7) | 319 (6.2) | −1.3 (−2.7 to 0.1) | .07 |
| Elevance Health Inc | 186 (10.8) | 150 (4.4) | 336 (6.6) | 6.3 (4.7 to 7.9) | <.001 |
| Other | 701 (40.6) | 1454 (43.0) | 2155 (42.2) | −2.5 (−5.3 to 0.4) | .09 |
| Overall Star rating | |||||
| <3.5 | 223 (12.9) | 480 (14.2) | 703 (13.8) | −1.3 (−3.3 to 0.7) | .20 |
| 3.5 to 4.0 | 938 (54.3) | 1895 (56.1) | 2833 (55.5) | −1.8 (−4.7 to 1.1) | .23 |
| 4.5 | 306 (17.7) | 732 (21.7) | 1038 (20.3) | −4.0 (−6.2 to −1.7) | .001 |
| 5.0 | 20 (1.2) | 67 (2.0) | 87 (1.7) | −0.8 (−1.5 to −0.1) | .03 |
| Unavailable | 240 (13.9) | 204 (6.0) | 444 (8.7) | 7.9 (6.0 to 9.7) | <.001 |
| Contract effective date | |||||
| <2006 | 621 (36.0) | 1573 (46.6) | 2194 (43.0) | −10.6 (−13.4 to −7.8) | <.001 |
| 2006 to 2013 | 370 (21.4) | 748 (22.1) | 1118 (21.9) | −0.7 (−3.1 to 1.7) | .56 |
| 2014 to 2019 | 247 (14.3) | 433 (12.8) | 680 (13.3) | 1.5 (−0.5 to 3.5) | .14 |
| >2020 | 489 (28.3) | 624 (18.5) | 1113 (21.8) | 9.8 (7.3 to 12.3) | <.001 |
| Enrollmentb | |||||
| <576 Individuals | 519 (30.1) | 759 (22.5) | 1278 (25.0) | 7.6 (5.0 to 10.2) | <.001 |
| 576-1800 Individuals | 402 (23.3) | 874 (25.9) | 1276 (25.0) | −2.6 (−5.1 to −0.1) | .04 |
| 1801-5586 Individuals | 382 (22.1) | 893 (26.4) | 1275 (25.0) | −4.3 (−6.8 to −1.9) | .001 |
| >5586 Individuals | 424 (24.6) | 852 (25.2) | 1276 (25.0) | −0.7 (−3.2 to 1.8) | .60 |
| SNP | |||||
| Yes | 1191 (69.0) | 229 (6.8) | 1420 (27.8) | 62.2 (59.8 to 64.5) | <.001 |
| No | 536 (31.0) | 3149 (93.2) | 3685 (72.2) | −62.2 (−64.5 to −59.8) | <.001 |
| SNP classification | |||||
| C-SNP | 313 (18.1) | 60 (1.8) | 373 (7.3) | 16.3 (14.5 to 18.2) | <.001 |
| D-SNP | 827 (47.9) | 57 (1.7) | 884 (17.3) | 46.2 (43.8 to 48.6) | <.001 |
| I-SNP | 51 (3.0) | 112 (3.3) | 163 (3.2) | −0.4 (−1.4 to 0.6) | .49 |
| Not an SNP | 536 (31.0) | 3149 (93.2) | 3685 (72.2) | −62.2 (−64.5 to −59.8) | <.001 |
| Mean monthly premium, $c | |||||
| 0 | 1673 (96.9) | 2801 (82.9) | 4474 (87.6) | 14.0 (12.4 to 15.5) | <.001 |
| 0.02 to 23.81 | 18 (1.0) | 193 (5.7) | 211 (4.1) | −4.7 (−5.6 to −3.8) | <.001 |
| 23.82 to 60.25 | 14 (0.8) | 196 (5.8) | 210 (4.1) | −5.0 (−5.9 to −4.1) | <.001 |
| 60.26 to 261.10 | 22 (1.3) | 188 (5.6) | 210 (4.1) | −4.3 (−5.2 to −3.4) | <.001 |
| Authority used to offer benefit | |||||
| VBID | 721 (41.7) | NA | 721 (41.7) | NA | NA |
| SSBCI | 926 (53.6) | NA | 926 (53.6) | NA | NA |
| Both | 80 (4.6) | NA | 80 (4.6) | NA | NA |
| Prior authorization | |||||
| Yes | 299 (17.3) | NA | 299 (17.3) | NA | NA |
| No | 1427 (82.7) | NA | 1427 (82.7) | NA | NA |
| Referral required | |||||
| Yes | 43 (2.5) | NA | 43 (2.5) | NA | NA |
| No | 1683 (97.5) | NA | 1683 (97.5) | NA | NA |
Abbreviations: C-SNP, Chronic Special Needs Plan; D-SNP, Dual Eligible Special Needs Plan; HMO, health maintenance organization; HMO-POS, health maintenance organization–point of service; I-SNP, Institutional Special Needs Plan; NA, not applicable; PFFS, private fee for service; PPO, preferred provider organization; SNP, Special Needs Plan; SSBCI, special supplemental benefits for the chronically ill; VBID, value-based insurance design.
Plans are described at the contract-plan level. Cells with NA are for variables unique to plans offering the Food and Produce benefit.
Categories defined by quartiles.
Categories defined by tertiles among plans with nonzero premiums.
In unadjusted models, a county with a 1-PP higher rate of food insecurity was associated with 2.5-PP (95% CI, 2.1-2.9 PP) higher enrollment in plans offering the Food and Produce benefit (P < .001) (Figure, A). Adjusting for plan characteristics did not meaningfully change the association (Figure, B). Additional adjustment for the percentage of plans offering Food and Produce benefits and the percentage that were SNPs attenuated the association by 26% (1.8 PP; 95% CI, 1.3-2.2 PP; P < .001) (Figure, C). Further adjustment for the percentage of MA beneficiaries enrolled in SNPs attenuated the association by an additional 86% (0.2 PP; 95% CI, 0.0-0.5 PP; P = .04) (Figure, D).
Figure. Association Between Enrollment in Medicare Advantage (MA) Plans Offering Food and Produce Supplemental Benefits and County-Level Food Insecurity.
Dots reflect mean percentage enrollment in MA plans offering the Food and Produce benefit across counties by ventile of county-level food insecurity, as measured by the 2025 Feeding America Map the Meal Gap6 project and weighted by county-level MA enrollment. B, Adjusted for basic county-level plan characteristics, including the number of MA plans and MA penetration (the proportion of Medicare beneficiaries enrolled in MA) as well as mean star ratings (in categories) and mean premiums (in quartiles) weighted by enrollment in each plan. C, Additionally adjusted for the proportion of MA plans offering the Food and Produce benefit and the proportion that were Special Needs Plans (SNPs). D, Additionally adjusted for the proportion of MA beneficiaries enrolled in SNPs.
Discussion
We found that SNP enrollment—rather than plan availability—largely explained the association between county-level food insecurity and enrollment in plans offering Food and Produce benefits. SNPs were more likely to offer such benefits, and individuals in counties with greater food insecurity were more likely to select these plans. The descriptive nature of our study precludes assessment of causality, and limited data availability prevents evaluation of benefit utilization.
Our study found little evidence that plans were preferentially offering Food and Produce benefits in more food insecure counties. Improved targeting of these benefits could better support communities most in need.
eMethods.
eReferences.
Data Sharing Statement
References
- 1.Centers for Medicare & Medicaid Services . 2022 MCBS PUF on financial well-being of Medicare beneficiaries living in the community. 2024. Accessed November 4, 2025. https://www.cms.gov/data-research/research/medicare-current-beneficiary-survey/data-tables/2022-mcbs-puf-financial-well-being-medicare-beneficiaries
- 2.Bleich SN, Leung CW, Hamad R. Implications of recent federal actions on food assistance programs. JAMA Health Forum. 2025;6(4):e251567. doi: 10.1001/jamahealthforum.2025.1567 [DOI] [PubMed] [Google Scholar]
- 3.Yang Z, Zhu E, Cheng D, et al. County-level enrollment in Medicare Advantage plans offering expanded supplemental benefits. JAMA Netw Open. 2024;7(9):e2433972. doi: 10.1001/jamanetworkopen.2024.33972 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Neuman T, Biniek F, Freed M. We need better data to measure the value of Medicare Advantage supplemental benefits. Health Affairs: Forefront. June 9, 2025. Accessed November 4, 2025. https://www.healthaffairs.org/content/forefront/medicare-advantage-supplemental-benefits-popular-but-data-needed-measure-their-value
- 5.Rowen NP, Stewart L, Saunders RS. Evaluation of supplemental benefits across Medicare Advantage plans and beneficiary demographic characteristics, 2019 to 2022. JAMA Netw Open. 2022;5(9):e2233020. doi: 10.1001/jamanetworkopen.2022.33020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Feeding America. Map the meal gap. Accessed November 4, 2025. https://map.feedingamerica.org/
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods.
eReferences.
Data Sharing Statement