Table 4. Estimated 10-Year Averted Hospitalizations and Net Policy Cost Savings Associated With Provision of MTMs, by Discounting Approach^a.

Insurance type	10-y Averted hospitalizations, No. (95% UI)b	10-y Net policy cost savings, 2019 $, in billions (95% UI)c
		No discounting of future costs	3% Discounting of future costsd	5% Discounting of future costs
Private	3 029 000 (1 812 000 to 4 385 000)	62.2 (−27.1 to 156.4)	45.5 (−32.8 to 127.3)	36.0 (−34.0 to 110.4)
Medicare	7 836 000 (5 008 000 to 11 144 000)	52.1 (−62.4 to 164.7)	30.2 (−69.4 to 128.2)	17.6 (−73.3 to 107.4)
Medicaid	2 226 000 (1 169 000 to 3 748 000)	31.1 (−12.1 to 81.5)	22.6 (−14.8 to 66.2)	17.7 (−16.4 to 57.4)
Dual eligible	5 166 000 (3 132 000 to 7 519 000)	115.2 (−16.9 to 260.0)	88.0 (−27.0 to 213.5)	72.4 (−32.0 to 186.6)
Total	18 257 000 (14 690 000 to 22 109 000)	260.7 (62.7 to 481.5)	185.1 (12.9 to 377.8)	143.7 (−11.8 to 319.4)

Abbreviations: MTM, medically tailored meal; UI, uncertainty interval.

^a In each of the 10 years, the eligible population was assumed to receive 8 months of medically tailored meals per year. Estimates are the mean of 1000 Monte Carlo simulations, with the 95% UI defined as the 2.5th percentile to the 97.5th percentile of the simulations. The policy simulation model ran 1000 Monte Carlo simulations using inputs and their uncertainties from the 2019 Medical Expenditure Panel Survey, relative risks of annual hospitalizations, and annual percentage change in health care expenditures, MTM receipt, screening costs, and meal costs. The policy simulation model was run separately and then summed for each of the 10 years (2019-2028) to obtain final estimates. Baseline distributions of hospitalizations and health care expenditures from 2020 to 2028 were estimated using the historical rate of change in population size and health care expenditures from 2010 to 2019 for the target population.

^b Rounded to the nearest 1000.

^c Rounded to the nearest $100 000 000.

^d Primary analysis.