Table 2. Changes in Suicide Rates in Medicaid Expansion and Nonexpansion States.

Characteristics	Suicide mortality rate per 100 000 individuals							Difference-in differences analyses
	All states, 2000-2018	Expansion statesa			Nonexpansion statesb			Unadjusted estimates	Adjusted estimatesc
		2000-2013d	2015-2018	Change	2000-2013d	2015-2018	Change		Estimate (95% CrI)	Pr(Est >0)
Total	16.82	13.94	16.49	2.56	16.67	19.77	3.10	−0.54	−0.40 (−0.66 to −0.14)	<.001
Age, y
20-29	14.72	11.92	14.95	3.03	14.09	18.10	4.02	−0.99	−0.52 (−1.00 to −0.05)	.02
30-44	16.18	13.48	15.77	2.29	16.46	19.15	2.68	−0.40	−0.30 (−0.72 to 0.09)	.07
45-64	18.74	15.60	18.16	2.56	18.41	21.40	2.99	−0.43	−0.04 (−0.40 to 0.31)	.42
Sex
Male	26.03	21.95	25.3	3.35	26.06	30.49	4.43	−1.07	−0.41 (−0.86 to 0.03)	.04
Female	7.71	6.02	7.70	1.68	7.44	9.19	1.75	−0.07	−0.17 (−0.41 to 0.06)	.29
Race
Black	7.90	7.25	8.61	1.37	7.21	8.69	1.48	−0.11	0.18 (−0.17 to 0.54)	.84
White	19.09	15.46	18.66	3.20	18.98	22.86	3.88	−0.68	−0.39 (−0.71 to −0.09)	.005
Othere	9.04	7.81	9.17	1.35	8.52	11.18	2.66	−1.30	−0.37 (−1.05 to 0.21)	.13

Abbreviations: CrI, credible interval; Pr(Est >0), 1-tailed probability of the estimate being greater than 0 (Pr[Est >0]<.025 indicates statistical significance).

^a Medicaid expansion states included Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Hawaii, Illinois, Iowa, Kentucky, Maryland, Massachusetts, Michigan, Minnesota, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, Oregon, Rhode Island, Vermont, Washington, Washington, DC, and West Virginia. Note that Alaska, Indiana, Louisiana, Montana, and Pennsylvania expanded in 2015 to 2018 and were included as Medicaid expansion states in our adjusted difference-in-differences analyses (with staggered postexpansion period definitions), although they were excluded from the suicide mortality rate information and unadjusted analyses because of differing postexpansion period definitions.

^b Nonexpansion states are those that have not expanded Medicaid or expanded after 2018, including Alabama, Florida, Georgia, Idaho, Kansas, Maine, Mississippi, Missouri, Nebraska, North Carolina, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, and Wyoming.

^c Adjusted estimates are based on a hierarchical bayesian linear regression model accounting for covariates including state fixed effects and state-specific trends, age, and state-level unemployment, poverty, educational level, opioid prescribing laws, and firearm laws. The model also accounts for correlation between repeated measures over time and age groups within state clusters.

^d Observations from 2014, the year during which most Medicaid expansion states enacted expansion, were included in our final adjusted models and in the summary statistics for 2000 to 2018.

^e Other subgroup includes data for the following racial groups: American Indian or Alaska Native, Asian, and Native Hawaiian or Other Pacific Islander.