TABLE 4.

Heterogeneity by treatment setting in the effect of ACA‐Medicaid expansions on accepted forms of payments and provision of discounted care using a differences‐in‐differences model: National Mental Health Services Survey 2010‐2018

Outcome	Medicaid	Private	Medicare	Cash or self‐pay	Discounted care
Hospitals
Proportion in expansion states, 2010‐12	0.8896	0.9631	0.9118	0.9257	0.7057
Expand	0.0065	−0.0113	0.0004	−0.0003	0.0050
Medicaid	[−0.0146,0.0277]	[−0.0279,0.0053]	[−0.0173,0.0182]	[−0.0236,0.0229]	[−0.0395,0.0495]
Observations	14 145	14 339	14 241	14 078	12 822
Residential providers
Proportion in expansion states, 2010‐12	0.7696	0.4969	0.3673	0.6595	0.4994
Expand	0.0304	0.0187	0.0082	−0.0228	−0.0189
Medicaid	[−0.0188,0.0795]	[−0.0240,0.0613]	[−0.0441,0.0605]	[−0.0680,0.0225]	[−0.0741,0.0362]
Observations	12 293	11 979	12 083	12 195	12 523
Outpatient providers
Proportion in expansion states, 2010‐12	0.8929	0.7671	0.6820	0.8454	0.8388
Expand	0.0136*	−0.0066	0.0247	0.0032	−0.0090
Medicaid	[−0.0002,0.0275]	[−0.0322,0.0189]	[−0.0050,0.0544]	[−0.0124,0.0188]	[−0.0339,0.0160]
Observations	55 845	55 684	55 392	55 711	56 339
All providers
Proportion in expansion states, 2010‐12	0.8727	0.7576	0.6710	0.8291	0.7623
Expand	0.0239***	−0.0087	0.0294**	0.0041	−0.0150
Medicaid	[0.0062,0.0417]	[−0.0371,0.0197]	[0.0004,0.0584]	[−0.0173,0.0256]	[−0.0387,0.0087]
*Hospital	−0.0254	−0.0002	−0.0661***	−0.0120	0.0459*
	[−0.0630,0.0122]	[−0.0562,0.0557]	[−0.1131, −0.0191]	[−0.0514,0.0274]	[−0.0051,0.0969]
*Residential	−0.0169	0.0522	0.0177	0.0036	0.0045
	[−0.0650,0.0311]	[−0.0329,0.1373]	[−0.0617,0.0971]	[−0.0610,0.0682]	[−0.0664,0.0754]
Observations	82 283	82 002	81 716	81 984	81 684

The unit of observation is a provider in a state in a year. Data are unweighted. “Expand Medicaid” is an indicator coded one if the state has expanded Medicaid in a state in a year and zero otherwise. All models estimated with a linear probability model and control for provider characteristics, state characteristics, state fixed effects, and year fixed effects. 95% confidence intervals account for within‐state clustering and are reported in square brackets. Discounted care = free care provided to patients who cannot pay and/or use of a sliding fee scale. National Mental Health Services Survey data are available in 2010, 2012, 2014, 2015, 2016, 2017, and 2018.

***, **, and * = statistically different from zero at the 1%, 5%, and 10% level.