Table 3.
Difference-in-Differences Estimates of Service Use and Perceived Unmet Need for Black, Hispanic, and Asian Adults with Heavy/Binge Drinking, by Gender
Post-ACA: 2011–2014 (vs. 2008–2009) |
Sub-Analysis Post ACA: 2014 (vs. 2011–2013) |
||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Black (vs. White) | Hispanic (vs. White) | Asian (vs. White) | Black (vs. White) | Hispanic (vs. White) | Asian (vs. White) | ||||||||||||||||||
Outcome | DiD Estimatea (%) | 95% CI | DiD Estimatea (%) | 95% CI | DiD Estimatea (%) | 95% CI | DiD Estimatea (%) | 95% CI | DiD Estimatea (%) | 95% CI | DiD Estimatea (%) | 95% CI | |||||||||||
Women | |||||||||||||||||||||||
SUD treatment | 0.72 | −0.64 | 2.09 | 0.48 | −0.64 | 1.44 | — | — | — | 1.19 | −0.78 | 3.18 | −0.03 | −1.19 | 1.11 | — | — | — | |||||
MH treatment | 0.35 | −3.92 | 4.61 | 1.95 | −2.42 | 6.39 | 2.94 | −4.13 | 10.02 | −1.23 | −6.66 | 4.19 | −2.60 | −7.78 | 2.57 | −3.79 | −13.66 | 6.09 | |||||
Unmet need-SUD treatment | 1.02* | 0.14 | 1.91 | −0.26 | −1.12 | 0.60 | — | — | — | −1.32* | −2.52 | −0.13 | −0.15 | −0.53 | 0.230 | — | — | — | |||||
Unmet need-MH treatment | 1.76 | −1.16 | 4.67 | 1.45 | −1.27 | 4.17 | 1.23 | −3.46 | 5.91 | −3.42 | −7.40 | 0.55 | −0.35 | −3.55 | 2.85 | −2.36 | −6.68 | 1.97 | |||||
Men | |||||||||||||||||||||||
SUD treatment | −1.19 | −3.62 | 1.25 | 0.76 | −0.41 | 1.93 | — | — | — | 1.45 | −0.94 | 0.038 | −0.66 | −1.97 | 0.7 | — | — | — | |||||
MH treatment | −1.48 | −4.71 | 1.75 | −0.04 | −2.45 | 2.37 | −7.10** | −12.40 | −1.78 | −2.40 | −6.15 | 1.34 | 1.05 | −2.40 | 4.53 | −5.08** | −9.32 | −0.83 | |||||
Unmet need-SUD treatment | −1.65* | −3.30 | −0.001 | −0.45 | −1.23 | 0.33 | — | — | — | 1.12* | 0.07 | 0.02 | 0.85 | −0.27 | 1.98 | — | — | — | |||||
Unmet need-MH treatment | 0.44 | −1.05 | 1.93 | 0.80 | −0.38 | 1.97 | 3.01 | −0.35 | 6.37 | 0.45 | −1.43 | 2.32 | −2.06** | −3.67 | −0.45 | 2.09 | −3.56 | 7.73 |
p<.05,
p<.01,
p<.001
Note: All models controlled for predisposing, enabling and need variables. The models estimating unmet needs for SUD treatment and MH treatment also controlled for past-year SUD treatment and past-year MH treatment, respectively.
The Difference-in-Differences (DiD) estimates are derived from the b coefficients (i.e., probabilities) from the regression models and presented as percentage points.