Recreational cannabis legalization and alcohol purchasing: a difference-in-differences analysis

Table 2 Monthly changes in alcohol purchasing (95% CI) per household by beverage category: policy states vs. control states

	Unrestricted sample		Restricted sample
	Single control state e^Beta (95% CI)	All non-policy states e^Beta (95% CI)	Single control state e^Beta (95% CI)	All non-policy states e^Beta (95% CI)
Colorado	(n = 7978)	(n = 171,829)	(n = 2843)	(n = 46,461)
All alcoholic products	0.90 (0.77, 1.06)	0.87 (0.77, 0.98)	0.92 (0.74, 1.14)	0.94 (0.80, 1.09)
Beer	0.93 (0.82, 1.05)	0.95 (0.87, 1.05)	0.96 (0.81, 1.13)	1.01 (0.90, 1.13)
Spirits	0.98 (0.88, 1.09)	0.96 (0.88, 1.04)	0.95 (0.82, 1.11)	0.97 (0.86, 1.08)
Wine	0.94 (0.86, 1.02)	0.94 (0.89, 0.99)	1.01 (0.90, 1.14)	0.99 (0.92, 1.06)
Washington	(n = 18,641)	(n = 172,694)	(n = 3556)	(n = 46,673)
All alcoholic products	1.07 (0.92, 1.25)	1.11 (0.97, 1.27)	1.08 (0.89, 1.31)	1.11 (0.94, 1.30)
Beer	1.05 (0.94, 1.18)	1.06 (0.96, 1.16)	1.04 (0.91, 1.19)	1.03 (0.92, 1.15)
Spirits	1.24 (1.12, 1.37)	1.25 (1.14, 1.37)	1.18 (1.02, 1.36)	1.21 (1.07, 1.36)
Wine	0.88 (0.79, 0.98)	0.91 (0.83, 1.00)	0.91 (0.80, 1.03)	0.94 (0.84, 1.05)
Oregon	(n = 6680)	(n = 170,736)	(n = 1095)	(n = 56,871)
All alcoholic products	1.06 (0.85, 1.34)	0.97 (0.81, 1.18)	0.97 (0.71, 1.33)	1.08 (0.86, 1.35)
Beer	1.05 (0.88, 1.25)	1.03 (0.89, 1.18)	1.01 (0.78, 1.32)	1.12 (0.94, 1.32)
Spirits	0.87 (0.77, 0.99)	0.85 (0.77, 0.95)	0.94 (0.79, 1.13)	0.96 (0.84, 1.09)
Wine	1.03 (0.88, 1.20)	0.97 (0.85, 1.11)	1.05 (0.87, 1.28)	1.03 (0.88, 1.21)

Models are fixed effects linear regression models with robust cluster standard errors, adjusted for household income, household size, marital status, and race. In each model, n is equal to the number of clusters (i.e., households) included for both the policy state and selected control(s). Estimates are exponentiated from log-transformed values to represent the average percent change in mean monthly alcohol purchasing per household following legalization of recreational cannabis. For example, a value of 0.87 corresponds to a 13% decrease while 1.25 corresponds to a 25% increase. The unrestricted sample is all households from 2004 to 2017. The restricted sample is limited to households from 2004 to 2017 that had data both before and after legalization of recreational cannabis in a given policy state. “Single control state” refers to models where a given policy state was compared to a single, matched control state. “All non-policy states” refers to models where controls were all states that did not legalize recreational cannabis. Bolded values represent p values < 0.05

ISSN: 2522-5782