. 2017 Jun 7;74(6):579–610. doi: 10.1001/jamapsychiatry.2017.0724

Table 1. Difference in Predicted Prevalences of Past-Year Cannabis Use and DSM-IV Cannabis Use Disorder, 1991-1992 to 2012-2013: Never-MML States and Ever-MML States (Model 1).

Outcome	Model 1, Prevalence (SE)^a
Outcome	Never-MML States^b (24 States)	Ever-MML States^c (15 States)
Cannabis use
1991-1992	4.54 (0.39)	5.55 (0.36)
2012-2013	6.70 (0.39)	9.15 (0.54)
Difference in prevalence	+2.17 (0.70)^d	+3.60 (0.79)^e
DiD (ever-MML vs never-MML states)	[Reference]	+1.43 (0.48)^f
Cannabis use disorder
1991-1992	1.35 (0.19)	1.48 (0.19)
2012-2013	2.30 (0.22)	3.10 (0.34)
Difference in prevalence	+0.96 (0.38)^g	+1.62 (0.47)^h
DiD (ever-MML vs never-MML states)	[Reference]	+0.66 (0.30)ⁱ

Abbreviations: DiD, difference-in-difference; MML, medical marijuana laws.

^{^a}

Prevalence based on participants in the 39 states included in all 3 surveys. States omitted: Alaska, Delaware, Hawaii, Idaho, Nebraska, New Hampshire, North Dakota, Rhode Island, South Dakota, West Virginia, and Wyoming. Prevalences back-transformed from marginal log-odds estimated from logistic regression interaction models (MML × survey) that included individual-level (sex, age, education, race/ethnicity, marital status, urbanicity, and poverty) and state-level (% male, % white, % <30 y, and % ≥25 y without a high school diploma) covariates. Predicted prevalences estimated by a model with a dichotomous MML variable (ever-MML states vs never-MML states).

^{^b}

Never-MML states (1996-2012): Alabama, Arkansas, Florida, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Minnesota, Mississippi, Missouri, New York, North Carolina, Ohio, Oklahoma, Pennsylvania, South Carolina, Tennessee, Texas, Utah, Virginia, and Wisconsin.

^{^c}

Ever-MML states (1996-2012): Arizona, California, Colorado, Connecticut, Maine, Maryland, Massachusetts, Michigan, Montana, Nevada, New Jersey, New Mexico, Vermont, Oregon, and Washington.

^{^d}

P = .003; change in never-MML states between 1991-1992 and 2012-2013.

^{^e}

P < .001; change in ever-MML states between 1991-1992 and 2012-2013.

^{^f}

P = .004, change in never-MML states between 1991-1992 and 2012-2013.

^{^g}

P = .01; change in never-MML states between 1991-1992 and 2012-2013.

^{^h}

P = .001; change in ever-MML states between 1991-1992 and 2012-2013.

^ⁱ

P = .03, change in never-MML states between 1991-1992 and 2012-2013.