Study: State Policies Legalizing Cannabis Show No Links to Psychosis-Related Outcomes

Investigators focus on whether legalization of the drug is associated with an increase in health care use for the psychotic disorder.

State policies legalizing cannabis were not linked with psychosis-related outcomes, according to the results of a study published in JAMA Network Open.

Further research on this topic could be beneficial and informative, investigators said.

Psychosis has been hypothesized to result from cannabis use, so investigators sought to determine if legalization of the drug could be associated with an increase in health care use for psychosis.

Investigators used the Optum Clinformatics Data Mart Database, which contains deidentified commercial and Medicare Advantage claims of more than 63 million unique individuals followed between January 1, 2003, and December 31, 2017.

The study data included diagnostic codes, member enrollment data, and pharmacy claims that were linked across file types with a unique patient identifier.

Investigators included all beneficiaries aged 16 years and older with at least 1 month of insurance eligibility during the study period.

The retrospective cohort study used a panel fixed-effects design, in which the state-month was the unit of analysis to evaluate the association of state cannabis policies with the rate of psychosis-related health care claims.

Investigators used the number of unique claims with psychosis-related diagnoses, prescribed, antipsychotics, and enrolled individuals for each state-month of follow-up.

They merged these data with time-varying categorical measures of state cannabis policy level and state-level demographic, economic, and social characteristics.

There were more than 7 million psychosis-related diagnoses and more than 20 million filled prescriptions for antipsychotics recorded over the study period. Additionally, 29 states adopted medical or recreational cannabis legalization policies. The results from fully adjusted models demonstrated that states with legalization policies experienced no statistically significant increases in rates of psychosis-related diagnoses or prescribed antipsychotics compared with those with no legalization policies.

In a secondary analysis, investigators found that rates of psychosis-related diagnoses increased slightly among men, including those aged 55 to 64 years and those who were Asian, in states with recreational policies compared with states with no polices.

Investigators used different types of cannabis that were permitted in the analysis, including medical or recreational, and also looked at whether retail outlets in the state were open and operational.

States without retail outlets only allowed home-grown cannabis or did not implement commercial sales. Each state was assigned state cannabis policy levels, which included medical only, no retail outlets; medical only, retail outlets; recreational, no retail outlets; no medical or recreational policy; or recreational, retail outlets.

Limitations of the study included that the findings only included insured individuals, so the data might not be generalizable to those without insurance.

Investigators aimed to minimize confounding by controlling state and time fixed effects, as well as time-varying state-level characteristics, but residual confounding by factors associated with broader policy environment, including expanded social safety net programs and rates of comorbid substance use, could be possible in the study.


Elser H, Humphreys K, Kiang MV, Mehta S, et al. State cannabis legalization and psychosis-related health care utilization. JAMA Netw Open. 2023;6(1):e2252689. doi:10.1001/jamanetworkopen.2022.52689

Related Videos
Burnout illustration | Image credit: Vikky Mir -
Public Health Matters podcast logo
Capitol Building | Image credit: lazyllama -
Smiling Asian female pharmacist working in chemist shop or pharmacy | Image Credit: Atstock Productions -
sofiko14 /
Mammogram scans | Image credit: okrasiuk -
WP_7824 /
Related Content
© 2023 MJH Life Sciences

All rights reserved.