In the U.S., legal cannabis is frequently advertised as an effective treatment for mental health problems such as anxiety and depression-particularly online. States that have legalized cannabis have not implemented regulations to address this type of advertising. This project aims to investigate the influence of psychotherapeutic advertising claims (PAC) and mental health warning labels (WL) on online cannabis purchasing behaviors among light-to-moderate cannabis users with symptoms of depression and/or anxiety.

The specific aims are to determine whether PAC increases cannabis purchasing intentions and whether a mental health WL can mitigate this effect. A realistic online cannabis shopping experience will be simulated using the digital Platform for Online Evaluation of Marijuana Marketing and Sales (POEMMS). The study will employ a between-subjects experimental design by randomizing 2,000 participants to one of four online stores that vary in PAC and WL content: (1) a control claims (CC)-only store, (2) a PAC store, (3) a WL store, and (4) a PAC + WL store. Participants will browse and select items as if making real purchases.

Analyses will examine three primary outcomes to determine the influence of PAC and WL on purchasing behaviors: (1) total milligrams of THC purchased, (2) average potency (%THC) of products, and (3) overall number of products purchased. The long-term objective is to inform evidence-based cannabis policy and regulatory strategies by understanding the impact of cannabis marketing on vulnerable populations.

This research is relevant to public health by addressing the potential risks associated with misleading cannabis marketing, which may lead to increased use and exacerbation of mental health symptoms among individuals with depression and anxiety. The project leverages a multidisciplinary team with expertise in addiction, mental health, epidemiology, and digital health technology. The findings have the potential to inform the development of targeted interventions and policies to reduce harms associated with cannabis advertising-ultimately contributing to better health outcomes and more effective regulation.

The U.S. legal cannabis industry increasingly uses psychotherapeutic advertising claims (PACs) on its websites to promote products that contain Δ⁹-tetrahydrocannabinol (THC; the primary intoxicant in cannabis) as effective treatments for mental health disorders-most notably depression and anxiety.

PACs employ mental health proxy terms (e.g., happy, calm) and symptoms (e.g., fatigue, stress) to imply or explicitly state that products alleviate specific mental health conditions (e.g., depression, anxiety). Although THC may temporarily relieve such symptoms, long-term use likely exacerbates depression and anxiety. The industry uses PACs to attract newer, less experienced (i.e., infrequent) consumers who often rely on industry-supplied "education" and marketing to make purchase decisions. Consequently, there is a need to examine the impact of PACs on cannabis purchasing among infrequent consumers with mental health symptomatology to understand whether such claims pose a risk of increased THC exposure and potential exacerbation of symptoms.

Warning labels (WLs) are an effective regulatory tool for reducing the purchase and use of harmful products (e.g., tobacco, alcohol, unhealthy foods). The WLs currently used by some U.S. states with legal cannabis focus on generic risks (e.g., "do not operate a motor vehicle") and do not address potential adverse impacts specific to mental health. Prior research has demonstrated that focused warnings about specific negative health effects improve consumer understanding of those effects. The industry's pervasive use of PACs and THC's association with worsened depression and anxiety underscore the need to investigate whether a mental health-focused WL can reduce THC purchasing among consumers with depression and anxiety.

The goal of the proposed project is to determine how PACs and a mental health WL impact cannabis purchase intentions among U.S. adults who are infrequent (1-100 days in the past year) cannabis consumers with symptoms of mild to severe depression and/or generalized anxiety. To achieve this goal, investigators will leverage a realistic online cannabis store platform developed in our laboratory. Participants (N=1,294) from two pilot studies (1) rated the online store as easy to navigate and highly realistic (e.g., realistic product types, potencies, names, prices); (2) rated the PAC stimuli as informative and helpful for making purchase decisions; and (3) spent total amounts on products that closely matched typical spending reported in prior studies.

The proposed between-subjects design study (N=2,000) will randomize participants to shop at one of four versions of the online cannabis store, each varying in the presence or absence of PACs and WLs:

1. Control Claims (CC) Store: includes only claims about a product's sensory (e.g., "Fruity") and quality (e.g., "Organic") attributes-a common alternative type of real-world cannabis advertising.
2. PAC Store: includes CC plus claims about a product's mental health effects (e.g., "Calming") and symptom relief (e.g., "Helps with stress").
3. WL Store: includes CC plus a WL about the negative impact of THC on anxiety and depression.
4. PAC+WL Store: includes PACs, WL, and CC.

Participants will be instructed to browse the store and buy products as if making real purchases. Each store will offer the same 30 products. Our specific aims are to test the effect of each store environment on three primary outcomes: (1) total milligrams of THC (mgTHC) purchased, (2) average potency (%THC) of products purchased, and (3) overall number of products purchased.

AIM 1: Test the effect of PACs on cannabis purchase intentions of infrequent users with anxiety or depression.

Hypothesis 1: Those exposed to the PAC Store will purchase (1) greater total mgTHC, (2) higher potency products, and (3) a greater overall number of products than those exposed to the CC Store.

AIM 2: Test the effect of a mental health WL on cannabis purchase intentions of infrequent users with anxiety or depression.

Hypothesis 2a: Those exposed to the WL Store will purchase (1) fewer total mgTHC, (2) lower potency products, and (3) a smaller number of products than those exposed to the CC Store.

Hypothesis 2b: Those exposed to the PAC+WL Store will purchase (1) fewer total mgTHC, (2) lower potency products, and (3) a smaller number of products than those exposed to PACs without the WL (PAC Store).

AIM 3: Test whether anxiety severity, depression severity, and past-year cannabis use frequency moderate the between-group relationships observed in Aim 1 and Aim 2.

Hypothesis 3: The observed effects of the PAC Store vs. CC Store (Aim 1) and WL Store vs. CC Store (Aim 2) will be greater for those with more severe depression or generalized anxiety and those with less frequent cannabis use. Notably, moderation effects may still exist even if investigators do not observe significant direct relationships as hypothesized in Aims 1 and 2. Thus, the feasibility and importance of testing the Aim 3 hypothesis do not depend on confirming Aims 1 and 2.

This proposal is responsive to NIDA NOT-DA-22-003. With expertise in addictive product advertising, cannabis policy, epidemiology and biostatistics, and co-occurring disorders, our team is uniquely qualified to conduct this project. Results will provide insights into how PACs and WLs alter cannabis purchasing and THC-related risks among novice consumers with anxiety or depression. Such information could help inform targeted regulations to protect this vulnerable group. This study will also lay the groundwork for future Cannabis Regulatory Science research exploring additional factors that influence cannabis purchasing and use patterns (e.g., advertising type, branding, demographics). This programmatic line of research will contribute to evidence-based cannabis policy that balances public health concerns with the evolving legal cannabis market.

Inclusion Criteria:

• Age 21 years or older
• Current residence in the United States
• Cannabis use on 1-100 days in the past year
• Current symptoms of mild to severe depression and/or generalized anxiety disorder, assessed by PHQ-8 (score ≥ 5) and GAD-7 (score ≥ 5)

Exclusion Criteria:

• Under 21 years of age
• Not currently residing in the United States

De-identified individual participant data (IPD) will be shared, including participants' product selections, browsing behaviors, purchasing decisions, demographics, cannabis use frequency, severity of anxiety and depression symptoms, and responses to pre- and post-shopping surveys. Data will be shared to promote transparency, reproducibility, and further research in cannabis regulatory science. All IPD collected throughout the trial that can be deidentified will be included. Data dictionaries, codebooks, and metadata describing study design, data collection, and processing will also be shared via The National Addiction & HIV Data Archive Program (NAHDAP) hosted by the Institute for Social Research at the University of Michigan

De-identified data and accompanying documentation will be made available after publication of each corresponding peer-reviewed paper and no later than the end of the project performance period. All cleaned and de-identified data will remain accessible indefinitely within The National Addiction & HIV Data Archive Program (NAHDAP). Thus, IPD sharing begins at the time of first publication and continues without an end date.

Researchers seeking access to the de-identified IPD must submit a formal data access request to the The National Addiction & HIV Data Archive Program (NAHDAP). Access will be limited to qualified investigators with institutional approval and a signed data use agreement. The NAHDAP will review all requests to ensure compliance with ethical and confidentiality standards. Approved users will receive controlled access to the requested data, metadata, and analysis code through the NAHDAP secure repository.