Abstract
Background
Evidence supporting cannabis substitution along with liberalized cannabis laws have left recovery homes such as sober living houses (SLHs) in a difficult position regarding policies relating to cannabis use among SLH residents. Moreover, there are few studies of cannabis use among SLH residents that can be used to inform cannabis use policies. Here we assess whether cannabis is related to alcohol use among SLH residents.
Methods
Data came from N = 205 SLH residents entering 28 SLHs in Los Angeles from 2021 to 2023. Interviews were at baseline and one-, two-, three-, and six-month follow-ups. All participants reported lifetime alcohol use disorder (AUD). The primary predictor was any past 30-day cannabis use. Past 30-day outcomes were any drinking, number of drinking days, and any alcohol problems. Longitudinal generalized estimating equation models tested associations between any past-30-day cannabis use and outcomes, adjusting for demographics, treatment, 12-step attendance, social network use, baseline drug use, and AUD severity.
Results
After adjustment for demographics and covariates, any past-30-day cannabis use was related to 7.02 times higher odds of any past 30-day drinking (OR=7.02, 95 % CI: 3.06, 16.12), 2.03 times more drinking days (IRR=2.03, 95 % CI: 1.01, 4.08), and 3.21 times higher odds of any past-30-day alcohol problems (OR=3.21, 95 % CI: 1.68, 6.14) vs. no past-30-day cannabis use.
Conclusions
Cannabis use is positively correlated with alcohol use and related problems in a sample of sober living house residents in Los Angeles.
Keywords: Cannabis, Alcohol, Drugs, Sober living house, Recovery residence
1. Introduction
Evidence is mixed regarding cannabis’ efficacy as a substitute for alcohol. How cannabis use is associated with other alcohol use among individuals with alcohol use disorders (AUD) also remains understudied. Here we assess how cannabis use is related to alcohol outcomes among individuals with AUD who reside in sober living houses.
1.1. Cannabis as a substitute for alcohol
Cannabis has been put forward as a safer substitute for other substances, including alcohol (Gunn et al., 2022, Lucas et al., 2016, Lucas et al., 2019, Reiman, 2009, Risso et al., 2020, Subbaraman, 2016). By definition, cannabis acts as a substitute for alcohol if it replaces the effects of alcohol, resulting in decreased use (Gunn et al., 2022). Some prior studies have examined cannabis substitution among individuals with AUD by assessing correlations between cannabis and alcohol use; a number of these studies have found that cannabis use is related to reduced alcohol use among those with AUD (Gunn et al., 2022, Subbaraman, 2016). Similarly, a study using daily use data found that individuals in AUD treatment drank significantly less and were less likely to binge-drink on days that cannabis was used (Karoly et al., 2021). Other studies have examined substitution by asking participants directly about how they use cannabis. For example, among medical cannabis patients, cannabis is often reported as a substitute for alcohol (Lucas et al., 2016, Lucas et al., 2019, Reiman, 2009). Thus, evidence supports that cannabis might act as a substitute for individuals with AUD.
1.2. Recovery residences
Historically, recovery residences have been abstinence-based environments for individuals in recovery to live with others in recovery. Though structure and level of service vary across types of recovery residences, common elements are peer support and abstinence from alcohol and drugs (National Alliance of Recovery Residences, 2012). However, the evidence regarding cannabis substitution along with rapidly changing laws have left recovery residences in a difficult position regarding policies related to cannabis use among residents. In response to a request for guidance from recovery residences, the National Alliance of Recovery Residences (NARR) developed a “Policy Guide on Medical Cannabis Use” that offers recommendations for forming cannabis-related policies (National Alliance for Recovery Residences, 2021). NARR’s policy guide maintains that in some states “medical cannabis is also included as an alternate/substitute for prescription opioids and/or opioid abuse and addiction treatment.” NARR also asserts that each state affiliate has the authority to decide whether to allow the use of medical cannabis among residents. Notably, NARR only represents a fraction of recovery residences in the US. Thus, the ultimate decision to allow cannabis use among residents likely varies across houses and often falls to individual house operators.
1.3. Sober living houses
Sober living houses (SLHs) in California are a type of recovery residence, typically located in residential neighborhoods, and managed by someone who is either paid or receives a reduction of fees (Polcin et al., 2023). SLHs do not provide onsite treatment or clinical services, and instead offer a social model approach to recovery that emphasizes peer support and attendance at mutual-help groups (Polcin et al., 2023). Studies of SLHs show significant, sustained improvements in multiple areas of functioning, including substance use, psychiatric severity, employment, and criminal justice involvement (Mericle et al., 2022; Polcin and Korcha, 2017). SLHs are particularly common in California, a state where adult cannabis use is legal.
Findings from a recent study of cannabis use in SLHs in Southern California show that almost one-fifth of residents sampled reported using cannabis in the past six months, and that this proportion continued over one year (Subbaraman et al., 2024). Furthermore, past six-month cannabis use was related to more alcohol use and problems (Subbaraman et al., 2024). Despite evidence for cannabis substitution in other populations, these novel findings suggest that cannabis use is positively correlated with alcohol use and related problems. However, this prior study did not require an AUD or SUD diagnosis, which might impact the association between cannabis use and other substance use due to increased severity of use. Given that replication of results across samples is crucial for establishing an evidence base, the current study builds on prior analyses by (1) focusing on alcohol outcomes and residents with AUD in an independent sample; and (2) using a more proximal past 30-day measure of cannabis use, which is likely more appropriate for examining questions related to cannabis substitution. Thus, the aim of this study is to assess whether past 30-day cannabis use is related to alcohol use and problems in a newly collected sample of SLH residents with AUD.
2. Methods
2.1. Sample and measures
Data came from N = 205 SLH residents entering 28 SLHs in LA County from 2021 to 2023. Houses were sampled from diverse areas of Los Angeles, CA and covered West Los Angeles (14 %, n = 4), Central Los Angeles (21 %, n = 6), South Bay/Long Beach (43 %, n = 12), and the San Gabriel and San Fernando Valleys (21 %, n = 6). All houses were members of the Sober Living Network (SLN), which is an association of SLHs located in California. SLN provides certification, consultation, and advocacy to member houses. For recruitment, managers or owners of houses registered with the SLN were invited to participate. Other houses were already known to the study team through participation in other studies. The final sample included 14 houses designated for men, 7 for women, and 7 for all genders, and had a range of 8–24 beds (mean=13.4, SD=3.5). On average, resident monthly fees were $1039, with a range from $500 to $4000.
New residents were invited to participate via information on posted flyers, the house manager, or prior study participants. Participants were required to (1) be 18 years of age or older; (2) be able to provide informed consent; and (3) meet DSM-5 criteria for a past-year AUD, which was determined by whether participants endorsed two or more DSM-5 criteria for AUD in the past 12 months. The mean and median number of participants per house was 7, ranging from 1 to 14. Baseline assessments were conducted within 30 days of entry into the SLH (mean=16.0 days, SD=9.0 days after entering the SLH). Follow-up interviews were completed at one, two, three, and six months. Follow-up rates were > 80 %, ranging from 83 % at six months to 93 % at one month; participants were followed regardless of whether they stayed in the SLH throughout the study period. All study procedures were approved by the Public Health Institute institutional review board (IRB).
2.2. Measures
The primary predictor, any past 30-day cannabis use, was assessed at each interview as part of the Addiction Severity Index (ASI). This widely used instrument measures the severity of problems in individuals with substance use disorders and has shown excellent reliability and validity (McLellan et al., 1992). Past 30-day outcomes were any drinking, number of drinking days, and any ASI alcohol problems. We focused on alcohol use outcomes because the sample eligibility criteria include having an AUD diagnosis. Any drinking (0 =no alcohol use on any day, 1 =alcohol use reported on at least one day) and number of drinking days were measured at each interview using Timeline Followback, a daily self-report collected at each interview. Alcohol problems were measured at each interview using the ASI alcohol subscale, which measures the level of alcohol-related issues for the prior 30 days. The alcohol composite score is on a scale of 0–1; we dichotomized it as 0 =no problems reported and 1 =any problems reported. Any score greater than zero was classified as having had a problem.
Demographic covariates were assessed at baseline and included age, gender, and race-and-ethnicity. Past-30-day inpatient SUD treatment and past-12-month AUD severity were also measured at baseline. Time-varying covariates measured at all interviews were 12-step meeting attendance and proportion (%) of people who use alcohol/drug heavily in the participant's social network. Finally, length of stay in the SLH (number of days) was assessed at the final timepoint and included as a covariate.
2.3. Attrition
We examined whether attrition differed by variables shown in Table 1. Attrition was significantly (Ps<0.05) greater among those reporting any (vs. no) alcohol use in the 30 days prior to baseline, among those with no (vs. any) drug or alcohol inpatient treatment in 30 days prior to baseline, and among those who did not (vs. did) attend at least one 12-step meeting in 30 days prior to baseline.
Table 1.
Descriptive statistics for a sample of sober living house residents, Los Angeles County 2021-2023.
| Overall (N = 205) |
No cannabis use in past 30 days (n = 155) |
Any cannabis use in past 30 days (n = 50) |
|
|---|---|---|---|
| Covariate values at baseline | |||
| Age⁎⁎ (%) | |||
| 18–29 | 28.3 | 23.2 | 44.0 |
| 30–38 | 23.4 | 22.6 | 23.4 |
| 39–49 | 27.8 | 29.7 | 27.8 |
| 50 and older | 20.5 | 24.5 | 20.5 |
| Gender (%) | |||
| Female | 33.2 | 36.1 | 24.0 |
| Male | 66.8 | 63.9 | 66.8 |
| Race-and-ethnicity (%) | |||
| Non-Hispanic White/Caucasian | 40.7 | 40.9 | 40.0 |
| Non-Hispanic Black/African American | 16.2 | 16.2 | 16.0 |
| Latinx/Hispanic | 36.3 | 35.1 | 40.0 |
| Other/Multiracial | 6.9 | 7.8 | 4.0 |
| Baseline inpatient SUD treatment past 30 days (%) | |||
| Yes | 42.4 | 45.8 | 32.0 |
| No | 57.6 | 54.2 | 68.0 |
| 12-step meetings in past 30 days (%) | |||
| Yes | 91.6 | 93.2 | 86.4 |
| No | 8.4 | 6.8 | 13.6 |
| M (SD) % of social network uses drugs/alcohol heavily⁎ | 13.1 (26.1) | 10.6 (21.8) | 20.8 (35.4) |
| M (SD) # non-cannabis drug use days, past 30 days | 0.9 (3.8) | 0.8 (3.6) | 1.2 (4.2) |
| Baseline AUD Severity | |||
| Mild | 2.0 | 0.6 | 6.0 |
| Moderate | 2.9 | 2.6 | 4.0 |
| Severe | 95.1 | 96.8 | 90.0 |
| Outcome values at baseline | |||
| Any drinking, past 30 days (%)⁎⁎⁎ | 29.8 | 21.9 | 54.0 |
| M (SD) # drinking days, past 30 days⁎⁎⁎ | 2.8 (5.8) | 1.8 (4.5) | 5.7 (8.0) |
| Any alcohol problems, past 30 days (%)⁎ | 75.6 | 72.3 | 86.0 |
| Any non-cannabis drug use problems, past 30 days (%)⁎⁎⁎ | 54.2 | 47.1 | 76.0 |
| Outcome values at six months | |||
| Any drinking, past 30 days (%) | 30.2 | 27.7 | 38.0 |
| M (SD) # drinking days, past 30 days⁎⁎ | 1.1 (3.4) | 0.6 (2.5) | 2.5 (4.9) |
| Any alcohol problems, past 30 days (%) | 47.3 | 46.0 | 51.1 |
| Any non-cannabis drug use problems, past 30 days (%) | 42.4 | 40.0 | 50.0 |
P < 0.05
P < 0.01,
P < 0.001
2.4. Statistical analyses
We first examined descriptive statistics for demographics and covariates overall and by cannabis use groups, i.e., those who have reported any cannabis use at any interview vs. those who report no cannabis use at any interview. We also tested differences in demographics, covariates, and outcomes between cannabis use groups using bivariate tests. We then used longitudinal generalized estimating equation (GEE) regression models with robust standard errors to test associations between any past-30-day cannabis use and outcomes, adjusting for time (interview), demographics, 12-step and treatment attendance, heavy use in social network, AUD severity, baseline drug use days for past 30 days, and length of stay in the SLH. We used logistic regression for dichotomous outcomes (any drinking, any alcohol problems) and Poisson regression for number of drinking days.
3. Results
Table 1 displays descriptive statistics for the sample overall and for each cannabis use group. In terms of significant (ps < .05) differences, a higher percentage of those who used cannabis were: 18–29 (p = .008); reported any drinking in the past 30 days at baseline (p < .001); had more drinking days in the past 30 days at baseline (p < .001); and had more alcohol problems (p = .049) at baseline vs. those who did not use cannabis. Those who used cannabis also had a higher percentage of individuals in their network who used drugs/alcohol heavily vs. those who did not use cannabis (p = .016). Regarding six-month outcomes, cannabis use groups only differed on drinking days, with those who used cannabis reporting more drinking days at the six-month interview vs. those who did not use cannabis in bivariate tests (p = .001). Results from GEE models (Table 2) show that after adjustment for covariates, any past-30-day cannabis use was related to 7.02 times higher odds of any past 30-day drinking (OR=7.02, 95 % CI: 3.06, 16.12), 2.03 times more drinking days (IRR=2.03, 95 % CI: 1.01, 4.08), and 3.21 times higher odds of any past-30-day alcohol problems (OR=3.21, 95 % CI: 1.68, 6.14) vs. no past-30-day cannabis use.
Table 2.
Results from generalized estimating equations testing associations between repeated measures of any past-30-day cannabis use vs. none and past-30-day outcomes among a sample of sober living house residents (N = 205).
| M = mean, SD = standard deviation | Any Drinking | Number of Drinking Days |
Any Alcohol Problemsa |
|---|---|---|---|
| Predictor and covariates | ORb (95 % CI) | IRR (95 % CI) | OR (95 % CI) |
| Any cannabis use vs. none | 7.02 (3.06, 16.12)⁎⁎⁎ | 2.03 (1.01, 4.08)⁎ | 3.21 (1.68, 6.14)⁎⁎⁎ |
| Interview (vs. Baseline) One month Two months Three months Six months |
0.20 (0.08, 0.47)⁎⁎⁎ 0.28 (0.13, 0.64)⁎⁎ 0.33 (0.16, 0.68)⁎⁎ 0.35 (0.15, 0.81)⁎ |
0.19 (0.07, 0.50)⁎⁎ 0.31 (0.12, 0.81)⁎ 0.33 (0.15, 0.73)⁎⁎ 0.30 (0.16, 0.60)⁎⁎ |
0.44 (0.31, 0.63)⁎⁎⁎ 0.29 (0.20, 0.42)⁎⁎⁎ 0.29 (0.18, 0.45)⁎⁎⁎ 0.32 (0.20, 0.53)⁎⁎⁎ |
| Age | 1.02 (0.99, 1.05) | 1.01 (0.99, 1.03) | 1.01 (0.98, 1.03) |
| Male vs. female | 1.22 (0.63, 2.36) | 1.64 (0.74, 3.66) | 0.74 (0.46, 1.22) |
| Non-Hispanic White vs. Other race-and-ethnicity | 0.95 (0.49, 1.85) | 1.23 (0.62, 2.44) | 0.93 (0.56, 1.54) |
| Baseline inpatient SUD treatment past 30 days vs. not | 0.78 (0.36, 1.66) | 0.55 (0.24, 1.24) | 0.92 (0.56, 1.53) |
| 12-step meetings in past 30 days vs. not | 0.45 (0.22, 0.90)⁎ | 0.44 (0.22, 0.89)⁎ | 1.07 (0.70, 1.64) |
| % % of social network uses drugs/alcohol heavily | 1.01 (1.00, 1.02)⁎ | 1.00 (1.00, 1.01) | 1.00 (1.00, 1.01) |
| Baseline # non-cannabis drug use days for past 30 days | 1.03 (0.97, 1.10) | 1.04 (0.98, 1.10) | 0.98 (0.94, 1.03) |
| Length of stay at SLH | 0.99 (0.99, 1.00)⁎⁎ | 0.99 (0.99, 1.00) | 0.99 (0.99, 1.00)⁎⁎ |
| Baseline AUD Severity (vs. None) Mild Moderate Severe |
0.22 (0.01, 3.77) 0.31 (0.04, 2.28) 0.18 (0.03, 0.97) |
0.17 (0.02,1.62) 0.22 (0.03, 1.68) 0.28 (0.05, 1.46) |
1.63 (0.25, 10.68) 1.73 (0.25, 11.94) 1.03 (0.23, 4.49) |
IRR = incidence rate ratio from Poisson regression, CI = confidence interval, SUD = substance use disorder
P < 0.05
P < 0.01
P < 0.001
Alcohol and non-cannabis drug problems measured using Addiction Severity Index
OR = odds ratio from logistic regression
4. Discussion
Here we examined whether cannabis is related to alcohol use among individuals with AUD residing in SLHs. We found that past-30-day cannabis use is related to higher odds of alcohol use, more drinking days, and higher odds of alcohol problems. While these results might be surprising given that cannabis use has been found to be associated with less alcohol use among some individuals with AUD (e.g., Karoly et al., 2021), they corroborate recent findings showing that past six-month cannabis use is related to worse outcomes in a separate sample of SLH residents (Subbaraman et al., 2024). Although the associations between cannabis use and increased odds of alcohol use and problems suggest that cannabis use is related to harm in this population and therefore not an effective substitute (Chick and Nutt, 2012, Subbaraman, 2014), future studies should replicate these findings with detailed data collected for the purpose of studying cannabis substitution; this could entail daily diary data and/or direct questions regarding substitution. Future studies should also identify the range of current cannabis use policies and practices in SLHs, as these have not been documented or tested. Until more research examines how SLH cannabis policies are related to outcomes among residents and how characteristics of cannabis use (e.g., frequency, medical use) are related to other substance use outcomes, we lack the information necessary to make definitive claims about cannabis substitution and needed by SLHs to create evidence-based house-level policies.
Though not a focus, attrition was significantly higher among those reporting past-30-day alcohol use and those reporting no 12-step meetings in the past 30 days; past analyses show that both leaving an SLH within six months and lack of 12-step meeting attendance are related to fewer days abstinent (Subbaraman et al., 2023), which might explain attrition seen here.
4.1. Limitations
Though we designed recruitment to optimize socioeconomic diversity, generalizability might be limited as all SLHs were in Los Angeles County, CA. Adult cannabis use is legal in California, so findings might not generalize to areas with stricter cannabis laws. We also do not have data on amounts of alcohol or cannabis used, which is often used to examine the presence of substitution. We did, however, examine alcohol problems as outcomes, which should also decrease if cannabis were acting as a substitute. We did not find that. Still, future studies should collect more detailed data on amounts of substances used as well as more information regarding recovery goals. Future studies should also examine reasons for and the social contexts of cannabis use as these might impact the relationship between cannabis and alcohol or other drug use. Finally, the parent study was not designed to examine cannabis use and is missing information regarding, e.g., cannabis use frequency, potency, strain, or having a medical recommendation. Future studies should examine these factors as both might moderate the relationship between cannabis and other substance use (Subbaraman et al., 2019, Subbaraman and Kerr, 2018).
CRediT authorship contribution statement
Douglas Polcin: Writing – review & editing, Supervision, Funding acquisition, Data curation. Amy Mericle: Writing – review & editing. Elizabeth Mahoney: Writing – review & editing, Project administration, Formal analysis, Data curation. Meenakshi Sabina Subbaraman: Writing – original draft, Methodology, Formal analysis, Conceptualization.
Declaration of Competing Interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Douglas Polcin reports financial support was provided by National Institute on Alcohol Abuse and Alcoholism. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was funded by the National Institute of Alcohol Abuse and Alcoholism at the National Institutes of Health (R01AA028252, PI Polcin).
Author disclosures
This work was funded by the National Institute of Alcohol Abuse and Alcoholism at the National Institutes of Health (R01AA028252, PI Polcin).
Contributor Information
Meenakshi S. Subbaraman, Email: msubbaraman@phi.org.
Elizabeth Mahoney, Email: lmahoney@bhrsca.org.
Amy Mericle, Email: americle@arg.org.
Douglas Polcin, Email: dpolcin@bhrsca.org.
References
- Chick J., Nutt D.J. Substitution therapy for alcoholism: time for a reappraisal? J. Psychopharmacol. 2012;26(2):205–212. doi: 10.1177/0269881111408463. [DOI] [PubMed] [Google Scholar]
- Gunn, R.L., Aston, E.R., & Metrik, J. (2022). Patterns of Cannabis and Alcohol Co-Use: Substitution Versus Complementary Effects. Alcohol Research: Current Reviews, 42(1), 04. https://doi.org/10.35946/arcr.v42.1.04 [DOI] [PMC free article] [PubMed]
- Karoly, H.C., Ross, J.M., Prince, M.A., Zabelski, A.E., & Hutchison, K.E. (2021). Effects of cannabis use on alcohol consumption in a sample of treatment-engaged heavy drinkers in Colorado. Addiction (Abingdon, England), 116 (9), 2529–2537. https://doi.org/10.1111/add.15407 [DOI] [PMC free article] [PubMed]
- Lucas P., Walsh Z., Crosby K., Callaway R., Belle-Isle L., Kay R., Capler R., Holtzman S. Substituting cannabis for prescription drugs, alcohol and other substances among medical cannabis patients: the impact of contextual factors. Drug Alcohol Rev. 2016;35(3):326–333. doi: 10.1111/dar.12323. [DOI] [PubMed] [Google Scholar]
- Lucas P., Baron E.P., Jikomes N. Medical cannabis patterns of use and substitution for opioids & other pharmaceutical drugs, alcohol, tobacco, and illicit substances; results from a cross-sectional survey of authorized patients. Harm Reduct. J. 2019;16(1):9. doi: 10.1186/s12954-019-0278-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- McLellan A.T., Kushner H., Metzger D., Peters R., Smith I., Grissom G., Pettinati H., Argeriou M. The Fifth Edition of the addiction severity index. J. Subst. Abus. Treat. 1992;9(3):199–213. doi: 10.1016/0740-5472(92)90062-s. [DOI] [PubMed] [Google Scholar]
- Mericle A.A., Patterson D., Howell J., Subbaraman M.S., Faxio A., Karriker-Jaffe K.J. Identifying the availability of recovery housing in the U.S.: The NSTARR project. Drug Alcohol Depend. 2022;230 doi: 10.1016/j.drugalcdep.2021.109188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- National Alliance for Recovery Residences. 2021. NARR Policy Guide on Medical Cannabis. https://narronline.org/wp-content/uploads/2021/08/NARR-Policy-Guide-on-Medical-Cannabis-07-15-2021.pdf
- National Alliance of Recovery Residences. 2012. A primer on recovery residences: FAQ. https://narronline.org/wp-content/uploads/2014/06/Primer-on-Recovery-Residences-09-20-2012a.pdf
- Polcin D.L., Korcha R. Housing status, psychiatric symptoms, and substance abuse outcomes among sober living house residents over 18 months. Addict. Disord. Their Treat. 2017;16(3):138–150. doi: 10.1097/ADT.0000000000000105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Polcin D.L., Mericle A.A., Braucht G.S., Wittman F.D. Moving social model recovery forward: recent research on sober living houses. Alcohol. Treat. Q. 2023;41(2):173–186. doi: 10.1080/07347324.2023.2167528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Reiman A. Cannabis as a substitute for alcohol and other drugs. Harm Reduct. J. 2009;6:35. doi: 10.1186/1477-7517-6-35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Risso C., Boniface S., Subbaraman M.S., Englund A. Does cannabis complement or substitute alcohol consumption? a systematic review of human and animal studies. J. Psychopharmacol. 2020;34(9):938–954. doi: 10.1177/0269881120919970. [DOI] [PubMed] [Google Scholar]
- Subbaraman M.S. Can cannabis be considered a substitute medication for alcohol? Alcohol. Alcohol. 2014;49(3):292–298. doi: 10.1093/alcalc/agt182. 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Subbaraman M.S. Substitution and complementarity of alcohol and cannabis: a review of the literature. Subst. Use Misuse. 2016;51(11):1399–1414. doi: 10.3109/10826084.2016.1170145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Subbaraman M.S., Kerr W.C. Alcohol use and risk of related problems among cannabis users is lower among those with medical cannabis recommendations, though not due to health. J. Stud. Alcohol Drugs. 2018;79(6):935–942. doi: 10.15288/jsad.2018.79.935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Subbaraman M.S., Barnett S.B., Karriker-Jaffe K.J. Risks associated with mid level cannabis use among people treated for alcohol use disorder. Alcohol., Clin. Exp. Res. 2019;43(4):690–694. doi: 10.1111/acer.13973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Subbaraman M.S., Mahoney E., Mericle A., Polcin D. Six-month length of stay associated with better recovery outcomes among residents of sober living houses. The American Journal of Drug and Alcohol Abuse. 2023;49:675–683. doi: 10.1080/00952990.2023.2245123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Subbaraman M.S., Mahoney E., Mericle A., Polcin D. Cannabis use and alcohol and drug outcomes in a longitudinal sample of sober living house residents in California. Journal of Substance Use and Addiction Treatment. 2024;165:209454. doi: 10.1016/j.josat.2024.209454. [DOI] [PMC free article] [PubMed] [Google Scholar]