Daily risk factors for using more alcohol and cannabis than intended: An examination of contextual and motivational risk factors among sexual minority women and gender diverse individuals

Abstract

Drinking more than intended is associated with alcohol consequences in the general population, and some risk factors have been identified. However, it is unclear whether these risk factors generalize to sexual and gender minority individuals (SGM) and whether unique SGM risk factors may also contribute to drinking more than intended. Research with the general population has also neglected the examination of using more cannabis than intended and the potential role of substance use motives. We aimed to address these gaps by examining contextual, motivational, and SGM-specific risk factors for using more alcohol or cannabis than intended at the daily level. We used ecological momentary assessment data from 366 sexual minority women and gender diverse individuals assigned female at birth. Social drinking was associated with drinking more than intended, consistent with prior work. Additional novel risk factors also emerged, including social, enhancement, and conformity motives. Several risk factors for drinking more than intended extended to using more cannabis than intended (i.e., social use; social, conformity, enhancement motives), while one cannabis-specific factor emerged (i.e., coping motives). Only one SGM-specific risk factor predicted drinking more than intended (i.e., drinking with SGM). Together, findings highlight several potential targets for interventions aiming to reduce consequences associated with drinking more than intended. For example, interventions may encourage SGM young adults to adjust their protective behavioral strategies so that they would be adequate if they drank more than intended.

Introduction

Sexual minority (e.g., lesbian/gay, bisexual, queer) and gender minority populations (i.e., individuals who identify with a gender other than that associated with their sex assigned at birth) are at elevated risk for alcohol and cannabis use disorders compared to heterosexual, cisgender populations.^1,2 Disparities are particularly pronounced for sexual minority cisgender women and gender diverse individuals (i.e., those who identify outside the gender binary) assigned female at birth.^3,4 These disparities have been attributed to two factors: 1) stressors experienced by sexual and gender minorities (SGM) as a result of stigma (i.e., minority stress)⁵ and 2) perceptions of substance use as normative among SGM communities.⁶ However, research on alcohol and cannabis use among SGM has focused almost exclusively on associations between minority stress and heavy episodic drinking,^7–9 neglecting other predictors and patterns of substance use.

Drinking more than intended is associated with alcohol consequences in the general population, and some risk factors have been identified.^10–12 However, it is unclear if these risk factors generalize to SGM. Minority stressors and SGM community norms around substance use may contribute to differences in the impact of risk factors identified in the general population among SGM. For example, the added burden of minority stress may lead to stronger effects of coping motives on unintended substance use than seen in the general population. Further, research has yet to determine whether unique SGM risk factors also contribute to drinking more than intended. Research with the general population has also neglected to examine using more cannabis than intended and the role of substance use motives. The current study aims to address these gaps by examining contextual, motivational, and SGM-specific risk factors for using more alcohol or cannabis than intended among a sample of sexual minority women and gender diverse individuals assigned female at birth (SMWGD), due to their elevated rates of alcohol and cannabis use disorders.¹³ We focus on alcohol and cannabis as these are the two most commonly used substances among SGM and rates of alcohol and cannabis use disorders are far higher than rates of other substance use disorders among SGM.^1,2,13,14

The theory of unplanned drinking behavior posits that people engage in protective behavioral strategies based on how much they intend to drink; thus, drinking more than intended may lead planned protective behavioral strategies to be inadequate.¹⁵ Consistent with this theory, drinking more than intended has been linked to consequences at daily and weekly levels.^10,11 Daily alcohol and cannabis consequences include negative physical (e.g., feeling nauseous, having a hangover), social (e.g., doing something embarrassing), and psychological consequences (e.g., feeling anxious [cannabis], feeling depressed [alcohol]) that occur frequently enough to exhibit variation at the daily level.¹⁶ Studying daily consequences can help to identify patterns of alcohol/cannabis consumption that contribute to these relatively minor, but common, consequences, leading to the identification of patterns of consumption that may be targets for harm reduction interventions. Few studies have explored daily risk factors for drinking more than intended, and results have been mixed. For example, Labhart, Anderson and Kuntsche ¹² linked social drinking with drinking more than intended, while Fairlie, Cadigan, Patrick, Larimer and Lee ¹¹ did not. Others have implicated friends’ encouragement to get drunk¹⁰ and drinking locations (e.g., clubs).¹² Given sparse and mixed results, further research is needed.

Notably, no studies have examined associations between drinking motives and drinking more than intended or examined using more cannabis than intended. Given the social nature of drinking more than intended,^10,12 social motives are likely to be associated with drinking more than intended. Similarly, conformity motives may contribute to drinking more than intended due to peers encouraging further consumption¹⁰ while enhancement may contribute to drinking more than intended due to a desire to further increase positive affect. On the other hand, coping motives, which are linked with drinking in solitary settings,¹⁷ may reduce the likelihood of drinking more than intended. To date, factors that may predict using more cannabis than intended have been neglected, making it unclear whether risk factors for drinking more than intended also apply to cannabis.

Elevated rates of alcohol and cannabis use disorders among SGM have been theorized to arise from two sources: 1) minority stressors, like microaggressions,⁵ and 2) normative perceptions of substance use among SGM communities.⁶ Minority stressors are theorized to overburden SGM individuals’ coping resources, increasing the likelihood that they will utilize substances to cope,¹⁸ a risk factor for substance use disorders.¹⁹ Research has linked minority stressors with heavier alcohol and cannabis use and consequences via coping motives,^7,20 supporting this theory. These added stressors and resulting depletion of coping resources may directly contribute to drinking more than intended and may lead coping motives to be associated with using more than intended among SGM.

Alcohol and cannabis use is perceived to be more normative among SGM than among heterosexual, cisgender individuals²¹ and this has been linked with SGM individuals’ own heavier alcohol and cannabis use.^22,23 These normative perceptions have also been theorized to lead drinking or using cannabis with other SGM to contribute to heavier use and studies have supported this theory.^24,25 However, it is unclear whether this may also contribute to drinking or using more cannabis than intended. Further, these SGM community norms may also lead to differences in associations between social motives/contexts and drinking or using more cannabis than intended in SGM compared to the general population, making it important to examine these associations among both SGM and cisgender, heterosexual populations.

The current study aimed to address gaps in research on drinking more than intended, including: 1) examining associations between motives and drinking more than intended; 2) testing SGM-specific risk factors for drinking more than intended; 3) determining if risk factors from the general population apply to SGM; and 4) expanding research to include using more cannabis than intended (i.e., being intoxicated for longer or becoming more intoxicated than intended). We utilized ecological momentary assessment data to minimize retrospective bias in reporting of alcohol and cannabis use intentions, actual use, motives, and contexts and to allow for the examination of within-person day-level associations. We focused on SMWGD because they experience a disproportionate burden of alcohol and cannabis use disorders. We hypothesized:

1. Drinking or using more cannabis than intended would be associated with more alcohol or cannabis consequences.

2. Drinking or using more cannabis than intended would be more likely to occur when alcohol or cannabis were used in social contexts and when social, enhancement, conformity, or coping motives were endorsed.

3. Microaggressions and using alcohol or cannabis with other SGM would be associated with using more alcohol or cannabis than intended.

Methods

Participants and Procedures

The current analyses used data from a study of substance use among 429 SMWGD conducted between August 2020 and May 2021. Participants were recruited via paid social media advertisements. Ads generated 618,815 impressions and 12,114 link clicks. 7,835 began the eligibility survey and 5,563 completed the eligibility survey. 2,009 were eligible, and 429 were ultimately enrolled in the study, with stratified recruitment quotas by race/ethnicity, gender identity, and sexual identity. Eligible participants were 1) U.S. residents, 2) 18–25 years old, 3) assigned female at birth, 4) identified as women or under the non-binary/gender diverse umbrella (e.g., non-binary, genderqueer, agender, gender fluid), 5) identified as lesbian, bisexual, pansexual, or queer, and 6) met alcohol or cannabis use criteria (i.e., reported having four or more drinks at least twice and/or using cannabis on at least three days in the past month).¹ Participants who appeared eligible were asked to verify demographic information via text message, which was cross-checked with their survey responses. The study included a 30-day EMA period with two surveys per day: one between 8:00am and 1:00pm and another between 6:00pm and 12:00am. Participants were paid up to $150. The study received IRB approval at Ohio State University. See Dyar et al., (2022)⁷ for details.

The analytic sample included 366 participants who reported planned alcohol or cannabis use during at least one EMA survey (Table 1) and was comprised predominately of people of color, with only 35.5% (n = 130) identifying as non-Latinx White. There was a sizeable number of gender diverse participants (24.3%; n = 89).

Table 1.

Demographics of Analytic Sample at Baseline (N = 366)

Demographic Variable	n	%
Sexual Identity
Lesbian	95	26.0%
Bisexual	100	27.3%
Pansexual	77	21.0%
Queer	94	25.7%
Race/Ethnicitya
White	209	57.1%
Black	75	20.5%
Latine	111	30.3%
Asian	50	13.7%
Other Race/Ethnicity	28	7.7%
Gender Identity
Cisgender Women	277	75.7%
Gender Minority	89	24.3%
Age (M, SD)	22.34 (1.96)

^aPercentages add up to more than 100% because participants could select multiple racial/ethnic identities.

EMA Measures

Alcohol and cannabis use intentions.

During evening surveys when participants did not report drinking or using cannabis, they were asked about their intentions to drink or use cannabis (e.g., “In the next 24 hours, do you plan to have at least one drink?”). When drinking was planned, they were asked “how much do you plan to drink?” (options: 0–25 or more in one drink increments). When cannabis use was planned, participants were asked “how high do you plan to get?” (0 “not at all” to 4 “extremely high”) and “how long do you plan to be high?” (0–12 or more hours in one-hour increments). We used subjective intoxication and duration of intoxication as proxies for quantity of cannabis consumed, as they function as parsimonious proxies for cannabis quantity that function across modes of use at the event-level.²⁶

Substance use motives.

When participants reported cannabis or alcohol use, they were asked about their reasons for use. Coping (3 items; e.g., “to reduce my anxiety”); social (1 item; “to be social or make a social gathering more fun”); enhancement (1 item; “to feel good or have fun”); and conformity motives (1 item; “because other were drinking/using marijuana”) were assessed. Response options were no (0) and yes (1). The three coping motive items were used to create a binary variable (0 = no coping motives, 1 = at least one coping motive).

Substance use context.

Participants were asked who they used alcohol or cannabis with. Options included: “I drank/used marijuana alone,” “friend(s),” “romantic or sexual partner, spouse, or date,” “coworker(s),” “relative(s),” and “stranger(s),” and participants could select multiple responses. Binary variables were created indicating if social alcohol or cannabis use occurred.

Sexual orientation and gender identity of substance use companions

were measured by asking “which of the following best describes those you were drinking/using marijuana with.” Response options: LGBTQ+ women, LGBTQ+ men, nonbinary individuals, straight women, and straight men. These were combined into variables indicating use with SGM individuals and heterosexual individuals.

Duration of cannabis intoxication

was measured when cannabis use was reported (i.e., “How many hours were you high?”). Responses were in one-hour increments from 0 to 11 hours and participants could indicate being high for 12+ hours. Duration of intoxication is a parsimonious proxy for cannabis quantity that functions across modes of use at the event-level.²⁶

Subjective intoxication

was assessed by asking “How high did you get when you used marijuana?” on a scale of 0 (not at all high) to 4 (extremely high).

Number of drinks

consumed was assessed by asking “How many drinks did you have since the last survey?” Participants could indicate the specific number of drinks they consumed from 0 to 24 or indicate that they consumed 25+ drinks.

Alcohol and cannabis consequences

were measured during morning assessments using five item measures for alcohol and cannabis derived from existing measures.^27–29 Participants were asked “Which of the following happened to you as a result of your marijuana use/drinking yesterday?” (e.g., “I said or did something that embarrassed me”). Participants could select multiple responses and a count of the number of consequences experienced was calculated. See Supplementary Materials for a list of consequences.

Microaggressions

were assessed by asking participants two questions. First, they were asked “Did you experience anything stressful or negative related to your sexual orientation since the last survey?” (yes or no).³⁰ Regardless of their response, participants were asked to “indicate which of the following events you have experienced since the last survey because of your sexual orientation” and provided with a list of 10 common microaggressions (e.g., “someone acted uncomfortable around me”).^31,32 Given that few participants endorsed multiple microaggressions on the same day, we created a binary variable indicating endorsement of either item.

Data Analysis

Analyses were conducted with Mplus 8.9. The median completion rate was 88.3% (M=74%, SD=28%). A total of 0.5% of data were missing within completed observations and were handled using Bayesian methods.³³ Analyses included 1,449 planned cannabis use occasions and 1,272 planned drinking occasions. Given that alcohol and cannabis use intentions asked about the next 24 hours, we calculated day-level aggregate values for all correlates.

Multilevel structural equation modeling (MSEM) with a Bayesian estimator and diffuse priors were used.^2,3 Probit regression was used to handle binary outcome variables. Count variables (i.e., number of drinks, duration of intoxication, consequences) were analyzed using a Polya-Gamma distribution, which is appropriate for count variables and can be implemented in MSEM.³⁴ Probit and Polya-Gamma analyses treat the outcome as being derived from a continuous underlying latent variable, providing regression coefficients that are interpreted in a similar manner as linear regression coefficients.³⁴

First, we examined whether using more alcohol or cannabis than intended predicted alcohol or cannabis consequences. In these models, within-person and between-person components of planned and reported consumption predicted within- and between-person components of alcohol or cannabis consequences. Within-person associations between intended and reported consumption and consequences were allowed to vary across individuals. In this and subsequent models, we controlled for day of assessment and assessment type (weekend/weekday) at the within-person level. We also controlled for the use of alcohol in models of cannabis consequences and vice versa to remove any effects of co-use on consequences. Age, sexual identity, gender identity, and race/ethnicity were included as covariates at the between-person level in all models.

Second, we examined whether characteristics of alcohol or cannabis use events were associated with using more than intended. We examined whether motives for cannabis use, contexts of cannabis use, or microaggressions predicted using more cannabis than intended (i.e., being intoxicated for longer than planned or getting more intoxicated than planned). Parallel analyses were conducted for drinking more than intended. In these models, within-person and between-person components of characteristics of alcohol or cannabis use predicted within- and between-person components of reported consumption/intoxication, while intended consumption/intoxication was included as a covariate. By controlling for intended consumption, the outcome in these models became the deviation between actual and intended consumption (i.e., the extent to which individuals drank or used more cannabis than intended). Within-person associations among characteristics of alcohol or cannabis use, planned consumption/intoxication, and reported consumption/intoxication were allowed to vary across individuals.

Results

Participants drank more than intended on 38.5% of days (n = 490) and drank an average of 1.95 more drinks than intended. Participants were intoxicated for longer than intended on 45.0% of days (n = 572) and were more intoxicated than intended on 38.9% of planned cannabis use days (n = 495). When participants used more cannabis than intended, they were intoxicated for an average of 2.27 additional hours and were 1.53 points higher on the subjective intoxication scale than intended. See Tables 2 and 3 for means, standard deviations, and intraclass correlations.

Table 2.

Prevalence of Drinking or Using More Cannabis Than Intended

Variable	n	%
Number of Drinks
Reported less than intended	336	26.4%
Equal to intended	448	35.2%
Reported more than intended	490	38.5%
Average additional drinks	1.95 (1.76)
Duration Intoxication (Cannabis)
Reported less than intended	282	22.2%
Equal to intended	417	32.8%
Reported more than intended	572	45.0%
Average additional hours of intoxication	2.27 (1.83)
Subjective Intoxication (Cannabis)
Reported less than intended	230	18.1%
Equal to intended	551	43.3%
Reported more than intended	495	38.9%
Average additional subjective intoxication	1.53 (.85)

Table 3.

Means, Variances, and Intraclass Correlations

	Mean	Standard Deviation	Range	Intraclass Correlation
Cannabis Variables
Intended Duration Intoxication	2.82	1.15	0–24	.56
Intended Subjective Intoxication	1.82	.55	0–4	.45
Duration of Intoxication	3.58	1.62	0–24	.43
Subjective Intoxication	2.21	.58	0–4	.25
Cannabis Consequences	.34	.66	0–6	.39
Cannabis Motives
Social	.18	.84	0–1	.18
Conformity	.10	1.18	0–1	.30
Enhancement	.49	1.10	0–1	.27
Coping	.80	1.04	0–1	.25
Cannabis Contexts
Social Context	.51	1.26	0–1	.33
SGM Companions	.74	1.99	0–1	.55
Heterosexual Companions	.44	1.87	0–1	.52
Alcohol Variables
Intended Number of Drinks	2.38	.84	0–25	.41
Number of Drinks	2.73	1.32	0–25	.37
Alcohol Consequences	.32	.35	0–5	.24
Alcohol Motives
Social	.40	.89	0–1	.19
Conformity	.23	1.08	0–1	.26
Enhancement	.50	.91	0–1	.20
Coping	.57	.79	0–1	.16
Alcohol Contexts
Social Context	.76	1.13	0–1	.28
SGM Companions	.70	1.45	0–1	.39
Heterosexual Companions	.62	1.37	0–1	.36
Other Variables
Microaggressions	.11	.87	0–1	.19

Using more than intended and substance use consequences

At the within-person level, the number of drinks consumed was associated with alcohol consequences when the intended number of drinks was controlled for (Table 4). The intended number of drinks did not predict consequences. This indicates that participants reported more consequences on days when they consumed more drinks than intended. In models of cannabis consumption, intending to be intoxicated for longer and marginally (p = .06) intending to become more intoxicated were associated with experiencing fewer consequences, while reported duration and subject intoxication were not associated with consequences.

Table 4.

Within-person associations between drinking and using more alcohol or cannabis than intended and consequences of use

Outcome	Predictors	b	95% CI	p
Alcohol Consequences	Intended Number of Drink	.08	−.05, .22	.23
	Reported Number of Drinks	.20	.12, .29	< .001
Cannabis Consequences	Intended Duration of Intoxication	−.16	−.32, −.004	.046
	Reported Duration of Intoxication	.05	−.04, .11	.21
Cannabis Consequences	Intended Subjective Intoxication	−.24	−.49, .01	.06
	Reported Subjective Intoxication	.09	−.05, .22	.20

Within-person covariates included in all models: day of assessment, assessment type (weekend/weekday), and alcohol use (in cannabis use model) or cannabis use (in alcohol use models). Between-person covariates included: age, sexual identity, gender identity, and race/ethnicity.

Using more than intended and contexts of use

At the within-person level, social, conformity, and enhancement motives were associated with consuming more alcohol than intended, being more intoxicated than intended, and being intoxicated for longer than intended (for cannabis use; Table 5). Coping motives were only associated with subjective intoxication, with coping motives for cannabis use being associated with becoming more intoxicated than intended.

Table 5.

Within-person predictors of drinking or using more cannabis than intended

	Deviation from Drinking Intentions			Deviation from Cannabis Intentions (Duration Intoxication)			Deviation from Cannabis Intentions (Subjective Intoxication)
Predictor	b	95% CI	p	b	95% CI	p	b	95% CI	p
Motives
Social	.27	.17, .37	< .001	.14	.05, .23	.004	.23	.05, .40	.01
Conformity	.24	.13, .35	< .001	.23	.10, .36	< .001	.40	.16, .65	.002
Enhancement	.37	.28, .46	< .001	.24	.17, .32	< .001	.49	.34, .63	< .001
Coping	.06	−.03, .15	.17	.10	−.01, .19	.06	.18	.01, .34	.03
Context
Social Context	.28	.16, .41	< .001	.20	.12, .29	< .001	.42	.28, .57	< .001
Using with SGM	.27	.14, .39	< .001	.001	−.16, .18	.99	−.14	−.43, .21	.44
Using with Heterosexuals	.20	.07, .32	.002	.12	−.02, .25	.09	.23	−.04, .53	.09
Microaggressions	.09	−.04, .21	.18	−.06	−.18, .05	.29	−.18	−.38, .03	.10

Within-person covariates included in all models: day of assessment, assessment type (weekend/weekday), intentions for number of drinks, duration of intoxication, or subjective intoxication. Between-person covariates included: age, sexual identity, gender identity, and race/ethnicity.

Social alcohol or cannabis use were associated with using more alcohol than intended and being intoxicated for longer and being more intoxicated than intended (for cannabis). Drinking with SGM or with heterosexuals was associated with drinking more than intended, but using cannabis with SGM or heterosexuals were not associated with using more cannabis than intended. Microaggressions were not associated with using more alcohol or cannabis than intended.

Discussion

The current study provides insights into risk factors for drinking and using more cannabis than intended among a population at high risk for alcohol and cannabis use disorders, SMWGD. Nearly 40% of observations included drinking more than intended, which is similar to estimates from other studies.^10,12 Drinking more than intended was a risk factor for alcohol consequences among SMWGD, while this was not the case for cannabis. Several risk factors for drinking and using more cannabis than intended were shared, including using in social contexts, and social, enhancement, and conformity motives. However, some risk factors were substance specific, including drinking with SGM and heterosexuals and coping motives for cannabis use. Together findings provide a foundation for future research on risk factors for using more cannabis than intended, motivational risk factors for using more than intended, and unique risk factors for SGM.

Consistent with prior research,^10,11 drinking more than intended was associated with more consequences, but results were more nuanced for cannabis. Intending to be intoxicated for longer and (marginally) intending to become more intoxicated predicted fewer consequences. One potential explanation for this finding is that individuals may have engaged in more protective behavioral strategies on days when they planned to engage in heavier cannabis use. Future research should explore factors that explain differences in results for drinking or using more cannabis than intended, such as planned or utilized protective behavioral strategies.

Social, enhancement, and conformity motives were associated with using more alcohol and cannabis than intended, while coping motives for cannabis use only predicted becoming more intoxicated than intended. This pattern is consistent with findings which highlight the social nature of drinking more than intended.^10,12 Associations with both social and conformity motives suggest that multiple motives may arise in social contexts, including desire to enhance a social gathering and because others were drinking or using cannabis. Coping motives for cannabis use were associated with becoming more intoxicated than intended but not becoming intoxicated for longer than intended. This suggests that coping motives may contribute to initial overconsumption of cannabis but may not contribute to continued heavier consumption throughout the day. Given that this was the first study we are aware of to examine associations between motives and using more than intended, future research should determine whether the same associations are found among cisgender, heterosexuals or if they are unique to SMWGD. As results for social and conformity motives are consistent with risk factors for drinking more than intended identified in other studies with the general population, these risk factors are likely to generalize to the general population, but this should be empirically tested as existing research has not yet examined associations between drinking motives and drinking more than intended in the general population.

Prior studies have produced mixed evidence for an association between drinking more than intended and social drinking. We found that social drinking was associated with drinking more than intended, consistent with Labhart, Anderson and Kuntsche ¹² but not Fairlie, Cadigan, Patrick, Larimer and Lee ¹¹. Different operationalizations of drinking more than intended may help to explain mixed findings. The current study and Labhart, Anderson and Kuntsche ¹² operationalized drinking more than intended as the number of additional drinks consumed above one’s planned drinks. Fairlie, Cadigan, Patrick, Larimer and Lee ¹¹, who did not find an association between social drinking and drinking more than intended, focused on occasions when individuals engaged in heavy episodic drinking that was not intended. These divergent findings suggest that drinking more than intended and unplanned heavy episodic drinking may have distinct risk factors. We also extended research on social drinking and drinking more than intended to cannabis use, demonstrating that social cannabis was associated with using more cannabis than intended.

We examined two potential SGM-specific risk factors, using substances with SGM and microaggressions. Drinking more than intended was more likely when participants drank with other SGM, as predicted. However, drinking with straight, cisgender individuals was also associated with drinking more than intended. This pattern is consistent with two prior studies that have examined these potential risk factors.^24,25 Researchers have proposed that drinking with SGM and cisgender, heterosexual individuals may both be associated with heavier use via different mechanisms (e.g., more permissive norms when drinking with SGM; using to cope with anxiety about stigma when drinking with heterosexuals).^24,25 Alternatively, this may represent an artifact of the influence of drinking with larger groups since larger groups may be more likely to include both SGM and heterosexual individuals and drinking in larger groups is associated with drinking more than intended.¹² Of note, using cannabis with SGM or heterosexual individuals did not predict using more cannabis than intended. We also found no support for an association between microaggressions and using more than intended. This finding is consistent with the nonsignificant associations between coping and most using more than intended variables, as coping is a mechanism through which microaggressions contribute to heavier substance use. Future research should explore potential factors that may moderate potential associations between microaggressions and using more than intended, such as motivational or contextual factors.

The current study should be considered in light of its limitations. First, only sexual minority cisgender women and gender diverse individuals assigned female at birth who used alcohol or cannabis regularly, lived in the US, and were between 18 and 25 were included. It is unclear whether similar patterns will be found among sexual minority men, those assigned male at birth, SGM who live outside the US, older individuals or adolescents, or those who use alcohol or cannabis less frequently. Second, alcohol and cannabis use intentions were only assessed during evening surveys when participants reported not using alcohol or cannabis since the prior survey. While we did not expect substance use intentions to differ on days when substances were used compared to when they were not, it is possible that this may have affected results. Third, cannabis quantity was not included, and thus, only subjective cannabis intoxication was examined.

Conclusions

Previously unexamined risk factors for drinking more than intended emerged from this study, including social, enhancement, and conformity motives. Only one SGM-specific risk factor predicted drinking more than intended (i.e., drinking with SGM). This study was the first to test whether risk factors for drinking more than intended also applied to using more cannabis than intended, identifying several shared risk factors and one unique risk factor for using more cannabis than intended. Future research with the general population or mixed SGM and cisgender, heterosexual samples should examine whether these associations are similar for these two populations and explore potential mechanisms underlying associations between risk factors and using more than intended.

Public Health Significance:

Study findings have potential implications for preventive and harm reduction intervention efforts for sexual minority women and gender diverse individuals. For example, interventions may encourage SGM young adults to adjust their protective behavioral strategies to reduce the potential for consequences even if they drink/use more cannabis than intended.