Unplanned vs. planned simultaneous alcohol and cannabis use in the daily lives of a predominantly White college student sample: What are the motives, contexts, and outcomes?

Abstract

Objective:

Simultaneous alcohol and cannabis use is common among college students and associated with more consumption and consequences compared to single-substance use. This study examined occasions of simultaneous use and compared planned occasions to unplanned occasions with respect to motives, contexts, consumption, and consequences.

Method:

College student simultaneous users (N=341; 53% women; 74% White) completed five daily surveys for 54 days. Mixed-effects models examined motives and contexts of simultaneous use occasions as a function of whether alcohol and cannabis use were 1) both planned vs. 2) unplanned, no use planned or 3) unplanned, single-substance use planned; and whether alcohol and cannabis consumption and negative simultaneous use-related consequences varied across planned vs. unplanned occasions.

Results:

Social and enhancement motives were related to planned simultaneous use; offered and coping motives were associated with planned single-substance use that became simultaneous use (vs. planned simultaneous use). Compared to unplanned use, planned simultaneous use was associated negatively with using at home or alone, and positively with using with others, more intoxicated people, and more people using cannabis. Planned simultaneous use was associated with more alcohol and cannabis consumption. No significant differences were found for negative consequences.

Conclusions:

Planned simultaneous use was motivated by social and enhancement reasons, whereas planned single-substance use that became simultaneous use was more likely motivated by offers or for coping. Planned simultaneous use resulted in greater consumption, but not negative consequences. Results provide specific motives and contexts associated with unplanned and planned simultaneous use to be incorporated into real-time interventions.

Public health significance:

This study highlights the importance of understanding predictors and outcomes of planned simultaneous alcohol and cannabis use compared to simultaneous use that was not planned (either no plans for any use or plans for either alcohol or cannabis but not both). Positive-reinforcement motives and social-orientated context may be important targets for risky planned simultaneous use occasions.

Simultaneous alcohol and cannabis use (also sometimes called simultaneous alcohol and marijuana use; SAM) is common in the U.S. (Subbaraman & Kerr, 2015, 2020), particularly among college students (White et al., 2019). Simultaneous use has been linked to increased alcohol and cannabis consumption and negative consequences compared to single-substance use (Ito et al., 2021; Lee et al., 2020; Linden-Carmichael et al., 2020; Metrik et al., 2018). However, almost no research has explored whether college students plan for simultaneous use or if it is unplanned and incidental, for example, unintentionally using cannabis after having been drinking, which has implications for its relations to etiological correlates (e.g., motives, contexts) and outcomes (Pearson & Henson, 2013).

The Theory of Planned Behavior posits that intentions immediately precede behavior (Ajzen, 1991). A growing body of research has studied the intention-behavior gap (Sheeran & Webb, 2016), that is, when a given behavior occurs without prior intention. Unplanned substance use has been studied using daily diary and ecological momentary assessment (EMA) for alcohol-only use (Fairlie et al., 2019; Lauher et al., 2020; Stevens et al., 2017) and for cannabis-only use (Emery et al., 2020). Recent work from our group examined outcomes of planned versus unplanned drinking, indicating that unplanned drinking was related to fewer drinks consumed and fewer negative consequences endorsed. These findings suggest that planned drinking events may result in riskier drinking and thus be important scenarios to target for interventions focused on reducing risky drinking. Therefore, understanding the motivational and contextual factors underlying planned simultaneous use (versus unplanned simultaneous use) may have implications for additional harm reduction strategies (Stevens et al., 2021a). In the first study of unplanned vs. planned simultaneous alcohol and cannabis use, Fairlie et al. (2021) found that 51.85% of simultaneous-use days were planned and that planned simultaneous use was associated with heavier consumption but not positive or negative consequences. However, this study did not consider plans for single-substance use that resulted in simultaneous use in their definition of unplanned simultaneous use or examine predictors of unplanned and planned simultaneous use.

Motives for substance use are considered the most proximal antecedent of substance use (Cooper et al., 2016), and a burgeoning literature demonstrates that motives for alcohol, cannabis, and simultaneous use vary day-to-day (Armeli et al., 2016; Buckner et al., 2019; Stevens et al., 2021b; Votaw & Witkiewitz, 2021, for a review). The Model of Unplanned Drinking Behavior postulates that motives may differentially relate to unplanned vs. planned use (Pearson & Henson, 2013). In an EMA study of college student drinkers, Stevens et al. (2021c) found that drinking to enhance positive affect was related to planned drinking. Stevens et al. (2021a) corroborated these relations in a multi-site sample of college students and found that drinking to cope or because alcohol was offered was related to unplanned drinking. Boyle et al. (2021) expanded upon this finding in a qualitative study of young adult heavy drinkers, showing that many participants did not plan to use cannabis during a drinking event but did so simply because it was available and offered to them. Other than this qualitative study, motives have not been investigated as predictors of unplanned or planned simultaneous use. Notably, simultaneous use could be explicitly motivated by plans to get “cross-faded” (i.e., to be both drunk and high; Patrick et al., 2020; Patrick & Lee, 2018), which is a motive that is unique to using multiple substances.

Physical and social contexts of unplanned and planned simultaneous use also have not been investigated but can inform the content and delivery of intervention targets. Planned drinking has been consistently linked to weekends (Fairlie et al., 2019; Lauher et al., 2020), celebrations (Fairlie et al., 2019), and to specific social and physical contexts (e.g., party, with friends; Stevens et al., 2021a). Unplanned drinking was associated with drinking at home, alone, or with roommates (Stevens et al., 2021a). However, it is necessary to understand whether these findings hold when examining unplanned and planned simultaneous alcohol and cannabis use, as it is possible, for example, that students who go to a party plan to consume only alcohol but end up also using cannabis because it is readily available in a social setting.

Present Study

Whereas several studies have examined predictors and outcomes of unplanned drinking, no quantitative study to our knowledge has examined predictors of simultaneous alcohol and cannabis use as a function of whether it was planned vs. unplanned. The purpose of the present study was to examine (1) the motives and contexts of unplanned and planned simultaneous use and (2) whether unplanned or planned simultaneous use was associated with alcohol and cannabis consumption and negative simultaneous use-related consequences. Our definition of unplanned simultaneous use included occasions where neither alcohol nor cannabis use was planned as well as occasions where only one substance was planned (i.e., plans for alcohol use but not cannabis and plans for cannabis use but not alcohol). Consistent with prior research, we expected that using for social reasons or to have fun would be related to planned simultaneous use, whereas using alcohol and/or cannabis because it was offered or to cope would be linked to unplanned simultaneous use. Likewise, we hypothesized that engaging in simultaneous use at a party, at a friend’s place, with friends, with more intoxicated people, and with more people using cannabis would be associated with planned simultaneous use, whereas using at home or alone would be related to unplanned simultaneous use. Consistent with Fairlie et al. (2021), we hypothesized that planned vs. unplanned simultaneous use would be linked to greater alcohol and cannabis consumption but not to negative simultaneous use-related consequences.

Methods

Design and Sample

Baseline/follow-up surveys.

Full-time college students (N=1,390) were recruited to participate in a larger parent study on simultaneous alcohol and cannabis use from universities in three states with legalized medical cannabis and varying recreational cannabis laws (School A: recreational use illegal; School B: recreational use decriminalized; School C: recreational use legal for adults age 21+). Eligible students (i.e., between ages 18-24, enrolled full-time in one of the three schools, endorsed past-year alcohol and cannabis use) first completed a baseline online survey for $25 compensation. Three months later students completed a follow-up online survey for $35 compensation. See White et al. (2019) for further details regarding initial screening, recruitment, and sample exclusions.¹

Repeated daily surveys.

Of those who completed the baseline survey, 693 used alcohol and cannabis at the same time ‘so that their effects overlapped’ (i.e., simultaneous use) within the past month, deeming them eligible for the repeated daily survey (RDS) phase. Invited participants were stratified based on frequency of past-month simultaneous use and sex assigned at birth by overenrolling frequent simultaneous users (i.e., three or more times in the past month) and male participants to ensure sufficient rates of simultaneous use and a balance of participants by sex. Of the 379 students invited, 343 students (90.5%) agreed to participate in this phase. The sample size for the RDS phase was determined from a formal power analysis in the original application for funding.

Data collection directly followed the baseline and three-month follow-up surveys and comprised 28 days of RDS at each burst (56 total days). Surveys were prompted at 9:00am, 2:00pm, 5:00pm, 8:00pm, and 11:00pm using a custom smartphone application (see Jackson et al., 2021). For the 9:00am survey, students reported on substance use from the last survey until bedtime and also information about the prior day such as consequences experienced. Participants were provided five hours to complete the 9:00am survey and two hours to complete all other surveys.² Compensation was $1 for each completed daily survey, with weekly and overall bonuses (for a total possible compensation of $200 per burst). Study procedures were approved by the coordinating university’s Institutional Review Board and a Certificate of Confidentiality was obtained.

Due to technical difficulties that occurred in the first two days, 54 study days and 341 students were retained (53% women; M_age=19.79; 74% White, 11% Asian, 9% bi- or multi-racial, 3% Black, 0.7% American Indian, 0.2% Pacific Islander, 2% other race; 10% Hispanic/Latinx). See Jackson et al. (2021) for a description of the full repeated daily survey sample. To address the present study’s aims, our analytic sample was restricted to simultaneous use occasions (n=2,267) nested within 275 students.³ Data and research materials are available upon request from the corresponding author. This study’s design and its analysis were not pre-registered.

Measures

Demographics.

Participants self-reported age and birth sex (‘1’ for male, ‘0’ for female) at the baseline survey.

Unplanned vs. planned alcohol and cannabis simultaneous use.

At each RDS, regardless of whether alcohol or cannabis were used, participants were asked: “Had you planned to drink (use marijuana) between [time X] and [time Y]?” Response options included yes or no. The timeframe for both questions was from the time the previous survey was submitted to the time the current survey began. Simultaneous-use occasions were classified into three categories: an unplanned simultaneous use occasion when no use was planned (referred to as no use planned) (‘0’) was defined by a participant endorsing ‘no’ to planning to drink and endorsing ‘no’ to planning to use cannabis (10% of all simultaneous-use occasions); an unplanned simultaneous use occasion when only single-substance use was planned (referred to as single-substance use planned) (‘1’) was defined by a participant endorsing ‘yes’ to either planning to drink or to use cannabis but not both (18%);⁴ and a planned simultaneous-use occasion (‘2’) was defined by a participant endorsing ‘yes’ to both planning to drink and to use cannabis (73%).

Simultaneous-use motives.

Participants were asked at each RDS that they endorsed simultaneous use: “What motivated you to drink and use marijuana between [time X] and [time Y]?” Participants were instructed to select all motives from the following: “to be social” (social; 45% of simultaneous-use occasions), “to cope” (coping; 12%), “it was offered” (offered; 28%), “to have fun” (enhancement; 85%), “to fit in” (conformity; 2%), “expand awareness” (expansion; 9%), “get higher from another drug” (cross-fading; 5%), and “was too high from the other drug” (offset; 0.71%). Motives were adapted from psychometrically-valid measures of drinking (Cooper, 1994), cannabis use (Simons et al., 1998), and simultaneous alcohol and cannabis use (Patrick et al., 2018) motives. All motives were analyzed in the present study except for conformity and offset motives (due to base rates <5%).

Simultaneous-use contexts.

Physical contexts. Participants reported on their physical contexts at each RDS: “Where were you while you were using alcohol and marijuana?” (select all that apply): home (52% of simultaneous-use occasions), friend’s place (36%), party (13%), bar/restaurant (7%), outside (3%), study space (0.22%), athletic facility (0.31%), elsewhere (5%). Given base rates, we only considered home, friend’s place, party, and bar/restaurant. Social contexts. Participants reported on their social context of simultaneous use at each RDS: “Who were you with while you were using alcohol and marijuana?” (select all that apply): alone (12% of simultaneous-use occasions), significant other (21%), roommate (32%), friend (67%), family (4%), strangers (8%), acquaintance (10%), and others (2%). We included all social contexts except for using with others or with family due to low base rates of these two contexts. Strangers/acquaintances were collapsed into a single social context because similar effects were found when examined separately. Number of people reported being intoxicated with the participant at a given RDS was examined, with options for none (30% of simultaneous-use occasions), some (22%), most (24%), and all (24%); number of people using cannabis with the participant at a given RDS was also examined (M=3.10; SD=3.13).

Alcohol and cannabis consumption.

At each RDS, participants indicated the number of drinks using a graphical interface, tapping on the timeline at each specific time a drink was consumed (see Jackson et al., 2021, for screenshots): “Tap your finger in the blue box each time you had a drink at the corresponding time” (M=3.63 drinks per occasion, SD=2.77). An analogous procedure was used for cannabis consumption, with participants instructed to tap at each specific time they used cannabis (M=2.91 cannabis uses per occasion, SD=3.23).

Negative simultaneous use-related consequences.

On the morning survey following a simultaneous-use day, participants indicated whether the following consequences occurred “because of yesterday’s use of alcohol and marijuana together”: hangover (17% of daily observations), nausea/vomiting (6%), hurt self (0.99%), drove car drunk/high (6%), blackout (2%), rude/aggressive (2%), and unwanted sex (0.53%). As part of the parent RDS study, we considered consequences across several validated measures (Kahler et al., 2005; Read et al., 2006; Simons et al., 2012; White et al., 2005; White & Labouvie, 1989), selecting acute consequences expected to vary at the daily level. We examined endorsement of any (yes/no) negative consequence on a given day, given the low endorsement of consequences (none=72.60% of simultaneous-use days; any=27.40%).

Covariates.

We included age, sex (male vs. female [ref]), school (i.e., recruitment site; ref=School C), weekend (i.e., Friday and Saturday vs. weekday [Sunday-Thursday]=ref), and other drug use (other than alcohol, cannabis, or tobacco) on a given occasion (ref=no) as covariates.

Analytic Strategy

Data management was conducted in SAS 9.4™ software (SAS Institute Inc., 2012). Analyses were restricted to simultaneous-use occasions (n=2,267; Level-1), nested within participants (N=275; Level-2).⁵ Generalized linear mixed-effects models (GLMMs) in SAS PROC GLIMMIX were used to account for the three-category nominal outcome (unplanned simultaneous use when no use was planned, unplanned simultaneous use when single-substance use was planned, and. planned simultaneous use). A similar approach was used for the follow-up analyses that also included the two distinct single-substance use planned categories (planned alcohol use-unplanned cannabis use and unplanned alcohol use-planned cannabis use). Consistent with recommendations, Level-1 occasion-level variables were person-mean centered, and Level-2 between-person variables were grand-mean centered (Curran & Bauer, 2011; Hox & Maas, 2005). Level-2 variables for binary indicators (motives, contexts) were defined as the proportion of endorsement across study days for a given person. Each motive, social context and physical context was tested in a separate model for interpretability. Significant findings and directions of effects were unchanged when tested in multivariate models. We also conducted sensitivity analyses using the BH procedure that controls for false discovery rate (Benjamini & Hochberg, 1995; Glickman et al., 2014; see Results below).

Models with alcohol consumption (negative binomial), cannabis consumption (negative binomial), and negative consequences (binomial) as outcomes included the three-level nominal planning variable as the independent variable. To disaggregate within- and between-person effects in these models, dummy codes were created for planned simultaneous use (vs. no use planned), and planned simultaneous use (vs. single-substance use planned). A proportion of endorsement of each dummy code across study days was constructed and then used as the Level-2 variable. All models included age, sex, and school as Level-2 covariates and weekend and other drug use as Level-1 covariates.

Results

See Table 1 for descriptive statistics of motives, contexts, alcohol and cannabis consumption, and simultaneous use-related consequences for simultaneous use occasions when no use was planned (unplanned simultaneous use-no use planned), only alcohol or only cannabis were planned (unplanned simultaneous use-single substance use planned), and use of both substances was planned (planned simultaneous use). Most planned and unplanned simultaneous use was reported at the 9:00am report, which assessed behavior from the 11:00pm survey until bedtime (43% of all simultaneous-use occasions), followed by the 11:00pm report (29%), 8:00pm report (19%), 5:00 pm report (7%), and the 2:00pm report which assessed behavior since wake time (2%).

Table 1.

Descriptive statistics of motives, contexts, consumption, and negative consequences by unplanned and planned simultaneous alcohol and cannabis use

	Unplanned Simultaneous Use- No Use Planneda n = 224	Unplanned Simultaneous Use- Single Substance Use Plannedb n = 398	Planned Simultaneous Usec n = 1,645
Motives
Social	74 (33.04%)	147 (36.93%)	807 (49.06%)
Coping	35 (15.63%)	57 (14.32%)	175 (10.64%)
Offered	68 (30.36%)	131 (32.91%)	425 (25.84%)
Enhancement	137 (61.16%)	312 (78.39%)	1,479 (89.91%)
Conformity	4 (1.79%)	5 (1.26%)	37 (2.25%)
Expansion	12 (5.36%)	38 (9.55%)	165 (10.03%)
Cross-fading	10 (4.46%)	14 (3.52%)	79 (4.80%)
Offset	3 (1.34%)	3 (0.75%)	10 (0.61%)
Physical Contexts
Home	113 (50.45%)	233 (58.54%)	837 (50.88%)
Friend’s place	74 (33.04%)	113 (28.39%)	624 (37.93%)
Party	12 (5.36%)	37 (9.30%)	248 (15.08%)
Bar/restaurant	11 (4.91%)	34 (8.54%)	121 (7.36%)
Outside	8 (3.57%)	15 (3.77%)	49 (2.98%)
Study space	1 (0.45%)	3 (0.75%)	1 (0.06%)
Athletic facility	1 (0.45%)	1 (0.25%)	5 (0.30%)
Elsewhere	12 (5.36%)	17 (4.27%)	90 (5.47%)
Social Contexts
Alone	41 (18.30%)	57 (14.32%)	185 (11.25%)
Significant other	39 (17.41%)	78 (19.60%)	361 (21.95%)
Roommate	43 (19.20%)	132 (33.17%)	554 (33.68%)
Friends	133 (59.38%)	229 (57.54%)	1,149 (69.85%)
Family	7 (3.13%)	20 (5.03%)	77 (4.68%)
Strangers	16 (7.14%)	24 (6.03%)	150 (9.12%)
Acquaintance	12 (5.36%)	41 (10.30%)	177 (10.76%)
Others	10 (4.46%)	13 (3.27%)	31 (1.88%)
Number of intoxicated people
‘None’	115 (51.34%)	151 (38.04%)	415 (25.23%)
‘Some’	36 (16.07%)	86 (21.66%)	372 (22.61%)
‘Most’	35 (15.63%)	75 (18.89%)	436 (26.50%)
‘All’	38 (16.96%)	85 (21.41%)	422 (25.65%)
Number of people using cannabis	M = 1.77 (SD = 2.16)	M = 2.53 (SD = 2.64)	M = 3.41 (SD = 3.28)
Alcohol and Cannabis Consumption and Negative Simultaneous Use-related Consequences
Number of drinks	M = 3.28 (SD = 2.35)	M = 3.25 (SD = 2.57)	M = 3.77 (SD = 2.85)
Number of cannabis uses	M = 2.52 (SD = 1.96)	M = 2.63 (SD = 2.26)	M = 3.03 (SD = 3.54)
Any negative consequence	40 (29.85%)	58 (24.68%)	261 (27.74%)

Note. n = 2,267 occasions.

^aUnplanned simultaneous use-No use planned involves unplanned alcohol use and unplanned cannabis use on a given occasion.

^bUnplanned simultaneous use-Single substance use planned involves either unplanned alcohol use and planned cannabis use or unplanned cannabis use and planned alcohol use on a given occasion.

^cPlanned simultaneous use involves planned alcohol use and planned cannabis use on a given occasion.

Predictors of Unplanned vs. Planned Simultaneous Use

Motives.

At the within-person level, using for social or enhancement reasons was associated with higher odds of planned simultaneous use vs. unplanned simultaneous use, no use planned and higher odds of planned simultaneous use vs. unplanned simultaneous use, single-substance use planned. Using because it was offered or for coping reasons was linked to higher odds of planned single-substance use that became simultaneous use compared to planned simultaneous use. All other within-person effects were not statistically significant. At the between-person level, using for social and enhancement reasons was related to higher odds of planned simultaneous use (vs. no use planned). Enhancement motives were also positively associated with planned simultaneous use (vs. single-substance use planned). Using because it was offered was linked to lower odds of planned simultaneous use (vs. single-substance use planned). Other between-person effects were not statistically significant (see Table 2). See Supplemental Table 1 for covariate effects.

Table 2.

Associations between motives and contexts and unplanned vs. planned simultaneous alcohol and cannabis use at the within-person (occasion) level and the between-person level.

Predictor	Planned simultaneous usea (Ref = Unplanned simultaneous use-No use plannedb)		Planned simultaneous use (Ref = Unplanned simultaneous use-Single substance use plannedc)
	OR	95% CI	OR	95% CI
Motives
Within-person level
Offered	0.90	(0.58, 1.40)	0.66	(0.48, 0.93)
Coping	0.66	(0.34, 1.30)	0.58	(0.36, 0.96)
Social	2.15	(1.41, 3.27)	2.47	(1.79, 3.40)
Enhancement	2.44	(1.54, 3.87)	2.62	(1.76, 3.90)
Expansion	2.29	(0.90, 5.81)	1.21	(0.69, 2.13)
Cross-fading	1.30	(0.55, 3.04)	1.11	(0.55, 2.25)
Between-person level
Offered	0.32	(0.13, 0.80)	0.34	(0.16, 0.71)
Coping	0.60	(0.17, 2.10)	1.05	(0.37, 2.97)
Social	2.50	(1.11, 5.63)	1.07	(0.56, 2.05)
Enhancement	47.20	(18.88, 118.04)	4.41	(1.80, 10.79)
Expansion	3.05	(0.65, 14.25)	0.85	(0.28, 2.58)
Cross-fading	1.17	(0.10, 13.73)	3.43	(0.40, 29.54)
Contexts
Within-person level
Physical Contexts
Home	0.55	(0.35, 0.85)	0.47	(0.34, 0.66)
Party	2.95	(1.41, 6.16)	2.18	(1.35, 3.53)
Bar/restaurant	1.89	(0.83, 4.29)	0.93	(0.03, 4.06)
Friend’s place	1.60	(1.05, 2.43)	1.93	(1.39, 2.68)
Social Contexts
Alone	0.43	(0.25, 0.74)	0.59	(0.38, 0.93)
Friends	2.45	(1.58, 3.80)	2.34	(1.69, 3.24)
Roommate	1.06	(0.62, 1.82)	0.78	(0.54, 1.12)
Significant other	1.76	(0.96, 3.24)	1.47	(0.92, 2.35)
Stranger/acquaintance	1.23	(0.71, 2.14)	1.16	(0.77, 1.73)
Number of intoxicated people	1.86	(1.54, 2.25)	1.62	(1.40, 1.87)
Number of people using cannabis	1.23	(1.13, 1.34)	1.13	(1.07, 1.19)
Between-person level
Physical Contexts
Home	1.26	(0.56, 2.82)	1.09	(0.57, 2.09)
Party	5.48	(1.44, 20.89)	1.14	(0.45, 2.90)
Bar/restaurant	2.15	(0.29, 15.95)	0.65	(0.14, 2.96)
Friend’s place	1.28	(0.55, 3.01)	1.38	(0.69, 2.76)
Social Contexts
Alone	0.39	(0.09, 1.62)	1.06	(0.29, 3.79)
Friends	1.30	(0.52, 3.27)	1.25	(0.59, 2.65)
Roommate	3.71	(1.59, 8.66)	1.21	(0.63, 2.29)
Significant other	1.33	(0.55, 3.24)	1.03	(0.51, 2.09)
Stranger/acquaintance	2.37	(0.58, 9.61)	1.16	(0.41, 3.30)
Number of intoxicated people	1.95	(1.31, 2.90)	1.02	(0.75, 1.38)
Number of people using cannabis	1.55	(1.33, 1.81)	1.14	(1.03, 1.27)

Note. OR = odds ratio.

^aPlanned simultaneous use involves planned alcohol and planned use cannabis use on a given occasion.

^bUnplanned simultaneous use-No use planned involves unplanned alcohol and unplanned cannabis use on a given occasion.

^cUnplanned simultaneous use-Single substance use planned involves either unplanned alcohol use and planned cannabis use or unplanned cannabis use and planned alcohol use on a given occasion. Each motive, social context, and physical context was tested in a separate model for interpretability. Significant findings and directions of effects were unchanged when tested in a multivariate model. All models included age, sex (male vs. female = ref), school (School A, School B vs. School C = ref), weekend (vs. weekday = ref), and other drug use (ref = no) as covariates. See Supplemental Tables S1 and S2 for covariate effects. Statistically significant effects (p < .05) are in bold typeface.

Contexts.

At the within-person level, using at home or alone was associated with lower odds of planned simultaneous use (vs. no use planned and vs. single-substance use planned). Using at a party, friend’s place, with friends, with more intoxicated people, and with more people using cannabis was linked to higher odds of planned simultaneous use (vs. unplanned use and vs. planned single-substance use). At the between-person level, using at a party, with a roommate, with more intoxicated people, and with more people using cannabis was associated with higher odds of planned simultaneous use (vs. no use planned). Using with more people using cannabis was also linked to higher odds of planned simultaneous use (vs. singe-substance use planned). All other within- and between-person effects were not statistically significant (see Table 2).⁶ See Supplemental Table S2 for covariate effects.

Follow-up Planned Single-Substance Use Comparisons

We also conducted follow-up analyses using a four-category outcome with unplanned simultaneous use when single-substance use was planned separated into unplanned alcohol use-planned cannabis use and planned cannabis use-unplanned alcohol use. At the within-person level, using for social or enhancement reasons was related to higher odds of planned alcohol use-unplanned cannabis use, whereas using for expansion reasons was linked to lower odds of planned alcohol use-unplanned cannabis use. Also, at the within-person level, using at home or alone was related to lower odds of planned alcohol use-unplanned cannabis use, whereas using at a party, with friends, with strangers/acquaintances, with more intoxicated people, and with more people using cannabis was associated with higher odds of planned alcohol use-unplanned cannabis use. A similar pattern of findings was evident at the between-person level (see Table 3).⁷ See Supplemental Table S4 for covariate effects.

Table 3.

Associations between motives and contexts and planned single-substance use on simultaneous alcohol and cannabis use occasions

Predictor	Planned alcohol use-Unplanned cannabis use (Ref = Planned cannabis use-Unplanned alcohol use)
	OR	95% CI
Motives
Within-person level
Offered	1.14	(0.61, 2.11)
Coping	0.75	(0.28, 1.97)
Social	2.66	(1.48, 4.80)
Enhancement	2.10	(1.06, 4.13)
Expansion	1.08	(0.30, 3.95)
Cross-fading	0.90	(0.24, 3.48)
Between-person level
Offered	1.81	(0.51, 6.41)
Coping	0.16	(0.02, 1.16)
Social	1.73	(0.56, 5.32)
Enhancement	0.59	(0.13, 2.69)
Expansion	0.01	(<0.01, 0.14)
Cross-fading	0.28	(0.01, 16.74)
Contexts
Within-person level
Physical Contexts
Home	0.52	(0.28, 0.96)
Party	3.25	(1.21, 8.70)
Bar/restaurant	2.15	(0.83, 5.58)
Friend’s place	1.21	(0.66, 2.20)
Social Contexts
Alone	0.19	(0.06, 0.55)
Friends	2.68	(1.44, 5.01)
Roommate	1.01	(0.49, 2.09)
Significant other	0.90	(0.39, 2.05)
Stranger/acquaintance	2.58	(1.21, 5.51)
Number of intoxicated people	2.80	(2.11, 3.71)
Number of people using cannabis	1.23	(1.11, 1.36)
Between-person level
Physical Contexts
Home	0.14	(0.05, 0.43)
Party	40.07 a	(6.59, 243.72)
Bar/restaurant	3.75	(0.28, 51.09)
Friend’s place	2.98	(0.89, 9.92)
Social Contexts
Alone	0.01	(<.01, 0.18)
Friends	7.48	(2.04, 27.37)
Roommate	0.48	(0.15, 1.49)
Significant other	0.51	(0.14, 1.81)
Stranger/acquaintance	2.98	(0.46, 19.51)
Number of intoxicated people	3.74	(2.21, 6.33)
Number of people using cannabis	1.26	(1.04, 1.53)

Note. OR = odds ratio. Each motive, social context and physical context was tested in a separate model for interpretability. Significant findings and directions of effects were unchanged when tested in a multivariate model. All models included age, sex (male vs. female = ref), school (School A, School B vs. School C = ref), weekend (vs. weekday = ref), and other drug use (ref = no) as covariates. Statistically significant effects (p < .05) are in bold typeface.

^aFew observations (n=8) were observed for those endorsing being at a party when only planning to use cannabis (but not alcohol), which resulted in a large odds ratio and a wide confidence interval. Statistically significant effects (p < .05) are in bold typeface. See Supplemental Table S3 for covariate effects.

Outcomes of Unplanned vs. Planned Simultaneous Use

At the within-person level, planned simultaneous use (vs. no use planned) was associated with greater alcohol consumption (incidence rate ratio [IRR]=1.32, p<.01) and greater cannabis consumption (IRR=1.35, p<01). Planned simultaneous use (vs. single-substance use planned) was linked to greater alcohol consumption (IRR=1.24, p<.01) at the within-person level and to greater cannabis consumption at the between-person level (IRR=1.53, p<.01). There was no statistically significant association between type of simultaneous use occasions (planned simultaneous use vs. unplanned use when no use was planned and unplanned use when single-substance was planned) and negative simultaneous use-related consequences.⁸ See Supplemental Table S4 for covariate effects.

Discussion

We examined the motives and contexts of simultaneous alcohol and cannabis use that was unplanned versus planned, as well as outcomes of these distinct simultaneous-use occasions. Our methodology extended prior work by examining plans for alcohol use and cannabis use separately, which identified a unique type of unplanned simultaneous use that has not been studied in prior work – a planned single-substance use occasion that becomes a simultaneous-use occasion. We also assessed intentions for use at the simultaneous occasion, which provided a detailed examination of these constructs at the occasion-level rather than at the day-level. Better understanding of the motives and contexts of these distinct simultaneous-use occasions will inform the content of interventions (i.e., targeting intentions when simultaneous use is planned vs. motives and contexts when simultaneous use is unplanned).

Corroborating prior research on unplanned drinking (Boyle et al., 2021; Stevens et al., 2021a; 2021c), plans for any substance use (whether it be planned simultaneous or planned single-substance) was motivated by social and enhancement reasons. These findings are especially relevant for unplanned simultaneous use occasions when only alcohol was planned vs. occasions when only cannabis was planned. Our findings provide additional context around competing theories about whether cannabis acts as a substitute (planned single-substance use) or complement (planned simultaneous use) to alcohol use (Risso et al., 2020; Subbaraman, 2016). Note, however, that because motives (and contexts) combined “drink and use marijuana” on occasions of simultaneous use, we were not able to parse out unique motivations for planned alcohol and unplanned cannabis use or the converse (or indeed whether there were unique motivations for alcohol or cannabis individually even when both were planned).

Our examination of simultaneous use context yielded findings consistent with the one study to date that has examined contexts of unplanned vs. planned drinking (Stevens et al., 2021a). Simultaneous use that was not planned (either no use planned or single-substance use planned) was more likely to occur at home or when alone. It is possible that an occasion of unplanned simultaneous use involved a user who was initially at home who ended up going out later and engaging in unplanned use. Using at home and alone was particularly likely on an unplanned simultaneous use occasion when only cannabis was planned as opposed to occasions when only alcohol was planned. Being at home, where alcohol may be readily available, may offer an opportunity to drink without any planning. Also, being home and not having to worry about transportation may also lead an individual to engage in unplanned drinking after engaging in planned cannabis use.

Planned simultaneous use was more likely to occur at a party or friend’s place and with friends, more intoxicated people, and more people using cannabis. Social events are often planned in advance (e.g., earlier in the day) and it might be that on planned use occasions participants are out with their friends because they had made plans to be outside of the home (e.g., for campus activities such as football games, or for celebrations such as birthdays or holiday events). Simultaneous use occasions where only alcohol use had been planned were also more likely to occur in social contexts (e.g., at a party, with friends, with others also drinking or using cannabis) than simultaneous use occasions where only cannabis use had been planned. Students may plan to attend social events that involve drinking; fortuitous use of cannabis use may then occur with little planning required. In addition, availability may be a factor in whether people engage in unplanned cannabis use. Party contexts where others may be supplying cannabis may present opportunities for individuals to engage in cannabis use when not planned. These findings could inform prevention efforts aimed at intervening in situations where unplanned substance use may occur and highlight the importance of providing students with refusal skills or psychoeducation around the impact of cannabis on subjective effects of alcohol.

Unique to simultaneous use, we did not find associations between cross-fading motives and unplanned or planned simultaneous use. This could be due to few simultaneous-use occasions being driven by cross-fading reasons in this sample (5%); future research should determine whether this null effect is replicable, as cross-fading motives implies some intentionality for simultaneous use and would be expected to relate to planned simultaneous use.

Extending the limited literature in this area, the findings suggest that a planned single-substance use event may become a simultaneous-use occasion at times when participants are using substances to cope and/or because alcohol or cannabis is offered to them. The move from planning to use either alcohol or cannabis (but not both) to using both substances also occurs when students are at home or alone. These specific motives and contexts could serve as in-the-moment targets to send protective behavioral strategies to individuals to reduce excess consumption and unwanted consequences on that occasion (Bravo et al., 2019). When using only one substance, coping motives and using alone have been identified as risk factors for negative consequences (Cooper et al., 2016; Mason et al., 2020; Waddell et al., 2021), so it is especially concerning that this motive and context might facilitate planned single-substance use becoming unplanned simultaneous use, which is inherently riskier than single-substance use (Jackson et al., 2020; Linden-Carmichael et al., 2020). Because we did not compare motives and contexts when planning to use a single substance (i.e., either alcohol or cannabis) and actually using only it to those when planning to use a single substance but using both, these suggestions are only tentative.

Engaging in simultaneous use because the product was offered may be difficult to target in the moment, suggesting that prevention messages be targeted to all users to prevent simultaneous use when only single-substance use was planned. Alternatively, future work in this area could identify the social and physical contexts that are most associated with offers for alcohol and cannabis and target these contexts in mobile health (mHealth) interventions to reduce unplanned simultaneous-use events. Planned single-substance use could result in significant ordering effects if another substance is used on that same occasion, with prior work showing that the first substance used on a given simultaneous-use occasion is consumed more heavily than the second substance (Gunn et al., 2020). When only single-substance use is planned, protective behavioral strategies could be tailored to whichever substance the individual intends to use, as it will likely be used first on this occasion. For example, if a simultaneous user endorses plans for alcohol but not cannabis use, strategies could focus on alcohol use to reduce consumption and unwanted consequences (e.g., alternating alcoholic and nonalcoholic beverages), with consideration of the contexts under which the intentional substance was used.

We supported findings by Fairlie et al. (2021) by showing that planned simultaneous use (vs. simultaneous use when both substances were unplanned) was associated with greater alcohol and cannabis consumption but was not linked to more negative simultaneous use-related consequences. The association between greater alcohol and cannabis consumption and planned simultaneous use corresponds with findings within the unplanned drinking literature, which have found planned drinking to be related to increased consumption (Stevens et al., 2021a). Regarding consequences, future research is needed to determine whether consuming more than intended is significantly related to negative consequences on an unplanned simultaneous use occasion, as has been shown for unplanned heavy drinking (Fairlie et al., 2019). It may also be that there are specific types of consequences that are more likely to be experienced (e.g., physical vs. social, or positive vs. negative) on planned vs unplanned occasions of simultaneous use; future research is needed to assess this possibility. Finally, it might also be that the amount of consumption is more important for experiencing simultaneous use-related consequences than planning or not planning to engage in simultaneous use (Fairlie et al., 2021).

Strengths, Limitations, and Future Directions

This is the first quantitative study to examine motives and contexts of unplanned and planned simultaneous alcohol and cannabis use and only the second study to investigate outcomes of unplanned and planned simultaneous use. We provided a nuanced examination of these constructs by studying occasion-level associations (~3-hour reporting window) and disaggregating within-person from between-person effects. We identified several motives and contexts that are linked to planned simultaneous use. When simultaneous use is planned, however, it may be more practical to target intentions for use on a given occasion to reduce alcohol and cannabis consumption. We also identified motives and contexts associated with unplanned simultaneous use that could serve as prevention and intervention targets in the absence of intentions for use.

Findings should be considered alongside their limitations. Data were drawn from a large sample of college students who were majority White; thus, our findings may not generalize to other racial/ethnic groups or non-college student populations. Data were self-report, and surveys required a degree of retrospection. Intentions for a specific amount of consumption were not assessed, and intentions for use were assessed at the time or after the use occurred. Future work should assess intentions prospectively and in greater detail, particularly plans for amount and intoxication. This would provide a better understanding of whether consequences vary as a function of consumption levels of simultaneous use, as has been shown for drinking (Fairlie et al., 2019). To reduce response burden, for simultaneous use occasions, we assessed motives for and contexts of using “alcohol and cannabis” and did not differentiate motives for and contexts of simultaneous use from those specific to using alcohol or those specific to cannabis. Thus, any motives and contexts specific to each substance could not be teased apart. Given the inevitable overlap between social and physical contexts, future research could extend this work by examining interactions between contexts, which may identify high-risk contextual combinations (e.g., using at home while alone) (Jackson et al., 2021).

We also did not examine number of contexts on a given occasion as an indicator of planned vs. unplanned simultaneous use, though prior research has shown that using in multiple contexts is associated with high-risk use (Terry-McElrath et al., 2017). This study was unable to determine whether motives and contexts shifted during a simultaneous-use occasion; future research could determine whether a shift (e.g., from coping early on to then being social later on or from being home to going to a party) explains why planned single-substance use becomes a simultaneous-use event. Research also is needed to determine whether planned simultaneous use is associated with positive consequences such as bonding with friends, as the present study did not assess positive outcomes of simultaneous use. Finally, the present study provided new information about simultaneous use occasions that were planned vs. unplanned; future research could compare different types of unplanned simultaneous use occasions as well as compare times when neither alcohol nor cannabis were planned but were used (unplanned simultaneous use) to occasions where neither substance was used, consistent with intention.