Effects of regular cannabis and nicotine use on acute stress responses chronic nicotine, but not cannabis use, is associated with blunted adrenocortical and cardiovascular responses to stress

Abstract

Rationale

Cannabis is one of the most prevalent substances used by tobacco smokers and, in light of the growing list of states and territories legalizing cannabis, it is expected that co-use of cannabis and nicotine will escalate significantly and will lead to continuing challenges with tobacco use.

Objectives

This study was conducted to examine the interactive effects of chronic cannabis and nicotine use on adrenocortical, cardiovascular, and psychological responses to stress and to explore sex differences in these effects.

Methods

Participants (N = 231) included cannabis-only users, nicotine-only users, co-users of both substances, and a non/light-user comparison group. After attending a medical screening session, participants completed a laboratory stress session during which they completed measures of subjective states, cardiovascular responses, and salivary cortisol during baseline (rest) and after exposure to acute stress challenges.

Results

Nicotine use, but not cannabis use, was associated with blunted cortisol and cardiovascular responses to stress across both men and women. Men exhibited larger cortisol responses to stress than women. Co-users had significantly larger stress-related increases in cannabis craving than cannabis-only users. Cannabis users reported smaller increases in anxiety during stress than cannabis non/light-users, and both male nicotine-only users and male cannabis-only users experienced significantly smaller increases in stress than their non/light-user control counterparts.

Conclusions

This study replicates and extends earlier research on the impacts of sex and nicotine use on stress responses, and it provides novel findings suggesting that when co-used with nicotine, cannabis use may not confer additional alterations to physiological nor subjective responses to stress. Co-use, however, was associated with enhanced stress-related craving for cannabis.

Introduction

The focus of this paper is to examine, for the first time, the interactive effects of chronic cannabis and nicotine use on adrenocortical, cardiovascular, and psychological responses to stress and to explore sex differences in these effects. A growing number of states and territories have legalized recreational cannabis, and most states and territories have legalized medical use of cannabis (National Conference of State Legislatures 2020). Cannabis is the second most prevalent substance of abuse among nicotine users in the USA, with close to half of nicotine users using cannabis regularly (Administration 2013a; Ramo et al. 2013). Moreover, its rate of use is expected to escalate significantly with changing cannabis legislation and regulations (Administration 2013b; Ahrnsbrak et al. 2016; Hurd et al. 2014; Lev-Ran et al. 2013; Leweke and Koethe 2008; Office 2012; Peters et al. 2012; Steigerwald et al. 2018). Simultaneous use of cannabis and nicotine is also growing (Akre et al. 2010; Barrett et al. 2006; Conway et al. 2018; Meier et al. 2020; Soldz et al. 2003), which is likely to perpetuate challenges in reducing the tobacco use rate.

Exposure to stressful events is one of the most common triggers of nicotine use and relapse (al’Absi 2006; 2007; Brown et al. 2009; Heishman 1999; Matheny and Weatherman 1998; Shiffman 2005). Stress increases the frequency of smoking, and it accelerates progression towards relapse (al’Absi 2006; Carey et al. 1993; Doherty et al. 1995; Kenford et al. 2002; Koval and Pederson 1999; Shiffman et al. 1996), especially in the presence of other negative affective states, such as anxiety, irritability, depression, and craving (Nakajima and al’Absi 2012; 2013; Pomerleau et al. 2005; Swan et al. 1996; Xu et al. 2008). Research also shows that nicotine use is associated with enhanced basal cortisol activity and attenuated cortisol responses to stress (al’Absi et al. 2003; Gilbert et al. 1997; Kirschbaum et al. 1994; Kirschbaum et al. 1993a, b; Tsuda et al. 1996). Studies suggest similar connections between stressful life events and cannabis use (Agrawal et al. 2012; El-Sawy and Abd Elhay 2011; Karriker-Jaffe 2013; Stone et al. 2012; Van Ryzin et al. 2012; Wu et al. 2014), and studies suggest that smoking nicotine or using cannabis can cause various psychobiological changes associated with emotion regulation and stress responses (Adamopoulos et al. 2008; al’Absi et al. 2005; Bruijnzeel et al. 2009; Childs and de 2009; Vanderkaay and Patterson 2006; Wise and Munn 1995).

Given the role of cannabis in stimulating endocannabinoids and the role of these chemicals in regulating stress responses (Hillard et al. 2016; Volkow et al. 2017), long-term cannabis use may influence stress responses. For example, there is research suggesting long-term changes in the endocannabinoid system in response to chronic cannabis use (Fox et al. 2013), although research on long-term effects of cannabis use on hypothalamic–pituitary–adrenal (HPA) responses is scant. Limited research indicates effects of regular cannabis use on heart rate (Li et al. 2005), subjective, and cortisol responses to stress (Cuttler et al. 2017; Somaini et al. 2012). Although other studies have found cortisol responses in the opposite direction (Fox et al. 2013) or not at all (Mizrahi et al. 2013). In light of the limited focus of early studies on cannabis use only, important questions related to co-use of cannabis with other substances, including nicotine, still remain. Therefore, we sought to examine the effects of regular cannabis-nicotine co-use on responses to acute stress.

Additional research indicates the importance of investigating sex differences when studying the impacts of cannabis and nicotine use on stress responses. For instance, there is evidence demonstrating sex differences in hormonal and cardiovascular responses to stress, indicating greater physiological responses to stress in men than in women (al’Absi et al. 1999; al’Absi et al. 2004b; Kirschbaum et al. 1995). Clinical and epidemiological researchers have also documented a greater frequency of stress reports among women with addiction problems compared to men with addiction problems (Hughes et al. 1991; Kalman 2002; Swan et al. 1993); stress is more frequently cited as a reason for drug use and relapse in women than in men (Alati et al. 2004; Green 2006; Green et al. 2004; Najavits et al. 1998). Cannabis use is also associated with greater negative symptom profiles in women (Lev-Ran et al. 2012), possibly due to women’s enhanced vulnerability to developing cannabis use disorder and their enhanced sensitivity to the subjective effects of cannabis compared to men (Cooper and Haney 2014). These findings highlight the need to examine sex differences when examining the impacts of cannabis and nicotine use on stress responses.

To date, no study has examined the effects of co-use of nicotine and cannabis on cortisol nor cardiovascular stress responses. In addition, while research has shown different patterns of associations between these substances and affect regulation and motivations for use (Farris et al. 2014; Martens and Gilbert 2008), we do not know the impact of co-use on affective and craving responses to acute stress. In this study, we examine the impacts of regular cannabis use, regular nicotine use, and regular co-use of cannabis and nicotine on the psychobiological mechanisms of stress responses. We also examine sex differences in stress responses in users of each substance and in co-users. We predicted that regular co-use of cannabis and nicotine would contribute to greater blunting in stress response measures compared to non/light-users of these substances and to users of either substance alone. We also predicted that females, compared to males, would exhibit greater negative affect and weaker stress responses.

Methods

Participants

Data from two similar stress-response studies were merged prior to testing our predictions. In both studies, adults aged 18–70 were recruited to participate in research that was advertised as examining stress responses of healthy adults who do or do not use cannabis regularly. For both studies, recruitment took place within two Minnesota communities (Duluth and Minneapolis), using flyers and online advertisements (Craigslist, Reddit, Facebook, local newspapers, etc.). Interested individuals completed a pre-screening phone interview or online survey to determine whether they met preliminary eligibility criteria. Those who met preliminary eligibility criteria were invited to an in-person medical screening to further assess eligibility via questionnaires on REDCap (Harris et al. 2009), interviews, and urine screens.

To be eligible, participants were required to have stable health (no negative changes during the last 6 months) and they could not drink an average of > 2 alcoholic drinks per day. Participants could not use illicit substances (other than cannabis), nor could they have recent history of drug/alcohol treatment, diagnosis of bipolar or psychotic disorder, pending legal action/supervision, liver disease, cancer, nor heart disease. To be eligible, participants had to either (a) use cannabis an average of 10 + days per month for 1 + years and have a positive urine screen for tetrahydrocannabinol (THC, > 50 ng/mL Onescreen 10-panel dip-card, American Screening Corporation) at the medical screening or (b) use cannabis 0–3 days per month for the last 1 + years and have a negative urine test for THC.

Procedures

Participants who met eligibility criteria after the in-person medical screening were scheduled for a future stress lab session that would begin at approximately 1 p.m. to control for diurnal variation in cortisol levels. Upon arriving at the lab, participants were fitted with a blood pressure cuff and then they completed a baseline resting period for 30 min during which they watched nature documentaries. Next, participants performed acute stress tasks (public speaking, mental arithmetic, and cold-pressor challenges) that have been used extensively in previous research on acute stress (al’Absi et al. 2019; al’Absi et al. 2015; Nakajima et al. 2019; Shaw and al’Absi 2008). For the public speaking task, participants were provided with a topic and given 4 min to prepare a speech that was then delivered for 4 min in front of the experimenter and a video camera for later evaluation. For the mental arithmetic challenge, participants were provided a 3-digit seed number (e.g., 132) and they were instructed to sum the three digits of that number (e.g., 1 + 3 + 2 = 6) and then add this sum to the seed number for their answer (e.g., 6 + 132 = 138). The participant’s answer (e.g., 138) would then serve as the new seed number, and they were to continue the series of calculations for an 8-min period. If an incorrect answer was provided, participants were instructed to try again beginning with their last correct response. For the cold-pressor challenge, participants submerged their hand (up to their wrist) in a bucket of ice water and they were asked to rate their pain every 15 s for 90 s (or until maximum tolerable pain was reached), after which they removed their hand and placed it on a towel while continuing to rate their pain ever 15 s for another 90 s. After the stress challenges, participants completed a 60-min recovery resting period during which they continued to watch nature documentaries.

Cardiovascular measures were assessed every 5 min during baseline and the first 30 min of the recovery resting periods and every 2 min during the speech and math stressor tasks. Participants provided saliva samples and completed measures of subjective states upon arrival at the lab, after the baseline resting period (pre-stress), immediately after completing the set of stress tasks, and after approximately 30 min of recovery.

Measures/apparatus

Participant characteristics

Participants self-reported demographic information (sex, age, race/ethnicity, and education), nicotine use habits (age began regular use, daily diary records, dependence via the Fagerström Test for Nicotine Dependence, FTND (Heatherton et al. 1991)), and cannabis use habits (age began using, number of use days per month). They also completed the 10-item perceived stress questionnaire (PSS; (Cohen et al. 1983)) and a depression inventory (CES-D; (Radloff 1977)). Lab staff also measured height and weight to calculate body mass index (BMI).

Cortisol

Participants provided saliva samples using Salivette® (Sarstedt) cotton swabs. Samples were centrifuged and aliquoted into cryovials for storage at − 20 °C until assayed in batches using luminescence immunoassay kits from IBL International. Inter- and intra-assay coefficients of variation ranged from 6.73 to 10.09% and from 5.99 to 8.49%, respectively.

Subjective states

Participants reported the extent to which they felt happy, cheerful, stressed, overwhelmed, and anxious as well as their desire to smoke (cigarettes) and their desire to use marijuana using 8-point scales that ranged from not at all to very strongly. Within each measurement timepoint, responses to the happy and cheerful items were averaged to create an index of positive mood (rs = 0.78 to 0.86); responses to the stressed and overwhelmed items were averaged to create an index of stressed mood (rs = 0.71 to 0.78) (DeAngelis and al’Absi 2020).

Cardiovascular measures

Systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), and mean arterial pressure (MAP) were measured using CARESCAPE™ V100 DINAMAP™ monitors. An index was created for each outcome within each of three time periods—baseline rest, during stress (speech prep, speech delivery, and mental math), and during recovery—by averaging measures within the period.

Statistical approach

Prior to analyses, participants were classified into two groups based on their cannabis use: participants were classified as regular cannabis users if they used cannabis 10 or more days per month, and all others were classified as cannabis “non/light-users” (i.e., used 3 days or less per month). Analysis using a continuous measure of cannabis use, number of days per month, was also conducted, and the results were consistent with the results obtained from the categorical comparisons. Participants were also classified into two groups based on their nicotine use: participants were classified as nicotine users if they used nicotine daily, and all others were classified as nicotine “non/light-users.” Due to the multi-modal nature of the data collected about nicotine use, which included qualitative data, it was not feasible to convert all sources of data into a continuous measure of nicotine use. Therefore, only the dichotomous nicotine variable was used to assess the impact of nicotine use.

Means, medians, standard deviations, and counts were used to summarize participant characteristics. Previous studies have found that age (Brindle et al. 2014; Uchino et al. 2010) and BMI (Garafova et al. 2014; Torres et al. 2014) are related to cardiovascular stress responses; therefore, we used ANOVA to examine potential differences in these participant characteristics based on sex, nicotine use, and cannabis use (see Supplementary Materials). We also examined potential differences in perceived stress and depression (see Supplement). Next, we examined whether baseline (pre-stress) values of the dependent measures differed based on sex, nicotine use, and cannabis use (see Supplement). Significant interactions were probed by examining simple effects and pairwise comparisons.

To examine potential group differences in response to stress, we created change scores for each of the dependent measures by subtracting pre-stress levels from levels immediately after (cortisol, mood, and craving) or during (cardiovascular) the acute stress tasks. Change scores were then examined using ANOVA (cortisol, mood, and craving) and ANCOVA (cardiovascular) with sex (2: male and female), nicotine use (2: user and non/light-user), and cannabis use (2: user and non/light-user) entered as fixed factors (and with age and BMI entered as covariates). Pearson correlations were used to examine correspondence among changes in the dependent measures.

Results

Participant characteristics

A total of 231 participants were included in these analyses. Table 1 presents a summary of participant characteristics with respect to cannabis/nicotine use and sex. Additional information is available in the Supplementary Materials.

Table 1.

Participant characteristics

Variable	Cannabis- only users	Co-users of nicotine and can nabis	Non/ light-user controls	Nicotine- only users	All females	All males	Full sample
Group size	68	86	46	31	108	123	231
Male sex: #	34	57	21	11	0	123	123
Hispanic: #	11	4	4	0	9	10	19
White/Caucasian: #	55	67	33	27	83	99	182
Age: M (SD)	29.0 (10.3)	30.6 (10.5)	35.9 (13.5)	39.1 (13.3)	33.5 (12.6)	31.3 (11.3)	32.3 (12.0)
Education beyond high school: #	63	69	43	21	94	102	196
BMI: M (SD)	25.8 (5.5)	28.4 (7.6)	26.0 (5.5)	30.1 (6.8)	27.7 (6.4)	27.1 (6.9)	27.4 (6.7)
Age began regular nicotine use (years): M (SD)	–	17.2 (4.4)	–	17.7 (4.7)	17.6 (6.1)	17.2 (2.8)	17.3 (4.5)
FTND: M (SD)	–	3.2 (2.3)	–	3.7 (2.1)	3.6 (2.5)	3.1 (2.1)	3.3 (2.3)
Age began using cannabis (years): M (SD)	17.2 (3.4)	15.7 (2.2)	–	20.8 (6.3)	17.1+ (3.4)	15.9+ (2.5)	16.3+ (2.9)
Annual cannabis use days: M (SD) Mdn	341 (53)	337 (65)	0 (0)	3 (8)	344+ (51)	335+ (65)	339+ (60)
	365	365	0	0	365	365	365

N sample size, # number/frequency count, M mean, SD standard deviation, Mdn median, FTND Fagerström Test for Nicotine Dependence (Heatherton et al., 1991)

⁺Includes only regular cannabis users (i.e., cannabis-only and co-user groups)

Cortisol, mood, cravings, and cardiovascular measures

Baseline measures

Baseline measures of the subjective reports showed significant group differences in positive mood (significant three-way interaction between sex, nicotine use, and regular cannabis use) and significant sex differences in baseline SBP. No other significant effects emerged in predicting baseline cortisol, subjective states, and cardiovascular measures. Details on these analyses and findings are provided in the Supplemental Materials.

Changes in cortisol

In predicting change in cortisol in response to acute stress, significant main effects emerged for nicotine use and for sex, Fs(1, 178) > 4.26, ps ≤ 0.04, partial η²s ≥ 0.02, and η²s ≥ 0.02. No other main effects nor interactions were significant, Fs < 2.90, ps > 0.09. Pairwise comparisons indicated that nicotine users had significantly smaller increases in cortisol (EMM = 0.24, SE = 0.32) than nicotine non/light-users (EMM = 1.09, SE = 0.26); females had significantly smaller increases in cortisol (EMM = 0.17, SE = 0.27) than males (EMM = 1.16, SE = 0.31).

Changes in mood

There was a main effect of nicotine use on change in positive mood, F(1, 221) = 4.38, p = 0.04, partial η² = 0.02, and η² = 0.02; non/light-users of nicotine experienced greater reductions in positive mood (EMM = − 1.12, SE = 0.13) than nicotine users (EMM = − 0.72, SE = 0.15). No other main effects nor interactions were significant in predicting change in positive mood, Fs(1, 221) < 2.79, ps > 0.09. Significant main effects of sex and regular cannabis use and significant 2-way interactions between nicotine use and both cannabis use and sex were qualified by a significant 3-way interaction between sex, nicotine use, and regular cannabis use in the model predicting change in stressed mood, F(1, 222) = 12.12, p < 0.001, partial η² = 0.05, and η² = 0.05. The simple effect of sex was significant for non/light-user controls and for nicotine-only users, Fs(1, 222) ≥ 5.07, ps < 0.03; the simple effect of cannabis use was significant for males who were non/light-users of nicotine, F(1, 222) = 16.79, p < 0.001; the simple effect of nicotine use was significant for males who were non/light-users of cannabis, F(1, 222) = 15.09, p < 0.001. Pairwise comparisons indicated that both male nicotine-only users (EMM = 0.32, SE = 0.46) and male cannabis-only users (EMM = 0.79, SE = 0.26) experienced significantly smaller increases in stress than their non/light-user control counterparts (EMM = 2.55, SE = 0.34). In addition, among non/light-user controls, males had significantly larger increases in stress (EMM = 2.55, SE = 0.34) compared to females (EMM = 1.52, SE = 0.31). Among nicotine-only users, males had significantly smaller increases in stress (EMM = 0.32, SE = 0.46) than their female counterparts (EMM = 2.34, SE = 0.35).

A main effect of cannabis use emerged for change in anxious mood, F(1, 222) = 5.63, p = 0.02, partial η² = 0.02, and η² = 0.02. No other main effects nor interactions were significant, Fs < 3.17, ps > 0.07. Cannabis users had significantly smaller increases in anxiety (EMM = 1.47, SE = 0.16) than cannabis non/light-users (EMM = 2.15, SE = 0.24).

Changes in craving

No significant effects emerged for sex, cannabis use, nor their interaction in predicting change in nicotine craving among nicotine users, Fs(1, 112) ≤ 1.80, ps > 0.18. A significant main effect of nicotine use emerged for change in cannabis craving among cannabis users, F(1, 149) = 4.28, p = 0.04, partial η² = 0.03, and η² = 0.03. No other main effects nor interactions were significant, Fs < 2.56, ps > 0.11. Nicotine users had significantly larger increases in cannabis craving (EMM = 0.54, SE = 0.17) than non/light-users of nicotine (EMM = 0.03, SE = 0.18).

Changes in cardiovascular measures

Significant main effects of nicotine use emerged for change in SBP, F(1, 186) = 5.57, p = 0.02, partial η² = 0.03, and η² = 0.03, change in HR, F(1, 185) = 5.80, p = 0.02, partial η² = 0.03, and η² = 0.03, and change in MAP, F(1, 187) = 4.33, p = 0.04, partial η² = 0.02, and η² = 0.02. No other main effects nor interactions emerged for these dependent measures (SBP Fs ≤ 1.92, ps > 0.16; HR Fs < 1.61, ps > 0.20; and MAP Fs ≤ 3.62, ps > 0.05); no significant effects emerged for change in DBP, Fs(1, 187) < 3.45, ps > 0.06. Compared to nicotine non/light-users, nicotine users had smaller increases in SBP (EMM_users = 13.3, SE = 1.3 and EMM_{non/light-users} = 17.2, SE = 1.0), HR (EMM_users = 7.2, SE = 0.8 and EMM_{non/light-users} = 9.8, SE = 0.7), and MAP (EMM_users = 11.5, SE = 0.9 and EMM_{non/light-users} = 14.0, SE = 0.8).

Associations among changes in cortisol, mood, cravings, and cardiovascular measures

As can be seen in Table 2, changes in cigarette craving were negatively related to both SBP and MAP (∣rs∣≥ 0.22, ps < 0.05), but there were no other relations between physiological measures and cravings for cigarettes or cannabis. Changes in cortisol were significantly related to changes in cardiovascular measures (rs ≥ 0.26, ps < 0.01), but not to changes in mood (∣rs∣≤ 0.04, ps > 0.54) nor in cravings (∣rs∣≤ 0.12, ps > 0.27). Among cannabis users, changes in cannabis craving were positively related to changes in stress and anxious moods (rs = 0.22, ps < 0.01); changes in cannabis craving were positively related to changes in nicotine craving among co-users (r = 0.26, p < 0.05). Among nicotine users, changes in cigarette craving were not significantly related to changes in mood (∣rs∣≤ 0.18, ps > 0.05).

Table 2.

Pearson correlations (N) among change in DVs

	Cortisol change	Positive mood change	Stressed mood change	Anxious mood change	Cigarette craving change	Cannabis craving change	SBP change	DBP change	HR change	MAP change
Cortisol change
Positive mood change	0.02 (186)
Stressed mood change	− 0.03 (186)	− .36** (230)
Anxious mood change	− 0.04 (186)	− .35** (230)	.59** (230)
Cigarette craving change	− 0.12 (89)	− 0.18 (116)	− 0.01 (116)	0.04 (116)
Cannabis craving change	− 0.01 (123)	− 0.08 (154)	.22** (154)	.22** (154)	.26* (86)
SBP change	.39** (160)	0.10 (197)	0.09 (197)	0.03 (197)	− .23* (96)	− 0.11 (132)
DBP change	.30** (161)	0.04 (198)	0.07 (198)	0.06 (198)	− 0.17 (97)	− 0.05 (134)	.66** (196)
HR change	.26** (161)	.14* (198)	0.06 (198)	0.05 (198)	− 0.04 (97)	− 0.03 (133)	.52** (196)	.32** (197)
MAP change	.37** (161)	0.04 (198)	0.13 (198)	0.08 (198)	− .22* (97)	− 0.07 (134)	.80** (196)	.84** (197)	.31** (197)

SBP systolic blood pressure, DBP diastolic blood pressure, HR heart rate, MAP mean arterial pressure

^*Correlation is significant at the 0.05 level (2-tailed)

^**Correlation is significant at the 0.01 level (2-tailed)

To explore the impact of time since last use of nicotine and cannabis prior to each session on cravings, we conducted a series of correlation analyses between the time since last use of nicotine and cannabis and respective craving measures and found no significant correlations (ps > 0.05). We also conducted a two-sample t-test to compare outcome measures (cortisol and craving) between those who consumed alcohol in the past 24 h and those who did not. In all cases, there was no evidence that consuming alcohol within the past 24 h had an impact on the outcome measures (ps > 0.05).

Discussion

This study examined the impact of regular cannabis use, nicotine use, and co-use on acute stress responses. To our knowledge, this is the first study to examine the impact of regular co-use of cannabis and nicotine on adrenocortical responses to stress. Although the results do not suggest additive nor the expected interactive effects of regular cannabis and nicotine use, they provide evidence that nicotine use, independent of cannabis use, was associated with blunted cortisol and cardiovascular responses to stress across both men and women. The results also show that regular cannabis use, independent of nicotine use, was associated with blunted affective responses to stress. Larger increases in cannabis craving were also observed among those who co-use cannabis and nicotine relative to cannabis-only users.

Our study provides a partial replication and an extension of recent studies on cannabis use (Cuttler et al. 2017; De Angelis and al’Absi 2020) and of studies on nicotine use (al’Absi 2018; al’Absi et al. 2003; Ginty et al. 2014; Lovallo et al. 2019; Sorocco et al. 2006) by examining the effects of co-use of these substances on acute stress responses. Chronic nicotine use has been associated with alterations in physiological systems involved in stress responses, with nicotine users showing decreased cortisol and SBP responses to acute stress (Badrick et al. 2007; Ginty et al. 2014; Rohleder and Kirschbaum 2006). This study replicated those findings, demonstrating that nicotine use was associated with blunted cortisol, SBP, HR, and MAP stress responses. These blunted stress responses among nicotine users may reflect long-term effects of nicotine use on stress response systems, leading to increased risk for maintaining use (to compensate for such dysregulation) and to increased vulnerability to stress-related disorders, such as depression (al’Absi 2020; al’Absi et al. 2005; al’Absi et al. 2004a; McEwen et al. 2016). Mechanisms for these blunted responses may involve disruption of systems that regulate the stress response, including the opioid system as well as the catecholaminergic and dopaminergic systems (al’Absi et al. 2020; Benitez-López et al. 2019; Burke and Miczek 2014; Sami et al. 2015).

Although acute stress in the current study produced the expected increases in cardiovascular and cortisol measures, unlike a previous study (Cuttler et al., 2017), we did not observe blunted stress responses in cortisol among regular users of cannabis, nor did we observe significant effects of cannabis use on cardiovascular measures. It is possible that our study found different results for cortisol stress reactivity due to the acute stress tasks or due to the nature of our within-subject study design. Previous investigators (Cuttler et al., 2017) used a different stress induction and a between-subject design, wherein the investigators found a significant between-group difference in cortisol responses for non-users who were in stress vs. no-stress groups, but no significant difference between cannabis users who were in a stress group vs. a no-stress group.

Though we did not find significant effects of regular cannabis use on cardiovascular nor cortisol stress responses, our finding that cannabis users experienced blunted affective responses (stressed mood and anxious mood) to stress is consistent with our previous assessment of the effects of regular cannabis use on stress responses (DeAngelis and al’Absi 2020). Contrary to expectations, this study did not find additive nor the expected non-additive effects of cannabis and nicotine on stress responses. Thus, cannabis use may not confer additional risk of stress-response dysregulation in those who are already affected by chronic nicotine use. Alternatively, the effect size for the impacts of cannabis use or co-use may have been too small to affect the physiological measures assessed here. These possibilities need to be addressed in future research.

One intriguing observation in this study was the larger increase in cannabis craving among those who co-use cannabis and nicotine relative to cannabis-only users. This suggests the possibility of an interactive effect of the two substances in inducing cannabis craving, although this was not the case for nicotine craving. It is worth noting that prior to the lab session, participants were not restricted from using either substance. Therefore, these results may be limited to absence of significant withdrawal effects of either substance, which require between 12 and 48 h to manifest (Budney et al. 2003; Tiffany and Drobes 1991; West and Hack 1991). Nevertheless, in light of the role of craving in increasing nicotine relapse (al’Absi et al. 2017; Lemieux et al. 2016; Nakajima and al’Absi 2012), it is important to follow up on enhanced cannabis cravings experienced by co-users in future studies and in the context of nicotine cessation.

This study provides evidence of sex differences in multiple measures. Our observation of smaller cortisol responses to stress in women compared to men is consistent with previous studies (Alabsi et al. 2015; Kotlyar et al. 2017; Lovallo et al. 2015; Stephens et al. 2016). In addition, we found that sex moderated stress-related changes in mood. Both male nicotine-only users and male cannabis-only users experienced significantly smaller increases in stressed mood than male non/light-users, while these differences were not observed in females. Also, among non/light-user controls, males had significantly larger increases in stress compared to females. Among nicotine-only users, males had significantly smaller increases in stressed mood than their female counterparts. In combination, these findings suggest that cannabis use in males, but not in females, was associated with smaller increases in negative affect in response to acute stress.

We should note limitations of this study. While carefully designed and implemented, the observed effects do not allow us to draw conclusions about causal relations between substance use and stress responses. This study also did not include a substance deprivation period prior to inducing acute stress; therefore, we cannot rule-out potential lingering effects of acute use, particularly during the initial baseline/rest measurement period. These limitations notwithstanding, we were able to assess stress responses with minimal impact of withdrawal, which, by itself, can cause changes in stress measures. Future research on co-use of nicotine and cannabis should examine the impact of withdrawal from either or both of these substances on mood and stress response measures. We note that the cannabis-only and non/light-user control groups included a few participants with very light use of nicotine. Though not likely, an effect of such inclusion could have reduced effect sizes for the between group comparisons. Of note, the nicotine effect was the most robust in the presented results. Furthermore, while we did not observe additive nor the expected non-additive effects of chronic cannabis and nicotine use on stress-related measures, it is possible that this may be attributable to our classification of individuals as cannabis users if they used 10 + cannabis use days per month (rather than requiring daily cannabis use, as was used for nicotine classification). It is also possible that the magnitude of substance-use effects varies across the two substances, potentially due to the neurobiological substrates of each substance (i.e., nicotine vs. cannabis). We acknowledge that these differences may be attributable to different patterns of use as well as motivators for use. Due to the observational nature of this study, we cannot compare directly the impact of the different levels of dependence between these two substances. Parametric and experimental studies are needed to address such a question. Finally, although we found no differences for changes in cortisol and craving between those who did and those who did not consume alcohol in the past 24 h, there may be long-term effects of alcohol use that could affect these responses to stress. We did not collect each participant’s alcohol history, which would be necessary to investigate these effects.

In summary, cannabis use alone or when combined with nicotine use was not associated with significant alterations in physiological responses to stress. Nicotine use, with or without cannabis co-use, was associated with blunted cortisol and cardiovascular responses to stress across both men and women. The results provide novel findings indicating that chronic cannabis use does not appear to exert additional disruption to cardiovascular and cortisol stress responses, emphasizing the importance of addressing the impact of nicotine addiction on physiological stress response dysregulation in nicotine-cannabis co-users.