Due to each follow-up visit occurring approximately 4-weeks apart, we potentially captured women around the same phase of the menstrual cycle, and thereby similar P4/E2 ratios, at each assessment. This was to some degree reflected in the average values of P4, E2, and the P4/E2 ratio of our sample. However, the CV estimates for our sample indicated moderate within-person variability across the active medication phase, which was also supported by the larger standard deviations when examining raw P4/E2 values, suggesting some degree of variation in P4/E2 between repeated measures during the active medication phase of this study. Future studies are needed to examine these effects in a larger sample with equal proportions of women in the luteal and follicular phase throughout the trial. Further, gathering assessments every 2 weeks to observe within person changes in response to the combination of varenicline plus naltrexone on smoking and drinking outcomes would be important to further understand the preliminary findings reported herein. The present study must be interpreted in light of strengths and weaknesses. Strengths include hormonal assays of both P4 and E2 at each follow-up appointment, comprising six hormonal assays per participant across the 12 weeks. Limitations include a small sample size and the assessment period of 4 weeks, which potentially limited within-person assessment across distinct phases of the menstrual cycle. While studies have found a relationship between sex hormones and smoking behavior , when summarized across a 4-week period,cannabis drying racks some of those nuanced effects may be less detectable. Another limitation includes the lack of objective measurements of alcohol consumption, such as EtG or CDT. The addition of these objective alcohol consumption measurements may have provided additional validation to self-report alcohol consumption.
In conclusion, this study provides preliminary evidence on the potential role of the effect of biomarkers of sex differences, namely P4/E2 ratio, among female heavy drinking smokers undergoing treatment. While a majority of analyses were null, there was an intriguing medication by hormone level interaction such that varenicline plus naltrexone increased percent days abstinent among women with greater P4/E2 ratio compared to the varenicline plus placebo condition. Additional studies of these effects in larger samples are warranted and sex hormones offer important information above and beyond comparing groups on the bases of sex.It is now widely recognized that cigarette smoking and drinking are behaviorally and clinically intertwined. Epidemiological data from the 2019 National Survey on Drug Use and Health found that 21.7% of adults reported using cigarettes over the last year . The same survey found 10.7% reported daily cigarette use and 6.3% reported heavy alcohol use over the past month . Individuals with a mild or moderate alcohol use disorder have exhibited a smoking prevalence 2 times higher than those without an AUD . The trend continues with those with a severe AUD having a smoking prevalence 3 times higher than those without an AUD . Conversely, nicotine use has also been shown to impact alcohol use. Individuals with any nicotine use disorder are 2.5 times more likely to meet criteria for a 12- month AUD diagnosis, and 3.2 times more likely to meet criteria for a lifetime AUD diagnosis . The effect of nicotine on alcohol use is so robust that even low levels of nicotine have been shown to increase quantity of alcohol consumption. In comparison to nonsmokers, non-daily smokers may experience an increased risk for hazardous drinking and for receiving a DSM-IV alcohol diagnosis . Previous studies have made salient how co-use of both substances may increase their acute rewarding effects , and promote cross-tolerance of both substances , further perpetuating co-use. The negative health consequences from cigarette use and alcohol use , including increased risk for various cancers and cardiovascular diseases, underscores the need to concomitantly reduce the use of both substances. Alcohol use hinders smoking cessation attempts. Over the last two decades, the number of adult cigarette smokers who have engaged in a smoking cessation attempt has increased . The average smoker attempts to quit smoking multiple times with estimates ranging from 6 to upwards of 30 or more attempts before one is successful in sustaining abstinence from cigarettes for at least 1 year .
Previous studies found that moderate and heavy alcohol use increases the risk of smoking lapses during a smoking cessation attempt . A recent study by Lynch and colleagues found that in comparison to nondrinkers, at 1-month post smoking cessation treatment, moderate drinkers experienced greater odds of continued smoking, however, at 7- months post-treatment, the odds of continued smoking were not different to that of nondrinkers. For heavy drinkers, in comparison to non-drinkers, the increased odds of continued smoking remained for 1-month and 7-months post-treatment . These studies underscore the importance of addressing alcohol co-use, particularly at higher levels, during an initial smoking quit attempt and thereafter. A previous clinical trial integrated a brief alcohol intervention in the context of smoking cessation treatment and found greater smoking abstinence among those who received integrated treatment compared to those who only received standard smoking cessation treatment . Transitioning results such as these into clinical practice, it is recommended that those trying to quit smoking limit or abstain from alcohol as much as possible . Taken together, these findings highlight the need for interventions that can simultaneously address smoking cessation and drinking reduction to target the amplified, negative effects of conjoint use in treatment. One of the main limitations of the field has been that clinical trials focus either on smoking or on drinking as a primary outcome, but rarely target both behaviors simultaneously. To address this limitation, our group has recently completed a 12-week clinical trial combining varenicline and naltrexone for smoking cessation and drinking reduction in a sample of heavy drinking daily smokers . Varenicline, a selective nicotinic acetylcholine partial agonist, is an FDA approved medication for smoking cessation . While naltrexone, a non-selective opioid receptor antagonist, is an FDA approved for the treatment of alcohol use disorder . This recently completed clinical trial provides a unique opportunity to examine behavior change across the two substances as both measures of alcohol and cigarette consumption were assessed concurrently throughout the clinical trial. This study randomized heavy drinking smokers to receive varenicline plus placebo or varenicline plus naltrexone.
Results indicated that smoking abstinence at 26 week follow-up was significantly higher in the varenicline plus placebo group, compared to the varenicline plus naltrexone group . For the primary drinking outcome of drinks per drinking day,pots for cannabis plants there was a main effect of medication in favor of the combined medication group at the 12 week end of medication phase; however, this effect was not sustained at the 26 week follow-up . These results shed light on the impact of combination pharmacotherapy in the context of smoking cessation and drinking reduction. The present study leverages data from the aforementioned clinical trial to interrogate the relationship between smoking and drinking across the treatment and follow-up periods . Using a cross-lagged panel model, we examine the directional influence that drinking and smoking variables have on each other over time. We were primarily interested in testing whether drinking outcomes mediate the relationship between pharmacotherapy and smoking outcomes. Through using cigarettes per smoking day as our smoking outcome, we were able to test how reductions in drinking are associated with reductions in smoking beyond a binary quit or no quit. Our primary drinking variable of interest was drinks per drinking day which aligns with the previously mentioned trial such that drinks per drinking day was the primary drinking outcome of that study. Our secondary drinking mediators of interest are percent heavy drinking days and percent days abstinent. These were also two secondary drinking outcomes in the primary trial . Based on the literature we hypothesize that compared to participants in the varenicline plus placebo condition, those in the varenicline plus naltrexone condition will experience greater reductions in drinking, thus leading to greater reductions in smoking. While the primary outcomes of the study found varenicline alone was associated with greater smoking abstinence, we hypothesized that the combined medication condition may be sensitive to a wider range of smoking behaviors, via cigarettes per smoking day. We also hypothesized that drinking and smoking will be related across the duration of the trial, and that reductions in drinking would lead to reductions in smoking. This study provides a unique contribution by extending beyond the main effect of medication on smoking and drinking outcomes and by interrogating mechanisms of action of the combination of varenicline plus naltrexone on drinking and smoking reduction. Participants were required to produce a breath alcohol concentration of 0.00 g/dl at all study visits and to test negative for all substances excluding cannabis. Participants deemed eligible after the in-person screening visit completed a physical exam to establish medical eligibility and were then randomized to one of two medications: 2mg of varenicline tartrate plus matching placebo pills or 2mg of varenicline tartrate plus 50mg of naltrexone. All participants took the first dose of medication under observation during the randomization visit. A detailed description of the study procedures, including medication titration procedures and monitoring of side effects, is provided in Ray et al. .
During the randomization visit, participants engaged in a 30–45-minute counseling session specifically for heavy drinking smokers , set a smoking quit date, and discussed a drinking goal of abstinence or reduction. Post-randomization, participants returned to the laboratory for in-person assessment visits at Weeks 4, 8, 12, 16, and 26. A series of individual differences measures were collected at the in-person screening visit, including: a) demographics questionnaire; b) Structured Clinical Interview for DSM-5 ; c) Fagerström Test of Nicotine Dependence ; d) Clinical Institute Withdrawal for Alcohol ; e) Timeline Follow-back to assess for past alcohol consumption and cigarette use ; and f) Smoking History Questionnaire to assess for past smoking behavior. During each post-randomization visit, research assessments were completed including the TLFB, along with carbon monoxide recordings. For the present study, the primary outcome measure was cigarettes per smoking day derived from the TLFB. The primary mediator of interested was drinks per drinking day also derived from the TLFB. The primary mediator was selected given that it represents the a priori registered drinking outcome for the trial. The secondary mediators of interest were two of the secondary registered drinking outcomes, namely percent heavy drinking days and percent days abstinent also derived from the TLFB. Both drinking and smoking outcomes derived from the TLFB were measured concurrently. We report how we determined our sample size, all data exclusions , all manipulations, and all measures in the study. This study was not preregistered and the data and study materials are not available online. All analyses were conducted in SAS University Edition version 9.4 . A cross-lagged panel model with PROC CALIS was used to test the proposed mediation models of drinking outcomes mediating the effects of medication on cigarettes per smoking day. We conducted a total of 2 cross-lagged panel models across 5 time points during the active medication phase and follow-up phase. The second model tested percent heavy drinking days and percent days abstinent as secondary mediators. A conceptual diagram of the cross-lagged panel models is presented in Figure 1A and Figure 1B for drinks per drinking day, and Figure 2A and Figure 2B for percent heavy drinking days and percent days abstinent. Our predictor across all models was medication condition and our outcome was cigarettes per smoking day. For each visit, cigarettes per smoking day was averaged across the 28 days leading up the visit. In other words, cigarettes per smoking day at Week 12 consisted of cigarettes per smoking day averaged across 28 days prior to week 12 of participant enrollment in the trial. The same approach was used to define drinks per drinking day, percent heavy drinking days, and percent days abstinent. For the functional relationships between observed variables in the cross-lagged panel models, the path from medication to drinking variables and cigarettes per smoking day during the active medication phase was a free parameter. All drinking and smoking variables had fixed first-order autoregressions from Week 4 through Week 16 due to equal spacing of 4-weeks between each measurement. We controlled for baseline drinking and smoking by allowing them to freely predict the respective variable at Week 4 .