Potential Moderating Effects of Sex/Gender on the Acute Relative Reinforcing and Subjective Effects of Reduced Nicotine Content Cigarettes in Vulnerable Populations

Abstract

Introduction

Reports in relatively healthy smokers suggest men are more sensitive than women to the subjective effects of reduced nicotine content cigarettes (RNCCs). We know of no reports examining sex differences in the relative reinforcing effects of RNCCs, an important outcome in assessing smoking’s addiction potential. The aim of the present study is to address this gap by examining sex/gender differences on reinforcing effects while examining whether sex differences in subjective effects are discernible in vulnerable populations.

Methods

Secondary analysis of a within-subject, double-blinded experiment examining acute effects of cigarettes varying in nicotine content (0.4, 2.4, 5.2, 15.8 mg/g) among 169 adult smokers with psychiatric conditions or socioeconomic disadvantage. Effects of dose, sex, and their interaction were examined on reinforcing (concurrent-choice and Cigarette Purchase Task [CPT] testing), and subjective effects (Cigarette Evaluation Questionnaire [CEQ] and craving/withdrawal ratings).

Results

Reducing nicotine content decreased the relative reinforcing effects of smoking in concurrent-choice and CPT testing (p’s < .05) with no significant effects of sex nor dose × sex/gender interactions. Reducing nicotine content decreased CEQ ratings with only a single significant effect of sex (higher Psychological Reward scores among women than men, p = .02) and no significant dose × sex/gender interactions. Results on craving/withdrawal paralleled those on the CEQ.

Conclusions

Reducing nicotine content decreases the addiction potential of smoking independent of sex in populations highly vulnerable to smoking and addiction, with no indication that women are less sensitive to subjective effects of RNCCs or would benefit less from a policy reducing the nicotine content of cigarettes.

Implications

A policy reducing the nicotine content of cigarettes has the potential to reduce the addiction potential of smoking across men and women who are especially vulnerable to smoking, addiction, and tobacco-related adverse health impacts.

Introduction

Cigarette smoking is responsible for substantial morbidity and mortality in the United States with nearly 500,000 deaths each year attributable to smoking.¹ Despite the significant adverse health effects associated with smoking, 15% of US adults are current cigarette smokers.² Vulnerable smoker populations such as those with concurrent substance use disorders (SUDs), non-SUD psychiatric disorders (e.g., depression), and socioeconomic disadvantage bear a disproportionate burden of smoking prevalence and related disease.^3–5

Furthermore, nationally representative data demonstrate that smoking prevalence rates are currently higher among men than women (16.7% vs. 13.6%, respectively),² though when examining sex/gender differences in smoking rates over time, the gap between men and women has narrowed substantially over the past decade.¹ In addition, a recent review concluded that although women are generally less successful in sustaining long-term smoking abstinence compared with men,⁶ these findings were mixed and differ based on the type of study examined (e.g., observational vs. intervention trials).

In terms of the mechanisms contributing to potential sex/gender differences in smoking, it has been suggested that nicotine may have greater primary reinforcing effects in men than in women.⁷ Specifically, a body of seminal work on this topic has posited that cigarette nicotine dose may be less influential in sustaining smoking among women versus men, with women being less sensitive to changes in nicotine dose and potentially more sensitive to the non-nicotinic conditioned effects of smoking.^7–10

Questions regarding sex/gender differences in the role that nicotine plays in promoting and sustaining cigarette smoking is a particularly timely topic as the US Food and Drug Administration (FDA) recently gained regulatory authority over the nicotine levels in cigarettes allowing them to reduce, although not eliminate, the nicotine content of cigarettes if doing so benefits public health.^11,12 Few studies have examined the effects of sex in response to reduced nicotine content cigarettes (RNCCs). We know of four prior reports on this topic in the past few years.^13–16 Faulkner et al. examined the effects of four different RNCCs on subjective effects among young adult smokers (ages 18–25),¹³ reporting that men were more sensitive than women to nicotine dose differences in ratings of craving, perceived nicotine content, and cigarette-liking ratings. Vogel et al. compared the effects of assigning participants to a RNCC alone or an RNCC in combination with a nicotine patch on smoking rate and nicotine withdrawal levels, noting that men, but not women, showed greater reductions in both outcomes when assigned to the combination compared with RNCC alone conditions.¹⁶ In the two studies examining effects of RNCCs alone among adult smokers, Perkins and colleagues reported that relative to men, women were less sensitive to effects of nicotine content on withdrawal relief, negative affect relief and subjective responses.^14,15 In sum, prior studies consistently indicate that men are more sensitive than women to changes in the nicotine content of cigarettes on smoking rate, craving reduction, withdrawal reduction and positive subjective responses to cigarettes. However, to our knowledge, these prior studies did not assess relative reinforcing effects and were conducted in relatively healthy populations of smokers. Thus, the aims of the current study are to (1) assess whether prior observations of sex/gender differences in sensitivity to the effects of altering the nicotine content of cigarettes have generality to measures of relative reinforcement and (2) determine whether prior reports of sex/gender differences in responsivity to nicotine extend to populations known to be especially vulnerable to cigarette smoking and addiction.

We recently completed a double-blinded, randomized controlled laboratory study examining the acute relative reinforcing and subjective effects of research cigarettes varying in nicotine content following overnight smoking abstinence in three vulnerable populations.¹⁷ The present study is a secondary analysis of that prior experimental study aiming to address the primary aims mentioned above. We hypothesized that consistent with the prior literature, men would be more sensitive than women to the relative reinforcing effects of nicotine dose.

Methods

Sample

Participants were 169 (120 Females, 49 Males) adult daily smokers enrolled in a multi-site, double-blinded, within-subject, clinical-laboratory study. Details of study methods have been reported previously.¹⁷ Briefly, participants were recruited via advertisements on Facebook, community bulletin boards, buses, and local newspapers from March 23, 2015 to April 25, 2016. Participants were recruited as exemplars of three different populations known to be especially vulnerable to cigarette smoking and addiction: Individuals with affective disorders, as an exemplar of those with mental illness (n = 56), individuals with opioid use disorder (OUD) as an exemplar of those with other SUDs (n = 60), and women of reproductive age (18–44 years old) with limited educational attainment, as an exemplar of those with socioeconomic disadvantage (n = 53).

Research Cigarettes

Cigarettes for this study were Spectrum research cigarettes obtained from the National Institute on Drug Abuse and manufactured by 22nd Century Group (Clarence, NY). Four nicotine doses were examined (0.4, 2.4, 5.2, and 15.8 mg of nicotine per gram of tobacco; milligrams/gram). Cigarettes were available in menthol or non-menthol flavor based on participant’s usual brand cigarette preference. The 15.8 mg/g cigarette is similar in nicotine content to commercially available cigarettes, thus functioning as a control condition. All experimental sessions were conducted under double-blind conditions with cigarettes referred to by arbitrary letter codes that varied randomly across participants.

Procedure

Study eligibility was determined during an intake assessment which consisted of various demographic, smoking, and psychiatric screening measures (i.e., Beck Depression Inventory [BDI] and Overall Anxiety Severity and Impairment Scale [OASIS]).^18,19 Participants completed a three-phase study spanning 14 experimental sessions. They were asked to abstain from smoking for 6–8 hours (breath carbon monoxide levels ≤ 50% of baseline levels) prior to all experimental sessions. Session 1 functioned as a baseline orientation session where participants smoked their own brand cigarette. Phase 1 (Sessions 2–5) focused on assessing subjective responses to a single cigarette smoked in the laboratory. Phase 2 (Sessions 6–11) assessed choices between cigarette puffs differing in nicotine content, when all were available for an equal response requirement. Phase 3 (Sessions 12–14) assessed choices between the highest and lowest doses, when the response requirement for the low dose was held constant and the response requirement for the high dose increased progressively. This report focuses on data from Phases 1 and 2 only.

In Phase 1 (Sessions 2–5), participants sampled each of the research cigarettes (one dose per session) in a random order. Participants smoked cigarettes ad libitum during these sessions. Before and every 15 minutes for an hour after smoking each cigarette, participants completed the Minnesota Nicotine Withdrawal Scale (MNWS)²⁰ and the Questionnaire on Smoking Urges-Brief version (QSU)²¹ to assess tobacco withdrawal and cigarette craving. In addition, after smoking each research cigarette, participants rated the subjective effects of smoking using the modified Cigarette Evaluation Questionnaire (mCEQ)²² and completed the Cigarette Purchase Task (CPT).^23,24 On the mCEQ, the following subscales are derived: Satisfaction, Psychological Reward, Enjoyment of Respiratory Tract Sensations, Aversion, and Craving Reduction. The CPT is a behavioral economic simulation task wherein participants estimate the number of cigarettes they would purchase to smoke within a 24-hour period across a range of cigarette prices. In the CPT, five indices are modeled: Intensity: daily cigarette smoking rate when cigarettes are available at no cost; O_max: maximum total expenditure on smoking in a 24-hour period; P_max: price at which maximum cigarette expenditure occurs and smoking rate begins decreasing; Breakpoint: price at which one would quit smoking rather than incur the cost; Elasticity: overall sensitivity of demand for increasing price.

Phase 2 (Sessions 6–11) directly tested the relative reinforcing effects of the different doses in the cigarettes by allowing participants to choose which cigarette they preferred to smoke. Each of the six possible cigarette dose pair combinations was tested once in separate sessions where participants were allowed to choose which dose cigarette they preferred to smoke during 3-hour concurrent-choice sessions. In all sessions, the two cigarettes being tested were each available at an equal response cost of 10 computer mouse clicks (Fixed Ratio-10), which earned the participant two controlled puffs (60 ml volume each) of the chosen cigarette. Participants were allowed to make as many or few choices to smoke during the 3-hour sessions as they wished.

Sex/gender

We used a binary variable for sex/gender where participants were asked to report their sex and were able to respond as “male” or “female.” This was assessed using a demographic questionnaire completed at the study intake session.

Statistical Methods

Sex/gender differences in demographic and smoking characteristics were compared using chi-square tests for categorical variables and t-tests for continuous variables. Significant differences in baseline characteristics by sex were included as covariates in subsequent statistical models. We combined the three vulnerable populations for the purposes of this secondary analysis.

Analyses of all Phase 1 (i.e., MNWS Total and Desire-to-Smoke scores, QSU Factors 1 and 2, mCEQ subscales, CPT indices) results examined differences in the nicotine dose of the research cigarettes, and whether these differences were affected by the sex/gender of the respondent. These were conducted using separate mixed model analyses of covariance (ANCOVA) to examine the effects of dose, time when relevant (i.e., QSU and MNWS scores), and sex/gender along with the interactions of dose and dose × time with sex/gender. In the absence of statistically significant interactions with sex/gender, analyses were repeated without this effect to assess the main effects of dose and sex/gender. All analyses were completed using maximum likelihood estimation, without imputation for missing values. Differences in the proportion of choices for the higher dose cigarette in Phase 2 concurrent-choice testing arrangements were examined in a similar manner. All models included fixed effects of age, cigarettes smoked/day, menthol smoker status, baseline BDI and OASIS total scores, session, and subpopulation as covariates. Because the research cigarettes were presented in random order using a Latin Square, sequence was included in the model as a random effect, as was the site at which the experiments were conducted. With respect to the CPT analyses, maximum expenditure, maximum price, breakpoint, and elasticity were log10 transformed to correct for skewness. Mean and standard error estimates were back-transformed into the original units for presentation. We reviewed CPT results and found systematic patterns²⁵ in 92.7% of demand curves; no data were excluded from analyses. In cases where participants reported zero consumption across all prices, curve fitting was not possible, so elasticity was not analyzed, and other demand indices were quantified as 0. In addition, we generated partial η ² to estimate effect sizes in our ability to detect sex/gender differences. Cohen suggests η ² values of 0.0099, 0.0588, and 0.1379 as small, medium, and large effects, respectively.

Finally, sensitivity analyses using mixed model ANCOVAs adjusting for age, menthol status, OASIS total score, session, dose and subpopulation (as these differed significantly by sex/gender) examined the moderating effects of sex/gender on all of the above outcomes with analyses restricted to the samples with OUD and affective disorders (i.e., excluding the group of socioeconomically disadvantaged women) to determine whether results extended to the samples which included both women and men.

All analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC).

Results

Participant Characteristics

Baseline participant characteristics are presented by sex/gender in Table 1. On average, men were more likely to be older, smoked more cigarettes per day, were more likely to be menthol smokers, and had higher levels of depression and anxiety than women (p’s < .05). There were also differences by sex/gender with regard to the vulnerable population for which participants were initially recruited (i.e., affective disorders, OUD, socioeconomically disadvantaged smokers), as only women of reproductive age were recruited into the socioeconomically disadvantaged group.

Table 1.

Participant characteristics by sex/gender

	Women (n = 120)	Men (n = 49)	P value
Subpopulation, n (%)			<.001
Economically disadvantaged women	53 (44)	0 (0)
Opioid use disorder	36 (30)	24 (49)
Affective disorders	31 (26)	25 (51)
Demographics
Age	33 ± 10	41 ± 14	<.001
Education, n (%)			.487
Some high school	21 (18)	6 (12)
High school graduate or equivalent	38 (32)	20 (41)
Some college	55 (46)	19 (39)
College graduate and above	6 (5)	4 (8)
Race, n (%)			.299
White	90 (75)	37 (76)
African-American	17 (14)	6 (12)
Other	13 (11)	6 (12)
Married, n (%)	23 (19)	4 (8)	.077
Smoking characteristics
Cigarettes smoked/day	15 ± 7	18 ± 8	.021
FTND score	5 ± 2	5 ± 2	.141
Age at first cigarette	16 ± 3	16 ± 6	.770
Menthol smoker, n (%)	49 (41)	12 (25)	.045
Psychiatric measures
BDI total score	11 ± 11	17 ± 12	.001
OASIS total score	5 ± 5	8 ± 5	<.001

Data points represent n (%) for subpopulation, education, race, and marital status and means ± SD for other characteristics. Bold values represent p < .05. FTND, Fagerstrom Test of Nicotine Dependence²⁶; BDI, Beck Depression Inventory¹⁸; OASIS, Overall Anxiety Severity and Impairment Scale.¹⁹

Concurrent-Choice Testing

Participants showed a significant preference for the higher over the lower nicotine doses across each of the six dose pairs examined in concurrent-choice testing (F(5, 831) = 6.57, p < .001, η ² = 0.032) with no significant main effects of sex/gender (F(1, 145) = 0.13, p = .72; η ² = 0.0001), nor dose × sex/gender interactions (F(5, 828) = 1.27, p = .27; η ² = 0.0063) noted (Figure 1).

Figure 1.

Bar graph showing results of concurrent-choice testing among women (gray bars) and men (black bars). Shown within each bar are the proportion of choices allocated toward the higher dose cigarette in comparisons across the four nicotine dose cigarettes during concurrent-choice testing sessions. Means represent least square means, and error bars represent SEM.

Sensitivity analyses examining the moderating effects of sex/gender only among the subpopulations with OUD and affective disorders produced largely consistent results. Specifically, there were no significant main effects of sex/gender (F(1, 109) = 0.00, p = .99), nor dose × sex/gender interaction (F(5, 562) = 0.92, p = .46) on concurrent-choice testing.

Cigarette Purchase Task

There was a main effect of nicotine dose across all five CPT indices (Intensity: F(3,498) = 5.51, p = .001, η ² = 0.023; O_max: F(3,498 = 13.14, p < .0001, η ² = 0.052; P_max: F(3, 498) = 12.56, p < .0001, η ² = 0.050; Breakpoint: F(3, 498) = 14.96, p < .0001, η ² = 0.058; Elasticity (F(3, 455) = 7.43, p < .01, η ² = 0.011). There were no significant main effects of sex/gender (Intensity: F(1,160) = 0.71, p = .40; O_max: F(1,160 = 0.52, p = .47; P_max: F(1, 159) = 0.53, p = .47; Breakpoint: F(1, 159) = 0.15, p = .70; Elasticity: F(1, 148) = 1.99, p = .16); all η ²’s < 0.01, nor interactions of dose × sex/gender (Intensity: F(3,495) = 1.25, p = .29; O_max: F(3,495 = 0.74, p = .53; P_max: F(3, 495) = 0.74, p = .53; Breakpoint: F(3, 495) = 1.03, p = .38; Elasticity: F(3, 451) = 0.58, p = .63; All η ²’s < 0.01) noted on any of the five CPT indices (Figure 2).

Figure 2.

Results from the Cigarette Purchase Task (CPT) simulating estimated demand for each of the different nicotine content cigarettes at escalating prices by sex. Demand intensity indicates estimated consumption at $0 price (higher scores indicating greater consumption when cigarettes are free); maximal expenditure, estimated maximal expenditure participants were willing to incur for smoking in 1 day (higher scores indicating greater expenditure); maximal price, estimated price at which demand begins to decrease proportional to price increases (higher scores signifying a greater cigarette unit price associated with unit elasticity for cigarettes); breakpoint, estimated price at which participants would quit smoking rather than incur its costs (higher scores indicating a greater cigarette unit price associated with discontinuation of smoking); and elasticity, estimated overall sensitivity of demand to price increases (higher scores indicating greater sensitivity to cigarette unit price increases). Data points represent least square means across participants and error bars represent standard error of the mean. Estimates for sensitivity to price, maximum expenditure, maximum price, and breakpoint relied on a logarithmic transformation, which have been back-transformed for presentation here.

In sensitivity analyses, similarly, there were no significant main effects of sex/gender, nor dose × sex/gender interactions on any indices of the CPT (Main Effects: Intensity: F(1,108) = 0.24, p = .62; O_max: F(1,109) = 0.61, p = .44; P_max: F(1, 109) = 0.68, p = .41; Breakpoint: F(1, 108) = 0.14, p = .71; Elasticity (F(1, 100) = 0.90, p = .34; Dose × sex/gender: Intensity: F(3, 336) = 1.45, p = .23; O_max: F(3, 336) = 0.48, p = .70; P_max: F(3, 336) = 0.32, p = .81; Breakpoint: F(3, 336) = 0.62, p = .60; Elasticity (F(3, 309) = 0.28, p = .84).

Modified Cigarette Evaluation Questionnaire

Reducing nicotine content decreased participant ratings on mCEQ in a dose-dependent manner (Satisfaction: F(3,498) = 40.41, p < .0001, η ² = 0.153; Psychological Reward: F(3,498) = 19.92, p<.0001, η ² = 0.082; Enjoyment of Respiratory Tract Sensations: F(3,498) = 26.92, p < .0001, η ² = 0.107; Aversion: F(3,498) = 7.08, p = .0001, η ² = 0.031; Craving Reduction: F(3,498) = 20.89, p < .0001, η ² = 0.085), with only a single significant main effect of sex/gender on the Psychological Reward subscale (F(1, 159) = 5.85, p = .02, η ² = 0.0086), and no significant dose × sex/gender interactions across any of the subscales (Satisfaction: F(3,495) = 2.33, p = .07; Psychological Reward: F(3,495) = 1.23, p = .30; Enjoyment of Respiratory Tract Sensations: F(3,495) = 2.32, p = .07; Aversion: F(3,495) = 0.27, p = .84; Craving Reduction: F(3,495) = 0.60, p = .62; All η ²’s < 0.01) (Supplementary Table 1). The effect of sex on the Psychological Reward subscale corresponds to women reporting higher levels of psychological reward than men across all cigarette doses (Figure 3).

Figure 3.

Bar graph showing results of the Modified Cigarette Evaluation Questionnaire (mCEQ) Psychological Reward subscale (left panel) and Questionnaire of Smoking Urges (QSU) Factor 2 scores (right panel) across research cigarettes by sex/gender. A significant main effect of sex/gender was observed for the Psychological Reward subscale of the mCEQ (F(1, 159) = 5.85, p = .02), and QSU Factor 2 (F(1, 159) = 5.26, p = .02). No interactions between sex/gender and dose or time were observed on these measures or the other mCEQ and QSU subscales. QSU Factor 2 scores are presented as change scores by subtracting the time point immediately following (i.e., 15 minutes after) smoking each dose cigarette from the presmoking baseline time point (prior to smoking research cigarettes, under conditions of acute abstinence). Means represent least square means, and error bars represent SEM.

In the sensitivity analyses, consistent with the findings above, there were no significant main effects of sex/gender, nor dose × sex/gender interactions observed on most of the mCEQ subscales (Main Effects: Satisfaction: F(1, 109) = 0.98, p = .32; Enjoyment of Respiratory Tract Sensations: F(1, 109) = 1.03, p = .31; Aversion: F(1, 109) = 0.05, p = .82; Craving Reduction: F(1, 109) = 0.09, p = .77; Dose × sex/gender: Satisfaction: F(3, 336) = 1.71, p = .16; Psychological Reward: F(3, 336) = 1.69, p = .17; Enjoyment of Respiratory Tract Sensations: F(3, 336) = 2.10, p = .10; Aversion: F(3, 336) = 0.28, p = .84; Craving Reduction: F(3, 336) = 0.57, p = .63). As reported in our primary analyses, there was a significant main effect of sex/gender on the Psychological Reward subscale of the mCEQ (F(1, 109) = 5.60, p = .02), but no dose × sex/gender interaction (F(3, 336) = 1.69, p = .17).

Questionnaire of Smoking Urges-Brief Version

There was a significant dose × time interaction on the QSU Factor 1 subscale (F(12, 2002) = 9.16, p < .0001; η ² = 0.032), such that higher doses reduced craving for a longer duration of time, but no significant main effect of sex/gender (F(1,159) = 2.09, p = .10; η ² = 0.0006), nor sex/gender × dose × time interaction (F(12, 1990) = 0.57, p = .87; η ² = 0.0020) (Supplementary Table 2).

For QSU Factor 2, there was a significant dose × time interaction (F(12, 2002) = 5.30, p < .0001; η ² = 0.019), and a main effect of sex/gender (F(1, 159) = 5.26, p = .02; η ² = 0.0016) (Figure 3), but the sex/gender × dose × time interaction on QSU Factor 2 was not significant (F(12, 1990) = 0.67, p = .79; η ² = 0.0024). The main effect of sex/gender corresponds to women reporting higher QSU Factor 2 scores than men.

In sensitivity analyses, consistent with our primary analyses, there were no significant main effects of sex/gender (F(1,109) = 1.84, p = .18), nor dose × time × sex/gender interaction (F(12, 1355) = 0.85, p = .60) observed on QSU Factor 1. There was a significant main effect of sex/gender on the QSU Factor 2 (F(1, 109) = 4.72, p = .03), but no dose × time × sex/gender interaction on QSU Factor 2 (F(12, 1355) = 0.88, p = .56).

Minnesota Nicotine Withdrawal Scale

There were significant dose × time interactions on the MNWS Desire-to-Smoke and Total Score ratings (Desire-to-Smoke: F(12, 2002) = 5.76, p < .0001; η ² = 0.020); Total Score: F(12, 2002) = 2.77, p < .001; η ² = 0.010)), but there were no significant main effects of sex/gender on either MNWS Desire-to-Smoke ratings (F(1, 146) = 1.23, p = .27; η ² = 0.0004) or Total Score ratings (F(1, 159) = 2.75, p = .10; η ² = 0.0008). The dose × time × sex/gender interactions on these measures were not significant (Desire-to-Smoke: F(12, 1990) = 0.60, p = .84; η ² = 0.0021; Total Score: F(12, 1990) = 0.83, p = .62; η ² = 0.0029) (Supplemental Tables 3 and 4).

In sensitivity analyses, there were no significant main effects of sex/gender (F(1,109) = 5.51, p = .12) nor dose × time × sex/gender interaction on MNWS Total Score (F(12, 1355) = 1.12, p = .34), and no significant main effect of sex/gender (F(12, 1355) = 0.80, p = .65) nor dose × time × sex/gender interaction (F(1,109) = 0.65, p = .42) on MNWS Desire-to-Smoke.

Discussion

Results from the present study do not indicate that sex/gender moderated the effects of reducing nicotine content on the relative reinforcing effects of smoking in the concurrent-choice or CPT-testing arrangements. Both women and men showed a clear preference for higher-nicotine content cigarettes in the concurrent-choice arrangement and CPT, consistent with the higher doses having greater relative reinforcing value or addiction potential.

Regarding our investigation of whether the observation that women are less sensitive than men to the subjective effects of nicotine dose differences noted in healthier populations^10–12 extend to these vulnerable populations, we saw no evidence in that direction. That is, there were no instances of significant interactions of nicotine dose and sex/gender in any of measures of subjective effects assessed. In the two instances where main effects of sex/gender were observed, they were in the direction of women reporting greater effects than men on the Psychological Reward subscale of the mCEQ and anticipation of greater negative reinforcement from smoking on the QSU Factor 2 scale. Those effects extend research on the potential of sex/gender differences in the subjective effects of smoking to vulnerable populations, but with no suggestion of women being less sensitive than men to reductions in the nicotine content of cigarettes as has been suggested in research in relatively healthier populations of smokers.^10–12

What may account for these between-study differences is unclear. One obvious difference between the present and prior studies is that participants in the present study were selected because they shared characteristics associated with greater vulnerability to smoking and addiction. Perhaps women with psychiatric conditions or socioeconomic disadvantage are more sensitive to the subjective effects of nicotine than women sampled from the general population. Certainly, it is well established that opioid use disorder, affective disorders, and socioeconomic disadvantage confer heightened vulnerability to smoking and nicotine dependence.^3,5,27 This is also a secondary analysis of a study that was not designed to examine sex/gender differences and thus did not attempt to recruit a study population with equal numbers of women and men or to keep other sociodemographic characteristics consistent across women and men. As women comprised 71% of the participants in this study, a larger percentage of men in the sample would have been more efficient from a statistical standpoint; nevertheless, the sample size of 169 individuals had 88% power to detect what Cohen²⁸ considers small effects sizes. We controlled for baseline sociodemographic characteristics that differed by sex in the statistical analyses but that does not rule out the presence of other potential confounders. Those study limitations notwithstanding, it is important to underscore what we believe is most important about the present results which, in our opinion, is not the between-study differences. Instead, what we deem most important is the positive evidence that the study provides indicating that women from populations especially vulnerable to persistent smoking and addiction are highly sensitive to reductions in the nicotine content of cigarettes. Reducing the nicotine content of cigarettes clearly decreased behavioral and subjective measures indicative of addiction potential in these vulnerable women without causing untoward levels of craving or withdrawal. These results suggest that if the FDA were to move forward with a policy of reducing the nicotine content in cigarettes to minimally addictive levels, it has the potential to benefit women and men alike from these vulnerable populations who disproportionately shoulder the burden of the adverse effects of cigarette smoking.

Funding

This project was supported by Tobacco Centers of Regulatory Science awards P50DA036114 /U54DA036114 from the National Institute on Drug Abuse (NIDA) and Food and Drug Administration (FDA), U54CA180908 from the National Cancer Institute (NCI) and FDA, Center of Biomedical Research Excellence award P20GM103644 from the National Institute of General Medical Sciences (NIGMS), and NIDA Institutional Training Grant T32DA007242. Support for JWT was also provided by U54DA031659. The content is solely the responsibility of the authors and does not necessarily represent the official views of NIDA, NCI, FDA, or NIGMS.

Declaration of Interests

JRH has received consulting and speaking fees from several companies that develop or market pharmacological and behavioral treatments for smoking cessation or harm reduction and from several nonprofit organizations that promote tobacco control. He also consults (without payment) for Swedish Match. All other authors have nothing to declare.