Sex differences in smoking cessation: a retrospective cohort study in a psychosocial care unit in Brazil

Abstract Introduction Despite the results of epidemiological and psychometric studies reporting comparable levels of tobacco dependence among males and females, some clinical studies have detected disparities. Some smoking cessation studies based on clinical setting programs reported poorer outcomes among women than men. Methods This retrospective cohort study aimed to compare treatment success and retention between men and women on a smoking cessation program (n = 1,014) delivered at a CAPS-AD unit in Brazil. The psychological intervention lasted 6 weeks for each group of 15 patients. Each patient had to participate in weekly group cognitive-behavioral therapy (CBT) sessions and individual medical appointments during this period. These appointments were focused on the possibility of prescribing pharmacological treatment (i.e., nicotine replacement therapy, bupropion, or nortriptyline) as adjuvants to group therapy. Results The women had lower smoking severity at baseline, more clinical symptoms, and lower prevalence of alcohol and drug use disorders and were older than the men. Females had significantly higher levels of success (36.6% vs. 29.7%) and retention (51.6% vs. 41.4%) than males. Sensitivity analysis showed that female gender was significantly associated with both retention and success, among those without drug use disorders only. Conclusion Depending on the smoking cessation setting (i.e., low and middle-income countries and mental health and addiction care units), females can achieve similar and even higher quit rates than males. Previous drug use disorder was an important confounding variable in the gender outcomes analyses. Future studies should try to replicate these positive smoking cessation effects of CBT-based group therapy plus pharmacotherapy in women.


Introduction
Worldwide, approximately 175 million women are daily smokers, with half of female smokers living in low and middle-income countries. 1Due to stigma, females may experience particular difficulties on smoking cessation programs, especially during pregnancy and motherhood. 2 A large number of factors may affect the outcome, such as caregiving demands, insufficient social backing, hormonal effects, a spouse who smokes, neurobiological variables determining dependence levels, and psychiatric comorbidities. 2,3n Brazil, according to the Vigitel/2019 survey (Vigilância de Fatores de Risco e Proteção para Doenças Crônicas por Inquérito Telefônico), 9.8% of adults are smokers, being 12.3% of males and 7.7% of females.4 Since 2002, the Brazilian Ministry of Health, in conjunction with state and municipal health departments, has organized a network within SUS (Sistema Único de Saúde) healthcare units, aiming to promote and offer smoking treatment.Health professionals administer treatment in individual consultations and support group sessions.
Patients who smoke should be helped to understand the role of cigarettes in their lives and be given guidance on how to quit smoking, how to resist the urge to smoke, and, especially, how to live without cigarettes.During the first four group meetings (or individual consultations), support manuals are provided with information on each of the sessions.Pharmacological treatment may also be provided to reduce the symptoms of nicotine withdrawal syndrome.
Despite results of epidemiological and psychometric studies reporting comparable levels of tobacco dependence among males and females, 5,6 some disparities have been detected in clinical studies. 7,8males more frequently seek treatment than males, 9 but employ anti-smoking medications less often. 102][13] Moreover, females report more concerns about weight gain related to smoking cessation. 14According to Cepeda-Benito et al., 15 nicotine replacement therapy (NRT), which is a treatment for tobacco dependence that is widely available as worldwide, seems to be less effective in females than in males.Moreover, smoking cessation studies based on programs in clinical settings reported poorer outcomes among women than men.
The literature on this topic presents mixed findings.Medical appointments were focused on the possibility of including pharmacological treatment (i.e., nicotine replacement therapy, bupropion, or nortriptyline) as an adjuvant to group therapy.These appointments also evaluated side effects and dose adjustments.
The physician could also refer the patient to another specialist or psychiatrist, if necessary, during or after the 6 weeks of treatment.Most of the investigators were psychiatry residents supervised by a qualified preceptor.

Sample
The participants were interviewed by health professionals (i.e., physicians, psychiatrists, and psychologists) at the service (who were trained by the director of the unit, who is a certified psychiatrist) at four times: t0, t1 (1 week), t2 (3 weeks), and t3 2) Patients who openly agreed to participate in the smoking cessation program.
3) Patients who completed the initial questionnaire.
Exclusion criteria: 1) Patients who did not live in São Caetano do Sul.
2) Patients younger than 18 years and older than 65 years.
3) Patients who did not complete the initial questionnaire.
4) Pregnant women.Treatment success was defined as when the patient (i) achieved abstinence of at least 28 days at the end of the treatment period and (ii) completed the 6 weeks of treatment.Both criteria were based on the self-report measures. 17In the present study, all patients who did not meet both criteria were considered treatment failures.In addition to treatment success, we also measured treatment retention as a secondary outcome of the present study.

Statistical analysis
Data were initially organized in a database and later analyzed using STATA version 11 software.The chisquare test was employed for descriptive analysis.All of the 33 variables were then analyzed separately when gender-related.We analyzed the linear correlations of variables with significant differences between genders to include them in the multivariate logistic regression model.Variables with significant correlations were excluded from the logistic regression model based on clinical relevance.Logistic regression models were estimated for successful treatment.Cox survival regression models were estimated for treatment retention.Initially, we employed multivariate survival variable models, including gender and all co-variables.
We employed logistic and survival bivariate models including gender and co-variables to investigate possible confounding factors of this co-variable.

Ethics approval
This study was approved by the Institutional Review Board at the FMABC University Center, with CAAE number 50004314.0.0000.0082.

Results
Figure (p = 0.0003) were both significantly different between men and women.
Table 2 shows the prevalence rates of types of adjuvant treatment employed: men used more pharmacological aids, nicotine patches (p = 0.086), nicotine gums (p = 0.636), bupropion (p = 0.879), and nortriptyline (p = 0.144) than women.We did not find any significant association between the type of treatment and success rates in either sex (p > 0.05).
In Table 3 In the univariate regression model, female sex was were statistically significant.However, problems with any drug were a confounding factor in this analysis (p = 0.099).
The survival analysis using multivariate Cox regression (Table 5) showed that female gender (aHR =1.24; 95%CI = 1.02-1.50;p = 0.024) was significantly associated with retention.We performed sensitivity analysis for logistic and survival regression models (Tables S1 and S2, available as online-only supplementary material), splitting the sample into those with and without other drug use disorders.The regression was performed using the variable failure as the dependent variable (with retention as the reference category).95%CI = 95% confidence interval; HR = hazard ratio.

Discussion
In the present study, women presented lower smoking severity at baseline, more clinical symptoms, and lower prevalence of alcohol and drug use disorders, and were older than men.Before adjusting for variables, women were more successful than men at completing the treatment and at quitting smoking.
This finding contradicted our initial hypothesis.In the multivariate regression, however, the association between female gender and treatment success was statistically non-significant.After adjusting for each variable, having previous issues with use of alcohol and other drugs was identified as a confounding factor in the analysis.
According to the literature, women having one or more children increases their odds of cessation compared to similarly aged women with no children. 18Moreover, the prevalence of smoking decreases substantially in the last three months of pregnancy. 19Other potentially important factors may include hormone variation and menstrual cycle, sex/gender differences in use and effectiveness of smoking cessation medication, and sex/gender differences in use of other tobacco products both prior to and subsequent to quitting smoking. 20The results of prospective observational and cross-sectional studies are mixed and demonstrate that bio-psychosocial variation in samples across place and time might determine whether or not women or men are less likely to quit smoking.
Having issues with use of other substances is a relevant indicator of vulnerability to smoke, and it has been associated with greater difficulty with quitting smoking. 21Regarding sex differences, literature shows that women with a substance use disorder diagnosis are more likely to be regular smokers when compared to men with the same diagnosis. 22cording to our experience, women may benefit more than men from a smoking cessation program even though several smoking cessation studies based on programs in clinical settings have described poorer outcomes for women when compared to men. 23,24wever, surveys in the general population have also suggested that this finding is not generalized to all women, since age could be an influencing variable, as demonstrated by Jarvis et al. 25 They used data from three major national surveys and showed that women under the age of 50 were more likely to have completely given up smoking than men, while among older age groups, men were more likely to have quit smoking than women.In our study, the mean age of both sexes was over 50 years, but women were nevertheless more successful.
Further studies are needed to understand the maintenance of gender-specific smoking patterns better and to develop more specific smoking cessation efforts for vulnerable smoker subgroups.Nonetheless, individuals with a history of substance use disorder are expected to report more difficulty in remaining abstinent after quitting smoking. 26Even after a successful treatment, former smokers with a history of other substance use should receive special follow-up attention regarding risk of relapse related to tobacco and other substances.
8][29] Despite seeking treatment more often, 30 females are more likely to receive less pharmacological treatments when compared to men. 10 Employing longitudinal data from the International Tobacco Control Four Country Surveys, conducted in Australia, Canada, the UK, and the US (ITC-4), Smith et al. 31 examined differences between genders in quitting attempts, reasons for quitting, use of smoking cessation medication, reasons for discontinuing smoking cessation medications, and rates of smoking cessation.
There were no sex differences in plans to quit, desire to quit, or quitting attempts.However, quitting success was lower among females who did not use any smoking cessation medication.
Mainly in females, there are promising findings on the use of bupropion, varenicline, contingency management, cognitive-behavioral therapy (CBT), and CBT plus pharmacotherapy (especially bupropion).
Castellani et al. 32 investigated short and longterm gender differences in smoking cessation using varenicline, finding similar treatment success rates between males and females, with females presenting more side effects due to medication than males.Waters et al. 33 reported that contingency management during the smoking cessation might produce higher self-efficacy in females when compared to males.Loreto et al. 34 evaluated a short-duration smoking cessation protocol using CBT plus pharmacotherapy, showing that females achieve more success than males, as reported in our study.Collins et al. 35 and Chatkin et al. 36 employed CBT plus bupropion and found similar success rates for both sexes.Corroborating these results, a meta-analysis investigating brief tobacco interventions in primary care found no significant gender differences. 37Finally, since weight gain is a major concern in this subgroup, any intervention addressing this issue might be promising. 38nfortunately, different authors report that a significant number of smokers receiving smoking cessation treatment relapse in the subsequent weeks and months. 39,40According to these authors' data, around half of the successful quitters in the present study should have relapsed.Generally, relapse rates range from 30 to 80% 39,40 during the following year.
Smith et al. 41

Conclusion
Our study may contribute more information to the current literature on the topic regarding those factors influencing the success rates of smoking cessation treatment programs in women.Depending on the smoking cessation setting (i.e., low and middle-income countries, mental health, and addiction care units), females could achieve similar and even higher quit rates than males.Previous drug use disorder was an important confounding variable in the gender outcomes analyses.Future studies should try to replicate these positive smoking cessation effects of CBT-based group therapy plus pharmacotherapy in women.
It was therefore necessary to conduct this study to compare success and retention between males and females participating in a smoking cessation program offered by a Psychosocial Care Center (Centro de Atenção Psicossocial -Álcool e Drogas -CAPS-AD) that treats alcohol and drug problems in the state of São Paulo, Brazil.In the present study, we hypothesized that males would have higher abstinence rates than females at the end of the treatment, as in previous studies with the general population.This retrospective cohort study aimed to compare treatment success and retention between men and women on a smoking cessation program offered at a CAPS-AD unit in Brazil.MethodsTreatment protocolOur sample was recruited from people enrolled on a Smoking Cessation Protocol delivered at a CAPS-AD treatment unit located in the city of São Caetano do Sul, in the state of São Paulo, Brazil.São Caetano do Sul has the highest Human Development Index (HDI) in Brazil, according to the Brazilian Institute of Geography and Statistics (IBGE).16The CAPS-AD units are well-known among health professionals and patients seeking addiction treatment (especially alcohol and marijuana) or needing smoking cessation treatment.As part of the protocol, the unit offers weekly group therapy, including motivational approaches for smoking cessation.Individuals interested in quitting smoking are then evaluated for further inclusion in the smoking cessation treatment.Patients may join the treatment protocol for smoking cessation in three different ways: 1) being referred by a Primary Care Unit (Unidade Básica de Saúde -UBS); 2) on-demand in a CAPS-AD; and 3) after treatment failure (patients who decide to repeat treatment).People in this category were only included in the present study at the time of their first contact with the smoking cessation treatment.Patients whose previous treatment has failed (after attending all six treatment appointments with no substance abuse cessation) are invited to join the initial motivational lecture again.The intervention lasted 6 weeks for each group of 15 patients.During this period, each patient had to participate in weekly group therapy sessions and attend individual appointments with a physician (t1 = 0, t2 = 1 week, t3 = 3 weeks, t4 = 6 weeks).The topics discussed in the group therapy sessions included risks of smoking, difficulties and benefits of quitting smoking, and relapse prevention.The groupconduction technique is based on the principles of cognitive-behavioral therapy (CBT).

( 6 1 )
weeks).During this period, weekly group sessions were held.The study sample comprised 1014 patients who completed the smoking cessation program from 2007 to 2016.Patients' data were collected using a structured questionnaire containing questions on their sociodemographic conditions and smoking habits.The following inclusion and exclusion criteria were applied: Inclusion criteria: Patients referred by a member of the smoking cessation team at the São Caetano do Sul CAPS-AD unit.

Table 1 -
Characteristics of a sample of 1,014 smokers treated at a CAPS-AD in São Caetano do Sul,SP, Brazil, 2007-2016

Table 2 -
Treatments used by a sample of 1014 smokers treated at an addiction care unit in São Caetano do Sul,SP, Brazil, 2007-2016 SE = standard error.

Table 3 -
Linear correlations among 9 variables in 1,014 smokers treated at a CAPS-AD unit in São Caetano do Sul, SP, Brazil, 2007-2016

Table 4 -
Results of Logistic Regression for treatment success of 1,014 smokers treated at a CAPS in São Caetano do Sul, SP, Brazil, 2007-2016

Table 5 -
Multivariate Cox Survival Regression for treatment retention: n = 1,014 smokers treated at a CAPS (São Caetano do Sul, SP, 42scribed 6-month abstinence rates as follows: bupropion = 16.8%;nicotinelozenge= 19.9%;patch=17.7%; patch plus lozenge = 26.9%;andbupropionplus lozenge = 29.9%.Limitations of this study may include that we did not perform a population-based study (selectionbias).Almost 15% of the women involved in the study reported any drug use disorder, which may not represent all women who seek treatment for smoking cessation in general services.Also, we did not consider the effect of the menstrual cycle on success rates, with premenstrual symptoms associated with higher relapse rates in the literature among women attempting smoking cessation.42