Background

The tobacco epidemic is one of the biggest public health threats today, killing close to six million people each year [1]. More than five million of these deaths are the result of direct tobacco use while more than 600,000 are the result of non-smokers exposed to second-hand smoke [1]. Unless urgent action is taken, the annual death toll could rise to more than eight million by 2030 [1]. In 2010, the Center for Disease Control and Prevention (CDC) reported that 68.8% of current smokers in the United State wanted to completely stop smoking but only 6.2% of smokers had successfully done so in the past 12 months [2]. Craving and withdrawal symptoms have been associated with smoking relapse [3]. A higher level of craving and withdrawal symptoms upon initiating abstinence has been associated with earlier relapse [4].

According to the clinical practice guidelines recommended by the US Public Health Service for treating tobacco use and dependence, a combination of counseling and medication is considered effective treatment [5]. However, previous studies have shown that long term abstinence rates for the combination of counseling and nicotine replacement therapy (NRT), varenicline and buproprion range from 6.5% –34.4%, 14.4%–34.6% and 6.3%–31.8%, respectively depending on dose/form/duration of medication and follow-up period [6,7,8,9,10,11,12,13,14,15,16,17,18,19]. Therefore, the effectiveness of the combination of counseling and medication for smoking cessation remains low. However, individually, the effectiveness of both counselling and medication for smoking cessation might also remain low.

Exercise is one choice of additional treatment for smoking cessation by relieving nicotine withdrawal symptoms and smoking craving and is a low cost treatment that is easy to access. Moreover, it can promote the health of the smoker. Acute bouts of exercise have been found to have a positive effect in the reduction of nicotine withdrawal symptoms and smoking craving [20, 21], which are important factors leading to smoking relapse [3, 22]. Therefore, exercise is an interesting treatment for smoking cessation. The possible mechanism of the effect of exercise on relieving nicotine withdrawal symptoms and smoking craving is including affect, biological, and cognitive hypotheses [21]. Evidence suggests that different types of exercise have different effects on these mechanisms. For the biological hypothesis, Goldfarb and Jamurtas [23] suggested that exercise-induced β-endorphins alterations are related to the type of exercise. Several studies revealed aerobic exercise on sufficient intensity increases β-endorphins in plasma [24,25,26,27], whereas there was controversy about the effect of resistance exercise on β-endorphins. Decrease and no change of β-endorphins after resistance exercise had been reported [28,29,30]. However, there were some studies reporting an increase of β-endorphins after resistance exercise [31]. For the affect hypothesis, a meta-analysis by Arent et al. showed that resistance training produced more improved mood in older adults in general than cardiovascular exercise [32]. Therefore, type of exercise might have effect on smoking cessation in a different way through these mechanisms.

To date, there has been no meta-analysis for the effect of exercise and exercise type on smoking cessation at the end of treatment and at the end of follow-up. However, there has been one systematic review on the effects of exercise on smoking cessation in general at follow-up [33]. Therefore, this study is the first meta-analysis to gain insight into which type of exercise is effective for smoking cessation at the end of treatment and at the end of follow-up.

Methods

Search strategy

Online searches were performed on PubMed, ScienceDirect, PEDro, Web of Science, Scopus and Cochrane Library databases up to November 2016. The following keywords were used: “physical activity”, “exercise”, “smoking”, “tobacco”, “cigarette”, “cessation”, “treatment” and “intervention” (Additional file 1). After inclusion of the articles based on the selection criteria, references were searched for additional articles.

Study selection

The search of electronic databases identified 8994 articles. TK selected relevant articles from those retrieved through the search strategy. The selection criteria were as follows:

  1. 1.

    The study design was a randomized controlled trial (RCT) that used exercise alone or as an adjunct program to smoking cessation intervention compared with smoking cessation intervention.

  2. 2.

    The article was a full report published in English. Letters, abstracts, books, conference proceedings, and posters were excluded.

  3. 3.

    The study samples were smokers who wished to quit or who were recent quitters.

  4. 4.

    Studies in populations with psychological problems or pregnant women were excluded.

  5. 5.

    The follow-up period was continued for at least 6 months after randomization.

Data extraction

Data extraction was performed independently by two authors (TK and ES). For each article, the characteristics of the participants, intervention parameters, outcomes, and results were extracted using a standardized form. Another author (PT) was consulted if disagreement between the two authors (TK and ES) persisted.

The following outcomes were examined at the end of the treatment and at the end of the follow-up: (a) point prevalence abstinence rate, and (b) continuous abstinence rate. The point prevalence abstinence was defined as not smoking for a few days before the follow-up, e.g. 7 days. Continuous abstinence was defined as not smoking throughout the follow-up period after the quitting date.

Risk of bias in individual studies

The methodological quality of the articles that met the selection criteria was evaluated independently by two authors (TK and ES). It was evaluated using Cochrane Collaboration’s tool which contained five fundamental bias domains: selecting bias, reporting bias, performance bias, detecting bias and attrition bias [34]. These five domains consisted of seven criteria: random sequence generation, allocation concealment, selective reporting, blinding participants and personnel, blinding of outcome assessment, incomplete outcome data and other bias. Each criteria had three rating categories: “low risk”, “high risk” and “unclear risk”. Studies were defined as “high risk” when at least three criteria were met as unclear risk and/or high risk. In contrast, studies were defined as “low risk” when less than or equal to two bias criteria were met as unclear risk and/or high risk. The rating for each bias criteria of the two authors was then compared. Disagreements between the two authors on individual bias criteria were identified and discussed in an attempt to reach a consensus. If agreement could not be reached, another author (PT) was consulted to reach a final judgment. Percentage agreement and Cohen’s kappa were calculated both before and after the consensus discussion.

Data analysis

Data were analyzed using Review Manager (RevMan5.3). All treatment effects were reported with a 95% confidence interval (CI). For dichotomous outcomes, the treatment effect was reported as relative risk (RR). An RR was estimated by using the following data: the number of participants who quit smoking and the total number of participants in each group. An RR of more than one indicated that exercise resulted in a greater chance of quitting smoking. Dichotomous outcomes were weighted using the Mantel–Haenszel method [35]. A random-effect model was conducted. Statistical heterogeneity was determined using I2 statistic. Funnel plots of the trial’s RR were evaluated for publication bias. Forest plots were generated to present the pooled estimates where there were two or more RCTs of sufficient clinical and statistical data. The effectiveness of exercise was reported in qualitative analysis, if the data were not quantified for meta-analysis (i.e. having only a single study or not providing any outcome data in a form that could be used).

The GRADE (Grades of Recommendation Assessment, Development and Evaluation) approach was used to assess the overall quality of the evidence for each outcome. The GRADE approach classified the quality of the evidence into four levels: high, moderate, low, and very low. A randomized controlled trial started as high-quality evidence and the quality of evidence was downgraded according to five domains. The five domains comprised:

  1. 1.

    Limitation of the study design (downgraded when more than 25% of the participants were from studies with a high risk of bias),

  2. 2.

    Inconsistency (downgraded when statistical significant heterogeneity was present),

  3. 3.

    Indirectness (downgraded when participants, intervention, outcomes or comparison of the study did not match with the objectives of this review),

  4. 4.

    Imprecision (downgraded when the number of events for each outcome was less than 300),

  5. 5.

    Publication bias (downgraded when an asymmetry of funnel plot was present).

Single studies (number of events less than 300) were considered inconsistent and imprecise and provided “low-quality evidence”, which could be further downgraded to “very low-quality evidence” if there existed limitations in the study design or indirectness.

The definitions of the quality of evidence were as follows [36]:

  • High quality Further research is very unlikely to change our confidence in the estimate of effect. All five domains are also met.

  • Moderate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. One of the five domains is not met.

  • Low quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Two of the five domains are not met.

  • Very low quality Any estimate of effect is very uncertain. Three of the five domains are not met.

Results

Search strategy

A total of 20 articles on 19 studies were judged to meet the selection criteria. However, two articles by Ussher et al. [37] and Ussher et al. [38] were identified as double publications with different follow-up periods. Consequently, these two articles were rated as one trial in this review. In total, 19 studies were included in the methodology quality assessment and data extraction (Fig. 1). All of the included studies came from peer-reviewed journals and one of these came from a fully open access journal. The number of pulled studies was less than a previous systematic review by Ussher et al. which 20 studies were included [33]. One article by Horn et al. was excluded from our review because there were some participants in the pre-contemplation stage (participants do not thinking about quitting) [39].

Fig. 1
figure 1

Flow diagram of data screening process

Full size image

Study characteristics

The characteristics of all studies are presented in Table 1. Of the 19 studies, 14 examined the effectiveness of an aerobic exercise program on smoking cessation [40,41,42,43,44,45,46,47,48,49,50,51,52,53]. One study each examined the effectiveness of a resistance training program [54], yoga [55], and a combined aerobic and resisted exercise program [56]. Two studies did not specify the precise type of exercise [37, 38, 57]. Thus, these two studies were classified as examining the effectiveness of physical activity on smoking cessation. Physical activity is defined as ‘any bodily movement produced by skeletal muscles that result in energy expenditure’ [58].

Table 1 Characteristics and results of included studies
Full size table

Of the included studies, eight provided supervised, group-based exercise at the research setting plus home-based exercise [40,41,42,43,44, 48, 49, 51]. Seven studies provided only supervised, group-based exercise at the research setting [45,46,47, 50, 54,55,56]. Three studies provided only home-based exercise [37, 38, 53, 57]. The remaining one, Taylor et al. provided home-based exercise or supervised, group-based exercise in each group [52].

Of the studies providing home-based exercise, the majority reported poor exercise adherence [37, 41, 44, 48, 57]. Kinnunen et al. reported that less than 50% of participants’ exercise met the prescription in the first 5 weeks and this dropped to 6.5% at the end of the treatment [44]. Marcus et al. reported that only an average of 15.2% of participants’ exercise met the prescribed requirements [48]. Bize et al. reported that 50% of participants of the physical activity group were classified as sedentary at the end of treatment [41]. McKay et al. reported that 38.0% of participants engaged in physical activity at a level of vigorous intensity and 79.4% participants engaged at a level of moderate intensity [57]. Ussher et al. reported that participants engaged in 30 min of moderate or vigorous exercise only 2.4–2.6 days per week even though exercise was prescribed 5 days per week [37].

Seven of the 19 studies assessed smoking status by point prevalence abstinence [42, 43, 45, 46, 49, 55, 57], three studies by continuous abstinence [37, 38, 41, 44], seven by both point prevalence abstinence and continuous abstinence [40, 47, 48, 50, 53, 54, 56] and two studies did not clearly state what outcomes they assessed [51, 52].

Risk of bias in individual studies

The rating of the two authors of the included studies before discussion had an agreement rate of 90.55% (121/133). The overall inter-rater agreement was k = 0.860 with an SE of measurement of 0.039. After discussion, the two authors had an agreement rate of 100% (133/133). Then, the overall inter-rater agreement was k = 1.00 with an SE of measurement of 0.00. This represents very good agreement between the two authors [59]. Disagreements were often related to reading errors or interpretation of the criteria list.

The results of the methodological quality assessment are presented in Fig. 2. Seven studies were rated as having a low risk of bias [37, 38, 40, 41, 53,54,55,56]. Twelve studies were rated as having a high risk of bias [42,43,44,45,46,47,48,49,50,51,52, 57]. The allocation concealment (selection bias) was rated as having unclear risk in all except two studies [37, 38, 40, 42,43,44,45,46,47,48,49,50,51,52, 54,55,56,57]. In addition, the blinding outcome assessment (detection bias) was rated as having a low risk in only one study [55].

Fig. 2
figure 2

Summary of risk of bias assessments for each study

Full size image

Analysis

The number of studies in each outcome regarding the type of exercise presents in Table 2. Tables 3, 4, 5, 6, 7, 8 present the analyzed results based on the GRADE approach. Two studies were excluded from the analysis because these studies did not clearly state what outcomes they assessed [51, 52].

Table 2 The number of studies in each outcome regarding the type of exercise
Full size table
Table 3 Summary of finding of evidence of the effectiveness of exercise on smoking cessation
Full size table
Table 4 Summary of finding of evidence of the effectiveness of aerobic exercise on smoking cessation
Full size table
Table 5 Summary of finding of evidence of the effectiveness of resistance exercise on smoking cessation
Full size table
Table 6 Summary of finding of evidence of the effectiveness of yoga on smoking cessation
Full size table
Table 7 Summary of finding of evidence of the effectiveness of a combined aerobic and resisted exercise program on smoking cessation
Full size table
Table 8 Summary of finding of evidence of the effectiveness of physical activity on smoking cessation
Full size table

Evidence of the effectiveness of exercise program on the point prevalence abstinence rate at the end of the treatment

Thirteen studies investigated the effectiveness of exercise on the point prevalence abstinence rate at the end of the treatment [40, 42, 43, 45,46,47,48, 50, 53,54,55,56,57]. The results indicated moderate quality evidence (13 RCTs; N = 4371; limitations in study design) that there was no significant difference between exercise and control condition on the point prevalence abstinence at the end of the treatment (RR 1.13, 95% CI 0.94–1.35) (Fig. 3).

Fig. 3
figure 3

The effectiveness of exercise on smoking cessation

Full size image

Regarding the effect of the type of exercise (Table 2), low quality evidence (1 RCT; N = 55; inconsistency, imprecision) was found for the positive effect of yoga on the point prevalence abstinence at the end of the treatment (RR 3.11, 95% CI 1.00–9.69).

Moderate quality evidence (9 RCTs; N = 1643; limitations in study design) was found for there being no effect of the aerobic exercise program on the point prevalence abstinence at the end of the treatment (RR 1.13 95% CI 0.89–1.44) (Fig. 4). The evidence for there being no effect the resistance training program (1 RCT; N = 25; inconsistency, imprecision), a combined aerobic and resisted exercise program (1 RCT; N = 330; inconsistency, imprecision) on the point prevalence abstinence at the end of the treatment was of low quality (RR 2.77 95% CI 0.69–11.17; and RR 0.91 95% CI 0.65–1.27, respectively). Very low quality evidence (1 RCTs; N = 2318; limitations in study design, inconsistency, imprecision) was found for there being no effect of physical activity on the point prevalence abstinence at the end of the treatment (RR 1.07 95% CI 0.84–1.37).

Fig. 4
figure 4

The effectiveness of aerobic exercise on smoking cessation

Full size image

Evidence of the effectiveness of exercise program on the continuous abstinence rate at the end of the treatment

Ten studies investigated the effectiveness of exercise on the continuous abstinence rate at the end of the treatment [37, 38, 40, 41, 44, 47, 48, 50, 53, 55, 56]. The results indicated moderate quality evidence (10 RCTs; N = 2810; limitations in study design) with there being no significant difference between exercise and control condition on the continuous abstinence rate at the end of the treatment (RR 1.03 95% CI 0.91–1.16) (Fig. 3).

Regarding the effect of the type of exercise (Table 2), moderate quality evidence existed for there being no effect of aerobic exercise (7 RCTs; N = 2156; limitations in study design) on the continuous abstinence rate at the end of treatment (RR 1.04 95% CI 0.91–1.19) (Fig. 4). Low quality evidence was found for there being no effect of the resistance training program (1 RCT; N = 25; inconsistency, imprecision), a combined aerobic and resisted exercise program (1 RCT; N = 330; inconsistency, imprecision) and physical activity (1 RCT; N = 299; inconsistency, imprecision) on the continuous abstinence rate at the end of treatment (RR 1.85 95% CI 0.19–17.84; RR 0.91 95% CI 0.53–1.55; and RR 1.03 95% CI 0.77–1.36, respectively).

Evidence of the effectiveness of exercise program on the point prevalence abstinence rate at the end of the follow-up

Twelve studies investigated the effectiveness of exercise on the point prevalence abstinence rate at the end of the follow-up [40, 42, 43, 45,46,47,48,49,50, 53,54,55]. Low quality evidence (12 RCTs; N = 1289; limitations in study design, imprecision) was found for there being no significant difference between exercise and control conditions on the point prevalence abstinence rate at the end of the follow-up (RR 1.14 95% CI 0.88–1.46) (Fig. 3).

Considering the effect of the type of exercise (Table 2), low quality evidence existed for there being no effect of the aerobic exercise program (9 RCTs; N = 879; limitations in study design, imprecision), the resistance training program (1 RCT; N = 25; inconsistency imprecision), yoga (1 RCT; N = 55; inconsistency, imprecision) and a combined aerobic and resisted exercise program (1 RCT; N = 330; inconsistency, imprecision) on the point prevalence abstinence at the end of the follow-up (RR 1.09 95% CI 0.77–1.54, Fig. 4; RR 2.31 95% CI 0.55–9.74; RR 1.44 95% CI 0.40–5.16; and RR 1.18 95% CI 0.73–1.89, respectively).

Evidence of the effectiveness of exercise program on the continuous abstinence rate at the end of the follow-up

Nine studies investigated the effectiveness of exercise on the continuous abstinence rate at the end of the follow-up [37, 38, 40, 41, 44, 47, 48, 50, 53, 55]. Low quality evidence (9 RCTs; N = 1904; limitations in study design, imprecision) was found for there being no significant difference between exercise and control conditions on the continuous abstinence rate at the end of the follow-up (RR 1.05 95% CI 0.79–1.39) (Fig. 3).

Considering the effect of the type of exercise (Table 2), low quality evidence was found for aerobic exercise(6 RCTs; N = 1250; limitations in study design, imprecision), a resistance training program (1 RCT; N = 25; inconsistency, imprecision), a combined aerobic and resisted exercise program (1 RCT; N = 330; inconsistency, imprecision), and physical activity (1 RCT; N = 299; inconsistency, imprecision) having no effect on the continuous abstinence rate at the end of the follow-up (RR 1.09 95% CI 0.73–1.63, Fig. 4; RR 1.85 95% CI 0.19–17.84; RR 1.81 95% CI 0.69–4.78; and RR 0.73 95% CI 0.38–1.42, respectively).

Discussion

This review evaluated the results of 20 articles (19 studies) on the effectiveness of exercise on smoking cessation. The exercise interventions reported in this review included aerobic exercise, resisted exercise, a combination of aerobic and resisted exercise, yoga, and physical activity. Four studies reported the positive effect of exercise on smoking cessation [45, 47, 49, 55]. This information was similar to a previous systematic review [33]. Only one study reported the positive effect of borderline significance on smoking cessation at 12-month follow-up [47] in our review which was different from a previous systematic review. One positive effect study was excluded from our review [39] because there were some participants in the pre-contemplation stage (participants not thinking about quitting). A previous systematic review by Ussher et al. reported very low quality evidence for whether an exercise program helps people to quit smoking [33]. Our review performed data analysis according to each exercise type in order to decrease treatment variability. The results indicated low to moderate quality evidence for there being no significant difference between aerobic exercise and control conditions on smoking cessation; very low to low quality evidence for there being no significant difference between physical activity and control conditions on smoking cessation; low quality evidence for there being no significant difference between a combined aerobic and resisted exercise program, resistance exercise and control conditions on smoking cessation. Low quality evidence was found for yoga having a statistically positive effect on smoking cessation at the end of the treatment when compared to control condition.

Study characteristics

Exercise types included aerobic exercise, resisted exercise, yoga, a combined aerobic and resisted exercise program and physical activity. Yoga differs from the other exercise types, which are considered to be only bodily exercises. Yoga comprises breathing exercises, meditative components and bodily exercise [55]. The breathing exercises and meditative components have a positive effect in several ways. Breathing exercises and meditative components have been shown to have positive effects on psychological health, such as stress, anxiety and depression reduction [60,61,62,63,64]. Previous studies showed that smoking craving and negative affect were reduced after breathing exercises in abstaining smokers [65, 66]. However, the yoga study included in this review used cognitive-behavioral therapy (CBT) as an adjunct program. CBT may have a different, more intense psychological approach to treatment than standard cessation counseling. CBT could enhance the psychological effect of yoga. Therefore, the positive effect of this study may have been due to psychological health improvement.

The effectiveness of the exercise program correlated to exercise adherence [67]. Access to exercise facilities was one of the factors associated with exercise adherence [68]. A home-based exercise program allows participants to easily access exercise facilities and evidence shows that participants in a home-based exercise program demonstrated higher exercise adherence than those in a group-based, supervised exercise program [69]. However, the rigor of exercise prescription should be considered. In a home-based exercise program, it is not certain that participants strictly follow the prescribed exercises. Of the included studies which used home-based exercise programs, a minority of participants were reported to have strictly followed the prescribed exercises. Thus, the positive effect of exercise may subside due to the exercise prescription not being followed rigorously. Several previous studies used supervised exercise programs which might increase rigorous commitment to the exercise program. The evidence revealed studies using supervised exercise programs reporting the positive effect of exercise when compared to a non-supervised exercise group [70, 71]. Therefore, supervised exercise programs should be considered for smoking cessation treatment and more research is needed on changing multiple behaviors, and achieving sufficient exercise dose through adherence.

Smoking status assessments also varied among the included studies; the point prevalence abstinence and the continuous abstinence. The point prevalence considerably overestimated the numbers who would continue to remain abstinent beyond the follow-up period. Reliance on point prevalence failed to capture the stated treatment goal. Continuous abstinence corresponded more closely to the treatment goal [72]. Therefore, continuous abstinence was recommended as the assessment of smoking cessation.

Methodological considerations

Of the 19 included studies, the one bias criteria rated as low risk in only one study was blinding outcome assessment. Blinding ensures that the apparent effect (or lack of effect) of interventions is not due to bias, which is important for internal validity. The blinding of all assessors eliminates the assessor’s biases [73, 74]. In addition, concealed treatment allocation is also important to prevent selection bias [75]. If treatment allocation is not concealed, the decision as to which group the participants are allocated could be influenced. Only two studies stated the concealment treatment allocation in their study. Therefore, concealment of treatment to minimize bias should be incorporated into the research and stated in the manuscript.

Evidence of the effectiveness of aerobic exercise programs for smoking cessation on the point prevalence and the continuous abstinence rate at the end of the treatment and the follow-up

This review showed low to moderate quality evidence that there was no effect of aerobic exercise on smoking cessation. Previous studies showed a positive effect of acute bouts of aerobic exercise on alleviating nicotine withdrawal symptoms and smoking craving [20, 21], which are important indicators for smoking relapse [3, 22]. However, Roberts et al. showed that the effect of aerobic exercise on decreasing nicotine withdrawal symptoms and smoking craving did not last beyond 20 min post-exercise [21]. Evidence has also suggested that the acute effect of a body scan and isometric exercise to decrease smoking craving was within the first 30 min post exercise in a laboratory setting but within the first 5 min in a normal environment [76]. Therefore, the short duration of the effect of aerobic exercise cannot relieve smoking craving and nicotine withdrawal symptoms throughout the day through just a single bout. Most of the studies in this review prescribed exercise programs only once per day and 3–5 days per week. Further studies should focus on an exercise program which consists of multiple bouts of exercise a day and its influence on smoking cessation.

Evidence of the effectiveness of other exercise types for smoking cessation on the point prevalence and the continuous abstinence rate at the end of the treatment and the follow-up

This review revealed there to be scarce evidence for any effect of other types of exercise on smoking cessation. More studies concerning other types of exercise on smoking cessation are needed before any final conclusions can be reached. However, of the other exercise types, yoga seemed to have a positive effect on smoking cessation. A single bout of yoga showed a positive effect on craving reduction, increased positive affect and decreased negative affect [77]. Smoking craving and negative affect are important indicators of smoking relapse [3, 22, 78]. In addition, the evidence suggested that yoga has a positive effect on stress reduction [79]. Perceived stress correlates with smoking and the ability to quit smoking [80]. Thus, providing a strategy to cope with smoking craving, negative affect and stress may help individuals to stop smoking.

The possible mechanism explained why yoga seems to have a positive effect on smoking cessation is yoga shifts the autonomic balance towards parasympathetic dominance. Several studies revealed that after performing breathing exercise and meditation, integral parts of yoga, parasympathetic activity increased and sympathetic activity decreased [81,82,83,84]. The increase in parasympathetic activity and decrease in sympathetic activity result in relaxation. However, the hypothesis was generated from a small sample of women only.

Two main methodological limitations of this systematic review are to be considered. First, the search strategy was limited to full reported publications in English. The possibility of publication and selection bias cannot be ruled out. This may have affected the results of this review. Second, the results of this review need to be interpreted with caution because some evidence was reached from only one study. Third, the yoga program has an influence on both physiological and psychological aspects. The results of this review cannot identify which aspect could be in charge of the revealed effects.

Conclusions

Twenty articles (19 studies) investigating the effectiveness of exercise for smoking cessation were reviewed and analyzed. The findings revealed low quality evidence for a positive effect on smoking cessation at the end of the treatment in the program where yoga plus CBT was used. However, which of the two work is still to be studied. Low to moderate quality evidence was found for there being no effect of aerobic exercise, resisted exercise, and a combined aerobic and resisted exercise. Very low to low quality evidence was found for there being no effect of physical activity on smoking cessation. Of 19 studies, 12 studies were rated as having high risk of bias. Thus, more high quality studies about the effectiveness of exercise on smoking cessation are needed. The design of future studies may be improved by addressing the number of methodological limitations, namely, the blinding of all assessors and concealing of treatment allocation.