Downloaded from bjsm.bmj.com on May 21, 2014 - Published by group.bmj.com
BJSM Online First, published on May 25, 2013 as 10.1136/bjsports-2013-092525 PEDro systematic review update This section features a recent systematic review that is indexed on PEDro, the Physiotherapy Evidence Database (www.pedro.org.au). PEDro is a free, web-based database of evidence relevant to physiotherapy.
Physical activity reduces cigarette cravings
STATISTICAL METHODS Both one-stage and two-stage meta-analyses were performed. The one-stage meta-analysis combined study data using a mixed linear regression model. The two-stage meta-analysis used individual participant data to derive an effect size in terms of mean difference between intervention and control groups for outcomes in each trial. Heterogeneity was identified by visual inspection of forest plots and with I2 and Q statistics.
RESULTS Haasova M, Warren FC, Ussher M, et al. The acute effects of physical activity on cigarette cravings: systematic review and meta-analysis with individual participant data. Addiction 2013;108:26–37.
BACKGROUND In the UK, over one-fifth of the adult population smokes. Many smokers (63%) wish to stop, but only 3–5% of unaided attempts are successful after 6–12 months and most people who stop smoking relapse within the first 8 days. Even with the best available pharmacological and behavioural support, lesser than 30% of smokers successfully stop.1–3 Physical activity (PA) is recommended as a smoking cessation aid.4 However, the evidence for the efficacy of PA in aiding cessation is limited.5
AIM The aim of this systematic review and individual patient data (IPD) meta-analysis was to investigate the acute effects of short bouts of PA on cigarette craving.
SEARCHES AND INCLUSION CRITERIA Six electronic databases (SPORTDiscus, MEDLINE, PubMed, Web of Science, EMBASE and PsycINFO) were searched. The Cochrane Tobacco Addiction Group specialised register, electronic theses and dissertations (ETD), Digital Library—Network Digital Library of Theses and Dissertations (NDLTD) and ProQuest Digital Dissertations were also searched. Search terms used were ‘smoking’ or ‘smoking cessation’, ‘exercise’ or ‘physical activity’ and ‘craving$’ or ‘withdrawal’. Relevant reference lists and meeting abstracts were hand searched. Eligible studies examined the acute effects of PA on the desire to smoke and/or strength of desire to smoke and included randomised crossover or parallel arm trials with a minimum abstinence period of 2 h prior to baseline measurement. Trials where participants were enrolled in cessation programmes or using nicotine replacement therapy were excluded. Both published and unpublished studies were included.
INTERVENTIONS PA interventions included moderate-intensity walking, running, light, moderate and vigorous intensity cycling and isometric exercise. Intensity was described using the rating of perceived exertion, percentage of heart rate (HR) maximum, HR reserve or a combination of these. Duration of the PA and control interventions ranged from 5 to 40 min. All control interventions were passive such as passive sitting, listening to an audio recording, a cognitive task or watching a video.
MAIN OUTCOME MEASURES Two outcome measures were used to assess cigarette craving, desire to smoke and strength of desire to smoke, both using a seven point Likert scale. Trials using a 0–5 scale had values adjusted up to match. Glass TW, et al. Br JArticle Sports Medauthor Month 2013 0 No 0employer) Copyright (orVoltheir
In total, 19 studies were included in the meta-analyses, of which 7 used a parallel-arm design and 12 used a crossover design. Strength of desire to smoke was the main outcome in 15 studies (797 participants) and desire to smoke was the main outcome in 17 studies (837 participants). The two-stage IPD random effects meta-analysis for strength of desire to smoke yielded an average standardised mean difference of −1.91 across studies (95% CI −2.59 to −1.22) and −2.03 for desire to smoke (95% CI −2.60 to −1.46), indicating benefits from the intervention (figure 1). Both analyses showed high heterogeneity between studies. The one-stage random effects meta-analysis for strength of desire to smoke yielded a fixed-effect size (mean difference) −1.89 (95% CI −2.53 to −1.26) and for desire to smoke, −2.03 (95% CI −2.54 to −1.51). When the two-stage meta-analysis included only moderate intensity PA interventions (walking or cycling), an average standardised mean difference across studies of −2.20 (95% CI −2.89 to −1.51) for strength of desire to smoke (13 studies) and −2.14 (95% CI −2.71 to −1.57) for desire to smoke (16 studies) was observed. Again, both showed high heterogeneity between studies. Results for the differing study designs were very similar, as were those from published and unpublished studies.
LIMITATIONS This review of studies on the acute effects of short bouts of PA on cigarette craving showed large reductions in craving following PA intervention. The treatment effect sizes in some cases exceeded 2 SDs—a magnitude rarely seen in clinical trials. When meta-analyses were limited to only moderate-intensity PA, effect sizes were noticeably larger, suggesting that the prospective moderating effects of PA intensity (and potentially type and duration) on cigarette craving require further investigation. IPD meta-analyses allow the combining of data from randomised controlled trials with the same outcome and may be beneficial when including unpublished studies or ‘grey literature’. IPD may also minimise publication bias, improve data and analysis quality and enhance the interpretation of findings. However, including ‘grey literature’ risks introduces other forms of bias. We note the limitations of self-reported measurement of cigarette craving, which is subject to social desirability bias. This may artificially inflate the estimates of intervention effect. There was a noticeable statistical heterogeneity (ie, effects varied widely between studies) which may be due to the variation between experimental protocols of the included studies. Future studies with larger samples should investigate the relative importance of type, duration and intensity of PA interventions as well as the long-term effects of acute craving reduction through PA intervention on smoking cessation, as current evidence is limited.5
2013. Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from bjsm.bmj.com on May 21, 2014 - Published by group.bmj.com
PEDro systematic review update
Figure 1 Individual participant meta-analysis data of all studies using a two-stage random effects regression model of post ‘desire to smoke’ with study and baseline adjustment. Negative effect size (ES) favours intervention and positive ES favours control (reproduced with permission from original paper6: Haasova M, Warren FC, Ussher M, et al. The acute effects of physical activity on cigarette cravings: systematic review and metaanalysis with individual participant data. Addiction 2013;108:26–37).
CLINICAL IMPLICATIONS
Competing interests None.
There is a growing body of evidence to support the use of short bouts of PA to reduce cigarette craving. The review highlights the potential of a single session of PA to reduce cravings, particularly when cravings are high, suggesting that primary care clinicians can assist their patients in a quit attempt by advising short bouts of PA, such as moderate walking or cycling, to reduce cigarette craving.
Provenance and peer review Commissioned; internally peer reviewed.
Timothy W Glass,1 Christopher G Maher2 1
Warwick Medical School, The University of Warwick, Coventry, UK Musculoskeletal Division, The George Institute for Global Health, Sydney, New South Wales, Australia
2
Correspondence to Dr Christopher G Maher, Musculoskeletal Division, The George Institute for Global Health, PO Box M201, Missenden Road, NSW 2050 Australia;
[email protected] Contributors TWG selected the systematic review and wrote the first draft of the manuscript. TWG and CGM contributed to interpretation of the data and revision of the final version of the manuscript, and both are guarantors.
2
To cite Glass TW, Maher CG. Br J Sports Med Published Online First: [ please include Day Month Year] doi:10.1136/bjsports-2013-092525 Accepted 20 April 2013 Br J Sports Med 2013;0:1–2. doi:10.1136/bjsports-2013-092525
REFERENCES 1 2 3 4
5 6
Robinson S, Harris H. Smoking and drinking among adults, 2009. A report on the 2009 General Lifestyle Survey. London, UK: Office for National Statistics, 2011. Hughes JR, Keely J, Naud S. Shape of the relapse curve and long-term abstinence among untreated smokers. Addiction 2004;99:29–38. Hughes JR. How confident should we be that smoking cessation treatments work? Addiction 2009;104:1637–40. Everson ES, Taylor AH, Ussher M. Determinants of physical activity promotion by smoking cessation advisors as an aid for quitting: support for the transtheoretical model. Patient Educ Couns 2010;78:53–6. Ussher MH, Taylor A, Faulkner G. Exercise interventions for smoking cessation. Cochrane Database Syst Rev 2012;1:CD002295. Haasova M, Warren FC, Ussher M, et al. The acute effects of physical activity on cigarette cravings: systematic review and meta-analysis with individual participant data. Addiction 2013;108:26–37.
Glass TW, et al. Br J Sports Med Month 2013 Vol 0 No 0
Downloaded from bjsm.bmj.com on May 21, 2014 - Published by group.bmj.com
Physical activity reduces cigarette cravings Timothy W Glass and Christopher G Maher Br J Sports Med published online May 25, 2013
doi: 10.1136/bjsports-2013-092525
Updated information and services can be found at: http://bjsm.bmj.com/content/early/2013/05/24/bjsports-2013-092525.full.html
These include:
References
This article cites 5 articles http://bjsm.bmj.com/content/early/2013/05/24/bjsports-2013-092525.full.html#ref-list-1
P
