Advertisement

Multiple Risk Behavior Interventions: Meta-analyses of RCTs

Open AccessPublished:February 28, 2017DOI:https://doi.org/10.1016/j.amepre.2017.01.032

      Context

      Multiple risk behaviors are common and associated with developing chronic conditions such as heart disease, cancer, or Type 2 diabetes. A systematic review, meta-analysis, and meta-regression of the effectiveness of multiple risk behavior interventions was conducted.

      Evidence acquisition

      Six electronic databases including MEDLINE, EMBASE, and PsycINFO were searched to August 2016. RCTs of non-pharmacologic interventions in general adult populations were selected. Studies targeting specific at-risk groups (such as people screened for cardiovascular risk factors or obesity) were excluded. Studies were screened independently. Study characteristics and outcomes were extracted and risk of bias assessed by one researcher and checked by another. The Behaviour Change Wheel and Oxford Implementation Index were used to code intervention content and context.

      Evidence synthesis

      Random-effects meta-analyses were conducted. Sixty-nine trials involving 73,873 individuals were included. Interventions mainly comprised education and skills training and were associated with modest improvements in most risk behaviors: increased fruit and vegetable intake (0.31 portions, 95% CI=0.17, 0.45) and physical activity (standardized mean difference, 0.25; 95% CI=0.13, 0.38), and reduced fat intake (standardized mean difference, –0.24; 95% CI=–0.36, –0.12). Although reductions in smoking were found (OR=0.78, 95% CI=0.68, 0.90), they appeared to be negatively associated with improvement in other behaviors (such as diet and physical activity). Preliminary evidence suggests that sequentially changing smoking alongside other risk behaviors was more effective than simultaneous change. But most studies assessed simultaneous rather than sequential change in risk behaviors; therefore, comparisons are sparse. Follow-up period and intervention characteristics impacted effectiveness for some outcomes.

      Conclusions

      Interventions comprising education (e.g., providing information about behaviors associated with health risks) and skills training (e.g., teaching skills that equip participants to engage in less risky behavior) and targeting multiple risk behaviors concurrently are associated with small changes in diet and physical activity. Although on average smoking was reduced, it appeared changes in smoking were negatively associated with changes in other behaviors, suggesting it may not be optimal to target smoking simultaneously with other risk behaviors.

      Context

      Physical inactivity, eating an unhealthy diet, smoking, and excessive alcohol consumption are associated with greater risk of developing cancers, cardiovascular diseases, and Type 2 diabetes
      • Strong K.
      • Mathers C.
      • Leeder S.
      • Beaglehole R.
      Preventing chronic diseases: how many lives can we save?.
      ; together, these conditions are estimated to account for more than 50% of preventable premature deaths globally. Studies suggest the majority of adults report two or more risk behaviors and approximately 25% of the adult population report three or more risk behaviors.
      • Schuit A.J.
      • van Loon A.J.
      • Tijhuis M.
      • Ocké M.
      Clustering of lifestyle risk factors in a general adult population.
      • Chou K.-L.
      The prevalence and clustering of four major lifestyle risk factors in Hong Kong Chinese older adults.
      • Baruth M.
      • Addy C.
      • Wilcox S.
      • et al.
      Clustering of risk behaviours among African American adults.
      Engaging in multiple risk behaviors is associated with greater risk of chronic disease and mortality compared with engaging in one or no risk behaviors.
      • Myint P.K.
      • Luben R.N.
      • Wareham N.J.
      • Bingham S.A.
      • Khaw K.
      Combined effect of health behaviours and risk of first ever stroke in 20,040 men and women over 11 years’ follow-up in Norfolk cohort of European Prospective Investigation of Cancer (EPIC Norfolk): prospective population study.
      Multiple risk behaviors are associated with health inequalities; people in unskilled work or with no qualifications are more likely to engage in two or more risk behaviors.
      • Poortinga W.
      The prevalence and clustering of four major lifestyle risk factors in an English adult population.
      • Lawder R.
      • Harding O.
      • Stockton D.
      • et al.
      Is the Scottish population living dangerously? Prevalence of multiple risk factors: the Scottish Health Survey 2003.
      • Buck D.
      • Frosini F.
      Clustering of Unhealthy Behaviours Over Time: Implications for Policy and Practice.
      High blood pressure and tobacco smoke are among the three leading risk factors for global disease burden, and unhealthy diet and physical inactivity accounted for 10% of all disease burden.
      • Lim S.S.
      • Vos T.
      • Flaxman A.D.
      • et al.
      A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010.
      Interventions supporting people to make healthy choices receive high priority in most high-income countries because of potential to improve population health and reduce future demand on health care. Given risk behaviors rarely occur in isolation, tackling multiple rather than single behaviors may be a more effective approach. Since 2000, there has been a steady increase in studies evaluating multiple risk behavior interventions, with a sharp rise from 2010 onward. Between 2010 and 2013, more than 100 studies were published.
      • King K.
      • Meader N.
      • Wright K.
      • et al.
      Characteristics of interventions targeting multiple lifestyle risk behaviours in adult populations: a systematic scoping review.
      Findings from these studies have not yet been synthesized and, to the authors’ knowledge, this is the first systematic review to evaluate which interventions are effective in changing which behaviors in adult, non-clinical populations. Which behaviors were targeted, by what type of intervention, and the achieved outcomes were investigated. This review also explored which factors were associated with improved outcomes.

      Evidence Acquisition

      Eligibility Criteria

      An iterative approach was used to determine inclusion criteria. This involved conducting a mapping exercise appraising the scope of the multiple risk behavior literature. No evidence was found suggesting restriction of studies to RCTs would limit types of eligible interventions.
      • King K.
      • Meader N.
      • Wright K.
      • et al.
      Characteristics of interventions targeting multiple lifestyle risk behaviours in adult populations: a systematic scoping review.
      Studies with the following characteristics were included.

      Population

      General adult (aged ≥16 years) or non-targeted subgroups of general adult populations (e.g., pregnant women, older adults, students) were included. Studies of targeted subgroups, where screening takes place to determine eligibility (e.g., to identify obesity, or those at risk of Type 2 diabetes), were excluded.

      Intervention

      Any non-pharmacologic intervention aiming to change at least two risk behaviors (risk behaviors were not determined a priori) was included. Studies of school- or family-based interventions were excluded to avoid duplication with registered protocol for a Cochrane systematic review.
      • MacArthur G.
      • Kipping R.
      • White J.
      • et al.
      Individual-, family-, and school-level interventions for preventing multiple risk behaviours in individuals aged 8 to 25 years.

      Comparator

      Any comparator (such as attention control, single risk behavior non-pharmacologic intervention) was included.

      Outcomes

      The primary outcome was change in risk behaviors. This included any behavior that entailed potential risk to participants’ health. Secondary outcomes were changes in weight, BMI, blood pressure, and cholesterol; intermediate outcomes included self-efficacy, attitudes, beliefs, and knowledge. Process-related outcomes were collected.

      Information Sources

      MEDLINE, EMBASE, PsycINFO, Science Citation Index, Cochrane Central Register for Controlled Trials, and Applied Social Sciences Index and Abstracts were searched from January 1990 to August 2016 with no language restrictions (Appendix, available online). Citation searches were carried out using Google Scholar, Scopus, Web of Science, and OVIDSP MEDLINE.
      • Wright K.
      • Golder S.
      • Rodriguez-Lopez R.
      Citation searching: a systematic review case study of multiple risk behaviour interventions.

      Selection of Studies, Data Collection Process, and Risk of Bias Assessment

      Final selection of studies, data extraction, and assessment of risk of bias was conducted by one reviewer and checked by a second. A modified version of the Cochrane Public Health Group’s data extraction template was used (piloted on five studies to ensure consistency) and the Behaviour Change Wheel (Table 1) to classify intervention content according to nine functions: education, persuasion, incentivization, coercion, training, enablement, modeling, environmental restructuring, and restrictions (no policy-level interventions were found).
      • Michie S.
      • van Stralen M.M.
      • West R.
      The Behaviour Change Wheel: a new method for characterising and designing behaviour change interventions.
      Table 1Summary of Behavior Change Wheel
      FunctionsDefinitionExamples
      Interventions
       EducationSeeking to provide or increase knowledgeEducational material provided through lectures, online, or written materials
       PersuasionSeeking to induce positive or negative feelings that impacts on behaviorUsing motivational interviewing to change behavior
       IncentivizationProviding positive reinforcement to change behaviorProviding vouchers contingent on engaging in a particular healthy behavior
       CoercionProviding negative reinforcement or punishment to change behaviorHaving to pay a fine for engaging in a risk behavior
       TrainingTraining participants to develop skills that help them to engage in healthy behaviorTeaching cooking skills to people who have an unhealthy diet
       RestrictionUsing rules to reduce or increase a particular behaviorProhibiting the use of novel psychoactive substances
       Environmental restructuringIntervening in the social or physical context to promote or reduce particular behaviorsIntegrating a health promotion program within the regular social activities of an African American church to encourage behavior change in their members
       ModelingProviding an example of someone engaging in a behavior or changing their behaviorRecruiting people who inject drugs and train them to promote use of clean needles within their social networks
       EnablementReducing barriers and providing support to help behavior changeProviding pedometers to help participants monitor their activity levels
      Policies
       Communication/marketingUsing media (e.g., newspapers, social media, TV) to promote healthy behaviorConducting mass media campaigns
       GuidelinesDeveloping guidance recommending engaging or not engaging in particular behaviorsNational guideline programs such as the National Institute for Health and Care Excellence
       FiscalTaxing unhealthy behaviors or offering subsidies to promote healthy behaviorIncrease taxes on tobacco, high sugar foods
       RegulationRules or principles that encourage healthy behaviorVoluntary agreements on advertising of unhealthy foods or drinks
       LegislationLegislating against unhealthy behaviorProhibiting the sale of tobacco to certain age groups
       Environmental/social planningPolicies related to the physical or social environmentTown planning to make cycling safer and more accessible to citizens
       Service provisionProviding a service that promotes healthy behaviorLocal authorities providing affordable and accessible gyms
      Source: Adapted From Michie et al.
      • Michie S.
      • van Stralen M.M.
      • West R.
      The Behaviour Change Wheel: a new method for characterising and designing behaviour change interventions.
      The Oxford Implementation Index was utilized to assess intervention characteristics and contextual factors.
      • Montgomery P.
      • Underhill K.
      • Gardner F.
      • et al.
      The Oxford Implementation Index: a new tool for incorporating implementation data into systematic reviews and meta-analyses.
      Both were adapted for the purposes of this review. The Cochrane Risk of Bias Tool was used to critically appraise included studies.
      • Higgins J.P.T.
      • Green S.
      Cochrane Handbook for Systematic Reviews of Interventions Version 5.
      For dichotomous outcomes, ORs and their 95% CIs were calculated, with values <1 favoring the intervention group. For continuous outcomes, standardized mean differences (SMDs) were calculated using Hedges’s g.
      • Hedges L.V.
      Distribution theory for Glass’s estimator of effect size and related estimators.
      Where a sufficient number of studies were available, mean differences were calculated on original scales (e.g., portions of fruit and vegetables).

      Statistical Methods

      Meta-analyses

      Random-effects meta-analyses using Review Manager, version 5, were calculated. Control conditions were grouped into three categories (minimal intervention, information provision, active control) to examine differences in effect estimates across these conditions.
      Heterogeneity assessment was based on visual inspection of forest plots and the I2 statistic.
      • Higgins J.
      • Thompson S.G.
      Quantifying heterogeneity in a meta-analysis.
      A Q-value (approximating χ2 distribution) of p<0.10 indicated statistically significant heterogeneity. Statistical heterogeneity was explored using meta-regression.

      Meta-regression analyses

      Mixed-effects meta-regression analyses (where there were at least ten studies for an outcome) were conducted to examine the influence of implementation factors on effectiveness based on criteria from the Oxford Implementation Index
      • Montgomery P.
      • Underhill K.
      • Gardner F.
      • et al.
      The Oxford Implementation Index: a new tool for incorporating implementation data into systematic reviews and meta-analyses.
      and the Behaviour Change Wheel.
      • Michie S.
      • van Stralen M.M.
      • West R.
      The Behaviour Change Wheel: a new method for characterising and designing behaviour change interventions.
      A permutation test adjusted p-values to reduce risk of false positives.
      • Higgins J.
      • Thompson S.
      Controlling the risk of spurious findings from meta-regression.
      Although meta-regression analyses were planned in advance, the findings should be considered exploratory given the large number of examined covariates.
      Covariates relating to intervention characteristics included number of intervention functions; specific intervention functions (as defined by the Behaviour Change Wheel); method of delivery; intervention duration; staff characteristics; participant characteristics; intervention setting; publication period; and duration of follow-up.

      Additional analyses

      Multivariate meta-analyses of correlated outcomes were compared with standard univariate meta-analyses. Subgroup analyses according to SES (studies with predominantly low SES versus mixed SES) and ethnicity (participants were predominantly from a black and minority ethnic population versus participants from majority and minority ethnic populations) were conducted.

      Evidence Synthesis

      Sixty-nine RCTs (comprising 73,873 participants) were included (Figure 1). Study quality was variable (Appendix Figure 1, available online). Blinding of participants and personnel was not included in the risk of bias assessment because it was not feasible given the nature of the interventions. Slightly more than half of the studies had high risk of bias for at least one domain: incomplete outcome data (attrition bias, n=27); other bias (n=8); blinding of outcome assessors (n=6); selective reporting (n=5); and allocation concealment (n=3).
      Contextual factors, participant characteristics, and intervention characteristics were extracted according to the Oxford Implementation Index (Appendix, available online). Most studies were conducted in the U.S. (n=34); United Kingdom (n=9); Netherlands (n=6); and Australia (n=5). Settings varied, including homes, community centers, churches, universities, primary care clinics, hospitals, and prisons. Few studies reported information about the wider environment in which the intervention took place or characteristics of the delivering organization. Data on other contextual factors such as occurrence of important external events at the time of intervention were limited.
      General adult populations were the focus in most studies (n=32). Others targeted students (n=13); older adults (n=8); pregnant women (n=4); and prisoners (n=1). Some specifically targeted those on low incomes (n=8) or black and minority ethnic groups (n=5). There is some overlap in categories; therefore, summing the totals exceeds the number of included studies.
      Most studies targeted two risk behaviors (n=32). Fewer studies targeted three (n=17); four (n=13); or five behaviors (n=2) (Appendix, available online). Most (72%) targeted diet and physical activity, with 46% focusing exclusively on these behaviors; 35% targeted diet and smoking but few focused exclusively on these behaviors; and 23% targeted alcohol and smoking but few focused exclusively on these behaviors.
      Number of intervention functions ranged from one (n=8) to five (n=4), with most including three functions (n=28). Coercion and restriction were not included as part of any intervention, and incentives and environmental restructuring were rarely used. Most (n=66) included an education function, and just more than half included education with training (n=39). Persuasion was also used in a number of studies (n=21). These functions are consistent with the mostly commonly adopted theoretic approaches, including Social Cognitive Theory,
      • Bandura A.
      Social Foundations of Thought and Action: A Social Cognitive Theory.
      the Health Belief Model,
      • Janz N.K.
      • Becker M.H.
      The Health Belief Model: a decade later.
      and the Theory of Planned Behaviour
      • Fishbein M.
      • Ajzen I.
      Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research.
      (the Appendix [available online] summarizes intervention functions classified using the Behaviour Change Wheel,
      • Michie S.
      • van Stralen M.M.
      • West R.
      The Behaviour Change Wheel: a new method for characterising and designing behaviour change interventions.
      reported theoretic approaches, and targeted risk behaviors).
      No clear patterns between particular risk behavior combinations and use of specific intervention functions were detected. Studies targeting a larger number of behaviors did not appear to adopt more intervention functions than studies targeting two behaviors.
      Although most studies (n=47) reported the theoretic basis of interventions, few reported examining changes in intermediate outcomes (e.g., attitudes, beliefs, and knowledge) predicted by theory to mediate behavior change (Appendix, available online).
      Twenty-three studies provided process evaluation data (Appendix, available online). These data were mostly related to participant uptake of materials such as pedometers, resistance bands, exercise calendars, and written/online materials. Only four studies reported analyses of intervention fidelity or challenges to implementation.
      Imperial Cancer Research Fund OXCHECK Study Group
      Effectiveness of health checks conducted by nurses in primary care: final results of the OXCHECK study.
      • McCambridge J.
      • Hunt C.
      • Jenkins R.J.
      • et al.
      Cluster randomised trial of the effectiveness of motivational interviewing for universal prevention.
      • Yanek L.
      • Becker D.
      • Moy T.
      • et al.
      Project Joy: faith based cardiovascular health promotion for African American women.
      • Hillier F.C.
      • Batterham A.M.
      • Nixon C.A.
      • et al.
      A community-based health promotion intervention using brief negotiation techniques and a pledge on dietary intake, physical activity levels and weight outcomes: lessons learnt from an exploratory trial.
      Most studies that requested participant feedback found high levels of satisfaction both with provided materials (e.g., pedometers, exercise calendars, educational materials) and content of the intervention.
      • Burke L.
      • Jancey J.M.
      • Howat P.
      • et al.
      Physical Activity and Nutrition Program for Seniors (PANS): process evaluation.
      • de Vries H.
      • Kremers S.P.J.
      • Smeets T.
      • et al.
      The effectiveness of tailored feedback and action plans in an intervention addressing multiple health behaviors.
      • Lee A.
      • Jancey J.
      • Howat P.
      • et al.
      Effectiveness of a home-based postal and telephone physical activity and nutrition pilot program for seniors.
      • van Assema P.
      • Steenbakkers M.
      • Kok G.
      • et al.
      Results of the Dutch Community Project “Healthy Bergeyk.”.
      • van Keulen H.M.
      • Mesters I.
      • Ausems M.
      • et al.
      Tailored print communication and telephone motivational interviewing are equally successful in improving multiple lifestyle behaviors in a randomized controlled trial.
      • Werch C.E.
      • Moore M.J.
      • Bian H.
      • et al.
      Are effects from a brief multiple behavior intervention for college students sustained over time?.
      • Wilcox S.
      • Parrott A.
      • Baruth M.
      • et al.
      The Faith, Activity, and Nutrition Program: a randomized controlled trial in African-American churches.
      Though some studies did not find differences in satisfaction between participants in the intervention group compared with controls,
      • Sikkema K.J.
      • Winett R.A.
      • Lombard D.N.
      Development and evaluation of an HIV-risk reduction program for female college students.
      • Ussher M.
      • West R.
      • McEwen A.
      • et al.
      Efficacy of exercise counselling as an aid for smoking cessation: a randomized controlled trial.
      • Werch C.
      • Bian H.
      • Moore M.J.
      • et al.
      Brief multiple behavior interventions in a college student health care clinic.
      these interventions were compared with relatively active control groups.
      Participant’s perceived effectiveness of interventions and engagement in behavior change were less positive. Only 55% of participants in one study
      • Burke L.
      • Jancey J.M.
      • Howat P.
      • et al.
      Physical Activity and Nutrition Program for Seniors (PANS): process evaluation.
      felt the intervention helped them to improve their diet. Similarly, another study
      • Lee A.
      • Jancey J.
      • Howat P.
      • et al.
      Effectiveness of a home-based postal and telephone physical activity and nutrition pilot program for seniors.
      found <50% of the participants considered the intervention effective in improving their diet and physical activity, with only 25% engaging in new activities. Although participants in another study
      • van Keulen H.M.
      • Mesters I.
      • Ausems M.
      • et al.
      Tailored print communication and telephone motivational interviewing are equally successful in improving multiple lifestyle behaviors in a randomized controlled trial.
      were satisfied with their motivational interviewing session, most provided neutral responses concerning perceived relevance. In another study,
      • Leslie W.S.
      • Koshy P.R.
      • Mackenzie M.
      • et al.
      Changes in body weight and food choice in those attempting smoking cessation: a cluster randomised controlled trial.
      several themes were identified in relation to participants’ perception of the intervention and subsequent impact on behavior change: the importance of group interaction (such as accountability, but also disappointment when group members stop attending); encouragement provided by advisors (such as improved motivation); and specific helpful aspects of the intervention (such as use of pedometers, goal setting).
      Summary estimates from the meta-analyses are presented in Table 2, Table 3. Subgroup analyses according to control group category (minimal intervention/information provision/active control) did not substantially change the pooled results and are not discussed further.
      Table 2Summary Point Estimates From the Meta-analyses of Multiple Risk Behavior Interventions (Primary Outcomes)
      Risk behavior outcomeSummary point estimateFollow-up time
      AllLow SESBME
      Dichotomous data
       Lack of fruit and vegetable intake: not adhering to FV recommendationsOR 0.62 (95% CI 0.51 to 0.76) I2=81%, K=11OR 0.65 (95% CI 0.44 to 0.83) I2=48%, K=3OR 0.65 (95% CI 0.38 to 1.12) I2=N/A, K=1Mean: 4 months Range: endpoint to 12 months
       Intake of fat/meat/dairy: not adhering to recommendationsOR 0.70 (95% CI 0.61 to 0.81) I2=0%, K=3OR 0.73 (95% CI 0.61 to 0.88) I2=N/A, K=1N/AMean: 5 months Range: endpoint to 8 months
       Physical activity: not adhering to physical activity recommendationsOR 0.73 (95% CI 0.65 to 0.83) I2=64%, K=19OR 0.85 (95% CI 0.72 to 1.00) I2=0%, K=4OR 0.58 (95% CI 0.38 to 0.87) I2=N/A, K=1Mean: 4 months Range: endpoint to 12 months
       SmokingOR 0.78 (95% CI 0.68 to 0.90) I2=63, K=17N/AN/AMean: 4 months Range: endpoint to 12 months
       Alcohol misuse: not adhering to alcohol intake recommendationsOR 0.84 (95% CI 0.65 to 1.08) I2=60%, K=5N/AOR 0.59 (95% CI 0.20 to 1.76) I2=N/A, K=1Mean: 5 months Range: endpoint to 12 months
      Continuous data
       Calorie intakeMD –83.37 (95% CI –148.54 to –18.20) I2=80%, K=9N/AN/AMean: 3 months Range: endpoint to 12 months
       Fruit and vegetable intake (post-intervention)SMD 0.17 (95% CI 0.11 to 0.23) I2=61%, K=22 Portions of fruit and vegetables: MD 0.31 (95% CI 0.17 to 0.45) I2=56%, K=13SMD 0.22 (95% CI 0.13 to 0.31) I2=0%, K=2 Portions of fruit and vegetables: MD 0.48 (95% CI 0.32 to 0.64) I2=0%, K=3SMD 0.14 (95% CI 0.06 to 0.22) I2=0%, K=3 Portions of fruit and vegetables: MD 0.37 (95% CI 0.15 to 0.59) I2=0%, K=2Mean: 5 months Range: endpoint to 12 months
       Intake of fat/meat/dairy (post-intervention)SMD –0.24 (95% CI –0.36 to –0.12) I2=82%, K=17SMD –0.14 (95% CI –0.22 to –0.06) I2=0%, K=3SMD –0.04 (95% CI –0.15 to 0.08) I2=0%, K=2Mean: 4 months Range: endpoint to 12 months
       Physical activity (post-intervention)SMD 0.25 (95% CI 0.13 to 0.38) I2=93%, K=27SMD 0.05 (95% CI 0.18 to 0.29) I2=56%, K=3SMD 0.12 (0.01 to 0.23) I2=32%, K=3Mean: 5 months Range: endpoint to 12 months
       Sexual risk behaviorsSMD –0.12 (95% CI –0.49 to 0.24) I2=32%, K=3N/AN/AMean: 4 months Range: 1–6 months
      Note: Results in bold are statistically significant.
      BME, black and minority ethnic groups; FV, fruit and vegetable intake; K, number of trials; MD, mean difference; N/A, not applicable; SMD, standardized mean difference.
      Table 3Summary Point Estimates From the Meta-analyses of Multiple Risk Behavior Interventions (Secondary and Intermediate Outcomes)
      OutcomeSummary point estimatesFollow-up time
      AllLow SESBME
      Self-efficacySMD –0.06 (95% CI –0.17 to 0.06) I2=71%, K=8N/ASMD 0.16 (95% CI 0.02 to 0.30) I2=N/A, K=1Mean: 4 months Range: endpoint to 9 months
      Weight (kg)MD –0.59 (95% CI –1.02 to –0.16) I2=57%, K=18MD –0.76 (95% CI –2.30 to 0.79) I2=41%, K=3MD –0.88 (95% CI –1.47 to –0.29) I2=N/A, K=1Mean: 5 months Range: endpoint to 12 months
      BMIMD –0.27 (95% CI –0.46 to –0.07) I2=65%, K=14MD –0.58 (95% CI –1.45 to 0.29) I2=7%, K=2MD –0.31 (95% CI –0.53 to –0.09) I2=N/A, K=1Mean: 5 months Range: endpoint to 15 months
      Systolic blood pressureSMD –0.11 (95% CI –0.19 to –0.04) I2=56%, K=13SMD 0.07 (95% CI –0.20 to 0.34) I2=N/A, K=1SMD –0.06 (95% CI –0.31 to 0.19) I2=N/A, K=1Mean: 6 months Range: endpoint to 24 months
      Diastolic blood pressureSMD –0.11 (95% CI –0.19 to –0.04) I2= 51%, K=13SMD 0.00 (95% CI –0.27 to 0.27) I2=N/A, K=1SMD –0.10 (95% CI –0.34 to 0.14) I2=N/A, K=1Mean: 6 months Range: endpoint to 24 months
      Total cholesterolSMD –0.17 (95% CI –0.27 to –0.06) I2= 81%, K=12SMD 0.00 (95% CI –0.27 to 0.27) I2=N/A, K=1SMD –0.09 (95% CI –0.34 to 0.16) I2=N/A, K=1Mean: 6 months Range: endpoint to 24 months
      HDL cholesterolSMD –0.13 (95% CI –0.31 to 0.05) I2= 87%, K=9SMD –0.12 (95% CI –0.39 to 0.15) I2=N/A, K=1SMD –0.11 (95% CI –0.35 to 0.13) I2=N/A, K=1Mean: 8 months Range: endpoint to 24 months
      LDL cholesterolSMD –0.17 (95% CI –0.34 to 0.00) I2= 85%, K=9SMD –0.04 (95% CI –0.31 to 0.23) I2=N/A, K=1SMD –0.09 (95% CI –0.33 to 0.15) I2=N/A, K=1Mean: 8 months Range: endpoint to 24 months
      Note: Results in bold are statistically significant.
      BME, black and minority ethnic groups; HDL, high-density lipoprotein; K, number of trials; LDL, low-density lipoprotein; MD, mean difference; N/A, not applicable; SMD, standardized mean difference.
      Compared with control groups, the intervention groups demonstrated a small increase in fruit and vegetable intake (0.31 portions, 95% CI=0.17, 0.45); a small reduction in calorie intake (–83.37, 95% CI= –148.54, –18.20) and fat intake (SMD= –0.24, 95% CI= –0.36, –0.12); and a small increase in physical activity (SMD=0.25, 95% CI=0.13, 0.38). However, the findings for physical activity were sensitive to an individual study
      • Foroushani A.R.
      • Estebsari F.
      • Mostafaei D.
      • et al.
      The effect of health promoting intervention on healthy lifestyle and social support in elders: a clinical trial study.
      where the intervention was more effective than in other studies (SMD=2.94). When this study was removed from the analysis, the effect estimate (SMD=0.15, 95% CI=0.09, 0.21) and heterogeneity (I2=61% vs 93% when considering all studies) were substantially reduced.
      Small to moderate improvements were found in overall diet score, fiber intake, calorie intake, sodium intake, alcohol use, and reduction of sexual risk behaviors, but there were few studies and some results lacked precision (i.e., wide CIs). There was also a statistically significant reduction in smoking (OR=0.78, 95% CI=0.68, 0.90).
      Two studies compared multiple and single risk behavior interventions.
      • Leslie W.S.
      • Koshy P.R.
      • Mackenzie M.
      • et al.
      Changes in body weight and food choice in those attempting smoking cessation: a cluster randomised controlled trial.
      • Bickmore T.W.
      • Schulman D.
      • Sidner C.
      Automated interventions for multiple health behaviors using conversational agents.
      One study
      • Leslie W.S.
      • Koshy P.R.
      • Mackenzie M.
      • et al.
      Changes in body weight and food choice in those attempting smoking cessation: a cluster randomised controlled trial.
      compared an intervention targeting two behaviors (smoking and diet) with an intervention targeting a single behavior (smoking). No statistically significant differences were found between groups for either smoking or diet. Another study compared physical activity alone, fruit and vegetable intake alone, combined physical activity and fruit and vegetable intake, and non-intervention control groups.
      • Bickmore T.W.
      • Schulman D.
      • Sidner C.
      Automated interventions for multiple health behaviors using conversational agents.
      The diet-alone intervention was effective in increasing fruit and vegetable intake; there was also supportive evidence for the physical activity intervention improving physical activity. However, it was inconclusive (as it was a relatively small study) whether the combined intervention improved either fruit and vegetable intake or physical activity.
      Minimal reductions in weight (–0.59 kg, 95% CI=–1.02, –0.16) and BMI (–0.27 points, 95% CI=–0.46, –0.07) were found, compared with control conditions. Small reductions were also found in systolic blood pressure (SMD=–0.11, 95% CI=–0.19, –0.04); diastolic blood pressure (SMD=–0.10, 95% CI=–0.16, –0.04); total cholesterol (SMD=–0.17, 95% CI=–0.27, –0.06); high-density lipoprotein cholesterol (SMD=–0.13, 95% CI=–0.31, 0.05); and low-density lipoprotein cholesterol (SMD=–0.17, 95% CI=–0.34, 0.00).
      Data on intermediate outcomes were very limited. The most commonly reported (eight studies) outcome was self-efficacy, where there was no evidence for improvement (SMD=–0.06, 95% CI=–0.17, 0.06). Table 3 provides further details on secondary and intermediate outcomes.
      A range of potential moderators of effectiveness were examined: intervention characteristics (e.g., follow-up time, intervention characteristics [content], sequential or simultaneous targeting of risk behaviors); contextual factors (e.g., setting, geographic location, significant external events occurring at time of intervention); and participant characteristics (e.g., ethnicity, income).
      Length of follow-up was a statistically significant predictor for meeting recommendations for fruit and vegetable intake, explaining all heterogeneity. Longer follow-up was associated with reduced effectiveness compared with post-intervention follow-up (<6-month follow-up: slope=1.68, 95% CI=1.31, 2.17, p=0.002); 6 to 12–month follow-up: slope=1.54, 95% CI=1.26, 1.97, p=0.005). Length of follow-up was not associated with any other outcome.
      Interventions including education, training, and enablement intervention content (slope=0.22, 95% CI=0.07, 0.38, adjusted r2=70.73%, adjusted p=0.015) and duration of intervention (slope=0.21, 95% CI=0.06, 0.36, adjusted r2=64.23%, adjusted p=0.009) were associated with increased physical activity.
      Enablement was associated with a reduced risk of smoking (slope=0.62, 95% CI=0.47, 0.81, adjusted p=0.007), and longer duration of intervention was associated with less effectiveness in reducing risk of smoking (slope=1.53, 95% CI=1.71, 2.01, adjusted p=0.001). Together, these factors explained 79.33% of heterogeneity.
      All studies examined simultaneous change of risk behaviors. Three studies
      • Vandelanotte C.
      • Reeves M.M.
      • Brug J.
      • et al.
      A randomized trial of sequential and simultaneous multiple behavior change interventions for physical activity and fat intake.
      • Schulz D.N.
      • Kremers S.P.
      • Vandelanotte C.
      • et al.
      Effects of a web-based tailored multiple-lifestyle intervention for adults: a two-year randomized controlled trial comparing sequential and simultaneous delivery modes.
      • King A.C.
      • Castro C.M.
      • Buman M.P.
      • Hekler E.B.
      • Urizar Jr, G.G.
      • Ahn D.K.
      Behavioral impacts of sequentially versus simultaneously delivered dietary plus physical activity interventions: the CALM Trial.
      compared simultaneous change with sequential change of risk behaviors and did not find statistically significant differences between interventions that aimed to changed diet and physical activity simultaneously and those that changed diet and physical activity sequentially. However, one study found that sequential interventions were more likely than simultaneous interventions to be effective in promoting smoking cessation (OR=1.51, p=0.004).
      • Schulz D.N.
      • Kremers S.P.
      • Vandelanotte C.
      • et al.
      Effects of a web-based tailored multiple-lifestyle intervention for adults: a two-year randomized controlled trial comparing sequential and simultaneous delivery modes.
      There was insufficient evidence to determine whether contextual factors had any impact on effectiveness for any outcomes.
      Overall, there were insufficient data to conclude whether effectiveness differs between lower- and higher-income groups, or between black and minority ethnic groups and majority ethnic groups (Table 2).
      Comparisons of univariate analyses with the multivariate analyses generally did not reveal substantial differences (Appendix, available online). The exception was the multivariate meta-analysis on smoking, meeting recommendations for fruit and vegetable intake, and meeting recommendations for physical activity. There was statistically significant evidence of improvement in fruit and vegetable intake in the univariate analyses (OR=0.62, 95% CI=0.51, 0.84, p<0.0001). However, in the multivariate meta-analysis, effectiveness in improving fruit and vegetable intake reduced substantially (OR=0.84, 95% CI=0.68, 1.03, p=0.09). This appears to be explained by a strong negative correlation between changes in smoking and fruit and vegetable intake (r=–0.95). In addition, changes in smoking behavior were negatively associated with improvements in physical activity although less strongly (r=–0.44).
      Moderate-sized correlations were found between all of the other behaviors included in the multivariate meta-analyses. Improvements in fruit and vegetable intake (r=0.53); calorie intake (r=0.56); fat intake (r=0.52); and physical activity (r=0.52) were all associated with weight loss in a similar magnitude, suggesting that all are important strategies for reducing weight.
      A stronger association was found between improvements in fruit and vegetable intake (r=0.56) and changes to total cholesterol than fat intake (r=0.41). Conversely, improvements in fat intake (systolic blood pressure, r=0.63; diastolic blood pressure, r=0.43) appeared to be more strongly associated with improvements in both systolic blood pressure and diastolic blood pressure than was fruit and vegetable intake (systolic blood pressure, r=0.52; diastolic blood pressure, r=0.23).
      Increased physical activity was strongly associated with changes to total cholesterol (r=0.87) and moderately associated with changes in systolic blood pressure (r=0.39), but there was no association with changes to diastolic blood pressure (r=0.05).

      Discussion

      A systematic review was conducted assessing effects of multiple risk behavior interventions in general adult populations. Studies specifically targeting at-risk populations, including those at risk of cardiovascular disease or who are obese, were excluded. Sixty-nine RCTs were included with a total of 73,873 participants. Diet and physical activity were most frequently targeted and interventions consisted mainly of education combined with skills training. All 69 trials examined the simultaneous change of behaviors, and three
      • Vandelanotte C.
      • Reeves M.M.
      • Brug J.
      • et al.
      A randomized trial of sequential and simultaneous multiple behavior change interventions for physical activity and fat intake.
      • Schulz D.N.
      • Kremers S.P.
      • Vandelanotte C.
      • et al.
      Effects of a web-based tailored multiple-lifestyle intervention for adults: a two-year randomized controlled trial comparing sequential and simultaneous delivery modes.
      • King A.C.
      • Castro C.M.
      • Buman M.P.
      • Hekler E.B.
      • Urizar Jr, G.G.
      • Ahn D.K.
      Behavioral impacts of sequentially versus simultaneously delivered dietary plus physical activity interventions: the CALM Trial.
      compared simultaneous with sequential change. Overall, small improvements in diet (e.g., fruit and vegetable, fat, and calorie intake); physical activity; and smoking were found, but effects diminished over time for fruit and vegetable intake. Multivariate analyses suggested weight loss was equally associated with improvements in fruit and vegetable intake, fat intake, calorie intake, and physical activity.
      Reductions in smoking were negatively associated with improvements in fruit and vegetable intake and physical activity. This is consistent with the finding that interventions that targeted smoking and other risk behavior sequentially are more effective than those that seek simultaneous change.
      • Schulz D.N.
      • Kremers S.P.
      • Vandelanotte C.
      • et al.
      Effects of a web-based tailored multiple-lifestyle intervention for adults: a two-year randomized controlled trial comparing sequential and simultaneous delivery modes.
      By contrast, no statistically significant differences were found in the three studies that compared sequential and simultaneous change of diet and physical activity.
      • Vandelanotte C.
      • Reeves M.M.
      • Brug J.
      • et al.
      A randomized trial of sequential and simultaneous multiple behavior change interventions for physical activity and fat intake.
      • Schulz D.N.
      • Kremers S.P.
      • Vandelanotte C.
      • et al.
      Effects of a web-based tailored multiple-lifestyle intervention for adults: a two-year randomized controlled trial comparing sequential and simultaneous delivery modes.
      • King A.C.
      • Castro C.M.
      • Buman M.P.
      • Hekler E.B.
      • Urizar Jr, G.G.
      • Ahn D.K.
      Behavioral impacts of sequentially versus simultaneously delivered dietary plus physical activity interventions: the CALM Trial.
      Most interventions were based on a Social Cognitive Theory approach, but intermediate outcomes were reported infrequently, which makes it difficult to assess the theoretic assumptions of these interventions. Self-efficacy is a key component of Social Cognitive Theory and was the most commonly reported intermediate outcome. In studies that reported this outcome, interventions did not appear to be effective in improving self-efficacy. More consistent reporting of intermediate outcomes is needed to comprehensively evaluate the effectiveness of multiple risk behavior interventions and to examine the validity of their theoretic assumptions.
      This systematic review adds to knowledge of multiple risk behavior change by providing a comprehensive evaluation of non-pharmacologic interventions targeting two or more risk behaviors in non-clinical adult populations. An earlier Cochrane review on multiple risk factor reduction assessed distal outcomes such as mortality and fatal and non-fatal coronary heart disease and found limited evidence of benefit from education and counseling interventions on these outcomes.
      • Ebrahim S.
      • Taylor F.
      • Ward K.
      • et al.
      Multiple risk factor interventions for primary prevention of coronary heart disease.
      Similarly, a recent review of non-pharmacologic multiple risk behavior interventions delivered in the workplace found small benefits in diet, physical activity, and smoking.
      • Bhattarai N.
      • Prevost A.T.
      • Wright A.J.
      • et al.
      Effectiveness of interventions to promote healthy diet in primary care: systematic review and meta-analysis of randomised controlled trials.
      • Osilla K.
      • van Busum K.
      • Schnyer C.
      • et al.
      Systematic review of the impact of worksite wellness programs.
      However, the review did not distinguish between studies targeting multiple and single behaviors.
      It was not possible to compare relative effectiveness of multiple and single risk behavior interventions as only two studies addressed this question. However, other systematic reviews have evaluated the effects of similar (non-pharmacologic) interventions on individual behaviors. Overall, the findings are comparable to those from this review. For example, interventions to improve diet in general populations increased servings of fruit and vegetables by a similar amount to the interventions included in this review (0.50 vs 0.31 more servings).
      • Bhattarai N.
      • Prevost A.T.
      • Wright A.J.
      • et al.
      Effectiveness of interventions to promote healthy diet in primary care: systematic review and meta-analysis of randomised controlled trials.
      • Johnson B.
      • Kanters S.
      • Bandayrel K.
      • et al.
      Comparison of weight loss among named diet programs in overweight and obese adults: a meta-analysis.
      Reviews focusing on physical activity
      • Hobbs N.
      • Godfrey A.
      • Lara J.
      • et al.
      Are behavioral interventions effective in increasing physical activity at 12 to 36 months in adults aged 55 to 70 years? A systematic review and meta-analysis.
      • Orrow G.
      • Kinmonth A.
      • Sanderson S.
      • et al.
      Effectiveness of physical activity promotion based in primary care: systematic review and meta-analysis of randomised controlled trials.
      • Cahill K.
      • Moher M.
      • Lancaster T.
      Workplace interventions for smoking cessation.
      reported small improvements (~SMD=0.20) similar to those found in this review (SMD=0.25).
      The large number of studies and consistency of findings argues against further trials focusing on the use of education and skills training to target risky behaviors. Similarly, the large number of trials focusing on simultaneous change of multiple behaviors suggests no further evidence is needed. By contrast, a key evidence gap relates to the sequencing of intervention components. Only three studies examined sequential change, and therefore findings are inconclusive. Evidence is lacking on how various intervention components might be ordered to maximize impacts on risk behaviors. Understanding how people approach behavior change, especially when multiple behaviors are involved, is important. A United Kingdom–based qualitative study found that people differ in their strategies for change, with some preferring to make changes simultaneously, viewing each behavior as part of a healthier lifestyle and others sequentially, seeing behaviors as discrete and easier to change when broken down into manageable chunks.
      • Koshy P.
      • Mackenzie M.
      • Leslie W.
      • et al.
      Eating the elephant whole or in slices: views of participants in a smoking cessation intervention trial on multiple behaviour changes as sequential or concurrent tasks..
      The present review indicates that interventions comprising education and skills training are associated with modest reductions in risk behaviors. At best, these interventions achieve small changes that may not translate into meaningful reductions in risk of mortality and cardiovascular disease–related mortality.
      • Vandelanotte C.
      • Reeves M.M.
      • Brug J.
      • et al.
      A randomized trial of sequential and simultaneous multiple behavior change interventions for physical activity and fat intake.
      Although information and skills are important, they should be considered alongside other factors that influence behavior. Lack of social support; cost of adopting healthy behaviors; balancing health behaviors with everyday life (e.g., routines, time management); cultural preferences; and environmental barriers are likely to be equally important.
      • Murray J.
      • Fenton G.
      • Honey S.
      • et al.
      A qualitative synthesis of factors influencing maintenance of lifestyle behaviour change in individuals with high cardiovascular risk.
      Individuals are influenced not only by their motivation and capability to make behavioral changes but also by opportunities afforded by the social and physical environment.
      • Michie S.
      • van Stralen M.M.
      • West R.
      The Behaviour Change Wheel: a new method for characterising and designing behaviour change interventions.
      The impact of the physical and social environment on behavior is increasingly recognized, and advocates of the Social Ecological approach
      • Sallis J.F.
      • Owen N.
      • Fisher E.B.
      Ecological models of health behavior.
      argue that risk behaviors need to be understood within the context of social and physical environmental factors. These include the home and workplace as well as broader societal factors such as income inequality that impact on individuals and groups.
      • Schneider M.
      • Stokols D.
      Multilevel theories of behavior change: a social ecological framework.
      However, the present systematic review identified few studies that incorporated environmental changes as part of the intervention package; where included, the focus was on the social rather than the physical environment. Despite the lack of evidence in support of environmental restructuring for changing health behaviors, findings from field and laboratory experiments suggest that human behavior is prompted by cues in the environment,
      • Lorenc T.
      • Petticrew M.
      • Welch V.
      • et al.
      What types of interventions generate inequalities? Evidence from systematic reviews.
      • Marteau T.M.
      • Hollands G.J.
      • Fletcher P.C.
      Changing human behavior to prevent disease: the importance of targeting automatic processes.
      and such approaches have been explored extensively in the discipline of environmental psychology. This promising approach to large-scale behavior change requires thorough evaluation through good-quality observational studies and RCTs where feasible.

      Limitations

      Strengths of this review include comprehensive and rigorous searching and the mapping exercise to determine inclusion criteria.
      • King K.
      • Meader N.
      • Wright K.
      • et al.
      Characteristics of interventions targeting multiple lifestyle risk behaviours in adult populations: a systematic scoping review.
      It was assessed whether restricting to RCTs would limit the type of interventions eligible for inclusion and found this was unlikely to be the case. This is a particular issue with reviews of public health interventions and has been referred to as an “inverse evidence law” whereby least is known about the effects of interventions most likely to influence whole populations because they tend to be evaluated using less rigorous methods.
      • Ogilvie D.
      • Egan M.
      • Hamilton V.
      • et al.
      Systematic reviews of health effects of social interventions: 2. Best available evidence: how low should you go?.
      Other strengths include use of the Behaviour Change Wheel to classify intervention components according to a standard set of functions. This enables identification of “active ingredients” across interventions and studies.
      Limitations include the variable quality of the RCTs. Slightly more than half of the studies had a high risk of bias for at least one of the assessed domains. Studies varied in the way they measured behaviors, particularly physical activity and alcohol intake, which made comparisons difficult. Reporting of intermediate outcomes such as self-efficacy, attitudes, and knowledge was limited and, importantly, few studies provided contextual information, for example, about important external events occurring at the time of the intervention.
      Few studies analyzed their results by subgroup. This is an important evidence gap: public health interventions, particularly those focusing on “downstream” interventions such as education and skills training, have the potential to increase health inequalities by disproportionately benefiting more-advantaged groups.
      • Lorenc T.
      • Petticrew M.
      • Welch V.
      • et al.
      What types of interventions generate inequalities? Evidence from systematic reviews.
      • Liu J.
      • Davidson E.
      • Bhopal R.
      • et al.
      Adapting health promotion interventions to meet the needs of ethnic minority groups: mixed methods evidence synthesis.
      Although most studies reported data on income, occupation, education, ethnicity, and gender, and a few specifically targeted low-income
      • Hillier F.C.
      • Batterham A.M.
      • Nixon C.A.
      • et al.
      A community-based health promotion intervention using brief negotiation techniques and a pledge on dietary intake, physical activity levels and weight outcomes: lessons learnt from an exploratory trial.
      • Burke L.
      • Lee A.
      • Jancey J.
      • et al.
      Physical activity and nutrition behavioural outcomes of home-based intervention program for seniors: a randomized controlled trial.
      • Staten L.K.
      • Gregory-Mercado K.Y.
      • Ranger-Moore J.
      • et al.
      Provider counseling, health education, and community health workers: the Arizona WISEWOMAN project.
      • Jackson R.
      • Stotland N.
      • Caughey A.
      • et al.
      Improving diet and exercise in pregnancy with Video Doctor counseling: a randomized trial.
      • Keyserling T.C.
      • Samuel Hodge C.D.
      • Jilcott S.B.
      • et al.
      Randomized trial of a clinic-based, community-supported, lifestyle intervention to improve physical activity and diet: the North Carolina enhanced WISEWOMAN project.
      • Weisman C.S.
      • Hillemeier M.M.
      • Symons Downs D.
      • et al.
      Improving women’s preconceptional health: long-term effects of the Strong Healthy Women Behavior Change Intervention in the Central Pennsylvania Women’s Health Study.
      • Peragallo N.
      • Gonzalez-Guarda R.M.
      • McCabe B.E.
      • et al.
      The efficacy of an HIV risk reduction intervention for Hispanic women.
      • Phillips G.
      • Bottomley C.
      • Schmidt E.
      • et al.
      Well London Phase-1: results among adults of a cluster-randomised trial of a community engagement approach to improving health behaviours and mental well-being in deprived inner-city neighbourhoods.
      or black and ethnic minority groups,
      • Yanek L.
      • Becker D.
      • Moy T.
      • et al.
      Project Joy: faith based cardiovascular health promotion for African American women.
      • Wilcox S.
      • Parrott A.
      • Baruth M.
      • et al.
      The Faith, Activity, and Nutrition Program: a randomized controlled trial in African-American churches.
      • Peragallo N.
      • Gonzalez-Guarda R.M.
      • McCabe B.E.
      • et al.
      The efficacy of an HIV risk reduction intervention for Hispanic women.
      • Campbell M.K.
      • James A.
      • Hudson M.A.
      • et al.
      Improving multiple behaviors for colorectal cancer prevention among African American church members.
      • Resnicow K.
      • Jackson A.
      • Blissett D.
      • et al.
      Results of the Healthy Body Healthy Spirit Trial.
      it was not possible to explore equity effects in a meaningful way.
      To ensure population homogeneity, this review focused on non-clinical adult populations, which means that a number of studies targeting specific at-risk populations, such as those who are obese or at high risk of cardiovascular disease, were excluded. Further systematic reviews are needed to address the effectiveness of multiple risk behavior interventions in these populations.

      Conclusions

      This is the first systematic review to provide overall estimates of the impact of non-pharmacologic interventions on multiple lifestyle risk behaviors in non-clinical, adult populations. Interventions, mainly consisting of education and skills training, targeting multiple risk behaviors resulted in small improvements in diet (e.g., fruit and vegetable intake) and physical activity and smoking. Such approaches result, at best, in small reductions in risk behaviors, which fail to translate into meaningful reduction in risk of overall mortality and cardiovascular disease–related mortality.

      Acknowledgments

      This study was funded by the Department of Health Policy Research Programme as part of the Public Health Research Consortium. The funder had no role in the design, management, data collection, analyses, or interpretation of the data or in the writing of the manuscript or the decision to submit for publication. We also thank Claire Khouja for assisting with data extraction and quality assessment.
      No financial disclosures were reported by authors of this paper.

      Appendix A. Supplementary material

      References

        • Strong K.
        • Mathers C.
        • Leeder S.
        • Beaglehole R.
        Preventing chronic diseases: how many lives can we save?.
        Lancet. 2005; 366: 1578-1582https://doi.org/10.1016/S0140-6736(05)67341-2
      1. WHO. World health statistics in 2012. www.who.int/gho/publications/world_health_statistics/2012.

        • Schuit A.J.
        • van Loon A.J.
        • Tijhuis M.
        • Ocké M.
        Clustering of lifestyle risk factors in a general adult population.
        Prev Med. 2002; 35: 219-224https://doi.org/10.1006/pmed.2002.1064
        • Chou K.-L.
        The prevalence and clustering of four major lifestyle risk factors in Hong Kong Chinese older adults.
        J Aging Health. 2008; 20: 788-803https://doi.org/10.1177/0898264308321082
        • Baruth M.
        • Addy C.
        • Wilcox S.
        • et al.
        Clustering of risk behaviours among African American adults.
        Health Educ J. 2011; 71: 565-575https://doi.org/10.1177/0017896911411761
        • Myint P.K.
        • Luben R.N.
        • Wareham N.J.
        • Bingham S.A.
        • Khaw K.
        Combined effect of health behaviours and risk of first ever stroke in 20,040 men and women over 11 years’ follow-up in Norfolk cohort of European Prospective Investigation of Cancer (EPIC Norfolk): prospective population study.
        BMJ. 2009; 338: b349https://doi.org/10.1136/bmj.b349
        • Poortinga W.
        The prevalence and clustering of four major lifestyle risk factors in an English adult population.
        Prev Med. 2007; 44: 124-128https://doi.org/10.1016/j.ypmed.2006.10.006
        • Lawder R.
        • Harding O.
        • Stockton D.
        • et al.
        Is the Scottish population living dangerously? Prevalence of multiple risk factors: the Scottish Health Survey 2003.
        BMC Public Health. 2010; 10: 330https://doi.org/10.1186/1471-2458-10-330
        • Buck D.
        • Frosini F.
        Clustering of Unhealthy Behaviours Over Time: Implications for Policy and Practice.
        The King’s Fund, London2012
        • Lim S.S.
        • Vos T.
        • Flaxman A.D.
        • et al.
        A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010.
        Lancet. 2012; 380: 2224-2260https://doi.org/10.1016/S0140-6736(12)61766-8
        • King K.
        • Meader N.
        • Wright K.
        • et al.
        Characteristics of interventions targeting multiple lifestyle risk behaviours in adult populations: a systematic scoping review.
        PLoS One. 2015; 10: e0117015https://doi.org/10.1371/journal.pone.0117015
        • MacArthur G.
        • Kipping R.
        • White J.
        • et al.
        Individual-, family-, and school-level interventions for preventing multiple risk behaviours in individuals aged 8 to 25 years.
        Cochrane Database Syst Rev. 2012; 6: CD009927https://doi.org/10.1002/14651858.cd009927
        • Wright K.
        • Golder S.
        • Rodriguez-Lopez R.
        Citation searching: a systematic review case study of multiple risk behaviour interventions.
        BMC Med Res Methodol. 2014; 14: 73https://doi.org/10.1186/1471-2288-14-73
        • Michie S.
        • van Stralen M.M.
        • West R.
        The Behaviour Change Wheel: a new method for characterising and designing behaviour change interventions.
        Implement Sci. 2011; 6: 42https://doi.org/10.1186/1748-5908-6-42
        • Montgomery P.
        • Underhill K.
        • Gardner F.
        • et al.
        The Oxford Implementation Index: a new tool for incorporating implementation data into systematic reviews and meta-analyses.
        J Clin Epidemiol. 2013; 66: 874-882https://doi.org/10.1016/j.jclinepi.2013.03.006
        • Higgins J.P.T.
        • Green S.
        Cochrane Handbook for Systematic Reviews of Interventions Version 5.
        Cochrane Collaboration, Oxford2011
        • Hedges L.V.
        Distribution theory for Glass’s estimator of effect size and related estimators.
        J Ed Behav Stats. 1981; 6: 107-128https://doi.org/10.3102/10769986006002107
        • Higgins J.
        • Thompson S.G.
        Quantifying heterogeneity in a meta-analysis.
        Stat Med. 2002; 21: 1539-1558https://doi.org/10.1002/sim.1186
        • Higgins J.
        • Thompson S.
        Controlling the risk of spurious findings from meta-regression.
        Stat Med. 2004; 23: 1663-1682https://doi.org/10.1002/sim.1752
        • Bandura A.
        Social Foundations of Thought and Action: A Social Cognitive Theory.
        Prentice Hall, Englewood Cliffs, NJ1986
        • Janz N.K.
        • Becker M.H.
        The Health Belief Model: a decade later.
        Health Educ Q. 1984; 11: 1-47https://doi.org/10.1177/109019818401100101
        • Fishbein M.
        • Ajzen I.
        Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research.
        Addison-Wesley, Reading, MA1975
        • Imperial Cancer Research Fund OXCHECK Study Group
        Effectiveness of health checks conducted by nurses in primary care: final results of the OXCHECK study.
        BMJ. 1995; 310: 1099-1104https://doi.org/10.1136/bmj.310.6987.1099
        • McCambridge J.
        • Hunt C.
        • Jenkins R.J.
        • et al.
        Cluster randomised trial of the effectiveness of motivational interviewing for universal prevention.
        Drug Alcohol Depend. 2011; 114: 177-184https://doi.org/10.1016/j.drugalcdep.2010.07.02
        • Yanek L.
        • Becker D.
        • Moy T.
        • et al.
        Project Joy: faith based cardiovascular health promotion for African American women.
        Public Health Rep. 2001; 116: 68-81https://doi.org/10.1093/phr/116.S1.68
        • Hillier F.C.
        • Batterham A.M.
        • Nixon C.A.
        • et al.
        A community-based health promotion intervention using brief negotiation techniques and a pledge on dietary intake, physical activity levels and weight outcomes: lessons learnt from an exploratory trial.
        Public Health Nutr. 2012; 15: 1446-1455https://doi.org/10.1017/S1368980011002862
        • Burke L.
        • Jancey J.M.
        • Howat P.
        • et al.
        Physical Activity and Nutrition Program for Seniors (PANS): process evaluation.
        Health Promot Pract. 2013; 14: 543-551https://doi.org/10.1177/1524839912461504
        • de Vries H.
        • Kremers S.P.J.
        • Smeets T.
        • et al.
        The effectiveness of tailored feedback and action plans in an intervention addressing multiple health behaviors.
        Am J Health Promot. 2008; 22: 417-425https://doi.org/10.4278/ajhp.22.6.417
        • Lee A.
        • Jancey J.
        • Howat P.
        • et al.
        Effectiveness of a home-based postal and telephone physical activity and nutrition pilot program for seniors.
        J Obes. 2011; 2011: 786827https://doi.org/10.1155/2011/786827
        • van Assema P.
        • Steenbakkers M.
        • Kok G.
        • et al.
        Results of the Dutch Community Project “Healthy Bergeyk.”.
        Prev Med. 1994; 23: 394-401https://doi.org/10.1006/pmed.1994.1054
        • van Keulen H.M.
        • Mesters I.
        • Ausems M.
        • et al.
        Tailored print communication and telephone motivational interviewing are equally successful in improving multiple lifestyle behaviors in a randomized controlled trial.
        Ann Behav Med. 2011; 41: 104-118https://doi.org/10.1007/s12160-010-9231-3
        • Werch C.E.
        • Moore M.J.
        • Bian H.
        • et al.
        Are effects from a brief multiple behavior intervention for college students sustained over time?.
        Prev Med. 2010; 50: 30-34https://doi.org/10.1016/j.ypmed.2009.12.010
        • Wilcox S.
        • Parrott A.
        • Baruth M.
        • et al.
        The Faith, Activity, and Nutrition Program: a randomized controlled trial in African-American churches.
        Am J Prev Med. 2013; 44: 122-131https://doi.org/10.1016/j.amepre.2012.09.062
        • Sikkema K.J.
        • Winett R.A.
        • Lombard D.N.
        Development and evaluation of an HIV-risk reduction program for female college students.
        AIDS Educ Prev. 1995; 7: 145-159
        • Ussher M.
        • West R.
        • McEwen A.
        • et al.
        Efficacy of exercise counselling as an aid for smoking cessation: a randomized controlled trial.
        Addiction. 2003; 98: 523-532https://doi.org/10.1046/j.1360-0443.2003.00346.x
        • Werch C.
        • Bian H.
        • Moore M.J.
        • et al.
        Brief multiple behavior interventions in a college student health care clinic.
        J Adolesc Health. 2007; 41: 577-585https://doi.org/10.1016/j.jadohealth.2007.06.003
        • Leslie W.S.
        • Koshy P.R.
        • Mackenzie M.
        • et al.
        Changes in body weight and food choice in those attempting smoking cessation: a cluster randomised controlled trial.
        BMC Public Health. 2012; 12: 389https://doi.org/10.1186/1471-2458-12-389
        • Foroushani A.R.
        • Estebsari F.
        • Mostafaei D.
        • et al.
        The effect of health promoting intervention on healthy lifestyle and social support in elders: a clinical trial study.
        Iran Red Crescent Med J. 2014; 16: e18399
        • Bickmore T.W.
        • Schulman D.
        • Sidner C.
        Automated interventions for multiple health behaviors using conversational agents.
        Patient Educ Couns. 2013; 92: 142-148https://doi.org/10.1016/j.pec.2013.05.011
        • Vandelanotte C.
        • Reeves M.M.
        • Brug J.
        • et al.
        A randomized trial of sequential and simultaneous multiple behavior change interventions for physical activity and fat intake.
        Prev Med. 2008; 46: 232-237https://doi.org/10.1016/j.ypmed.2007.07.008
        • Schulz D.N.
        • Kremers S.P.
        • Vandelanotte C.
        • et al.
        Effects of a web-based tailored multiple-lifestyle intervention for adults: a two-year randomized controlled trial comparing sequential and simultaneous delivery modes.
        J Med Internet Res. 2014; : e26https://doi.org/10.2196/jmir.3094
        • King A.C.
        • Castro C.M.
        • Buman M.P.
        • Hekler E.B.
        • Urizar Jr, G.G.
        • Ahn D.K.
        Behavioral impacts of sequentially versus simultaneously delivered dietary plus physical activity interventions: the CALM Trial.
        Ann Behav Med. 2013; 46: 157-168https://doi.org/10.1007/s12160-013-9501-y
        • Ebrahim S.
        • Taylor F.
        • Ward K.
        • et al.
        Multiple risk factor interventions for primary prevention of coronary heart disease.
        Cochrane Database Syst Rev. 2011; 1: CD001561https://doi.org/10.1002/14651858.cd001561.pub3
        • Bhattarai N.
        • Prevost A.T.
        • Wright A.J.
        • et al.
        Effectiveness of interventions to promote healthy diet in primary care: systematic review and meta-analysis of randomised controlled trials.
        BMC Public Health. 2013; 13: 1203https://doi.org/10.1186/1471-2458-13-1203
        • Osilla K.
        • van Busum K.
        • Schnyer C.
        • et al.
        Systematic review of the impact of worksite wellness programs.
        Am J Manag Care. 2012; 18: e68-e81
        • Johnson B.
        • Kanters S.
        • Bandayrel K.
        • et al.
        Comparison of weight loss among named diet programs in overweight and obese adults: a meta-analysis.
        JAMA. 2014; 312: 923-933https://doi.org/10.1001/jama.2014.10397
        • Hobbs N.
        • Godfrey A.
        • Lara J.
        • et al.
        Are behavioral interventions effective in increasing physical activity at 12 to 36 months in adults aged 55 to 70 years? A systematic review and meta-analysis.
        BMC Med. 2013; 11: 75https://doi.org/10.1186/1741-7015-11-75
        • Orrow G.
        • Kinmonth A.
        • Sanderson S.
        • et al.
        Effectiveness of physical activity promotion based in primary care: systematic review and meta-analysis of randomised controlled trials.
        BMJ. 2012; 344: e1389https://doi.org/10.1136/bmj.e1389
        • Cahill K.
        • Moher M.
        • Lancaster T.
        Workplace interventions for smoking cessation.
        Cochrane Database Syst Rev. 2014; 2: CD003440https://doi.org/10.1002/14651858.cd003440.pub4
        • Koshy P.
        • Mackenzie M.
        • Leslie W.
        • et al.
        Eating the elephant whole or in slices: views of participants in a smoking cessation intervention trial on multiple behaviour changes as sequential or concurrent tasks..
        BMC Public Health. 2012; 12: 500https://doi.org/10.1186/1471-2458-12-500
        • Murray J.
        • Fenton G.
        • Honey S.
        • et al.
        A qualitative synthesis of factors influencing maintenance of lifestyle behaviour change in individuals with high cardiovascular risk.
        BMC Cardiovasc Disord. 2013; 13: 48https://doi.org/10.1186/1471-2261-13-48
        • Sallis J.F.
        • Owen N.
        • Fisher E.B.
        Ecological models of health behavior.
        in: Glanz K. Rimer B.K. Viswanath K. Health Behavior and Health Education: Theory, Research and Practice. Jossey-Bass, San Francisco, CA2008
        • Schneider M.
        • Stokols D.
        Multilevel theories of behavior change: a social ecological framework.
        in: Shumaker S.A. Ockene J.K. Riekert K.A. The Handbook of Health Behavior Change. 3rd ed. Springer, New York, NY2009: 85-106
        • Lorenc T.
        • Petticrew M.
        • Welch V.
        • et al.
        What types of interventions generate inequalities? Evidence from systematic reviews.
        J Epidemiol Community Health. 2013; 67: 190-193https://doi.org/10.1136/jech-2012-201257
        • Marteau T.M.
        • Hollands G.J.
        • Fletcher P.C.
        Changing human behavior to prevent disease: the importance of targeting automatic processes.
        Science. 2012; 337: 1492-1495https://doi.org/10.1126/science.1226918
        • Ogilvie D.
        • Egan M.
        • Hamilton V.
        • et al.
        Systematic reviews of health effects of social interventions: 2. Best available evidence: how low should you go?.
        J Epidemiol Community Health. 2005; 59: 886-892https://doi.org/10.1136/jech.2005.034199
        • Liu J.
        • Davidson E.
        • Bhopal R.
        • et al.
        Adapting health promotion interventions to meet the needs of ethnic minority groups: mixed methods evidence synthesis.
        Health Technol Assess. 2012; 16: 1-469https://doi.org/10.3310/hta16440
        • Burke L.
        • Lee A.
        • Jancey J.
        • et al.
        Physical activity and nutrition behavioural outcomes of home-based intervention program for seniors: a randomized controlled trial.
        Int J Behav Nutr Phys Act. 2013; 10: 14https://doi.org/10.1186/1479-5868-10-14
        • Staten L.K.
        • Gregory-Mercado K.Y.
        • Ranger-Moore J.
        • et al.
        Provider counseling, health education, and community health workers: the Arizona WISEWOMAN project.
        J Womens Health. 2004; 13: 547-556https://doi.org/10.1089/1540999041281133
        • Jackson R.
        • Stotland N.
        • Caughey A.
        • et al.
        Improving diet and exercise in pregnancy with Video Doctor counseling: a randomized trial.
        Patient Educ Couns. 2011; 83: 203-209https://doi.org/10.1016/j.pec.2010.05.019
        • Keyserling T.C.
        • Samuel Hodge C.D.
        • Jilcott S.B.
        • et al.
        Randomized trial of a clinic-based, community-supported, lifestyle intervention to improve physical activity and diet: the North Carolina enhanced WISEWOMAN project.
        Prev Med. 2008; 46: 499-510https://doi.org/10.1016/j.ypmed.2008.02.011
        • Weisman C.S.
        • Hillemeier M.M.
        • Symons Downs D.
        • et al.
        Improving women’s preconceptional health: long-term effects of the Strong Healthy Women Behavior Change Intervention in the Central Pennsylvania Women’s Health Study.
        Womens Health Issues. 2011; 21: 265-271https://doi.org/10.1016/j.whi.2011.03.007
        • Peragallo N.
        • Gonzalez-Guarda R.M.
        • McCabe B.E.
        • et al.
        The efficacy of an HIV risk reduction intervention for Hispanic women.
        AIDS Behav. 2012; 16: 1316-1326https://doi.org/10.1007/s10461-011-0052-6
        • Phillips G.
        • Bottomley C.
        • Schmidt E.
        • et al.
        Well London Phase-1: results among adults of a cluster-randomised trial of a community engagement approach to improving health behaviours and mental well-being in deprived inner-city neighbourhoods.
        J Epidemiol Community Health. 2014; 68: 606-614https://doi.org/10.1136/jech-2013-202505
        • Campbell M.K.
        • James A.
        • Hudson M.A.
        • et al.
        Improving multiple behaviors for colorectal cancer prevention among African American church members.
        Health Psychol. 2004; 23: 492-502https://doi.org/10.1037/0278-6133.23.5.492
        • Resnicow K.
        • Jackson A.
        • Blissett D.
        • et al.
        Results of the Healthy Body Healthy Spirit Trial.
        Health Psychol. 2005; 24: 339-348https://doi.org/10.1037/0278-6133.24.4.339