UBC Faculty Research and Publications

Validity of the Stages of Change in Steps instrument (SoC-Step) for achieving the physical activity goal… Rosenkranz, Richard R; Duncan, Mitch J; Caperchione, Cristina M; Kolt, Gregory S; Vandelanotte, Corneel; Maeder, Anthony J; Savage, Trevor N; Mummery, W. K Nov 30, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12889_2015_Article_2539.pdf [ 550.21kB ]
JSON: 52383-1.0307461.json
JSON-LD: 52383-1.0307461-ld.json
RDF/XML (Pretty): 52383-1.0307461-rdf.xml
RDF/JSON: 52383-1.0307461-rdf.json
Turtle: 52383-1.0307461-turtle.txt
N-Triples: 52383-1.0307461-rdf-ntriples.txt
Original Record: 52383-1.0307461-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessValidity of the Stages of Change in Stepsinstrument (SoC-Step) for achieving thephysical activity goal of 10,000 steps perdayRichard R. Rosenkranz1,2*, Mitch J. Duncan3, Cristina M. Caperchione4, Gregory S. Kolt2, Corneel Vandelanotte5,Anthony J. Maeder2, Trevor N. Savage2 and W. Kerry Mummery6AbstractBackground: Physical activity (PA) offers numerous benefits to health and well-being, but most adults are notsufficiently physically active to afford such benefits. The 10,000 steps campaign has been a popular and effectiveapproach to promote PA. The Transtheoretical Model posits that individuals have varying levels of readiness forhealth behavior change, known as Stages of Change (Precontemplation, Contemplation, Preparation, Action, andMaintenance). Few validated assessment instruments are available for determining Stages of Change in relation tothe PA goal of 10,000 steps per day. The purpose of this study was to assess the criterion-related validity of theSoC-Step, a brief 10,000 steps per day Stages of Change instrument.Methods: Participants were 504 Australian adults (176 males, 328 females, mean age = 50.8 ± 13.0 years) from thebaseline sample of the Walk 2.0 randomized controlled trial. Measures included 7-day accelerometry (ActigraphGT3X), height, weight, and self-reported intention, self-efficacy, and SoC-Step: Stages of Change relative toachieving 10,000 steps per day. Kruskal-Wallis H tests with pairwise comparisons were used to determine whetherparticipants differed by stage, according to steps per day, general health, body mass index, intention, and self-efficacy to achieve 10,000 steps per day. Binary logistic regression was used to test the hypothesis that participantsin Maintenance or Action stages would have greater likelihood of meeting the 10,000 steps goal, in comparison toparticipants in the other three stages.Results: Consistent with study hypotheses, participants in Precontemplation had significantly lower intention scoresthan those in Contemplation (p = 0.003) or Preparation (p < 0.001). Participants in Action or Maintenance stages weremore likely to achieve ≥10,000 steps per day (OR = 3.11; 95 % CI = 1.66,5.83) compared to those in Precontemplation,Contemplation, or Preparation. Intention (p < 0.001) and self-efficacy (p < 0.001) to achieve 10,000 steps daily differedby stage, and participants in the Maintenance stage had higher general health status and lower body mass index thanthose in Precontemplation, Contemplation and Preparation stages (p < 0.05).Conclusions: This brief SoC-Step instrument appears to have good criterion-related validity for determining Stages ofChange related to the public health goal of 10,000 steps per day.Trial registration: Australian New Zealand Clinical Trials Registry reference: ACTRN12611000157976 World HealthOrganization Universal Trial Number: U111-1119-1755Keywords: Motivation, Transtheoretical Model, Targeted intervention, Sedentary lifestyle, Walking, Pedometer* Correspondence: ricardo@ksu.edu1Kansas State University, Manhattan, USA2Western Sydney University, Sydney, AustraliaFull list of author information is available at the end of the article© 2015 Rosenkranz et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Rosenkranz et al. BMC Public Health  (2015) 15:1197 DOI 10.1186/s12889-015-2539-yBackgroundMaintaining a physically active lifestyle offers numerousbenefits to health and well-being [1]. A large proportionof adults, however, are not sufficiently physically activeto afford such benefits [2]. Although physical activityguidelines are typically presented in terms of minutesper week [1], communicating physical activity in termsof steps per day has been offered as an easily understoodstandard that can be assessed with pedometers or otherinexpensive physical activity monitors [3]. Pedometerscan be useful tools for motivation, goal setting, and self-monitoring; a systematic review of physical activity pro-motion studies employing pedometers showed that theiruse, within a broader behavior change intervention, wasassociated with increased physical activity, as well as im-provements in blood pressure and body mass index [4].Many public health interventions have employed pedom-eters or other step-counters while targeting the promotionof adults’ physical activity levels, and 10,000 steps per daycampaigns have been popular and effective approaches inthese efforts [5–9]. Although the 10,000 steps per day goalis not universally appropriate across various levels of age,gender, and physical function, the goal is deemed to be areasonable and motivating target for healthy adults [3].Physical activity intervention studies employing the 10,000steps per day goal have shown weight loss, improvedglucose tolerance, and reduced blood pressure among theoutcomes from increased physical activity toward achievingthis goal [10, 11].Researchers and public health practitioners have rec-ognized that participants in health promotion programshave varying levels of readiness to undertake potentiallychallenging lifestyle changes, such as increasing physicalactivity to a level of 10,000 steps per day [12–16]. Con-sistent with the Transtheoretical Model [12], interven-tions can target various subgroups, based on level ofreadiness for health behavior change, and interventionscan be tailored to differing needs or preferences of theparticipants [13–17]. Studies have shown that suchtargeted approaches are effective [13], cost-effective[14, 15] and often lead to greater improvements in healthoutcomes compared to non-targeted or non-tailored ap-proaches [13, 16].Tailored interventions are increasingly common, and haveshown success in increasing physical activity [13–17]. Fewvalidated assessment instruments are currently available fordetermining Stages of Change according to the Trans-theoretical Model, and none has been based around the10,000 steps per day public health physical activity goal.The purpose of this study was to assess the criterion-related validity of a brief 10,000 steps per day Stages ofChange in Steps (SoC-Step) instrument. To achieve this, weused baseline data from the Walk 2.0 Study, a 3-arm ran-domized controlled trial addressing the effects of Web 2.0applications on engagement, retention, and physicalactivity behavior change [18].HypothesesAccording to the Transtheoretical Model, individuals canbe categorized into one of five stages, based on the combin-ation of behavioral intention and previous behavior. There-fore, our main hypothesis testing focused on each of theseaspects from the SoC-Step instrument: whether or notparticipants indicated that they intended to be, and/or indi-cated currently being, active at a level of taking 10,000 stepsper day. Consistent with previous literature [12, 19, 20], itwas hypothesized that participants who self-categorized asbeing in Precontemplation on the SoC-Step instrumentwould have significantly lower intention scores com-pared to participants who self-categorized into Con-templation or Preparation stages. Regarding previousbehavior, it was hypothesized that those self-categorizedinto Action and Maintenance stages would have greaterlevels of accelerometer-measured steps per day, comparedto those in Precontemplation, Contemplation, or Prepar-ation stages. In a similar vein, it was hypothesized that ifthe SoC-Step instrument had good criterion validity, par-ticipants in the Action and Maintenance stages, as com-pared to those in Precontemplation, Contemplation, orPreparation stages, would have significantly higher likeli-hood to reach the 10,000 steps per day goal.In addition to the core elements of intention andprevious behavior, the Transtheoretical Model suggeststhat self-efficacy increases as individuals move fromContemplation to Preparation and Action stages. There-fore, we hypothesized that self-efficacy to be physicallyactive at a level of taking 10,000 steps per day woulddiffer among these three stages, with Action showing thehighest levels of self-efficacy. Also, given that there is anabundance of literature on the health benefits of regularphysical activity [1], we hypothesized that participants inthe Maintenance stage would have significantly higherlevels of general health and lower body mass indexvalues, as compared to those in Precontemplation,Contemplation, and Preparation.MethodsParticipantsParticipants were 504 Australian adults (176 males, 328females, mean age = 50.8 ± 13.0 years) from the baselinesample of the Walk 2.0 randomized controlled trial [18].This three-arm randomized controlled trial aimed toevaluate the effectiveness of two web-based physicalactivity interventions, compared to a control group(using a physical activity log book). Both interventionsencouraged meeting the 10,000 steps per day goal (or anequivalent level of any physical activity), but also allowedfor individual goals above or below that standard. FullRosenkranz et al. BMC Public Health  (2015) 15:1197 Page 2 of 10details of recruitment methods and the study protocolare published elsewhere [18]. The study received ethicsapproval from the Human Research Ethics Committeesof the University of Western Sydney (Reference numberH8767) and CQUniversity (H11/01-005).To be eligible, participants were residents of WesternSydney (New South Wales) or Rockhampton (Queensland),Australia and were able to speak and read English. Eligibil-ity requirements also included: being free of a medicalcondition which could be exacerbated by physical activity;being over 18 years of age; having access to the Internet butnot a member of the website www.10000steps.org.au; andnot currently meeting physical activity guidelines, butwilling to increase activity levels.Psychosocial measurementThe current study draws upon questionnaire data thatwere obtained from an online survey that participantscompleted during baseline assessments. The online ques-tionnaire included items and scales from the Transtheore-tical Model [12, 20], Social Cognitive Theory [21, 22], andTheory of Planned Behavior [23], all relevant to the 10,000steps per day goal. From the Transtheoretical Model,Stages of Change were determined using the SoC-Stepinstrument by asking whether participants were currentlytaking 10,000 steps per day, how long they had been doingso, or if they were not yet doing so, whether they intendedto, or were preparing or beginning to take 10,000 stepsper day. The SoC-Step instrument (see Additional file 1)was modelled on a previously published scale [20], butadapted to the 10,000 steps goal for the Walk 2.0 project.Relevant to Social Cognitive Theory, the online ques-tionnaire included 4 items (on a 0–100 scale; Cron-bach’s α = 0.848) to assess self-efficacy for physicalactivity [22] and 10 items (on a 0–100 scale; Cronbach’sα = 0.923) to assess self-efficacy to overcome commonbarriers to physical activity [22]. Self-efficacy itemsrequired participants to rate their degree of confidencefor a set of incremental physical activity behavioraltargets (2,000 steps per day; 6,000 steps per day; 10,000steps per day; 14,000 steps per day) by recording anumber from 0 (cannot do at all) to 100 (highly certaincan do). Relevant to the Theory of Planned Behavior,the online questionnaire included 2 items (on a 1–5scale; Cronbach’s α = 0.832) on intention and 4 items(on a 1–5 scale; Cronbach’s α = 0.772) on attitudesrelated to being physically active at a level of taking10,000 steps per day. These were based on a set ofpreviously published items [24], but were modified forWalk 2.0 project needs toward the behavioral goal of10,000 steps. Intention items rated on a 5-point scalewere, “I intend to be physically active at a level oftaking 10,000 steps on most days, if not all days of theweek, for the next month” and “I will try to be physicallyactive at a level of taking 10,000 steps on most days, if notall days of the week, for the next month.”Other online questionnaire scales included the RAND36 item Short Form Health Survey (SF-36) to assessquality of life [25]. This instrument covers 8 dimensionsof health, including limitations in physical activities andusual role activities due to health problems, bodily pain,general mental health, and vitality (energy and fatigue). TheSF-36 has demonstrated adequate validity in Australianpopulations [26], and is suitable for use in the generalpopulation of adults [27, 28].Physical activity monitoringPhysical activity was assessed with ActiGraph GT3Xphysical activity monitors (ActiGraph, Shalimar, FL,USA). ActiGraph monitors have demonstrated reliabil-ity and validity for the measurement of physical activ-ity, including step counts, in free-living environments[29–31]. Participants were provided with an ActiGraph(fastened to an elastic belt) which they were asked towear on their right hip. Participants were instructed towear the ActiGraph continuously for 7 full days duringwaking hours, unless they planned to swim, bathe, orplay contact sports. Participants were asked to complete aphysical activity log that was used to record activitiesundertaken when the ActiGraph was not worn and theduration of non-wear.ActiGraph data were collected in 1-s epochs, includingstep counts. When participants returned to completebaseline assessments (at least 8 days after receiving theActiGraph) the physical data were downloaded and stored.Wear time was assessed using the criteria of 60 min ofconsecutive zero data with a 2-min spike tolerance [32].Minimum valid wear time was set at 600 min of wear timeper day across a minimum of 5 days. Participants withinsufficient valid wear time were asked to wear theActiGraph for an additional 7 days. Those not returningan ActiGraph that contained at least minimal validwear time data in three attempts (n = 39) were excludedfrom analysis. Mean wear time among participants was867.2 (SD = 73.7) minutes per day, recorded across 5.9(SD = 0.7) valid days. Outcome variables for the presentstudy consisted of steps per day (adjusted for totalaccelerometer wear time), and whether or not the dailystep count met the 10,000 steps per day goal.AnthropometryHeight, weight, and waist circumference were mea-sured by project staff in a research setting at baselineassessments. Weight was measured in light clothingand without shoes using Seca 700 mechanical scales(Seca Corp., Hamburg, Germany). Height was measuredwith feet together and head held in the Frankfurtplane, via Seca 220 measuring rod (Seca Corp., Hamburg,Rosenkranz et al. BMC Public Health  (2015) 15:1197 Page 3 of 10Germany). Waist circumference was measured usingSeca 203 measurement tape (Seca Corp., Hamburg,Germany) at the plane aligned with both iliac crests,in accordance with the National Institutes of Healthprotocol [33].Statistical analysisAll analyses were conducted using SPSS for Windows(Version 22.0), with alpha set at <0.05. Variables werechecked for normality and other parametric assumptions,and non-parametric tests were used when assumptionswere violated. Descriptive statistics included frequencies,percentages, mean, median, standard deviation, and inter-quartile range.Independent samples Mann–Whitney or Kruskal Wallistests (including pairwise comparisons where necessary)were used to test:1) The hypothesis that participants inPrecontemplation would have significantly lowerintention scores compared to participants inContemplation or Preparation stages;2) The hypothesis that those in Action orMaintenance stages would have greater levels ofaccelerometer-measured physical activity comparedto those in Precontemplation, Contemplation, orPreparation stages;3) The hypothesis that self-efficacy would differ amongContemplation, Preparation, and Action stages, withAction showing the highest levels of self-efficacy;4) The hypothesis that participants in the Maintenancestage would have significantly higher levels ofgeneral health and lower body mass index values, ascompared to those in Precontemplation,Contemplation, and Preparation stages.Binary logistic regression, with odds ratios and95 % confidence intervals, was used to test the hy-pothesis that participants in Maintenance or Actionstages would have greater likelihood of meeting the10,000 steps goal, in comparison to participants inthe other three stages.ResultsDemographics for the sample are shown in Table 1. Par-ticipants comprised mostly middle-aged adults (meanBMI over 29 kg/m2; mean age around 51 years old); themajority of participants were women (65 %) and wereoverweight or obese (75 %). More participants were re-cruited from the Rockhampton area (n = 311) than frommajor metropolitan area of Western Sydney (n = 193).The sample was diverse with regard to education, in-come, employment, and physical activity level, but theTable 1 Descriptive characteristics of Walk 2.0 study participantsN Mean ± SDAge (years) 504 50.8 ± 13.1Height (m) 504 1.67 ± 0.09Weight (kg) 504 81.8 ± 18.9Body mass index (kg/m2) 504 29.3 ± 6.0Waist circumference (cm) 504 99.9 ± 15.0Physical activity level (unadjusted steps/day) 465 7,248 ± 2,424N PercentageSexMales 176 34.9 %Females 328 65.1 %Weight StatusUnderweight 6 1.2 %Normal weight 117 23.2 %Overweight 181 35.9 %Obese 200 39.7 %EducationPostgraduate degree 40 7.9 %Graduate certificate or diploma 38 7.5 %Bachelor degree 93 18.5 %Advanced diploma/diploma 75 14.9 %School certificate 118 23.4 %School education 140 27.8 %Household combined annual income (AUD)$0—$41,599 111 20.0 %$41,600—$77,999 117 23.2 %$78,000—$129,999 115 22.8 %$130,000+ 103 20.4 %Did not report income 68 13.5 %Area of residenceSydney area 193 38.3 %Rockhampton area 311 61.7 %EmploymentFull time employment 234 46.4 %Employed part-time/casual 111 22.1 %Retired/Pensioner 96 19.0 %Other 63 12.5 %Country of birthAustralia 398 79.0 %United Kingdom 35 6.9 %India 8 1.6 %Other 63 12.5 %Speak language other than English at home 75 14.9 %Rosenkranz et al. BMC Public Health  (2015) 15:1197 Page 4 of 10vast majority was born in Australia (79 %) and spokeEnglish at home (85 %).Table 2 displays participants’ health-related character-istics by Stages of Change. Descriptive statistics includemeans and standard deviations. Inferential statisticsinclude the Kruskal-Wallis H test of distributions. TheseKruskal-Wallis tests revealed that the distribution of allvariables differed among the five Stages of Change(p < 0.05). The following are results of the specifictests of our hypotheses used to examine the validityof the SoC-Step instrument.Intention to be physically active at a level of 10,000 stepsper dayConsistent with hypotheses, there were significant dif-ferences in intention scores between non-intenders (“Ido NOT intend to be physically active at a level oftaking 10,000 steps…” = Precontemplation stage) andthose who intended (“I do intend to be physicallyactive at a level of taking 10,000 steps…” = Contem-plation and Preparation stages) to be more physicallyactive (H = 70.9, df = 2, p < 0.001). Specifically, thoseparticipants who self-categorized into Precontempla-tion (median = 2.3, IQR = 1.1, 3.9), had significantlylower intention scores than those in Contemplation(median = 4.0, IQR = 3.5, 4.0; H = 3.3, df = 1, p = 0.003)or Preparation (median = 4.0, IQR = 4.0, 5.0; H = 5.9,df = 1, p < 0.001).Previous physical activity behavior: accelerometer-measured steps per dayConsistent with study hypotheses, results showed sig-nificant differences in accelerometer-measured stepsper day between participants who self-categorizedinto Action or Maintenance stages (“Currently, I takeenough steps [10,000 steps per day] to receive healthbenefits”), and those who self-categorized into Pre-contemplation, Contemplation, or Preparation (“Cur-rently, I do NOT take enough steps [10,000 steps perday] to receive health benefits.”) stages (U = 17,273,df = 1, p < 0.001). Specifically, those participants whoself-categorized into Action or Maintenance stages(median = 7,654, IQR = 6,386,10,172), had significantlyhigher step counts than those in Contemplation orPreparation (median for “intenders” = 6,724, IQR =5,594, 8,295; H = 68.2, df = 1, p < 0.001) but not Pre-contemplation (median for “non-intenders” = 6,222;IQR = 4,863, 9,605; H = 83.1, df = 1, p = 0.124).Also consistent with the study hypothesis, analysis re-vealed that those in Action or Maintenance stages weremore likely to achieve 10,000 steps per day, as compared tothose in Precontemplation, Contemplation, or Preparationstages (27 vs. 11 % of participants meeting the 10,000 stepsgoal; OR = 3.11; 95 % CI = 1.66, 5.83, p < 0.001).Physical activity self-efficacyFigure 1 shows self-efficacy to achieve 10,000 steps perday by Stages of Change. The stages differed in measuresTable 2 Participants’ health-related variables by Stages of Change (n = 506)Precontemplationn = 16Mean ± SDContemplationn = 201Mean ± SDPreparationn = 214Mean ± SDActionn = 13Mean ± SDMaintenancen = 60Mean ± SDp-valueaSteps per dayb 7,147 ± 3,166 6,998 ± 2,119 7,077 ± 2,210 7,864 ± 2,550 8,464 ± 2,917 p = 0.015Waist circumference 102.7 ± 13.5 100.6 ± 14.8 101.3 ± 15.2 100.5 ± 17.5 91.8 ± 12.8 p < 0.001Body mass index 30.7 ± 5.5 29.7 ± 6.1 29.9 ± 6.0 28.3 ± 4.8 25.2 ± 4.1 p < 0.001Intention 2.5 ± 1.3 3.8 ± 0.7 4.3 ± 0.6 4.0 ± 0.7 4.4 ± 0.6 p < 0.001Self-efficacy PA 6.2 ± 2.8 8.1 ± 1.8 8.9 ± 1.6 9.9 ± 0.8 10.0 ± 1.1 p < 0.001Self-efficacy barriers to PA 5.1 ± 2.4 6.5 ± 1.9 7.0 ± 1.9 7.5 ± 1.9 7.8 ± 1.6 p < 0.001Attitudes 3.1 ± 1.0 4.0 ± 0.5 4.1 ± 0.5 4.1 ± 0.6 4.1 ± 0.7 p < 0.001Physical functioning 76.3 ± 27.2 82.6 ± 18.2 85.9 ± 15.0 87.3 ± 17.5 94.8 ± 7.9 p < 0.001Role physical 68.8 ± 38.2 78.7 ± 34.5 82.8 ± 30.5 76.9 ± 33.0 92.9 ± 16.7 p = 0.035Energy fatigue 52.8 ± 17.9 53.2 ± 18.8 55.7 ± 20.5 66.2 ± 15.6 64.7 ± 18.7 p = 0.001Emotional wellbeing 68.3 ± 19.8 75.4 ± 16.5 78.7 ± 14.8 78.8 ± 14.5 81.1 ± 13.8 p = 0.041Role emotional 56.3 ± 48.3 78.3 ± 33.6 82.9 ± 30.6 87.2 ± 32.0 87.8 ± 28.1 p = 0.021Social functioning 77.3 ± 25.1 85.1 ± 19.0 86.7 ± 19.0 81.7 ± 22.6 91.3 ± 17.0 p = 0.033Pain score 72.7 ± 27.4 78.0 ± 20.9 77.7 ± 21.9 72.3 ± 29.8 87.6 ± 15.0 p = 0.015General health 50.9 ± 16.1 62.0 ± 19.9 63.6 ± 20.0 68.8 ± 11.9 79.1 ± 15.0 p < 0.001Note: Means and SD shown for descriptive purposesaExact p value is from Kruskal-Wallis test of overall difference among Stages of Change distributions (not based on means, due to data skewness)bSteps per day adjusted for total accelerometer wear timeRosenkranz et al. BMC Public Health  (2015) 15:1197 Page 5 of 10of self-efficacy to achieve 10,000 steps per day (H = 34.8,df = 2, p < 0.001). As hypothesized, pairwise comparisonsshowed that Action participants (median = 10.0; IQR = 9.4,10.5) had significantly higher self-efficacy to achieve 10,000steps per day than those in Contemplation (median = 8.5;IQR = 6.8, 9.5; H = 4.1, df = 1, p < 0.001), or Preparation(median = 9.3; IQR = 8.0, 10.0; H = 2.4, df = 1, p < 0.001).Preparation stage participants also had significantlygreater self-efficacy than those in Contemplation (H =4.9, df = 1, p < 0.001).Similarly, stages differed in measures of self-efficacy toovercome common barriers to physical activity (H = 8.4,df = 2, p = 0.015). As hypothesized, pairwise comparisonsshowed that Preparation stage (median = 7.1; IQR = 5.7,8.5) participants had higher self-efficacy than those inContemplation (median = 6.4; IQR = 5.3, 7.9; H = 2.5, df = 1,p < 0.035). Action stage participants (median = 7.1; IQR =6.5, 8.9), however, did not differ significantly from those inPreparation (H = 1.8, df = 1, p = 0.196) or Contemplationstages (H = 0.9, df = 1, p = 0.983)Body mass index and self-reported general healthFigure 2 displays median body mass index, and Fig. 3shows self-reported general health values, across the fivestages of change. Analyses showed that there was asignificant difference across stages for body mass index(H = 24.2, df = 4, p < 0.001), as well as for general health(H = 31.2, df = 4, p < 0.001).Consistent with hypotheses, those in Maintenance stage(median = 24.7 kg/m2, IQR = 22.6, 27.8) had lower bodymass index values than participants in Precontemplation(median = 32.7 kg/m2, IQR = 27.1, 34.6; H = 3.9, df = 1,p = 0.001), Contemplation (median = 28.8 kg/m2, IQR =25.4, 33.3; H = 5.6, df = 1, p < 0.001), and Preparationstages (median = 29.0 kg/m2, IQR = 25.9, 33.3; H = 5.8,df = 1, p < 0.001). Participants in the Maintenance stagewere not significantly different from those in Actionstage for body mass index or general health (p > 0.05).As hypothesized, participants in the Maintenancestage (median = 80.0, IQR = 70.0, 90.0) reported bettergeneral health than their counterparts in Precontem-plation (median = 55.0, IQR = 45.0, 58.8; H = 5.4, df = 1,p < 0.001), Contemplation (median = 65.0, IQR = 50.0,75.0; H = 6.1, df = 1, p < 0.001), and Preparation stages(median = 70.0, IQR = 50.0, 80.0; H = 5.4, df = 1, p < 0.001).DiscussionThis study used the baseline questionnaire, activity monitor,and anthropometric data from the Walk 2.0 Study to assessthe criterion-related validity of a brief 10,000 steps per dayStages of Change (SoC-Step) instrument, and foundsubstantial evidence for its validity. As hypothesized in01234567891011Contemplation* Action†Self-Efficacy for10,000 steps per day Preparation^Fig. 1 Self-Efficacy by Stages of Change. As hypothesized based on the Transtheoretical Model, self-efficacy levels differed across Contemplation,Preparation, and Action Stages (H = 34.8, df = 2, p < 0.001). * Contemplation differed significantly from Preparation (H = 2.4, df = 1, p < 0.001) andAction (H = 4.1, df= 1, p < 0.001) stages. ^ Preparation differed significantly from Contemplation (H = 4.9, df= 1, p < 0.001) and Action (H = 2.4, df = 1,p < 0.001) stages. † Action differed significantly from Contemplation (H = 4.1, df= 1, p < 0.001) and Preparation (H = 4.9, df = 1, p < 0.001) stagesRosenkranz et al. BMC Public Health  (2015) 15:1197 Page 6 of 10accordance with the Transtheoretical Model, there weresignificant differences among the Stages of Change in stepsper day, in likelihood of meeting the 10,000 steps per daygoal, and in intention and self-efficacy related to achieving10,000 steps per day.According to Devon and colleagues [34], criterion-related validity is evidenced by the relationship betweenthe characteristics of a measurement instrument andperformance on another performance measurement.Concurrent criterion-related validity refers to the corres-pondence between scores from two instruments thatwere measured at the same point in time [34]. A meas-ure achieves its degree of criterion validity (also knownas empirical or statistical validity) to the extent that itcorresponds with another observation that accuratelymeasures the variable under study [35].One key element of the SoC-Step instrument is thereported intention to “be physically active at a level oftaking 10,000 steps or more on most days, if not all daysof the week.” The Theory of Planned Behavior two-itemscale of intention scores provided a concurrent measureof this construct. Consistent with theory, these intentionscores varied as a function of the Stages of Change inthe hypothesized direction, such that Precontemplationstage participants had lower levels of intention thanthose in Contemplation or Preparation. These findingssupport the concurrent validity of SoC-Step with regardto intention to meet the 10,000 steps per day standard.0102030405060Precontemp on* Contemp on* Preparation* Ac on MaintenanceBodyMassIndex(Kg/m2 )Fig. 2 Body Mass Index by Stages of Change. As hypothesized, body mass index values differed by Stages of Change (H = 24.2, df = 4, p < 0.001).* Maintenance stage differed significantly from Precontemplation (H = 3.9, df = 1, p = 0.001), Contemplation (H = 5.6, df= 1, p < 0.001) and Preparationstages (H = 5.8; df = 1, p < 0.0010102030405060708090100Precontemplation* Contemplation* Preparation*  Action      MaintenanceSelf-Rated General HealthFig. 3 General Health by Stages of Change. As hypothesized, self-reported general health differed by Stages of Change (H = 31.2, df = 4,p < 0.001). * Maintenance stage differed significantly from Precontemplation (H = 5.4, df = 1, p = 0.001), Contemplation (H = 6.1, df = 1, p < 0.001)and Preparation stages (H = 5.4, df = 1, p < 0.001Rosenkranz et al. BMC Public Health  (2015) 15:1197 Page 7 of 10In the present study, our criterion of physical activitywas the daily step count obtained from ActiGraph activ-ity monitors. The Stages of Change measure was signifi-cantly related to daily step count, as hypothesized. Mostgermane to the issue, those participants who indicatedthat, “Currently, I take enough steps (10,000 steps perday) to receive health benefits,” on the SoC-Step instru-ment (i.e., those classified in Action or Maintenancestages) had the highest step counts, as well as the high-est likelihood of meeting the 10,000 steps per day stand-ard. Together, these findings suggest that the SoC-Stepinstrument shows good concurrent validity with theActigraph physical activity monitor criterion variable.Self-efficacy is a central construct within both SocialCognitive Theory and the Transtheoretical Model. As hy-pothesized, the measures of physical activity self-efficacypertaining to participants’ confidence to achieve varyinglevels of steps per day varied as a function of Stages ofChange. The measure of self-efficacy pertaining to partici-pants’ confidence to be active when faced with variouschallenges (self-efficacy to overcome common barriers tophysical activity) was also related to Stages of Change inthe hypothesized manner, although Action stage partic-ipants were not significantly different from those inContemplation or Preparation (possibly due to the smallnumber in Action stage). Overall, these findings providefurther evidence for the criterion-related validity of thebrief SoC-Step instrument.The evidence for validity of the investigated SoC-Stepinstrument parallels that of other researchers who havepreviously developed and validated brief Stages of Changeinstruments [19, 20, 36–38]. The present instrument,however, is unique in its focus on the 10,000 steps goal.Reed and colleagues [37] investigated various algorithmsto ascertain optimal characteristics of staging regular exer-cise behavior. These authors found that precise definitionsof exercise, including all parameters needed to meet acriterion, were helpful to participants who were assessingthemselves relative to a stage of change [37]. Furthermore,a five-choice format was endorsed as effective in assessingstage. The present study’s instrument conforms with theserecommendations, in that extensive information with aprecise behavioral definition is provided to the participant,along with a five-choice response format. The presence ofsuch recommended characteristics may partially explainthe substantial concordance observed between the SoC-Step and the criterion measure.The observed relationships between stage classificationand health are similar to systematic review findings byBize et al. [39], along with those of Laforge et al. [40],who found that exercise stage was associated with self-perceived quality of life. In particular, Laforge et al. foundthat general health was lowest in Precontemplation, andhighest in the Maintenance stage. Similarly, Cardinal andcolleagues [36] found a relationship between Stages ofChange in exercise and body mass index. The presentstudy, however, is unique in its framing of Stages ofChange around a public health physical activity goal, andin using additional objective health-related comparisonssuch as activity monitor data and clinically measured bodymass index.One limitation with regard to validity of this instrumentwas observed in that many participants who categorizedthemselves in Action or Maintenance stages for a physicalactivity level of 10,000 steps per day did not meet thisstandard based on objectively measured accelerometerdata. It is important to note, however, that accelerometersdo not capture all types of physical activity, and may pro-vide a conservative estimate of step counts in some cases[31, 41]. In addition, our reliance on self-report instru-ments for both Stages of Change and intention may resultin bias associated with common method variance, althoughundertaking measurement of participant intention withoutreliance on self-report would likely present alternate formsof bias.Another limitation is the small number of participantsclassified in Precontemplation, which then hindered ourability to detect differences in relevant pairwise compari-sons, due to low power. Having small numbers in Precon-templation is expected here, however, given that our studysample comprised those successfully recruited to a behav-ior change study. Although it may appear incongruousthat participants recruited into a physical activity promo-tion study could classified in the Precontemplation stage,there are two issues to consider: 1) Stages of Change isspecific to a target behavior [12], and while our behavioraltarget was 10,000 steps daily (at any intensity) for categor-izing participants, they were originally included in theintervention if they “were currently engaging in less than ahalf an hour (30 min) of moderate-to-vigorous (e.g., walk-ing, running or playing sport) physical activity on five ormore days of the week;” 2) Stages of Change suggests thatparticipants frequently regress into earlier stages [12], andthat could have been a factor in the present study. Last,our sample was delimited to middle-aged Australiansfrom two regions participating in an intervention (pre-dominantly female, mostly overweight and obese, of anarrow age range, and mostly categorized in contempla-tion and preparation stages), so our findings may notapply to demographically different populations. Futurestudies should assess validity of the SoC-Step Instrumentin younger adults.Despite the aforementioned limitations, the presentstudy possessed a number of strengths, including a largeand diverse sample of middle-aged Australian adults whomay represent a larger population of those who would bewilling to participate in public health physical activityinterventions. These participants provided not only self-Rosenkranz et al. BMC Public Health  (2015) 15:1197 Page 8 of 10report data, but also objective measures of physical activityand body mass index, which contribute to the rigor of thisstudy’s methods.ConclusionsIn this study, a cohort of mostly middle-aged, over-weight and obese, Australian adults showed variationsin objectively measured steps, body mass index, self-reported intention, and self-efficacy, as a function ofself-reported Stages of Change toward the public healthgoal of achieving 10,000 steps daily. These variationscorresponded with the hypotheses derived from theTranstheoretical Model, and thereby provided supportfor the criterion-related validity of the SoC-Step instru-ment. This brief instrument appears to have goodcriterion-related validity for determining Stages ofChange related to the public health goal of 10,000 steps,and could be useful in tailored intervention efforts thatcould help lead to improvements in health-related qualityof life.Additional fileAdditional file 1: The SoC-Step, a Stages of Change instrumentrelevant to the physical activity goal of 10,000 steps per day.(PDF 175 kb)AbbreviationsSoC-Step: The Stages of Change in Steps instrument related to 10,000 stepsper day; SF-36: Short Form Health Survey.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsRRR conceived of the study, analyzed data, and wrote the first draft of themanuscript. WKM, GSK, AJM, CV, MJD, and CMC conceived the Walk 2.0 Projectand procured the project funding. GSK led the coordination of the trial. TNSmanaged the project, including data collection. AJM developed the IT platformfor the project. All authors read, edited, and approved the final manuscript.AcknowledgementsThis trial is being funded by the National Health and Medical Research Council(Project Grant number 589903). The funder does not have any role in the studyother than to provide funding. CV was supported by a National Health andMedical Research Council (#519778) and National Heart Foundation of Australia(#PH 07B 3303) post-doctoral research fellowship during the conception of theresearch project and recruitment. CV is currently supported by a Future LeaderFellowship (ID 100427) from the National Heart Foundation of Australia.MJD is supported by a Future Leader Fellowship (ID 100029) from theNational Heart Foundation of Australia. Publication of this article was fundedin part by the Kansas State University Open Access Publishing Fund.Author details1Kansas State University, Manhattan, USA. 2Western Sydney University,Sydney, Australia. 3University of Newcastle, Newcastle, Australia. 4University ofBritish Columbia, Kelowna, Canada. 5Central Queensland University,Rockhampton, Australia. 6University of Alberta, Edmonton, AB, Canada.Received: 18 February 2015 Accepted: 23 November 2015References1. US Department of Health and Human Services (USDHHS). Physical ActivityGuidelines Advisory Committee Report, Washington, DC; 2008.2. Hallal PC, Andersen LB, Bull FC, Guthold R, Haskell W, Ekelund U. Globalphysical activity levels: surveillance progress, pitfalls, and prospects. Lancet.2012;380:247–57.3. Tudor-Locke C, Craig CL, Brown WJ, Clemes SA, De Cocker K, Giles-Corti B,et al. How many steps/day are enough? For adults. Int J Behav Nutr PhysAct. 2011;8:79.4. Bravata DM, Smith-Spangler C, Sundaram V, Gienger AL, Lin N, Lewis R,et al. Using pedometers to increase physical activity and improve health: asystematic review. JAMA. 2007;298:2296–304.5. Brown WJ, Mummery WK, Eakin E, Schofield G. 10,000 Steps Rockhampton:evaluation of a whole community approach to improving population levelsof physical activity. J Phys Act Health. 2006;3:1–14.6. De Cocker KA, De Bourdeaudhuij IM, Brown WJ, Cardon GM. Effects of “10,000steps Ghent”: a whole-community intervention. Am J Prev Med. 2007;33:455–63.7. Davies C, Corry K, Van Itallie A, Vandelanotte C, Caperchione C, MummeryWK. Prospective associations between intervention components andwebsite engagement in a publicly available physical activity website: thecase of 10,000 steps Australia. J Med Internet Res. 2012;14:e4.8. Mummery WK, Schofield G, Hinchliffe A, Joyner K, Brown W. Disseminationof a community-based physical activity project: the case of 10,000 steps.J Sci Med Sport. 2006;9:424–30.9. Brown W, Eakin E, Mummery WK, Trost S. 10,000 Steps Rockhampton:establishing a multi-strategy physical activity promotion project in acommunity. Health Promot J Austr. 2003;14:95–100.10. Schneider PL, Bassett Jr DR, Thompson DL, Pronk NP, Bielak KM. Effects of a10,000 steps per day goal in overweight adults. Am J Health Prom. 2006;21:85–9.11. Swartz AM, Strath SJ, Bassett Jr DR, Moore JB, Redwine BA, Groër M, et al.Increasing daily walking improves glucose tolerance in overweight women.Prev Med. 2003;37:356–62.12. Prochaska JO, DiClemente CC. Toward a comprehensive model of change.In: Miller WR, Heather N, editors. Treating Addictive Behaviors: Processes ofChange. 1st ed. New York: Plenum Press; 1986. p. 3–27.13. Krebs P, Prochaska JO, Rossi JS. A meta-analysis of computer-tailoredinterventions for health behaviour change. Prev Med. 2010;51:214–21.14. Golsteijn RH, Peels DA, Evers SMAA, Bolman C, Mudde AN, et al. Cost-effectiveness and cost-utility of a Web-based or print-delivered tailoredintervention to promote physical activity among adults aged over fifty: aneconomic evaluation of the Active Plus intervention. Int J Behav NutrPhys Act. 2014;11:10–1186.15. Van Keulen H, Bosman J, van Tulder M, Severens J, de Vries H, Brug J, et al.Cost-effectiveness of tailored print communication, telephone motivationalinterviewing, and a combination of the two: results of an economicevaluation alongside the Vitalum randomized controlled trial. Int J BehavNutr Phys Act. 2010;7:64.16. Lustria MLA, Noar SM, Cortese J, Van Stee SK, Glueckauf RL, Lee J. A meta-analysisof web-delivered tailored health behavior change interventions. J Health Comm.2013;18:1039–69.17. Peels DA, Bolman C, Golsteijn RHJ, De Vries H, Mudde AN, Van Stralen MM,et al. Long-term efficacy of a tailored physical activity intervention amongolder adults. Int J Behav Nutr Phys Act. 2013;10:104.18. Kolt GS, Rosenkranz RR, Savage TN, Maeder AJ, Vandelanotte C, Duncan MJ,et al. WALK 2.0-using Web 2.0 applications to promote health-relatedphysical activity: a randomised controlled trial protocol. BMC Public Health.2013;13:436.19. Lippke S, Fleig L, Pomp S, Schwarzer R. Validity of a stage algorithm for physicalactivity in participants recruited from orthopedic and cardiac rehabilitationclinics. Rehab Psych. 2010;55:398–408.20. Ronda G, Van Assema P, Brug J. Stages of change, psychological factors andawareness of physical activity levels in the Netherlands. Health Promot Int.2001;16:305–14.21. Bandura A. Social foundations of thought and action: a social cognitivetheory. Englewood Cliffs, NJ: Prentice-Hall; 1986.22. Bandura A. Guide for constructing self-efficacy scales. In: Pajares F, Urdan TC,editors. Self-Efficacy Beliefs of Adolescents. Volume 5. Greenwich: InformationAge Publishing; 2006.23. Ajzen I. The theory of planned behavior. Organ Behav Hum. 1991;50:179–211.Rosenkranz et al. BMC Public Health  (2015) 15:1197 Page 9 of 1024. Caperchione CM, Duncan MJ, Mummery WK, Steele R, Schofield G.Mediating relationship between body mass index and the direct measuresof the theory of planned behaviour on physical activity intention. PsycholHealth Med. 2008;13:168–79.25. Ware Jr JE, Sherbourne CD. The MOS 36-item short-form health survey(SF-36): I. Conceptual framework and item selection. Med Care. 1992;30:473–83.26. McCallum J. The SF-36 in an Australian sample: validating a new, generichealth status measure. Aust J Public Health. 1995;19:160–6.27. Stevenson C. SF-36: Interim Norms for Australian Data. Australia: AustralianInstitute of Health and Welfare; 1996.28. Australian Bureau of Statistics. National health survey: SF-36 populationnorms; Australia. Canberra: ABS; 1995.29. Sasaki JE, John D, Freedson PS. Validation and comparison of ActiGraphactivity monitors. J Sci Med Sport. 2011;14:411–6.30. Vanhelst J, Mikulovic J, Bui-Xuan G, Dieu O, Blondeau T, Fardy P, et al.Comparison of two ActiGraph accelerometer generations in the assessmentof physical activity in free living conditions. BMC Res Notes. 2012;5:187.31. Esliger DW, Probert A, Connor GS, Bryan S, Laviolette M, Tremblay MS.Validity of the Actical accelerometer step-count function. Med Sci SportsExerc. 2007;39:1200–4.32. Healy GN, Matthews CE, Dunstan DW, Winkler EA, Owen N. Sedentary timeand cardiometabolic biomarkers in US adults: NHANES 2003–06. Eur Heart J.2011;32:590–7.33. NHLBI Obesity Education Initiative Expert Panel. Treatment guidelines. In:Clinical Guidelines on the Identification, Evaluation, and Treatment ofOverweight and Obesity in Adults—the Evidence Report. Washington D.C.:National Institutes of Health; 1998. p. 56–94.34. DeVon HA, Block ME, Moyle‐Wright P, Ernst DM, Hayden SJ, Lazzara DJ,et al. A psychometric toolbox for testing validity and reliability. J NursScholarship. 2007;39:155–64.35. Kaplan RM, Bush JW, Berry CC. Health status: types of validity and the indexof well-being. Health Serv Res. 1976;11:478.36. Cardinal BJ. Construct validity of stages of change for exercise behavior. AmJ Health Prom. 1997;12:68–74.37. Reed GR, Velicer WF, Prochaska JO, Rossi JS, Marcus BH. What makes a goodstaging algorithm: examples from regular exercise. Am J Health Prom. 1997;12:57–66.38. Marttila J, Nupponen R. Assessing stage of change for physical activity: howcongruent are parallel methods? Health Educ Res. 2003;18:419–28.39. Bize R, Johnson JA, Plotnikoff RC. Physical activity level and health-relatedquality of life in the general adult population: a systematic review. PrevMed. 2007;45:401–15.40. Laforge RG, Rossi JS, Prochaska JO, Velicer WF, Levesque DA, McHorney CA.Stage of regular exercise and health-related quality of life. Prev Med. 1999;28:349–60.41. Rosenkranz RR, Rosenkranz SK, Weber C. Validity of the Acticalaccelerometer step-count function in children. Ped Exerc Sci. 2011;23:355.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Rosenkranz et al. BMC Public Health  (2015) 15:1197 Page 10 of 10


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items