An Internet-based intervention for promoting and maintaining physical activity: a randomized controlled trial.

Delivered by Publishing Technology to: University of California - San Francisco IP: 169.230.243.252 on: Tue, 09 Dec 2014 14:05:28 Copyright (c) PNG Publications. All rights reserved.

An Internet-Based Intervention for Promoting and Maintaining Physical Activity: A Randomized Controlled Trial Sonthaya Sriramatr, PhD; Tanya R. Berry, PhD; John C. Spence, PhD

Objectives: To evaluate the efficacy of a Social Cognitive Theory-based Internet intervention designed to promote and maintain leisure-time physical activity in university-aged female students in Thailand. Methods: The 3-month intervention was delivered through a website and e-mails with a follow-up evaluation 3 months after the end of the intervention. Female students (N = 220) were allocated to 4 parallel groups. Results: Significant increases in steps/ day, weekly leisure-time activity score, self-efficacy, outcome expectations, and self-regulation, and reduced resting

T

hai female students are less likely to participate in leisure-time physical activity for enhancing health benefits.1 Only 8.9% of Thai female adolescents met physical activity recommendations compared to 13.1% from the Philippines and 17.1% from Indonesia.2 In countries such as Japan (~8%) and Korea (~18%), the prevalence of leisure-time physical activity at recommended levels was higher than for Thai female students--only about 2% of whom were active.1 It is well known that physical activity can be increased through theoretical based interventions.3 Theoretically-based intervention programs are more efficacious4,5 and associated with larger effect sizes6 than atheoretical interventions. Social Cognitive Theory (SCT) is potentially useful4,5,7,8 as it is often used in intervention research9-11 and highly associated with physical activity change.12 For example, a SCT Internet-based intervention based increased modSonthaya Sriramatr, Assistant Professor and Chair, Department of Sports Science, Faculty of Physical Education, Srinakharinwirot University, Thailand. Tanya R. Berry, Associate Professor and Canada Research Chair, Physical Activity Promotion, and John C. Spence, Professor and Associate Dean, Research, Faculty of Physical Education and Recreation, University of Alberta, Canada. Correspondence Dr Sriramatr; [email protected]

430

heart rate were observed at the end of the intervention. With the exception of resting heart rate, the intervention effects on these variables also remained at the follow-up. Conclusions: The Internet intervention program was effective in promoting and maintaining leisure-time physical activity in university-aged female students. Key words: social cognitive theory, self-efficacy, outcome expectations, selfregulation, steps, weekly leisure-time activity score Am J Health Behav. 2014;38(3):430-439 DOI: http://dx.doi.org/10.5993/AJHB.38.3.12

erate physical activity in college female students in the United States.13 Similarly, a SCT-based intervention increased targeted constructs (self-efficacy and outcome expectations) for exercise, physical activity, and physical performance in older Thai adults with knee replacement.14 SCT constructs include self-efficacy (SE), which is the confidence to perform a particular behavior, including confidence in overcoming the barriers to performing behavior.15 Outcome expectations (OE) are expectations that a given behavior will produce a particular outcome,16 and self-regulation (SR) variables include an individual’s ability to set achievable goals, use effective strategies for attaining goals, and self-monitor to evaluate the success in attaining the goals.17 Studies have shown that these SCT constructs play an important role in physical activity initiation and adherence.18,19 Bandura20 posited that SE and OE influence physical activity directly and through the development and use of SR. Also, many studies have found that by enhancing SE, OE, and SR, physical activity can be increased and maintained.7,16,17 Nonetheless, there are no such studies in young Thai adults. One SCT-based intervention was conducted with Thai elders undergoing knee replacement13 and several other descriptive studies have been published.13,21,22 In general, these studies found positive


Sriramatr et al effects of SCT constructs on physical activity.13,21,22 Therefore, using SCT as a theoretical framework to promote and maintain physical activity in Thai female students is warranted. The Internet is a promising tool to promote physical activity due to its increased usage.23,24 Over 30% of Thais had access to the Internet in 2012.23 The Internet allows users to access information quickly and conveniently.25,26 The Internet also may reduce barriers to participation such as travel, time, and costs.27 Internet-based physical activity interventions have been reported as cost-effective methods for promoting physical activity,27 and are related to significant increases in physical activity.5,9,26,28 Such interventions are appealing for use in universities that have extensive technological networks that allow Internet access for students. In Thailand, 90% of undergraduate students have access to the Internet, and 75% of students accessed the Internet at university 1-3 hours a time and 1-3 times a week.29 Therefore, physical activity in Thai female students might effectively be promoted and maintained through an Internet intervention. This study used SCT and intervention mapping as frameworks for developing an Internet-based intervention physical activity program. A Solomon 4-group design was employed to investigate the effects of the program on primary outcomes: SCT constructs (ie, SE, OE, and SR), steps/day, and weekly leisure-time activity score (LTAS), and secondary outcomes: VO2max and resting heart rate. The specific purpose was to evaluate the efficacy of a SCT-based Internet intervention designed to promote and maintain leisure-time physical activity in university-aged female students in Thailand. It is hypothesized that participants in the intervention groups would have significant improvements in outcome variables compared with those in the control groups. It was further hypothesized that there would be no significant pretest sensitization effects on the outcome variables because of previous studies that did not report the influence of pretest sensitization effects.30,31 METHODS Participants and Selection Procedure This report follows CONSORT guidelines32 but was not registered in a publicly accessible clinical trials database. The number of participants was based on a power analysis conducted on the intervention factor (2 levels), the pretest factor (2 levels) and the interaction effect of the intervention by pretest. Because intervention and truecontrol groups were compared, a medium effect size (Cohen’s f=.25) was assumed for the interaction group. Setting a power of 0.80, the effect size of 0.25,27 and an alpha level of .05 for the main and interaction effects, 32 participants per group were required. A review of the literature of Internetbased physical activity interventions found that the average attrition rate was about 30%.5Thus, we recruited a sample that permitted assignment of at

Am J Health Behav.™ 2014;38(3):430-439

least 42 participants per group. Participants were recruited by advertising placed in newsletters, on university notice boards, the university website, and through the university email. Participants were screened for eligibility via interview. Eligibility criteria included: female students aged 18-24 years, able to use a computer with minimal assistance to complete an Internet survey, were not enrolled in a physical activity program, would not participate in other physical activity programs during the study, and responded “no” to all questions on the Physical Activity Readiness Questionnaire indicating no health risks for participating in physical activity.33 Overall, 224 female students expressed interest but 4 students were ineligible due to health problems. Thus, 220 students participated in this study. Eligible participants completed informed consent procedures, which had been reviewed and approved by the appropriate university Research Ethics Boards. Internet-based Intervention Physical Activity Program The program was developed based on SCT and the intervention mapping framework. Intervention mapping increases the link between theory and practice34 and contributes to more effective programs.35 It contains 6 steps.36 In the first step, a needs assessment determined that low rates of physical activity was a problem among Thai students. Specific program objectives, behavioral outcomes, and health outcomes of the intervention were determined. SCT was chosen as an appropriate theory and if SCT variables (ie, SE, OE, and SR) change as a result of the intervention, the behavioral outcomes and health outcomes should also change.37 Thus, the matrix of change objectives was developed in the second step (Table 1) and searching and selecting theoretical methods of behavioral change and translating these methods into practical applications was done in the third step (Table 2). In the fourth step, the intervention program was set and adopted and implemented by the researchers in the fifth step. In the last step, evaluation planning was undertaken. Because indicators of success are improvements in determinants of behavioral outcomes, behavioral outcomes themselves, and health outcomes, both direct and indirect measures were used during evaluation. Information was delivered through a website (http://www.sport-exercise.com) and e-mails. E-mails directed participants to the website, gave personal feedback, and provided physical activity of role models. Research Design A 3-month randomized control trial intervention was conducted because it has been found that 3-month Internet physical activity interventions are more effective than longer ones.5 The Solomon 4-group design was used because increases

DOI:

http://dx.doi.org/10.5993/AJHB.38.3.12

431

An Internet-Based Intervention for Promoting and Maintaining Physical Activity: A Randomized Controlled Trial

Table 1 Matrixes of Change Objectives Behavioral Outcome: PA Level Changes


Performance Objectives

Determinants SE

OE

SR

Participate in MVPA at least 30 minutes per day for 3 days a week

SE.1: Express confidence in ability to participate in MVPA

OE.1: Expect that participation in MVPA can improve outcomes

SR.1: Set PA goals and monitor PA each week

Increase time spent in MVPA at least 9 minutes per week

SE.2: Express confidence in ability to increase the duration in MVPA

OE.2: Expect that increasing the duration in MVPA can improve outcomes

SR.2 Set PA goals and monitor PA each week

Note. SE = Self-efficacy; OE = Outcome Expectations; SR = Self-regulation; MVPA = Moderate-to-Vigorous Physical Activity; PA = Physical Activity

in physical activity in control group participants may occur in randomized controlled physical activity promotion trials,31 a phenomenon known as pretest sensitization.38 This effect may occur because by completing a questionnaire, new cognitions may be created which may subsequently influence behavior.39 Thus, the pretest may serve as an intervention itself, which will compromise the internal validity of the actual intervention.31 The Solomon 4-group design controls for this possibility by including groups that have not completed the pretest.40 Therefore, the 220 eligible students were randomly allocated (using a computer-generated list of random numbers) to 4 parallel groups: intervention with pretest (I-P) and no pretest (I-NP) groups and control with pretest (C-P) and no pretest (C-NP) groups (Figure 1). The intervention groups. Face-to-face orientation to the website and self-monitoring information methods were taught to participants in the intervention groups (I-P and I-NP group) before starting the study. All participants were instructed on website navigation and login procedures and received a pedometer. During the 3-month intervention period, participants accessed the website and recorded their average physical activity (ie, steps/day and minutes/day), set their physical activity goals (ie, minutes/day) for the next week, and identified their SE and OE on physical activity goals that they had set every Saturday. Weekly e-mails were sent to each participant every Sunday for a total of 12 e-mails sent during the intervention period. At the start of the intervention, participants were advised to accumulate at least 90 minutes of moderate-to-vigorous physical activity per week (ie, 30 minutes on 3 days) and to increase by at least 9 minutes per week (ie, 3 minutes on 3 days). If they were to adhere to this advice, they would meet the recommendation of physical activity for health benefits at the eighth week (ie, 150 minutes per week)41 and they would spend more than 180

432

minutes per week being active (ie, 60 minutes on 3 days) by the end of the intervention. Participants could be active in any way they chose. During the 3-month follow-up period, 3 e-mails during the fourth week of each month asking participants to report their weekly physical activity. Retention strategies included monetary payment and direct contact with participants. Participants in the intervention and control groups were paid 900 Baht (~$30 CAD) and 300 Baht (~$10 CAD), respectively to compensate them for their time. The payments did not appear to influence quality or completeness of data collection and retention rate of participants in the intervention compared to the control groups (Figure 1). Participants were contacted directly via e-mail or telephone when they did not access the website during a given week. The mobile phone was used when participants did not respond to the e-mail after 24 hours. On average, 5-10 persons were sent reminder e-mails and 1-2 persons were called each week. The control groups. Participants in the control groups (C-P, C-NP) did not receive any treatments, but they were tested on the dependent variables and received computer instruction and pedometers. Data collection. Participants in the I-P and C-P groups completed the measures at the pretest, the end of the intervention (ie, 12 weeks after the start of the intervention), and the 3-month follow-up. Participants in the I-NP and C-NP groups completed the measures at the end of the intervention and the 3-month follow-up. All data collection was conducted by female research assistants in the same testing facility and on the same schedule. Physical activity variables. Physical activity variables include total weekly LTAS and steps/day. The total weekly LTAS were measured using the Thai version of the Godin-Shephard Leisure-Time Physical Activity Questionnaire42 which includes 3 questions: “considering a 7-Day period (a week),

Sriramatr et al

Table 2 Frameworks of Program Developments Determinants Methods


SE

OE

SR

Performance accomplishments

Practical Strategies

How it can be accomplished through Internet

Students receive information about their behavioral and health outcomes at pretest and the end of the intervention. Students start PA at minimum level. Students set PA goals by increasing 9 minutes each week. Students record their PA (steps & minutes) every week. Students receive positive feedback about their PA levels each week.

Show a graph summarizing individual’s behavior and health outcomes at pretest and the end of the intervention. --Goal setting logs. PA logs. Receive information by e-mails.

Vicarious experience

Students receive information about PA of role models. Receive information by e-mails. Students received information about their PA. Show a graph of PA goals and actual PA in each week.

Verbal persuasion

Students receive positive feedback about their PA levels each week.

Receive information by e-mails.

Improve physical and emotional state

Students participate in PA at minimum levels and increase PA at lower rate each week.

---

Increasing awareStudents receive information about PA. ness of the potential Students receive information about PA workouts. benefits of PA Students receive general PA guidelines. Students know their OE on PA goals each week.

Post in the website. Post in the website. Post in the website. Show a graph of OE on PA goals in each week.

Creating favorable PA outcome expectancies

Students choose their own activities Students start PA at minimum level Students increase PA at lower rate each week

-------

Goal setting

Students set PA goals by increasing 9 minutes each week Students receive positive feedback about their PA Students receive information about their PA goals

Goal setting logs.

Self-monitoring

Receive information by e-mails. Show a graph of PA goals and actual PA in each week.

Students record their PA (steps & minutes) each week PA logs.

Note. SE = Self-efficacy; OE = Outcome Expectations; SR = Self-regulation; PA = Physical Activity

how many times on average do you do the following kinds of physical activity (ie, strenuous, moderate and mild) for more than 15 minutes during your free time.” The total weekly LTAS is calculated by multiplying weekly frequencies of mild, moderate, and strenuous activities by 3, 5, and 9 respectively which correspond to metabolic equivalent categories of the activities listed and summing the products.43 The Thai version of the Godin-Shephard Leisure-Time Physical Activity Questionnaire has shown test-retest reliability and construct validity with Thai female undergraduate students.42 Steps were measured with a Yamax Digi-Walker SW-701 which accurately measures steps.44 Participants were asked to wear the pedometers for 3 days on weekdays during each testing week. Physical fitness variables. Resting heart rate was measured after participants had sat quietly for 5 minutes. The wrist pulse rate was measured for one minute. Body weight (kg) was measured with a digital Tettler Toledo—Wildcat® weighting machine. Height was measured to the nearest 0.1 cm using

a stadiometer. Both body weight and height were measured twice and the average used to calculate body mass index (BMI). The Queen’s College Step Test was used to measure VO2max because it has been shown to be a valid and reliable measure of VO2max in female university students45 and a strong correlation (r = -0.83) between VO2max and heart rate following the step test has been shown in females aged 22 years.46 SCT variables. Thai versions of the Multidimensional Self-efficacy for Exercise Scale (MSES), Outcome Expectations Questionnaire (OEQ), and Selfregulation Questionnaire (SRQ)42 were used. The MSES contains 9 items that measure task, coping, and scheduling efficacy for performing physical activity.47 Each item starts with the statement: “How confident are you that you can...”, and follows with statements to measure task (eg, follow directions), coping (eg, when lacking energy), and scheduling (eg, in your daily routine) efficacies. All items were rated from 0% (no confidence) to 100% (complete confidence). A previous study in Thai female un-

Am J Health Behav.™ 2014;38(3):430-439

DOI:


433


Figure 1 Flow Diagram of Participants through Trial

Assessed for eligibility N = 224


Enrollment

I-P Randomized to grp: N = 55 Received intervention: n = 50 Discontinued Int: N = 5

C-P Randomized to grp: N = 55 Discontinued: N = 1 Injury from accident: N = 1

Drop out (lack of time): N = 5

Followed up: N = 55

Followed up: N = 55

Completed: N = 48

Completed: N = 50

Did not complete: N = 7


Refusal (lack of time): N = 7


Followed up: N = 55

Followed up: N = 55

Completed: N = 45

Completed: N = 42




Included in Intention-totreat Analysis: N = 55

Random Assignment N = 220

End of the intervention

3-Month Follow-up

Analysis

dergraduate students reported a Cronbach’s alpha of 0.93 for task, coping, and scheduling efficacies.42 The OEQ contains 9 items regarding the beliefs and values of outcomes of being active (eg, weight control, fun, fitness). Each item was rated on a 5-point belief scale ranging from disagree a lot (1) to agree a lot (5), then on a 5-point value scale ranging from very unimportant (1) to very important (5).48 Cronbach’s alphas ranging from 0.82 to 0.84 for mental, social benefits, and physical outcomes of physical activity have been reported with female Thai undergraduate students.42 The SRQ contains the 10-item Exercise Goal-Setting Scale (EGS) and the 10-item Exercise Planning and Scheduling Scale (EPS).49 The EGS includes items related to goal setting, self-monitoring, and problem solving. The EPS includes items related to scheduling and planning exercise as part of one’s daily routine. All items were scored on a 5-point scale ranging from 1 (does not describe) to 5 (describes completely). A previous study in Thai female undergraduate students reported good internal reliabilities (Cronbach’s alphas = 0.72 - 0.89).42 Data analysis. A true intention-to-treat analysis was used in which all randomized participants were included in the groups that they were allocated and multiple imputation was used for han-

434

I-NP Randomized to grp: N = 55

C-NP Randomized to grp: N = 55

Received intervention: n = 49 Discontinued Int: N = 6

Drop out (lack of time): N = 6

Followed up: N = 55

Followed up: N = 55

Completed: N = 45

Completed: N = 53





Followed up: N = 55

Followed up: N = 55

Completed: N = 43

Completed: N = 47






Excluded - Did not meet eligibility criteria: N = 4




dling missing data using a 3-step process previously reported.50 The significance level was set at .05. To assess the presence of pretest sensitization and the main effects of the intervention, statistical methods for the Solomon 4-group design provided by Braver and Braver40 were used such that a series of 2 (intervention or control) x 2 (pretest or no pretest) between-groups analysis of variance (ANOVA) tests were performed on each of the posttest dependent variable scores. Based on the outcomes of the interaction between the intervention groups and the pretest, the main effects of the pretest and the intervention could be interpreted (see Braver and Braver40 for more information). To aid in the interpretation of findings, Cohen’s categorization of partial eta-squared as small (0.01), medium (0.06), and large (0.16) effects was adopted.51 RESULTS Descriptive Statistics The number of participants retained for each stage of the study is shown in Figure 1. The average age, weight, height, and BMI of participants in each group were about 19 years, 52 kg, 159 cm, and 21 kg/m2, respectively. There were no significant differences between the intervention and control groups on any of the demographic, physical

Sriramatr et al

Table 3 Mean Score of Outcome Variables


Intervention with Pretest

Steps LTAS RHR VO2max SE OE SR

Control with Pretest

Pretest


Follow-up

Pretest


Follow-up

7375.29 ± 2944.18 43.14 ± 23.43 80.72 ± 10.14 37.43 ± 3.22 63.14 ± 12.15 82.73 ± 5.57 60.34 ± 9.52

11002.88 ± 2059.49 51.89 ± 21.89 78.54 ± 8.52 37.79 ± 3.44 64.86 ± 10.93 86.66 ± 7.66 61.97 ± 10.36

11654.56 ± 1982.56 54.14 ± 11.36 80.11 ± 12.63 37.44 ± 2.53 64.59 ± 8.28 88.25 ± 7.85 63.66 ± 7.71

7274.92 ± 2380.27 41.45 ± 17.30 82.57 ± 7.80 37.90 ± 3.33 57.95 ± 19.92 79.59 ± 20.11 59.93 ± 11.57

7126.88 ± 2265.00 35.07 ± 17.42 81.41 ±7.44 37.43 ± 3.15 54.65 ± 16.62 82.91 ± 8.45 59.04 ± 8.18

8194.50 ± 2303.04 37.58 ± 12.46 81.98 ± 7.60 37.04 ± 2.17 59.78 ± 7.99 82.74 ± 8.48 61.46 ± 8.99

Note. The data are shown as the mean ± standard deviation. LTAS = Leisure-time Activity Score; RHR = Resting Heart Rate; SE = Self-efficacy; OE = Outcome Expectations; SR = Self-regulation

activity, physical fitness, or SCT variables at pretest (all p > .05, see Table 3 for means and standard deviations). On average, 90%-95% of participants accessed to website, recorded physical activity, and set physical activity goals each week, and, 85% met physical activity goals. The attrition rate was about 21% for the intervention groups and 18% for the control groups. At the End of the Intervention A series of 2 (intervention or control) x 2 (pretest or no pretest) between-groups ANOVA tests showed no significant interactions for steps/day, LTAS, VO2max, resting heart rate, SE, OE, and SR (all p > .05). Therefore, pretest sensitization did not occur for any of the outcome measures (see Tables 3-4 for descriptive statistics). However, there was a significant pretest effect for SE (p < .05, eta = 0.03). The main effects of the intervention on steps/ day, LTAS, and resting heart rate were significant. The intervention group participants recorded more steps/day than those in the control (p < .01, eta = 0.31), greater LTAS (p < .01, eta = 0.13), and a lower resting heart rate (p < .05, eta = 0.026). No main effect of the intervention was observed for VO2max, (p > .05, eta = 0.001). For the SCT constructs, significant main effects of the intervention were found for SE (p < .01, eta = 0.12), OE (p < .01, eta = 0.032), and SR (p < .01, eta = 0.035), with participants in the intervention groups reporting higher SE, OE, and SR than those in the control groups. Because of concerns about limited power provided by main effects tests, Braver and Braver40 suggest further analysis is warranted for non-significant main effects. Thus, we tested the intervention effects in the 2 pre-post (I-P and C-P) groups with a 2-group analysis of covariance (ANCOVA) on the posttest scores for VO2max, while co-varying pretest scores. No significant effects of the intervention were observed for VO2max (p = .40). Independent-sample t tests comparing the 2 postonly (I-NP and C-NP) groups indicated no signifi-

Am J Health Behav.™ 2014;38(3):430-439

cant difference for VO2max (p = .86). Braver and Braver40 recommend a final analysis that takes full advantage of the power provided by the multiple groups included in this design. The probability level from the ANCOVA and t test for VO2max were each converted to a normal deviate (z) value, and then the resulting z’s were combined into a single zmeta which was then converted back into a p-value from which significance was determined. No significant p-value was found for VO2max (zmeta = 0.725, p = .468) indicating no effect on VO2max at the end of the intervention. At the 3-Month Follow-up A series of 2 (intervention or control) x 2 (pretest or no pretest) between-groups ANOVA tests showed no significant interactions for steps/day, LTAS, VO2max; resting heart rate, for SE; OE, and SR (all p > .05). Therefore, pretest sensitization did not occur for any of the outcome measures. However, there was a significant effect of the pretest condition for steps/day (p < .01, eta = 0.036). The main effect of the intervention on steps/day and LTAS was significant. Intervention group participants had more steps/day than those in the control groups (p < .01, eta = 0.35) and greater LTAS (p < .01, eta = 0.30). No main effects were observed for VO2max or resting heart rate (both p > .05). For the SCT constructs, significant main effects of the intervention were found for SE (p < .01, eta = 0.10), OE, (p < .01, eta = 0.068), and SR (p < .01, eta = 0.06), with participants in the intervention groups reporting higher SE, OE, and SR than the controls. The intervention effects in the 2 pretest groups were tested with a 2-group analysis of covariance (ANCOVA) on the follow-up scores for VO2max and resting heart rate, while co-varying pretest scores. No significant effects of the intervention were observed for VO2max (p = .36) or RHR (p = .61). Independent-sample t tests comparing the 2 no pretest groups indicated no significant difference between the groups for VO2max (p = .31), or resting heart

DOI:


435



Table 4 Mean Score of Outcome Variables

Steps LTAS RHR VO2max SE OE SR

Intervention with No Pretest

Control with No Pretest


Follow-up


11442.59 ± 4452.12 52.83 ± 24.73 78.14 ± 8.44 37.52 ± 4.81 59.99 ± 14.78 84.16 ± 9.00 64.45 ± 11.14

10601.18 ± 2883.21 59.50 ± 19.63 80.16 ± 5.02 37.51 ± 2.41 66.54 ± 7.33 85.33 ± 10.96 68.28 ± 11.11

7470.60 ± 2341.38 39.09 ± 16.11 80.38 ± 7.04 37.40 ± 1.43 50.14 ± 11.62 81.04 ± 12.64 60.15 ± 8.63

Follow-up 7570.25 ± 1380.78 39.32 ± 11.22 80.03 ± 7.91 37.09 ± 1.92 60.83 ± 8.14 80.82 ± 9.84 61.57 ± 7.54

Note. The data are shown as the mean ± standard deviation LTAS = Leisure-time Activity Score; RHR = Resting Heart Rate; SE = Self-efficacy; Outcome Expectations; SR = Selfregulation

rate (p = .92). With zmeta test, a non-significant pvalue was found for VO2max (zmeta = 1.36, p = .17) and (zmeta = 0.43, p = .67). Therefore, the intervention had no effect on VO2max or resting heart rate at the 3-month follow-up. DISCUSSION This is the first reported study using the Internet as a tool to promote physical activity in Thailand, and in particular, with Thai female students. Overall, the intervention was successful with large effect sizes. The SCT-based Internet intervention program was effective in promoting and maintaining leisure-time physical activity in Thai undergraduate female students. The intervention had significant effects on steps/day, LTAS, and SCT variables at the end of the intervention and the 3-month follow-up in the absence of pretest sensitization. Though the intervention successfully improved resting heart rate at the end of the intervention, this was not sustained at the 3-month follow-up and the intervention had no effect on VO2Max. This study used a true intention-to-treat analysis such that data on all participants randomly allocated to the intervention and the control groups were analyzed. Demographic data and dependent variables at the baseline were not significantly different between participants who completed the study and those who dropped out from the study. Thus, the outcomes at the end of the intervention and the follow-up might be similar for these participants. Effects on Physical Activity and SCT Variables The success of the intervention in changing behavior and SCT variables is consistent with previous studies with female university students13,52 and with review and meta-analysis studies that have reported the efficacy of Internet-based interventions in promoting physical activity.5,9,25,28 However, those reviews and meta-analyses reported

436

small effect sizes,5,9,28 whereas this study found large effect sizes. Thus, a SCT-based Internet intervention is a useful tool in the promotion and 3-month maintenance of leisure-time physical activity in Thai female students. Of note, this study found that steps/day of participants in the control group remained at the low-end of steps recommendations (ie, < 7500).53 The maintained changes after 3 months are similar to ones reported with Taiwanese female university students who maintained increased physical activity at 5 months post-intervention.52 Contrary to this, the physical activity of female university students in North America was not maintained after 4 months.13 Also, there is research with adult populations reporting that an Internet-based physical activity intervention was efficacious in the short-term but did not produce longer-term adherence to physical activity.5,28,54 However, participants in those studies13,54 were not contacted or did not have access to interactive features during the follow-up period, whereas participants in the present research received follow-up contact and access to interactive features. Marcus et al55 suggest that the lack of contact with participants during follow-up might play role in physical activity relapse. Thus, the physical activity relapse in previous studies may be because most Web-based programs are not interactive enough during the follow-up period to engage participants fully.54 Future studies should examine the influence of continued contact or website access post-intervention on the maintenance of physical activity. This research highlights possible cross-cultural considerations of using SCT and the Internet to promote physical activity. The influence of Thai culture may explain some of the maintenance of physical activity during the follow-up period. Cultural background and experiences can influence the self; moreover, the way that people live, think,


Sriramatr et al and behave are influenced by culture.56,57 People in Western countries are more likely to be independent (behavior is organized primarily by reference to one’s own thoughts, feelings, and actions) whereas people in Asian countries are more likely to be interdependent (behavior is determined by the perceived thoughts, feelings, and actions of others).56 The interdependent self of Thai students may have influenced physical activity maintenance, an idea corroborated by the low attrition rate (~20%) in this study compared to previous Western studies (~30%).5,28 It is also possible that the relationship between the self and SCT variables in Thai students may be stronger than for Western students. According to Bandura,20 SCT operates at interpersonal levels. SCT assumes humans are social beings who develop their sense of self and personal efficacy from others through interpersonal exchanges, and that the interpersonal environment is critical in affecting and predicting one’s health behavior and, in turn, health outcomes. Therefore, the interdependent self in Thai students may have a stronger influence on SCT variables than for Western students who may be more independent. However, the influence of culture on starting and maintaining physical activity was not examined in this study; thus, this suggestion needs verification and future studies should examine the specific influence of culture on the efficacy of SCT-based physical activity programs. Effects on VO2max and Resting Heart Rate Despite changes in physical activity and SCT variables, the intervention had no effect on VOmax. Resting heart rate was decreased at the 2 end of the intervention but the decrease was not maintained at follow-up. Previous research with adult women has also shown that although a 12week e-mail-based walking programs could improve 1-mile walk test time, it could not improve VO2max.58 It is known that physical activity participation is one important factor for determining cardiovascular fitness as measured by VO2max59 and many studies report a significant correlation between physical activity level and cardiovascular fitness.60,61 However, this association is stronger for higher intensity physical activity.62 Participants in the current research could participate in the physical activity of their own choosing and were asked to increase duration of physical activity by 3 minutes per week. The results showed that participants participated in an average of about 52 minutes or 12,160 steps per day during the intervention period and about 62 minutes or 14,080 steps per day during the follow-up period. Thus, although participants met our recommendations, it is likely that the physical activity may not have been of sufficient intensity to improve cardiovascular fitness. However, the increase in physical activity may still have influenced health outcomes, particularly in the area of metabolic fitness, an area for future research.63

Am J Health Behav.™ 2014;38(3):430-439

Pretest Sensitization Effects As found in previous studies,30,31 even though participants in the pretest groups were measured on a number of objective and self-report variables, completing the pretest measures did not influence the cognitions and behavior of participants. These results support the suggestion of Spence et al31 that the pretest sensitization effect is less relevant for experimental designs.31 Although there was a significant effect of the pretest condition for SE and steps, the lack of significant interactions between the pretest condition and the intervention condition suggested this effect was similar for both the intervention and the control groups. Strengths and Limitations This study has some significant methodological and theoretical strengths including the use of a Solomon 4-group design, and SCT and intervention mapping in developing the intervention program. However, the research was limited to female university students who may be more likely to have access to the Internet compared to other Thai populations. Thus, the results are not generalizable to other Thai population groups. Also, the Queen’s College Step Test was used to measure VO2max and accurate measurement of the carotid pulse rate is critical for valid testing. Future studies should confirm these findings with other indirect measures of VO2max and include a measurement of the intensity of the activity undertaken. Also, it is difficult to disentangle the effects of the self-monitoring, e-mail contact, and interaction, because a true control group was included that did not receive any treatments. CONCLUSIONS In conclusion, SCT and intervention mapping can be used as frameworks for developing physical activity interventions. A SCT-based Internet intervention program successfully promoted and maintained leisure-time physical activity in universityaged female students in Thailand, which is consistent with previous studies conducted in Western and Asian countries. The Internet is a potentially useful tool for delivering a SCT-based intervention physical activity program. However, the intervention should be tested in a variety of population over a longer period before wide scale implementation. Human Subjects Statement This study involved human participants and was reviewed and approved by the Research Ethics Boards of the University of Alberta and the Srinakharinwirot University prior to the study implementation. Conflict of Interest Statement No conflict of interest. Acknowledgements The first author acknowledges the Srinakharin-

DOI:


437


wirot University and the University of Alberta for financial support during his doctoral studies. REFERENCES


1. Haase A, Steptoe A, Sallis JF, Wardle J. Leisure-time physical activity in university students from 23 countries: associations with health beliefs, risk awareness, and national economic development. Prev Med. 2004;39(1):182-190. 2. Lau XC, Chong KH, Poh BK, Ismail, MN. Physical activity, fitness and the energy cost of activities: implications for obesity in children and adolescents in the tropics. Adv Food Nutr Res. 2013;70:49-101. 3. Baranowski T, Anderson C, Carmack C. Mediating variable framework in physical activity interventions: how are we doing? How might we do better? Am J Prev Med. 1998;15(4):266-297. 4. Biddle SJH, Hagger MS, Chatzisarantis NLD, Lippke S. Theoretical frameworks in exercise psychology. In Tenenbaum G, Eklund RC, (Eds). Handbook of Sport Psychology. New York: Wiley; 2007:537-559. 5. Vandelanotte C, Spathonis KM, Eakin EG, Owen N. Website-delivered physical activity interventions: a review of the literature. Am J Prev Med. 2007;33(1):54-64. 6. Webb TL, Joseph J, Yardley J, Michie S. Using the Internet to promote health behavior change: a systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. J Med Internet Res. 2010;17;12(1):e4. 7. Anderson ES, Wojcik JR, Winett RA, Williams DM. Social-cognitive determinants of physical activity: the influence of social support, self-efficacy, outcome expectations, and self-regulation among participants in a church-based health promotion study. Health Psychol. 2006;25(4):510-520. 8. Petosa RL, Suminski R, Hortz B. Predicting vigorous physical activity using social cognitive theory. Am J Health Behav. 2003;27(4):301-310. 9. Norman GJ, Zabinski MF, Adams MA, et al. A review of eHealth interventions for physical activity and dietary behavior change. Am J Prev Med. 2007;33(4):336-345. 10. Belanger-Gravel A, Godin G, Vezina-Im LA, et al. The effect of theory-based interventions on physical activity participation among overweight/obese individuals: a systematic review. Obes Rev. 2011;12(6):430-439. 11. Marcus BH, Williams DM, Dubbert PM, et al. Physical activity interventions: what we know and what we need to know. A scientific statement from the American Heart Association Council on nutrition, physical activity, and metabolism. Circulation. 2006;114(24):2739-2752. 12. Hallam JS, Petosa R. The long-term impact of a foursession work-site intervention on selected social cognitive theory variables linked to adult exercise adherence. Health Educ Behav. 2004;31(1):88-100. 13. Wadsworth DD, Hallam JS. Effect of a web site intervention on physical activity of college females. Am J Health Behav. 2010;34(1):60-69. 14. Harnirattisai T, Johnson RA. Effectiveness of a behavioral change intervention in Thai elders after knee replacement. Nurs Res. 2005;54(2):97-107. 15. Glanz K, Lewis FM, Rimer BK. Health Behavior and Health Education: Theory, Research, and Practice. San Francisco: Jossey-Bass; 2002. 16. Williams DM, Anderson ES, Winett RA. A review of the outcome expectations construct in physical activity research. Ann Behav Med. 2005;29(1):70-79. 17. Schnoll R, Zimmerman BJ. Self-regulation training enhances dietary self-efficacy and dietary fiber consumption. J Am Diet Assoc. 2001;101(9):1006-1011. 18. Gao Z, Xiang P, Lee AM, Harrison L Jr. Self-efficacy and outcome expectations in beginning weight training class:

438

their relations to students’ behavioral intention and actual behavior. Res Q Exerc Sport. 2008;79(1):92-100. 19. Stadler G, Oettingen G, Gollwitzer PM. Physical activity in women: effects of a self-regulation intervention. Am J Prev Med. 2009;36(1):29-34. 20. Bandura A. Self-efficacy: The Exercise of Control. New York: WH Freeman; 1997. 21. Poomsrikaew O, Berger BE, Kim MJ, Zerwic JJ. Age and gender differences in social-cognitive factors and exercise behavior among Thais. West J Nurs Res. 2012;34(2):245264. 22. Piaseu N, Schepp K, Belza B. Causal analysis of exercise and calcium intake behaviors for osteoporosis prevention among young women in Thailand. Health Care Women Int. 2002;23(4):364-376. 23. Internet World Stats. Usage and population statistics/ Asia stats/Thailand (Online). Available at: http://www. internetworldstats.com/stats3.htm. Accessed October 1, 2013. 24. Marshall AL, Owen N, Bauman AE. Mediated approaches for influencing physical activity: update of the evidence on mass media, print, telephone and website delivery of interventions. J Sci Med Sport. 2004;7(1 Suppl):74-80. 25. Marcus BH, Ciccolo JT, Sciamanna CN. Using electronic/computer interventions to promote physical activity. BJSM online. 2009;43(2):102-105. 26. Napolitano MA, Marcus BH. Targeting and tailoring physical activity information using print and information technologies. Exerc Sport Sci Rev. 2002;30:122-128. 27. Dunton GF, Robertson TP. A tailored Internet-plus-email intervention for increasing physical activity among ethnically-diverse women. Prev Med. 2008;47(6):605-611. 28. Davies CA, Spence JC, Vandelanotte C, et al. Meta-analysis of Internet-delivered interventions to increase physical activity levels. Int J Behav Nutr Phys Act. 2012;9:52. 29. Kitikannakorn N, Sitthiworanan C. Searching for health information on the Internet by undergraduate students in Phitsanulok, Thailand. Int J Adolesc Med Healt. 2009;21(3):313-318. 30. Spence JC, Burgess J, Rodgers W, Murray, T. Effect of pretesting on intentions and behavior: a pedometer and walking intervention. Psychol Health. 2009;24(7):777789. 31. van Sluijs EM, van Poppel MN, Twisk JW, van Mechelen W. Physical activity measurements affected participants’ behavior in a randomized controlled trial. J Clin Epidemiol. 2006;59(4):404-411. 32. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomized trials. Open Med. 2010;4(1):e60-68. 33. Thomas S, Reading J, Shephard RJ. Revision of the physical activity readiness questionnaire (PAR-Q). Can J Sport Sci. 1992;17(4):338-345. 34. Bartholomew LK, Parcel GS, Kok G. Intervention mapping: a process for developing theory-and evidencebased health education programs. Health Educ Behav. 1998;25(5):545-63. 35. Bartholomew K, Parcel G, Kok G, et al. IM: Developing Theory and Evidence-based Health Education Programs. Mountain View, CA: Mayfield; 2001. 36. Bartholomew LK, Parcel GS, Kok G, et al. Planning Health Promotion Programs: An Intervention Mapping Approach. San Francisco, CA: A Wiley Imprint; 2011. 37. Bandura A. Health promotion by social cognitive means. Health Educ Behav. 2004;31(2):143-164. 38. Campbell DT, Stanley JC. Experimental and Quasi-experimental Designs for Research. Chicago, IL: Rand-McNally; 1963. 39. Ogden J. Some problems with social cognition models: a pragmatic and conceptual analysis. Health Psychol. 2003;22(4):424-428. 40. Braver MC, Braver SL. Statistical treatment of the Solo-


Sriramatr et al mon four-group design: a meta-analytic approach. Psychol Bull. 1988;104(1):150-154. 41. Tremblay MS, Warburton DE, Janssen I, et al. New Canadian physical activity guidelines. Appl Physiol Nutr Metab. 2011;36(1):36-46. 42. Sriramatr S, Berry TR, Rodgers WM. Validity and reliability of Thai versions of questionnaires measuring leisure-time physical activity, exercise-related self-efficacy, outcome expectations and self-regulation. Pacific Rim Int J Nurs Res. 2013;17(3):203-216. 43. Godin G. The Godin-Shephard leisure-time physical activity questionnaire. Health and Fitness Journal of Canada. 2011;4(1):18-22. 44. Crouter SE, Schneider PL, Karabulut M, Bassett DRJr. Validity of 10 electronic pedometers for measuring steps, distance, and energy cost. Med Sci Sports Exerc. 2003;35(8):1455-1460. 45. McArdle WD, Katch FI, Pechar GS, et al. Reliability and interrelationships between maximal oxygen intake, physical work capacity and step-test scores in college women. Med Sci Sports. 1972;4(4):182-186. 46. Chatterjee S, Chatterjee P, Bandyopadhyay A. Validity of Queen’s College step test for estimation of maximum oxygen uptake in female students. Indian J Med Res. 2005;121(1):32-35. 47. Rodgers WM, Wilson PM, Hall CR, et al. Evidence for a multidimensional self-efficacy for exercise scale. Res Q Exerc Sport. 2008;79(2):222-234. 48. Dishman RK, Motl RW, Sallis JF, et al. Self-management strategies mediate self-efficacy and physical activity. Am J Prev Med. 2005;29(1):10-18. 49. Rovniak LS, Anderson ES, Winett RA, Stephens RS. Social cognitive determinants of physical activity in young adults: a prospective structural equation analysis. Ann Behav Med. 2002;24(2):49-56. 50. Polit DF, Gillespie BM. Intention-to-treat in randomized controlled trials: recommendations for a total trial strategy. Res Nurs Health. 2010;33(4):355-368. 51. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Erlbaum Associates; 1988. 52. Huang SJ, Hung WC, Chang M, Chang J. The effect of an Internet-based, stage-matched message intervention

on young Taiwanese women’s physical activity. J Health Commun. 2009;14(3):210-227. 53. Tudor-Locke C, Craig CL, Thyfault JP, Spence JC. Stepdefined sedentary lifestyle index:

An adaptive physical activity intervention for overweight adults: a randomized controlled trial.

A Lifestyle Physical Activity Intervention for Patients with Chronic Obstructive Pulmonary Disease. A Randomized Controlled Trial.

Reflective and impulsive processes explain (in)effectiveness of messages promoting physical activity: a randomized controlled trial.

Promoting Physical Activity With the Out of School Nutrition and Physical Activity (OSNAP) Initiative: A Cluster-Randomized Controlled Trial.

Changes in physical activity during a weight loss intervention and follow-up: a randomized controlled trial.

Effects of an educational intervention of physical activity for children and adolescents with cystic fibrosis: a randomized controlled trial.

Pasos Hacia La Salud: a randomized controlled trial of an internet-delivered physical activity intervention for Latinas.

Activity monitor intervention to promote physical activity of physicians-in-training: randomized controlled trial.

mActive: A Randomized Clinical Trial of an Automated mHealth Intervention for Physical Activity Promotion.

Intervention effects on adolescent physical activity in the multicomponent SPACE study: a cluster randomized controlled trial.

Efficacy and causal mechanism of an online social media intervention to increase physical activity: Results of a randomized controlled trial.

An Intervention to Increase Physical Activity in Children: A Randomized Controlled Trial With 4-Year-Olds in Preschools.

Maintaining physical activity during head and neck cancer treatment: Results of a pilot controlled trial.

Effect of Moderate to Vigorous Physical Activity Intervention on Improving Dementia Family Caregiver Physical Function: A Randomized Controlled Trial.

Application of objective physical activity measurement in an antenatal physical activity consultation intervention: a randomised controlled trial.

Radon balneotherapy and physical activity for osteoporosis prevention: a randomized, placebo-controlled intervention study.

Self-directed physical activity intervention in older adults undergoing adjuvant chemotherapy for colorectal cancer: Design of a randomized controlled trial.

Physical activity during pregnancy and maternal-child health (PAMELA): study protocol for a randomized controlled trial.

Long term effects of self-determination theory and motivational interviewing in a web-based physical activity intervention: randomized controlled trial.

Psychosocial telephone intervention for dementia caregivers: A randomized, controlled trial.

Bari-Active: a randomized controlled trial of a preoperative intervention to increase physical activity in bariatric surgery patients.

Effectiveness of a low-threshold physical activity intervention in residential aged care--results of a randomized controlled trial.

Comment on: Bari-Active: a randomized controlled trial of a preoperative intervention to increase physical activity in bariatric surgery patients.

A Randomized trial of an Asthma Internet Self-management Intervention (RAISIN): study protocol for a randomized controlled trial.