Purpose: Previous studies have indicated a relationship between the use of commercial physical activity apps (e.g., Fitbit, Strava) and physical activity engagement. Use of social components of such apps, in particular app-specific communities (connecting with other app users) and existing social media platforms (e.g., Facebook) have the potential to enhance physical activity. This study aimed to explore the psychological mechanisms underlying the relationship between the use of commercial physical activity apps (and their social components) and physical activity engagement.

Method: An online cross-sectional survey assessed physical activity, engagement with commercial physical activity apps (and their associated social components), and psychological constructs (social support, self-efficacy, motivation, trait competitiveness, trait social comparison). The SPSS macro PROCESS was used to conduct mediation (Model 4) and moderation (Model 1) analyses. Alpha was set at 0.05.

Results: Participants were 1206 adults aged 18-83 years (M_age = 34.0 ± 13.5 years, 88.5% female). App use was positively associated with physical activity engagement (p < .001, d = 0.40). The relationship between app use and physical activity was fully mediated by social support (b = 8.7, CI 3.3, 14.7), self-efficacy (b = 21.7, CI 14.2, 30.7), intrinsic motivation (b = 9.4, CI 1.4, 17.5) and identified regulation (b = 34.8, 2, CI 24.7, 48.5). Trait competitiveness (b = 25.3, p < .05) but not trait social comparison (b = 15.6, p = .372) moderated the relationship between app use and physical activity. In addition, the relationships between features of app-specific communities (providing/ receiving encouragement) and existing social media platforms (sharing posts, providing/ receiving encouragement and engagement in comparisons) influenced physical activity via social support, self-efficacy, and identified regulation.

Conclusions: The relationship between the use of commercial physical activity apps (and their social components) and physical activity is underpinned by social support, self-efficacy and motivation (intrinsic and identified). This highlights that commercial physical activity apps may be fundamental in promoting physical activity, given their capacity to influence psychological constructs associated with physical activity. Future research should further explore the use of commercial physical activity apps and their associated social components to increase physical activity engagement.

Purpose: For the first time in the literature we develop and validate behavioral and psychosocial text messages to be used in educational interventions to promote self-care in Brazilian patients with type 2 diabetes mellitus (T2DM).

Methods: The messages were developed based on the scientific literature, as well as on the guidelines of the Brazilian Diabetes Society and the Food Guide for the Brazilian Population. In addition, 35 patients with T2DM from a public health center in Brazil were consulted through discussion groups about their main barriers in managing T2DM. This consultation was approved by the Research Ethics Committee of the Federal University of Minas Gerais (UFMG-Brazil). In the validation process the Delphi technique was adopted. Six diabetes experts from differing areas in Health Sciences (nursing, nutrition, pharmacy, psychology, and odontology) and one linguist were selected to evaluate the messages that were developed. All experts had master's and/or doctorate degree and had been involved in a research project about T2DM in the last 5 years. The evaluation criteria used to classify the messages were relevance, intelligibility, and readability. Experts rated each message with grades assigned from 1 to 3 in an online questionnaire. A comment field was also available. The validation process took place in two rounds until the messages exceeded the minimum Content Validity Index (CVI) recommended by the literature.

Results: Based on the behavioral and psychosocial aspects of T2DM, the themes chosen for the developed messages were medication use, eating habits, physical activity, emotions, and perceived support (friends, family, and medical staff). 34 messages were developed of which 33 were validated. At the end of the second round, consensus was achieved on 91%.

Conclusions: 33 messages were validated and considered relevant, intelligible, and readable for patients with T2DM. Based on this, the authors recommend their use in educational interventions since these are reliable messages with high potential to promote self-care.

Purpose: Physical activity (PA) interventions are often limited in engagement strategies (e.g., social competition and collaboration, and effective feedback loops) that may be essential for successful behavioural change. These strategies are core elements of massive open online courses (MOOC) that are free and with unlimited participation. However, there have been few attempts to use MOOCs for the purpose of promoting health behaviours of learners. The aim of this study was to evaluate the effect of PA promotion delivered via a MOOC on concern about current levels of PA, readiness, importance and confidence to change learners’ PA behaviour.

Methods: The Sit less, get active MOOC consists of three weeks of core PA-related learning material, followed by weekly PA promotional messages and monthly PA promotional videos for six months. Learners who volunteered for the evaluation component completed PA-and health-related surveys: before the course started, upon the completion of the core course, and after 6 months during the time-period October 2016 and March 2018. The effect of the 3-week course on learners’ concern about their current levels of PA, and their readiness, importance and confidence to change PA behaviour was explored using repeated measures ANCOVA adjusted for age, sex, and time from baseline to completing the follow up questionnaire.

Results/findings: 530 learners (mean age±SD of 48±14 years, 78% women, 89% with college/university degree) had complete data on variables of interest at baseline and 3-week follow up. After completing the course the learners expressed less concern about their PA levels, and reported being more confident in success to increase their PA and to incorporate extra PA throughout the daily routine. They were also more likely to report doing enough PA to stay healthy and that they intend to do 30 minutes or more of moderate PA at least 5 times per week (p<0.001 for all).

Conclusions: The results indicate that PA promotion delivered via a MOOC improved confidence and readiness of learners to increase their PA level. MOOCs could help scale up PA promotion strategies; and could aid delivery and evaluation of PA promotion among health professionals, patients, work force, and general public.

Purpose: ​The rise of health apps created novel prospects for behaviour change. Despite the popularity of publicly available physical activity (PA) apps, little is known about their effectiveness. This ​study aimed to investigate the potential of the apps to increase PA. The objectives were: 1) to determine the feasibility and acceptability of the trial, 2) explore the effects of the apps on behavioural and psychological outcomes​

Methods: Feasibility randomised crossover trial of two popular apps. Inactive adults residing in London (UK) were eligible. The two interventions were: App A - 7 Minute Workout Challenge by Fitness Guide Inc., App B - One You Couch to 5k by Public Health England. After 1-week run-in period, participants were randomly allocated to one of the two assessment sequence (App A, B or App B, A). The primary outcomes were feasibility and acceptability of the trial measured using recruitment and retentions rates. Secondary outcomes included the effectiveness of the apps on objectively measured PA using accelerometer, self-reported PA, and psychological outcomes: exercise self-efficacy, intentions, PA outcome expectancy. These were assessed at baseline, 1 week post-baseline, 3- and 5-week follow-ups with interviews conducted following the completion of the trial. ​

Results:  The trial methods were feasible and acceptable to participants. Out of 104 participants who were eligible and consented, 63.5% (66/104) were enrolled and randomised. The primary analysis of the accelerometer data showed that there were no significant differences between baseline and the interventions using the continuous variables. Sixteen of 51 participants (31.4%) increased their time in moderate to vigorous PA (MVPA) by 20% from baseline following the introduction of the intervention (95% CI= 19.1% to 45.39). Self-reported PA outcomes showed significant increase and sedentary behaviour decreased. Exercise self-efficacy and intentions increased whilst PA outcome expectancy decreased. There were no significant differences between the two apps.

Conclusions: The recruitment and retention rates found in this crossover trial suggest that this design to study digital interventions was feasible and acceptable to the participants. The impact of this two PA apps showed promising results with an impact observed for 20% increase in MVPA self-reported PA, intentions and exercise self-efficacy.

Purpose:Web-based interventions that use computer-tailoring have shown to be effective, though people tend to ‘skim’ and ‘scan’ text on the Internet rather than thoroughly read it. The use of online videos is, however, popular and engaging. Therefore, the aim of this 3-group RCT was to examine whether using personally-tailored videos in a web-based physical activity intervention is more effective compared to traditional personally-tailored text and a control group.

Methods: 501 Australians were randomised into: text-tailored, video-tailored, or control. The control group only received access to web-based physical activity articles. Over a 3-month period, the intervention groups additionally received access to 8 sessions of personalised and theory-based (constructs from TPB, SCT, SDT) physical activity advice based on responses to online surveys. Only the web-based delivery method (either personalised-text or personalised-video) differed between intervention groups. Intervention groups were also able to complete action plans. The primary outcome was a 7-day measure of physical activity using waist-worn Actigraphs. Secondary outcomes include self-reported physical activity, meeting activity recommendations, steps, sitting time and website engagement. Outcomes were assessed at baseline, 3-, and 9-months. Data were analysed using generalized linear mixed models with intention-to-treat using multiple imputation. 

Results: Attrition was high with only 186 participants remaining at 9-months; 72% of participants were female, the average age was 44(±13). Actigraph measured physical activity (min/wk) significantly increased for the control group (at 3-months: 1.23(1.03-1.41),p=0.02; at 9-months: 1.33(1.09-1.62),p<0.001) and for the text-tailored group (9-months: 1.22(1.01-1.47),p=0.04), though no between group differences were found. Likewise, few between group differences were observed for secondary outcomes. Text-tailored participants took more steps compared to video-tailored participants (3-months: 0.88(0.79-0.97),p=0.01), self-reported more physical activity compared to control (3-months: 1.73(1.30-2.30),p<0.01) and were more likely to meet activity recommendations compared to control (3-months: 2.17(1.06-4.45),p=0.03). Video-tailored participants spent more time on the website compared to text-tailored participants (90 vs. 77 minutes, p=0.02).

Discussion: The lack of an intervention effect is in contrast with pilot and other study outcomes. Possible explanations include applying a more rigorous methodology, ceiling effects in psychosocial correlates, or intervention content not being perceived as engaging. Process evaluation and mediation analyses will provide further insights.

Purpose: Using pooled data from two randomized controlled trials (RCT) that applied the same intervention to improve physical activity and sleep behaviours, the purpose was to examine if a composite activity-sleep behaviour index (ASI) mediates the relationship between the intervention and symptoms of depression, anxiety, or stress, or quality of life in Australian adults.

Methods: REFRESH: Physically inactive adults (40–65 years) who reported poor sleep quality were recruited for a three-arm RCT (Physical Activity and Sleep Health (PAS; n=110), Sleep Health-Only (SO; n=110) or Wait-list Control (CON; n=55) groups). SYNERGY: Physically inactive adults (18–65 years) who reported poor sleep quality were recruited for a two-arm RCT ((PAS; n=80), or CON; n=80) groups). Assessments were conducted at baseline, 3 months (primary time-point), and 6 months. The PAS groups received a pedometer, and accessed a smartphone/tablet “app” using behaviour change strategies (e.g., self-monitoring, goal setting, action planning), with additional email/SMS support. Mental health outcomes were assessed using DASS-21, SF-12 and SF-36. The ASI score comprised self-reported moderate-to-vigorous-intensity physical activity, resistance training, sitting time, sleep duration, sleep efficiency and sleep quality. Data from the PAS and CON groups were pooled for mediation analysis (n=325), which was performed using SEM and the product-of-coefficients test, with bias-corrected confidence intervals (p<0.05).

Results/findings: At 3 months, compared with CON, the PAS group showed significant improvements in ASI scores, and also significant total effects on stress ((b, p-value) -1.27, p=0.041), and energy and fatigue (3.99, p=0.027); but not depression (-1.11, p=0.113), anxiety (-0.70, p=0.107), QOL-physical health (0.72, p=0.405), or QOL-mental health (2.15, p=0.058). At 3 months, improved ASI scores were associated with improvements in depression (-0.16, p=0.002), anxiety (-0.09, p=0.009), stress (-0.21, p<0.001), QOL-mental health (0.36, p<0.001), and energy and fatigue (0.55, p<0.001). Improved ASI scores mediated statistically significant proportions of the intervention effects on stress (57%) and energy and fatigue (48%). Associations at 6 months were attenuated but remained statistically significant and followed a similar pattern.

Conclusions: The intervention significantly improved overall physical activity and sleep behaviours in adults, and these improvements significantly mediated the intervention effect on stress and ratings of energy and fatigue.

Purpose: Developing novel treatment approaches and treatment combinations for increasing moderate-to-vigorous physical activity (MVPA) is imperative for preventing chronic diseases. Adaptive goal setting and the use of micro-incentives are relatively novel approaches to increase MVPA. This presentation reports on the main outcomes of the WalkIT Arizona trial testing adaptive goals combined with financial rewards for increasing adults’ MVPA. We hypothesized stronger improvements for MVPA among adaptive-goal (versus static) setting and immediate (versus delayed) reinforcement groups over 12 months.

Methods: Insufficiently-active adults (N=518, 64.3% female, mean age=45.3±9.2 yrs.) from Phoenix, Arizona, USA were randomized into one of four treatments to compare goal tye (adaptive vs. static) and financial reward type (immediate vs. delayed non-contingent) for promoting MVPA in a 2x2 factorial trial. Participants wore an ActiGraph GT9X daily to assess MVPA bout-mins, yielding 135,190 total valid wear days. Generalized linear mixed models tested for treatment group x phase (time) interactions, adjusting for accelerometer wear time and design variables. Due to zero-inflated MVPA bout-min values, we examined baseline-to-intervention phase changes in (a) the odds of engaging at least one bout of MVPA on each day and (b) on days with at least one bout, how the daily total of MVPA bout mins differed across goal and reward type conditions (i.e., Group x Phase interactions).

Results: Among static-goal participants, the odds of engaging any bouts during the intervention phase was greater than during the baseline phase (i.e., OR=1.92). Among adaptive-goal participants, the intervention vs. baseline difference was significantly stronger (OR=2.74, p<.001 for Goal x Phase interaction). For the immediate reward group, the intervention vs. baseline difference in odds of MVPA was stronger than in the delayed reward group (OR=2.51 vs. OR=2.10; p=.027 for interaction). On days with at least one MVPA bout, the immediate reward group had 66% more MVPA minutes during the intervention than baseline, while the delayed reward group had 44% more (p<.001 for interaction). No other interactions were significant.

Conclusions: Adaptive goals outperformed static goals, and immediate rewards outperformed delayed-non-contingent financial rewards for MVPA adoption.