Purpose: Adolescents’ diet is frequently suboptimal which has a triple impact on their immediate health, future health, and the health of the next generation. Interventions that focus on adolescents’ values have shown a positive impact on motivation for health behaviour change. However, very few studies have explored underlying motivations of adolescents’ food choices. This study aimed to explore what factors motivate adolescents’ food choices, what is the perceived impact of each factor in relation to the others, and how these factors relate to each other.

Methods: Adolescents age 12 to 17 years old participated in 26 semi-structured individual qualitative interviews to understand their motivation for food choices. During the individual interviews, a rudimentary conceptual map was created with each participant. The individual maps were then combined into age group specific maps (12-13, 14-15 and 16-17 years-old). Focus groups were then conducted to explore relevance of the different factors and connections between them. Data analysis is ongoing using grounded theory and the maps will be supplemented with data from individual interviews.  

Results/findings: Preliminary analysis show that parents (what they eat and cook), friends (what they eat), convenience (quick, easy, nearby and time it takes to prepare) and what they like/want were key factors that influence food choices. During individual interviews, environmental concerns (climate change and global warming) and social media (influencers and their diets/bodies) were frequently mentioned but were discussed to a lesser extent in the group interviews. Adolescents associated cravings with both their likes/wants and the convenience of what they are craving.

Conclusions: The use of conceptual maps allows to increase young people to voice their own experience of choosing food for themselves. Finding the influences of adolescent’s food choices and how they relate to each other has the potential to design more engaging and effective interventions to improve diet in adolescents.

Purpose: Sport can provide many health benefits to university students, yet participation rates in transitioning students tend to decline. The Theory of Planned Behavior (TPB; Ajzen, 1991) has been successfully adopted to understand participation in many health behaviors, such as sport. Despite its utility, the inclusion of past behavior has been shown to have significant effects on social cognitive constructs and future behavior. Such effects have been suggested to represent automatic, habitual processes. The purpose of the study was to understand the relationship between social cognition constructs and past behavior relating to sports participation in transitioning university students.

Methods: A prospective correlation design was used with two waves of data collection. First-year undergraduate students (N = 286) completed assessments of TPB constructs and past behavior at Time 1. Four-weeks later at Time 2 participants reported their participation in sport. Two structural equation models were conducted; Model 1 tested the influence of TPB constructs on behavior and Model 2 included past behavior.

Results: Model 1 accounted for 59% of the variance in intention and 42% in behavior, which increased to 68% and 43%, respectively, in Model 2. Model 1 demonstrated all three antecedents of intention to be significant and intention to predict behavior. Intention also mediated the effects of attitude, subjective norm, and perceived behavioral control on behavior. Model 2 showed similar direct and indirect effects. Moreover, past behavior had a total effect on behavior and a direct effect on all TPB constructs. The effects of past behavior on intention and behavior were mediated by TPB constructs and there was no direct effect from past behavior to behavior.

Conclusions: The study found the TPB to explain transitioning university students’ participation in sport. The study also showed the effects of past behavior to be mediated through social cognition constructs. Study findings suggest interventions should focus on the conscious, deliberative factors underlying sports participation rather than habitual, automatic factors. This could be due to the unstable environments transitioning students navigate when starting university.

Purpose: According to the WHO, increasing physical activity levels should be a priority, worldwide. Many applications (Apps) in the market offer asynchronous training plans which could potentially help lead more active lifestyles. However, attrition rates to training apps tend to be high and behavioural change remains a challenge. The goal of this project is to design a computerized predictive model to help identify App user behaviour, before it occurs. Eventually, motivation actions will be undertaken to promote adherence to healthy patterns.

Methods: This was an observational, retrospective pilot study to assess adherence and usage patterns of Mammoth Hunters, an App that personalizes exercise plans with the aim of breaking the barriers of lack of time, knowledge or motivation to increase activity. Data on demography and App usage were collected (n= 777). Correlations between the available data were extracted by using three clustering models: K-means, Balanced Iterative Reducing and Clustering Using Hierarchies (BIRCH) and Agglomerative Clustering. The clusters were validated using the Silhouette score and Calinski-Harabasz Index.

Results: With demographic details and usage data we extracted the patterns and factors related to adherence in App users. Findings were consistent across all of the applied clustering models. Three similar clusters based on user BMI and the mean workout frequency per month were identified, with a silhouette score and Calinski-Harabasz Index of 0.27 and 264 respectively for K-means clustering, 0.24 and 210 respectively for BIRCH clustering, and 0.23 and 212 respectively for the Agglomerative clustering. Additionally, a motivational questionnaire was validated in a subsample of 222 users.

Conclusions: Mammoth Hunters App users have been clustered according to their demographic details and usage habits. We will next study the correlations between their motivation (intrinsic vs extrinsic) and their levels of engagement to their exercise plan, by using artificial intelligence and machine learning. This will enable us to identify user patterns and predict their behaviour. By doing so, we plan to develop a series of motivational interventions to promote user adherence to exercise and reduce App attrition.

Purpose:Preventing weight regain after weight loss is a major challenge that could partly be addressed by identifying predictors of long-term weight management.This study aimed to examine, in a hierarchical fashion, self-regulatory and motivational predictors of 12- and 18-month successful weight loss maintenance (WLM) in adults who achieved clinically significant (≥5%)weight loss. 

Methods:Data reported here is from the NoHoW 2x2factorial randomized controlled trial, which tested the efficacy of a theory-based digital toolkit for WLM. A total of 1263 and 1180 participants (68.4% women; 45±11.7y; 29.4±5.0kg/m² at study entry) completed the 12-month and 18-month assessments, respectively. Signal detection analysis identified self-regulatory/motivational variables that best predicted <3% weight change (weight loss plus weight maintenance) vs. ≥3% weight change (weight regain), from baseline to 12-month and 18-month measurement. A set of 25 self-regulatory/motivational predictor variables (6-month changes from baseline) plus study arm and compliance, were examined. To compare the self-regulatory/motivational profile of the most and least successful subgroups emerging from signal detection analysis, independent-sample t tests, corrected for multiple testing using the Bonferroni’s procedure, were used.

Results/findings:Participants with higher changes in self-monitoring, self-efficacy for WLM and competence need satisfaction were more likely to achieve success at 12 months (83.3%;p<0.001). Higher changes in integrated regulation for eating behaviors proved somewhat compensatory for individuals with lower changes in self-efficacy for WLM (70.6% vs. 53.4% for individuals with lower changes in integrated regulation;p<0.01). Similarly, individuals with higher changes in self-monitoring and competence need satisfaction and lower changes in amotivation for healthy eating behaviors were significantly more likely to achieve <3% weight change at 18 months (74.7%;p<0.01). Participants with lower changes in self-monitoring skills, higher changes in introjected regulation and lower changes in integrated regulation for eating behaviors were the least successful at 18 months (only 30% achieved success). The most successful subgroups at both 12 and 18 months showed a significantly more positive profile concerning the majority of self-regulatory/motivational predictor variables.

Conclusions:Improving self-monitoring skills, self-efficacy for WLM and promoting competence need satisfaction are promising and priority intervention targets in order to promote clinically significant WLM.

Purpose. Recent definitions propose a distinction between ‘habit’ – a psychological process whereby exposure to cues triggers learned cue-behaviour associations, which in turn activate a non-conscious impulse to act – and ‘habitual behaviour’, which denotes action brought about by the habit process. This distinction has important implications for weight loss maintenance, as it allows for the possibility that people may successfully discontinue unwanted habitual behaviour – by, for example, avoiding cues, or wilfully inhibiting unwanted habit impulses – yet retain the underlying habit associations that generate such impulses. In such instances, these ‘dormant habits’ – i.e. learned, automatic habit associations that have not been acted upon for some time due to avoidance of cues or wilful inhibition – may later re-emerge, derailing behaviour change attempts. People trying to maintain weight loss may thus relapse into old patterns of behaviour upon exposure to certain cues, or when their willpower is depleted. This paper argues that dormant habits pose a risk to weight loss maintenance but have been overlooked as intervention targets.

Methods: Two literature reviews were undertaken, by locating existing reviews and coding eligible studies from those reviews. Review 1 identified qualitative studies of post-intervention experiences of weight loss maintenance, which were subsequently coded for key indicators of the ongoing influence of dormant habits among study participants. Review 2 described the behaviour change content of weight loss interventions, and specifically, the presence or absence of six techniques conducive to breaking habit associations (e.g. habit reversal, behavioural substitution).

Results/findings: In Review 1, 18 (69%) of 26 studies reported some evidence of participants experiencing unwanted, environmentally-dependent and automatic tendencies to engage in behaviours antithetical to weight loss, suggesting that many were struggling to overcome dormant habits. In Review 2, only five (4%) of 130 interventions used techniques suited to breaking dormant habit associations.

Conclusions: Interventions largely fail to acknowledge that people trying to maintain weight loss may experience the continued influence of unwanted dormant habits. Intervention developers should adopt strategies that not only inhibit engagement in unwanted habitual behaviours, but also break underlying cue-response associations that may undermine weight loss maintenance.

 

Strong evidence demonstrates that more time spent in sedentary behaviour (SB) is related to greater all-cause mortality, cardiovascular disease mortality, type-2 diabetes, and cancer. Current evidence supports the use of multi-level interventions developed using participative approaches targeted at specific at-risk subgroups. This study aimed to establish the acceptability and feasibility of a multicomponent theory-led intervention incorporating behaviour change techniques to reduce workplace sedentary behaviour and increase physical activity

Methods

A pilot feasibility study using a cluster randomised-controlled crossover design was used. The intervention targeted factors at the individual (mHealth using a Garmin watch), the environment (provision of an under-desk pedal machine), and the organisational levels (management recruitment to the study), that influence occupational SB. The primary outcomes of the study were acceptability and feasibility of assessments, study procedures and processes from an employee and management perspective (focus groups, semi-structured interview); recruitment and retention; and a qualitative evaluation of participants’ perspectives of the intervention overall. Secondary outcomes included SB and PA (objective (ActivPAL) and subjective (EMA)), and work engagement (UWES-9).

Results

The results showed that the intervention, as well as the trial processes were acceptable and feasible to conduct (n=21). The main intervention benefit was an increase of awareness of the dangers of SB through participation in the study, at both management and staff level, as well as increased productivity from a management perspective. The main intervention barriers were time priorities and the lack of assistance with the ergonomic set up of the pedal machine. Recruitment rate was 40% at cluster level, and 80% at an individual level; retention was 95% overall from baseline to post-intervention (8-weeks). Mean cycling time was 27 minutes/day (SD 10.23) in the intervention period. Workplace SB reduced by 20.4 minutes/day, and total weekday SB reduced by 45.7 minutes/day.

Conclusion

This study demonstrated the acceptability of a multicomponent intervention to reduce workplace SB in professional men. The implications of such an intervention on SB and PA requires further research, but the present evidence suggests that it is possible to reduce workplace SB using an under-desk pedal machine and mHealth by increasing cycling time.