Live Research Spotlight 11: Physical Activity & Sedentary Behavior

Background: Many physical activity (PA) promotion campaigns have been developed and implemented to improve health and well-being. Historically, these campaigns have been evaluated with social cognitive approaches that are limited in explaining the PA intention-behaviour gap and distinguishing between behavioural adoption and maintenance. Habit or identity have been proposed as proxy indicators of PA maintenance. We proposed the development and testing of a simple algorithm that integrates action control and dual process theories with four self-reported items (PA intentions, behaviours, habit, and identity) to classify campaign audiences and target PA promotion content.
Methods: A profile algorithm was first developed on one dataset (n = 1,350, representative of a general Canadian population) and subsequently tested on two additional datasets (n = 376 and n = 6,396, both representative of populations involved in Canada-wide public education campaigns). Our a priori criterion for success was for the four profiles to individually account for at least 10% and collectively account for at least 85% of the sample. The four expected profiles were: 1) non-intenders with low habit and identity who were not meeting PA guidelines (i.e., “non-intenders”); 2) intenders with low habit and identity who were not meeting PA guidelines (i.e., “unsuccessful adopters”); 3) intenders with low habit and identity who were meeting PA guidelines (i.e., “successful adopters”); and 4) intenders with high habit and/or identity who were meeting PA guidelines (i.e., “successful maintainers”).
Results: The initial algorithm produced four meaningful profiles (“non-intenders”, “unsuccessful adopters”, “successful adopters”, “successful maintainers”) which collectively accounted for 87.6% of the sample. The two test samples revealed two profiles (“successful adopters”, “successful maintainers”) that together accounted for 69.8% and 70.9% of the samples, respectively. The “non-intenders” and “unsuccessful adopters” each accounted for 7.7-9.2% and 9.2% of the samples, respectively.
Conclusions: The algorithm produced four meaningful intention-behaviour profiles; however, only the “successful adopters” and “successful maintainers” were revealed as meaningful in the test samples. This may be a consequence of individuals from the test samples already being actively involved in PA promotion campaigns. Future research should assess profile distributions across different campaign audiences.

Presenter - Heather Hollman, MRSc,
University of Victoria

Co-Author - Leigh M. Vanderloo, PhD,
ParticipACTION, University of Western Ontario

Co-Author - Markus J. Duncan, PhD,
ParticipACTION

Co-Author - Guy Faulkner, PhD,
University of British Columbia

Co-Author - Eun-Young Lee, PhD,
Queen's University

Co-Author - Mark S. Tremblay, PhD,
Children's Hospital of Eastern Ontario Research Institute

Co-Author - John C. Spence, PhD,
University of Alberta

Co-Author - Ryan E. Rhodes, PhD, FSBM,
University of Victoria

Background: Emerging adulthood (18 and 26-years old) is critical for the development of long-term health behaviors, like physical activity (PA). Only 25% of emerging adults (EAs) meet the recommended 150 minutes per week of moderate to vigorous physical activity (MVPA). In rural communities, where residents have less access to health resources and healthcare, fewer than 20% of adults meet PA recommendations but EA data are limited. Further, data are lacking on PA mediators: motivation, intention, and planning in rural EAs. Therefore, the purpose of this study is to describe PA engagement, motivation, intention, and planning in a sample of rural EAs.
Methods: Participants were recruited via a nationwide survey panel. Inclusion criteria were: 18 to 26 years old, able to read and understand English, and live in a rural country as defined by the Rural Urban Commuting Area Codes 4-9 in Colorado, Utah, New Mexico, or Arizona. We assessed PA with the Global Physical Activity Questionnaire, Self-Determination Theory motivation with the Behavioral Regulations in Exercise Questionnaire v.3, and intention and planning with the Behavioral Intentions Scale. Participants reported minutes of MVPA accumulated through work and recreation, and we calculated total MVPA minutes. Using the Shapiro-Wilk test for normality, we found the MVPA variables were not normally distributed (p<.0001 for all) as a high number of participants reported zero MVPA minutes. We created binary variables of “Meets Recommendations” (≥ 150 minutes of MVPA per week) vs. “Does Not Meet Recommendations” (≤ 149 minutes of MVPA per week) for total and recreational MVPA.
Results: Our sample included 141 respondents; 115 (81.56%) female and 25 (17.73%) male, aged 22.00±2.63 years. With minutes of work MVPA included, 75 (53.19%) met MVPA recommendations whereas without, only 32 (22.70%) met recommendations. Participants reported low levels of amotivation (1.25±0.92), external (1.39±1.02), introjected (2.22 ±1.01), identified (2.43±0.73), and integrated regulation (1.82±0.80), and intrinsic motivation (2.29±0.86). They also reported low PA intention (4.21±1.61, 7-pt Likert scale) and planning (4.27±1.61, 7-pt Likert scale).
Discussion: More EAs met MVPA recommendations when work minutes were included. Rural EAs have low PA motivation, intention, and planning. Researchers should investigate barriers to PA in this group as they have high instances of health disparities and low access to health resources.

Author - Kayla Nuss, PhD,
Klein Buendel

Co-Author - Julia A. Berteletti,
Klein Buendel, Inc.

Co-Author - Amanda N. Brice,
Klein Buendel

Co-Author - Alishia A. Kinsey,
Klein Buendel, Inc.

Co-Author - Noah Chirico,
Klein Buendel

Co-Author - Sierra Held,
Klein Buendel

Purpose: Sedentary behavior (SB) is high in older adults. Whether SB reduction interventions have long term maintenance in older adult populations is unknown. We conducted a 12-month randomized controlled trial of sitting reduction in older adults and examined behavior change as well as blood pressure.
Methods: Participants (N = 283) were randomized to the sitting reduction intervention (I-STAND) or a healthy living control condition for 6-months and a sub-sample from both arms (N = 164) had 12-month follow-up data. The theory based I-STAND intervention included 10 health coaching contacts, feedback, setting goals to reduce sitting, standing desks, and fitness tracker prompts. The attention control group received health coaching on general healthy aging topics. At 6-months, I-STAND participants were re-randomized to 5 booster contacts or no further contact while control group participants were given no further contact and all were assessed at 12-months. activPAL accelerometers measured behavior change outcomes and blood pressure (BP) was taken at baseline and 3, 6, and 12-months. Differences between group mean changes from baseline were estimated using linear regression models with GEE adjusted for baseline outcomes, covariates (e.g. demographics, health conditions), and whether randomization occurred before or after pandemic onset.
Results: Participants (mean age = 68.3 (6.5) years, 65% women, 3% Asian, 14% Black, 77% White, mean BMI = 35 (4.8)) sat for 10.9 hrs/day at baseline. The booster group was not different from the no booster group. Combined, the I-STAND arms had significant reductions in sitting (-59 mins/day, p < 0.001) but not BP (-2.4mmHg, p=0.247) compared to the control group. The I-STAND arm had significantly more standing time (+51 mins/day, p < 0.001), stepping time (7 mins/day, p = 0.04), and shorter sitting bout duration (-2.5 mins/day, p = 0.003) and decreased time to complete 5 chair stands (-2.8 seconds, p = 0.01) compared to the control group.
Conclusion: Intervention participants were able to maintain reductions in sitting time over 12-months even without additional coaching contacts. They also were more active and had better physical function than the control group at follow-up. BP reduced but not significantly. Long-term sitting reduction could be a viable health promotion strategy in older populations.

Author - Dori Rosenberg, PhD, MPH, FSBM,
Kaiser Permanente Washington Health Research Institute

Co-Author - Lauren Fink,
Kaiser Permanente School of Medicine

Co-Author - Weiwei Zhu, MS,
Kaiser Permanente Washington Health Research Institute

Co-Author - Mikael Anne Greenwood-Hickman, MPH,
Kaiser Permanente Washington Health Research Institute

Co-Author - Andrea Cook, PhD, PhD,
Kaiser Permanente Washington Health Research Institute

Co-Author - David Arterburn, MD, MPH,
Kaiser Permanente Washington Health Research Institute

Co-Author - Jennifer McClure, PhD,
Kaiser Permanente Washington Health Research Institute

Co-Author - David Dunstan, PhD,
Baker Heart and Diabetes Institute

Co-Author - Neville Owen, PhD, FSBM, PhD, FSBM,
Baker Heart and Diabetes Institute

Co-Author - Beverly B. Green, MD, MPH, MD, MPH,
Kaiser Permanente Washington Health Research Institute

Background: Given low rates of physical activity (PA) among individuals with overweight/obesity, identifying differences between those who choose to be active versus inactive could provide intervention targets for increasing PA. The current study compared regular exercisers (≥150 min/wk of PA) and non-exercisers (<30 min/wk) on affective response to laboratory-based moderate-intensity PA, memory bias (e.g., discrepancy between remembered and actual affect), and anticipated affective response towards an upcoming PA session. An exploratory aim was to pilot a brief intervention for non-exercisers focused on improving exercise-related affect and memory bias.
Methods: A total of 59 weight-loss seeking individuals with overweight/obesity (Age=47.1 ±10.3; BMI= 32.1±3.3 kg/m2; 79.7% Female; 94.9% non-Hispanic; 91.5% White) completed two PA sessions in which they walked on a treadmill for 30 minutes at moderate intensity. Participants reported their anticipated affect prior to exercise via visual analogue scale (VAS) and momentary affect before, during, and after exercise using the Feeling Scale (-5 to +5). One, 3, and 7 days following this session participants reported their remembered affect via VAS. After PA session 2, non-exercisers were randomized and received an affect-based intervention (n=15) or exercise information (n=15) and completed a third PA session >7 days after the intervention.
Results: Remembered affect more closely resembled post-exercise affective response (versus ‘during’ exercise; p=0.001) for both exercisers and non-exercisers. Memory bias was greater among non-exercisers (p=0.04), such that non-exercisers remembered exercise less favorably. Anticipated ‘during’ exercise affect was more positive for exercisers than non-exercisers (Ex=71.3±18.7, Non-ex=54.1±21.3, p=0.002), with a similar trend observed for anticipated ‘post-exercise’ affect (Ex=81.0±15.3, Non-ex=73.5±16.4, p=0.073). Non-exercisers receiving the intervention exhibited significantly less memory bias than those randomized to the control (p=0.04).
Conclusions: Results suggest that reducing memory bias for PA may be an important intervention target. Maximizing positive affect towards the end of the PA session and immediately after activity concludes may most effectively influence remembered affect for a PA session. Finally, providing information about affective responses to PA and focusing on the most positive aspects of PA can reduce negative memory bias among non-exercisers.

Presenter - Katrina M. Oselinsky, PhD,
The Miriam Hospital and Warren Alpert Medical School of Brown University

Author - Kathryn E. Demos, PhD,
Brown University

Author - Shira Dunsiger, PhD,
Brown University, School of Public Health

Author - David M. Williams, PhD, FSBM,
Brown University

Author - Jessica L. Unick, PhD,
Brown University

Evidence highlights the importance of 24-hour movement behaviors (i.e., sleep, sedentary behavior, physical activity) in protecting against anxiety and depressive symptoms during adolescence. However, most studies are cross-sectional, rely on self-reported data, and/or use statistical methods that do not account for the interdependent nature of movement behavior data within a 24-hour day. This study examined prospective associations between device-assessed 24-hour movement behaviors and anxiety/depressive symptoms in adolescents using compositional data analysis. This study used data from three waves of the Adolescent Brain Cognitive Development (ABCD) Study. A broadly representative sample of 4,997 adolescents ages 10-13 years (49% female) wore a Fitbit Charge 2 for three weeks to capture sleep, sedentary behavior, light physical activity (LPA), and moderate-to-vigorous physical activity (MVPA) during the 2-year follow-up (2018-2020). Parents completed the Child Behavior Checklist during the 3-year (2019-2021) and 4-year (2020-2022) follow-up to assess anxiety/depressive symptom severity. Weighted compositional negative binomial regression and isotemporal substitution models were computed to examine associations between each movement behavior and anxiety/depressive symptom severity, adjusted for sociodemographic confounders. The average daily 24-hour movement composition for the sample consisted of 36.0% sleep, 40.9% sedentary behavior, 20.5% LPA, and 2.6% MVPA. A significant inverse association was observed between MVPA and anxiety/depressive symptom severity one-year (B = -0.09; 95% CI: -0.05, -0.13) and two-years later (B = -0.19; 95% CI: -0.15, -0.23). No significant associations were found for the other movement behaviors at both follow-up points. Reallocating 15 min of sleep, sedentary behavior, or LPA to MVPA was associated with significantly lower anxiety/depressive symptom severity at both follow-up points (all ps < .05). Engaging in MVPA appears to play a protective role in buffering against the development of anxiety/depressive symptoms during adolescence. Results align with the “every move counts” public health messaging, but emphasize the importance of engaging in higher intensity activities at the expense of other movement behaviors. Future research investigating which anxiety/depressive symptoms benefit most from MVPA is warranted to inform precision medicine interventions for adolescents during this vulnerable developmental stage.

Co-Author - Ethan Hunt, MPH, PhD,
The University of Texas Health Science Center at Houston, School of Public Health in Austin

Co-Author - Christopher D. Pfledderer, PhD,
The University of Texas Health Science Center at Houston, School of Public Health

Co-Author - Augusto César F. De Moraes, PhD,
The University of Texas Health Science Center at Houston, School of Public Health in Austin

Co-Author - Carah D. Porter, MS,
Kansas State University

Co-Author - Claire I. Groves, MS,
The University of Texas at San Antonio

Background: Physical activity (PA) promotion campaigns aspire to promote long-lasting PA behavior. Recently, our team developed an algorithm to determine the proportion of PA promotion campaign audiences who were enacting and/or maintaining their PA intentions. The algorithm uses four self-reported PA-related constructs – intentions, behaviors, habit, and identity – to determine four profiles: “non-intenders”, “unsuccessful adopters”, “successful adopters”, and “successful maintainers”. Characterizing these profiles by sociodemographic characteristics would inform more tailored and targeted PA promotion content. Consequently, the purpose of this study was to assess differences in sociodemographic characteristics across the algorithm profiles. Methods: We combined three datasets (total n = 8,122); one representative of the general Canadian population and the other two of Canada-wide public education campaign audiences. The PA algorithm was employed to determine the four profiles and odds ratios (OR) were calculated using logistic regression analyses with the following sociodemographic variables: age (18-64 compared to 65+ years), gender, sexual orientation, newcomer status (<10 years in Canada), race, education, and income. Results: The majority of the sample were “successful maintainers” (51.94%), followed by “successful adopters” (15.87%), “non-intenders” (10.83%), and “unsuccessful adopters” (9.35%). Compared to the “successful maintainers”, the remaining three profiles had higher odds of identifying as an adult aged 18-64 years (OR: 1.30-1.88), woman or gender diverse (except for “non-intenders”; OR: 1.21-1.24), 2SLGBTQIA+ (OR: 1.59-2.20), a newcomer to Canada (except for “successful adopters”; OR: 1.96-1.98), racialized (except for “successful adopters”; OR: 1.24-1.47), and reporting a disability (OR: 1.47-2.24) or low income (except for “unsuccessful adopters”; OR: 1.21-1.70). “Successful adopters” had higher odds than “successful maintainers” of reporting a post-secondary degree (OR: 1.19). Conclusions: Individuals were less likely to be translating and/or maintaining their PA intentions if they were younger (18 to 64 years), identified as women or gender diverse, 2SLGBTQIA+, racialized, or a newcomer to Canada; or if they reported a disability or low income. Future PA promotion campaigns should consider tailoring content to these underrepresented audiences with considerations of unique barriers and facilitators they may face.

Presenter - Heather Hollman, MRSc,
University of Victoria

Co-Author - Leigh M. Vanderloo, PhD,
ParticipACTION, University of Western Ontario

Co-Author - Markus J. Duncan, PhD,
ParticipACTION

Co-Author - Guy Faulkner, PhD,
University of British Columbia

Co-Author - Eun-Young Lee, PhD,
Queen's University

Co-Author - John C. Spence, PhD,
University of Alberta

Co-Author - Mark S. Tremblay, PhD,
Children’s Hospital of Eastern Ontario Research Institute

Co-Author - Ryan E. Rhodes, PhD, FSBM,
University of Victoria