Psychological mechanisms in a digital intervention to improve physical activity: A multicentre randomized controlled trial

• Published in: British journal of health psychology Vol. 23; no. 2; pp. 296 - 310
• Main Authors: Schwarzer, Ralf, Warner, Lisa, Fleig, Lena, Gholami, Maryam, Salvatore, Sergio, Cianferotti, Luisella, Ntzani, Evangelia, Roman‐Viñas, Blanca, Trichopoulou, Antonia, Brandi, Maria L.
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01-05-2018
• Subjects: Adult; Aged; Change agents; Clinical trials; digital intervention; Exercise; Exercise - psychology; Female; Follow-Up Studies; Greece; habit; Habits; Health Behavior; Health Promotion - methods; Humans; Individual differences; Internet; Intervention; Italy; Life Style; Lifestyles; Longitudinal Studies; Male; Middle Aged; Motivation; Physical activity; planning; Program Evaluation - methods; Psychological mechanisms; Self Report; self‐monitoring; Sex Factors; Spain; Time use; Women; Young Adult; Greece; Italy; Spain; planning; habit; digital intervention; self-monitoring; physical activity
• ISSN: 1359-107X; 2044-8287
• DOI: 10.1111/bjhp.12288

Objectives
The randomized controlled trial examined factors that might be responsible for individual differences in physical activity change among men and women who participated in a lifestyle intervention. The main purpose of the analyses regarded the role of psychological mechanisms involving motivation, planning, self‐monitoring, and habit strength.
Design
A two‐arm digital intervention was conducted in Italy, Spain, and Greece to improve physical activity levels, with follow‐ups at 3 and 6 months after baseline assessment.
Methods
Participants were 1,564 adults at baseline, n = 638 at 6‐month follow‐up. Linear mixed models examined the intervention effects, and a two‐group longitudinal structural equation model explored which psychological constructs (motivation, planning, self‐monitoring, habit strength) were associated with changes in physical activity.
Results
In addition to an overall increase in self‐reported activity, there were interactions between time and sex and between time and experimental groups, and a triple interaction between time, sex, and experimental groups, indicating that men reported an increase in activity independent of groups, whereas women in the active control group did not benefit from the intervention. Planning, self‐monitoring, and habit strength mediated sequentially between initial motivation and follow‐up physical activity.
Conclusions
Although the intervention produced overall improvements in physical activity, the time‐by‐treatment interaction emerged only for women. The mechanism included a sequence leading from motivation via planning, self‐monitoring, and habit strength towards physical activity.
Statement of contribution
What is already known on this subject?
Digital lifestyle interventions can be effective in terms of physical activity performance gains.
Men are on average more physically active than women.
Long‐term adherence rates to digital interventions are usually low.
What does this study add?
Giving users of an online platform more interactive options did not make a difference.
Women gained more than men from adaptive, dynamic online platform content.
Individual characteristics (motivation, planning, self‐monitoring, habit) were more important than online treatment features.