The effect of implementation intentions on physical activity among obese older adults: A randomised control study

Abstract

Objectives: Physical activity is a key factor for healthy ageing and obesity management. The aim of this study was to test the effect of implementation intentions (IIs) on physical activity and body weight among inactive obese older adults.

Design and measures: At baseline, 101 obese (classes I and II) older adults were randomised to an experimental or a control condition. IIs were delivered in the experimental condition in addition to the common intervention activities. Physical activity (primary outcome) was assessed by means of pedometers. The effect of IIs on body weight (secondary outcome) was also evaluated.

Results: No main effects for IIs were observed for all outcomes (ps > 0.21). A significant time × condition interaction was observed for the number of steps per day (p = 0.01). At the 6-month follow-up, the improvement in physical activity was greater among participants in the experimental condition (d = 0.59). None of the measured cognitions moderated or mediated IIs.

Conclusion: Results suggest that IIs could be appropriate to favour the maintenance of physical activity among inactive obese older adults. However, this technique seems to have limited impacts over and above common intervention activities on weight loss.

Keywords:

• physical activity
• implementation intentions
• obesity
• older adults
• weight loss

Acknowledgements

Ariane Bélanger-Gravel is supported by a doctoral award from the Canadian Institutes of Health Research. Paul Poirier is a senior clinical researcher of the Fonds de recherche en santé du Québec (FRSQ). We would like to express our gratitude to the study participants for their contribution to this work and to the staff for their dedicated work.

Notes

Notes

1. Participants underwent anthropometric measurement and a physical fitness test at a three-month follow-up meeting. However, due to logistic constraints, pedometers data were not obtained. Thus, this measurement period was not considered in the present statistical analyses.

2. Invalid days of recording correspond to data lost during the return of pedometers by postal mail or when pedometers were not wore by the participants (this latter information was confirmed with the participants log provided with the pedometer).

3. An intention-to-treat analysis was performed by means of the application of a multiple imputation approach. According to this analysis, the ‘time × condition’ interaction effect was no longer significant (p = 0.61), as well as the significant effect of time (p = 0.09), for which a statistical trend was observed following the intention-to-treat analysis.

4. The results of the physical fitness test revealed that participants increased their aerobic capacity during the study following a similar pattern of results than those observed for the pedometers steps per day. Indeed, a significant effect of time was observed (p < 0.0001), as well as a statistical trend for the time × condition interaction (p = 0.17).