Context: The transition to retirement has been recognized as a turning point in determining physical activity and may present a critical "window" for promoting physical activity. This systematic review examined changes in physical activity across the retirement transition, whether these changes vary by SES, and what is known about predictors of these changes.
Evidence acquisition: Peer-reviewed articles and gray research literature, published between January 1980 and July 2010 in any country or language, were identified. Longitudinal and cross-sectional observational studies were included. Study selection, quality assessment, data extraction, and synthesis were performed between July 2010 and March 2011. A harvest plot approach to visualizing the findings was combined with a narrative synthesis.
Evidence synthesis: Of the 19 included studies, 11 examined changes in exercise, or leisure-time physical activity, or both; seven, changes in total physical activity; and one study, both. Most studies used single-item measures of physical activity (n=9) or custom questionnaires (n=6). Results suggested that exercise and leisure-time physical activity increased after the retirement transition, whereas findings regarding total physical activity were inconsistent. SES moderated the association, with low SES being associated with a decrease and high SES with an increase in physical activity. Evidence on predictors of change was scarce and methodologically weak.
Conclusions: Evidence suggests that exercise and leisure-time physical activity increases after the retirement transition, but whether and how total physical activity changes is unclear. Imprecise physical activity measures used in primary studies limit conclusions, and this highlights the need for further research.
Copyright © 2012 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.