Abstract
Background: Understanding socio-economic differences in the factors influencing physical activity among older adults is essential for developing comprehensive interventions. We aimed to quantify the associations of modifiable correlates and determinants on physical activity among older adults of lower versus higher socio-economic status in the United Kingdom.
Methods: In this systematic review and meta-analysis, we searched MEDLINE, Embase, Web of Science, Cochrane Central Register of Controlled Trials (CENTRAL), and Scopus from inception to December 2023, for peer-reviewed studies published in English, investigating associations between a modifiable factor as an independent variable and physical activity as a dependent variable, by socio-economic status (defined according to various area- and individual-level indicators, including neighbourhood deprivation, wealth or income, education, and occupational class), in samples of community-dwelling older adults aged 60+ years in the United Kingdom. Hospitalised and institutionalised populations were excluded. Random effects meta-analyses were performed separately for people of lower and higher socio-economic status. Risk of bias was assessed with the Mixed Methods Appraisal Tool. This study was registered with the International prospective register of systematic reviews (PROSPERO; CRD42022351708).
Results: Searches identified 11,472 references; seventy-seven studies met the selection criteria, of which fifty-one contributed to meta-analyses (N range = 134–29,280). Of the exposures positively associated with physical activity, physical function, social participation, and perception of general health had the largest effect sizes (standardised mean difference [SMD] range = 0.53–0.81; I2 range = 54.81–91.00%). Estimates were comparable among older adults of lower and higher socio-economic status, except for the presence of built physical activity facilities, access to walking and cycling infrastructure, and less smoking, which were positively associated with physical activity only among individuals of lower socio-economic status.
Conclusions: Our results suggest researchers need to better understand discrepancies in the prevalence of the assessed correlates (e.g., fewer participants of lower socio-economic status reported good physical function) to inform policies that reduce inequalities in older adults’ physical activity levels. However, most studies were cross-sectional. Future longitudinal and experimental research should gauge the suitability of these correlates as intervention targets.
Methods: In this systematic review and meta-analysis, we searched MEDLINE, Embase, Web of Science, Cochrane Central Register of Controlled Trials (CENTRAL), and Scopus from inception to December 2023, for peer-reviewed studies published in English, investigating associations between a modifiable factor as an independent variable and physical activity as a dependent variable, by socio-economic status (defined according to various area- and individual-level indicators, including neighbourhood deprivation, wealth or income, education, and occupational class), in samples of community-dwelling older adults aged 60+ years in the United Kingdom. Hospitalised and institutionalised populations were excluded. Random effects meta-analyses were performed separately for people of lower and higher socio-economic status. Risk of bias was assessed with the Mixed Methods Appraisal Tool. This study was registered with the International prospective register of systematic reviews (PROSPERO; CRD42022351708).
Results: Searches identified 11,472 references; seventy-seven studies met the selection criteria, of which fifty-one contributed to meta-analyses (N range = 134–29,280). Of the exposures positively associated with physical activity, physical function, social participation, and perception of general health had the largest effect sizes (standardised mean difference [SMD] range = 0.53–0.81; I2 range = 54.81–91.00%). Estimates were comparable among older adults of lower and higher socio-economic status, except for the presence of built physical activity facilities, access to walking and cycling infrastructure, and less smoking, which were positively associated with physical activity only among individuals of lower socio-economic status.
Conclusions: Our results suggest researchers need to better understand discrepancies in the prevalence of the assessed correlates (e.g., fewer participants of lower socio-economic status reported good physical function) to inform policies that reduce inequalities in older adults’ physical activity levels. However, most studies were cross-sectional. Future longitudinal and experimental research should gauge the suitability of these correlates as intervention targets.
| Original language | English |
|---|---|
| Article number | 83 |
| Journal | International Journal of Behavioral Nutrition and Physical Activity |
| Volume | 22 |
| Early online date | 23 Jun 2025 |
| DOIs | |
| Publication status | Published - 23 Jun 2025 |
Data Availability Statement
The datasets generated and analysed during the current study, analytic code for the meta-analyses and extraction of additional data from publicly available datasets, results of sensitivity and sub-group analyses, funnel plots, and Egger’s regression tests, as well as a complete record of independent variables, higher-order exposures, and studies that were included in this review but not meta-analysed are available in the GitHub repository, https://github.com/OliviaMalkowski/Meta-analysis-PA-SES. To adhere to statistical disclosure standards, all data cells pertaining to a sample size of 30 or below have been censored. These data are however available from the authors upon reasonable request and with permission of the data owners, as outlined in the data availability statements of the original studies included in the systematic review.Acknowledgements
We gratefully acknowledge the support of the following researchers in providing supplementary information or data towards this systematic review: Julii Brainard, Sabina M. De Geest, Jan Koetsenruijter, Kevin L. Lamb, Joreintje D. Mackenbach, Jenna Panter, Chrissy Roberts, Paschal Sheeran, and Christian von Wagner. We would also like to thank all researchers who took the time to respond to a request for supplementary information or data towards this systematic review, or a request for clarification on a study’s methods or participant sample, even if only to let us know that the respective study did not meet our inclusion criteria or that it was not possible to provide us with the supplementary information or data that we required. Collaborating authors in the Physical Activity Inequalities in Later Life consortium: Ahmad Alkhatib, Kyle J. Bourassa, Hannah Burnett, Mia Cajita, Simon Cook, Anne Corbett, Janie Corley, Simon R. Cox, Byron Creese, Katie Dudman, Joanne K. Garrett, Adam Hampshire, Coral L. Hanson, Melvyn Hillsdon, Jeroen Lakerveld, Agnieszka Lemanska, Laura Macdonald, Jane Maddock, Adilson Marques, Irene Mosca, David Ogilvie, Déborah Oliveira, Tytti P. Pasanen, Ilaria Pina, Alex V. Rowlands, Shaun Scholes, Lindsey G. Smith, Emma Solomon-Moore, Emmanuel Stamatakis, Yannick Stephan, Clare Stevinson, Tessa Strain, Cecilie Thøgersen-Ntoumani, and Carri WestgartFunding
| Funders | Funder number |
|---|---|
| Economic and Social Research Council | ES/P000630/1 |
