Abstract

--- - Background The focus of most epidemiological studies has been mortality or clinical events, with less information on activity limitations related to basic daily functions and their consequences. Standardised data from multiple countries at different economic levels in different regions of the world on activity limitations and their associations with clinical outcomes are sparse. We aimed to quantify the prevalence of activity limitations and use of assistive devices and the association of limitations with adverse outcomes in 25 countries grouped by different economic levels. Methods In this analysis, we obtained data from individuals in 25 high-income, middle-income, and low-income countries from the Prospective Urban Rural Epidemiological (PURE) study (175 660 participants). In the PURE study, individuals aged 35-70 years who intended to continue living in their current home for a further 4 years were invited to complete a questionnaire on activity limitations. Participant follow-up was planned once every 3 years either by telephone or in person. The activity limitation screen consisted of questions on self-reported difficulty with walking, grasping, bending, seeing close, seeing far, speaking, hearing, and use of assistive devices (gait, vision, and hearing aids). We estimated crude prevalence of self-reported activity limitations and use of assistive devices, and prevalence standardised by age and sex. We used logistic regression to additionally adjust prevalence for education and socioeconomic factors and to estimate the probability of activity limitations and assistive devices by age, sex, and country income. We used Cox frailty models to evaluate the association between each activity limitation with mortality and clinical events (cardiovascular disease, heart failure, pneumonia, falls, and cancer). The PURE study is registered with ClinicalTrials.gov, NCT03225586. Findings Between Jan 12, 2001, and May 6, 2019, 175 584 individuals completed at least one question on the activity limitation questionnaire (mean age 506 years [SD 98]; 103 625 [59%] women). Of the individuals who completed all questions, mean follow-up was 107 years (SD 44). The most common self-reported activity limitations were difficulty with bending (23 921 [136%] of 175 515 participants), seeing close (22 532 [134%] of 167 801 participants), and walking (22 805 [130%] of 175 554 participants); prevalence of limitations was higher with older age and among women. The prevalence of all limitations standardised by age and sex, with the exception of hearing, was highest in low-income countries and middle-income countries, and this remained consistent after adjustment for socioeconomic factors. The use of gait, visual, and hearing aids was lowest in low-income countries and middle-income countries, particularly among women. The prevalence of seeing close limitation was four times higher (6257 [165%] of 37 926 participants vs 717 [40%] of 18 039 participants) and the prevalence of seeing far limitation was five times higher (4003 [106%] of 37 923 participants vs 391 [22%] 2 2%] of 18 038 participants) in low-income countries than in high- income countries, but the prevalence of glasses use in low-income countries was half that in high-income countries. - Walking limitation was most strongly associated with mortality (adjusted hazard ratio 132 [95% CI 125-139]) and most consistently associated with other clinical events, with other notable associations observed between seeing far limitation and mortality, grasping limitation and cardiovascular disease, bending limitation and falls, and between speaking limitation and stroke. Interpretation The global prevalence of activity limitations is substantially higher in women than men and in lowincome countries and middle-income countries compared with high-income countries, coupled with a much lower use of gait, visual, and hearing aids. Strategies are needed to prevent and mitigate activity limitations globally, with particular emphasis on low-income countries and women.