Article Text

Download PDFPDF

Disability and capability
Physical capability and subsequent mortality: a systematic review and meta-analysis of published data
  1. R. Cooper,
  2. D. Kuh,
  3. R. Hardy
  1. MRC Unit for Lifelong Health and Ageing, Department of Epidemiology and Public Health, Royal Free and University College Medical School, London, UK

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


    Evidence suggests that measures of physical capability may be predictive of subsequent disability, health and mortality, but the results from existing published studies have not been synthesised.


    To review published literature examining the associations between physical capability and subsequent all-cause mortality.


    Systematic review and meta-analysis.


    Medline (from 1950 to 2008/05), EMBASE (from 1980 to 2008/02) and manual searches of the reference lists of eligible papers.

    Included Studies

    Any study examining the association between at least one of the specified measures of physical capability (grip strength, chair rises, standing balance, walking speed/time) and mortality in a population who were healthy and community-dwelling at baseline.


    Of 2240 unique records screened, 20 papers met the inclusion criteria of the review. An additional 3 papers were identified from reference lists. Within the 23 papers, there were 17 published analyses of grip strength, 9 of walking time/speed, 2 of standing balance and 1 of chair rises. The mean age of participants ranged from 36 to 80 y but the majority of study populations were aged 60 y and above. Of the papers investigating the association between grip strength and mortality, 6 presented Hazard Ratios for a unit change in grip strength with basic adjustments (total n = 10 171, deaths = 4267). The pooled Hazard Ratio for mortality per 5 kg increase in grip strength, from a random effects meta-analysis, was 0.90 (95% CI 0.87 to 0.93). Heterogeneity between studies (I2 = 65.2%) was largely explained by a single study which had weaker results. The pooled HR when this study was removed was 0.89 (0.86 to 0.92) (I2 = 25.7%). Other papers, where comparisons between categories of grip strength had been performed, also generally found weaker grip strength to be associated with higher rates of mortality but these analyses were too heterogeneous to be combined using meta-analysis. Due to differences in the ways in which walking speed had been measured and analysed it was not possible to perform meta-analyses of these data, but the results from all 9 studies showed that slower walking speed/longer walking time was significantly associated with increased mortality risk.


    This review demonstrates that grip strength and walking time/speed are useful markers of subsequent mortality risk in community-dwelling populations. The underlying explanations of these associations needs to be elucidated and further research is necessary to assess whether other measures of physical capability, including chair rises and balance, have similar predictive value.