Physical capability predicts mortality in late mid-life as well as in old age: Findings from a large British cohort study

Victoria L Keevil,Robert Luben,Shabina Hayat,Avan A Sayer,Nicholas J Wareham,Kay-Tee Khaw

doi:10.1016/j.archger.2017.10.001

Abstract

IntroductionLow physical capability predicts mortality, perhaps by association with co-morbidity. However, few studies include participants <70years old with lower co-morbidity burdens compared to older adults. We examined relationships between usual walking speed (UWS), timed chair stands speed, grip strength, standing balance and all-cause mortality in 8477 participants aged 48–92years enrolled in the European Prospective Investigation of Cancer-Norfolk study. MethodsParticipants (55.1% female) were followed up for 6.0 years (inter-quartile range 4.6, 7.5). Associations were examined using Cox proportional hazards regression by age-group (<70years versus ≥70years) and then in the whole cohort adjusted for age, sex, anthropometry, history of diabetes/stroke/myocardial infarction/cancer, smoking, alcohol intake, socioeconomic status, television viewing time and physical activity. ResultsAge and sex adjusted associations were similar in younger and older participants (Pinteraction all >0.05) and those with lower physical capability had higher mortality risk. For example, in those <70years old hazard ratios (95% confidence interval) for mortality in the third, second and lowest sex-specific quartiles of UWS compared to the highest were 1.21 (0.75, 1.96), 2.11 (1.35, 3.28) and 2.91 (1.84, 4.62) and in participants ≥70years old were 1.19 (0.73, 1.95), 2.09 (1.35, 3.24) and 2.64 (1.73, 4.02) respectively. In the whole cohort, strong associations between all physical capability tests and mortality persisted after multivariable adjustment and after excluding participants with co-morbidity. ConclusionsPhysical capability was independently predictive of future mortality risk with similar associations in late mid-life, when co-morbidity burden is lower, as at older age.

Highlights

Low physical capability predicts mortality, perhaps by association with co-morbidity
The ability to carry out everyday activities, can be objectively measured using simple tests such as grip strength (GS), timed chair stands speed (TCSS), usual walking speed (UWS) and standing balance (SB)
Exploring a range of physical capability measures (Leong et al, 2015; Ortega & Silventoinen, 2012; Rantanen et al, 2012). Emerging evidence from these studies suggests that the association is weaker in younger adults (Cooper et al, 2010) and that there may be a threshold effect, with only the very lowest performers experiencing increased risk of mortality (Cooper et al, 2014; Elbaz et al, 2012; Katzmarzyk & Craig, 2002; Ortega & Silventoinen, 2012), rather than the linear dose-response relationship described in older adults

Summary

Introduction

Low physical capability predicts mortality, perhaps by association with co-morbidity. These studies, similar to those included in the previous meta-analysis (Cooper et al, 2010), under-represent women (Elbaz et al, 2012; Ortega & Silventoinen, 2012; Rantanen et al, 2012) and often only evaluate associations between mortality and grip strength rather than exploring a range of physical capability measures (Leong et al, 2015; Ortega & Silventoinen, 2012; Rantanen et al, 2012) Emerging evidence from these studies suggests that the association is weaker in younger adults (Cooper et al, 2010) and that there may be a threshold effect, with only the very lowest performers experiencing increased risk of mortality (Cooper et al, 2014; Elbaz et al, 2012; Katzmarzyk & Craig, 2002; Ortega & Silventoinen, 2012), rather than the linear dose-response relationship described in older adults. None of the studies included were from the United Kingdom and only 2 included adults with a mean age of < 70 years

Methods

Results

Conclusion