What physical performance measures predict incident cognitive decline among intact older adults? A 4.4 year follow up study

Highlights

• Few studies assessed if low physical performance predicts poor cognitive status.

• In 1249 participants followed for 4.4 years slow gait speed predicted cognitive decline.

• Slow gait speed predicted the onset of cognitive impairment at follow-up.

Abstract

Reductions in physical performance, cognitive impairment (CI) and decline (CD), are common in older age, but few prospective cohort studies have considered the relationship between these domains. In this study we investigated whether reduced physical performance and low handgrip/lower limbs strength, could predict a higher incidence of CI/CD during a 4-year follow-up among a cohort of elderly individuals. From 3099 older community-dwelling individuals initially enrolled in the Progetto Veneto Anziani (PRO.V.A.) study, 1249 participants without CI at the baseline were included (mean age 72.2 years, 59.5% females). Physical performance measures included the Short Physical Performance Battery (SPPB), 4 m gait speed, chair stands time, leg extension and flexion, handgrip strength, and 6-Minute Walking Test (6MWT), categorized in gender-specific tertiles. CI was defined as a Mini-Mental State Examination (MMSE) score below 24; CD a decline of 3 or more points in the MMSE without CI. At baseline, participants developing CI during follow-up scored significantly worse across all physical performance measures compared to those that retained normal cognitive status. After adjusting for potential confounders, a significant trend for MMSE changes was noted for all physical performance tests, except for the SPPB and chair stands time. Multinomial logistic regression revealed that slow gait speed at baseline significantly predicted CD at follow up. Poor SPPB performance and slower gait speed predicted the onset of CI at the follow-up. In conclusion, slow walking speed appears to be the best independent predictor of poor cognitive status over a 4.4-year follow-up, while other items of SPPB were also significantly associated with CI.

Introduction

Cognitive impairment (CI) is a common condition in older people, affecting about 20% of individuals over 70 years old, and is associated with increased disability, healthcare utilization and mortality (Plassman et al., 2008). Several risk factors have been elucidated for CI and decline (CD), including aging, low level of formal educational, and several chronic diseases (Alfaro-Acha et al., 2006).

Recently, interest has intensified in understanding modifiable lifestyle factors that might influence the development of CD and CI. Among those that have received considerable attention is physical (in) activity (Norton et al., 2014). Data on physical activity levels are generally collected by means of questionnaires, however, there are inherent difficulties relying purely upon self-report physical activity and function, which are subject to strong recall bias in older people (Schlosser Covell et al., 2015). Data on physical activity whilst useful, indicates how much activity an individual is engaging in and does not reflect an individual's physical performance. Nevertheless, two meta-analyses recently demonstrated that self-reported reduced physical activity level is an independent risk factor for the onset of poor cognitive status (Sofi et al., 2011 Jan, Hamer and Chida, 2009 Jan). The mechanisms linking declining physical and cognitive performance include lower cognitive reserves, changes in the neurotrophic factors in the hippocampus, and the exacerbation of atherosclerotic processes in the cerebral vessels (Schlosser Covell et al., 2015, Blondell et al., 2014, Ahlskog et al., 2011, Buchman et al., 2012, Bennett et al., 2005). Whilst these previous studies have been helpful, assessing actual physical performance and muscle strength with objective measurements might offer a more accurate indicator of actual performance, particularly in older people who might have memory complaints and difficulty accurately recalling activity levels (Mitchell et al., 2014).

To date, a paucity of research has considered objective physical performance as a risk factor for future cognitive status in community dwelling older adults. The research has typically been cross sectional, or prospective with short duration follow up, although walking speed has been the most common physical performance measure considered (Ojagbemi et al., 2015 Feb, Mielke et al., 2013 Aug). Two prospective studies have found that handgrip strength was associated with a greater decline in memory tests (Alfaro-Acha et al., 2006, Christensen et al., 1994), but another study was unable to confirm these findings (Albert et al., 1995). There is a paucity of lower limb function tests (e.g. Short Physical Performance Battery) considering CD/CI. To our knowledge one cross-sectional study demonstrated that poor lower limb function correlated strongly with poor cognitive status (Rosano et al., 2005), but these results were only partially confirmed by a prospective study (Gallucci et al., 2013). Moreover, no comprehensive prospective representative study has considered the range of physical performance measures (gait speed, handgrip strength, lower limbs strength) in one study over a period of 4 years or more. Therefore, the data regarding physical performance and the prediction of CD/CI is equivocal and it remains unclear which measure might be the optimal predictors and offer most clinical utility.

Given the aforementioned limitations and gaps in the literature, the aim of the present study was to examine which objective physical performance measures provide an optimal predictor of incident CD and CI in a representative cohort of elderly individuals over a follow-up of 4.4 ± 1.2 years.