Blodgett JM, Cooper R, Davis DHJ, Kuh D and Hardy R (2020) Associations Between Factors Across Life and One-Legged Balance Performance in Mid and Later Life: Evidence From a British Birth Cohort Study. Front. Sports Act. Living 2:28. doi: 10.3389/fspor.2020.00028
“The life course epidemiology of balance performance has been under-investigated compared with other measures of physical capability such as grip strength and chair rise performance.”
Population: 3,111 British participants in the MRC National Survey of Health and Development born in 1946 and followed through time
The questions covered:
The MRC National Survey of Health and Development is a British cohort study composed of individuals born in 1946 and followed through their life course. Starting at age 53, participants in the study were assessed for static balance performance using a timed one-legged, eyes-closed balance protocol. In this study, the investigators evaluated the association between balance performance and a variety of risk factors, including indicators of socioeconomic (in early and later life), anthropometric, behavioral, health, and cognitive status. They used statistical models to determine which of these factors was significantly correlated with balance performance, and whether the associations varied by age or sex.
The results of this analysis indicate a large number of risk factors from across the life course are associated with balance performance in older age. Specifically, participants from lower socioeconomic statuses (both in childhood and adulthood), those with higher BMI, those who were current or former smokers, and those with a history of health conditions (diabetes, cardiovascular disease, respiratory symptoms, knee pain, or depression/anxiety) had worse balance times. Taller participants, those who reported engaging in more frequent physical activity, and those with better memory assessment results had improved balance times. Female participants had worse balance performance than male participants, but none of the estimated associations between the risk factors of interest and balance varied by sex. As participants aged, the correlation of balance performance with socioeconomic factors, memory assessment, and frequency of physical activity decreased, while the correlation with knee pain and depression increased. When the risk factors of interest were combined into a single large statistical model, the direction of all associations remained the same and the majority of the factors remained statistically significant.
This study provides evidence that balance performance in older adults is highly correlated with a variety of diverse factors across an individual’s lifetime, and that these correlations are largely independent of one another. It also suggests that the relationships between these factors and balance performance change as people age, but are uniform for both men and women. Taken as a whole, these results support the idea that balance is complex and multifactorial, and has the potential to serve as an indicator for a large number of social, biological, and health conditions, even when measured through a fairly coarse assessment protocol.
Introduction: Despite its associations with falls, disability, and mortality, balance is
an under-recognized and frequently overlooked aspect of aging. Studies investigating
associations between factors across life and balance are limited. Understanding the
factors related to balance performance could help identify protective factors and
appropriate interventions across the life course. This study aimed to: (i) identify
socioeconomic, anthropometric, behavioral, health, and cognitive factors that are
associated with one-legged balance performance; and (ii) explore how these associations
change with age.
Methods: Data came from 3,111 members of the MRC National Survey of Health and
Development, a British birth cohort study. Multilevel models examined how one-legged
standing balance times (assessed at ages 53, 60–64, and 69) were associated with
15 factors across life: sex, maternal education (4 years), paternal occupation (4 years),
own education (26 years), own occupation (53 years), and contemporaneous measures
(53, 60–64, 69 years) of height, BMI, physical activity, smoking, diabetes, respiratory
symptoms, cardiovascular events, knee pain, depression and verbal memory. Age and
sex interactions with each variable were assessed.
Results: Men had 18.8% (95%CI: 13.6, 23.9) longer balance times than women at age
53, although this difference decreased with age (11.8% at age 60–64 and 7.6% at age
69). Disadvantaged socioeconomic position in childhood and adulthood, low educational
attainment, less healthy behaviors, poor health status, lower cognition, higher body mass
index (BMI), and shorter height were associated with poorer balance at all three ages.
For example, at age 53, those from the lowest paternal occupational classes had 29.6%
(22.2, 38.8) worse balance than those from the highest classes. Associations of balance
with socioeconomic indicators, cognition and physical activity became smaller with age,
while associations with knee pain and depression became larger. There were no sex
differences in these associations. In a combined model, the majority of factors remained
associated with balance.
Discussion: This study identified numerous risk factors across life that are associated
with one-legged balance performance and highlighted diverse patterns of association
with age, suggesting that there are opportunities to intervene in early, mid and later life.
A multifactorial approach to intervention, at both societal and individual levels, may have
more benefit than focusing on a single risk factor.
Note: The direction of effects associated with health status risk factors shown in Table 2 and Figure 6 do not match the description of the results. Per correspondence with the first author, the coefficients for the mean difference in % balance time for the following health factors should be negative: history of diabetes, history of cardiovascular events, respiratory symptoms, knee pain, and symptoms of anxiety/depression. The coefficients for the interaction effects in Table 2 are correct.