Effects of Balance Training on Balance Performance in Healthy Older Adults: A Systematic Review and Meta-analysis

Melanie Lesinski, Tibor Hortobágyi, Thomas Muehlbauer, Albert Gollhofer, Urs Granacher

The effects of balance training (BT) in older adults on proxies of postural control and mobility are well documented in the literature. However, evidence-based dose-response relationships in BT modalities (i.e., training period, training frequency, training volume) have not yet been established in healthy older adults. The objectives of this systematic literature review and meta-analysis are to quantify BT intervention effects and to additionally characterize dose-response relationships of BT modalities (e.g., training period, training frequency) through the analysis of randomized controlled trials (RCTs) that could maximize improvements in balance performance in healthy community-dwelling older adults. A computerized systematic literature search was performed in the electronic databases PubMed and Web of Science from January 1985 up to January 2015 to capture all articles related to BT in healthy old community-dwelling adults. A systematic approach was used to evaluate the 345 articles identified for initial review. Only RCTs were included if they investigated BT in healthy community-dwelling adults aged ≥65 years and tested at least one behavioral balance performance outcome (e.g., center of pressure displacements during single-leg stance). In total, 23 studies met the inclusionary criteria for review. Weighted mean standardized mean differences between subjects (SMDbs) of the intervention-induced adaptations in balance performance were calculated using a random-effects model and tested for an overall intervention effect relative to passive controls. The included studies were coded for the following criteria: training modalities (i.e., training period, training frequency, training volume) and balance outcomes [static/dynamic steady-state (i.e., maintaining a steady position during standing and walking), proactive balance (i.e., anticipation of a predicted perturbation), reactive balance (i.e., compensation of an unpredicted perturbation) as well as balance test batteries (i.e., combined testing of different balance components as for example the Berg Balance Scale)]. Heterogeneity between studies was assessed using I (2) and Chi(2)-statistics. The methodological quality of each study was tested by means of the Physiotherapy Evidence Database (PEDro) Scale. Weighted mean SMDbs showed that BT is an effective means to improve static steady-state (mean SMDbs = 0.51), dynamic steady-state (mean SMDbs = 0.44), proactive (mean SMDbs = 1.73), and reactive balance (mean SMDbs = 1.01) as well as the performance in balance test batteries (mean SMDbs = 1.52) in healthy older adults. Our analyses regarding dose-response relationships in BT revealed that a training period of 11-12 weeks (mean SMDbs= 1.26), a frequency of three training sessions per week (mean SMDbs= 1.20), a total number of 36-40 training sessions (mean SMDbs = 1.39), a duration of a single training session of 31-45 min (mean SMDbs = 1.19), and a total duration of 91-120 min of BT per week (mean SMDbs = 1.93) of the applied training modalities is most effective in improving overall balance performance. However, it has to be noted that effect sizes for the respective training modalities were computed independently (i.e., modality specific). Because of the small number of studies that reported detailed information on training volume (i.e., number of exercises per training session, number of sets and/or repetitions per exercise, duration of single-balance exercises) dose-response relationships were not computed for these parameters. The present findings have to be interpreted with caution because we indirectly compared dose-response relationships across studies using SMDbs and not in a single controlled study as it is difficult to separate the impact of a single training modality (e.g., training frequency) from that of the others. Moreover, the quality of the included studies was rather limited with a mean PEDro score of 5 and the heterogeneity between studies was considerable (i.e., I (2) = 76-92 %). Our detailed analyses revealed that BT is an effective means to improve proxies of static/dynamic steady-state, proactive, and reactive balance as well as performance in balance test batteries in healthy older adults. Furthermore, we were able to establish effective BT modalities to improve balance performance in healthy older adults. Thus, practitioners and therapists are advised to consult the identified dose-response relationships of this systematic literature review and meta-analysis. However, further research of high methodologic quality is needed to determine (1) dose-response relationships of BT in terms of detailed information on training volume (e.g., number of exercises per training session) and (2) a feasible and effective method to regulate training intensity in BT.