Andre Lacroix, Tibor Hortobagyi, Rainer Beurskens, Urs Granacher

• Objectives The objective of this systematic review and meta-analysis was to quantify the effectiveness of supervised vs. unsupervised balance and/or resistance training programs on measures of balance and muscle strength/ power in healthy older adults. In addition, the impact of supervision on training-induced adaptive processes was evaluated in the form of dose-response relationships by analyzing randomized controlled trials that compared supervised with unsupervised trials. Data Sources A computerized systematic literature search was performed in the electronic databases PubMed, Web of Science, and SportDiscus to detect articles examining the role of supervision in balance and/or resistance training in older adults. Study Eligibility Criteria The initially identified 6041 articles were systematically screened. Studies were included if they examined balance and/or resistance training in adults aged >= 65 years with no relevant diseases and registered at least one behavioral balance (e.g., time during single leg stance) and/or muscleObjectives The objective of this systematic review and meta-analysis was to quantify the effectiveness of supervised vs. unsupervised balance and/or resistance training programs on measures of balance and muscle strength/ power in healthy older adults. In addition, the impact of supervision on training-induced adaptive processes was evaluated in the form of dose-response relationships by analyzing randomized controlled trials that compared supervised with unsupervised trials. Data Sources A computerized systematic literature search was performed in the electronic databases PubMed, Web of Science, and SportDiscus to detect articles examining the role of supervision in balance and/or resistance training in older adults. Study Eligibility Criteria The initially identified 6041 articles were systematically screened. Studies were included if they examined balance and/or resistance training in adults aged >= 65 years with no relevant diseases and registered at least one behavioral balance (e.g., time during single leg stance) and/or muscle strength/ power outcome (e.g., time for 5-Times-Chair-Rise-Test). Finally, 11 studies were eligible for inclusion in this meta-analysis. Study Appraisal Weighted mean standardized mean differences between subjects (SMDbs) of supervised vs. unsupervised balance/resistance training studies were calculated. The included studies were coded for the following variables: number of participants, sex, age, number and type of interventions, type of balance/strength tests, and change (%) from pre- to post-intervention values. Additionally, we coded training according to the following modalities: period, frequency, volume, modalities of supervision (i.e., number of supervised/unsupervised sessions within the supervised or unsupervised training groups, respectively). Heterogeneity was computed using I 2 and chi(2) statistics. The methodological quality of the included studies was evaluated using the Physiotherapy Evidence Database scale. Results Our analyses revealed that in older adults, supervised balance/resistance training was superior compared with unsupervised balance/resistance training in improving measures of static steady-state balance (mean SMDbs = 0.28, p = 0.39), dynamic steady-state balance (mean SMDbs = 0.35, p = 0.02), proactive balance (mean SMDbs = 0.24, p = 0.05), balance test batteries (mean SMDbs = 0.53, p = 0.02), and measures of muscle strength/power (mean SMDbs = 0.51, p = 0.04). Regarding the examined dose-response relationships, our analyses showed that a number of 10-29 additional supervised sessions in the supervised training groups compared with the unsupervised training groups resulted in the largest effects for static steady-state balance (mean SMDbs = 0.35), dynamic steady-state balance (mean SMDbs = 0.37), and muscle strength/power (mean SMDbs = 1.12). Further, >= 30 additional supervised sessions in the supervised training groups were needed to produce the largest effects on proactive balance (mean SMDbs = 0.30) and balance test batteries (mean SMDbs = 0.77). Effects in favor of supervised programs were larger for studies that did not include any supervised sessions in their unsupervised programs (mean SMDbs: 0.28-1.24) compared with studies that implemented a few supervised sessions in their unsupervised programs (e.g., three supervised sessions throughout the entire intervention program; SMDbs: -0.06 to 0.41). Limitations The present findings have to be interpreted with caution because of the low number of eligible studies and the moderate methodological quality of the included studies, which is indicated by a median Physiotherapy Evidence Database scale score of 5. Furthermore, we indirectly compared dose-response relationships across studies and not from single controlled studies. Conclusions Our analyses suggest that supervised balance and/or resistance training improved measures of balance and muscle strength/power to a greater extent than unsupervised programs in older adults. Owing to the small number of available studies, we were unable to establish a clear dose-response relationship with regard to the impact of supervision. However, the positive effects of supervised training are particularly prominent when compared with completely unsupervised training programs. It is therefore recommended to include supervised sessions (i.e., two out of three sessions/week) in balance/resistance training programs to effectively improve balance and muscle strength/power in older adults.…