The Effects of a Multicomponent Lower Extremity Training Technique on Physical Function in Healthy Older Adults: A Randomized Controlled Trial

To the Editor-in-chief,

The authors have gone through this article with great interest and gathered valuable information. We would like to congratulate Hungu Jung et al. for their valuable study published in the ”Gerontology & Geriatric Medicine” titled “The Effects of a Multicomponent Lower Extremity Training Technique on Physical Function in Healthy Older Adults: A Randomized Controlled Trial” on volume 6, page 1–10, and year 2020. The authors clearly explain about that aging can lead to reduce muscular strength, range of motion and loss of balance etc. Physical activity and exercises help to improve these age related dysfunction. Moreover, the exercises are designed to execute without the assistance of any equipments and also are cost effective. The authors had mentioned about the exercises clearly with images, which is appreciative.

But there are some methodological issues in this paper which needs to be highlighted. First, they had mentioned the calculated sample of, n = 14, in each group, as per the effect size = 1.3, α-error = .05, and power = 0.95. However, based on the above mentioned criteria, when we calculated the sample size using the G*Power software (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany; http://www.gpower.hhu.de/), the sample size was 17 in each group. Also they had mentioned that the number of sample was increased above 14 to avoid dropouts. But they did not mention about the percentage of dropout they had taken. Next, in statistical analysis part, the author did not mention about the normal distribution of data. For meaningful result of any study, normality assumption should be followed. Continuous data following normal distribution should be presented in mean and p-values within and between the groups should be estimated using the values of mean. However, if the data is not normally distributed, the median is used to represent the data. Proper selection of representative data and calculation of significant value is essential to avoid wrong interpretation. Therefore, the testing of normal distribution of data is important regardless of sample size (Mishra et al., 2019). Based on that, parametric and non-parametric test can be used. If the data is normally distributed, parametric test can be used, and nonparametric tests are used for not normality distributed data (Mishra et al., 2019). Last, in the CONSORT flow chart, the authors had mentioned twice about the drop-out under follow-up section, which are not appropriate (Scazufca et al., 2019). We appreciate the authors as they had explained clearly about the protocol design and weekly modification of intervention.