Comment on ‘Effects of Vivifrail multicomponent intervention on functional capacity: A multicentre randomized controlled trial’ by Casas-Herrero et al.

We read with great interest the recent article by Álvaro Casas-Herrero et al. entitled ‘Effects of Vivifrail multicomponent intervention on functional capacity: a multicentre, randomized controlled trial’.¹ The article validated that the Vivifrail multicomponent exercise training programme is an effective and safe intervention for improving functional capacity in community dwelling frail/prefrail older patients with MCI or mild dementia, which provides strong evidence for the validation of physical intervention. However, here are some questions I would like to discuss with the authors.

The first question is about the missing data: The dropout rate was as high as 37% in general, especially 48% in the intervention group at the end of the third month. It is not a low level for a randomized controlled trial especially under such a small sample size. In clinical research, missing data often result in biased results and may even completely reverse the research conclusion.² Therefore, I wonder whether this effect had been taken into account in data processing? Tipping-point³ would be recommended to test the effect of missing data. If it had been used in this study, will the result be reversed at a certain level?

Secondly, as more patients discontinued the study in the intervention group compared with the control group, could you please describe the reasons in detail? If the exercise programme itself is difficult for the old, will the promotion be restricted in the future?

Lastly, is there any differences in the characteristics of missing patients between the control group and the intervention group? It is possible that those who can persist 3 months in the intervention group already have better functional capacity. So the significance of intervention observed in the study might partly resulted from the follow-up bias. Especially on the premise of the small sample and high rate of dropouts, the consistency of results might be affected.

In summary, it is hoped that the above-mentioned issues can be pondered in favour of consolidating the findings of Álvaro Casas-Herrero et al. and can better guide clinical decision making.