Inter-rater reliability of the instrumented lean-and-release (iLEAN) test for manual assessment of reactive balance: a pilot study in simulated older adults.

Everyone tested was a healthy young woman of average age 20.7 years wearing a weighted vest, ankle weights and knee braces. The authors state that these results may not generalise to older adults or clinical populations. The same weights were worn by everyone, so participants of different body size were not impaired to the same degree. Only 16 people were tested, and the confidence intervals around the agreement figures are wide. Agreement in the forward direction could be as low as 0.17. The same person was not tested twice, so the study says nothing about whether a single assessor gets the same score on a different day (no test-retest reliability). Only women took part. The authors say future work should include men with a range of body sizes. Forward and backward were tested only on the dominant leg, which may miss differences between sides in people with stroke or Parkinson's disease. The foam block used in the forward and lateral trials may not be necessary and stopped four participants from completing the left and right trials at the starting load. The authors suggest it may be better left out, particularly in people whose balance is already impaired. The authors state that an initial target load of 5% of body weight may be too high for frail individuals, and suggest lower starting loads such as 3%.

The authors declare no conflicts of interest. The paper was also posted as a preprint on medRxiv. The device used was a commercially available hand-held dynamometer from Anima Corp., and the authors report no funding or sponsor involvement in the design, analysis or writing of the study.

Do not read this as evidence that the iLEAN test is reliable in the older or clinical patients you treat. The two assessors tested 16 healthy young women wearing weights to mimic ageing, and in the forward direction agreement was only moderate by the authors' own cut-offs, with a confidence interval wide enough to include poor agreement.