Isometric Hand Grip Strength Measured by the Nintendo Wii Balance Board

A Reliable New Method

A. W. Blomkvist; S. Andersen; E. D. de Bruin; M. G. Jorgensen


BMC Musculoskelet Disord. 2016;17(56) 

In This Article


To the best of our knowledge, this is the first report measuring HGS using the WBB. The results demonstrated good reproducibility for the WBB in measuring isometric HGS with ICC values similar to JD,[23,28–33] the gold standard for measuring HGS. In addition, SEM and LOA values were comparable to or lower than[34] those observed with the JD, reflecting an acceptable absolute reproducibility. Moreover, we found a high concurrent validity between the JD and WBB with Pearson's product–moment correlation averaging 0.85. As expected, the Pearson correlation was somewhat lower than reported correlations between the JD and other handgrip dynamometers.[35,36] Still, the correlation was higher than that found between other HGS measurement techniques, such as the sphygmomanometer,[37] grip-ball[32] and vigorimeter,[38] and comparable to stationary alternatives, such as the BTW work simulator.[39]

On the other hand, the ICC values between the JD and WBB were a bit weaker than the Pearson's correlations, ranging between fair and good (0.793–0.803). The confidence intervals for the ICC validity analysis were significantly wider than for the reproducibility analysis, spanning from 0.525 to 0.901 and thus encompassing the full range from poor to high correlations. Hence, we have less confidence on the ICCs for the validity analysis, which must be interpreted with caution. The result from our study also indicate lower errors of measurement when considering the mean value of two or three measurements rather than one measurement or the maximum of three measurements, which is consistent with other studies.[40]

Although the WBB showed a acceptable correlation with the JD, there was a systematic difference in the results. On average this was 15.4 ± 5.5 kg for the dominant hand and 11.9 ± 5.5 kg for the non-dominant hand with the JD giving higher values. Hence, there is an inter-instrument difference between the WBB and the JD, and the instruments are not interchangeable. However, this lack of agreement has also been found between different dynamometers[35,41,42] and even between different models of JD.[43] These studies[35,41,42] are consistent with our results in that the JD tends to give higher values when compared to other instruments.

Comparing our results from the dominant and non-dominant hand there is similar reproducibility between hands, but a somewhat better correlation with the JD for the dominant side, both with Pearson's correlation (0.87 vs 0.82 on average) and ICC (.793 vs .768). The cause for this difference cannot be deduced from our results, but it might be expected that the strength difference between the two sides will result in lower correlation for the weaker side, i.e. non-dominant side (about 5 % weaker according to our results), since the JD allows assessment to the nearest kilogram, while the WBB allows assessment down to the nearest 100 gram. In combination with the increased measurement error for lower loadings with the JD, this may explain the difference observed.

One limitation with the WBB method for measuring HGS is the lack of adjustable handles to accommodate different hand sizes. Thus, the effect of hand size on our method is unknown. Still, this is the first investigation of HGS using the WBB. Compared to the above mentioned techniques the WBB has advantages in that it is a low-cost, portable and wide-spread tool. Furthermore, it has the potential for multiple roles in the clinical setting. This may include objectified measurements of reaction-time,[19] balance[18] and lower limb muscle strength.[21] The data presented here demonstrate the applicability of an additional facility, the HGS. The WBB has also been successfully used as an intervention tool for balance in healthy eldery,[44] as well as in chronic diseases[45,46] and for physical rehabilitation.[47–49] Finally, the multiple use of one instrument with rapid and automatic transferral of measurements to the computer system prevents loss of results and error in reporting, and it supports optimal use of staff time.

This study has both strengths and weaknesses. Firstly, we did not investigate inter-rater reliability. The impact on WBB results is likely to be limited as the results are read automatically and hand positioning is likely to be similar between raters. Secondly, the participants positioning deviates from the standardised procedure recommended by American Society of Hand Therapists,[16] but only by having a slightly more extended elbow position. The magnitude of this limitation is likely to be minimal. Thirdly, we only investigated independent older adults with a high level of functioning, as evaluated by their physical activity. Caution should be taken for generalizing the results to other age groups and populations. Still, this study had a sufficient number of participants for the purpose under study, and the methods have been reported in sufficient detail with the relevant statistics according to the GRRAS.