Validity and test-retest reliability of a novel simple back extensor muscle strength test
Amy T Harding1, Benjamin Kurt Weeks1, Sean A Horan1
1Menzies Health Institute Queensland, Griffith University, Gold Coast campus, QLD, Australia; School of Allied Health Sciences, Griffith University, Gold Coast campus, QLD, Australia.
Objectives:
To develop and determine convergent validity and reliability of a simple and inexpensive clinical test to quantify back extensor muscle strength.
Methods:
Two testing sessions were conducted, 7 days apart. Each session involved three trials of standing maximal isometric back extensor muscle strength using both the novel test and isokinetic dynamometry. Lumbar spine bone mineral density was examined by dual-energy X-ray absorptiometry. Validation was examined with Pearson correlations (r). Test-retest reliability was examined with intraclass correlation coefficients and limits of agreement. Pearson correlations and intraclass correlation coefficients are presented with corresponding 95% confidence intervals. Linear regression was used to examine the ability of peak back extensor muscle strength to predict indices of lumbar spine bone mineral density and strength.
Results:
A total of 52 healthy adults (26 men, 26 women) aged 46.4 ± 20.4 years were recruited from the community. A strong positive relationship was observed between peak back extensor strength from hand-held and isokinetic dynamometry (r = 0.824, p < 0.001). For the novel back extensor strength test, short- and long-term reliability was excellent (intraclass correlation coefficient = 0.983 (95% confidence interval, 0.971-0.990), p < 0.001 and intraclass correlation coefficient = 0.901 (95% confidence interval, 0.833-0.943), p < 0.001, respectively). Limits of agreement for short-term repeated back extensor strength measures with the novel back extensor strength protocol were -6.63 to 7.70 kg, with a mean bias of +0.71 kg. Back extensor strength predicted 11% of variance in lumbar spine bone mineral density (p < 0.05) and 9% of lumbar spine index of bone structural strength (p < 0.05).
Conclusion:
Our novel hand-held dynamometer method to determine back extensor muscle strength is quick, relatively inexpensive, and reliable; demonstrates initial convergent validity in a healthy population; and is associated with bone mass at a clinically important site.
