attacks against Coeffecient of variance and REG test
Tue, 01/13/2009 - 1:25pm
I have been doing a lit review, and there has been in the literature, some arguments agains using Coeffecients for grip, SD for grip, as well as arguments against the rapid exchange grip test. Mostof these articles are by an OT out of U of florida by the name of Shechtman, along with his colleagues. They present some good arguments, and have conducted some of their own research, albeit with limited sample size, looking at the statistics of the CV, SD, seeing if it is valid. Also looking at sensitivity and specificity of the CV and the REG. The results don't look good, but there werre some desing flaws with the study. Studies have been published in Journal of Hand Therapy over the last few years. He brings up some valid arguments, but wonder if he has an alterior motive. I do believe he is designing his own test for effort using Grip, which may be why he is using the last 4 years of his life downplaying CV and REG test.
Has anyone else looked at, ran across these articles and if so, what do you feel about these papers? I think they draw some generaliztions and their methodology was off, but it might just be enough to throw doubt into a jury or judge's mind, if you catch my drift.
Thu, 01/15/2009 - 10:55am
#2
attacks against Coeffecient of variance and REG test
Hello!
I forwarded your question to our clinical advisor and although Roy already answered you, I am also going to post what Jim had to say as well since he sent me an answer. It falls in line with the Matheson philosophy, so the answer is not too much different than Roy's:
We are always searching for the best evidence regarding validity of effort tests. Using CV's as a basis for effort has always been weighted with a certain amount of suspicion, but is still considered a good indicator of effort. As an experienced evaluator, I have on occasion explained away less value on a "failed" MVE and/or REG test when high levels of effort was supported by my observations of other, more durable signs of effort (Jay, Spine 2000). Often this arises after a frank discussion of low effort findings right after the MVE and/or REG tests. Oddly enough, if the evaluator goes back and re-tests for MVE and REG, the subject passes. This seems to lend support to the intrinsic and extrinsic value in us using this test, and early on. Hope this helps.
Good luck!
Jenn
Wed, 01/14/2009 - 10:40am
#3
attacks against Coeffecient of variance and REG test
roymatheson wrote:
First, Schechtman is a "she", not a "he".
As with all advances in or critiques of the methods we use one should read the actual article before commenting; as I have not read the article I will be careful in my response.
Keep in mind that Matheson does not rely on one test or one instrument to assess physical effort. The grip dynamometer is one tool in a battery of effort tests. Assuming that someone does flawless research that indicates that the dynamometer should not be used for effort testing then we, as an industry, would have to rethink the issue. In the meantime, we should use all of the tools at our disposal to address issues of effort.
Thanks,
And that is going to be our stance. I can e-mail you the articles.
We are meeting as a group to discuss these articles along with some other topics in a work journal review meeting coming up, and our stance is going to be that we use a cluster of tests to lok at issue not just one partclar stance. Schectman article discuss seperately, the validity of the SD, validity of CV, validity of REG, but do not look at what happen to validity or sen/spec when the tests are used as a cluster which may also change things. Our stance is going to be that we use a cluster of the above to look at effort, along with CPT and HR if it is ever brought up. Also we are reviewing some epi papers which look solely at research in general and support improving sen/spec as the number of diagnostic tests are used as a battery versus one single test. Hopefully that will help us on stand, as most of our FCE's go to court, if our documentation is good, we don't have to go, which is a Godsend.
Tue, 01/13/2009 - 2:37pm
#4
attacks against Coeffecient of variance and REG test
First, Schechtman is a "she", not a "he".
As with all advances in or critiques of the methods we use one should read the actual article before commenting; as I have not read the article I will be careful in my response.
Keep in mind that Matheson does not rely on one test or one instrument to assess physical effort. The grip dynamometer is one tool in a battery of effort tests. Assuming that someone does flawless research that indicates that the dynamometer should not be used for effort testing then we, as an industry, would have to rethink the issue. In the meantime, we should use all of the tools at our disposal to address issues of effort.
I would have liked to have quoted this directly before, but did not have access to this literature. This comes from a document by Dr. Matheson, I believe it is from the introduction to the AMA Guide To Disability and Impairment, but may be wrong. It is titled as "AMA FCE (2003)" at the EPIC website. Not only is this a great summary of pertinent issues in our field, but lends significant credence to the RMA approach to FCE. An excerpt I find most relevant to our goal of conducting valid FCE's includes this quote;
"It is important to recognize that almost any indicator of less than full effort can be volitionally defeated, and that some tests are more robust than others. The ease with which a person can misrepresent ability varies with the volitional control and transparency of the attribute being measured and with the method of the measurement. The easiest method to contravene is one that is most transparent, such as a grip strength test or pulmonary function measure."
This implies to me there is less value in grip strength testing/CV's and more on the "thinking evaluator". This follows my previous point that the MVE and REG tests are a good place to start, but I would much rather testify about what I saw than what the trend in research indicates-often with conflicting conclusions. This lends more and more credence being lent to what RMA teaches as a philosophy than a tool.