Paul W Stratford1. 1. School of Rehabilitation Science, McMaster University, Hamilton, Ont.
Abstract
Purpose: The past several decades have seen considerable interest in identifying and applying threshold change values with outcome measures commonly used by physiotherapists. The crucial question of interest to clinicians is, To what extent can valid inferences be drawn from an outcome measure's change or improvement score? To date, typical reporting by researchers includes the presentation of a validity coefficient, often in the form of the area under a receiver operating characteristic curve, and a threshold change or improvement value. A limitation of existing work is that it does not convey the confidence that a clinician can have in a decision based on applying the proposed threshold change value. Methods: This knowledge translation article presents three questions, or building blocks, to consider when making a judgment about a patient's change status: (1) to what extent does a measure assess change in the context of interest, (2) what is the threshold change value, and (3) how confident can a clinician be in making the correct decision about a patient's change status when applying the threshold change value? Results: This article provides a process for combining clinical expertise with the results from threshold value studies to enhance confidence in clinical decisions about individual patients' change status. Conclusions: The article shows how a graph can be used to efficiently translate the results from threshold value studies to convey the chance of making the correct decision about a patient's change status.
Purpose: The past several decades have seen considerable interest in identifying and applying threshold change values with outcome measures commonly used by physiotherapists. The crucial question of interest to clinicians is, To what extent can valid inferences be drawn from an outcome measure's change or improvement score? To date, typical reporting by researchers includes the presentation of a validity coefficient, often in the form of the area under a receiver operating characteristic curve, and a threshold change or improvement value. A limitation of existing work is that it does not convey the confidence that a clinician can have in a decision based on applying the proposed threshold change value. Methods: This knowledge translation article presents three questions, or building blocks, to consider when making a judgment about a patient's change status: (1) to what extent does a measure assess change in the context of interest, (2) what is the threshold change value, and (3) how confident can a clinician be in making the correct decision about a patient's change status when applying the threshold change value? Results: This article provides a process for combining clinical expertise with the results from threshold value studies to enhance confidence in clinical decisions about individual patients' change status. Conclusions: The article shows how a graph can be used to efficiently translate the results from threshold value studies to convey the chance of making the correct decision about a patient's change status.
Entities:
Keywords:
diagnostic test; outcome assessment; outcome measures; reproducibilty of results
Authors: Thomas J Hoogeboom; Rob A de Bie; Alfons A den Broeder; Cornelia H M van den Ende Journal: BMC Musculoskelet Disord Date: 2012-07-02 Impact factor: 2.362