Article

The purpose was to review
the available literature for evidence on the reliability and measurement error of
protractor-based goniometry assessment of the finger joints. METHODS: Databases
were searched for articles with key words "hand," "goniometry," "reliability,"
and derivatives of these terms. Assessment of the methodological quality was
carried out using the Consensus-Based Standards for the Selection of Health
Measurement Instruments checklist. Two independent reviewers performed a best
evidence synthesis based on criteria proposed by Terwee et al (2007). RESULTS:
Fifteen articles were included. One article was of fair methodological quality,
and 14 articles were of poor methodological quality. An acceptable level for
reliability (intraclass correlation coefficient > 0.70 or Pearson's correlation >
0.80) was reported in 1 study of fair methodological quality and in 8 articles of
low methodological quality. Because the minimal important change was not
calculated in the articles, there was an unknown level of evidence for the
measurement error. DISCUSSION: Further research with adequate sample sizes should
focus on reference outcomes for different patient groups. For valid therapy
evaluation, it is important to know if the change in range of motion reflects a
real change of the patient or if this is due to the measurement error of the
goniometer. Until now, there is insufficient evidence to establish this cut-off
point (the smallest detectable change). CONCLUSION: Following the Consensus-Based
Standards for the Selection of Health Measurement Instruments criteria, there was
limited level of evidence for an acceptable reliability in the dorsal measurement
method and unknown level of evidence for the measurement error. LEVEL OF
EVIDENCE: 2a.
CI - Copyright (c) 2017 Hanley & Belfus. Published by Elsevier Inc. All rights
reserved.

Langue : ANGLAIS