Staes et al (2009), on the other hand, reported better reliability for end-feel assessment of accessory intercarpal motion as compared to mobility classifications.

With respect to spinal movement, Haneline et al (2008) similarly found somewhat higher reliability for measurement of end-feel. We hypothesise that measuring physiological movement for joints with large ranges of motion using goniometers or inclinometers, and measuring end-feel for joints with limited range of motion will lead to more reliable decisions about joint restrictions in clinical practice. Since Selleckchem FK228 few studies have investigated reliability of measurement of end-feel or accessory movements in upper extremity joints, future research should focus on the inter-rater reliability of these measures compared with measurements of physiological movements within the same sample of participants and raters. In this review, we found studies investigating inter-rater reliability of upper extremity joint motion examination to have been poorly conducted. Only one study satisfied all external validity criteria Selleck Vemurafenib and only two met all internal validity criteria. None of the included studies was both externally and internally valid. This finding

is no different from that of reviews of reliability of measurements of spinal movement (Seffinger et al 2004, Van Trijffel et al 2005). The majority of the studies in our review met the criterion concerning blinding procedures. However, criteria about the stability of participants’ and raters’ characteristics during the study were often either unmet or unknown. Instability of the participants’ characteristics under investigation, in this case joint range of motion or end-feel, may be caused

by changes in the biomechanical properties of connective tissues as a result of natural variation over time or the effect of the measurement procedure itself (Rothstein and Echternach 1993). Similarly, instability of the raters, in this case their consistency in making judgments, may be caused by mental fatigue. Instability of raters’ or participants’ characteristics can lead to underestimations of reliability, whereas a lack of appropriate these blinding of raters can lead to overestimation. In the presence of all of these methodological flaws, direction of risk of bias is difficult to predict. Factors about internal validity are closely linked to issues of generalisation of results. For instance, performing several measurements on a large number of participants in a limited time period is not only susceptible to bias but also does not reflect clinical practice. Reliability of measurements varies across populations of participants and raters (Streiner & Norman 2008).