Confidence in clinical reasoning

Article Outline

• Copyright

Subjective history and clinical tests are the backbone of our diagnostic skills, and these skills take many years to learn and develop. Apart from experiential learning, how confident can we be of the results of clinical tests we employ? Most clinicians will probably be familiar with, and know, the sensitivity and specificity of commonly used clinical tests. However, the interpretation of these can be confusing when there is a large discrepancy between these two values. The Odds ratio is another statistic that can be employed for diagnostic accuracy, and this tends to be more commonly reported in the medical rather than physiotherapy literature. In the first paper presented in this issue, Walton and Sadi use the positive likelihood ratio (PLR), which employs both sensitivity and specificity in its calculation and it is an easy statistic for the clinician to interpret. The PLR is calculated by dividing the sensitivity by (1—specificity) and a value of between;

Additionally, methodological issues that can influence studies reporting diagnostic accuracy, such as internal validity and how to avoid missing false negatives, are discussed. Good internal validity is essential when deciding how robust the results of a study are. Other interesting concepts, such as publication bias, are considered and how these can be accounted for. Finally, at the end of this article, an excellent example of a case study in clinical reasoning is presented. The paper raises many issues surrounding clinical testing that all clinicians employing sound clinical reasoning should consider.

The number of diagnostic criteria and tests for shoulder pathologies seems to be increasing. Whether this is due to a limitation of existing clinical tests and/or an improved understanding of shoulder pathology and biomechanics is unclear. Walton and Sadi present a systematic review and meta-analysis of diagnostic tests for SLAP lesions and report that 14 different tests had been described in the literature. Only five of these tests qualified for further analysis, and of these, only Yergason's test was robust enough to have an influence on clinical decision making with a PLR of 2.29.

Continuing on the shoulder theme, Whitely and colleagues present a paper investigating the relationship between humeral torsion and proprioception of the shoulder in young male baseball players. Humeral retrotorsion is associated with increased range of movement of external rotation of the shoulder, necessary in throwing actions, and an increased throwing velocity. The authors found a strong and significant correlation between retrotorsion and active proprioceptive acuity in the non-dominant arm but not in the dominant arm in this cohort. The authors suggest a cognitive processing capacity model whereby greater retrotorsion reduces processing requirements, an interesting concept.

The next two papers both investigate the biomechanics of landing, among other parameters, in healthy and ACL deficient subjects, respectively. Plyometrics is a technique employed by coaches and trainers to increase jump performance, and eccentric training (a large component of plyometric training) has also been used to decrease landing forces in female athletes. It is intuitive that decreasing these forces will contribute to injury prevention. In a small controlled trial, Vescovi and colleagues investigated the effect of a 6-week plyometric training programme in female collegiate basketballers. A clinically meaningful reduction in vertical ground reaction force was found in the training group but without any increase in jump performance. The authors suggest that different regimens should be adopted if the training benefits are expected to include improved performance. In the next paper, Phillips and van Deursen found significant differences in parameters assessed for two functional activities, namely a running task and a single leg hop in ACL deficient subjects. The authors discuss the results with respect to motor control theory and discuss the concept that performance should not just be seen as ‘how fast’ or ‘how far’. An argument is presented for including measures of the quality of movement and skill level in addition to performance outcome to enhance confidence in return to sport decisions. The tests employed in the study could potentially be used to identify ACL deficient subjects who are non-copers.

The concept of ‘passive’ tissues, such as neural tissue and the fascia, being able to influence range of movement is not new. However, in the final article Mitchell and colleagues investigate the effect of neutral and anterior pelvic position on the range of ankle dorsiflexion during three different postures. The range of dorsiflexion varied considerably between the three postures but this did not relate to pelvic posture. The authors suggest that clinicians may want to consider the starting position for techniques aimed at mobilising dorsiflexion at the ankle joint. A systematic approach to mobilisation of the ankle is suggested to target different tissues in different positions, with a caveat that further research be undertaken to support this theory.

This is the last issue for 2008 and we would like to take this opportunity of thanking all our reviewers who are listed in the back of this issue. We really do appreciate all your time, effort and constructive comments, without which the peer-review process would not succeed.

A fun-filled festive season to all our readers and contributors, and we look forward to 2009.