Abstract
Relatively little information on the value of physical examination is available, as this is a difficult subject to research. The research that has been done often shows high intra-rater and inter-rater agreement. A study (or combination of studies) yields a number (e.g. kappa or LR), which is referred to as the ‘point estimate’. The certainty or uncertainty of the point estimate is expressed as a p-value or confidence interval. Sensitivity and specificity are test properties that indicate how well a test differentiates between people who do and do not have the particular disease. These test properties are influenced by the degree of unwellness and the cut-off point selected for the test. The likelihood ratio (LR) combines sensitivity and specificity and is an important measure of a test or examination, indicating the extent to which the odds (and hence likelihood) of having the disorder are changed by the result of the test. The predictive value of a test, as well as the quality of the test, depends on the prior probability of having the disorder, and it can be difficult to estimate.