Diagnostic odds ratio

Log-DiagnosticOddsRatio

Diagnostic odds ratio (DOR) is a measure used in epidemiology and evidence-based medicine to evaluate the effectiveness of a diagnostic test. It is defined as the ratio of the odds of the test being positive in subjects with the disease relative to the odds of the test being positive in subjects without the disease. Mathematically, it can be expressed as:

\[ \text{DOR} = \frac{\text{Sensitivity} / (1 - \text{Sensitivity})}{\text{Specificity} / (1 - \text{Specificity})} \]

Alternatively, using the terms of a confusion matrix (true positives, false positives, true negatives, and false negatives), the DOR can be calculated as:

\[ \text{DOR} = \frac{\text{TP} \times \text{TN}}{\text{FP} \times \text{FN}} \]

Where:

TP = True Positives
TN = True Negatives
FP = False Positives
FN = False Negatives

The DOR is a single indicator of test performance that combines the properties of both sensitivity and specificity. A DOR of 1 indicates that a test does not discriminate between patients with the condition and those without it. A DOR greater than 1 suggests that the test is good at discriminating between patients with and without the condition, with higher values indicating better discriminatory test performance. Conversely, a DOR less than 1 indicates poor test performance.

Advantages and Limitations[edit | edit source]

One of the main advantages of the DOR is that it is a single summary measure that does not change with the prevalence of the condition. This makes it particularly useful for comparing diagnostic tests across studies and settings where the disease prevalence might differ.

However, the DOR has limitations. It can be undefined when either the sensitivity or the specificity is 1, as this would involve division by zero. Additionally, the DOR does not provide information on the absolute performance of a test, such as how many positive test results are actually true positives (positive predictive value) or how many negative test results are true negatives (negative predictive value). These limitations mean that while the DOR is useful for comparing the discriminatory power of tests, it should not be the sole measure used to assess test performance.

Applications[edit | edit source]

The DOR is widely used in meta-analysis of diagnostic test performance, where individual studies reporting sensitivity and specificity are combined to provide overall estimates of test effectiveness. It is particularly useful in situations where direct comparison of test performance is complicated by varying disease prevalence across studies.