Diagnostic studies
Diagnostic studies are conducted to evaluate the accuracy and effectiveness of diagnostic tests, which are used to determine the presence or absence of a particular disease or condition. In diagnostic studies, a group of individuals suspected of having the disease or condition are tested using the diagnostic test, and the results are compared to a gold standard (a definitive test or criteria for diagnosis).
Common performance measures for diagnostic tests include:
Sensitivity: The probability of an individual testing positive given the presence of disease.
Sensitivity = TP / (TP + FN)
Specificity: The probability of an individual testing negative given the absence of disease.
Specificity = TN / (TN + FP)
Positive predictive value (PPV): The probability that an individual who tests positive actually has the disease.
PPV = TP / (TP + FP)
Negative predictive value (NPV): The probability that an individual who tests negative actually does not have the disease.
NPV = TN / (TN + FN)
Positive likelihood ratio (LR+): The probability of an individual with the disease testing positive divided by the probability of an individual without the disease testing positive.
LR+ = sensitivity / (1 - specificity)
Negative likelihood ratio (LR-): The probability of an individual with the disease testing negative divided by the probability of an individual without the disease testing negative.
LR- = (1 - sensitivity) / specificity
Accuracy: The proportion of correct test results among all individuals tested.
Accuracy = (TP + TN) / (TP + FP + TN + FN)
Receiver operating characteristic (ROC) curves are used to evaluate the overall performance of diagnostic tests by plotting the true positive rate (sensitivity) against the false positive rate (1-specificity) at various threshold values of the test. A diagonal line from the bottom left corner to the top right corner of the plot represents a test with no discrimination ability, while a curve that approaches the top left corner of the plot represents a test with high discrimination ability.
The positive predictive value (PPV) and negative predictive value (NPV) of a diagnostic test can be affected by changes in disease prevalence. When the prevalence of a disease is low, false positives can be more common than true positives, which can lower the PPV (and raise the NPV) of the test. Conversely, when the prevalence of a disease is high, false negatives can be more common than true negatives, which can lower the NPV (and raise the PPV) of the test. Therefore, the prevalence of a disease should be taken into account when interpreting the results of a diagnostic test.
The Google Sheet above allows you to enter a 2 x 2 table to calculate diagnostic performance measures. In the second half of the sheet, you can adjust various parameters, such as disease prevalence, pretest probabilities, and sample size, to obtain other diagnostic performance measures as well as a visual representation of the 2 x 2 table (TN, TP, FN, and FP) in the provided chart. If you would like to access an editable version, click on the top right corner of the sheet above to open it in Google sheets. You can then download the file and enable editing on your device.
References: