Diagnostic Testing
Establish evidence for tests that diagnose disease and healthy people
The Diagnostic Testing decision tree will help researchers navigate the applied decision making processes associated with the establishing the utility and interpretability of diagnostic tests.
In order to conduct a study on a diagnostic test, researchers first have to choose the current "gold standard" to compare the diagnostic test of interest to when establishing evidence.
Each participant is given the diagnostic test and then the findings, either positive or negative, are compared to the findings of the "gold standard" method of diagnosis.
The prevalence of the disease state being detected by the diagnostic test and the "gold standard" also has to be taken into consideration. With higher rates of prevalence, researchers will get more "positive" results simply due to the fact that more cases exist in the population.
Click on a button below to continue.
In order to conduct a study on a diagnostic test, researchers first have to choose the current "gold standard" to compare the diagnostic test of interest to when establishing evidence.
Each participant is given the diagnostic test and then the findings, either positive or negative, are compared to the findings of the "gold standard" method of diagnosis.
The prevalence of the disease state being detected by the diagnostic test and the "gold standard" also has to be taken into consideration. With higher rates of prevalence, researchers will get more "positive" results simply due to the fact that more cases exist in the population.
Click on a button below to continue.
Diagnostic testing calculations
Sensitivity is the ability of a diagnostic test to detect disease.
Specificity is the ability of a diagnostic test to identify the healthy.
Researchers want to set the optimum cut-point for maximizing both sensitivity and specificity with a diagnostic test that uses a continuous scale of measurement. Researchers may also want to compare the diagnostic abilities of several diagnostic tests concurrently using Area Under the Curve (AUC or c-statistic).
Positive predictive value is how believable a "positive" test result is in a given population.
Negative predictive value is how believable a "negative" test result is in a given population.
Diagnostic accuracy is the overall ability of a test to correctly diagnose disease states.
Predict for future outcomes or events using pretest probabilities, clinical factors, and likelihood ratios.
The Youden index gives an overall measure of the ability of a given diagnostic test based sensitivity and specificity.
Downloads for diagnostic testing
Click on the Download Database for a free Excel database formatted for diagnostic testing data.
Download a free Excel calculator for diagnostic testing calculations.
Hire A Statistician - Statistical Consulting for Students
DO YOU NEED TO HIRE A STATISTICIAN?
Eric Heidel, Ph.D. will provide the following statistical consulting services for undergraduate and graduate students at $75/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.
- Statistical Analysis
- Research Design
- Sample Size Calculations
- Diagnostic Testing and Epidemiological Calculations
- Survey Design and Psychometrics
