Java Source Code

Applet 1. Effect of Referral Bias on Apparent Sensitivity and Specificity

Your browser can't run Java 1.1 applets.
Otherwise, you would have seen a calculation applet which looks like the following
.

FN = false negative, FP = false positive, PV = predictive value, Sens. = sensitivity, Spec. = specificity, TN = true negative, TP = true positive.
The values in this applet may be changed by entering new numeric text (strike the [Enter] key to see the effect of changes) or by moving the slide bar. Entering new values in the fields "Numbers of Patients," "Disease Prevalence," "True Sensitivity," "True Specificity," "Apparent Sens.," "Apparent Spec.," "Positive Test," and "Negative Test" automatically changes the TP, FP, TN, and FN numbers. The direction of the arrows (in the center frame, above "Referral Rate") indicates whether changes in the referral rates will affect the true or the apparent sensitivity and specificity values. When the arrow points to the right, changes in referral rates (for positive and/or negative tests) affect the apparent sensitivity and specificity. Conversely, when the arrow points to the left, changes in referral rates (for positive and/or negative tests) affect the true sensitivity and specificity values. When any value other than the referral rate is changed, the direction of the arrows does not affect the output. Note that the positive and negative predictive values are independent of referral biases based on test outcome. This applet requires that your browser run Java 1.1. Refreshment of values may be slower on earlier implementations of the Java virtual machine.

Applet 2. Effect of Referral Bias and Adequacy of Stress on Apparent Sensitivity and Specificity

Your browser can't run Java 1.1 applets.
Otherwise, you would have seen a calculation applet which looks like the following.

HR = heart rate, MPHR = age-predicted maximum heart rate, NPV = negative predictive value, PPV = positive predictive value.
The values in this applet may be changed by entering new numeric text (strike the [Enter] key to see the effect of changes) or move the slide bar. Entering new values in the fields "Numbers of Patients," "Disease Prevalence," "True Sensitivity," "True Specificity," "Apparent Sens.," "Apparent Spec.," "Positive Test," and "Negative Test" automatically changes the TP, FP, TN, and FN values. The direction of the arrows (in the center frame, above "Referral Rate") indicates whether changes in the referral rates will affect the true or the apparent sensitivity and specificity values. When the arrow points to the right, changes in referral rates (for positive and/or negative test results) affect the apparent sensitivity and specificity. Conversely, when the arrow points to the left, changes in referral rates (for positive and/or negative test results) affect the true sensitivity and specificity values. When any value other than the referral rate is changed, the direction of the arrows does not affect the output. Note that the positive and negative predictive values for each subgroup are independent of referral biases based on test outcome. For the whole population, however, the respective positive and negative predictive values are affected by referral biases. This applet requires that your browser run Java 1.1. Refreshment of values may be slower on earlier implementations of the Java virtual machine.

This Article

	Abstract
	Full Text
	Submit a response

Services

	Email this article to a friend
	Alert me to new issues of the journal