This calculator estimates the true prevalence of disease, given a) the test characteristics from a validation (test accuracy) study, followed by b) test results in a sample from the target population.
a) Validation study. Run a test T on cases known to have disease (D+) to estimate sensitivity and controls known not to have disease (D-) to estimate specificity.
b) Prevalence study. Now apply the test to a representative sample from the population of interest. To estimate true prevalence, the proportion of the sample with a positive test (T+) must be adjusted for the test's imperfect sensitivity and specificity.
Instructions: Enter parameters in the green cells. Answers will appear in the blue box below.
Test validation
DV+
Number of cases known to have disease in the validation study
TV+ & DV+
Number of positive test results among those cases
Sensitivity:
%
Confidence interval:
Lower bound*
Upper bound*
DV−
Number of known non-disease controls in validation study
TV− & DV−
Number of negative test results among those controls
Specificity:
%
Confidence interval:
Lower bound*
Upper bound*
*Confidence intervals for sensitivity and specificity are calculated using normal approximation.
Prevalence
Sample size
Number of people tested in sample from target population.
T+
Number of positive test results in population.
P(T+)
Raw prevalence
CL
%
Confidence level
Point estimate of true prevalence in population:
Confidence interval per Reiczigel
Lower bound:
Upper bound:
Confidence interval per Rogan and Gladen
Lower bound:
Upper bound:
References:
Rogan WJ, Gladen B. Estimating prevalence from the results of a screening test. Am J Epidemiol. 1978;107(1):71-6.
Lang Z, Reiczigel J (2014) Confidence limits for prevalence of disease adjusted for estimated sensitivity and specificity, Preventive Veterinary Medicine, 113, 13-22.