Effect Measures of Diagnostic Accuracy Test Calculated Based on 2x2 Table
Result based on new test (T) Result based on golden standard (D) Total
 Subjects with the disease (D+)  
 Subjects without the disease (D-)
Positive（T+）
Negative(T-)
Total
Confidence Level
Distribution for confidence intervals
Measure Measure Point estimation Standard error Lower limit Upper limit
Accuracy Evaluation Sensitivity , %
False negative rate, %
Specificity , %
False positive rate, %
Diagnostic Value Evaluation Positive Predictive Value, %
Negative Predictive Value, %
Comprehensive Evaluation Accuracy , %
Diagnostic Odds Ratio
Positive Likelihood Ratio
Negative Likelihood Ratio
Bayesian Calculation pre-test probability
Post-test positive probability
Post-test negative probability
Consistency Evaluation
Kappa value Standard error Lower limit Upper limit
Hypothesis Tes for Consistency Evaluation

Kappa test for Consistencye

   Kappa Test

Mcnemar test for Consistency

McNemar test

Explanation:

  • The effect measures with a "%" symbol are single-group rates, and this page uses the asymptotic method. You can choose other methods to calculate the confidence intervals, click here to jump.
  • Pre-test and post-test probabilities are not multiplied by 100%.
  • For Kappa value, it ranges from 0 to 1, and calculation of confidence interval uses the Cohen (1960) method.
    • Negative Kappa indicates complete inconsistency.
    • Kappa < 0.4 indicates poor consistency.
    • 0.4 < Kappa < 0.6 indicates moderate consistency.
    • 0.6 < Kappa < 0.8 indicates good consistency.
    • Kappa > 0.8 indicates excellent consistency.
  • Hypothesis Test for Consistency:
    • Kappa value test: When P < significance level, it indicates Kappa is not equal to 0.
    • McNemar consistency test: When P < significance level, it indicates the results of the two methods are inconsistent."

Reference:

[1] Eusebi P. Diagnostic accuracy measures. Cerebrovasc Dis. 2013;36(4):267-272. doi:10.1159/000353863.
[2] Berkman ND, Wallace I, Watson L, et al. Screening for Speech and Language Delays and Disorders in Children Age 5 Years or Younger: A Systematic Review for the U.S. Preventive Services Task Force [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2015 Jul. (Evidence Syntheses, No. 120.)
[3]Ying GS, Maguire MG, Glynn RJ, Rosner B. Calculating Sensitivity, Specificity, and Predictive Values for Correlated Eye Data. Invest Ophthalmol Vis Sci. 2020 Sep 1;61(11):29. doi: 10.1167/iovs.61.11.29. PMID: 32936302; PMCID: PMC7500131.
[4] https://www.ncss.com/wp-content/themes/ncss/pdf/Procedures/PASS/Confidence_Intervals_for_One-Sample_Sensitivity.pdf