Formula for calculating positive predictive value (PPV):
High-Yield Explanation
Ans: (a) a/a+bRef: Park, 24th edition, Page 149Basic measures used in evaluation of a diagnostic test are:a. Sensitivity and specificityb. Predictive values - positive predictive value and negative predictive valueSensitivityAbility of the test to identify true positivesSpecificityAbility of the test to identify true negativesPositive predictive valueProbability that a patient with positive test result truly has the diseaseNegative predictive valueProbability that a patient with negative test result truly does not have the disease Screening testDia<gnosisTotalDiseasedNon diseasedPositivea (True Positive)b (False Positive)a+bNegativec (False Negative)d (True Negative)c+dTotala+cb+da+b+c+dUsing above tableSensitivity =a/(a+c)x100Specificity =d/(b+d)x100Predictive value of a positive test =a/(a+b)x100Predictive value of a negative test =d/(c+d)x100Percentage of false negatives =c/(a+c)x100Percentage of false positive =b/(b+d)x100Tips to avoid confusions and to be clear of this concept:Whatever question is asked about these indicators1. First make the table with disease in columns and test in the rows2. Remember the rule that sensitivity and specificity are column percentages and predictive values are row percentages.3. Always, true values in numerator.In the below tables, concentrate on the shaded areas Sensitivity DiseasedNon-DiseasedTotalTest+a (True Positives)b (False Positives)a+bTest-c (False Negatives)d (True Negatives)c+dTotala+cb+da+b+c+dSensitivity =a/a+c*100 Specificity DiseasedNon-DiseasedTotalTest +a (True Positives)b (False Positives)a+bTest-c (False Negatives)d (True Negatives)c+dTotala+cb+da+b+c+dSpecificity =d/b+d*100Positive Predictive Value DiseasedNon-DiseasedTotalTest +a (True Positives)b (False Positives)a+bTest -c (False Negatives)d (True Negatives)c+dTotala+cb+da+b+c+dPositive predictive value =a/a+b*100 Negative Predictive Value DiseasedNon-DiseasedTotalTest +a (True Positives)b (False Positives)a+bTest-c (False Negatives)d (True Negatives)c+dTotala+cb+da+b+c+dNegative predictive value =d/c+d*100