In: Statistics and Probability
A screening test wished to improve the diagnostic ability to identify Zika-infected fetuses in pregnancy rather than after birth. A total of 330 pregnant women (carrying one baby each) in Miami were tested. A total of 36 babies had Zika; 24 of whom tested positive. A total of 300 babies tested negative for the virus. What is the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the test? Please match the measures to their correct answers below.
20.0% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
94.5% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
82.6% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
33.3% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
66.7% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
96.0% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
98.0% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
80.0% Choose 1… Accuracy Sensitivity Positive Predictive Value Specificity Negative Predictive Value None of these
We are given here that:
n(Total) = 330,
n(Zika and +) = 24
n(Zika and -) = 36 - 24 = 12
n(-) = 300, therefore n(+) = 330 - 300 = 30
Therefore, P(no Zika and +) = n(+) - n(Zika and +) = 30 - 24 =
6
n(no Zika and -) = n(-) - n(Zika and -) = 300 - 12 = 288
The parameters are now computed here as:
a) Sensitivity is computed here as:
= n(+ and Zika) / n(Zika)
= 24 / 36 = 2/3
Therefore 2/3 is the required probability here.
Therefore 66.67% is the required sensitivity here.
b) Specificity is computed here as:
= n(- and no Zika) / n(no Zika)
= 288 / 294
= 0.9796
Therefore 97.96% is the required specificity here.
c) The positive predicted value here is computed as:
= P(+ and zika) / P(+)
= 24/30
= 0.8
Therefore 80% is the positive predicted value here.
d) The negative predictive value here is computed as:
= P( - and no Zika) / n(-)
= 288 / 300
= 0.96
Therefore 96% is the negative predicted value here.
e) The accuracy is computed as:
= Total correct predictions / Total tests
= (12 + 288)/330
= 30/33
= 0.9090
Therefore 90.91% is the accuracy here.