The Los Angeles County Health Department was chosen to evaluate the validity of
ID: 3197837 • Letter: T
Question
The Los Angeles County Health Department was chosen to evaluate the validity of a new diabetes screening test. The department recruited 1000 participants for the study. 800 participants were eligible for the screening test and were subjected to the state-of the art glucose dipstick screening test. The results showed that 69 participants tested positive out of 688 participants who did not have diabetes. If the sensitivity of the test was 90%, calculate the prevalence per 1000 people, positive predictive value and negative predictive value and specificity.
Explanation / Answer
Disease
(number)
Non-disease
(number)
Total
(number)
Positive
(number)
A
(True Positive)
B
(False Positive)
Negative
(number)
C
(False Negative)
D
(True Negative)
Given:
Total Participants who took the test = 800
Number of people did not have diabetes= TNon-Disease = 688
Number of people who tested positive but did not have diabetes = B (false positive) = 69
Sensitivity = 90%
Inference:
TNon-Disease = 688 , Total = 800
Thus, TDisease = 800 - 688 = 112
Referring to table above, B + D = TNon-Disease and B = 69
Therefore D = 688 - 69 = 619
Assumptions:
?Let People with diabetes and tested positive (A) = x
And since Total number of people with diabetes = A + C = TDisease
x + C = 112
Thus, C = 112 - x
Inserting above Data in the table, we get:
Disease
(number)
Non-disease
(number)
Total
(number)
Positive
(number)
Negative
(number)
Sensitivity (%) = A / (A + C) * 100
90 = x/112 * 100
x = (90 * 112) / 100
x = 100.8 But since People cannot be fractions,
x = 100 -> A = 100
Therefore, C = 112 - 100
C = 12
Now,
TTest Positive = A + B
TTest Positive = 100 + 69
TTest Positive = 169
Therefore,
TTest Negative = Total - TTest Positive = 800 - 169
TTest Negative = 631
Filling in the data above in the 2nd Table:
Disease
(number)
Non-disease
(number)
Total
(number)
Positive
(number)
Negative
(number)
= 100 / 169 * 100
= 69.17%
?= 619 / 631 * 100
= 98.09%
?= 619 / 688 * 100
= 89.97%
*** I am not sure about the prevalence per 1000 people , apologies if i am incorrect ***
?= 112 / 800 * 100
= 14 %
Disease
(number)
Non-disease
(number)
Total
(number)
Positive
(number)
A
(True Positive)
B
(False Positive)
TTest PositiveNegative
(number)
C
(False Negative)
D
(True Negative)
TTest Negative TDisease TNon-Disease TotalRelated Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.