Epidemiology Table. Results of a screening test for diabetes compared to the tru
ID: 143732 • Letter: E
Question
Epidemiology Table. Results of a screening test for diabetes compared to the true diagnosis.
True Diagnosis
Test Results
Diabetic
Nondiabetic
Total
Positive
47
18
65
Negative
153
4,782
4,935
Total
200
4,800
5,000
In above Table, assume the prevalence of diabetes increases from 4% to 8%. Given the same sensitivity and specificity at blood glucose level of 180 mg/dl:
i) generate a new 2-by-2 table of the screening test against the true diagnosis;
ii) compute the positive predictive value of the test and
iii) explain your results in relation to this prevalence increase (in one or two sentences).
True Diagnosis
Test Results
Diabetic
Nondiabetic
Total
Positive
47
18
65
Negative
153
4,782
4,935
Total
200
4,800
5,000
Explanation / Answer
In the given table, sensitivity is measured by,
= True positive / (true positive + false negative)
= 47/(47+147)
= 47/200
= 0.235
Specificity is calculated by,
= True negative / (true negative + false positive)
= 4782 /(4782 + 18)
= 4782/4800
= 0.99625
The prevalence of diabetes is increased from 4% to 8%.
If sensitivity remains same, true positive is,
= 0.235 × 800
= 188
False negative is,
= 800 - 188
= 612
If specificity remains same, true negative is calculated as follows,
= 0.99625 × 4200
= 4184
False positive is,
= 4200 - 4184
= 16
I) The new table is,
II) Positive predictive value when prevalence is 4% is calculated by,
= True positive / (true positive + false positive)
= 47/(47+18)
= 47/65
= 0.7231
Positive predictive value when prevalence is 8% is,
= 188 / (188+16)
= 188/204
= 0.9216
Iii) When the prevalence of the disease increases, the positive predictive value of the test increases.
True diagnosis Test results Diabetic Non diabetic Total Positive 188 16 204 Negative 612 4184 4796 Total 800 4200 5000Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.