Skip to main content

Table 2 Dmin scenario estimated mean and 95% confidence intervals of the sensitivity and specificity and the proportion that failed to meet the lower bound of the desirable TPP criteria confidence interval (sensitivity 93%; specificity 97%) assuming the test achieved minimum performance for the desirable TPP (Dmin: 97% sensitivity and 99% specificity) in the evaluation sample, for a simulated population size of 5,000

From: Examining the effect of evaluation sample size on the sensitivity and specificity of COVID-19 diagnostic tests in practice: a simulation study

Evaluation sample size

Observed TP/TN in evaluation

Theoretical

Simulation

Prevalence: 1%

Prevalence: 5%

Prevalence: 10%

Prevalence: 15%

Mean

95% CI

Failure (%)

Mean

95% CI

Failure (%)

Mean

95% CI

Failure (%)

Mean

95% CI

Failure (%)

Mean

95% CI

Failure (%)

Sensitivity

30

30

96.9%

88.8%

99.9%

10.5%

96.9%

86.5%

100.0%

13.5%

96.9%

88.5%

100.0%

11.3%

96.9%

88.5%

100.0%

10.7%

96.9%

88.5%

100.0%

11.1%

50

49

96.2%

89.6%

99.5%

11.9%

96.2%

86.8%

100.0%

17.7%

96.1%

88.8%

100.0%

14.0%

96.1%

89.3%

99.6%

13.2%

96.2%

89.3%

99.6%

12.4%

100

97

96.1%

91.6%

98.9%

7.1%

96.1%

87.8%

100.0%

17.2%

96.0%

90.6%

99.5%

10.3%

96.1%

91.2%

99.2%

8.6%

96.1%

91.3%

99.1%

8.1%

150

146

96.7%

93.4%

98.9%

1.7%

96.8%

89.8%

100.0%

10.7%

96.7%

92.5%

99.6%

3.7%

96.7%

92.8%

99.2%

2.9%

96.7%

93.1%

99.1%

2.3%

200

194

96.5%

93.6%

98.6%

1.1%

96.5%

89.6%

100.0%

12.1%

96.5%

92.7%

99.2%

3.5%

96.5%

93.0%

99.0%

2.6%

96.5%

93.3%

98.8%

1.8%

250

243

96.8%

94.3%

98.6%

0.3%

96.8%

90.4%

100.0%

9.6%

96.8%

93.1%

99.2%

2.3%

96.8%

93.7%

99.0%

0.9%

96.8%

93.9%

98.9%

0.6%

500

485

96.8%

95.1%

98.2%

0.0%

96.8%

90.7%

100.0%

8.4%

96.8%

93.8%

99.2%

0.8%

96.8%

94.3%

98.8%

0.2%

96.8%

94.6%

98.6%

0.1%

1000

970

96.9%

95.7%

97.9%

0.0%

96.9%

91.1%

100.0%

7.5%

96.9%

94.3%

99.1%

0.3%

96.9%

94.9%

98.6%

0.0%

96.9%

95.1%

98.4%

0.0%

2000

1940

97.0%

96.2%

97.7%

0.0%

97.0%

91.4%

100.0%

6.9%

97.0%

94.4%

98.9%

0.2%

96.9%

95.1%

98.5%

0.0%

97.0%

95.4%

98.3%

0.0%

Specificity

30

30

96.9%

88.8%

99.9%

38.9%

96.9%

88.8%

99.9%

39.5%

96.9%

89.1%

99.9%

38.7%

96.9%

88.9%

99.9%

39.1%

96.9%

88.8%

99.9%

39.5%

50

50

98.1%

93.0%

100.0%

21.2%

98.1%

93.1%

100.0%

20.1%

98.1%

93.1%

100.0%

21.7%

98.1%

92.9%

100.0%

21.5%

98.1%

93.0%

100.0%

20.7%

100

99

98.0%

94.6%

99.8%

19.0%

98.0%

94.5%

99.8%

19.4%

98.0%

94.7%

99.8%

19.1%

98.1%

94.5%

99.8%

18.5%

98.0%

94.5%

99.8%

18.9%

150

149

98.7%

96.4%

99.8%

5.7%

98.7%

96.4%

99.9%

5.6%

98.7%

96.4%

99.9%

5.8%

98.7%

96.4%

99.9%

6.1%

98.7%

96.3%

99.9%

5.7%

200

198

98.5%

96.5%

99.7%

5.8%

98.5%

96.4%

99.7%

6.3%

98.5%

96.4%

99.7%

6.4%

98.5%

96.4%

99.7%

6.0%

98.5%

96.4%

99.7%

6.3%

250

248

98.8%

97.2%

99.8%

1.9%

98.8%

97.1%

99.8%

2.1%

98.8%

97.1%

99.8%

2.2%

98.8%

97.1%

99.8%

2.0%

98.8%

97.1%

99.8%

2.3%

500

495

98.8%

97.7%

99.6%

0.2%

98.8%

97.6%

99.6%

0.3%

98.8%

97.6%

99.6%

0.3%

98.8%

97.7%

99.6%

0.3%

98.8%

97.6%

99.6%

0.4%

1000

990

98.9%

98.2%

99.5%

0.0%

98.9%

98.1%

99.5%

0.0%

98.9%

98.1%

99.5%

0.0%

98.9%

98.1%

99.5%

0.0%

98.9%

98.1%

99.5%

0.0%

2000

1980

99.0%

98.5%

99.3%

0.0%

98.9%

98.3%

99.4%

0.0%

98.9%

98.4%

99.4%

0.0%

98.9%

98.4%

99.4%

0.0%

99.0%

98.3%

99.4%

0.0%

  1. TP = true positives (in sensitivity), TN = true negatives (in specificity)