Table 1

Algorithm summary performance statistics for false alarm rates approximating 0.005

Algorithm

false alarm rate

median (mean) sensitivity

Day 1 sensitivity

Day 2 sensitivity

Days 1–7 sensitivity

CARL

median (mean) timeliness

median (mean) adjusted timeliness


EARS C1

0.0049

1 (0.53)

0.13

0

0.13

1.0

18.5 (54.6)

34.0 (78.8)

EARS C2

0.0042

1 (0.53)

0.33

0

0.33

1.0

0.0 (24.1)

32.0 (71.1)

EARS C3

0.0049

1 (0.60)

0.40

0

0.40

1.0

0.0 (17.9)

32.0 (68.1)

NBC 7-day

0.0049

1 (0.87)

0.2

0.07

0.40

2.0

12.0 (16.3)

14.0 (18.5)

NBC 14-day

0.0049

1 (0.93)

0.2

0.07

0.33

1.6

9.5 (13.9)

11.0 (15.2)

NBC 28-day

0.0049

1 (1.00)

0.27

0.07

0.60

2.9

5.0 (8.5)

5.0 (8.5)

NBC 56-day

0.0042

1 (0.93)

0.13

0

0.27

3.0

14.5 (19.1)

14.0 (18.8)

p-value¥

-

0.0002

-

-

-

-

0.71

0.008


Test threshold that produced a false alarm rate ≤ 0.005

Detection on the first outbreak day is equivalent to a timeliness of 0 days

¥Friedman rank sum test

CARL: Conditional Average Run Length – conditional on the detection of the outbreak during the first 7 days

EARS: Early Aberration Reporting System

NBC: Negative binomial cusum with an out of control state defined as 2 standard deviations greater than the mean

Watkins et al. BMC Medical Informatics and Decision Making 2008 8:37   doi:10.1186/1472-6947-8-37

Open Data