Skip to main content

Table 1 Number of deaths prevented and their 95% confidence intervals (CI) for the three scenarios and their various assumptions

From: Estimating effects of health policy interventions using interrupted time-series analyses: a simulation study

Year of implementation

Approach

Number of deaths preventeda

95% CI

Number of deaths preventeda

95% CI

Number of deaths preventeda

95% CI

Scenario 1

Effect size

  

−5

− 10

− 15

 

True value

60

 

120

 

180

 

5

Estimated

56

(1, 112)

122

(71, 173)

179

(126, 233)

 

Predicted

55

(−43, 153)

118

(24, 213)

176

(77, 275)

 

True value

60

 

120

 

180

 

9

Estimated

61

(0, 122)

121

(54, 188)

181

(118, 245)

 

Predicted

61

(−18, 140)

120

(41, 200)

181

(101, 261)

 

True value

50

 

120

 

180

 

13

Estimated

59

(1, 117)

119

(62, 176)

179

(120, 238)

 

Predicted

59

(−15, 133)

118

(44, 192)

179

(102, 256)

Scenario 2

Effect size

  

−5

−10

−15

 

True value

52

 

112

 

172

 

5

Estimated

52

(7, 96)

112

(67, 156)

171

(127, 215)

 

Predicted

50

(−50, 150)

111

(10, 212)

171

(71, 271)

 

True value

52

 

112

 

172

 

9

Estimated

52

(22, 82)

112

(80, 145)

172

(142, 202)

 

Predicted

53

(−26, 131)

111

(30, 191)

172

(94, 251)

 

True value

52

 

112

 

172

 

13

Estimated

52

(21, 82)

113

(81, 144)

171

(141, 202)

 

Predicted

51

(−23, 124)

111

(35, 187)

179

(94, 247)

Scenario 3

Effect size

  

−5

−10

−15

 

True value

1

 

17

 

33

 

5

Estimated

0

(− 28, 28)

17

(−11, 45)

33

(4, 62)

 

Predicted

−1

(− 93, 91)

16

(−75, 108)

32

(− 59, 124)

 

True value

1

 

17

 

33

 

9

Estimated

1

(−19, 20)

17

(−3, 37)

33

(13, 52)

 

Predicted

0

(−74, 75)

15

(− 57, 88)

34

(−41, 108)

 

True value

1

 

17

 

33

 

13

Estimated

0

(−20, 21)

17

(−2, 35)

33

(14, 52)

 

Predicted

0

(− 68, 67)

15

(−53, 83)

32

(−38, 102)

  1. CI Confidence interval
  2. aIn the 12 months following the intervention