Skip to main content

Table 7 Results of Simulations for Scenarios 5 and 6

From: A comparison of estimators from self-controlled case series, case-crossover design, and sequence symmetry analysis for pharmacoepidemiological studies

   

Results of simulations

Scenarios

Methods (Effect measures)

Settings

Mean of estimates (Ratio scale)

Bias (Log scale)

Empirical standard error

Mean squared error

Mean standard error

Coverage (%)

5

SCCS (IRR)

P C a = 0.3

3.02

0.0055

0.1019

0.0104

0.0983

93.9

 

CCO (OR)

 

3.02

0.0053

0.2065

0.0426

0.1974

94.2

 

SSA (ASR)

 

4.39

0.3799

0.2764

0.2206

0.2628

72.8

 

SCCS (IRR)

P C a = 0.6

3.06

0.0202

0.1046

0.0113

0.1001

93.1

 

CCO (OR)

 

3.02

0.0053

0.2065

0.0426

0.1974

94.2

 

SSA (ASR)

 

7.87

0.9647

0.3638

1.0630

0.3397

0.1

 

SCCS (IRR)

P C a = 1.0

3.13

0.0421

0.1088

0.0136

0.1027

90.3

 

CCO (OR)

 

3.02

0.0053

0.2065

0.0426

0.1974

94.2

 

SSA (ASR)

 

NA

NA

NA

NA

NA

NA

6

SCCS (IRR)

Restricted to patients who did not experience the event before the first exposure

5.03

0.5164

0.1040

0.2775

0.1002

0.0

 

CCO (OR)

3.02

0.0053

0.2065

0.0426

0.1974

94.2

 

SSA (ASR)

NA

NA

NA

NA

NA

NA

  1. Abbreviations ASR Adjusted sequence ratio, CCO Case-crossover, IRR Incident rate ratio, NA Not applicable, OR Odds ratio, SCCS Self-controlled case series, SSA Sequence symmetry analysis
  2. aProbability of censoring upon event occurrence