Skip to main content

Table 2 The impact of the choice of one-sample estimator on AE frequency categories for the maximal follow-up time

From: Survival analysis for AdVerse events with VarYing follow-up times (SAVVY)—estimation of adverse event risks

  

Very rare

Rare

Uncommon

Common

Very common

(a) Gold-standard AJE

Incidence

Very rare

6

    

proportion

Rare

 

0

   
 

Uncommon

  

6

  
 

Common

   

86

2

 

Very common

    

86

Probability

Very rare

6

    

transform

Rare

 

0

   

incidence

Uncommon

  

3

  

density

Common

  

3

51

 

ignoring CE

Very common

   

35

88

1-Kaplan-

Very rare

6

    

Meier

Rare

 

0

   
 

Uncommon

  

4

  
 

Common

  

2

72

 
 

Very common

   

14

88

Probability

Very rare

6

    

transform

Rare

 

0

   

incidence

Uncommon

  

4

  

density

Common

  

2

79

1

accounting

Very common

   

7

87

for CE

      

Aalen-

Very rare

6

    

Johansen

Rare

 

0

   

death only

Uncommon

  

5

  
 

Common

  

1

73

 
 

Very common

   

13

88

(b) AJE death only

Incidence

Very rare

6

    

proportion

Rare

 

0

   
 

Uncommon

  

5

1

 
 

Common

   

73

15

 

Very common

    

86

  1. Deviations from the AJE are the non-diagonal entries. The first rows consider the gold-standard AJE and the last five rows the comparison of the incidence proportion and the AJE (death only) estimator. Diagonal entries are set in bold face. Non-diagonal zeros are omitted from the display