Skip to main content

Table 1 Characteristics of participants in each simulated dataset

From: Does pattern mixture modelling reduce bias due to informative attrition compared to fitting a mixed effects model to the available cases or data imputed using multiple imputation?: a simulation study

Phase

5

7

9

Smoking status, n (%)

 Non-smoker

4940 (49.4)

4940 (49.4)

4940 (49.4)

 Ex-smoker

4343 (43.4)

4436 (44.4)

4557 (45.6)

 Current smoker

717 (7.2)

624 (6.2)

503 (5.0)

Age Category (year), n (%)

 < 50

2420 (24.2)

  

 50 and < 55

2967 (29.7)

  

 55 and < 60

2010 (20.1)

  

 60 and < 65

1896 (19.0)

  

 65

707 (7.1)

  

Employment Grade, n (%)

 High

5812 (58.1)

  

 Intermediate

3878 (38.8)

  

 Low

310 (3.1)

  

Education, n (%)

 None

555 (5.6)

  

 School

4675 (46.8)

  

 University

4770 (47.7)

  

Standardised cognitive function (SD), mean (SD)

 Global

0.00 (0.78)

−0.22 (0.78)

− 0.42 (0.79)

 Memory

0.05 (0.71)

− 0.10 (0.71)

− 0.26 (0.72)

Participation status, n (%)

 Response

8709 (87.1)

7850 (78.5)

7549 (75.5)

 Died

0

240 (2.4)

560 (5.6)

 Non-response

1291 (12.9)

1485 (14.9)

1289 (12.9)

 Attrition

0

425 (4.3)

602 (6.0)