Skip to main content

Table 2 Trial characteristics

From: A review of the handling of missing longitudinal outcome data in clinical trials

  

Method of handling missing data

Acceptable method?1

  

No. of papers

Complete cases

Simple2

Multiple

Mixed models

No missing data

Unclear

Other3

Yes

No

Unclear

Number of patients

1-100

48

17

8

0

4

8

3

8

18

9

13

 

101-200

22

8

2

2

5

1

0

4

9

11

1

 

201-300

12

4

2

0

4

0

0

2

8

4

0

 

301-400

7

1

1

1

1

0

0

3

3

3

1

 

400+

11

1

1

1

4

0

0

4

6

3

2

Country of publication

USA

53

20

7

2

10

4

0

10

20

18

11

 

UK

40

8

6

2

7

3

3

11

20

12

5

 

Denmark

3

2

0

0

0

1

0

0

2

0

0

 

Netherlands

3

1

1

0

1

0

0

0

2

0

1

 

Japan

1

0

0

0

0

1

0

0

0

0

0

Number of time points

3

38

13

4

3

10

1

1

6

14

16

7

 

4

28

9

4

1

4

2

2

6

14

4

8

 

5

16

4

3

0

3

2

0

4

10

3

1

 

6

6

2

0

0

0

2

0

2

1

3

0

 

7

4

1

0

0

0

2

0

1

1

1

0

 

8+

8

2

3

0

1

0

0

2

4

3

1

Year

2005-06

19

9

2

0

1

2

1

4

7

4

6

 

2007-08

27

8

3

0

5

4

0

7

11

10

2

 

2009-10

25

7

3

1

4

3

1

6

9

8

5

 

2011-12

29

7

6

3

8

0

1

4

17

8

4

Clinical area

Mental health

13

2

4

2

1

1

1

2

7

2

3

 

Cancer

11

4

1

1

4

0

1

0

5

4

2

 

Rheumatology

10

4

1

0

2

1

0

2

3

5

1

 

Infectious diseases

8

4

1

0

1

1

0

1

4

2

1

 

Heart and circulation

7

1

2

0

1

0

0

3

3

2

2

 

Dentistry/oral health

6

3

1

0

0

2

0

0

3

1

0

 

Neurology

6

2

0

0

2

0

0

2

2

2

2

 

Anaesthesia and pain

6

3

0

0

1

0

0

2

2

2

2

 

Other4

33

8

4

1

6

4

1

9

15

10

4

Dropout reasons recorded?

Yes

35 (39.8%)5

15

7

0

7

NA

NA

6

20

11

4

 

Partial information

25 (28.4%)

7

5

1

7

NA

NA

5

15

9

1

 

No

28 (31.8%)

10

2

3

4

NA

NA

9

9

10

9

  1. 1Based on the 91 papers which had missing data.
  2. 2One paper which used a complete case analysis also used simple imputation as a secondary analysis. In Table2, this paper is included in the simple imputation section.
  3. 3Comparison of means e.g t-test, RMANOVA.
  4. 4A full list of medical areas is included in Table3.
  5. 5Disregarding the 9 papers without missing data and 3 papers where the missing data handling method was unclear.