Table 2 Methods for presenting, testing, measuring, and communicating variability in results (n=53)

  No plots

5

9 %

  Forest Plots

34

64 %

  ROC

40

75 %

  Both

26

49 %

Statistical

 Cochran’s Q test

28

53 %

 I²

31

58 %

  Confidence Intervals presented

7

13 %

 τ²

7

13 %

  From a univariate analysis

6

11 %

  From a bivariate analysis

1

2 %

 Prediction intervals, ellipses, or bands

3

6 %

 Provided a definition for significant variability

24

49 %

  Cochran’s Q test

10

19 %

  I²

7

11 %

  Cochran’s Q test or I²

7

13 %

 Whether to perform a meta-analysis in the first place^a

1^a

4 %^a

 Whether to use a fixed or a random effects model

16

30 %

 Whether to investigate sources of heterogeneity

4

8 %

How variability in results is mentioned in the abstract and discussion and/or conclusions

Discussion/Conclusions

Abstract

 Any mention of variability listed below

29 (55 %)

15 (28 %)

 A vague discussion variability^b

17 (32 %)

10 (19 %)

 Reported results of a statistical test or measurement of variability

N/A

4 (8 %)

 Variability in results precludes firm conclusions or is a study limitation

13 (25 %)

2 (4 %)

 Despite variability in results, a conclusion could still be made

7 (13 %)

3 (6 %)

 There was no relevant variability in results

1 (2 %)

2 (4 %)