Skip to main content

Table 2 Meta-regression estimates of the association of the proportion of studies comprising a summary effect size based on trial classification

From: Influence of pilot and small trials in meta-analyses of behavioral interventions: a meta-epidemiological study

Trial classification

Independent variable

Estimate

SE

(95CI)

Includes only studies with N > 100

Linear effect

 − 0.31

0.08

(− 0.47,

 − 0.15)

Excludes N ≤ 100 (i.e., small studies) and studies classified as pilot/feasibility

0–20%

0.46

0.08

(0.30,

0.62)

20–40%

0.12

0.04

(0.04,

0.19)

40–60%

0.13

0.04

(0.05,

0.21)

60–80%

0.11

0.04

(0.04,

0.18)

80–100%

0.02

0.04

(− 0.06,

0.10)

Includes only studies with sample sizes > 370

Linear effect

 − 0.48

0.12

(− 0.71,

 − 0.24)

Excludes N ≤ 370 and studies classified as pilot/feasibility

0–20%

0.30

0.06

(0.19,

0.41)

20–40%

0.20

0.05

(0.09,

0.31)

40–60%

0.04

0.05

(− 0.06,

0.15)

60–80%

0.08

0.06

(− 0.04,

0.21)

80–100%

0.01

0.15

(− 0.29,

0.32)

Includes studies with N > 100 and are classified as pilot/feasibility

Linear effect

0.39

0.10

(0.19,

0.59)

Excludes N ≤ 100 (i.e., small studies)

0–20%

0.03

0.03

(− 0.04,

0.09)

20–40%

0.10

0.03

(0.04,

0.17)

40–60%

0.11

0.04

(0.02,

0.19)

60–100%

0.29

0.06

(0.16,

0.41)

Includes all studies except for studies classified as pilot/feasibility

Linear effect

0.16

0.09

(− 0.03,

0.34)

 

0–20%

0.02

0.02

(− 0.02,

0.05)

20–40%

0.08

0.02

(0.04,

0.12)

40–60%

0.06

0.04

(− 0.02,

0.14)

60–100%

0.02

0.06

(− 0.10,

0.13)

  1. Note: Effects presented based on the comparison of the original standardized mean difference (SMD) and the re-estimated SMD using the random effects model