Skip to main content

Table 2 Summary of iterative models predicting odds of publication based on abstracts characteristics

From: The mysterious case of the disappearing pilot study: a review of publication bias in preliminary behavioral interventions presented at health behavior conferences

Variable

Model 1

Model 2

Model 3

Model 4

Model 5

OR

95% CI

OR

95% CI

OR

95% CI

OR

95% CI

OR

95% CI

Randomized multi-arm trial

1.52

1.09

2.11

1.49

1.07

2.08

1.30

0.92

1.85

0.96

0.57

1.62

0.77

0.21

2.82

Significance

   

1.17

0.81

1.70

1.12

0.76

1.66

1.03

0.56

1.92

0.86

0.37

1.99

Sample size (n ≤ 24 referent)

 25–49

      

1.89

1.19

3.02

1.88

1.18

2.99

1.96

0.63

6.16

 50–99

      

2.01

1.25

3.25

2.00

1.24

3.23

1.10

0.36

3.39

  ≥ 100

      

1.82

1.09

3.03

1.81

1.08

3.01

1.01

0.34

2.97

Randomized multi-arm trial × significance

      

1.40

0.67

2.95

1.01

0.22

4.75

Randomized multi-arm trial × sample size

 25–49

            

0.90

0.14

5.88

 50–99

            

2.22

0.38

13.06

  ≥ 100

            

1.66

0.28

9.75

Significance × sample size

             

 25–49

            

0.86

0.22

3.38

 50–99

            

0.96

0.23

3.93

  ≥ 100

            

1.84

0.46

7.41

Randomized multi-arm trial × significance × sample size

 25–49

            

1.87

0.21

16.89

 50–99

            

1.87

0.22

15.94

  ≥ 100

            

1.07

0.12

9.67

Conferencea

0.73

0.51

1.05

0.71

0.49

1.03

0.78

0.53

1.15

0.77

0.52

1.14

0.73

0.49

1.08

Yearb

1.04

0.98

1.09

1.03

0.98

1.09

1.04

0.98

1.10

1.04

0.98

1.10

1.04

0.99

1.11

  1. Bold text indicates statistically significant, p ≤ 0.05
  2. aISBNPA referent
  3. bYear label represents year of abstract presentation