Skip to main content

Table 3 Univariable and multivariable logistic regression derived odds ratios (OR) and 95% confidence intervals (CI) for the effect of year of publication, journal, continent of corresponding author, ethical approval, involvement of statistician, trial registration, significance of results, conflict of interest, funding, impact factor and number of authors on the likelihood of SR inclusion

From: Are orthodontic randomised controlled trials justified with a citation of an appropriate systematic review?

Predictor variables

Category

Univariable analysis

Multivariable analysis

OR (95%CI)

p-value

OR (95%CI)

p-value

Journal

AJODO

Reference

   

ANGLE

1.50 (0.59, 3.78)

0.390

  

EJO

1.27 (0.54, 3.01)

0.587

  

OCR

1.20 (0.32, 4.47)

0.786

  

AOJ

Not estimable

  

JOO

Dropped (exactly same % with AJODO)

  

JO

1.00 (0.17, 5.88)

1.00

  

Continent of corresponding author

Europe

Reference

   

Americas

1.45 (0.50, 4.18)

0.491

1.22 (0.41, 3.61)

0.72

Asia or other

0.41 (0.21, 0.78)

0.007

0.36 (0.18, 0.71)

0.003

Ethical approval

No approval or exempt from approval

Reference

   

Ethical approval obtained

1.69 (0.54, 5.27)

0.367

  

Involvement of statistician

Not reported

Reference

   

Reported

1.08 (0.66, 1.76)

0.755

  

Trial registration

No

Reference

   

Yes

1.41 (0.75, 2.67)

0.284

  

Significance of results

Not significant

Reference

   

Significant

1.00 (0.55, 1.83)

1.000

  

Conflict of interest

Conflicts exist and declared

Reference

   

No conflicts to declare

1.09 (0.56, 2.04)

0.782

  

Not clearly declared

1.79 (0.20, 15.92)

0.600

  

Funding

Industry funded and declared

Reference

   

No industry sponsorship/funding to declare

0.57 (0.29, 1.11)

0.100

  

Not clearly declared

0.89 (0.33, 2.38)

0.820

  

Year of publication

Per unit

1.07 (0.96, 1.18)

0.220

1.09 (0.98, 1.22)

0.11

Impact factor

Per unit

1.52 (0.76, 3.06)

0.237

  

Number of authors

Per unit

1.07 (0.93, 1.22)

0.339