Skip to main content

Table 5 Regression Analysis

From: An evaluation of the completeness of safety reporting in reports of complementary and alternative medicine trials

 

Increase in percentage of words devoted to the reporting of safety in the results sectionŦ

Adequacy of reporting

Predictors

Least-square regressions (Adjusted¥) Effect Size (95% CI)

Logistic regressions (Adjusted) OR (95% CI)

Sample Size (per 10 unit increase)

0.02 (-0.04-0.07)

1.00 (1.00-1.00)

Double Blind

2.86 (-1.04-6.76)

1.05 (0.47-2.31)

Significant results for efficacy

-1.93 (-5.36-1.49)

1.18 (0.03-1.40)

Industry Funded

0.27 (-4.04-4.58)

2.49 (0.98-6.32)

Longest duration to follow-up, ≥ 6 m

-0.07 (-4.04-3.91)

0.99 (0.42-2.33)

Paediatric population

-4.50 (-11.07-2.07)

0.56 (0.11-2.90)

Therapeutic studies

3.21 (-0.84-7.26)

1.81 (0.73-4.48)

Combined studies

-0.32 (-4.03-3.39)

0.79 (0.34-1.81)

Multicentre

4.02 (0.12-7.92)*

1.89 (0.84-4.27)

2009 Journal Impact Factor (per 10 unit increase)

1.05 (-6.01-8.18)

1.19 (0.98-1.43)

  1. *Statistically significant, p = 0.045, ¥All regressions adjusted for CAM intervention, ŦMultiple least-squares regressions for each predictor detailed. Adjusted forced entry (p < 0.1) and stepwise elimination and entry models retain multicenter (p = 0.03) and therapeutic studies (0.02).