Discontinuation* | Non-publication | |||||||
Univariable | Multivariable† | Univariable | Multivariable‡ | |||||
OR (95% CI) | p Value | OR (95% CI) | p Value | OR (95% CI) | p Value | OR (95% CI) | p Value | |
SNSF RCTs vs investigator-initiated RCTs without SNSF support§ | 1.18 (0.59 to 2.34) | 0.632 | 1.05 (0.51 to 2.11) | 0.898 | 0.93 (0.49 to 1.75) | 0.832 | 1.18 (0.54 to 2.55) | 0.681 |
SNSF RCTs vs industry-initiated RCTs¶ | 5.34 (2.63 to 10.94) | <0.001 | 3.84 (1.68 to 8.74) | <0.01 | 1.64 (0.91 to 2.92) | 0.096 | 1.12 (0.50 to 2.43) | 0.772 |
*Only RCTs discontinued due to slow recruitment.
†Adjusted for single-centre (vs multicentre) and planned sample size.
‡Adjusted for discontinuation (due to slow recruitment), single-centre (vs multicentre) and planned sample size.
§Complete-case multivariable logistic regression with 22 missings for completion status analysis and 25 missings for publication status analysis, from a total of 179.
¶Complete-case multivariable logistic regression with 20 missings for completion status analysis and 23 missings for publication status analysis, from a total of 313.
RCT, randomised clinical trial; SNSF, Swiss National Science Foundation.