How often does unclear allocation concealment in randomised trials reflect inadequate reporting of adequate methods?

Article type
Authors
Pildal J, Chan A, Hrobjartsson A, Forfang E, Altman D, Gotzsche P
Abstract
Background: It is not known to which extent the association between unclear reporting of allocation concealment in randomised controlled trials (RCTs) [1,2,3,4] and inflation of the effect of the experimental intervention reflects inadequate allocation concealment and ensuing selection bias, or other biases. An observational study of RCTs performed by the Radiation Therapy Oncology Group found that unclear concealment in the trial report did not reflect inadequate concealment according to the protocol; however, all included trials regarded radiation therapy and employed central randomisation 5. Thus, this finding is unlikely to be generalisable[1]. Here we address the same question using a more representative sample.

Objectives: Primary: to estimate how often RCTs reports with unclear allocation concealment had adequate allocation concealment according to the protocol.

Methods: Two pairs of authors independently categorized allocation concealment in 102 sets of trial protocols and corresponding reports as adequate, unclear, inadequate, or no available information according to two sets of criteria outlined in table 1. The protocols were of randomised trials (except dentistry) approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg in 1994 and 1995.

Results: According to predefined strict criteria, only 3 trial reports described adequate allocation concealment and 1 described inadequate concealment. 14 of the other 98 trials had adequate concealment according to the protocol (14% (95% confidence interval 9% to 23%)). Even when post hoc defined loose criteria were applied, 19 trial reports had adequate and 1 had inadequate allocation concealment, while only 33 of the remaining 82 trials with unclear concealment had adequate concealment according to the protocol (40% (29% to 51%)).

Conclusion: Most RCTs with unclear allocation concealment in the trial report also had unclear allocation concealment according to the protocols. This is compatible with the notion of inadequate reporting reflecting inadequate methods, but does not prove it. The disclosed lack of clarity in the protocols calls for dual action: more explicit Good Clinical Practice guidelines for protocols writing and public access to protocols by legislation.

Reference: 1. Soares HP, Daniels S, Kumar A et al. Bad reporting does not mean bad methods for randomised trials: observational study of randomised controlled trials performed by the Radiation Therapy Oncology Group. BMJ. 2004; 328(7430):22-4. 2. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA. 1995; 273(5):408-12. 3. Moher D, Pham B, Jones A et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses? Lancet. 1998; 352(9128):609-13. 4. Kjaergard LL, Villumsen J, Gluud C. Reported methodologic quality and discrepancies between large and small randomized trials in meta-analyses. Ann Intern Med 2001; 135(11):982-9. 5. Juni P, Altman DG, Egger M. Systematic reviews in health care: Assessing the quality of controlled clinical trials. BMJ. 2001; 323(7303):42-6.