Subversion of allocation concealment in a randomised controlled trial: a historical case study

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Medical Research Integrity

Background:

• Randomization is crucial for unbiased clinical trials, but subversion can introduce selection bias.
• Sealed opaque envelopes, despite known vulnerabilities, are still used in over 10% of trials.
• Lack of empirical data on subversion effects contributes to the continued use of less secure methods.

Purpose of the Study:

• To empirically quantify the effects of subversion in clinical trial randomization.
• To compare the security of sealed envelope randomization versus centralized telephone allocation.
• To provide evidence for discontinuing insecure randomization methods.

Main Methods:

• A historical before-and-after study design was employed.
• Compared sealed envelope (local) vs. centralized telephone (remote) randomization in a multi-center surgical trial (654 patients).
• Analyzed key prognostic variables, specifically patient age at trial entry, between randomization methods.

Main Results:

• The sealed envelope system led to significantly lower median patient age in the experimental group for overall and specific clinician cohorts (p < 0.01).
• This indicates corrupted randomization due to inadequate allocation concealment with sealed envelopes.
• No significant age differences were observed between allocation groups using the centralized system.

Conclusions:

• Inadequate allocation concealment with sealed envelopes corrupted the randomization process for some patients.
• Centralized randomization provides secure and verifiable treatment allocation.
• The use of sealed envelopes in clinical trials should be discontinued unless absolutely necessary.