Elsevier

The Lancet

Volume 359, Issue 9305, 9 February 2002, Pages 515-519
The Lancet

Series
Generation of allocation sequences in randomised trials: chance, not choice

https://doi.org/10.1016/S0140-6736(02)07683-3Get rights and content

Summary

The randomised controlled trial sets the gold standard of clinical research. However, randomisation persists as perhaps the least-understood aspect of a trial. Moreover, anything short of proper randomisation courts selection and confounding biases. Researchers should spurn all systematic, non-random methods of allocation. Trial participants should be assigned to comparison groups based on a random process. Simple (unrestricted) randomisation, analogous to repeated fair coin-tossing, is the most basic of sequence generation approaches. Furthermore, no other approach, irrespective of its complexity and sophistication, surpasses simple randomisation for prevention of bias. Investigators should, therefore, use this method more often than they do, and readers should expect and accept disparities in group sizes. Several other complicated restricted randomisation procedures limit the likelihood of undesirable sample size imbalances in the intervention groups. The most frequently used restricted sequence generation procedure is blocked randomisation. If this method is used, investigators should randomly vary the block sizes and use larger block sizes, particularly in an unblinded trial. Other restricted procedures, such as urn randomisation, combine beneficial attributes of simple and restricted randomisation by preserving most of the unpredictability while achieving some balance. The effectiveness of stratified randomisation depends on use of a restricted randomisation approach to balance the allocation sequences for each stratum. Generation of a proper randomisation sequence takes little time and effort but affords big rewards in scientific accuracy and credibility. Investigators should devote appropriate resources to the generation of properly randomised trials and reporting their methods clearly.

Section snippets

Non-random methods masquerading as random

Ironically, many researchers have decidedly non-random impressions of randomisation.8, 9, 10 They often mistake haphazard approaches and alternate assignment approaches as random.11 Some medical researchers even view approaches antithetical to randomisation, such as assignment to intervention groups based on preintervention tests, as quasirandom.12 Quasirandom, however, resembles quasipregnant, in that they both elude definition. Indeed, anything short of proper randomisation opens limitless

Simple (unrestricted) randomisation

Elementary yet elegant describes simple randomisation (panel 3).21 Although the most basic of allocation approaches, analogous to repeated fair coin-tossing, this method preserves complete unpredictability of each intervention assignment. No other allocation generation approach, irrespective of its complexity and sophistication, surpasses the unpredictability and bias prevention of simple randomisation.22

The unpredictability of simple randomisation, however, can also be a disadvantage.23 With

Restricted randomisation

Restricted randomisation procedures control the probability of obtaining an allocation sequence with an undesirable sample size imbalance in the intervention groups.20 In other words, if researchers want treatment groups of equal sizes, they should use restricted randomisation.

Stratified randomisation

Randomisation can create chance imbalances on baseline characteristics of treatment groups.28 Investigators sometimesavert imbalances by use of prerandomisation stratification on important prognostic factors, such as age or disease severity. In such instances, researchers should specify the method of restriction (usually blocking). To reap the benefits of stratification, investigators must use a form of restricted randomisation to generate separate randomisation schedules for stratified subsets

Separation of generation and implementation

Investigators often neglect, usually unintentionally, one other important element of randomised controlled trial design and reporting. With all approaches, the people who generated the allocation scheme should not be involved in ascertaining eligibility, administering treatment, or assessing outcome. Such an individual would usually have access to the allocation schedule and thus the opportunity to introduce bias.8 Faults in this trial component might represent a crevice through which bias

Conclusion

Randomised controlled trials set the methodological standard of excellence in medical research. The key word is randomised, which must be done properly. Generation of a randomisation sequence takes little time and effort but affords big rewards in scientific accuracy and credibility. Investigators should devote appropriate resources to doing the generation properly and reporting their methods clearly.

We thank Willard Cates and David L Sackett for their helpful comments on an earlier version of

References (31)

  • Streptomycin treatment of pulmonary tuberculosis

    BMJ

    (1948)
  • R Peto

    Why do we need systematic overviews of randomized trials?

    Stat Med

    (1987)
  • KF Schulz et al.

    Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials

    JAMA

    (1995)
  • KF Schulz

    Subverting randomization in controlled trials

    JAMA

    (1995)
  • KF Schulz

    Unbiased research and the human spirit: the challenges randomized controlled trials

    CMAJ

    (1995)
  • Cited by (577)

    View all citing articles on Scopus
    View full text