SeriesGeneration of allocation sequences in randomised trials: chance, not choice
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)
Statistical properties of randomization in clinical trials
Control Clin Trials
(1988)- et al.
Randomization in clinical trials: conclusions and recommendations
Control Clin Trials
(1988) Properties of simple randomization in clinical trials
Control Clin Trials
(1988)- et al.
Properties of permuted-block randomization clinical trials
Control Clin Trials
(1988) - et al.
Properties of the urn randomization in clinical trials
Control Clin Trials
(1988) - et al.
The CONSORT statement: revised recommendations for improving the quality of reports or parallel-group randomised trials
Lancet
(2001) The clinical trial
N Engl J Med
(1952)The Design of Experiments
(1935)The role of randomization in clinical trials
Stat Med
(1982)Controlled trials: the 1948 watershed
BMJ
(1998)
Streptomycin treatment of pulmonary tuberculosis
BMJ
Why do we need systematic overviews of randomized trials?
Stat Med
Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials
JAMA
Subverting randomization in controlled trials
JAMA
Unbiased research and the human spirit: the challenges randomized controlled trials
CMAJ
Cited by (577)
A comparison of 2 doses of antenatal dexamethasone for the prevention of respiratory distress syndrome: an open-label, noninferiority, pragmatic randomized trial
2024, American Journal of Obstetrics and GynecologyComparison of Transversus Thoracis Muscle Plane Block and Pecto-Intercostal Fascial Plane Block for enhanced recovery after pediatric open-heart surgery
2023, Anaesthesia Critical Care and Pain MedicineEffect of anodized titanium abutment collars on peri-implant soft tissue: A split-mouth clinical study
2023, Journal of Prosthetic Dentistry