Review articles

By Dr. William Kent
Corresponding Author

Submitting Author Dr. William Kent
Other Authors Dr. William Kent
Royal Sussex County Hospital, Brighton, - United Kingdom


Clinical trials

Kent W. Randomisation and blinding in clinical trials. WebmedCentral CLINICAL TRIALS 2012;3(5):WMC003289
doi: 10.9754/journal.wmc.2012.003289

This is an open-access article distributed under the terms of the Creative Commons Attribution License(CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Submitted on: 13 May 2012 06:15:37 PM GMT
Published on: 14 May 2012 12:19:38 PM GMT


High quality clinical trials are designed to reduce bias, to increase the confidence that the reader can place in the conclusions of the trial. This article evaluates the role of randomisation and blinding to prevent bias and how to best implement them in a trial.


The quality of a trial can be described as the confidence that the design, conduct, report, and analysis restrict bias.[1] Bias is defined as systematic errors in clinical trials that result in a difference between the results of a study and reality.[2,3] This can lead to inaccurate results and wrong conclusions about the effectiveness of an intervention. Good study design is required to minimise bias4 because it can be introduced at any stage of trial and cannot be measured or controlled for by statistical methods.[2] This article will focus on two methods of preventing bias in clinical trials: randomisation and blinding.


Three possible explanations can account for any difference between groups at the end of a trial: 1) a real effect; 2) chance; and 3) a systematic difference between the groups other than the intervention (bias).[5] Randomisation is a technique used to reduce this third possibility by preventing selection bias and confounding factors. Participants are allocated in a random and unpredictable way to the different groups in the trial.[6] Providing enough participants are involved in the trial the random allocation produces groups which on average are as alike as possible. The assumption being that along with known, measured characteristics, unknown and unmeasured characteristics will also be equally distributed between the groups eliminating their possible influence on the results.[1,2,7] Randomisation also facilitates blinding of treatments which can reduce bias after intervention allocation. It also permits the use of probability theory to express the likelihood that any difference in outcomes between the groups is due to chance.[8]
If randomisation is not used then predictive factors tend to be unevenly distributed between the groups in a trial.[1] This is the result of selection bias through conscious or unconscious manipulation of the allocation sequence by the researchers. For example if a researcher allocates people with a less favourable prognosis to the control group, then (assuming that the intervention is not significantly worse than the control) a positive outcome can be expected for the intervention group irrespective of whether it is in reality any better or not. This uneven allocation may occur because of financial motivation or because of good intentions of the researchers to try and get the best treatments for a patient. Irrespective of the underlying motivation bias is introduced which can invalidate the results of a trial. To avoid this, rigorous methods to prevent researchers from knowing or manipulating the allocation sequence should be implemented. Successful randomisation therefore involves three stages: random sequence generation, concealed allocation, and implementation. Each stage should be adequately described in the methods of a trial so that the likelihood of bias in group allocation can be evaluated.[2],  

Random sequence generation
True randomisation is necessary because some methods of allocation such as alternate allocation, or methods based on patient characteristics (Table 1) are not reliably random and are easily predicted and manipulated.[2] In general terms two different types of randomisation exists:
Fixed randomisation: the randomisation method is defined and sequences set up before the start of the trial (eg simple and restricted randomisation).
Adaptive randomisation: randomisation is adjusted as the participants are recruited to account for imbalances in numbers or characteristics of the participants (eg minimisation).[2]
Simple randomisation is equivalent to fair coin tossing whereby each participant has an equal chance of being allocated to each group. However, manual methods such as coin tossing and dice rolling are not used due to issues with concealment, validation and reproducibility. Instead random number sequences generated by a computer or a textbook are usually used. These methods are easy, unpredictable, reliable and have the advantage of producing an audit trial.[8]  
Restricted randomisation is used to control randomisation to achieve balance between groups in size (block randomisation) or in size and specified characteristics (stratified randomisation).[8] It is especially relevant for sample sizes of less than 200 as due to the play of chance simple randomisation may not produce groups with similar numbers and comparable participant characteristics.[6] However, these methods do not offer any more protection against bias then simple randomisation.[8]
Adaptive or dynamic random allocation methods seek to have balanced groups throughout the study. Minimisation is the only alternative to randomisation recognised by the CONSORT group.[6] The first participant is randomly allocated then the following participants are allocated so as to minimise the imbalance between groups at that time of the study. It ensures balance between groups for size and identified important participant characteristics at all times of the study and is the only method that can guarantee similar groups in a trial.[9]

Allocation concealment
Allocation concealment is used to prevent selection bias before the assignment of interventions and can always be implemented.[10,11] It is different from blinding which is a method used to protect against bias after intervention allocation and cannot always be implemented.[6] Without adequate allocation concealment even random, unpredictable allocations can be corrupted.[1,6] Each investigator in the trial will have an agenda grounded in their personal opinions about the intervention and instances of researchers sabotaging the concealed allocation to subvert the random allocation have been described.[10] The reasons given varied from an intellectual challenge or doing what they believed was right for a patient to deliberate sabotage or manipulation to confirm a belief.[10] Well meaning or not, such tampering undermines the validity of a trial. Most examples of subverted randomisation occur because the methods used to conceal the allocation were inadequate. Robust allocation concealment removes the opportunity for the conscious or unconscious manipulation of participant allocation should therefore be implemented (table 2).

Implementation & Empirical evidence
Strong empirical evidence has been reported that inadequate sequence generation and allocation concealment are associated with overestimates of treatment effects.[1,7,11] These findings support the use of randomisation in clinical trials and as a criterion in critical appraisal. It is therefore important that the methods of randomisation and how it was implemented are adequately described.

Blinding is the method of keeping participants, and the researchers involved in a trial unaware of the assigned interventions after allocation. It is utilised to avoid participants’ or investigators’ expectations impacting upon the results.[1] Knowledge, experiences and beliefs can all influence physical and psychological responses to an intervention.[12] Its use is supported by research comparing adequately blinded trials with inadequately blinded trials which found that knowledge of intervention allocation is associated with larger estimates of treatment effects.[13]  
Different levels of blinding have been described (figure 2) but their meanings are poorly understood[13-15] and a lack of consistency in their implementation adds to this confusion.[1,13-15] Therefore, the methods used should be explicitly stated to allow readers to assess the adequacy of the blinding and evaluate any impact that bias may have had on the trial.[15] In an attempt to increase and improve the use of blinding, descriptions of blinding methods have been published for both pharmacological[13] and non pharmacological interventions[14] and guidelines have been written for the correct reporting of blinding.[6]
Blinding of different individuals involved in a trial has many potential benefits which can improve the credibility of trial conclusions.[12] Blinding makes unintentional or intentional bias of results difficult, improves compliance, and improves participant retention.[15] Careful consideration needs to given to the intervention procedures so that differences do not cause unblinding. For example, pharmacological trials should include a placebo or alternative treatment that is identical in appearance (size, colour, weight, feel, odour etc) and route of administration.[4] Nonpharmacological interventions are often harder to blind.[14] The lack of blinding does not automatically indicate a methodologically unsound trial as the design may prevent bias in other ways (eg objective outcome measures).[12,13] Furthermore the influence of blinding on reducing bias has yet to be fully ascertained. Studies are required to assess its influence on preventing bias because its actual impact may be considerably different to its theoretical influence.[15]

Discussions and Conclusion

All trials have limitations but good study design reduces the potential for bias and combined with the good reporting increases our confidence in the conclusions of a trial. Randomisation can be used to prevent selection bias and confounding factors in a trial. Whereas, blinding reduces the opportunity for bias to be introduced after the interventions have been allocated. However, randomisation and blinding do not guarantee a methodologically sound trial especially as these techniques are often poorly understood and implemented. To be successful these methods need to be robustly implemented to prevent even determined individuals from influencing the results of a trial. The methods used must also be explicitly reported so that the overall quality and credibility of a trial and its conclusions can be evaluated with confidence.


1. Gluud LL. Bias in clinical intervention research. American journal of epidemiology. 2006; 163(6) 493-501.
2. Gosall NK, Gosall GS. The doctor’s guide to critical appraisal. PasTest: Knutsford, UK. 2006.
3. Petrie A, Sabin C. Medical statistics at a glance. 2nd edition. Blackwell: oxford. 2005.
4. Forder PM, Gebski VJ, Keech AC. Allocation concealment and blinding: when ignorance is bliss. MJA. 2005; 182(2):87-89.
5. Beller EM, Gebski V, Keech AC. Randomisation in clinical trials. MJA. 2002; 177; 565-567.
6. Altman DG, Schulz KF, Moher D, et al. for the CONSORT Group. The revised CONSORT statement for reporting randomised trials: explanation and elaboration. Ann Intern Med 2001; 134: 663-694.
7. Kunz R, Vist G, Oxman AD. Randomisation to protect against selection bias in healthcare trials. Cochrane database of systematic reviews 2007, issue 2.
8. Schultz KF, Grimes DA. Generation of allocation sequences in randomized trials: chance, not choice. Lancet. 2002; 359: 515-519.
9. Treasure T, MacRae KD. Minimisation: the platinum standard for trials? Randomisation doesn't guarantee similarity of groups; minimisation does. BMJ. 1998; 317:362-3.
10. Schultz KF, Grimes DA. Allocation concealment in randomized trials: defending against deciphering. Lancet. 2002; 359: 614-618.
11. 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;273:408-12.
12. Day SJ, Altman DG. Blinding in clinical trials and other studies. BMJ. 2000; 321:504.
13. Boutron I, Guillet L, Estellat C, Moher D, Hrobjartsson A, Ravaud P.  Reporting methods of blinding in randomized trials assessing pharmacological treatments: Asystematic review. PLOS Medicine. 2006; 3(10): 1931-1939.
14. Boutron I, Guillet L, Estellat C, Moher D, Hrobjartsson A, Ravaud P.  Reporting methods of blinding in randomized trials assessing nonpharmacological treatments. PLOS Medicine. 2007; 4(2): 370-380.
15. Schultz KF, Grimes DA. Blinding in randomized trials: hiding who got what. Lancet. 2002; 359: 696-700.

Source(s) of Funding


Competing Interests



This article has been downloaded from WebmedCentral. With our unique author driven post publication peer review, contents posted on this web portal do not undergo any prepublication peer or editorial review. It is completely the responsibility of the authors to ensure not only scientific and ethical standards of the manuscript but also its grammatical accuracy. Authors must ensure that they obtain all the necessary permissions before submitting any information that requires obtaining a consent or approval from a third party. Authors should also ensure not to submit any information which they do not have the copyright of or of which they have transferred the copyrights to a third party.
Contents on WebmedCentral are purely for biomedical researchers and scientists. They are not meant to cater to the needs of an individual patient. The web portal or any content(s) therein is neither designed to support, nor replace, the relationship that exists between a patient/site visitor and his/her physician. Your use of the WebmedCentral site and its contents is entirely at your own risk. We do not take any responsibility for any harm that you may suffer or inflict on a third person by following the contents of this website.

2 reviews posted so far

Randomisation and blinidng in clinical trials
Posted by Dr. Abdul M Ruknudin on 28 May 2012 02:35:42 PM GMT

Randomisation and blinding in clinical trials
Posted by Dr. Karim Hmadcha on 15 May 2012 05:52:57 PM GMT

0 comments posted so far

Please use this functionality to flag objectionable, inappropriate, inaccurate, and offensive content to WebmedCentral Team and the authors.


Author Comments
0 comments posted so far


What is article Popularity?

Article popularity is calculated by considering the scores: age of the article
Popularity = (P - 1) / (T + 2)^1.5
P : points is the sum of individual scores, which includes article Views, Downloads, Reviews, Comments and their weightage

Scores   Weightage
Views Points X 1
Download Points X 2
Comment Points X 5
Review Points X 10
Points= sum(Views Points + Download Points + Comment Points + Review Points)
T : time since submission in hours.
P is subtracted by 1 to negate submitter's vote.
Age factor is (time since submission in hours plus two) to the power of 1.5.factor.

How Article Quality Works?

For each article Authors/Readers, Reviewers and WMC Editors can review/rate the articles. These ratings are used to determine Feedback Scores.

In most cases, article receive ratings in the range of 0 to 10. We calculate average of all the ratings and consider it as article quality.

Quality=Average(Authors/Readers Ratings + Reviewers Ratings + WMC Editor Ratings)