Atormac
Neurology India
menu-bar5 Open access journal indexed with Index Medicus
  Users online: 2733  
 Home | Login 
About Editorial board Articlesmenu-bullet NSI Publicationsmenu-bullet Search Instructions Online Submission Subscribe Videos Etcetera Contact
  Navigate Here 
 Search
 
  
 Resource Links
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
    Article in PDF (484 KB)
    Citation Manager
    Access Statistics
    Reader Comments
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this Article
   References

 Article Access Statistics
    Viewed2248    
    Printed14    
    Emailed0    
    PDF Downloaded20    
    Comments [Add]    

Recommend this journal

 


 
Table of Contents    
CORRESPONDENCES
Year : 2018  |  Volume : 66  |  Issue : 2  |  Page : 583-584

Bias in randomized trials


Department of Community Medicine, Dr. Rajendra Prasad Government Medical College, Tanda, Kangra, Himachal Pradesh, India

Date of Web Publication15-Mar-2018

Correspondence Address:
Dr. Sunil K Raina
Department of Community Medicine, Dr. Rajendra Prasad Government Medical College, Tanda, Kangra, Himachal Pradesh
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/0028-3886.227290

Rights and Permissions



How to cite this article:
Raina SK. Bias in randomized trials. Neurol India 2018;66:583-4

How to cite this URL:
Raina SK. Bias in randomized trials. Neurol India [serial online] 2018 [cited 2019 Oct 23];66:583-4. Available from: http://www.neurologyindia.com/text.asp?2018/66/2/583/227290




I went through with interest the article entitled “A randomized placebo-controlled trial of progesterone with or without hypothermia in patients with acute severe traumatic brain injury”, published in Neurology India (2017;65:1304-11).[1] The authors deserve credit for their efforts.

Randomized controlled trials (RCTs) with a robust study design are viewed as the “gold standard” for comparing different interventions, as when performed correctly, randomization guarantees an unbiased estimate of the treatment effect. The authors seem to have taken care of the robustness of the design by performing “blinding and allocation concealment” in recruitment. However, the study design of this study used restricted randomization as the authors recruited 107 patients in four groups in a ratio of 1:1.

I would like to share my concerns with the authors regarding the design of this study. RCTs despite using adequate allocation concealment suffer from selection bias. This selection bias arises from restricted randomization (as used by authors in this study). Restricted randomization forces the treatment groups to be similar in some manner, for example, by requiring that a similar number of patients are assigned to each treatment group. This means that the imbalance between groups (intervention and control) can never be more than two patients. It has been reported that when the imbalance between groups is 1, the recruiter will be able to correctly guess the next allocation with 83% probability; when the imbalance is 2, this probability increases to 100%.[2] The answer lies in conducting simple randomization (sometimes also referred to as “complete” or “unrestricted” randomization). Simple randomization works by assigning each patient to one of the treatment groups with a certain probability (usually 50%). This probability (50%) is the same for every patient, regardless of previous allocations. For example, in a trial where 30 of the first 40 patients were assigned to the intervention group and only 10 to the control group, when the 31st patient presents for randomization, the patient would still have an equal chance of being assigned to either treatment groups, regardless of the imbalance in numbers. As the probability stays the same, recruiters or allocators will not be able to guess with any accuracy which treatment the patient will be assigned to (as they would essentially be trying to guess the results of a coin-flip). This reduces selection bias. Agreed that simple randomization is infrequently used in practice probably to provide a balance in the number of patients assigned to each treatment group,[3],[4] the fact remains that, provided the overall sample size is not too small, this lack of balance has only a very small impact on power and should not be used as a reason to avoid simple randomization.[5]

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Sinha S, Raheja A, Samson N, Goyal K, Bhoi S, Selvi A, Sharma BS. A randomized placebo-controlled trial of progesterone with or without hypothermia in patients with acute severe traumatic brain injury. Neurol India 2017;65:1304-11.  Back to cited text no. 1
[PUBMED]  [Full text]  
2.
Kahan BC, Rehal S, Cro S. Risk of selection bias in randomised trials. Trials 2015;16:405.  Back to cited text no. 2
    
3.
Kahan BC, Morris TP. Reporting and analysis of trials using stratified randomisation in leading medical journals: Review and reanalysis. BMJ 2012;345:e5840.  Back to cited text no. 3
    
4.
Hewitt CE, Torgerson DJ. Is restricted randomisation necessary? BMJ 2006;332:1506-8.  Back to cited text no. 4
    
5.
Rosenberger WF, Lachin JM. Randomization in clinical trials. Chichester: Wiley; 2005.  Back to cited text no. 5
    




 

Top
Print this article  Email this article
   
Online since 20th March '04
Published by Wolters Kluwer - Medknow