Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 29.
Published in final edited form as: Phytother Res. 2019 Jun 13;33(8):1955–1956. doi: 10.1002/ptr.6394

Replacement of dropouts may bias results: Comment on ‘The effect of green tea ointment on episiotomy pain and wound healing in primiparous women: A randomized, double‐blind, placebo‐controlled clinical trial’

Lilian Golzarri-Arroyo 1, Stephanie L Dickinson 1, David B Allison 1
PMCID: PMC8240096  NIHMSID: NIHMS1712911  PMID: 31192504

Randomized Controlled Trials (RCT) are important for making unbiased estimates of treatment effects, but need to be performed carefully with proper analysis. While it is common to have subjects drop out of a study, this could lead to bias if they are not accounted for correctly in the statistical analysis (Wood et al., 2004; Juni, et Egger, 2005). Bell et al. show how even if the dropout rates are similar between study arms, the results for treatment effects can be biased; while having unequal dropout rates does not mean that the results will be biased (Bell et al., 2013). Therefore, it is necessary to account for the dropouts correctly, which is why Intention to Treat (ITT) analysis is preferred to provide unbiased results (Elobeid et al., 2009; Gupta, 2011; Hollis, & Campbell, 1999; Newel, 1992).

Shahrahmani et al. conducted an RCT on the effect of green tea ointment on episiotomy pain and wound healing in primiparous women (Shahrahmani et at., 2017), and in doing so, they replaced the dropouts with new participants. This presumably alleviates the problem of losing power, but it does not eliminate the potential bias of missing data from dropouts (Juni et al., 2001). Moreover, replacing dropouts can introduce new bias. The method of replacement used by Shahrahmani et al. is not clear. Referring to the ten subjects lost to follow-up (3 from the green tea group, 3 from the placebo ointment, and 4 from the group with no ointment), the authors state “subjects were replaced by other eligible women”, where the original number of 33 per group is maintained for analysis. If the next person enrolled was put into the treatment group of the most recent dropout, this is not a random process. If additional subjects are needed, the next participant enrolled should be assigned randomly to one of the three groups with equal probability, in the same manner as the rest of the study participants, thus maintaining the randomization allocation ratio (Shibadas, & Vinu, 2018). Compromising the randomization interferes with the validity of causal inferences. Additionally, assigning a new participant to the treatment group of a dropout outside the randomization procedure might mean that the allocation is not concealed if study staff know the dropout’s treatment group, which can lead to selection bias (Day, & Altman, 2000).

It is also puzzling that the researchers replaced the dropouts after they intentionally recruited extra participants to account for 20% potential attrition. The protocol submitted in the Iranian Registry of Clinical Trials (IRCT2016011225983N) does not contain any plan to replace subjects who dropout.

Because the 10 replacements were not truly randomized in Shahrahmani et al. (if our understanding is correct), they should not be included in the final analysis. An ITT analysis should be performed including the 10 participants who dropped out after randomization and excluding the 10 replacements if they were not randomly assigned to their treatment group according to the protocol.

The authors were contacted, but they were not willing to comment or share the data.

Funding:

This work was funded in part by the National Institute of Health (NIH): R25HL124208, R25DK099080, P30AG050886 and U24AG056053. The opinions expressed are those of the authors and not necessarily of the NIH or any other organization.

Conflict of Interest: Dr. Allison has received personal payments or promises for same from: American Society for Nutrition; American Statistical Association; Biofortis; Columbia University; Fish & Richardson, P.C.; Frontiers Publishing; Henry Stewart Talks; IKEA; Indiana University; Laura and John Arnold Foundation; Johns Hopkins University; Law Offices of Ronald Marron; MD Anderson Cancer Center; Medical College of Wisconsin; National Institutes of Health (NIH); Sage Publishing; The Obesity Society; Tomasik, Kotin & Kasserman LLC; University of Alabama at Birmingham; University of Miami; Nestle; WW (formerly Weight Watchers International, LLC). Donations to a foundation have been made on his behalf by the Northarvest Bean Growers Association. Dr. Allison is an unpaid member of the International Life Sciences Institute North America Board of Trustees. Dr. Allison’s institution, Indiana University, has received funds to support his research or educational activities from: NIH; Alliance for Potato Research and Education; American Federation for Aging Research; Dairy Management Inc; Herbalife; Laura and John Arnold Foundation; Oxford University Press.

Footnotes

The other authors declare that they have no conflict of interest.

Compliance with Ethical Standards:

Ethical approval: This article does not contain any studies with human participants or animals performed by any of the authors.

References:

  • 1.Wood AM, White IR, & Thompson SG. (2004). Are missing outcome data adequately handled? A review of published randomized controlled trials in major medical journals. Clinical Trials, 1, 368–76. [DOI] [PubMed] [Google Scholar]
  • 2.Juni P, & Egger M. (2005). Commentary: Empirical evidence of attrition bias in clinical trials. International Journal of Epidemiology, 34(1), 87–88. [DOI] [PubMed] [Google Scholar]
  • 3.Bell ML, Kenward MG, Fairclough DL, & Horton NJ. (2013). Differential dropout and bias in randomised controlled trials: when it matters and when it may not. BMJ, 346, e8668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Elobeid MA, Padilla MA, McVie T, Thomas O, Brock DW, Musser B, Lu K, Coffey CS, Desmond RA, St-Onge MP, Gadde KM, Heymsfield SB, & Allison DB. (2009). Missing Data in Randomized Clinical Trials for Weight Loss: Scope of the Problem, State of the Field, and Performance of Statistical Methods. PLoS One, 4(8), e6624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gupta SK. (2011). Intention-to-treat concept: A review. Perspect Clin Res, 2(3), 109–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Hollis S, & Campbell F. (1999). What is meant by intention to treat analysis? Survey of published randomised controlled trials. BMJ, 319, 670–674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Newel DJ. (1992). Intention-to-treat analysis: Implications for quantitative and qualitative research. Int J Epidemiol, 21, 837–41. [DOI] [PubMed] [Google Scholar]
  • 8.Shahrahmani H, Kariman N, Jannesari S, Rafieian‐Kopaei M, Mirzaei M, Ghalandari S, Shahrahmani N, & Mardani G. (2017). The effect of green tea ointment on episiotomy pain and wound healing in primiparous women: A randomized, double‐blind, placebo‐controlled clinical trial. Phytotherapy Research. 2018;32:522–530. [DOI] [PubMed] [Google Scholar]
  • 9.Juni P, Altman DG, & Egger M. (2001). Assessing the quality of controlled clinical trials. BMJ, 323:42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Shibadas B, & Vinu MJ. (2018). An Overview of Clinical Trial Operation: Fundamentals of clinical trial planning and management in drug development Edition 2. Shibadas Biswal. 211–213. [Google Scholar]
  • 11.Day SJ, & Altman DG. (2000). Blinding in clinical trials and other studies. BMJ, 321: 504. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES