This trial describes effects of scalp cooling on hair protection (HP) and regrowth (HR) in breast cancer patients [1].
The authors did not describe the rationality of using 2:1 instead of 1:1 randomization. The 2:1 randomization needs higher sample size and reduces validity. The ratio of 2:1 may be advantageous in certain situations but this study had none [2].
The PI assessed the outcome despite being aware of the treatment status introducing an observer bias with unpredictable impact on the odds ratio (underestimation or overestimation). Blinded assessment by the independent observer could avoid this bias [3].
Randomization did not create balanced groups probably due to small sample size [4]. The ‘scalp cooling” group was younger, and the control group had more patients with hypothyroidism. Age and hypothyroidism are potential confounders [5]. This author’s analysis shows that the proportion of hypothyroid patients are comparable (p = 0.27). The ‘age’ variable could not be analyzed. Finally, the authors concluded that anthracycline arm had inferior outcome without reporting odds ratio, 95% CI, and p values of the regression model. Authors could have included age in the model to assess confounding. To conclude, the study design and biases reduce the validity and generalizability.
Funding
None.
Ethical approval
This is a correspondence letter only. There is no human or animal research involved; thus, ethical committee approval is not required.
Declaration of competing interest
Author declare no conflict of interest.
References
- 1.Bajpai J., Kagwade S., Chandrasekharan A., Dandekar S., Kanan S., Kembhavi Y. Randomised controlled trial of scalp cooling for the prevention of chemotherapy induced alopecia. Breast. 2019 Dec 12;49:187–193. doi: 10.1016/j.breast.2019.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hey S.P., Kimmelman J. The questionable use of unequal allocation in confirmatory trials. Neurology. 2014 Jan 7;82(1):77–79. doi: 10.1212/01.wnl.0000438226.10353.1c. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Szklo M., Nieto F.J. Understanding lack of validity:bias. In: Szklo M., Nieto F.J., editors. Epidemiology : beyond the basics. third ed. Jones & Bartlett Learning; Burlington, Mass: 2014. pp. 120–121. [Google Scholar]
- 4.Gordis L. Assessing preventive and therapeutic measures: randomized trilas. In: Gordis L., editor. Epidemiology. fifth ed. Elsevier; Philadelphia, Pennsylvania: 2014. p. 145. [Google Scholar]
- 5.Phillips T.G., Slomiany W.P., Allison R. Hair loss: common causes and treatment. Am Fam Physician. 2017;96(6):371. [PubMed] [Google Scholar]