We thank Dr Etminan (1) for interest in our study. We agree that the case number is small, and as the first report of this association, caution is warranted. However, we would like to clarify that the analysis of frequent use was based on 26 cases (Table 3) who were frequent users [not 14 as reported by Etminan (1)], including confirmed and self-reported cases without a medical record because the positive predictive value of self-report was 85%. Further, the risk estimate for ever use (hazard ratio = 1.80, 95% confidence interval = 1.12 to 2.88; Table 2) was based on a slightly larger number of exposed cases (n = 38).
With respect to the latent period for uterine cancer, we included a sensitivity analysis excluding the first year of follow-up and observed similar results (frequent user hazard ratio = 2.65, 95% confidence interval = 1.47 to 4.80; Supplementary Table 3, available online). Although we only assessed hair product use in the previous 12 months, we considered this to be a proxy for longer-term exposure that may be representative of use over several years or perhaps longer given that women initiate using these products at young ages (2). We encourage future studies to capture hair product use over the life course.
Residual confounding is always a concern in observational studies. Although there were imbalances in the distribution of oral contraceptive and hormone replacement therapy use by straightener and/or relaxer use, differences in the estimates with or without adjusting for these variables were negligible. Propensity score matching is an alternative to Cox regression for confounding control; however, it may also magnify existing biases (3) or unnecessarily inflate variances (4). Other approaches for addressing confounding, including propensity score matching, would be useful for triangulation of evidence but would not necessarily improve the estimates in our study.
Last, we did not consider cardiovascular disease (CVD) to be a confounder, as it is unlikely that CVD would lead to changes in hair product use. CVD-related death is a competing risk because death, a censoring event, would preclude us from observing future uterine cancer diagnoses. Potential bias related to competing risks could occur when there is a strong joint risk factor for CVD-related death and uterine cancer, beyond the adjusted covariates. We assume that censoring by competing risks is ignorable in our analyses given the exposures and included covariates in the Cox models and cumulative risk estimates. Thus, our approach is robust to associations between measured exposures, covariates, and competing risks such as CVD-related death.
No single study can definitively establish a causal relationship between straightener and/or relaxer use and uterine cancer. However, our findings, including the monotonic dose-response relationship, relatively large magnitude of effects, no evidence of confounding biasing results away from the null, along with consistency with associations seen for other hormonal cancers (2,5,6), provide strong evidence for a potential effect. Further, the disproportionate burden of aggressive uterine cancer subtypes experienced by Black women (7), who also more frequently use hair straighteners and/or relaxers, provides further justification for future studies investigating the relationship between straighteners and/or relaxers and uterine cancer incidence.
Contributor Information
Che-Jung Chang, Epidemiology Branch, Department of Health and Human Services, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC, USA.
Katie M O’Brien, Epidemiology Branch, Department of Health and Human Services, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC, USA.
Alexander P Keil, Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA.
Symielle A Gaston, Epidemiology Branch, Department of Health and Human Services, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC, USA.
Chandra L Jackson, Epidemiology Branch, Department of Health and Human Services, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC, USA; Intramural Research Program, National Institute on Minority Health and Health Disparities, Bethesda, MD, USA.
Dale P Sandler, Epidemiology Branch, Department of Health and Human Services, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC, USA.
Alexandra J White, Epidemiology Branch, Department of Health and Human Services, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC, USA.
Funding
This work was supported by the Intramural Research Program of the National Institute of Health, National Institute of Environmental Health Sciences (grant number Z01-ES044005, Z1AES103332-01).
Notes
Role of the funder: The funder had no role in the writing of this response or the decision to submit it for publication.
Disclosures: The authors have no conflicts of interest to disclose.
Author contributions: Che-Jung Chang, PhD (Conceptualization; Writing—original draft; Writing—review & editing); Katie M. O’Brien, PhD (Conceptualization; Writing—review & editing); Alexander P. Keil, PhD (Conceptualization; Writing—original draft; Writing—review & editing); Symielle A. Gaston, PhD (Writing—review & editing); Chandra L. Jackson, PhD (Writing—review & editing); Dale P. Sandler, PhD (Writing—review & editing); Alexandra J. White, PhD, MSPH (Conceptualization; Writing—review & editing).
Data availability
No new data were generated or analyzed in this response.
References
- 1. Etminan M. RE: Use of straighteners and other hair products and incident uterine cancer. J Natl Cancer Inst. 2022. doi: 10.1093/jnci/djac234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. White AJ, Gregoire AM, Taylor KW, et al. Adolescent use of hair dyes, straighteners and perms in relation to breast cancer risk. Int J Cancer. 2021;148(9):2255-2263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Pearl J. Invited commentary: understanding bias amplification. Am J Epidemiol. 2011;174(11):1223-1227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Brookhart MA, Schneeweiss S, Rothman KJ, Glynn RJ, Avorn J, Stürmer T.. Variable selection for propensity score models. Am J Epidemiol. 2006;163(12):1149-1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. White AJ, Sandler DP, Gaston SA, Jackson CL, O’Brien KM.. Use of hair products in relation to ovarian cancer risk. Carcinogenesis. 2021;42(9):1189-1195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Coogan PF, Rosenberg L, Palmer JR, Cozier YC, Lenzy YM, Bertrand KA.. Hair product use and breast cancer incidence in the Black Women’s Health Study. Carcinogenesis. 2021;42(7):924-930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Abel MK, Liao C-I, Chan C, et al. Racial disparities in high-risk uterine cancer histologic subtypes: a United States Cancer Statistics study. Gynecol Oncol. 2021;161(2):470-476. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No new data were generated or analyzed in this response.