The Development and Validation of the Hair Shedding Visual Scale for Asian Women (HSVS‐A)
Xiaojin Shi,Myung-Ha Song,Yi Zhao,Hsiaohan Tuan,Xiaona Na
DOI: https://doi.org/10.1111/1346-8138.16989
2023-01-01
Abstract:Female pattern hair loss (FPHL) and telogen effluvium (TE) are the most common forms of non-scarring hair loss in women,1 which have adverse effects on appearance, quality of life, and mental health.1-3 In the early stages of FPHL, increased hair shedding is a common and obvious symptom as well as a reduction in hair density.2 Various classification methods of FPHL, such as the Ludwig scale and the Sinclair scale,4, 5 focused on the hair density of the central area over the mid scalp and were of limited help in the assessment of hair shedding. Previous studies have designed and verified visual scales to assess hair shedding in FPHL in Caucasian women.6, 7 However, hair features, such as caliber, curliness, and color, can vary by ethnicity and affect the visual assessment of hair loss. Moreover, patients with TE who complain of excessive hair shedding often struggle to quantify the exact number of hairs shed.3 Therefore, we aimed to establish and validate the Hair-Shedding Visual Scale for Asian Women (HSVS-A) to distinguish between healthy subjects and women with FPHL and/or TE. We developed the HSVS-A by collecting naturally shed hair from Asian women with short, medium, and long hair (Figure 1), and measured the length, caliber, and curliness of the hairs (Table S1). Each hair length group (short, medium, and long) was divided into 10 grades according to the number of shed hairs. We curled the hair into clumps of the same diameter and photographed 10 bundles of hair placed on the same white A4 paper (Figure 1). There were 205 adult Chinese women enrolled by convenience sampling from communities and one clinic in Beijing, China (Figure S1). All the subjects received a trichoscopic examination and were diagnosed with a type of alopecia by professional dermatologists. The details of the study design are shown in the Supporting Information Data S1. This study was approved by the ethics committee of Tsinghua University. We conducted a two-part, diagnostic study to assess the reliability and validity (Part 1) and the viability (Part 2) of the HSVS-A. On the initial visit, all subjects were asked to select the picture in HSVS-A that best correlated with the extent of hair shedding during an average shampooing day. Next, we required 187 subjects in Part 1 to wash their hair at least once within 48 h and select again the best correlated picture during an average shampooing day (second visit). The reliability of the HSVS-A score was evaluated by the Spearman's rank correlation coefficient (ρ) and Cohen's kappa coefficient (κ) between the two visits. Validity was evaluated by constructing receiver operating characteristic (ROC) curves and calculating the area under curve (AUC) about the self-reported HSVS-A score and the hair loss diagnosis. Additionally, to evaluate the consistency between visual and actual hair loss, we asked 18 subjects in Part 2 to wash their hair every other day for seven consecutive times and to collect the hair shed during each shampoo in separate zip-lock bags. HSVS-A scores for the hairs in all the bags were also reported (Table S4). Two operators manually counted the number of hairs in parallel. Viability was evaluated by conducting a Pearson's correlation analysis between the HSVS-A score and manual counting of hair loss (details are described in Supporting Information Data S1). The mean shedding scores were 2.32–3.13 in healthy women, 3.61–4.20 in FPHL, 4.13–5.43 in TE, and 4.67–5.17 in combined FPHL and TE (Tables S2, S3). Spearman's coefficient of test–retest was 0.649–0.864, and significant. ROC for the HSVS-A produced an AUC of 0.717 (95% CI 0.629–0.802) for FPHL and 0.872 (95% CI 0.809–0.934) for TE, with an optimal cut-off score of 3.25 for both (Figure S2). Moreover, Figure S3 and Table S4 show the distributions and strong correlations between HSVS-A results and manual counting. Pearson's correlation coefficients for subjects with short, medium and long hair were 0.818, 0.950 and 0.799, respectively (p < 0.001). To the best of our knowledge, the HSVS-A is the first visual scale to reflect hair shedding for Asian women with black and straight hair and showed good reliability and validity to identify FPHL and TE and had good concordance with the actual hair loss. Unlike the previous studies,6, 7 we further considered the scale utility for TE and conducted ROC analysis to calculate the best cut-off scores for Asian patients with TE or FPHL. Our study shows that the HSVS-A is an effective, fast, and simple tool that could be used in clinics as a quick screening tool for Asian females with different hair lengths. Given the differences in hair features among Asians, scale validation is still required in varied Asian populations. None declared. This study was reviewed and approved by the ethics committee of Tsinghua University (project number: 20220079). The authors confirm that the ethical policies of the journal have been adhered to and appropriate ethical review committee approval has been received. Data S1: Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.