Correlates of hair loss in adult women of African descent in London, U.K.: findings of a cross‐sectional study

DEAR EDITOR, The U.K. has seen a doubling of the national proportion of black Africans between 2001 and 2011, and in the capital city London, 15 6% of the population is of full or partial Afro-Caribbean descent. Given the high burden of hair and scalp disorders in this cohort, it is likely that these patients will comprise an increasingly significant part of the workload of dermatologists. However, there remains a paucity of research originating from the U.K. in relation to this challenging area of dermatology. In light of this, we undertook a cross-sectional study to determine the prevalence and establish correlates of hair loss in women of African descent living in London. Women of African descent (≥ 18 years) were recruited from staff, students, patients and visitors (including female friends and relatives) at a hospital in London over a 9-month period (Fig. 1) following ethical approval (Research and Ethics Committee Reference: 12/LO/1758). After providing informed written consent, subjects were asked to complete a validated, 36-point questionnaire (Table 1). A total of 248 subjects were recruited for the study; the final data analysis consisted of 242 subjects (six subjects were excluded due to uncertainty about their hair-loss status). Using the self-reported presence or absence of hair loss as the dependent variable, univariate logistic regression analysis was performed for each of the questionnaire items as independent predictors. These comparisons were based on those with known values of a particular factor. As the probability of hair loss was associated with age, multiple logistic regression analysis was used to assess the relation of each independent factor concurrently, with age as a continuous covariate. A P-value of < 0 05 was considered significant. The odds ratios (ORs) cited represent the age-adjusted analyses in models where age was significant and unadjusted univariate ORs were used where age was not a significant factor in the model. In keeping with other published epidemiological studies, we observed a significant burden of hair loss in our cohort (a sample proportion of self-reported hair loss was 54%). We also identified several statistically significant correlates of hair loss in this cohort (Fig. 2). Increasing age was positively correlated with hair loss; subjects with hair loss had a greater mean age [44 8 years (range Table 1 An abridged version of the London Hair Grooming Assessment Instrument