Yes, let's abandon race—It does not accurately correlate with hair form

To the Editor: I read with interest the recent reports by Bigby et al in the Journal1, 2, 3 and thank them for raising this important issue. There is no doubt that there are variable human traits which probably resulted from environmental adaptation,4 and that some traits are likely to cluster in “race” groups that were under similar environmental factors as people who migrated “out of Africa.” The most obvious of these would be skin color and hair form.

Probably the most exhaustive study of hair form examined 8 parameters in hair from 7 populations.5 Participants of Asian ancestry were found to have the largest hair diameter and most consistent medulla. East African hair had the highest kink, curvature, crimp, and ratio of natural to straight hair.5 Computer-aided reconstruction of scalp biopsies demonstrated the spiral and straight nature of “African” and “Asian” follicles, respectively.6 These findings were recently confirmed by data demonstrating that asymmetric differentiation of layers of the bulb result in the curved follicle that is typical of “African hair.”7 Thus, the most significant findings to date, whether using hair form characteristics,5 computer-aided reconstruction,6 or immunohistochemistry7 have been that “African” and “Asian” hair represent extremes of the spectrum of hair morphology. The latter also possibly explains the pragmatic classification of hair into 3 major groups: African, Asian, and European, which is currently used in dermatology texts.

External features, although impressive, are not necessarily markers of other traits or even response to disease—they may be proxies of variable accuracy. For example, although people of African ancestry are reported to have higher mean blood pressures than white Americans, this race-linked association falls flat when one considers that the mean blood pressure in parts of Europe is significantly higher than that of African Americans.8

Even the recent testing—and subsequent targeted US Food and Drug Administration approval—of BiDil (NitroMed, Inc., Lexington, Mass) for hypertension in African Americans does not mean that this is the only group that can benefit from its use.9 In fact, components of BiDil (hydralazine and isosorbide dinitrate) have been known to be effective across populations for years. Current evidence suggests that for common diseases10, 11, 12 in populations, environmental factors are more important disease determinants than genetic traits. It is now also known that many previous associations of race with disease, like assessments of intellect, were in fact confounded by nurture (eg, socioeconomic status13).

Race is not a quantifiable biologic entity, and there is more variation within than between groups, yet its social effects are not insignificant.14 While race has been used to do untold harm—and as a South African, the experience is recent—its use in dermatology is a vexed question. For example, the definition of “African hair” is important because this hair form is associated with a higher prevalence of specific disorders (eg, acne keloidalis15). The latter may seem like a contradiction, especially in view of what I said above, until one realizes that race is in fact only a proxy for hair form. Within groups of “African” and “Asian” hair reside extremes of the spectrum of hair morphology, with the majority, typical of human variation, falling within these extremes. Even indigenous Africans in Africa display variations in hair form. I agree with Bigby et al that race should be abandoned, and whilst it is a useful, albeit nonspecific, surrogate for hair form, its use suggests a lack of rigor on our part.

There have been at least three previous attempts, as cited by Hrdy,5 to classify hair form. They have included: straight, wavy, frizzy, and woolly; straight, wavy, kinky, and woolly; and lastly straight, wavy, and helical. These classifications have all become historic records which may be a sign of our comfort with current terminology, the lack of representative participation in their development, or perhaps a case of teaching old dogs new tricks! There is need for a creative, inclusive, classification of hair form which should be user-friendly and could be an adaptation of one of the above. Fitzpatrick's16 “or other classifications” could also be adapted and evaluated for reliability as replacement for race, only if skin color is relevant in clinical reports. A credible development of such classifications could ensure their international acceptance. If it were not for disorders that are prevalent in specific hair forms and skin colors, attempts to clarify terminology would be unnecessary. Unlike other medical disciplines, where traits are often invisible polymorphisms, we have a unique opportunity in dermatology. Why should we settle for proxies and surrogates when we can be more specific?