Sharma Megha, Brown Andrew W, Chung Simon, Sexton Kevin W, Schootman Mario, Mourani Peter M
Division of Neonatology, Department of Pediatrics, University of Arkansas for Medical Sciences, Little Rock, AR; Arkansas Children's Research Institute, Little Rock, AR.
Arkansas Children's Research Institute, Little Rock, AR; Department of Biostatistics, University of Arkansas for Medical Sciences, Little Rock, AR.
J Pediatr. 2025 Jul 10;286:114720. doi: 10.1016/j.jpeds.2025.114720.
To examine the objective differences and overlap in skin pigment among neonates born to mothers of different racial groups.
We conducted a prospective study assessing neonatal skin tone by spectrophotometry using SkinColorCatch (Delfin, Miami, FL) at the pulse oximetry site (dorsum of foot) in neonatal and cardiac intensive care units. Colorimetric indices were compared between non-Hispanic Black (NHB) and non-Hispanic White (NHW) neonates using Welch's t test. Distribution overlap was evaluated using the overlap coefficient and the probability of a higher value. Bivariate regression assessed the relationships between colorimetric indices and neonatal characteristics, including gestational age and birth weight. A P value of <.05 was considered significant.
Among 100 neonates (33 NHB, 57 NHW, 7 Hispanic, 3 other), skin darkness measures (melanin index, perceived lightness, and individual typology angle [ITA°]) were significantly different between NHB and NHW infants. However, substantial overlap was observed: overlap coefficient of 0.66 (95% CI, 0.51-0.83) for the melanin index and 0.73 (95% CI, 0.57-0.90) for the ITA°. The probability of higher value was 0.27 (melanin index) and 0.69 (ITA°), further demonstrating overlap. Lower gestational age and birth weight correlated with darker skin tone characteristics and this association did not differ significantly by race.
Despite statistical differences, NHB and NHW neonates exhibit substantial overlap in objectively assessed skin color. These findings support the use of direct skin pigmentation measurements in device accuracy studies rather than race-based classifications, helping to decrease bias in medical algorithms and device research.
研究不同种族母亲所生新生儿皮肤色素的客观差异与重叠情况。
我们进行了一项前瞻性研究,在新生儿重症监护病房和心脏重症监护病房,使用SkinColorCatch(Delfin,迈阿密,佛罗里达州)在脉搏血氧测定部位(足背)通过分光光度法评估新生儿肤色。使用韦尔奇t检验比较非西班牙裔黑人(NHB)和非西班牙裔白人(NHW)新生儿的比色指数。使用重叠系数和更高值的概率评估分布重叠情况。双变量回归评估比色指数与新生儿特征(包括胎龄和出生体重)之间的关系。P值<0.05被认为具有统计学意义。
在100名新生儿(33名NHB、57名NHW、7名西班牙裔、3名其他种族)中,NHB和NHW婴儿的皮肤暗度测量值(黑色素指数、感知亮度和个体类型角度[ITA°])存在显著差异。然而,观察到大量重叠:黑色素指数的重叠系数为0.66(95%CI,0.51 - 0.83),ITA°的重叠系数为0.73(95%CI,0.57 - 0.90)。更高值的概率分别为0.27(黑色素指数)和0.69(ITA°),进一步证明了重叠。较低的胎龄和出生体重与较深的肤色特征相关,并且这种关联在不同种族之间没有显著差异。
尽管存在统计学差异,但NHB和NHW新生儿在客观评估的肤色上表现出大量重叠。这些发现支持在设备准确性研究中使用直接的皮肤色素沉着测量,而不是基于种族的分类,有助于减少医学算法和设备研究中的偏差。