Whether there is racial bias in health monitoring by wearable devices (such as smart watches, fitness bracelets, etc.) is an issue that has attracted widespread attention in recent years. Existing studies and cases show that ** The technical design or algorithm training of certain equipment may have prejudices against specific ethnic groups**, which are mainly reflected in the following aspects： --- ### **1. Limitations of sensor technology** Wearable devices rely on optical sensors (such as photoelectric plethysmography, PPG) to measure indicators such as heart rate and blood oxygen saturation (SPO₂). These sensors penetrate the skin by emitting specific wavelengths of light (such as green, red, or infrared light) to detect changes in blood flow. however： -**Effects of skin color differences**: Melanin in dark skin absorbs more light, resulting in reduced signal strength, which may affect measurement accuracy. For example： -**Blood oxygen monitoring**: A 2022 study in Nature Medicine pointed out that dark-skinned people use pulse oximetry to time their blood oxygen levels, and the probability of misjudging low blood oxygen levels is 3 times that of light-skinned people (the true value may be underestimated), which may lead to clinical decision-making errors. -**Heart rate monitoring**: Some smartwatches have higher heart rate data errors in dark-skinned users (such as earlier versions of Apple Watches). -**Differences in hair and skin structure**: Curly hair or thicker epidermis may further interfere with the sensor signal. --- ### **2. Deviation of algorithm training data** The development of most health algorithms is based on data sets for specific populations. If there is a lack of diversity, it may lead to a decrease in applicability to other races.： -**Predominantly white data sets**: For example, the exercise recognition algorithms of early fitness equipment were mostly based on data from European and American populations, and the exercise pattern recognition of Asian or African-American populations may not be accurate enough. -**Deviation of the disease risk model**: Some algorithms for health risk assessment (such as cardiovascular disease) rely on the physiological characteristics of specific ethnic groups and may not be applicable to other groups. --- ### **3. Potential deviation of health index standards** The definition of some health indicators itself may have racial limitations： -**Blood pressure and heart rate standards**: The current standards are mostly based on Western population studies and may ignore physiological differences between races (for example, the blood pressure baseline of African-American populations is usually higher). -**Sleep monitoring**: Some devices judge the sleep stage by movement and heart rate variability, but differences in sleep patterns or physiological responses of different races may affect accuracy. --- ###**Actual cases and evidence** -**Controversy over blood oxygen monitoring**: In the 2020 New Crown epidemic, the error of pulse oximeter in dark-skinned patients was confirmed to cause delays in treatment. The U.S. FDA issued a warning in 2021, requesting a reassessment of the racial sensitivity of related equipment. -**Smart watch research**: A 2021 Lancet Digital Health study found that some devices have a heart rate error rate of up to 20% in dark-skinned users. -**Motion tracking deviation**: A 2023 University of California study pointed out that some devices have lower step counting accuracy for African-American users than white users. --- ###**Industry improvements and solutions** -**Technical optimization**: Improve sensor design (such as using multi-wavelength light sources) and develop algorithms to adapt to different skin tones. -**Data diversity**: Include more ethnic, gender, and age samples in algorithm training. -**Transparent testing and supervision**: Manufacturers are required to disclose the racial sensitivity test results of their devices and include them in medical device approval standards (for example, the FDA is promoting relevant regulations). -**User education**: Inform users of the limitations of equipment and avoid excessive dependence on a single data. --- ###**Conclusion** There is indeed potential racial bias in the health monitoring of wearable devices, mainly due to the historical limitations of technology design, data bias, and standard-setting. Although the industry is gradually improving, users still need to be cautious when relying on this data for health decisions (especially for medical purposes) and prioritize diversified and proven devices. At the same time, academia and regulatory agencies need to continue to promote research on technological fairness and reduce systemic biases in health technology.