Anjar Salim
Nearly 107 million Americans currently rely on wearable fitness technology to track everything from daily step counts and sleep quality to complex cardiovascular recoveries and exertion thresholds. For millions of consumers, the sleek displays on their wrists or fingers serve as the primary arbiters of health, offering real-time feedback designed to optimize athletic performance and catch early signs of distress. However, a persistent technical hurdle continues to cast doubt on the reliability of these devices for a significant portion of the population. Particularly for individuals with darker skin tones, a fundamental question lingers: Can the data provided by these commercial health monitors truly be trusted?
Recent scientific literature has brought this issue to the forefront of the consumer electronics industry. A comprehensive review published in 2025 examined the accuracy of smartwatches across diverse demographic groups and uncovered pronounced performance gaps. While certain top-tier devices maintained reliable heart rate measurements within five beats per minute across all skin tones, others faltered significantly. During moderate-to-vigorous physical exertion, some popular smartwatches underestimated the heart rates of darker-skinned users by ten to fifteen beats per minute. Similar discrepancies emerged during cycling sessions and resistance training, exposing a systemic vulnerability in optical health monitoring.
The implications of these tracking errors extend far beyond miscalculated calories or flawed workout summaries. Accurate heart rate metrics are critical for gauging exercise intensity and monitoring cardiovascular recovery rates, but they can also serve as lifesaving early-warning systems. Undetected irregular heartbeats can indicate serious underlying conditions—ranging from dangerous blood clots and strokes to advanced heart failure—that require immediate medical intervention. When a wearable device generates false readings or misses these vital physiological signals, the consequences can be genuinely perilous.
This technological shortcoming is hardly a recent discovery within the consumer technology sector. Major manufacturers have long understood that extracting precise data through highly pigmented skin presents a complex engineering challenge. In response to mounting scientific scrutiny and consumer advocacy, leading firms have intensified their research and development efforts to bridge the accuracy gap. Prominent outlets recently reached out to eight major wearable technology companies—including industry heavyweights like Apple, Oura, WHOOP, Google, and Polar—to evaluate how manufacturers are addressing historical biases and refining their hardware and software to ensure equitable performance across all user demographics.
The Root Cause: How Photoplethysmography Interacts with Melanin
To understand why smartwatches and fitness rings frequently struggle with darker skin tones, it is necessary to examine the foundational physics behind the technology. The vast majority of modern consumer wearables rely on photoplethysmography (PPG) to monitor physiological metrics. PPG systems utilize optical sensors comprising light-emitting diodes (LEDs) and photodetectors to measure dynamic changes in blood volume beneath the skin’s surface.
A standard PPG device functions by flashing green light directly into the user’s tissue. Within human blood, hemoglobin—a specialized protein responsible for transporting oxygen from the lungs to peripheral tissues—absorbs a predictable portion of this light. The remaining light reflects backward, striking the device’s optical sensor. As the heart beats and blood surges through the vascular network, the volume of blood changes, creating fluctuating patterns in the reflected light that the device translates into a real-time heart rate.
The central complication arises from melanin, the natural pigment responsible for determining skin, hair, and eye color. Melanin possesses a high capacity for absorbing and scattering light across the visible spectrum, including the green wavelengths traditionally utilized by optical sensors. Consequently, the higher the concentration of melanin in a user’s skin, the greater the absorption of the incoming and reflected light. This physical barrier diminishes the return signal received by the sensor, rendering darker skin tones significantly more vulnerable to data loss and signal degradation.
For years, the dermatology and medical research communities have attempted to standardize skin tone classification to study these disparities objectively. The Fitzpatrick Scale has long served as the industry standard, categorizing human skin into six distinct groups—ranging from Type I (very fair skin that burns easily and never tans) to Type VI (deeply pigmented skin that tans easily and rarely burns). However, modern researchers increasingly criticize the Fitzpatrick scale for its subjectivity, noting that its reliance on self-reported sun sensitivity, alongside cultural and societal factors, introduces avoidable inaccuracies into clinical studies and technology development.
Beyond Pigmentation: The Omnipresent Threat of Signal Noise
While skin pigmentation is a primary factor in optical sensor discrepancies, researchers emphasize that it is far from the only variable influencing accuracy. Dr. Michael Rosenberg, a cardiac electrophysiologist and researcher in the division of cardiology at the University of Colorado Anschutz Medical Campus, notes that signal noise represents the single greatest obstacle to precise wearable monitoring.
According to Dr. Rosenberg, environmental and physiological interference can severely disrupt sensor performance. This noise category encompasses tattoos, dense body hair, fluctuating hydration levels, heavy perspiration, and residual buildup from lotions, sunscreens, or topical medications. Furthermore, anatomical and physiological factors—such as wrist circumference, local skin temperature, and even the distribution of freckles—can alter how light interacts with the skin, resulting in skewed algorithmic interpretations.
Evolution of the Issue: A Timeline of Scrutiny and Improvement
The debate surrounding racial bias in optical health sensors did not emerge in a vacuum. The trajectory of this issue spans more than a decade of academic research, technological adaptation, and growing consumer awareness.
- 2010s: Early commercial smartwatches and fitness bands enter the mainstream market, primarily utilizing green-light PPG sensors. Initial validation studies focus overwhelmingly on light-skinned, healthy, young adult populations, inadvertently laying the groundwork for future demographic blind spots.
- Late 2010s to Early 2020s: Independent academic researchers begin publishing independent critiques highlighting performance disparities in pulse oximeters and optical heart rate monitors among patients and users with darker skin tones, sparking initial internal reviews among major tech companies.
- 2025: A landmark systematic review evaluates numerous smartwatch accuracy studies, formally documenting widespread variability in heart rate tracking across diverse skin tones, specifically noting severe under-reporting during vigorous exercise in darker-skinned cohorts.
- 2026: Subsequent dermatological and cardiological publications criticize reliance on outdated classification systems like the Fitzpatrick scale, prompting leading wearable manufacturers to accelerate the adoption of advanced objective measurement tools, such as spectrophotometers and the Monk Skin Tone scale, alongside aggressive algorithmic recalibrations.
Industry Responses: How Manufacturers Are Engineering Greater Equity
Recognizing that consumer trust depends entirely on data reliability, major wearable companies have aggressively mobilized to eliminate demographic performance gaps through a combination of advanced hardware architecture, algorithmic compensation, and more inclusive testing protocols.
WHOOP, known for its screenless performance tracker and subscription-based analytics platform, has implemented rigorous demographic representation standards during its internal data collection phases. Emily Capodilupo, senior vice president of research, algorithms, and data at WHOOP, emphasizes that the company intentionally overrepresents specific populations—including varied skin tones, genders, and age groups—whenever performance vulnerabilities are identified.
To counteract the light-absorbing properties of melanin, WHOOP developed a proprietary algorithm that constantly monitors signal strength and dynamically adjusts LED brightness in real-time. If the sensor detects a weakened return signal due to high skin pigmentation, the device automatically increases LED intensity. Capodilupo acknowledges the physical trade-offs involved in this approach, noting that higher-intensity illumination consumes more battery life. However, she asserts that the company views this as a necessary compromise to ensure uncompromising data integrity.
Apple employs a similarly multifaceted strategy, addressing optical limitations at both the software and hardware levels. The technology giant incorporates adaptive algorithms that regulate light transmission and amplify weak return signals across successive generations of the Apple Watch. Furthermore, Apple integrates lab-grade spectrophotometers during research and development phases. Unlike subjective visual scales, spectrophotometers provide precise, objective measurements of light reflection at varied wavelengths, ensuring that hardware engineers can calibrate sensors to deliver equivalent signal strength across the entire spectrum of skin tones.
Oura has leveraged a different anatomical approach by utilizing smart rings worn on the finger rather than traditional wrist-based form factors. Shyamal Patel, PhD, senior vice president of science at Oura, notes that the palmar side of the finger features more consistent pigmentation than the wrist. Additionally, the finger provides proximity to larger digital arteries, which naturally produce a stronger pulsatile signal.
To further optimize accuracy, Oura utilizes "smart sensing" technology capable of routing light through multiple pathways within the skin, continuously selecting the pathway that yields the highest-fidelity signal. In parallel, Patel confirmed that Oura is actively exploring advanced classification frameworks, such as the Monk Skin Tone (MST) scale—a ten-tone visual scale designed specifically to capture a comprehensive spectrum of deeper skin tones—while executing global testing initiatives to guarantee robust demographic diversity in its validation cohorts.
Polar, long recognized for its athletic watches and chest-strap monitors, highlights the ongoing evolution of sensor ergonomics and processing algorithms. Raija Laukkanen, PhD, director of science collaborations at Polar, emphasizes that while optical sensing remains highly practical for everyday continuous tracking, consumers requiring absolute precision during high-intensity training should look toward alternative technologies.
Specifically, Laukkanen points to electrocardiography (ECG) as the ultimate gold standard in heart rate monitoring. Unlike PPG, which measures reflected light through the skin, ECG sensors—such as those found in the Polar H10 chest strap or on-demand features on select Apple, Polar, and WHOOP devices—measure the heart’s electrical impulses directly. Because electrical signals pass through skin tissue without optical interference, ECG technology remains entirely unaffected by skin pigmentation, tattoos, or surface melanin concentration.
Google Health has similarly reinforced its commitment to demographic equity, issuing formal statements detailing its continuous validation testing. A Google spokesperson noted that the company routinely recruits diverse testing populations from across the globe during the development cycles of Fitbit and Pixel Watch devices to ensure rigorous performance standards across varied user bases.
Practical Guidance for Consumers
Despite significant engineering strides, experts caution that commercial wearables remain consumer electronics rather than diagnostic medical devices. While they excel at tracking broad longitudinal health trends and activity patterns, occasional discrepancies are inevitable.
To maximize the accuracy and reliability of wearable devices, cardiologists and industry researchers recommend several best-practice maintenance steps:
- Ensure a snug, secure fit so the optical sensor maintains continuous contact with the skin without shifting during movement.
- Position the device slightly above the wrist bone rather than directly on top of it to avoid bone interference.
- Routinely clean the underside of the device to eliminate accumulated sweat, dead skin cells, and residue from lotions, oils, or sunscreens.
- Avoid placing optical sensors directly over dense tattoos, as heavy ink severely disrupts light transmission.
- Install all software and firmware updates promptly to benefit from ongoing algorithmic improvements.
The ongoing evolution of wearable technology highlights a crucial truth: health-tracking devices must be engineered for everyone. As manufacturers expand their research methodologies, diversify their testing cohorts, and refine their optical and electrical capabilities, the industry moves closer to a future where every consumer can look at their wrist or finger with absolute confidence in the numbers staring back at them.
Chandra Hamzah
Tatiana Tantri Indah Sari
Budi Mulyono
Lukman Taruna Persada Ramadan