Skin-tone bias in pulse oximetry for investors

Decision-ready map

• Clinical risk: missed hypoxemia → outcomes + liability

• Regulatory momentum: stronger evaluation across skin tones

• Diligence: stratified performance + external validation

• Value: procurement trust + lower adverse-event risk

(1) What it is

For investors, skin‑tone bias in pulse oximetry is a material technical and regulatory risk. Because SpO₂ influences clinical escalation and at-home decisions, systematic overestimation can translate into missed deterioration events, creating liability exposure, procurement pushback, and reputational harm.

(2) Who it helps

This brief helps venture and strategic investors evaluate device and wearable companies whose product performance depends on SpO₂, and to translate “equity claims” into diligence questions and milestones.

(3) What evidence exists

Large paired datasets show occult hypoxemia occurs more frequently for Black patients than White patients at similar SpO₂ ranges. Reproducibility has been reported in VHA and other health-system studies, and outcome associations (organ dysfunction and mortality) have been identified. A 2024 systematic review finds overestimation in darker skin tones is commonly reported across studies. FDA publications signal updated expectations for performance evaluation considering skin pigmentation and for improved labeling.

(4) Translation barriers

Evidence is device-specific: performance varies by model and context. Collecting paired reference data is costly, making underpowered subgroup analyses common. Additionally, companies may make broad “all skin tones” claims without transparent stratified performance, and post-market change control (updates, component substitutions) can degrade parity.

(5) Equity/safety checks

Require stratified performance distributions and an explicit pigmentation measurement method. Verify the reference standard (SaO₂ co‑oximetry) and pairing approach. Confirm real-world stress tests (low perfusion, motion). Evaluate governance: monitoring dashboards, drift detection, and revalidation triggers after updates, aligned with FDA signals.

(6) Decision questions

• Is the company’s claim category appropriate (medical monitoring vs wellness), and does evidence match?

• Is there external validation across sites/devices and subgroup parity metrics?

• Does the company have a post-market plan for model updates and performance monitoring by subgroup?

• Could parity evidence create procurement advantage or reduce downside risk?

(7) Practical next steps

Add an “equity evidence” section to IC memos: subgroup metrics, external validation, and monitoring plan. Make completion of representative paired validation a financing milestone before scaling commercialization. Encourage health-system partnerships to produce publishable evidence and procurement-ready documentation.

(8) References

https://doi.org/10.1056/NEJMc2029240
https://doi.org/10.1016/j.bja.2024.01.023
https://doi.org/10.1016/j.eclinm.2022.101428
https://doi.org/10.1016/j.chest.2021.09.025
https://doi.org/10.1001/jamanetworkopen.2021.31674
https://doi.org/10.1136/bmj-2021-069775
https://www.fda.gov/media/175828/download
https://www.fda.gov/news-events/press-announcements/fda-proposes-updated-recommendations-help-improve-performance-pulse-oximeters-across-skin-tones