ISSUE 009 · Oct 11, 2026 · 8 min read

What if tomorrow’s pain crisis appeared on the wrist tonight?

Turn nocturnal vasoconstriction trajectories in sickle cell disease into a 72-hour risk window—and make the product accountable for same-day confirmation and access, not merely an alert.

The issue in one sentence

Turn nocturnal vasoconstriction trajectories in sickle cell disease into a 72-hour risk window—and make the product accountable for same-day confirmation and access, not merely an alert.

Diagram 01

From a nocturnal signal to faster pain care

01Nocturnal raw PPG→

02Personal baseline→

03AI risk window→

04Patient check-in→

05Same-day hematology→

06Day hospital · home plan→

07Pain · ED outcomes

ANM proposed model. The small 2026 PPG feasibility signal is not treated as evidence of clinical benefit; prospective silent validation must precede any care activation.

01Clinical problem · Evidence

Pain crises feel sudden—and care starts even later.

Vaso-occlusive pain episodes are a major driver of emergency visits and admissions in sickle cell disease. Even when patients sense deterioration, deciding when home care is no longer enough is difficult, and delays and inequities in first analgesia persist in emergency departments.

02Technology stack · Proposal

Read nocturnal vasoconstriction as a within-person change.

Combine raw-access wrist PPG with heart-rate variability, skin temperature, movement, sleep, medication, hydration, and self-reported pain. AI estimates a 72-hour risk window from each person’s baseline rather than a universal cutoff, withholding alerts on low-quality nights.

03Research question · Hypothesis

Can the signal predict acute care before a pain diary does?

Test whether a model including nocturnal vasoconstriction trajectories predicts adjudicated vaso-occlusive acute-care episodes within 72 hours better than a symptom-and-utilization baseline, while keeping false-alert burden tolerable.

04First study · Proposal

Six centers, 240 adults, nine months of silent prospective validation.

Adults with at least two crises in the prior year wear the sensor nightly, while scores remain hidden from patients and clinicians. The comparator uses pain diary, prior utilization, and weather; the test model adds raw-PPG vasoconstriction, HRV, skin temperature, and sleep. The single primary outcome is the patient-level externally validated difference in AUPRC for adjudicated crises within 72 hours. Prespecified success requires ΔAUPRC ≥0.10, PPV ≥25% at 70% sensitivity, and no more than two false alerts per patient-month. Performance by skin tone, sex, genotype, hydroxyurea use, and transfusion is prespecified exploratory analysis.

05Defensible data · Hypothesis

The moat is not the waveform—it is where the patient went next.

Link raw PPG and signal quality, objectively measured skin tone, personal baselines, pain, medication, hydration, weather, contact success, day-hospital capacity, time to analgesia, ED visits, and admissions. Only a subsequent randomized study should test whether alert → nurse confirmation → same-day pain pathway changes utilization and outcomes.

06Venture thesis · Assumption

Sell guaranteed access to a pain pathway—not a wristband.

Initial customers are comprehensive sickle cell centers and Medicaid managed-care organizations. Bundle device rental, silent quality audit, nurse confirmation, day-hospital or telehealth routing, and time-to-analgesia reporting per member-month. Shared savings follows only if a randomized trial shows fewer unplanned ED visits or admissions.

Evidence map

Feasibility exists; prediction remains unproven

12 people

People with sickle cell disease in the 2026 wrist-PPG study

79 nights

Nocturnal wrist-PPG recordings collected from patients

0.24–0.70

Overnight wrist-to-finger PPG correlation range across nine nights

19 people

Single-center 2023 sample exploring wearable pain models through 30 days after discharge

Evidence

Observed signals versus the untested care thesis

Established: a January 2026 study collected wrist PPG from 12 people with sickle cell disease and six healthy controls. Nocturnal vasoconstriction metrics were feasible across 131 nights, including 79 patient nights; simultaneous wrist-to-finger comparison covered only nine nights in three healthy controls. It did not test prediction of imminent crises or clinical outcomes. Two 2023 single-center studies of 20 and 19 people likewise modeled pain that had already occurred. ASH now frames time from ED arrival to first analgesic as an equity quality measure. Prediction within 72 hours, fewer ED visits through same-day routing, and a viable per-member-month service therefore remain hypotheses.

2026 nocturnal wrist-PPG study 2023 30-day wearable pain study Blood-flow and pain mechanism registry ASH emergency pain-care quality measures FDA optical-sensor skin-tone guidance

Failure conditions and safety, ethics, and regulatory risks

Prediction failure condition

Do not activate clinical alerts if the test model fails to improve AUPRC by at least 0.10, yields PPV below 25% at 70% sensitivity, or exceeds two false alerts per patient-month.

Optical bias

Melanin, low perfusion, sensor pressure, and motion can alter optical waveforms. Report unreadable rates and performance by objective skin tone, perfusion, and device; halt launch if sensitivity differs by more than ten percentage points in the darkest-skin group.

False reassurance and over-triage

A low score must never delay care for fever, chest pain, dyspnea, neurologic symptoms, or rapidly escalating pain. A high score triggers patient confirmation and clinical assessment—not automatic fluids or opioids.

Access after the alert

Prediction without day-hospital capacity, after-hours contact, or transport may only add anxiety. Do not sell to sites unable to contact within two hours and offer same-day in-person or telehealth assessment.

Surveillance and discrimination

Pain, medication, location, and biosignals are sensitive in insurance and employment. Minimize collection, limit purpose, allow deletion and withdrawal, and contractually bar use of risk scores for coverage decisions or medication-compliance policing.

Regulatory boundary

Existing consumer-wearable functions are not authorization to predict sickle-cell crises. A disease-risk score tied to care routing may be regulated device software, requiring early FDA engagement, analytical and clinical validation, cybersecurity, and human-factors review.

Bibliography

References

Established facts are separated from research and venture hypotheses that still require validation.

01
Shah P, Sy S, Chen M, Khoo MCK, Coates TD, Veluswamy S. Wearable Device Photoplethysmography As a Viable Tool to Longitudinally Monitor Vasoconstriction Biomarkers for Predicting Vaso-Occlusive Crisis in Sickle Cell Disease. JMIR Hum Factors. 2026;13:e75465.
A feasibility study of 12 patients, six controls, and 131 nights. Simultaneous wrist-to-finger comparison covered only nine nights in three controls; imminent crisis prediction was not tested.PMID 41494152 · DOI 10.2196/75465
02
Vuong C, Utkarsh K, Stojancic R, et al. Use of consumer wearables to monitor and predict pain in patients with sickle cell disease. Front Digit Health. 2023;5:1285207.
Nineteen adults were followed for a median 28 days, yielding 2,395 pain observations. This small single-center model did not establish pre-crisis prediction.PMID 37954032 · DOI 10.3389/fdgth.2023.1285207
03
Stojancic RS, Subramaniam A, Vuong C, et al. Predicting Pain in People With Sickle Cell Disease in the Day Hospital Using the Commercial Wearable Apple Watch. JMIR Form Res. 2023;7:e45355.
A 20-person day-hospital feasibility study with median wear time of 2 hours 17 minutes; it modeled pain during treatment rather than validating an early warning.PMID 36917171 · DOI 10.2196/45355
04
ClinicalTrials.gov. Blood Flow and Pain Crises in People With Sickle Cell Disease.
A registered mechanistic study of peripheral blood flow and pain crises; it informs biological plausibility but is not a wearable-product effectiveness trial.NCT01568710
05
American Society of Hematology. Timely and Equitable Pain Management for Sickle Cell Disease in the Emergency Department. Hematology Quality Measures.
Defines facility-level measures for median time from ED arrival to first analgesic and the gap versus renal-colic care.Accessed Oct 11, 2026
06
U.S. Food and Drug Administration. Pulse Oximeters for Medical Purposes: Non-Clinical and Clinical Performance Testing, Labeling, and Premarket Submission Recommendations. Draft Guidance. January 2025.
A nonbinding draft noting that optical-sensor performance can be affected by skin pigmentation; it does not authorize this PPG algorithm.FDA-2023-N-4976

The next issue publishes every day at 9 AM KST.All issues