Hazard¶

DO NOT EDIT BELOW THIS LINE UNLESS YOU KNOW WHAT YOU ARE DOING

Hazard name¶

Avatar gradient insufficient for patient distinction

General utility label¶

[2]

Likelihood scoring¶

TBC

Severity scoring¶

TBC

Description¶

Avatar gradient colors may not provide sufficient visual distinction for rapid patient identification in high-pressure clinical environment due to limited color variations, color-blind accessibility issues, or display contrast problems.

Causes¶

Only 30 gradient variations (0-29) means collisions likely with >30 patients
Color-blind clinicians cannot distinguish similar hues
Low contrast on different displays or lighting conditions
Random generation can assign same gradient to different patients

Effect¶

Clinician relies on visual color cue but confuses two patients with similar or identical gradient colors.

Hazard¶

Wrong patient selected based on visual recognition instead of reading name carefully.

Hazard type¶

WrongPatient
IncorrectResult

Harm¶

Clinical action performed on wrong patient including medication administration, procedure scheduling, or discharge planning.

Existing controls¶

None identified during initial analysis.

Assignment¶

Clinical Safety Officer

Labelling¶

TBC (awaiting scoring)

Project¶

Clinical Risk Management

Hazard controls¶

Design controls (manufacturer)¶

Expand avatar gradient palette to 100+ distinct combinations using HSL color space with maximum perceptual distance. Ensure minimum 15-degree hue separation and saturation/lightness variation to prevent similar-looking gradients.
Add patient initials or first two letters of surname overlaid on gradient circle in contrasting white/black text (calculated for WCAG AAA contrast). Provides text-based identifier in addition to color.
Implement deterministic gradient assignment: hash patient FHIR ID to consistent gradient index, ensuring same patient always gets same color across sessions and devices.
Add colorblind-safe mode setting in user preferences: alternative patient identifier system using patterns (stripes, dots, checks) overlaid on gradient for deuteranopia/protanopia users.
Include accessibility indicator: if two patients in visible list have similar gradients (hue difference <30 degrees), display warning icon next to both: "Similar colors - verify name carefully."

Testing controls (manufacturer)¶

Unit test: Generate gradients for 100 patients, calculate minimum perceptual distance between any two gradients. Assert minimum distance >15 CIELAB ΔE units to ensure human-perceivable distinction.
Visual regression test: Render patient list with 10 patients on various displays (high contrast, low contrast, different color gamuts). Verify gradients remain distinguishable across display types.
Colorblind simulation test: Apply deuteranopia filter to patient list, verify initials remain readable with sufficient contrast. Assert pattern overlays (when enabled) provide additional distinction.
Unit test: Create patient with FHIR ID "abc123", generate gradient, destroy component, recreate with same ID. Assert identical gradient generated (deterministic hashing).

Training controls (deployment)¶

Train clinicians to never rely solely on avatar color for patient identification. Emphasize: "Always read patient name and NHS number. Avatar color is supplementary visual aid only."
Document clinical workflow: "Before selecting patient from list, verbally state patient name aloud and verify against expected patient to prevent visual recognition errors."
Include in medication administration training: "Five Rights of Medication Administration: verify patient name, DOB, and NHS number from wristband match patient on screen. Do not rely on avatar color alone."

Business process controls (deployment)¶

NHS Trust policy: Avatar colors are visual aids only and must not be used as sole patient identifier. All clinical actions require verification of name, DOB, and NHS number.
Clinical governance: Incident reporting for near-misses where clinician almost selected wrong patient due to similar avatar colors. Analyze trends to identify high-risk scenarios.
Accessibility policy: Clinicians with colorblindness must declare condition during IT onboarding to enable alternative patient identifier modes. No stigma or penalty for declaration.

Residual hazard risk assessment¶

TBC — awaiting initial controls implementation.

Hazard status¶

Draft from LLM

Code associated with hazard¶

frontend/src/components/patients/PatientsList.tsx
frontend/src/lib/fhir-patient.ts
backend/app/utils/colors.py