Before & After

Improving Accuracy in Patient Merge Workflow

Patient merge is a high-risk administrative workflow where incorrect decisions can lead to duplicated records, lost clinical history, or patient safety issues.
Users must compare two patient records quickly, accurately, and with confidence.

Before
  • Dense, form-like layout required users to manually scan every field

  • Matching and non-matching values were visually similar

  • Important discrepancies (MRN, DOB, SSN) were easy to miss

  • No clear prioritization of high-risk fields

  • Cognitive load increased with record length and scrolling

After
  • Grouped information into clear, semantic sections (Demographics, Contact, Problems, etc.)

  • Introduced explicit match / no-match indicators at the field level

  • Used visual cues to surface discrepancies immediately

  • Reduced visual noise by simplifying layout and spacing

  • Clearly labeled “New Patient” vs “Old Patient” context

  • Consolidated decision action into a single, clear confirmation step

How the redesign reduced friction:

  • Users can identify mismatches at a glance instead of reading every field

  • High-risk discrepancies surface visually without interpretation

  • Grouped sections reduce scrolling and mental context switching

  • Visual confirmation reduces hesitation and second-guessing

  • Clear merge action reinforces confidence in the final decision

UX shift:

From carefully reading to avoid mistakes → to seeing discrepancies and acting with confidence.

Why this matters:

  • Patient merge errors can affect clinical safety and data integrity

  • Administrative users work under time pressure with high accountability

  • Reducing cognitive load lowers the likelihood of costly mistakes

  • Clear visual guidance supports consistent decision-making across users

  • Improved confidence reduces rework and audit risk

  • Match indicators reduce reliance on memory and visual scanning alone

  • Structured grouping improves readability and focus

  • Consistent patterns support learnability for infrequent users

Accessibility considerations