When classifying your work felt like a maze
Replacing friction with a clear, structured flow
Occupational Data Health
The NIOSH ODH initiative aimed to build a working prototype to demonstrate how patients could enter or update work information before or during visits, while capturing occupation, industry, job history, and work-related exposures to support diagnostics, care decisions, and public health research.
The goal wasn’t to build a full product, but to create a prototype that demonstrated the flow to vendors so they could integrate it into their systems.
Inherited Situation
Before I joined,
the project had stalled.
Years of requirements gathering and planning led to overly detailed requirements and a simplistic, unusable prototype focused on classification accuracy - not real user behavior.
The Reality Check
The result was a simplistic translation of requirements - an idealized flow that relied on perfect matches to census classifications for industry and occupation.
Rigid classification
Patients were expected to match to exact census categories.
No guidance
The system expected answers, but didn’t help users get there.
Language mismatch
People don’t describe their work in standardized census terms.
Imagine sitting at a doctor’s office, not feeling your best, filling out routine forms. Then you’re asked to describe your job in standardized terms on top of everything else.
There’s no guidance, only guesswork. If it doesn’t match exactly, the system fails you.
The Insight
Users wouldn’t fail.
The system would fail them.
My Role
I walked into a broken system and had to rebuild it quickly under constraints. The stakeholders were running out of time to produce a viable prototype. As the only UX professional on the team, I took on a wide range of responsibilities to support the project’s needs.
Product Strategy Support
Helped prioritize the prototype around the most critical user flows.
Product Experience Architecture
Defined how the vision became a usable experience.
Design Lead
Owned end-to-end UX/UI design and rebuilt the experience from the ground up.
Research and Usability Testing
Led user research and usability testing to shape design direction.
Cross-Functional Facilitator
Helped align stakeholders and the team around a viable solution.
Implementation Handoff
Created interactive HTML/JavaScript prototypes that accelerated development handoff.
Me
UX Designer
Project Manager
Strategy & SME
Developer
Engineering
The Context I Inherited.
No prior user research
Without any user research, issues had to be identified through the UI.
Dense documentation
Required filtering and prioritizing what actually mattered.
Limited time to explore
Had to be rebuilt on a tight deadline, with little room to test ideas or refine the approach.
Stakeholder Alignment
Multiple stakeholders influenced priorities, making product decisions slower and harder to align.
The Pivot
The existing prototype was not usable and needed to be rebuilt quickly. To meet the goal, the new solution had to be drastically different.
Break it into steps
Turn complex decisions into simple choices.
Prevent failure early
Guide people instead of demanding perfect input.
Show progress
Give clear feedback at every step so progress stays visible.
A preview of key improvements
A preview of key improvements
Understanding the Users
Two very different user groups shaped the redesign. With no prior research available, the personas were synthesized from stakeholder and SME conversations, then sharpened through usability testing.
People at a doctor’s office trying to complete required forms quickly before their visit.
Needs
Challenges
Provide systems used in medical offices to capture and manage patient information.
Needs
Challenges
In response to constraints, stakeholder input, and user needs, the following principles guided the redesign.
Prevent failure early
Break decisions into steps
Guide selection, not search
Explore without penalty
Provide clear feedback always
Always allow a way forward
The Changes
The redesign process began in low fidelity, focusing on how users move through the workflow before visual design was introduced. The goal was to simplify how users search, narrow options, and keep track of progress while working through complex job selection tasks.
Instead of forcing users into dead ends or overwhelming result lists, the system introduces guided refinement and controlled exploration from the very first interaction.
The workflow shifts from deep trees and long dropdowns to progressive narrowing, revealing relevant occupations only after users establish broader intent.
The system preserves previously explored categories and selections, allowing users to compare options without repeating work or relying on memory.
Usability Testing
The sessions revealed consistent patterns in how users approached decisions and interacted with the system, highlighting where additional guidance and feedback were needed.
Testing highlighted several expectations users had throughout the flow:
Final Design
The redesigned prototype turned a confusing, failure-prone process into a guided experience that helped patients identify their work information step by step without needing to know exact industry or occupation classifications.
The redesigned flow guided users from broader industry and occupation categories into more specific selections, reducing guesswork and helping users narrow choices progressively without needing the “correct” answer upfront.
Persistent section indicator keeps users oriented. Placeholder rationale.
"My job is a(n)" replaces clinical "Occupation" for a conversational tone. Placeholder rationale.
Shows the number of matches before committing. Placeholder rationale.
Action framed as collaborative help, not navigation. Placeholder rationale.
"1. General Occupation" structures the decision into two steps. Placeholder rationale.
Always-available browsing history affordance. Placeholder rationale.
Breaks 15 categories into manageable chunks. Placeholder rationale.
A single emphasized card previews selection. Placeholder rationale.
"I am done" activates only after a valid selection. Placeholder rationale.
Description appears right where the choice happens. Placeholder rationale.
Step 2 continues inline — no page jump. Placeholder rationale.
Picking the specific match finalizes the choice. Placeholder rationale.
History accumulates automatically as users browse. Placeholder rationale.
"8 browsed" surfaces progress through the search space. Placeholder rationale.
One click jumps back to any prior browse. Placeholder rationale.
Selection stays highlighted across sections. Placeholder rationale.
The Patient Work Summary brings occupations, industries, exposures, and work history into one place, helping users clearly see progress, review information, and identify what was still missing.
"Main Summary" anchors users across the multi-section flow. Placeholder rationale.
Sections named in everyday terms, not clinical jargon. Placeholder rationale.
One obvious "+ Add new" per empty section. Placeholder rationale.
Section identity preserved across states. Placeholder rationale.
Each job uses the same card structure, no new layout to learn. Placeholder rationale.
Collapsed by default, expandable on demand. Placeholder rationale.
The redesigned flow allowed patients to add household members and related work exposure information, helping capture possible occupational exposures brought into the home.
Placeholder rationale for marker 1.
Placeholder rationale for marker 2.
Placeholder rationale for marker 3.
Placeholder rationale for marker 4.
Placeholder rationale for marker 1.
Placeholder rationale for marker 2.
Placeholder rationale for marker 3.
Placeholder rationale for marker 4.
Placeholder rationale for marker 1.
Placeholder rationale for marker 2.
Placeholder rationale for marker 3.
Placeholder rationale for marker 4.
Looking Back
Adoption Outcome
The prototype improved usability and showed a clear path forward, but it was not adopted by vendors.
Outcome
Limitation
