CASE STUDY — ケーススタディ
ク リ ニ カ ル リ サ ー チ プ ラ ッ ト フ ォ ー ム
Leading design for an AI-powered clinical research platform — from brand identity to a two-sided enterprise product connecting sites and CROs.
ROLE
Design Lead, team of 3
TIMELINE
1–2 Years
STATUS
Pilot / Early Customers
COMPLIANCE
HIPAA + 21 CFR Part 11
問題 THE PROBLEM
Clinical trial recruitment is quietly failing the people it’s supposed to help.
MISS RATE
80%
of clinical trials fail to meet their initial enrollment timelines - costing pharma millions per day in opportunity cost alone
TERMINATION CAUSE
57%
of clinical trial terminations are due to enrollment difficulties - meaning the research just stops and research funds go down the drain
Behind these numbers are delayed cures, patients who lose access to life-saving treatments, and drug costs that rise to cover inefficient research. Clinical staff burn out under inefficiencies that ripple across the entire ecosystem.
リサーチ RESEARCH & MANDATE
I was brought in to understand a space the previous design lead couldn’t.
Clinical research has its own language, its own conventions, and workflows that are deeply entrenched for good reason. You can't just parachute in and start designing — you have to actually understand what's happening or you'll break something. I interviewed clinical researchers, PIs, site coordinators, and monitors across all kinds of institutions, accounting for blinding protocols, randomization structures, cohort configurations, and how things change depending on whether the sponsor is a major CRO like IQVIA or a more in-house operation.
The complexity wasn’t just in the features - it was in understanding which edge cases were universal enough to design for, and which were site-specific customizations. Getting that wrong would have produced a tool that worked for one workflow and failed everyone else.
I can now speak clinical research terminology fluently. But it took a lot of interviews across the US, with researchers at different institution types, major CROs, and boutique sponsors - before I felt like I had a wide enough picture to actually design from.
プロダクト THE PLATFORM
One platform. Every stakeholder in the trial process.
What makes Centrova different isn't any single feature - it's that all of these things actually exist together. Sites currently patch this together with 2–3 tools that weren't built for each other. We replaced that entirely.
SMARTMATCH
Bidirectional matching engine — find the right patients for a trial, or the right trials for a patient, using a proprietary tagging database built from clinicaltrials.gov and patient records.
AI SCREENING
Bulk patient-trial evaluation with source-linked AI decisions. Every recommendation shows the exact line from the patient file used to justify it. Researcher makes the final call.
TRIAL MANAGER
Browse, configure, and manage all trials. Role-based access, double-blind support, enrollment tracking, and direct criteria editing when clinicaltrials.gov is inaccurate.
DATA HUB
Shared file storage for visit data, trial documentation, deviation logs, and direct data entry. One-click export to sponsors.
CLINICAL TEAMS
Trial channels, direct messages, and monitor-to-site lines for submitting queries and receiving site reports. Separate blinded and unblinded chats for double-blind study integrity.
MONITOR PORTAL
A dedicated section for CRO monitors to review evaluations, finalize enrollment, and request compliance documents without having to schedule a remote visit or send a single email.
デザイン KEY DESIGN DECISIONS
The hardest problems weren’t visual. They were structural.
01
Adding Monitors to the Platform
This was probably the biggest product direction call I made, and it came entirely from research. Monitors and sites are supposed to be deeply connected - they're working on the same trials - but the way current workflows are set up, they're constantly jumping through hoops just to share information. Email chains, phone calls, remote visits. It's wildly inefficient for how often they actually need to communicate.
What I realized through interviews is that Centrova sits in a pretty unique position: we hold patient data relevant to screening, but we're not an EHR. That means we can give monitors the access they need without the same tight restrictions that make everything so slow - and we can do it in a way that's still HIPAA-compliant and deidentified appropriately.
I took this to the CEO and COO with the interview data to back it up. It was outside the original scope, but they got on board - and it ended up being one of the features monitors responded to most positively in pilot.
02
Redesigning the Screening Flow
The pre-match score existing IN patient selection turns a blank selection screen into a discovery tool - surfacing the most likely candidates visually before the researcher has done anything. It's one of the most powerful features in the platform.
03
AI Transparency + Human Authority
In a regulated clinical environment, you can't just let an algorithm make enrollment decisions and expect people to trust it. So we designed around a structure where the AI does the heavy lifting - screening all selected patients against all trial criteria simultaneously - but each decision it makes is linked back to the exact line in the patient's file it used to justify it. Researchers have to make a final call on each criterion themselves. The AI recommends. The human decides. This isn't just a trust thing though, it's a compliance thing. Clinical responsibility stays with qualified staff, and the double-check is mandatory by design.
04
Visual Identity — Research-Focused, Not Clinical
We held a full brand meeting with the design team before touching anything. The direction we landed on was research-focused rather than clinical - there's a difference, and it matters to the people using it. Researchers don't see themselves as hospital administrators. We also made a deliberate call to avoid looking overly tech-forward. In high-stakes environments, being too flashy about your AI capabilities actually works against you, reducing trust rather than building it. Large typography, top nav, content front and center. Nothing that feels like it was generated from a Shadcn template.
成果 OUTCOMES
From 90 minutes to 5. Streamlined clinical comms. We’re on the path to a faster cure.
~90 minutes
5–10 minutes
Email, phone, remote visits
Unified in-platform
2–3 disconnected tools
1 integrated platform
Manual, inconsistent
Auto-generated PDF
Pilot feedback from monitors described the centralization of communications, screening history, and file access as dramatically simpler than their current multi-tool workflow. PIs responded positively to the role-based access system and the transparency of AI screening decisions.
The 90 minute to 5–10 minute improvement isn’t inconsequential. It’s an order-of-magnitude change that makes bulk screening of large patient populations feasible for the first time without equally scaling staff.
振り返り REFLECTION
The most impactful thing I did wasn’t a screen.
It was the research that led me to advocate for including monitors in the platform. That decision changed the product’s scope, its value proposition, and ultimately its ability to solve the communication fragmentation problem at the heart of clinical research. It came from listening carefully enough to understand a problem the original brief hadn’t accounted for.
The screening workflow redesign also reinforced something important: simplicity isn’t the absence of features — it’s the right structure for the features you have. The wizard felt simple on paper but failed in practice. The Screening Center feels more complex on paper but maps directly to how researchers actually work, which makes it feel simpler in use.
Centrova is still in pilot. But the infrastructure being built — a unified, compliant, AI-augmented platform connecting every stakeholder in the clinical trial process — is the kind of foundation that can meaningfully accelerate the timeline between medical discovery and patient access.