Case Study: Civis Health — Financial Assistance Within Reach

Project: Civis Health Role: Product Architect (PharmD) Stack: Next.js 14, TypeScript, Tailwind CSS, pdf-lib Live: civis-health.vercel.app

The Problem: A Hidden Safety Net

Millions of patients in the United States are eligible for hospital financial assistance programs — but most don't know these programs exist or how to determine if they qualify. The application process is fragmented: each hospital has its own eligibility criteria, its own forms, and its own submission process. For patients already navigating a medical crisis, this bureaucratic complexity becomes a second barrier to care.

The result is a massive underutilization of financial assistance — programs designed to help the most vulnerable go unused because the pathway to access is opaque.

The Solution: Automated Eligibility & PDF Generation

Civis Health is a free, open-source web application that automates the eligibility determination process and generates pre-filled PDF applications. The entire flow is designed around three principles: simplicity, privacy, and zero cost.

Visit Live Application: civis-health.vercel.app →

How It Works:

1. Select a Hospital — Choose from a data-driven registry of hospitals and their specific financial assistance programs.
2. Enter Household Information — Annual income and household size are the only inputs required.
3. Instant Eligibility Determination — The system calculates the patient's Federal Poverty Level (FPL) percentage and compares it against the hospital's specific thresholds.
4. Download a Pre-Filled Application — Eligible patients receive a server-generated PDF ready for submission.

Technical Architecture:

• Flexible Eligibility Schema: Hospitals use different eligibility models — some define limits as a percentage of FPL (e.g., Kaiser at 350% FPL), others use flat dollar amounts by household size (e.g., Sutter Health). The type-safe IncomeLimit union type handles both models seamlessly.
• FPL Calculator: Core algorithm uses 2024 HHS Federal Poverty Level guidelines ($15,060 base + $5,380 per additional person) to calculate a household's FPL percentage.
• Server-Side PDF Generation: Uses pdf-lib to generate application PDFs entirely on the server via Next.js API routes — no external services, no cloud dependencies, near-zero cost.
• Data-Driven Hospital Registry: Hospital data (eligibility criteria, program names, PDF field mappings) lives in a structured JSON registry, making it trivial to add new hospitals without code changes.

GENERATING_DIAGRAM...

Design Decisions: Privacy-First, Non-Profit Aesthetic

Civis Health was designed with deliberate constraints:

• Privacy by Design: All FPL calculations happen client-side. No user data is stored — not in a database, not in cookies, not in analytics. The form data travels only via URL query parameters, making the entire system stateless.
• Zero Cost to Users: No accounts, no fees, no upsells. The tool is free and open-source.
• Institutional Branding: The Navy + Teal color palette was chosen to signal trust and non-profit credibility — deliberately avoiding the "startup" aesthetic. The design communicates: this is a public service, not a product.
• Extensible Data Model: The pdf_field_map in each hospital record is pre-wired for future PDF form-filling automation. When real hospital PDF forms are obtained, the system can programmatically fill fields via pdf-lib's form API without architectural changes.

The Strategic Implication

Civis Health demonstrates a principle I've carried from oncology marketing into product development: the most impactful work often addresses access barriers, not just scientific innovation.

In pharma, we invest billions in drug development and clinical trials — but patients who can't navigate the financial assistance landscape never access the treatments at all. Civis Health attacks this access gap directly, with a lean tool that automates what previously required a social worker, a stack of paperwork, and weeks of processing time.

The same constraint-driven, domain-informed approach that drives my work at Pfizer — where I translate clinical data into commercial strategy — applies here. The difference is the output: instead of a brand strategy, it's a tool that puts a pre-filled application in a patient's hands in under 60 seconds.

Authored by Daniel Tran — Product Architect & PharmD.