← Work

Process Engineering  ·  Thesis Project

Healthcare Reimbursement Process Standardization

OAMal Art. 71  ·  Ente Ospedaliero Cantonale (EOC)

Outcome

Delivered 88 validated request templates, a centralized KPI tracking system, and a governance model — without waiting for a multi-year IT roadmap. Established case-level measurability where none existed. Reduced the two structural causes of hospital-side rework to addressable process failures with specific interventions.

Context

Oncology drug reimbursement workflow within the Ente Ospedaliero Cantonale (EOC), governed by OAMal Art. 71. Approval timing directly impacts patient access to therapy. Five stakeholder classes: patients, physicians, administrative staff, insurers, and pharmaceutical companies. Legacy EHR (GECO) with no structured field for clinical justification — free text only.

Problem

The process ran on fragmented data capture and tacit clinical knowledge. Rework cycles were unpredictable. Root causes were invisible because the unit of analysis was wrong — event-based logs distorted patient cases, inflating counts and masking where failures actually occurred.

Constraints

  • Data window: Jan–Dec 2024  ·  965 requests
  • Event-based log distortion: average 2–3 extra rows per patient; over 50% of cases affected
  • Mean lead time: 17.16 days across 965 requests
  • Rework rate: ~10% of requests triggered additional information cycles
  • GECO integration: multi-year dependency with competing IT priorities — ruled out as a solution path
  • Sensitive data: highly confidential medical history and correspondence throughout

Approach

01 —

Unit of analysis redefinition

Reconstructed patient-level histories from event-based logs. Redefined the unit of analysis from communication events to patient case. This was the prerequisite for any meaningful measurement — without it, duplicate event rows made rework rates and lead times unreadable.

02 —

KPI framework

Built a KPI framework across all 965 cases: Lead Time, Rework Rate, and a tunable weighted Friction Index. Applied to the reconstructed patient-level dataset.

03 —

Root cause identification

Identified that 83% of hospital-side rework (excluding insurer administrative delays) had two structural causes: incomplete documentation and normative or indication ambiguity. Not volume. Not staff capacity. Two addressable failure modes.

04 —

Standardization layer

Designed 88 request templates compatible with GECO's free-text constraints. Each template standardizes upstream data entry for a specific clinical scenario, reducing the ambiguity that triggers rework loops.

05 —

Tracking infrastructure

Built a centralized tracking table with standardized inputs and automated KPI calculation to enable ongoing monitoring without manual aggregation.

06 —

Governance model

Proposed a Knowledge Manager role to sustain standards, maintain the template knowledge base, and manage continuous improvement. The GECO integration path was ruled out explicitly — a multi-year IT dependency with competing priorities would not have delivered a deployable intervention within the project window.

Deliverable

88 validated request templates deployable within GECO's current constraints. Centralized tracking system with automated KPI output. Governance model with defined Knowledge Manager role. Case-level measurement framework replacing the distorted event-based view.