Mohamad-Ali Salloum is a Pharmacist and science writer. He loves simplifying science to the general public and healthcare students through words and illustrations. When he's not working, you can usually find him in the gym, reading a book, or learning a new skill.
Stop Blaming People, Start Fixing Systems: Human Error vs. System Error in Clinical Research
Share
Stop Blaming “Human Error”: How ICH GCP E6(R3) Reframes Quality in Clinical Trials. Clinical trials still see the same recurring deviations—late SAE reporting, missed visit windows, outdated ICF versions, missing data. And too often, the explanation offered is simply “human error.” But the newly finalized ICH GCP E6(R3) makes it clear: lasting quality comes from system design , not blame. E6(R3) modernizes GCP with flexible, risk‑based, technology‑enabled expectations and explicitly strengthens Quality by Design (QbD), proportionality, and management of critical‑to‑quality factors.
If “human error” keeps recurring, the real cause isn’t the person—it’s the system they’re working in.
The Two Lenses: Person vs. System
The Person Approach: Retrain and Move On
This outdated model treats deviations as individual lapses—forgetfulness, miscalculations, oversight. The typical response is retraining or reminders. The outcome? The error returns because nothing structural changed.
The System Approach: Design for Reliability
E6(R3) reinforces that quality must be designed into trials from the start, not inspected in afterward. The system approach emphasizes usability of workflows, automation and digital tools, clarity of roles, proportional controls, critical‑to‑quality focus, and better data governance. That aligns with modern patient‑safety science and high‑reliability industries.
Reason’s Swiss Cheese Model still applies: every layer—protocol, SOPs, training, systems—has holes. Incidents occur when holes align. Strengthen each layer and misalign the holes.
What ICH GCP E6(R3) Expects
- Quality by Design (QbD): Identify critical‑to‑quality (CTQ) factors and build controls where they matter most.
- Proportionality and Fitness‑for‑Purpose: Right‑size controls; not all data must be error‑free if reliability is preserved.
- Modern Data Governance: Robust oversight of computerized systems, audit trails, eConsent, and remote processes—media‑neutral by design.
- Strengthened Sponsor Responsibilities: Demonstrate process‑level oversight and rationale for decisions across the trial lifecycle—not just at closeout.
- Support for Innovative Designs: Decentralized, digital, and pragmatic trials are enabled when risks are proactively managed.
Tip: Document why your controls are sufficient for each CTQ—this is often what inspectors want to see.
Plain‑English Definitions
Human Error: A slip or mistake by a person. Often a symptom , not the root cause.
System Error: A condition in the environment, process, tools, or workload that increases error probability across people and time.
Fixing the system reduces repeat deviations dramatically.
A CRA’s Four‑Question Decision Aid
- Task design: Is the task dependent on memory or manual calculation? → System.
- Environment/workload: Were interruptions, stress, or understaffing involved? → System.
- Information/interface: Are SOPs clear and tools user‑friendly? If not → System.
- Controls: Are there alerts, double‑checks, dashboards, or QTLs? If missing → System.
Even if a person contributed, E6(R3) expects you to assess and strengthen system barriers.
Field Scenarios: How E6(R3) Should Change Your Response
1) Late SAE Reporting
Old explanation: Coordinator forgot to report.
E6(R3) approach:
- Process Daily AE review checklists and backup coverage.
- Tools EMR keyword alerts and a centralized SAE tracker with due‑date warnings.
- Oversight Weekly PI/CRA timeliness review with QbD‑aligned CTQs and QTLs.
Result: A workflow that makes timely reporting the default.
2) Out‑of‑Window Visits
Old explanation: “Staff miscalculated.”
E6(R3) approach: Validated visit‑window calculator, automatic calendar reminders, double‑checks during scheduling, and capacity planning to reduce overload—Quality by Design in action.
3) Wrong ICF Version Used
Old explanation: “Nurse grabbed the wrong form.”
E6(R3) approach: Single digital source of truth for ICFs; watermarked versions; removal of obsolete copies; eConsent with automated version control; and a CRA‑review queue for new versions.
This enhances compliance and inspection readiness.
What Inspectors Commonly Cite Under E6(R3)
- Repeated deviations “fixed” only with retraining
- “Human error” listed repeatedly as the root cause
- Weak justification for controls or missing decision rationale
- CAPAs missing effectiveness verification or closed late
Actions You Can Take This Month
- Map a high‑risk process (e.g., SAE → reporting) and identify memory‑based steps; add automation or checklists.
- Define critical‑to‑quality (CTQ) factors and proportional controls aligned with E6(R3).
- Document decision‑making—not just outcomes—consistent with R3 expectations for process rationales.
- Strengthen eTMF/CTMS oversight to ensure end‑to‑end visibility of essential records.
- Prepare for inspections by demonstrating real‑time oversight, rationale, data governance, and continuous risk management.
Quick Knowledge Check
~3 minutesList of Services
ABOUT THE AUTHOR
Mohamad-Ali Salloum, PharmD
Share
Recent articles:
















