Why Retention Matters More Than Ever
GLP-1 programs often include extended ocular safety follow-up (e.g., 24–36 months). Sustained retention preserves statistical power and reduces bias; across clinical research, drop-out rates of ~25–30% are common, with higher rates reported in some settings.
Regulators are watching ocular safety closely: on June 6, 2025, EMA’s PRAC concluded that NAION is a very rare side effect of semaglutide and recommended stopping semaglutide if NAION occurs; product information should be updated accordingly. The AAO highlighted this determination and advises patients to seek prompt eye care and discuss any changes with their prescriber.
Evidence on semaglutide and NAION remains mixed: several cohorts and reviews show increased risk (e.g., HR≈2–4), while others show no association—underscoring the need for careful documentation and long-term follow-up rather than assumptions.
Bottom line: strong retention protects endpoint interpretability, reduces bias, and supports regulatory confidence.
Where traditional approaches break down
- Generic engagement that ignores changing motivations over time
- Site-to-site variability in process, experience, and scheduling
- Reactive outreach (intervening only after a missed visit)
- No long-term architecture for months 18–36, when “engagement fatigue” peaks
These patterns drive preventable attrition and cost.
The 20/20 Onsite approach to retention
We’re built for long-horizon ophthalmic follow-up. Our point-of-need model provides standardized, patient-centric assessments to sponsors and sites, reducing the burden and keeping participants engaged.
- Risk-based engagement: Structured check-ins and reminders, escalating support for participants with emerging risk (e.g., schedule instability, transport barriers).
- Multi-site consistency: Calibrated equipment, certified staff, and harmonized visit flows reduce variability and “experience gaps” that increase dropout risk.
- Burden reduction at point-of-need: Flexible scheduling and on-location assessments minimize travel/time costs.
- Regulatory-ready documentation: Clear tracking of contacts, withdrawals, and continued use of already-accrued data, aligned to FDA guidance.
GLP-1–specific considerations for eye safety follow-up
|
Challenge |
Why it drives dropout |
How we address it |
|
24–36-month horizons |
Motivation wanes; life changes accumulate |
Long-term engagement plans with milestone touchpoints and “why this visit matters” education |
|
Multi-site networks |
Inconsistent experience/scheduling |
Standardized SOPs, equipment, and visit design across locations |
|
Safety-signal news cycle (e.g., NAION headlines) |
Anxiety → missed visits/withdrawal |
Clear, balanced patient messaging referencing current guidance; rapid routing for clinical concerns. |
|
Competing time/transport demands |
Practical barriers compound over time |
Point-of-need assessments, extended hours, and concierge scheduling. |
Patient Retention Toolkit
There’s no universal “acceptable” dropout rate, but higher retention improves data power and credibility. Plans should proactively minimize attrition, document reasons for withdrawal, and preserve analyzable data per FDA/OHRP guidance.
Your retention toolkit should include:
- Clear risk flags and escalation playbooks
- A pre-scheduled reminder cadence for the full monitoring duration
- Standardized visit flows and patient-facing materials across sites
- Documentation for withdrawals and continued use of already-accrued data (as permitted)
Talk to us
Planning a GLP-1 eye safety study or rescuing one at risk of attrition? We’ll help you operationalize a retention-first plan that reduces burden, keeps patients engaged, and protects ocular endpoints.
“Patient safety is at the heart of what we do at 20/20 Onsite. By improving access to care, we're helping to gather the data that guides patient safety long-term.”
-Dr. Katherine Harkins, Director, Optometry, 20/20 Onsite
Book a call → We’ll map where and why participants tend to drop, then configure a point-of-need plan aligned with current guidance.