Implementation and operations

How to increase retina follow-up adherence in people with diabetes: reminders, risk segmentation, and referral closure

Practical clinical and digital strategies to help more people with diabetes complete annual fundus screening, with simple metrics and without overloading ophthalmology.

How to increase retina follow-up adherence in people with diabetes: reminders, risk segmentation, and referral closure

In diabetes care, diabetic retinopathy often progresses without symptoms for years. That is why the most common problem is not “lack of treatment,” but lack of timely follow-up: patients who do not complete annual fundus screening, or who are lost after a high-risk result.

This article is a practical guide for teams in: - Ophthalmology and retina - Internal medicine/diabetology - Hospital and clinic operations - Health insurers/payers - Systems and innovation

Goal: increase adherence and close referrals without saturating specialist capacity.

Clinical recommendations for frequency and criteria:
- ADA Standards of Care - Retinopathy (updated annually): https://diabetesjournals.org/care/article/49/Supplement_1/S261/163919/12-Retinopathy-Neuropathy-and-Foot-Care-Standards
- AAO Preferred Practice Pattern - Diabetic Retinopathy: https://www.aao.org/education/preferred-practice-pattern/diabetic-retinopathy-ppp
- Practical summary (EyeWiki): https://eyewiki.org/Diabetic_Retinopathy_Screening


1) The real funnel: where patients are lost

In practice, loss happens at 3 points:

1) Before screening (not scheduled / no-show)
2) After screening (result is available but no one acts on it)
3) At referral (specialist follow-up is indicated, but appointment is never completed)

The solution is not a single “you are due” reminder. It is a workflow with clear ownership and measurement.


2) Reminders that work: what evidence says (and how to apply it pragmatically)

Evidence shows that SMS/messaging reminders can increase attendance for appointments and follow-up in healthcare. A Cochrane review found that text messaging improves attendance compared with no reminders or postal reminders, with low-to-moderate quality evidence (but consistent direction) and low cost.
Source: https://www.cochrane.org/evidence/CD007458_mobile-phone-messaging-reminders-attendance-healthcare-appointments

In retina/diabetes programs, studies also report improved screening attendance with SMS reminders. For example:
- Koshy et al. (2008) describes the cost-effectiveness potential of SMS to reduce no-shows in retina screening: https://pmc.ncbi.nlm.nih.gov/articles/PMC2438329/
- Chen et al. (2018) reports improved attendance using informative SMS reminders in rural clinics: https://www.sciencedirect.com/science/article/abs/pii/S0002939418303878

Simple rules so reminders do not become spam

  • Short message + explicit action: “Your annual retina screening is pending. Book now / attend on date / reply 1 to coordinate.”
  • Two touches, not ten: one reminder (7-14 days before) + one reinforcement (24-48 h before).
  • Minimal personalization: name + location + how to confirm/reschedule.
  • Offline alternative: phone call or PHC coordination for patients without mobile messaging access.

3) Risk segmentation: the key to improve adherence without collapsing retina services

Not every patient needs the same urgency. Segmentation helps you: - intensify outreach where clinical benefit is highest, - reduce pressure on ophthalmology for lower-risk cases.

Example operational scheme: - Red (high priority): referable DR-compatible result / critical finding / repeated “ungradable”
- target: confirmed appointment within 7-14 days (or sooner per clinical protocol). - Yellow (medium priority): mild/moderate changes without urgent criteria
- target: appointment or follow-up in 30-90 days. - Green (lower risk): annual follow-up planned
- target: ensure next screening happens without friction.

This aligns with triage logic to avoid service saturation, also covered in:
/en/blog/teleophthalmology-implementation-models-hospitals-clinics-campaigns/


4) “Loop closure”: the metric that separates real programs from endless pilots

Coverage matters. But in diabetes care, outcomes improve when referrals are actually completed.

Minimum monthly dashboard metrics

1) Coverage: eligible vs screened
2) % gradable studies: gradable/ungradable images (and causes)
3) Time to result: capture -> pre-triage -> report (SLA)
4) Referral rate: % requiring specialist follow-up
5) Loop closure: % referred who were actually seen
6) Time to effective appointment: referred -> seen
7) Loss to follow-up: not contacted / no-shows / reschedules
8) Quality/Audit: second reads / discordances where applicable


5) Implementation checklist (hospital, clinic, payer)

A) Hospital / public network (PHC + referral center)

  • Define who schedules and who confirms appointments.
  • Reserve protected slots for red-priority referrals.
  • Create a contact pool (PHC / call center / community health workers) to recover no-shows.
  • Contingency plan for ungradable studies (fast recapture).

B) Clinic / private network

  • Offer screening as a “15-minute service” (capture) + remote report.
  • Explicit result SLA (for example 24-72 h by severity).
  • Automated reminders + simple rescheduling (without endless phone loops).

C) Health insurer / payer

  • Named list: members with diabetes and no fundus screening in 12 months.
  • Segmented campaign (green/yellow/red) with different contact intensity.
  • Provider agreements and incentives to absorb high-risk referrals.

Useful framework for digital interventions (including targeted communication and reminders):
WHO - Recommendations on Digital Interventions for Health System Strengthening: https://www.who.int/publications/i/item/9789241550505


6) Practical example: how Retinar helps with this challenge

In real programs, Retinar typically contributes in two areas that directly affect adherence and loop closure:

1) Reduce friction on day 0 (useful capture on first attempt)
- Capture quality control and standardization to reduce ungradable studies and unnecessary repeat visits.

2) Prioritization and actionable lists
- A high-risk result can become a priority contact list, with status traceability: pending -> contacted -> appointment confirmed -> attended.
- This list can integrate into local operations (scheduling, call center, PHC), so results do not remain trapped in a PDF or an unowned inbox.

If your organization is also defining how to combine automation with professional review, this framework can help:
/en/blog/human-in-the-loop-ai-in-healthcare/


7) Message templates (ready to use)

Reminder 1 (7-14 days before)
> Hi {Name}. Your annual diabetes retina screening is due. You can complete it at {Location}. Reply “YES” and we will coordinate your appointment, or “CHANGE” to reschedule.

Reminder 2 (24-48 h before)
> Hi {Name}. This is a reminder for your retina screening at {Location} on {Day} at {Time}. If you cannot attend, reply “CHANGE”.

Priority referral (red)
> Hi {Name}. Your retina study requires specialist follow-up. We can prioritize your appointment at {Location}. Reply “YES” and we will call you today.


Closing

If your diabetes retina program is not improving despite “having technology,” review this: clear scheduling ownership + risk segmentation + loop-closure dashboard. With those three elements, reminders stop being isolated messages and become measurable operations.

If useful, we can run a technical-clinical demo and build a phased plan (capture, triage, referral, follow-up) adapted to your hospital, clinic, or payer with metrics from month one. Contact us through retinar.com.ar.

Implement Retinar and reduce blindness cases due to diabetes

Share your institutional context and we will propose an implementation plan.

Contact us