Patient Recall Systems: How They Work, Message Samples, and What to Automate

Updated 1 November 2026. Operational guide, not clinical or legal advice.

What a recall system does

1. Computes due dates. For every patient and every recall-eligible service, the next due date is the last completed service plus the interval your clinical guidelines set (annual visit, periodic screening, follow-up after a procedure, chronic-care check).
2. Checks status. Before any message: not booked already, not declined this cycle, not deceased or transferred, not do-not-contact, consent on file for the channel.
3. Runs a cadence. Typically one message 4–6 weeks before the due date, one at the due date, one 4 weeks after; then a pause and a hand-off to reactivation if the patient lapses.
4. Books. A one-click link to online scheduling or a reply option; a call for patients who prefer it.
5. Reminds. Once booked, the appointment-reminder sequence takes over (two touches, confirmation required; see the evidence review).
6. Measures. Recall completion rate by service and cohort; attendance; time from due date to visit.
7. Resets. A completed service writes the next due date.

Recall, reminder, reactivation: three systems, one profile

Scroll for more → On small screens, choose cards for wide tables.

All three read the same patient profile: identity, consent per channel, status, last service, next due date. When they live in three tools with three copies of the patient, the status layer drifts and the patient gets a recall for a service they declined last month. That is the problem EngageHealth's unified patient profile exists to remove.

Message samples (original copy; no procedure names in subject lines)

Recall, 4–6 weeks before due

Recall, at the due date

Recall, 4 weeks after due

Rules: the practice's usual sender; a monitored reply-to; one-click unsubscribe headers (required by Gmail and Yahoo for bulk senders); the booking page first-party and free of third-party pixels; the word "routine" rather than the service name in anything visible on a lock screen. Detail belongs in the portal.

What to automate, what to keep human

Scroll for more → On small screens, choose cards for wide tables.

EngageHealth automates the first column. Unified patient profile with consent per channel, due-date and status fields from the EHR, cadence with stop rules, aggregate attendance reporting; BAAs confirmed per deployment. Request early access →

Stop rules and consent

Scroll for more → On small screens, choose cards for wide tables.

Channel consent is per channel, not global: a patient who agreed to SMS reminders has not agreed to email newsletters. Record the consent source and date for each.

Measuring recall

• Recall completion rate: patients who completed the due service within the cycle window ÷ patients recalled, by service and by month of due date.
• Booking rate from recall messages: bookings within 30 days ÷ recalled, from the scheduling system, not from clicks.
• Attendance of recall-generated appointments (no-shows subtract; the evidence review has the levers).
• Time from due date to completed service, median.
• Opt-outs and complaints per batch.
• Coverage: share of eligible patients with a valid consented channel. Low coverage, not low open rates, is the usual reason recall underperforms.
• Aggregate only; no individual engagement leaves the BAA-covered system.

Common failures

1. Three copies of the patient. EHR, practice-management and engagement tools disagree on status; the patient gets recalled for a declined service.
2. Recall from the marketing list. The marketing list lacks the status gates; see the reactivation guide.
3. Service names on lock screens. "Your mammogram is due" in a subject line is a disclosure to anyone who sees the phone.
4. No stop after booking. The patient books on day 2 and still receives touches 2 and 3.
5. Measuring opens. The outcome is the completed service.
6. One channel for everyone. Patients without email or who declined email never get recalled; coverage is the metric to watch.

FAQ

What is a patient recall system? A process and tooling that computes when each patient is due for a service, checks their status and consent, sends a short cadence of reminders by the patient's chosen channel, books the visit, and measures completion. It is distinct from appointment reminders (for booked visits) and reactivation (for lapsed patients).

Is patient recall a marketing communication under HIPAA? No. Recall for the provider's own services is a treatment communication; no authorization is needed. The patient list is PHI, so the platform that sends recall needs a BAA and the messages carry no clinical detail.

How many recall messages should a practice send? Three per cycle: before the due date, at it, and after it; then pause or hand off to reactivation. Any booking, decline or reply ends the cycle.

What should a recall email say? That a routine visit is due, how to book in one click, and how to stop reminders. Not the service name in the subject line, not the reason for the last visit.

What recall completion rate is good? There is no published benchmark worth quoting outside dental vendor material. Measure completion by service and cohort month over month, and watch coverage (patients with a valid consented channel) first.

Should recall use email, text or calls? The patient's recorded preference. Trial evidence shows text reminders perform about as well as calls at lower cost; email is the cheapest first touch; calls go to patients with prior no-shows or no digital channel.

Running recall across several clinics or EHRs? EngageHealth unifies the patient profile, consent and due-date status across systems. Contact us →

Sources

• 45 CFR 164.501 (treatment and health care operations communications excluded from marketing); 45 CFR 164.312(e). https://www.law.cornell.edu/cfr/text/45/164.501
• Gurol-Urganci I et al., Cochrane Database of Systematic Reviews 2013, CD007458: text reminders RR 1.14 vs none; RR 0.99 vs phone; lower cost. https://www.cochrane.org/CD007458
• Dantas LF et al., Health Policy 2018: lead time and prior no-show history as strongest predictors; ~23% average no-show.
• MGMA Stat, Patient access priorities for 2026 (Dec 2025): no-shows the top priority at 27%; Patient no-shows in 2025 and 2026 polls. https://www.mgma.com/mgma-stat/patient-access-priorities-for-2026
• Google, Email sender guidelines; Yahoo, Sender Best Practices: one-click unsubscribe, complaint limits.
• EngageMKTG, EngageHealth, Patient Reactivation for Clinics and Health Systems, No-Show Rates and What Reminders Really Change, HIPAA-Compliant Email Marketing (internal).