> ## Documentation Index
> Fetch the complete documentation index at: https://docs.morf.health/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Sending Appointment Communications
> Build a full appointment communication workflow with time-window filters, deduplication, and cancellation handling.
This doc is for anyone building workflows that send communications around appointments — reminders, cancellation follow-ups, or post-appointment review requests. Appointment communication in Morf is flexible: you control exactly when messages fire and what conditions have to be true first. Use the expressions below to build a workflow that fits your practice needs.
***
## Filter by event type
***Use this when** you want the workflow to run only when an appointment is booked or changed, not on every appointment event.*
```
morf_event_type.contains("SCHEDULED") || morf_event_type.contains("UPDATED")
```
Use these sample prompts with Flo to generate a CEL expression that fits your goals.
```
Only run this workflow when an appointment is scheduled or updated.
```
```
Continue when an appointment is done.
```
***
## Check the appointment is still in the future
***Use this when** you want to make sure you're not sending a reminder for an appointment that has already passed.*
```
appointment_datetime.isAfter(morf.now())
```
Use these sample prompts with Flo to generate a CEL expression that fits your goals.
```
Only send this reminder if the appointment hasn't already happened.
```
```
Skip this appointment if the appointment date has already passed.
```
***
## Filter to a specific time window
***Use this when** you want to send a reminder a specific amount of time before an appointment — like 24 hours before or 7 days before.*
**Start here — common time windows:**
**More than 7 days away:**
```
appointment_datetime.isAfter(morf.now().add(parseDuration("167h")))
```
**Within 24 hours:**
```
morf.now().isAfter(appointment_datetime.sub(parseDuration("24h")))
```
**Within 1 hour:**
```
morf.now().isAfter(appointment_datetime.sub(parseDuration("1h")))
```
**Duration reference:**
| Window | Expression |
| ---------- | ----------------------- |
| 10 minutes | `parseDuration("10m")` |
| 1 hour | `parseDuration("1h")` |
| 24 hours | `parseDuration("24h")` |
| 3 days | `parseDuration("72h")` |
| 7 days | `parseDuration("168h")` |
| 30 days | `parseDuration("720h")` |
**Days don't have a shortcut.** Write hours instead: 3 days = `"72h"`, 7 days = `"168h"`. If you use `"7d"`, the expression won't work.
Use these sample prompts with Flo to generate a CEL expression that fits your goals.
```
Write a filter that passes only if the appointment is between 24 and 72 hours from now.
```
```
I want to send a reminder when there are exactly 7 days until the appointment.
```
```
I want to send a review request a day after the appointment is complete.
```
***
## Timezone-aware time windows
***Use this when** your patients are in different time zones and you want the reminder timing to respect their local time — for example, only sending if the appointment is more than 72 hours away as of 9 AM in their timezone.*
**Advanced — use only if basic time windows aren't precise enough for your needs:**
```
appointment_updated_at.isBefore(
parseTimestamp(
appointment_datetime.customFormatInTimezone("2006-01-02", appointment_patient_timezone_identifier.orValue("America/Phoenix"))
+ "T09:00:00"
+ appointment_timezone_offset
).sub(parseDuration("72h"))
)
```
**Annotated breakdown:**
```
appointment_updated_at.isBefore( // Was the appointment updated before the cutoff?
parseTimestamp( // Convert the cutoff string back to a timestamp
appointment_datetime
.customFormatInTimezone( // Format the date in the patient's local timezone
"2006-01-02",
appointment_patient_timezone_identifier.orValue("America/Phoenix")
)
+ "T09:00:00" // Anchor to 9:00 AM local time
+ appointment_timezone_offset // Apply the UTC offset for that timezone
).sub(parseDuration("72h")) // Go back 72 hours from that 9 AM point
)
```
Replace `"America/Phoenix"` with the appropriate default timezone for your patient population, and change `"T09:00:00"` to the local time you want to anchor to. `appointment_patient_timezone_identifier` is populated automatically by Healthie.
Use these sample prompts with Flo to generate a CEL expression that fits your goals.
```
Write a filter that checks if the appointment is more than 72 hours away, anchored to 9 AM in the patient's timezone.
```
```
Only send if the appointment is at least 3 days away, based on the patient's local time.
```
```
Use the patient's timezone to check if the appointment is more than 48 hours from now.
```
***
## Check the appointment hasn't been cancelled
***Use this when** you want to verify the appointment is still active before sending a reminder or confirmation.*
**Start here — check the status field directly:**
```
appointment_status.orValue("") != "Cancelled" &&
appointment_status.orValue("") != "Re-Scheduled"
```
**Advanced — check against a live lookup result:**
```
list_appts.exists(appt,
appt.appointment_id == appointment_id &&
appt.appointment_status == "Confirmed"
)
```
Use these sample prompts with Flo to generate a CEL expression that fits your goals.
```
Don't send a reminder if the appointment has been cancelled or rescheduled.
```
```
Only send if the appointment status is Confirmed.
```
***
## Ready to build?
Start building your appointment reminder workflow now.
Want a walkthrough or help getting started? Book a demo with us.
***
## Further reading
For the complete CEL function reference — including all available operators, string functions, date helpers, and optional chaining syntax — see the [CEL Reference](/docs/libraries/cel).