Dental and medical practices often ask about two related problems: missed appointments and the staff time spent on manual reminder calls. The baseline varies by practice, so measure it before deciding what the automation needs to improve.
An automated sequence can send a reminder, collect a confirmation or cancellation, and update the schedule where the EHR supports it. The workflow is only useful if the communication and every connected service are configured for the data involved.
Here's what actually matters from a HIPAA standpoint, and what doesn't.
What HIPAA actually says about automated messages
HIPAA treats appointment reminders as part of treatment, so they can be sent without a separate HIPAA authorization. HHS states this directly in its appointment-reminder guidance. Providers still need reasonable safeguards and must accommodate reasonable confidential-communication requests.
A short message containing only what the patient needs to act is safer than one that names a diagnosis or procedure. Even the practice name can reveal a specialty, so message content should reflect the patient's communication preferences and the practice's risk assessment. HHS recommends limiting what is disclosed in reminder messages in its patient-message guidance.
A message that includes diagnosis information, insurance details, or treatment specifics in the SMS body is a different conversation. Don't do that.
The practical checks:
- Confirm the patient-provided contact details and communication preferences
- Limit the message to the information needed for the reminder
- Apply the relevant HIPAA, telecom, carrier, and state-law requirements rather than treating them as one consent rule
- Put a BAA in place with each vendor that acts as a business associate and confirm that the exact service is covered
- Configure access and audit controls for systems that contain or use ePHI
Check every service in the data path
Twilio requires an eligible edition, an executed BAA, an eligible service, and a HIPAA-enabled account. Its current setup requirements are documented under HIPAA Accounts. A general-purpose automation platform or any other service in the data path also needs to be reviewed separately; one vendor's BAA does not cover another vendor.
A vendor's general security page is not a substitute for a signed BAA and eligible-services list. The deployment also remains a shared responsibility: the practice controls what is sent, who can access it, and how responses return to the EHR.
The workflow that works
Here's the version I typically build for clinics:
- 1Appointment created in the EHR (Dentrix, Athenahealth, eClinicalWorks)
- 248-hour reminder fires via SMS: time, date, provider, confirmation link
- 3Patient replies YES/NO or clicks a link
- 4YES: appointment confirmed, no action needed
- 5NO: slot marked as available, front desk gets a notification, waitlist patient contacted if applicable
- 6No reply by 24 hours: second reminder fires
- 7Day-of: 2-hour reminder with office address and a link to any digital intake forms
The system logs every send, delivery, and reply. Confirmation status is visible in a dashboard the front desk can see in real time.
The goal is fewer calls to voicemail, fewer sticky notes, and a visible exception queue when someone still needs staff follow-up.
What this can do to no-show rates
In one 60-day dental engagement, the practice's recorded no-show rate fell from 18% to 7.4% after an SMS confirmation workflow was introduced. That is one engagement, not a guaranteed benchmark. Randomized studies have found that text reminders can reduce nonattendance, but the effect varies by setting and baseline process; one trial reported an 8.9% nonattendance rate with SMS versus 11.9% under standard care (PubMed).
Use the practice's own no-show rate, appointment value, cancellation-fill rate, message volume, software costs, and support costs to model the return.
If you're thinking about this for your practice
Start with the EHR's built-in reminder tools. Check whether they support the required message controls, two-way confirmation, schedule writeback, waitlist routing, auditability, and patient preferences. Consider a custom integration only for gaps that matter enough to justify its implementation and maintenance cost.
The build time for a custom integration depends on the EHR's access model, supported operations, vendor approval, and testing requirements. Scope those constraints before promising a launch date.
