Luma Health and Central both answer patient calls with AI, but they are built for very different buyers. Luma sells Operational AI to health systems that want access automated across call centers, referrals and EHR workflows. Central’s AI receptionist picks up calls, answers questions and books appointments for practices that mainly need the phone covered.
A feature checklist makes them look closer than they are. Both handle scheduling calls, both work after hours and both speak more than one language. What separates them is the kind of project you are running.
Are you redesigning patient access across a health system, or closing the gap between the calls coming in and the staff available to answer them? Whichever it is, the practice management principles that govern records, confidentiality and communication still apply when software takes the call.
A simple way to compare the two
Look at Central when
The priority is making sure no call goes unanswered. Central answers inbound calls around the clock, including after-hours, weekend and overflow calls. It books appointments during the conversation and hands off to a person when the AI cannot resolve a request.
Look at Luma Health when
The goal is enterprise access orchestration. Luma describes itself as Operational AI for healthcare, with its Spark AI core coordinating capabilities across products, agents and workflows. Its Navigator voice agent writes appointment and refill changes back into Epic, and Luma’s analysts support Navigator implementations with Epic, Oracle Health and other major EHRs.
That difference shapes everything else. A health system AI program and a phone coverage rollout have different owners, budgets and success measures.
Key differences at a glance
| Central | Luma Health | |
| Buying context | Evaluated when a practice needs every call answered around the clock | Evaluated as an Operational AI platform across access, referrals, faxes and call center workflows |
| Core strength | Answering, qualifying and booking calls 24/7 with a written summary after each conversation | Orchestrating access tasks at scale through Spark, Navigator and Fax Transform |
| Scheduling and EHR | Books live during the call through calendar tools such as Cal.com, Calendly and Google Calendar | Changes appointments and writes updates back into Epic |
| Channels and languages | Phone with booking links sent by SMS on request, 30+ languages | Phone with an optional switch to SMS, multiple languages |
| Escalation and access fit | Warm transfer to a live receptionist, connecting only if that person is available | At UAMS, deployed to complement the existing contact center and patient portal |
| HIPAA position | States full HIPAA compliance, BAAs with all customers handling PHI and ISO 27001:2022 certification | Describes Navigator as HIPAA-compliant, zero-retention agentic AI |
| Who should look closer | Independent practices and smaller groups with phone-heavy front desks | Health systems and large groups running EHR-integrated access programs |
The workflow difference most practices miss
On paper, both platforms take scheduling calls with AI. The practical difference is where the appointment lands and how much of the organization has to change to get there.
Luma is designed for health systems where the schedule lives inside Epic or another major EHR. Navigator can cancel and change appointments inside that record, which is what large access centers need. Luma also supplies resources to help teams evaluate Navigator with an AI governance committee.
Many practices are not there yet. They need the phone answered at 7 p.m., the appointment booked and a clear summary waiting for staff in the morning. Central is built for that narrower layer, and its receptionist page lists calendar tools such as Cal.com, Calendly and Google Calendar for booking.
Capabilities that matter in this comparison
Central
Live booking. The receptionist checks time zones and open slots, then books onto a connected calendar while the patient is still on the line. Practices comparing patient scheduling software for after-hours and overflow calls can judge Central on one job. Does the call end with a confirmed appointment in the calendar your staff check?
Knowledge training. Staff add a website URL or upload PDFs, Word files and other documents. The receptionist then answers common questions from that material.
Escalation and transfers. When the AI gets stuck or a caller asks for a person, it warm-transfers to a live receptionist and supports multi-department transfers. The handoff only connects if someone is available, so a defined fallback still matters.
Call records. Every conversation produces a written summary with key details and next steps. Calls can be categorized and tagged for analytics, and caller details are captured in a free built-in CRM.
Luma Health
Navigator. A patient-facing voice AI concierge that talks with patients by phone, can move them to SMS and understands multiple languages. Patients can change appointments or refill prescriptions without logging in or waiting on hold.
Outbound workflows. Navigator can also call patients to complete tasks such as pre-procedural guidance and readmission follow-up.
Fax Transform. Spark-powered processing that parses structured data from faxed referrals, DME requests and prescription refills.
Enterprise integration. Navigator writes appointment and refill information back into Epic, so changes appear in the record staff already use.
Evidence snapshots
Luma publishes customer outcomes, and they should be read as vendor-reported results rather than independent proof. Luma reports that UAMS automated 95% of inbound calls to its after-hours line with Navigator. The same case study reports more than 800 fewer call center hours a year spent processing cancellations and nearly 10,000 calls handled annually without staff intervention.
That case study is scoped to after-hours cancellations inside an Epic environment. Luma states UAMS saw results in under one month.
Central does not publish a healthcare outcome study on its receptionist page. Practices can generate their own evidence instead through its 10-day free trial, using real call scripts and their own appointment types.
Set a baseline before any pilot. Useful measures include missed and abandoned calls, voicemails left after hours, share of calls that end with a booking and time staff spend returning calls. Track the same measures afterward alongside errors and patient complaints.
Practice scenarios where the difference becomes clear
Health system access centers should confirm how much of their volume is routine cancellations and changes that could write straight back to the EHR. That is the workflow Luma’s UAMS case study covers.
Independent and specialty practices should look at where calls are lost. If the problem is evenings, weekends and overflow during busy hours, a focused receptionist is the smaller project.
Multi-site groups sitting between the two should map which workflows truly need EHR writeback and which only need a booked slot and a summary.
Where HIPAA terms fit
Both vendors publish a HIPAA position, and a published claim is only the starting point for due diligence. Luma describes Navigator as HIPAA-compliant, zero-retention agentic AI.
Central’s trust and security page states it is fully HIPAA compliant and executes Business Associate Agreements with every customer handling protected health information. The same page states customer data processed by its AI features is never used for model training.
How to decide
Evaluate Central more closely if
- Missed calls after hours and during busy periods are the main access problem.
- Your practice books through Cal.com, Calendly or Google Calendar.
- You want call summaries, lead capture and live booking without running an enterprise AI program.
Evaluate Luma Health more closely if
- Your schedule lives in Epic, Oracle Health or another major EHR and writeback is a hard requirement.
- You want referrals, faxes and outbound follow-up automated alongside inbound calls.
- An AI governance process already exists to review new patient-facing workflows.
Implementation notes
Software cannot fix unclear scheduling rules. Simplify appointment types where sensible and decide which requests the AI may handle before you switch it on.
Pilot on one line or one site first. Build the call scripts with the staff who answer the phone today, and define what happens when a transfer finds nobody available.
Test multilingual calls with real speakers and decide whether the AI should introduce itself as AI. Review call summaries daily during the first weeks.
Questions to ask before you choose
- Does the booking need to land in the EHR, or is a connected calendar enough?
- Who owns the rollout: access leadership, operations, IT or the practice manager?
- What happens to a clinical or urgent call the AI should not handle?
- How will success be measured: calls handled, bookings made, staff hours or all three?
Frequently asked questions
Is Central’s AI receptionist an Operational AI platform like Luma Health?
No. Central’s AI receptionist is a front-desk tool focused on answering calls, booking appointments and escalating to a person. Luma positions itself as an Operational AI platform coordinating access workflows across a health system.
Which platform makes more sense for enterprise health systems?
Luma is the stronger fit when EHR writeback, referral automation and contact center scale are priorities. Central is the stronger fit when a practice needs round-the-clock phone coverage without redesigning its access operation.
What should a practice confirm before an AI receptionist handles patient calls?
Confirm the vendor will sign a Business Associate Agreement before any patient information moves. HHS guidance on business associates explains why covered entities need that contract in place. Also confirm where bookings land, how urgent callers reach a person and what happens when no one picks up a transfer.
How should the two platforms be compared?
Start with the problem you need solved. A team fixing call center access across dozens of clinics has a different buying problem than a practice losing calls after 5 p.m.
The bottom line
Luma is a strong candidate for health systems that want AI woven through access, referrals and EHR workflows. Central is a strong candidate for practices that want every call answered and booked without an enterprise rollout.
The deciding evidence should come from a pilot using your own call volume, appointment rules and patients, not from a feature list or another organization’s results.
