How LMH Health answers every patient call on the first ring
Building the foundation for a system-wide, AI-enabled call center
of 16,000+ inbound calls answered on the first ring
of calls routed directly to clinical and operational teams
of callers engaged with the AI Agent
of inbound calls handled without reaching staff
At OrthoKansas, one of LMH Health 30 clinics, the front desk got roughly 300–400 calls on busy weekdays while also serving patients in person. The team was budgeted for 9 FTEs but was chronically short-staffed, and the live answer rate was essentially 0%. Every inbound call went to voicemail.
"Probably the biggest complaint that I get all the time is that I want to talk to a live person." – Candice Bonnel, Practice Manager, OrthoKansas
Front desk staff then handled every step of each call themselves. They worked through voicemails, returned calls, and routed each patient to the right team: scheduling, clinical, billing, records, or a provider's care team. Without a central call center, that work added up to a backlog, inconsistent routing, and lost voicemails. Patients waited at least one business day for a callback, and up to five days when it took three attempts.
About 75% of those voicemails were scheduling requests. That meant revenue was leaking from appointments that never got booked, and patients had a poor experience just trying to get care.
"I hear that all the time – 'I've called but nobody returned my call.' Well, maybe that voicemail got lost in translation or it got dropped." – Candice Bonnel
The problem spanned all 30 clinics, each running its own phone tree on an outdated system. LMH Health leadership recognized a need to address this problem in order to maintain a positive patient experience across its system.
With a partnership already in place with Notable, LMH Health turned to automation to alleviate this burden on staff and improve the patient experience with a scalable, repeatable Voice AI workflow. Together they built a Voice AI Agent that handles:
- Answers to common questions, such as clinic address and fax number
- Routing to scheduling, billing, medical records, and clinical teams
- 24/7 coverage, with after-hours transfer to the answering service for urgent clinical needs
The workflow was rolled out in phases across two pilot sites.
Phase 1, OrthoKansas: The agent went live in 10 weeks end to end, covering content build, integration testing, and user acceptance testing, followed by four weeks of monitoring. In its first 24 hours it handled nearly 500 calls, more than a typical busy weekday at the clinic.
Phase 2, Baldwin City Primary Care: The second site reused OrthoKansas's prompt architecture, routing patterns, and testing approach, so it could go live even faster.
LMH Health reviewed 100% of the agent's calls through its quality assurance workflow, with human review by site leadership and support teams.
LMH Health designed this first phase so that every call is answered right away and sent to the right place. Early results at both pilot sites show that foundation is in place.
Every call answered on the first ring. The agent has answered 100% of more than 16,000 inbound calls, at any hour, so patients hear a response right away instead of an unanswered ring. That includes 282 after-hours calls, where patients are prompted to leave a message or reach the answering service. 78.4% of callers engaged with the AI Agent, a sign that patients are comfortable getting help this way.
Patients routed directly to the right team. About 40% of calls go directly to provider care teams, triage nursing, billing, records, or the after-hours service instead of passing through the front desk first. The remaining 60% go to the front desk already triaged.
Less noise for the front desk. 24.3% of inbound calls ended without staff involvement, giving the team more time to focus on the ones that need their attention most.
What's next:
With its prompt architecture, routing patterns, and testing approach now reused across two sites, LMH Health has a foundation to build on. LMH Health and Notable are refining the current phase and preparing to launch outbound calls for referrals. From there, LMH Health plans to expand to Neurology, then to rural health clinics and more specialties, with the goal of bringing all 30 clinics under one AI-enabled access model.
LMH Health's long-term ambition is to automate up to 95% of call center functions system-wide, for a $10–15 million annual operating margin impact.