Healthcare support is one of the harder problems in AI voice. Patients calling about appointments, prescriptions, and test results are often anxious. They want to speak in their first language. They need to trust that the system understands them correctly. And the margin for error — wrong appointment time, misheard prescription details — is genuinely higher than in most other industries.
Lumen Labs runs a network of diagnostic labs and outpatient clinics across Maharashtra and Gujarat. 18 facilities, 400,000 patient contacts annually, a support team of 40 agents stretched thin across a 16-hour call window. When they deployed AIVA in November, they were handling 12,000 calls a month and their average wait time was 6.4 minutes.
The problem with IVR
Lumen's original IVR system was a DTMF tree — press 1 for appointments, press 2 for test results, press 3 for billing. It was universally despised. Patients calling about test results had to navigate four levels of menu before reaching a human. Patients who spoke Marathi or Gujarati and didn't feel confident with the English-language menu options often hung up and called back, hoping for a different agent.
The IVR was also creating work, not saving it. Agents spent the first 90 seconds of every call figuring out what the patient had already done in the IVR system and what they actually needed. Information gathered by the IVR wasn't usable because it wasn't surfaced to the agent in any structured way.
Lumen's head of operations, Vikram Shah, told us: "Our IVR was a waiting room, not a support system. It held patients in a queue but didn't help anyone."
The integration
Deploying AIVA at Lumen required connecting to three systems: their appointment scheduling platform, their lab results database (with strict access controls and PHI handling), and their billing system for payment queries.
The PHI access was the most careful part. We worked with Lumen's data security team to implement field-level access controls — AIVA can confirm whether test results are ready and direct patients to the right portal to view them, but it never reads or speaks the results themselves. For anything requiring actual result discussion, AIVA routes to a medical professional. This boundary is hard-coded and not configurable.
Setup took eight days — longer than a typical AIVA deployment, mostly due to the access control configuration and the testing required to verify PHI boundaries were working correctly in every scenario we could construct.
If you're deploying AIVA in healthcare, medical, or financial contexts: we require a security review before go-live. The PHI boundary configuration is standard for any deployment involving health or financial data.
The results
After 90 days of operation:
- Average wait time: 6.4 minutes → 52 seconds
- First-contact resolution rate: 38% (human) → 89% (AIVA-handled calls)
- Calls requiring human escalation: 62% → 11%
- Patient satisfaction (post-call survey): 3.1 → 4.4 / 5
- Agent utilisation: shifted from tier-one volume to complex patient queries and callbacks
The most striking finding from Lumen's own patient survey data: 71% of patients who interacted with AIVA reported preferring the experience to their previous IVR interactions. Patients didn't care whether they were talking to an AI. They cared whether they got a fast, accurate answer in a language they were comfortable in.
Vikram Shah's summary: "We thought the barrier to AI in healthcare was trust. It turns out the barrier was IVR. Patients will happily talk to an AI that listens. They won't tolerate a menu tree that doesn't."