Your Patients Aren't Chatbot Tickets

Eighty-four percent of health insurers are already using AI and machine learning. UnitedHealth is pushing to route more than half its customer calls through chatbots. The patient's first contact with the healthcare system is increasingly a machine. For practitioners who built their model on the opposite principle, that shift changes the landscape around you.

The insurance industry's AI bet

The numbers aren't subtle. Eighty-four percent of health insurance companies are using AI or machine learning in some capacity. UnitedHealth Group -- the largest health insurer in the country -- is actively targeting more than 50% of customer service calls to be handled by chatbots. Oscar Health is automating over 4,000 support tickets per month. Industry projections suggest 95% of insurance customer interactions will be AI-facilitated by the end of 2026.

This isn't coming. It's here.

What it means in practice: the patient's first point of contact with the healthcare system -- the call about a prior authorization, the question about coverage, the appeal of a denied claim -- is increasingly handled by a machine. Before a patient ever reaches a physician, they've likely already interacted with an algorithm.

Patients are fighting bots with bots

Here's where it gets interesting. Patients have noticed what's happening, and they're responding in kind.

Tools like DoNotPay and specialized AI appeal generators are helping patients fight automated claim denials with their own automated appeals. It's AI versus AI, with the patient's health coverage hanging in the balance. When an insurer uses an algorithm to deny a prior authorization in seconds, patients are using AI to generate an appeal letter in minutes.

This is the healthcare interaction model that's emerging: machines talking to machines about whether a human gets the care their doctor ordered. Nobody planned this system. It's what happens when automation optimizes for cost on both sides.

What this means for independent practices

If you're a specialist or a Creighton Model specialty care provider, you already know the billing landscape is difficult. specialty evaluations don't fit neatly into standard CPT codes. The time-intensive nature of the work -- the charting instruction, the cycle-by-cycle review, the surgical planning based on months of observed biomarkers -- doesn't align with insurance models built around 15-minute encounters.

Now add chatbots to that equation.

The same insurance systems that already underpay for specialty evaluations are investing billions in AI to further reduce the cost of processing claims. When a prior authorization request for a diagnostic laparoscopy lands in a system optimized to find reasons to deny, the nuance of why this particular patient needs this particular procedure based on three months of Creighton charting data doesn't translate well to an algorithm's decision tree.

Automated systems are built to process volume. specialty care is built to address the individual. Those two design principles are fundamentally in tension.

The contrast is the point

There's a reason patients find their way to specialty medicine. Many of them have already been through the conventional system. They've been the ticket in someone's queue. They've received the form letter, the automated denial, the chatbot response that didn't address their actual question.

They chose specialty because someone was willing to sit with them, look at their chart, and treat them as a person with a specific history and specific goals. The Creighton Model charting system requires sustained human observation and interpretation. specialty evaluations take time, conversation, and context. These aren't inefficiencies. They're the method.

When a patient who's been bounced between chatbots for three weeks finally sits down with a practitioner who actually listens to them for 45 minutes, that experience isn't a throwback. It's a differentiator. And increasingly, it's the thing patients are willing to travel across state lines to access.

The relationship is the instrument

There's a temptation to see the automation trend as someone else's problem. Your practice doesn't use chatbots for patient interactions. You aren't automating triage. You're not the ones replacing human contact with algorithms.

But the ecosystem around you is changing. Your patients are arriving at your office after navigating an increasingly automated healthcare system. Their expectations are being shaped by those interactions -- sometimes lowered, sometimes made more cynical, sometimes made more desperate for something different.

Understanding that context matters for how you position your practice, how you communicate your value, and how you explain to patients why your approach takes longer and costs what it costs. The answer isn't complicated: the physician-patient relationship isn't a legacy workflow to be optimized. It's the clinical instrument.

The practices that can articulate that clearly -- on their website, in their patient communications, in how they present themselves to the world -- are the ones that will stand out as the rest of healthcare moves in the other direction.