Practicing physicians know all too well how prior authorization can disrupt the delivery of smooth, timely patient care. It’s a common cause of delays, administrative hurdles, and immense frustration for those trying to do what’s best for their patients. While efforts to address these challenges have been made in recent years, problems remain. The sheer number of authorizations needed for everyday services and treatments has surged, with practices now facing an average 39 weekly requests per physician, further adding to their administrative load and driving burnout.
Many of us who entered medicine to help patients now face the frustrating reality of being pulled away from care to navigate manual, time-consuming processes. But a new chapter is unfolding – one that improves clinical productivity by combining the speed and transformational potential of AI with the judgment and oversight of experienced clinicians. This physician-led shift reimagines how prior authorization should work – and how AI can support care delivery.
Moving beyond rigid automation
The initial promise of electronic prior authorization was streamlined processes and faster decisions. But too often, the technology relied on rigid algorithms that couldn’t reflect the nuances of real clinical practice. Automation alone falls short; physicians know that care rarely follows a predictable path. Care plans are often complex, individualized, and deeply nuanced.
Encouragingly, that’s starting to change. In my own experience working with health plans, I’m seeing a growing push to bring physicians and our expertise directly into the process, building guidelines and tools that reflect both clinical best practices and real-world patient needs. That means using our clinical expertise to help shape how AI-powered solutions are designed from the ground up.
By directly embedding clinicians’ expertise into these tools, we’re able to shape the very policies that guide the automated systems real providers will eventually use. The outcomes are faster approvals, better clinical productivity, and care decisions that align with the latest evidence and clinical guidelines we trust. In this approach, clinicians aren’t just reacting to automated decisions – we are helping shape the technology that drives them.
Training AI with a physician lens
As the adoption of AI in healthcare grows, so too does the scrutiny around it. But one thing has become clear: these tools require transparency and clinical oversight. That’s especially true for intelligent prior authorization. AI should never be used to replace clinicians – it should support them. That’s why clinical intelligence is embedded both at the point of care – boosting overburdened providers’ clinical productivity – and behind the scenes in AI-supported decision-making.
Physicians review and validate the guidelines and model outputs, and ensure algorithms reflect real, evidence-based care. In practice, it means AI can partner with providers to help handle time-consuming administrative tasks, surface key data, and point providers toward the best next steps. It’s a meaningful shift in a process that’s long been ripe for change: clinicians stay in control, while getting the right support to be more efficient and provide the best care.
This kind of clinician-in-the-loop AI that augments automation is already seeing results: faster decision turnaround times, fewer denials, and better provider satisfaction. For physicians, that means we can spend more time on patient care, and less on paperwork.
The rise of physician-guided clinical intelligence and responsible AI
At the heart of this shift is a more collaborative approach, with clinical intelligence tools shaped by those who understand patient care firsthand. Rather than fully automating care decisions, the goal is a collaborative model where technology bolsters clinical productivity. Our expertise defines and refines the guidelines, ensuring decisions are always clinically informed and reflect current best practices.
As a surgeon, the stakes are high. Prior authorizations can delay care, lead to treatment abandonment, and cause unnecessary pain and risks. Prior authorization delays can also cause patients to abandon recommended treatment altogether. But with the right tools, that can all change and when we’re involved in building the systems that support surgeons and our teams, the changes are clear: faster turnarounds, fewer inappropriate denials, and better alignment with medical society standards.
More importantly, these responsible AI models aren’t static. They’re continuously refined by physicians and nurses, ensuring their decision logic holds up in the real world. This means they’re not just faster – they’re more accurate, traceable, and trustworthy. When physicians help develop these tools, automation becomes a partner in care and clinical productivity – not a barrier.
A collaborative model for improved outcomes
We’re now seeing what it looks like when prior authorization becomes a collaborative effort that works with providers to support care delivery. By involving physicians in shaping our AI systems, we can turn utilization management into a benefit for all – patients, providers, and health plans.
Today’s best-performing models are collaborative by design. They codify clinical guidelines, pull the right data from patient records, and deliver real-time approvals when possible. And when they can’t? They escalate the decision to a real clinician for review. This collaborative approach accelerates access to care while respecting clinical judgment.
These AI-powered solutions don’t just speed things up, they help clinicians stay focused on what only they can do – caring for patients. That’s true clinical productivity in action, and the results speak for themselves.
Enhancing provider satisfaction and patient outcomes
When physicians help build the tools we use in practice, prior authorization shifts from a bureaucratic burden to a support system. We bring the clinical context and nuance that algorithms alone can’t see. The more we can be involved in guiding how these tools are built and used, the more time we get back for complex cases that truly need our attention. That’s how we make real gains in both clinical productivity and care quality.
A call to physicians
To unlock the full potential of AI in prior authorization, physicians must stay engaged – not just as users, but as architects. Clinical expertise should remain at the heart of how AI systems are created and refined, with the goal of realizing automation that supports sound medical decisions – a balance critical for the best outcomes and provider experience.
This ongoing partnership – between technology developers, health plans, and physicians – is key to getting it right. By keeping our AI and ML grounded in real-world medicine, we can shape a future where prior authorization works smarter, faster, and better for everyone involved.
Ultimately, this is about restoring the doctor-patient relationship to the heart of healthcare. When prior authorization respects clinical judgment instead of standing in its way, we can reclaim the time and focus we need to focus on what matters most: our patients.
Photo: Thai Noipho, Getty Images
Dr. Paul Johnson currently sits on Cohere Health’s Medical Advisory Board and is an experienced orthopedic spine specialist who has dedicated over 30 years to providing exceptional care to patients in the Knoxville area. He earned his medical degree from Northwestern University and completed his residency in orthopedics at Vanderbilt University. Dr. Johnson then pursued a spine fellowship at Emory University, further developing his expertise in spinal care.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
