PARTNERED WITH
INEFFICIENCY INSIGHTS
AI-Powered Discharge: Transforming Patient & PCP Handoffs
In August 2024, I wrote about a critical gap in patient care: the disconnect between the detailed discharge summaries we write for other providers and the generic, jargon-filled papers patients receive. At the time, I proposed generative AI as a theoretical solution to transform complex medical documentation into patient-friendly formats.
Fourteen months later, I'm writing to tell you that the solution works—and it's better than I imagined.
Since that article, I've integrated AI tools, particularly DoximityGPT, into my daily discharge workflow. The transformation has been remarkable, not just in theory but in practice. I'm now routinely converting physician-directed discharge summaries into two critical documents:
Patient-friendly hospital course summaries
Warm, succinct handoff letters for primary care physicians
But here's what I've learned that I couldn't have predicted from the outside: the problem isn't just about making discharge summaries readable. The problem is that we've been creating two fundamentally inadequate documents—one that's too complex for patients and another that's too sparse for effective transitions of care.
Consider what actually happens during a typical discharge. I write a comprehensive discharge summary filled with medical terminology, clinical reasoning, and detailed treatment plans. This document is designed for the next healthcare provider—the PCP, the specialist, or the facility accepting the patient. Meanwhile, the patient receives auto-generated discharge papers that list medications, appointments, and generic instructions but provide virtually no narrative about what actually happened during their hospital stay.
What about the PCP?
They often receive nothing until the discharge summary eventually appears in their inbox. And when it arrives, it's a dense, clinical document that requires significant time to parse. In the dangerous window of early post-discharge, the PCP is operating blind, waiting for the patient to show up or the hospital records to arrive.
The result is a three-way communication failure:
The patient leaves confused about their hospital course and what comes next
The PCP lacks timely, actionable information to ensure safe transitions
The system suffers from preventable readmissions and adverse events during the highest-risk period—those first few days after discharge
Transitions of care remain one of the most dangerous periods in healthcare. Studies consistently show that communication breakdowns during handoffs contribute to medical errors, readmissions, and patient harm. Yet we continue to treat discharge communication as an afterthought—a box to check rather than a critical intervention.
The tools to solve this problem now exist. The question is whether we'll use them.
Insights
Root Cause Analysis: 5 Whys
The 5 Whys process in root cause analysis involves repeatedly asking "Why?" five times to drill down into the root cause of a problem by exploring the cause-and-effect relationships underlying the issue.
The problem: Patients leave the hospital without adequate understanding of their care, and PCPs lack timely, actionable information for safe transitions—despite having the clinical documentation needed to create both resources.
Why?: We produce only one primary document (the discharge summary) optimized for provider-to-provider communication, not patient education or PCP handoff efficiency.
Why?: The discharge workflow treats documentation as a single-purpose deliverable rather than source content that could generate multiple, audience-specific outputs.
Why?: Traditional documentation systems lack the capability to efficiently transform clinical content into different formats for different audiences—requiring physicians to manually create multiple documents if they want to serve multiple needs.
Why?: Healthcare technology has historically focused on structured data capture and regulatory compliance rather than communication optimization and information accessibility.
Why (root cause)?: Healthcare workflows evolved in an era when document transformation required manual rewriting—a time burden no one could justify—so we accepted that each stakeholder would receive inadequate information rather than burden physicians with creating multiple versions. We've simply never had the tools to solve this efficiently until now.
Impact Analysis
Impact analysis is the assessment of the potential consequences and effects that changes in one part of a system may have on other parts of the system or the whole.
Patient: Walks out the hospital doors with auto-generated discharge papers that list medications and appointments but provide no coherent narrative of what happened to them. They can't explain their hospital course to family members, don't understand why certain medications were started or stopped, and lack context for warning signs that should prompt them to seek care. This information gap breeds anxiety, reduces adherence to treatment plans, and leaves them unprepared to be active participants in their recovery. When complications arise, they often can't provide their PCP with adequate information about their hospital stay, delaying appropriate intervention.
Clinician or Provider: Discharging physicians write detailed summaries knowing that patients won't understand them and PCPs won't receive them (or can’t review them) promptly. PCPs operate in a dangerous information vacuum during the critical post-discharge window—they may receive a call from a confused patient before they receive any hospital documentation. When the discharge summary finally arrives, it requires significant cognitive effort to extract the essential handoff information buried in clinical detail. Meanwhile, both parties know the current system is inadequate but lack efficient tools to create additional documentation. The result is professional dissatisfaction, medico-legal risk, and the persistent feeling that we're failing our patients at a critical transition point.
System: Poor discharge communication drives preventable readmissions, emergency department visits, and adverse events during the highest-risk period of care transitions. The 30-day readmission rate remains stubbornly high partly because patients and PCPs lack the information needed for safe transitions. Every readmission represents not just a clinical failure but a massive cost to the system—Medicare penalizes hospitals for excess readmissions, and the total cost of a readmission often exceeds the original admission. Beyond direct costs, inadequate discharge communication erodes patient trust in the healthcare system and perpetuates health inequities, as patients with limited health literacy or language barriers suffer disproportionately. We've accepted this dysfunction for so long that it feels like an immutable feature of hospital care rather than a solvable problem.
Solution
In my August 2024 article, I cited the Zaretsky et al. (2024) study, which demonstrated that large language models can transform discharge summaries to a sixth-grade reading level. At the time, this was promising research. Today, it's my daily workflow.
Here's what I actually do…
For Patients: The Plain-Language Hospital Course
After completing my discharge summary, I copy it into DoximityGPT with a simple prompt:
Convert this discharge summary into a patient-friendly explanation of what happened during their hospital stay. Use plain language, avoid medical jargon, and organize it as a clear narrative that helps them understand their care.
The transformation takes 30 seconds. What emerges is a document that:
Explains their diagnosis in terms they can understand
Describes what we found and why we did each test
Clarifies the treatment rationale in plain language
Outlines what to watch for at home
Makes the follow-up plan concrete and actionable
I print this alongside their standard discharge papers. The response has been remarkable. Patients actually read it. Family members reference it when asking questions. And critically, patients can now explain their hospital course to others—their PCP, their family, their home health nurse.
For PCPs: The Warm Handoff Letter
The second transformation addresses an equally critical gap: the PCP handoff. While my discharge summary will eventually reach the PCP through official channels, I now generate a concise, collegial handoff letter that I can send immediately—often the same day as discharge.
The prompt is straightforward:
Transform this discharge summary into a warm, concise handoff letter to the patient's primary care physician. Focus on key clinical points, follow-up needs, and outstanding issues that require PCP attention. Keep it to one page and write in a collegial, physician-to-physician tone.
The result is a document I would want to receive if I were the PCP:
Brief clinical summary hitting the essential points
Clear follow-up tasks and timeline
Outstanding test results or pending issues
Medication changes with rationale
Red flags to watch for
Warm, personal tone that builds collegial relationships
I send this via secure message or fax the same day. The PCP now has actionable information during the highest-risk window, without waiting for the full discharge summary to route through the system.
Why This Works
Three factors make this solution sustainable:
Time efficiency: Both transformations take under a minute total. I'm not writing new content—I'm transforming existing documentation into audience-appropriate formats. The time investment is negligible compared to the communication value.
Quality preservation: The AI uses my clinical content directly, maintaining accuracy while optimizing presentation. I review each output quickly, but the base content is reliable because it's derived from my own documentation.
Multiplicative impact: One discharge summary now serves three distinct purposes: regulatory documentation, patient education, and PCP handoff. The same clinical work supports multiple communication needs without additional cognitive burden.
What's Still Missing
This is a workaround, not a solution. I'm using a third-party AI tool outside my EHR because health systems haven't integrated this capability into their workflows. Every transformation requires copy-paste operations and manual distribution.
The real solution requires:
EHR integration: Automated generation of patient-friendly summaries and PCP handoff letters directly from discharge summary documentation
Multi-format output: Systems that recognize discharge summaries as source content capable of generating multiple audience-specific outputs
Seamless distribution: Automatic delivery of patient summaries (printed and digital) and immediate PCP notification through existing secure channels
Language accessibility: Automated translation of patient summaries into preferred languages
Accessibility features: Audio versions for patients with visual impairments or limited literacy
The technology exists. The clinical benefit is proven. The time burden is minimal. What we need now is health system leadership willing to integrate these capabilities into standard workflows.
THE MIDDLEMEN SERIES
Healthcare Clearinghouses: The Hidden Middlemen Powering Medical Billing
Healthcare clearinghouses are the digital middlemen of the revenue cycle, sitting between providers and payers to standardize, scrub, and route billions of insurance claims. They make the system work—but also add cost and complexity, reflecting the fragmentation of U.S. healthcare.
👉 Read more details here.
HUDDLE #TRENDS
Reimagining Care Utilization in the Age of Algorithms
We spend nearly $5 trillion on healthcare every year—but up to 15% of that is wasted on administrative bloat.
Think: prior auths, claims, referral faxes, and inboxes full of chart reviews. Not care. Just the stuff around it.
AI may finally be the tool to automate that waste away.
But which AI tools are actually moving the needle? What’s hype vs. reality? And how close are we to fully automated workflows like dynamic scheduling, ambient documentation, and claims adjudication?
👉 Full breakdown in my latest Huddle+ article here.
HUDDLE UNIVERSITY
Healthcare 101 Course
We were trained to take care of patients. But no one ever taught us how the system actually works:
How hospitals get paid.
Why your patient’s prior auth was denied.
What Medicare Advantage is doing to care delivery.
And why everything feels harder than it should.
That’s why I built this course: How Healthcare Really Works.
It’s short, visual, and built for people like us — physicians trying to make sense of the business and policy forces shaping our day-to-day work.
