By Dr. Roger Seheult, MedCram Contributor
Disclosure: I have no financial relationship with Doximity or any other AI platform mentioned in this article.
As healthcare providers, we’re all too familiar with the relentless demands of clinical documentation. Between patient rounds, consults, and urgent requests, finding time to complete progress notes, histories, physicals, and consult reports can feel like an insurmountable task. The administrative burden often spills into late hours, delaying note availability for care teams and cutting into time for direct patient care. But what if there was a way to reclaim that time while maintaining the highest standards of patient privacy? Enter HIPAA-compliant artificial intelligence (AI), a game-changer for providers. In this post, I’ll share my personal experience with Doximity GPT, a HIPAA-compliant AI tool, and explore how it—and other compliant AI solutions—can transform your workflow, with practical prompts to get you started.
My Struggle with Documentation Overload
Picture this: I’m on morning rounds, juggling updates from nurses, respiratory therapists, and pharmacists while typing furiously into the electronic health record (EHR). I’d capture the raw data—vital signs, lab results, medication changes—but the critical assessment and plan for each patient’s progress note? That had to wait. By the time rounds ended, new consults and patient requests would pile up, pushing note completion to the end of my shift. I’d spend an hour or more after hours crafting those notes, exhausted, while other team members went home without access to my updates. It was inefficient, frustrating, and worst of all, it delayed patient care.
Then I discovered Doximity GPT, a HIPAA-compliant AI engine designed for clinicians. This tool has revolutionized my workflow, and I’m excited to share how it’s made me a more efficient and effective provider.
Doximity GPT: A HIPAA-Compliant Lifesaver
Doximity GPT leverages advanced AI models to generate comprehensive clinical documentation—full histories and physicals, progress notes, consults, and even predictions of likely in-hospital complications—while adhering to HIPAA privacy standards. With the right prompts, it produces tailored outputs that integrate seamlessly into your EHR, saving time and mental energy. For me, the biggest game-changer is real-time note completion. During rounds, I input updates from the care team directly into Doximity GPT, use a prompt to generate an updated assessment and plan, and review the output in seconds. Once I’ve verified its accuracy, I sign the note, and it’s done—before I move to the next patient.
This has slashed my documentation time dramatically. Instead of staying late to write notes, I’m done by the end of rounds. My notes are immediately available to the care team, improving communication and coordination. Most importantly, I have more time to focus on patients—seeing consults promptly, addressing concerns, and avoiding delays that leave patients waiting. It’s not just about efficiency; it’s about being a better doctor.
A Word of Caution: Review Is Essential
Let’s be clear: Doximity GPT is not a magic bullet. The AI-generated notes are impressive, but they’re not perfect. Human oversight is critical. You must review every note for accuracy, completeness, and clinical appropriateness before signing. The beauty, though, is that editing a pre-generated note is far easier than writing one from scratch. Think of it as a highly skilled scribe who gets you 90% of the way there, leaving you to fine-tune the details. This balance of automation and human judgment is what makes HIPAA-compliant AI so powerful.
Why HIPAA Compliance Matters
In healthcare, patient privacy is non-negotiable. The Health Insurance Portability and Accountability Act (HIPAA) sets strict standards for protecting protected health information (PHI). Standard AI tools like ChatGPT are not HIPAA-compliant, as their developers (e.g., OpenAI) typically don’t sign Business Associate Agreements (BAAs) and may use input data to train models, risking PHI exposure. Using such tools for clinical tasks could lead to privacy breaches and legal consequences.
HIPAA-compliant AI, like Doximity GPT, is designed with privacy first. It operates in secure environments (e.g., AWS GovCloud for some platforms), employs robust encryption, and ensures zero data retention for sensitive inputs. Doximity signs BAAs, guaranteeing compliance with HIPAA’s Privacy, Security, and Breach Notification Rules. This allows you to use AI confidently, knowing your patients’ data is protected.
Sample Prompts for Doximity GPT
To help you hit the ground running, here are some prompts I use with Doximity GPT. These are tailored to generate specific outputs but can be modified for your specialty or workflow. Always ensure inputs comply with HIPAA by using de-identified data or secure platforms.
-
-
History and Physical (ICU)
Link:
-
This link will take the information that you give it afterwards and sort it into the traditional structure of an H+P but the assessment will be in a ICU systems style. It will also give you the ICD-10 codes and give you the estimated length of stay based on the patient that you describe.
-
-
History and Physical (Floor)
Link:
-
This link will take the information that you give it afterwards and sort it into the traditional structure of an H+P. It will also give you the ICD-10 codes and give you the estimated length of stay based on the patient that you describe.
-
Generate today’s Assessment and Plan based on yesterdays and todays updates
-
This link will tell AI that you want to first copy and paste yesterday’s A/P then after hitting enter, copy and paste in or type today’s updates. A new updated A/P will come out that you can paste into todays note.
Other HIPAA-Compliant AI Options
Not everyone has access to Doximity GPT, but several other HIPAA-compliant AI tools can streamline clinical workflows while safeguarding PHI. Here are a few worth exploring:
-
Hathr AI: Powered by a HIPAA-compliant version of Claude AI, Hathr AI automates tasks like summarizing clinical notes, generating insurance pre-authorizations, and suggesting billing codes. It operates in FedRAMP-certified environments and offers a flexible API for EHR integration. Pricing starts at $45/month, making it accessible for smaller practices.
-
Hippocratic AI: Specializes in AI-powered chatbots for patient engagement and triage. It ensures HIPAA compliance through secure data handling and BAAs, ideal for practices looking to automate non-clinical tasks like appointment scheduling.
-
eClinicalWorks: Integrates AI into its EHR platform to summarize records, identify high-risk patients, and match documents automatically. Its comprehensive approach suits larger healthcare systems.
-
Nuance Dragon Ambient eXperience (Dax Copilot): A voice-activated AI tool that generates clinical documentation from patient conversations. It integrates with Epic and uses Microsoft Azure OpenAI with HIPAA-compliant guardrails.
Each tool has unique strengths, so research which aligns best with your practice’s needs. Always verify that the vendor signs a BAA and adheres to HIPAA’s security standards before implementation.
The Future of AI in Healthcare
HIPAA-compliant AI is more than a time-saver—it’s a paradigm shift. By automating repetitive tasks, tools like Doximity GPT free us to focus on what matters most: patient care. My experience is proof: I’m no longer tethered to my computer after hours, and my patients benefit from faster, more coordinated care. But this is just the beginning. As AI evolves, we can expect even smarter tools that predict outcomes, personalize treatments, and reduce clinician burnout, all while prioritizing patient privacy. To be clear, most of what I am describing here doesn’t dictate the management from AI. It’s simply helping out in terms of making a document but things are likely going to change in the future.
If you’re ready to explore HIPAA-compliant AI, start with Doximity GPT’s free access for U.S. clinicians and medical students. Experiment with the prompts above, review the outputs diligently, and see how it transforms your workflow. For those without Doximity access, consider Hathr AI, Hippocratic AI, or others mentioned. The key is to choose a solution that’s secure, compliant, and tailored to your needs.
At MedCram, we’re committed to empowering providers with practical, evidence-based tools to enhance patient care and medicine explained clearly! HIPAA-compliant AI is a powerful ally in that mission. Share your experiences with AI in the comments below, or join our community to discuss how technology is shaping the future of medicine. Let’s embrace innovation responsibly and keep patient care at the heart of everything we do.
Disclaimer: Always consult your institution’s policies and legal counsel before adopting new technology to ensure compliance with HIPAA and other regulations.
References:
-
Doximity GPT: Your Free, HIPAA-Compliant Workflow Assistant.
-
HIPAA Compliant LLM for Healthcare – Hathr AI.
-
10 Top AI Tools in Healthcare for 2025 | Informa TechTarget.
-
13 AI Healthcare Companies to Watch in 2025 | Informa TechTarget.
-
Is ChatGPT HIPAA Compliant? Updated for 2025.
Stay tuned to MedCram for more insights on leveraging technology to elevate clinical practice.