Once upon a time, clinicians didn’t have enough data about their patients.
To put it mildly, things have changed. Today we’re drowning in data. As a practicing emergency physician, it is a reality I see every day.
In an emergency department (ED) environment characterized by constant time pressure, staffing constraints, and high patient acuity, clinicians sometimes must wade through reams of electronic health records (EHR) data to find critical patient information that has become buried among endless details.
Insights gained in the ED are highly relevant to primary care, where analogous data challenges evolve over longer periods and across multiple touchpoints. Primary care physicians today are experiencing their own challenges with data overload.
For example, PCPs spend roughly between one-third and one-half of every patient encounter on EHR work, rather than directly interacting with patients, a 2025 Journal of General Internal Medicine study found. This imbalance reduces the time available for meaningful clinical conversations and contributes to clinician burnout.
Adding to the challenges, the percentage of Medicare beneficiaries who visited five or more providers per year increased from 17.5 percent in 2000 to 30.1 percent in 2019, according to a study in the Annals of Internal Medicine. With every additional provider a patient sees, the likelihood of fragmented data, duplicative tests and procedures, and delayed follow-up increases.
However, emerging AI tools offer physicians some hope of relief from data overload. By helping doctors organize and interpret the large volumes of information surrounding each patient encounter, these tools support clinical judgment. The result is greater focus on the most relevant details within the workflows that practices already rely on.
Deeper insights in less time
The practice of medicine demands efficiency, and no area more so than emergency medicine. ED clinicians must quickly synthesize and interpret information, identify risks, and take decisive action. In my more than 15 years of practicing emergency medicine, I’ve witnessed how better data presentation improves clinicians’ efficiency.
For example, AI can summarize and structure patient data in ways that support clinical reasoning, reducing complexity for doctors. By advancing small gains in clarity and speed, AI tools create opportunities for improvement.
In my ED experience, AI-supported summaries have accelerated the speed at which care teams can assess patients admitted from skilled nursing facilities. In moments, we can assess changes in vital signs, treatment updates, and care plans, enabling us to reduce duplicative testing, speed up evaluation, and drive quicker, more-informed decisions about whether patients can return to their prior settings.
Additionally, AI-generated predictive insights support better clinical decision-making. With risk-scoring models, physicians can identify patients most likely to deteriorate following acute events. As a result, these models drive earlier interventions and stronger coordination between acute, post-acute, and primary care providers, helping clinicians reduce preventable readmissions.
However, it is essential that AI does not function as a black box. Clinicians should understand how results are produced, be able to validate source documentation, and maintain full control over care decisions. Transparency and validation foster confidence and support safe deployment across varied patient populations.
Better visibility, better decisions
EDs often function as a crossroads for acute, post-acute, and community-based care. Limited visibility across these settings drives avoidable readmissions and repeat emergency visits.
Similarly, PCPs need visibility as patients move across different care settings, including data on discharge timing, medication changes, recent interventions, and urgent concerns. With timely data, PCPs can identify potential issues and intervene before conditions grow worse.
AI-based tools support these needs by integrating and standardizing data from hospitals, post-acute facilities, and home-based providers, helping PCPs gain a full 360-degree view of patient health.
When practices consider adopting these technologies to support future care delivery, they should seek solutions that ease administrative demands, fit within existing workflows, and facilitate cross-setting information exchange. Responsible AI principles, transparency, and clear physician oversight of all clinical decisions should remain central requirements.
Conclusion
As data volumes continue to grow across health care, AI offers a practical way to help physicians surface what matters most without adding complexity to already demanding workflows. When thoughtfully designed, these tools can reduce administrative burden, improve clinical clarity, and strengthen coordination across care settings. Ultimately, AI’s value lies in supporting clinical judgment, enabling physicians to deliver more informed, efficient, and patient-centered care.
Hamad Husainy is an emergency physician and physician executive.
