Siloed health records are harming patients. ‘One patient, one record’ is the answer.

Making the Case for a Unified Longitudinal Health Record 

By Marcus Schabacker, MD, PhD, President and CEO, ECRI 

A 58-year-old man sees his primary care doctor for fatigue and unexplained weight loss. His doctor orders bloodwork and refers him to an oncologist at a different health system. The oncologist's office cannot access his prior labs, his PCP's notes, or the CT scan he had six months earlier at an urgent care center for an unrelated issue. Rather than build on what already exists, the oncology team requests records by fax, waits several business days, and when the images do not transfer cleanly, orders a new CT scan rather than delay further. Three weeks pass between the first abnormal bloodwork and the start of any treatment plan, not because the cancer was hard to find, but because no one could see the full picture at once. 

Medical Records Travel Poorly – with Tragic Outcomes 

This is a routine feature of healthcare in America: patients move from one provider to another far more easily than their medical information does. When patients transition across care settings, critical clinical information does not reliably travel with them. The result is one patient with dozens of incomplete, unreconciled records. That is dangerous for patients and inefficient for clinicians and health systems. No amount of clinician training, vigilance, or best practice adoption can fully compensate for an information pipeline that systematically withholds critical patient data at the point of care.  

Evidence Shows the Toll of Fragmentation 

The harm and inefficiencies caused by fragmented medical records and poor interoperability are well-documented. Fragmented care and poor EHR interoperability are sometimes associated with more inappropriate medication use and patient safety events, with lower continuity of care with a treating physician also linked to higher mortality. Patients transferred between hospitals that cannot electronically share records undergo more repeat imaging and testing. Primary care clinicians report that missing clinical information is common and increases the risk of delayed care and unnecessary services. 

There is no available evidence proving a unified medical record architecture saves lives. AHRQ's evidence report in 2015 concluded that no studies reported on outcomes of mortality and morbidity; and studies published in Health Affairs had conflicting findings in 2015 and 2019 and neither reported on mortality benefits. But evidence reinforcing other benefits of interoperability is compelling enough to act upon.  

Better electronic information sharing between providers is associated with fewer hospital readmissions and, among vulnerable older patients, significantly lower in-hospital mortality. A unified health record gives clinicians immediate access to complete patient information, reducing time spent hunting for records, reconciling charts, and re-entering data across disconnected systems.  

Three Possible Paths – Why “One Patient, One Record” Stands Apart 

In the search for solutions, variations of three possible paths often emerge. 

#1 --- The first potential solution is where the U.S. is already headed: towards slow, steady, incremental improvement. This leaves the solution largely in the hands of the manufacturers and EHR vendors, with the government creating a regulatory architecture requiring vendors to create interoperable systems. However, this approach hasn’t worked yet. The proprietary nature of the model will remain an obstacle, where each EHR vendor strives to be the best and only provider.  

If we remain on this path, hospitals and software vendors would keep adopting a shared technical standard called Fast Healthcare Interoperability Resources (FHIR), which lets one system request specific pieces of a patient's chart from another. The federal government would keep enforcing the 21st Century Cures Act rules that make it illegal to deliberately block the sharing of patient data. The Trusted Exchange Framework and Common Agreement (TEFCA) continues to connect more regional record-sharing networks to one another. This is progress, but it falls short of the swift and comprehensive fixes needed.  

#2 --- The second possible solution is more ambitious and calls for government action: health records that follow a patient smoothly across any hospital, clinic, or software vendor. Getting there would take real federal investment in the information pipeline underneath. That means a reliable way to confirm that the John Smith in one hospital's system is the same John Smith in another's, universal use of the same technical standard rather than each vendor's own version of it, and regional networks that assemble a patient's full history over time instead of a snapshot from a single visit. In this scenario, the government creates the regulatory architecture that mandates a single standardized identify for each patient and then requires adoption and compliance. This model moves us towards one patient, one record, but leaves a somewhat proprietary model in place. 

Either path would be an improvement, but neither solves the whole problem. Even in the more ambitious scenario, someone still must reconcile conflicting records, deciding which medication list is current when two hospitals disagree. If there are gaps and errors in the underlying data, those remain and just become faster to transmit. And the duplicated tests, repeated imaging, and staff hours spent chasing down records that already exist somewhere else would continue. 

#3 --- The "One Patient, One Record” framework would feature a unified, longitudinal digital health record that aggregates a patient’s complete medical history across all care settings. The record would act as a single source of truth; a chronological timeline including comprehensive medication and allergy lists, historical lab results and diagnostic imaging, real-time care plans and surgical histories, plus demographics and social determinants of health. 

The bedrock of a unified record is the universal adoption of the Fast Healthcare Interoperability Resources (FHIR) mentioned earlier. This does not necessarily require a single global database. Instead, it relies on a federated, highly interoperable network governed by strict data standards.  

Lessons Learned – Internationally and in America 

The “one patient, one record” model is a demonstrated operational reality. Estonia, Denmark, and Canada have all implemented national or system-level health data sharing architectures, confirming the model is technically and operationally feasible. Plus, the U.S. government has already built and operated a unified longitudinal health record at national scale: at the Veteran’s Health Administration (VHA).

• Estonia’s system demonstrates that a nationwide, seamless, interoperable health record is achievable and can give clinicians access to a patient’s complete health information across care settings, sharing through a secure, federated system with mandatory participation. 
• Denmark shows us the safety potential of interoperability, giving prescribers real-time access to complete medication histories across care settings, but that technology must be paired with strong governance and consistent record maintenance. 
• Canada’s EHR demonstrates significant cost savings and fewer unnecessary healthcare visits, but voluntary adoption is not enough; national standards, sustained funding, mandates, and enforcement are needed to achieve widespread interoperability.
• America’s VHA record system led to measurable quality improvement and efficiencies. A rocky rollout, however, did lead to missing clinical orders and adverse events. Nationwide integration is achievable, but only with oversight, testing, safety safeguards, and phased implementation.  

Common Rebuttals 

“One patient, one record” has been met with a few consistent rebuttals over the years – namely data privacy, cybersecurity, and vendor resistance.  

• Health privacy in the US is already failing. Data that’s not HIPAA protected, from symptom web searches, period trackers, wearables, and mental health apps, moves through a legal market in which brokers have been documented selling lists of named individuals identified as having depression, bipolar disorder, PTSD, or a history of sexual assault, with minimal screening of who is buying. Patients today have neither privacy nor control. 
• Centralizing data does heighten the cybersecurity risk. ECRI has long called for cybersecurity safeguards including specific safety strategies that will become even more critical in the future. To address cybersecurity, use current cybersecurity defense technology, employ a zero-trust architecture, end-to-end encryption, and multi-factor authentication to safeguard records against breaches.
• Some have voiced concern that a single health record could be misused as a national surveillance system. Patients with immigration concerns, substance use histories, HIV, or mental health diagnoses have rational reasons to fear a complete, permanent, centrally reachable record. The system would need enforceable legal limits on who can reach into the records; statutory protections must be a prerequisite.
• Commoditizing a unified record is not in the commercial interest of the EHR vendors. Realistically, we must stop expecting voluntary adoption. This requires regulation, payment conditions, or both. Vendors move when federal conditions change. We should use government policy and financial incentives to penalize information blocking and reward open data sharing. 

Call to Action 

A unified patient health record would benefit every corner of American healthcare: giving clinicians the information they need to make better decisions with less administrative burden; helping health systems reduce readmissions, duplicate work, and wasted resources; and giving payers and policymakers new opportunities to lower costs. Interoperability coalitions and data-sharing initiatives have made important progress and should continue, but incremental improvement is not enough. It is time to commit to something more ambitious: national policy, championed by federal healthcare leaders, that makes a unified, longitudinal patient record a fundamental expectation of American healthcare.