mCODE: Learning from Cancer Patients

L.A treasure trove of data locked behind the firewalls of proprietary systems could help us diagnose heart disease, diabetes, cancer, and other conditions more quickly and accurately, and treat them better. However, electronic health record infrastructure is largely underutilized because it was not designed to allow organizations to easily share data.

Electronic health records were first developed in the 1960s But they didn’t go mainstream until about 12 years ago when the federal government offered incentives for their use. At the time, expectations were high that it would be a solution for collecting and sharing valuable patient data seamlessly and securely. to multiple databases manually.

No country yet. The rush to develop electronic medical records has spawned unique and competing data systems customized for hospitals, clinics, and other medical institutions. Additionally, it took years to develop a software standard that would allow data sharing between systems.Thanks to organizations such as Health Level Seven International and the Office of the National Coordinator for Health Information Technologythere has been some progress.

advertisement

The Covid disruption is a result of the lack of standards and the ease with which electronic health records, such as making shareable patient data readily available so doctors can assess which treatments have worked for which patients, have not. I highlighted what I can’t do.

At the height of the pandemic, as Mayo Clinic doctors fought to save the lives of their patients, medical staff often had to pause to fill out long hours. REDCap survey To inform Minnesota health officials of Covid-19 cases. Because the state-mandated investigation is outside of Mayo’s regular EHR system, recording Covid case information into a giant Excel spreadsheet required painstaking manual work. “When we were booming, we were drowning and weighed down,” Priya Sampaskumar, an infectious disease and critical care specialist at the University of Mayo, told my team. Miter, the non-profit research and development organization I work for. System A was unable to communicate with System B.

advertisement

Given how electronic medical records are currently structured, there are different medical Sharing and analyzing high-quality data from millions of patients in a system is difficult, if not impossible. .

For example, most of the data leading to new cancer treatments today comes from clinical trials. This is a problem. Because he makes up less than 6% of adult American cancer patients participating in these trials. The proportion of children is even smaller. In other words, there is limited information about which treatments work for which patients. Clinicians and researchers do not have ready access to data from the majority of cancer patients. For example, she is a 52-year-old Hispanic woman with diabetes and breast cancer, and a 70-year-old man with stage 3 lung cancer.

Fortunately, the quest to extend standards and interoperability across electronic health record systems to improve the quality, safety, and effectiveness of patient care did not start from scratch. Fast Healthcare Interoperability Resources (FHIR) standard By defining how health information is exchanged between different computer networks, it facilitates the sharing of data regardless of how those systems store it.

2019, Miter and several other nonprofits are working to develop a common cancer care standard and language that can be incorporated into an EHR and used to capture the characteristics, treatments, and outcomes of all cancer patients. It was started. We built the standard on existing best practices, including his HL7 experience developing the Fast Healthcare Interoperability Resources standard.

result, mCODE (short for minimal Common Oncology Data Elements) are currently being tested by over 60 healthcare organizations and other stakeholders (including EHR vendors), and see the potential to learn from the experiences of millions of patients. increase. Cancer was chosen to test the hypothesis that only minimal and significant information is required to obtain valuable results comparable to those found in clinical trial reports.

We also learned the need to involve the community in building and driving consensus around new standards. mCODE comprises subject matter experts, including cancer clinicians, informatics, health services researchers, data standards and interoperability experts, and cancer patients, under the auspices of MITER and the American Society of Clinical Oncology Developed by an interdisciplinary group.

mCODE aims to standardize health records rather than focus on data exchange standards. This will enable diverse stakeholders to share information in meaningful ways to achieve outcomes at scale, such as more efficient research, faster trial matching and more personalized care. . Using mCODE’s common data language and open-source, non-proprietary model, organizations can access and analyze data from various EHR systems. This includes important data that is difficult to find today, such as the stage of a patient’s cancer or the outcome of a particular treatment.

In its first pilot project, the mCODE team worked with a clinical trial group testing new uses for existing drugs to treat breast cancer. Initial results of preliminary data, which have not yet been published, indicate that the accuracy of the mCODE results match that of the clinical trial team 95% of the time. Since then, mCODE has been incorporated into several other clinical trials and is being tested for other uses, including cancer registries and pre-approval of treatments. The team also freely shares its expertise and open source technology with organizations involved in cardiology, genomics, and dementia that wish to use the mCODE approach to develop and test standards for their specialty. I’m here.

A standards-based approach like mCODE puts valuable information about a patient’s disease and possible treatments at the point of care, readily available to all physicians. This insight has the potential to improve patient care and shared decision-making, drive innovation, and lay the foundation for a national cancer health learning system.

The possibilities for improving patient care and research through shareable data are endless. But to make that happen, we need the entire community: electronic medical record vendors, healthcare systems, payers, researchers, patients, and more. It will also require changes in incentives. Many of today’s players, including EHR vendors and healthcare systems, are making patient data their own, believing it will give them a competitive advantage. That is no longer the case because no single organization can have enough data to solve big problems.

Physicians, scientists, and patients desperate for effective treatments recognize the clear benefits of having access to large-scale data and will drive this approach. Unlocking the potential of these proprietary systems couldn’t be more important.

Jay J. Schnitzer is a pediatric surgeon and senior vice president, chief medical officer and chief technology officer at MITER.

---

First Opinion Newsletter: If you love reading opinion and perspective essays, get our weekly First Opinion Roundup delivered to your inbox every Sunday. SIGN UP HERE.