As consumers increasingly interact with diagnostic data that originates outside of the clinical laboratory, healthcare systems are faced with the challenge of how to integrate this data in the electronic health record (EHR) and make it useful to clinicians.
Currently, integration of mobile health (mHealth) technology data into the EHR is immature, with few health systems throughout the country having the ability to bring patient-generated health data from remote patient monitoring tools into the medical record, according to Juan Espinoza, MD, chief research informatics officer with Lurie Children’s Hospital of Chicago.
Examples of patient-generated data include that taken from blood glucose monitors, blood pressure readings using home health equipment, or exercise and diet tracking using a mobile app or wearable device.
Typically, patient-generated data or test results that come from remote monitoring tools are printed and scanned into the medical record, noted James Nichols, PhD, DABCC, FADLM, medical director of chemistry and point-of-care testing at Vanderbilt University School of Medicine in Nashville, Tennessee. Patients are primarily responsible for capturing or recording this data, and patients also decide how to share it with healthcare providers.
In some cases, such as continuous glucose monitors (CGMs), a physician will simply look at a patient’s app on their phone. Sometimes a middleware platform, such as Apple HealthKit, is used to aggregate and integrate patient-generated data with an EHR.
“Continuous glucose monitors collect a wealth of data, which can show trends and estimate the rise and fall of glucose over time,” Nichols said. “Each CGM manufacturer has a different application that interacts with the cloud. What they do with that information is proprietary.”
Early integration efforts for remote monitoring devices largely have focused on CGMs, and their use has increased substantially in recent years. “A handful of academic medical centers have built interoperability interfaces for CGMs, but this type of integration is still in its infancy,” said Espinoza, who has led efforts to integrate CGM data into health records. “CGM devices were never designed for EHR integration. So we had to figure it out after the fact.”
Barriers to Integration Abound
In studies, healthcare providers have reported that disruption to clinical workflow and time constraints are top barriers to integrating patient-generated data into EHRs. A December 2020 report published in JAMIA Open concluded that, while work on integrating patient-generated data into EHRs appears to be at an early stage, this data has the potential to close healthcare gaps and support personalized medicine (doi: 10.1093/jamiaopen/ooaa052).
The authors called for more efforts to understand how to optimize data integration, standards for EHR delivery, and clinical workflows.
Some EHR manufacturers, such as EPIC, are working on CGM modules, Espinoza said. However, myriad technical issues must be dealt with. One example is ensuring the correct patient identity, such as in the case of a child whose device is registered to a parent or who might have a different last name than the parent. There are also operational, legal, compliance, and financial problems.
Government Initiatives Offer a Nudge
The federal government is attempting to tackle some of these problems. The 21st Century Cures Act, signed into law in 2016, includes provisions on data sharing and interoperability to encourage the access, exchange, and use of electronic health data. In addition, a proposed rule from the Centers for Medicare and Medicaid Services requires healthcare facilities to implement technologies that support open application programming interfaces (APIs) that allow real-time, bidirectional data exchange between patients and providers.
The Office of the National Coordinator for Health Information Technology published a white paper in 2018 highlighting some of the challenges of collecting and using patient-generated data. These include the technical challenges related to accuracy of measurement, data provenance, and privacy and security concerns.
The report concluded that clinicians and researchers should prioritize health conditions where the use of patient-generated data could have the greatest impact, and payers should expand reimbursement models to cover use of this data to drive positive health outcomes.
Sizing Up the Benefits of CGM Data Integration
The benefits of integrating CGM data into the medical record are many, according to Espinoza. “Any time a clinician has to leave the EHR to go look for data elsewhere, it’s lost time and lost efficiency,” he said. “Because it’s such a burden to leave the EHR to find data, some providers won’t take that extra step. Integration will make it easier to look at the data, which in turn will improve patient care and decrease the likelihood of mistakes occurring.”
Nichols agreed that integration of CGM data into the medical record would help clinicians improve patient care. “It would certainly personalize a patient’s experience,” he said. “Rather than use general reference intervals, doctors could rely on a patient’s own data to make decisions.”
In 2021, Espinoza was instrumental in forming a group to focus on the challenges involved in integrating CGM data into EHRs. Led by the Diabetes Technology Society, the group—which included more than 140 participants from industry, government, and academia—published a report with a set of data standards in November 2022 (www.diabetestechnology.org/icode).
According to the report, there are no common data management systems among CGM manufacturers. The lack of standards and failure to integrate with other healthcare data inside the EHR renders CGM data less useful than it could be. The iCoDE project was tasked with developing technical specifications to integrate CGM data into the EHR and creating workflows and guidelines to facilitate data integration efforts.
Pilot Project Shows Promise
There is some evidence that integrating CGM data into the EHR improves patient care. A 2020 pilot project between HealthPartners Institute’s International Diabetes Center (IDC) in Minneapolis and Abbott Diabetes Care allowed for CGM data to transfer from Abbott’s cloud-based system, LibreView, to an EHR, allowing physicians to view patients’ CGM data along with their laboratory results.
“CGM data provides a wealth of information, but without easy access, clinicians can’t fully leverage this information for their discussions with patients and clinical recommendations,” said IDC Medical Director Amy Criego, MD, at the American Diabetes Association’s 81st Scientific Sessions in 2021. “We demonstrated that there’s an effective way for clinicians to both view and track this data over time in the EHR, which we expect will improve how they’re able to support their patients.”
Beware Data Fatigue
Although glucose monitors clearly provide useful data, there are questions about the value of other types of data, such as heart rate or temperature, from mHealth devices.
“Data by itself is not always useful unless there’s an evidence-based protocol about what to do with it,” said Ji Yeon Kim, MD, physician director, esoteric chemistry & immunology, special coagulation, and lab informatics for Kaiser Permanente Southern California Regional Reference Laboratories. Excess data can lead to data fatigue, which can be counterproductive, she said.
Kim suggested that health leaders think carefully about the types of data they want brought into the medical record and establish a protocol for how the data will be seen and used by clinicians.
Nichols acknowledged that figuring out just what data should be integrated into an EHR is something that clinicians are struggling with. “We are all still learning about what to do with all the data generated by at-home monitoring equipment and wearable sensors,” he said. “It’s all very new.”
Lab Professionals Contribute to the Conversation
Both Espinoza and Nichols believe that there will come a day when integration of patient-generated data into the medical record is commonplace, but they say that time is still a way off.
“We have a long way to go,” Nichols said, who noted that clinical laboratories can play a role in determining what types of data are selected for integration. He advises that laboratorians stay abreast of developments in this area, as well as contribute to the conversation by sitting on relevant work teams and advising clinicians. “Laboratorians are in a great position to do this,” he said.
Kimberly Scott is a freelance writer who lives in Lewes, Delaware. [email protected]