The Food and Drug Administration (FDA) recently requested $100 million in its FY 2019 budget “to advance the use of real-world experience to inform patient care and provide efficient and potentially lower cost ways to develop clinical data to expedite medical product development.”
With this proposed budget, the FDA plans to build a system that includes data from electronic health records (EHRs) to help better evaluate medical products, like devices, with real-world data.
This move signals the need for healthcare stakeholders to connect the dots between patients’ disjointed data. But would this potential system from the FDA really solve the issue at hand? How effective would it be at comparing measures across devices?
The Limitations of EHR Data
What the FDA intends to do with this proposed budget is build a data warehouse and analytics platform that would allow for reporting on safety and outcomes of medical products. And while expanding beyond just claims data to collect data from the EHR is certainly a step in the right direction, there are limitations around using EHR data for the effective measurement and evaluation of medical devices.
The first challenge the FDA will face is ensuring data interoperability, particularly when looking to collect and merge data from entirely different EHR systems where the data model is the vendor’s secret sauce. For example, a diabetic reading of an HbA1c test in one system may not be equivalent to an HbA1c reading in another system. We need to make sure the units are equivalent, such as percentage, mg/dL, or mmol/L, and that any circumstances that may change the reading – like chronic bleeding, anemia, if the person has an uncommon form of hemoglobin type, or perhaps has had a recent blood transfusion – are recorded in a universal manner. It is also likely that these important factors are not available in the existing dataset. As the FDA builds its new data warehouse, this lack of data interoperability is likely something they’ll have to work around.
More importantly, the types of device data available in the EHR are very limited – specifically, it won’t provide meaningful context around how medical devices are used. Today, users of medical devices frequently receive large degrees of variability in advice, training and outcomes. The EHR will contain data around what happened to a patient during a physician visit – capturing their labs, orders and vitals, demographics, and general health, for instance. But in order to really measure the impact of a medical device, we need to monitor device data and the context of its use (such as whether the device is used with the assistance of a clinician or self-administered) along with outcomes and how quickly they are achieved. This data isn’t currently available in the EHR.
Getting that data means they will need to track device and patient outcome data, applying measurement across the entire duration of the patient journey as the device is being used.
Capturing Meaningful Data
One way to begin capturing meaningful data that sits outside of the EHR is through the use of wearables and fitness trackers. These devices generate data that measure a patient’s condition and changes in their health. Capturing this data over long periods of time – and supplementing EHR data with it – can bring the FDA one step closer to incorporating contextual data into this new system.
The best way to capture meaningful data, however, would be by implementing device care pathways which help gather device and usage data throughout time, whilst also providing patient context. Care pathways (or clinical pathways) are longitudinal workflows that can be applied to medical devices to gather patient and device data and operationalize activity around the device. Pathways ensure the device is being used optimally along patient journey – from the first appointment through to rapid recovery. An airplane needs a flight plan, ensuring it arrives optimally from point A to B, with least cost fuel and avoiding adverse weather conditions on-route. In a similar way, care pathways assist in ensuring the patient and care team use the device optimally throughout the care journey – information that is critically important to achieving better outcomes, more rapidly.
Further, comparing medical devices isn’t as simple as comparing apples to apples – we really need to understand who is eating the apple, how they are eating it, and what their eating patterns are. To take the diabetes example and truly evaluate medical devices, we would need to understand simple usage concepts, like whether the patient has received diabetic device training and understands how to clean the prick area to avoid contamination, and other issues that may lead to inconsistent readings, such as frailty and if the person needs a care giver. This richer context around a medical device’s longitudinal use is crucial to truly understanding the device.
Fast Forward
The FDA’s move to implement EHR data into a system that seeks to evaluate medical products is a giant step towards improved safety and reduced costs for medical devices – but, whether or not it’s successful all comes down to the data. The sooner we can find ways to measure and incorporate more meaningful data, beyond the EHR, the more impactful this new system will be in assessing medical devices in real-world settings.
By Rick Halton, Vice President, Marketing & Product, Lumeon