Oakland, Calif.-based Kaiser Permanente uses EHR data to track the quality and safety of medical devices, aiming to provide better, more personalized patient care.
Liz Paxton, PhD, director of medical device surveillance and assessment at Kaiser Permanente, told Becker's the device surveillance program has caught safety issues and helped improve surgical processes within the 40-hospital system — and beyond.
Becker's interviewed Dr. Paxton about how and why her health system's EHR tracks over 4.2 million medical devices.
Question: How does Kaiser Permanente use EHR data to identify high-performing medical devices and improve surgical outcomes and processes?
Liz Paxton: In 2001 we first developed our total hip-and-knee registry, and that focused on patient characteristics, the implant surgical techniques and primarily revision as the outcome of assessment. And we started that project in San Diego and rolled it out to all of our medical centers nationwide. It started off as a paper-based process, and then we transitioned it and integrated a standardized documentation form into the electronic health record.
So we really leveraged existing data within our administrative databases or EHR, and then added a core group of variables that could help, not only with documentation of the procedure in the chart, but also to help with our registry evaluation. So it serves really a dual purpose. Fortunately, we have a module in Epic that captures the barcode of the implants that we're able to extract that information from, along with the core data that we extract from the smart form which is integrated into the EHR, as well as integrate on that with our other data available within the system.
We also focus on looking at specific devices in their performance, and we do that on an annual basis, unless a physician flags a certain device that they're concerned about that we look into immediately.
But for our system, we can capture every point within care that the patient has a visit or procedure, and so that's very useful in terms of being able to use devices. We also have really low attrition so, for our patients, over 20 years only 2% have left the health plan, and that allows us to track those patients longitudinally.
We set up this program as a quality assurance program, so that we wanted to first of all be able to identify higher and lower-performing implants. We wanted to be able to identify clinical best practices. We also wanted to identify patients at risk for certain outcomes so they could be optimized prior to the surgery, and really look at comparative effectiveness research. And so we established this program with the inter-regional chiefs, and we work very closely with them. We find a clinical best practice. We work with the chiefs' group to set a target and a goal and then provide feedback on utilization patterns and outcomes to change that clinical practice and advanced quality of care. And as a result, we've seen numerous examples of enhanced quality of care and changes in clinical practices.
Q: How do you use this technology to personalize care for patients?
LP: One of the areas we have focused on is the development of patient risk factors: calculators that identify, for each individual, patient their personalized risk for complications, readmissions, ED visits. So the physician can enter that patient's personal information and come out with a risk of complication. We also, for our anterior cruciate ligament registry, identified the best graft for the patient based on their demographics, and so that's used as a tool at the point of care for clinical decision-making, really shared decision-making between the patient and the physicians.
We also look at, on a broader level, certain patient populations that are at higher risk, and then provide that information to our clinicians who develop protocols and pathways based on patient characteristics and really focus on optimizing patients for the surgical procedure. For example, let's say they have high BMI. We find that high BMI is associated with an increased risk of infection. The physicians would work with them to let them know about that risk, as well as try to mediate that risk through either some sort of weight loss program or working with the patient directly.
Q: How does this research benefit other patients around the country?
LP: Our registry research impacts not only our patients, but patients nationally and internationally. We share our findings at international conferences. We publish our findings.
We also work directly with the FDA, when we find a problem with an implant that has a performance that's not up to par. So for example, we were working with our physicians. They identified an endovascular graft they thought there was a problem with. So we looked at it, and absolutely there was an issue with a higher complication rate. So we notified the FDA. They used that in their decision-making for advisory and recall.
Q: What has been the biggest challenge in this line of work?
LP: When we first integrated our process into the EHR, it was quite challenging. We had a paper-based process, and we were concerned that integrating it into clinical workflow in the EHR would result in individuals participating less. But fortunately, that wasn't the case. We were able to overcome that by providing feedback to surgeons on their participation rates and getting them to complete forms retroactively if necessary.
Getting the registry off the ground in the beginning, we had to do a lot of socializing the concept and demonstrating the value of the registries and post-market surveillance and immediately being able to identify patients with recalled devices, so the physician has a true benefit to the patient, helping them with their clinical expertise.
Q: Where do you see this technology going next?
LP: First of all, we have transitioned away from the concept of registries, where we're pulling data out, cleaning it, constantly updating it, and following those patients for a lifetime.
We started a new model just recently where we're taking advantage of existing data in the EHR, so we're not asking for additional documentation from the physicians or frontline staff, and we're using that to answer clinical questions and then moving on to the next clinical question, so that we can support multiple specialties and answer lots of questions about devices within our organization and externally as well.
I see that as the basis potentially for a national surveillance system within the United States. I believe there's an opportunity to link data from multiple systems and healthcare programs to evaluate devices and work collaboratively, increasing the sample size for small adverse events, which allows you to check problems with the devices earlier, and really have the opportunity to enhance patient safety for patients nationally and internationally.
Q: Is there anything else you'd like to mention?
LP: It's really critical that we focus on a national implant surveillance system for patient safety and quality care, and it's really important that physicians have ownership of that process, as well as the data, so that they trust the information and act upon it.
So having the infrastructure, as we do within our system, is really critical, with clinical leadership really guiding what we're looking at, as well as taking the data and acting upon it, because that's really the purpose of the work that we do, is providing feedback to change clinical practice and enhance outcomes.