NeuroVoices: Andrea Synowiec, DO, on Using a Large-Scale Data Registry to Capture Patient-Reported Outcomes

Andrea Synowiec, DO

As technology has improved, the ability to collect wide-scale, multilevel data and incorporate it back into practice has become more available. For Allegheny Health Network, a healthcare system that includes 11 hospitals, data sharing is critical. With the help of their clinician staff, Allegheny’s Headache Center created a headache registry that not only captures data on treatment responses, but also monitors more patient-centric quality of life outcomes.

Navigating the migraine treatment landscape can be difficult, in part because the selection is ever-expanding. Calcitonin-gene related (CGRP) inhibitors, a relatively new medication class introduced in 2018, differ from other therapeutics such as beta-blockers or anticonvulsants in that they were developed solely for preventing chronic and episodic migraine headaches. Most of the medications in this new class, except for eptinezumab (Vyepti; Lundbeck) and rimegepant (Nurtec ODT; Biohaven), are administered subcutaneously. Although similar in mechanism, each of these medications have differing safety profiles and adverse effects.

Andrea Synowiec, DO, assistant director of the Allegheny Headache Center, believes that the holistic, patient-centric data the registry captures will help clinicians make more accurate treatment decisions. In a new iteration of NeuroVoices, Synowiec discussed the origins of the registry, reasons for the inclusion of certain outcomes, and where the focus is from both a patient and clinician perspective.

NeurologyLive®: How was the headache registry created? Why did you specifically design it the way you did?

Andrea Synowiec, DO: We are excited to talk about our registry, which is still fairly new. We began with the idea in the beginning of 2019, not knowing what the year coming would hold. We started having discussions about how we might look better at our patients and decide what our outcomes are like if we’re providing good care. It took us an entire year, coupled with a few months of turmoil from the pandemic, to have a functioning registry. The idea behind what we’re doing here is to see if what we’re doing is helpful in an area where the amount of technology, devices we have available, the medications we have to treat patients, are rapidly coming into the market. They’re coming into a place where we can give them to patients and yet, using these medications and devices on certain types of patients, or even in combination, is something that hasn’t been published information.

Despite that, we have a wealth of information when we look at the people we’re treating. The idea arose, what if we look at our own group, see how we’re practicing, and then see if these patients are getting better, or if certain combinations of therapies might work well. We decided to create the registry within the electronic health record system (EHR), which for us is Epic. We created an Epic-based system where nobody has to put in any information. The EHR will automatically look at the registry and run a report pulling all the information that we need so we can try to answer the clinical questions we have.

What are some of its most impactful capabilities?

When we initially ran the registry, it had identified somewhere around 25,000 patients. We’re a big healthcare system. We have 11 hospitals at this point and are rapidly growing. When we looked at our population, we were stuck with the question of, “Do we want to look at only some patients or do we want to look at everybody?” We decided that the best thing to do was to get the widest lens and focus in as we felt we needed to answer the questions that we have.

Essentially, what happened is we can begin to ask questions like, "Do we think this new injectable medication and this old injectable medicine will work well together?" We can query our patient population, look for exposure between the 2 medications and see the dates of exposure. Alongside this, just in our regular practice, we’ve been collecting patient disability scores. This includes things like the MIDAS score, which is a well-validated migraine disability score that asks questions about if a person with migraine can go to work, can they take care of their kids, etc. These are meaningful things for patients.

We also look at things like depression scores, anxiety scores, and we also have a measure for fatigue. I love these measures of outcomes because to our patients, those are the things that matter. It matters if you’re tired, it matters if you can’t take care of your kids. Those are data points that unless we make an effort to remember, aren’t typically asked in a routine visit. It’s easy to forget that the side effects of a medicine or condition, if not managed well, can make huge changes on a patient’s lifestyle and quality of life.

As we’ve routinely collected these data, it became a cool idea to look at these patient-centered outcomes, and to see if what we’re doing is making our patients actually better, as opposed to the kind of stuff you might find in a study, which is based on numbers and hard data. It’s been a long time in the process, and we’ve run the database for a couple of clinical queries, but we’re hoping this data could change how we look at this patient population, how we elicit data, and how we decide whether a treatment is working or not. Then how do we use that information to help patients understand how they might be able to partner with us, and even potentially, can we convince payers or insurance companies that things that they might think are expensive or should be held down the road should be offered sooner.

Do clinicians and patients have differing opinions on which data may be more important to collect?

It was an interesting problem when we created our database because things like total headache days in a month or how many times a patient takes their medicine, are numbers that are selected for research studies that may bring a drug to FDA approval. We love that hard data. But I feel like that does not tell the story for a lot of conditions, especially migraine, because there may be severity or additional symptoms that are hard to quantify.

When we look at our registry, the thing that we don’t have to this point is that hard data that tells how many days a patient has been affected. That is not input the same way into our record system. We can search for that by identifying patients and looking through the record, but the stuff that is heavily monitoring is the quality of life and disability aspects. This includes the changes in mood and other metrics like changes in fatigue, which actually may be more represented than anything else. I freely tell patients I get migraine too, and I’m a patient in my own headache center, my partner takes care of me. Those are the things that matter to me as a patient. Can I go to work? Can I come home and take care of my kids? While we want to include that hard numerical data, which we will over time, I’m proud that it’s not about that right now.

Transcript edited for clarity. For more NeuroVoices, click here.