Patient-provided information for improving real-world evidence: an interview with Elisabeth Oehrlein
In this feature, Elisabeth Oehrlein, (National Health Council, DC, USA) discusses the importance of patient engagement and leveraging patient-provided information to improve real-world evidence (RWE).
Please could you introduce yourself and your role at the National Health Council?
I am the Senior Director of Research and Programs at the National Health Council (DC, USA). Our mission is to provide a united voice for patients with chronic diseases and disabilities, their families and caregivers.
We are a membership organization with approximately 140 members; the largest category of members is patient organizations and the remaining are other stakeholders from across the health care ecosystem – including payers, biopharma and health professional organizations. We envision a society in which all people have access to quality, affordable health care that reflects personal goals and aspirations to promote patients best possible health outcomes.
What is the importance of patient-provided information for improving real-world evidence (RWE)?
Patient-provided information is data that stem directly from patients. However, there are two levels of patient-provided information. First, patient-provided information can take the form of information that patients voice or report themselves – information about outcomes, preferences, perspectives. This is often simply referred to as patient input. Second, patient-provided information can also refer to data derived from patients’ wearable devices, such as Fitbits. This is often termed patient-generated data.
At the National Health Council, a lot of our work relates to patient engagement; not only capturing or gathering meaningful patient input through patient engagement, but also how patient input can be applied. For example, right now we have an ongoing project about how we apply patient input to improve a real-world data study design – to make sure that real-world evidence reflects real-world patient experiences.
…we have an ongoing project about how we apply patient input to improve a real-world data study design – to make sure that real-world evidence reflects real-world patient experiences.”
For example, we are working to understand how patient input may be useful for assembling study cohorts, identifying important covariates, and so forth. We know that electronic medical records and administrative claims data are often utilized for RWE generation. However, researchers often face a lot of uncertainty when trying to figure out exactly what is going on in those datasets. If there are gaps or unexpected trends in data, patients can be really helpful for supplementing explanations for these that cannot be inferred from studying the traditional dataset alone.
At ISPOR (18–22 May 2019, New Orleans, LA, USA), we presented the example of atrial fibrillation (AF) where there is often a time lag between when a patient might first be diagnosed with a condition and when they actually get to see a specialist for it. During this lag period, the patient may not receive a prescription, but they may also not understand why they may need a prescription in the first place, as stroke risk is often not adequately explained to them. This has huge implications for researchers with respect to setting index or follow-up dates for a study. If these are not set for the right time period, the study findings may be skewed.
Another important way that patient engagement can help is aiding researchers in coming up with the right research questions in the first place. For example, a lot of my past work was in AF. I found it interesting that – when I was speaking with AF patients and looking online at AF patient forums – there was a huge amount of patient discussion concerning triggers of AF episodes. However, when I looked in the literature, there had not been much research into AF triggers.
By contrast, patient-generated data can be extremely helpful for filling in gaps where there is missing data, or for supplementing data that researchers might be interested in but don’t have access to. For example in claims datasets, we often do not have information about patients’ functional status. In instances where this information may be helpful, a Fitbit might be able to supplement this additional data. However, we still face challenges with regard to linking Fitbit and wearable data with other types of data.
How may RWE guide regulatory decision-making and value assessment, and why is the patient perspective important in this?
In previous research that we conducted concerning patients’ perspectives on RWE being used in regulatory decision making, we found that patients very much feel that RWE can be extremely helpful in this regard. RWE can promote label expansion by informing new indications for an approved drug product. Last year, Pfizer utilized RWE concerning a breast cancer drug, which was already approved for use in women, to demonstrate the drug had similar effectiveness amongst men. Since, the indication for that drug has been expanded. This is the type of example that we found to be very reflective of what our qualitative research with patients told us – patients are willing to accept this kind of use of RWE. They still want clinical trials for new (unapproved) drugs, but they see a clear place for RWE use too.
…we still want clinical trials, but we see a clear place for RWE use too.”
A further aspect of patient-provided information, as it relates to value assessment specifically, is that a lot of patient groups have robust registries of longitudinal data about their constituents. There has been a really nice example within the multiple sclerosis (MS) field, where patient groups came together and surveyed their MS patients. Within approximately four days, the groups were able to collect data from thousands of MS patients concerning a question that one of the US value assessment bodies had. This demonstrates the remarkable opportunity that exists for compiling data - nearly in real time - when value assessors believe they have insufficient data for their work. To take advantage of these opportunities, value assessors – and other decision-makers – need to partner with patient groups. To build trust, they also need to demonstrate how the data that was collected was used in the value assessment process.
Do you think the quality of some of the data is a concern? How can we work towards generating RWE and data that is increasingly high quality and valid?
There is certainly a huge amount of data variability. With respect to how we may further improve data quality, I think we are putting a lot of work, research and investment into the research infrastructure that is collecting patient-generated and real-world data. We must work to collect meaningful data, not just data for the sake of accumulating data.
We must work to collect meaningful data, not just data for the sake of accumulating data…”
We must ensure that the patient-provided data we are collecting and the questions that we are asking patients, are actually on topics patients care about and want used. If we simply just ask patients to provide information to us, but they don’t really care about the information that they are providing, then we are not advancing the field. This is a key area where patient engagement is important. We are not just telling patients to wear Fitbits because we can get their data from it. Instead, we are engaging patients and asking: do you care about the number of steps you take and what might be an indicator of a good day for you with respect to that number of steps?
How do you think the inclusion and use of patient-generated information will change and evolve in the future?
I think the use of patient-provided and -generated information will continue to increase. Already, we are seeing groups that have never really been interested in patient engagement becoming increasingly interested. For example, international groups, such as the Council for International Organizations of Medical Sciences (CIOMS; Geneva, Switzerland) – which is well known for its work on drug safety – now has a working groups on patient involvement in the development of medicines. Regulators in Japan have also started explaining how patient input can be incorporated.
My hope is that over the next few years, we see a blending of patient-centered drug development and RWE initiatives – right now they’re pretty siloed and they shouldn’t be."
I think the FDA, European Medicines Agency (EMA; Amsterdam, Netherlands) and some health technology assessment bodies have laid the groundwork and demonstrated, along with the patient community, that patients are experts on their conditions and the information they provide is really useful. My hope is that over the next few years, we see a blending of patient-centered drug development and RWE initiatives – right now they’re pretty siloed and they shouldn’t be.
The opinions expressed in this feature are those of the interviewee/author and do not necessarily reflect the views of The Evidence Base® or Future Science Group.