How NICE is approaching guidelines for RWE generation: Q&A with Dr. Páll Jónsson
In its recently announced five-year strategy, the UK’s National Institute for Health and Care Excellence (NICE) articulated how it plans to use real-world evidence (RWE) to “resolve gaps in knowledge and drive forward access to innovations for patients.”
This comes as NICE and other health technology assessment (HTA) agencies have expressed interest in using RWE to inform decision-making, as well as in generating new RWE to supplement the evidence submitted by manufacturers, or that is available in existing literature.
We spoke with Páll Jónsson, Ph.D., Associate Director for Science Policy and Research at NICE, to learn about the organization’s RWE goals and guidelines, and how he sees RWE continuing to impact the HTA process.
Dr. Jónsson is trained in biochemistry and bioinformatics, and prior to NICE he worked across academia, small biotech, and large pharmaceutical organizations. His previous roles often focused on drug discovery in oncology, a perspective that he finds useful to understanding “the long journey that it takes to get a drug through from development to patients in practice.” Read on to learn more.
Responses have been edited for clarity and length.
Q: Can you share some information about the work you do at NICE?
A: NICE assesses the clinical and cost effectiveness of new drugs and other medical technologies. But the HTA process is just one piece of our work; we also develop guidelines to inform clinical care, public health, and social care.
My team works with other NICE teams to identify challenges, ranging from a need to respond to novel scientific developments to changes to the health care system, then we work together to develop solutions. My team has a broad portfolio of work; for instance, we’re currently addressing how to measure quality of life in people who receive treatments, and how to incorporate environmental sustainability into our recommendations.
We’ve also explored how NICE can assess histology-independent cancer drugs—which are interventions used to treat various tumor types that have a common genetic mutation. These mutations are often rare, so the clinical evidential effectiveness of these drugs is based on immature data and very small sample sizes. This makes it challenging to assess whether the drug will provide value to the health system. We’re working with NICE colleagues to find the best way to assess these drugs, and it’s likely that we’ll need RWE to do so.
Q: How has NICE’s use of real-world data (RWD) and RWE evolved over time?
A: NICE has always tried to take a pragmatic approach to evidence in our HTA processes. And because the majority of the evidence we receive was generated to support regulatory approvals—and therefore largely comes from randomized controlled trials (RCTs) that establish the relative efficacy of new treatments—we’ve relied on RWE to inform many other aspects of our work. For instance, we use RWE to better our understanding of the natural history of diseases, as well as in our economics modeling to characterize patient populations.
We’re increasingly seeing new drugs come to market supported by single-arm studies, particularly for rare diseases, which have small patient populations and potentially limited RCT data. Inevitably, because of those external factors, we have to rely more on RWD to support these types of evaluations. We also use RWE to support assessments for diagnostics, medical devices, interventional procedures, and, most recently, digital health technologies, which traditionally have had a less developed evidence base.
The bottom line is that we are seeing more RWD and RWE submitted alongside RCTs than we did before, and I think we’re getting better at understanding where RWE is appropriate for informing decision-making. At the same time, we’re also getting better at appraising the quality of RWE.
Q: How has COVID-19 impacted NICE’s priorities in the last year, particularly with respect to RWD and RWE?
A: As you would imagine, COVID-19 had a significant and disruptive impact on everything we did over the last 12 months. We shifted our core focus to developing guidance for treating COVID-19, and our Data & Analytics team—who were responsible for the RWE methods and standards programs—provided support to the development of our rapid COVID-19 clinical guidelines. As part of that, our teams dealt with an incredible amount of emerging evidence from various sources. This demonstrates the completely different landscape we’re in at the moment; we have so much data, generated so quickly in a variety of sources. We need to understand how to navigate these novel challenges, because they ultimately impact our RWE capabilities.
The pandemic itself has brought to us valuable experience, albeit through difficult situations. It has enabled us to, for instance, develop a pipeline of living guidance for COVID, which is intended to be a more dynamic, more frequently updated guidance and takes into account the learnings from the evidence generated as we go along. This experience with COVID-19 is also informing our thinking around how we use RWE more broadly.
Now we are moving into the COVID-19 recovery phase in our work program, and we’re refocusing our efforts on the RWE methods and standard programs, which form a key part of NICE’s strategic ambition to demonstrate leadership in data, research, and science. We will work with external partners—including international consortia, industry, and others—to enable more rapid development of high quality output.
Q: Can you give an example of an initiative that you see having a significant impact on NICE’s use of RWE?
A: We recently announced our five-year strategy—the first time that we’ve created a strategy document—to outline how we’ll develop our work over the next five years. One of the work packages detailed within this strategy is for NICE to “develop a world-leading capability and standards for routinely using RWD and RWE to inform all aspects of our work.”
So, what will this mean for NICE? We will have increasing ability to link RWE with evidence-based practice, which means we can shift from producing recommendations at a single, “static” point in time to more dynamic, living guidance. In terms of health technology assessments of new drugs, we will look to robust ways of using RWE to resolve issues of uncertainty at the time of launch so that we can improve access to new innovations for patients.
Q: Globally, how do you see RWE continuing to inform HTA processes?
A: RWE can play a pivotal role in managed access agreements, which are the UK’s method of providing early access to promising new treatments whilst continuing to collect data to resolve gaps in the evidence base. I think the use of RWE here will continue to expand.
There are also opportunities for living guidelines beyond COVID-19, including for clinical care that takes evidence created outside of clinical trials into account. Coming alongside an improved data infrastructure, this will allow us to access and harness the valuable data that’s generated in the health service.
Q: What more needs to be done to advance RWE use by HTA agencies?
A: HTAs are part of the health care environment, but they’re not the only part. It’s important to have a dialogue across sectors and stakeholders to ensure that the data being generated is coordinated between regulators and downstream decision makers, such as HTAs, payers, providers, and patients.
Data quality and evidence synthesis are two important aspects to consider going forward in this journey. We need to understand how to generate high quality RWE by using the most appropriate methods for synthesizing data into decision-quality evidence. To get there, we need to drive the research agenda and funding priorities through new collaborations with academia, government bodies, or industry to work towards the practical application of RWD, and towards understanding the strengths and limitations of new data sources.
Q: Across the health care landscape, stakeholders are collaborating on projects to advance RWE methods and learnings. Which of these initiatives are you watching the most closely, and which do you think will most impact the RWE knowledge base?
A: As a decision maker, NICE sees the lack of transparency regarding how evidence is generated as a major barrier for using RWE. The Real-World Evidence Transparency Initiative is doing great work to guide us toward more transparency in evidence generation, so I’m watching their work quite closely.
NICE is also a partner in the European Health Data Evidence Network (EHDEN), which is generating Europe’s largest federated network of health data. The goal is to standardize the data network to a common data model, and then develop analytical tools. This is attractive, because it addresses many challenges around data protection and access, and it enables us to analyze data across different jurisdictions without moving data from the site where it was generated. Of course, there are pros and cons associated with the federated approach compared with pooling data into a single, large repository. But I’m excited to test the approach in the context of an HTA agency to see where that takes us.
Opposite to the concept of the federated network, the National Health Service is developing the concept of trusted research environments (TREs). This service will provide approved researchers with access to de-identified, linked data from health and care systems in the UK. The idea is to offer a flexible solution that provides a safe and secure way to access and analyze data. This service has proven useful in helping us understand the impact of COVID-19, but it could be scaled across other areas to generate information on guidance in health and care.
Q: Through work with RWE partners, what does NICE hope to achieve?
A: Generally speaking, I’m keen to continue stress testing data and analytics in the context of our decision-making frameworks. If we start to do something new, be it using new data or new analytics or new ways of synthesizing data, we need to ensure that it drives towards reducing uncertainty in decision making. That’s a key point in our decision frameworks; it’s the end goal. If we can get there, even one step at a time, we’re in a great place.