In this QCast episode, co-hosts Jullia and Tom explore health economics and outcomes research. They explain how outcomes research looks at the real world end results of healthcare, how health economics and outcomes research combines those outcomes with costs and resource use, and why this evidence matters for pricing, reimbursement, and access decisions. They walk through typical HEOR study designs, the role of real world data and longitudinal databases, and the collaboration needed between sponsors, statisticians, and programmers to deliver decision-grade evidence.
🎧 Listen to the Episode:
Key Takeaways
What Health Economics and Outcomes Research Is and Why It Matters
Health economics and outcomes research, or HEOR, assesses the value of treatments by looking at clinical outcomes, quality of life, and costs together. It considers how medicines perform in routine care, not just in controlled trials, and weighs benefits, harms, and resource use. This helps payers and health systems decide which treatments offer the best value for patients and budgets, and guides manufacturers on how to position and support their products.
Using Real World Data and Longitudinal Databases
HEOR often relies on large real world datasets such as claims, electronic health records, and hospital databases that follow patients over time. These sources show how diseases are managed, how medicines are used, and what healthcare resources are consumed. Because the data are large and complex, teams typically use dedicated data warehouse environments, apply consistent coding and pre-processing, and work in SQL to define cohorts and variables before creating smaller, analysis-ready datasets for modelling.
Designing and Delivering HEOR Studies
A structured approach to HEOR study design helps ensure reliable results. First, define clear inclusion and exclusion criteria, baselines, and look-back periods, and specify how to identify disease, exposure, and outcomes. Next, run analyses that match the question, starting with descriptive summaries and, where needed, moving to comparative and adjusted models. Finally, prepare curated outputs that can feed into cost-effectiveness or budget impact models. Throughout, methods should be chosen to address a specific decision, such as understanding burden of disease or supporting a pricing and access discussion.
Collaboration, Quality Control, and Transparency
Effective HEOR work depends on close collaboration between sponsors, programmers, and statisticians, backed by strong quality control. Clear cohort definitions, code lists, and flow diagrams make it obvious who is included and why. Independent programming checks reduce errors, while thorough documentation supports transparency and reproducibility. This level of clarity helps payers, regulators, and other stakeholders understand how the evidence was generated and increases confidence in the conclusions.
Practical Takeaways for Teams Planning HEOR Work
HEOR is most useful when it is anchored to a specific decision about value, pricing, access, or policy. Teams should respect both the strengths and limitations of real world data, invest in suitable infrastructure and programming skills, and treat quality control as essential. Thinking about payers, clinicians, and regulators from the start helps shape analyses and reporting so that HEOR becomes a strategic tool across the product lifecycle rather than a one-off exercise.
Full Transcript
Jullia
Welcome to QCast, the show where biometric expertise meets data-driven dialogue. I’m Jullia.
Tom
I’m Tom, and in each episode, we dive into the methodologies, case studies, regulatory shifts, and industry trends shaping modern drug development.
Jullia
Whether you’re in biotech, pharma or life sciences, we’re here to bring you practical insights straight from a leading biometrics CRO. Let’s get started.
Tom
Today we are focusing on health economics and outcomes research. Jullia, before we get into databases and programming, can you set the scene? What do we mean by outcomes research, and how does that link to health economics and outcomes research as a whole in drug development?
Jullia
So, outcomes research is essentially about the real-world end results of healthcare. Instead of asking only whether a treatment works under ideal trial conditions, it asks what happens to patients in routine care. We look at clinical outcomes, patterns of use, and how care affects quality of life and resource use when people are treated in everyday settings. Health economics and outcomes research, or HEOR, brings that together with economic thinking. It evaluates the value of treatments by looking at their benefits, harms, and costs side by side. That includes direct medical costs, like hospital stays and drugs, as well as things like time off work or the burden on caregivers. In practice, HEOR gives decision makers a structured way to judge whether a new treatment offers better value than current options, not just whether it is effective.
Tom
So, from what you’re saying, it’s not just an academic exercise. How does HEOR influence decisions across the product lifecycle, particularly around pricing, reimbursement, and access for patients in different health systems?
Jullia
You’re right, it’s highly applied. HEOR runs through the whole lifecycle of a medicine. Early on, teams use outcomes and economic thinking to identify which patient groups may see the largest incremental benefit. That can inform dose selection, inclusion criteria, and which endpoints to prioritise. As evidence matures, HEOR helps calculate cost effectiveness or budget impact compared with current practice, which is central to health technology assessments by bodies such as NICE in the UK and other agencies globally. These organisations increasingly expect a mix of trial data and real-world evidence when judging value for money. HEOR outputs then feed into pricing and reimbursement negotiations, value dossiers, and launch strategies. For manufacturers, it supports decisions on which indications to pursue, how to position a treatment, and how to communicate value to payers, clinicians, and patients. For health systems, it offers a way to allocate constrained budgets towards treatments that deliver meaningful outcomes per unit of spend.
Tom
I noticed you mentioned real world evidence there. Day to day, what does outcomes research actually look like in practice? If we were to zoom in on the programming side, what kinds of projects are teams working on and what are they trying to answer?
Jullia
So, from a programming perspective, outcomes research often centres on large retrospective, non-interventional studies using existing healthcare databases. Instead of building a new trial, we work with de identified data from claims, electronic health records, or hospital systems that follow patients over many years. Sponsors might ask us to measure disease prevalence in a given population, track how a drug is used over time, or compare treatment patterns between products. We may analyse healthcare resource use, such as hospitalisations and outpatient visits, alongside associated costs. Another common task is assessing whether there is a sufficient real-world population that matches planned inclusion and exclusion criteria for an upcoming trial. In all of these cases, programmers need a solid grasp of database structure, coding rules, and statistical concepts. Our role is to translate broad business questions into concrete cohorts, variables, and analysis datasets that can support robust outcomes and economic evaluations.
Tom
However, the thing about those datasets is they can be huge and are often messy. Can you talk about the types of longitudinal databases that are common in HEOR and how to efficiently handle the technical challenges of working with that scale of data?
Jullia
Absolutely. So, longitudinal databases are designed to follow patients over time, often across different care settings. Examples include administrative claims databases, where we see diagnoses, procedures, and pharmacy dispensings, and clinical databases from primary care or hospital networks. These resources can cover millions of individuals over decades, so naively extracting everything into standard datasets can be slow, expensive, and hard to manage. To deal with that, we usually build a dedicated data warehouse environment tailored for big data. Raw vendor data are first pre-processed to harmonise coding, resolve inconsistencies, and integrate multiple sources. We then use database technologies, such as Teradata or similar platforms from Oracle, IBM, or Microsoft, to push as much processing as possible inside the database. Analysts work in SQL to define cohorts and derive variables, and only bring smaller, analysis ready extracts into tools like SAS for modelling. This approach keeps runtimes manageable, reduces storage overhead, and makes it easier to rerun or adapt studies when assumptions change.
Tom
Once you have that environment set up, design choices become critical. How do you typically design an outcomes research study in this setting, and what are the main analysis approaches you see in HEOR work?
Jullia
We usually think in three broad phases. First, we define and extract the cohort. That means setting clear inclusion and exclusion criteria, applying look back periods, and establishing baselines. You want to be explicit about how you identify disease, exposure, and outcomes using codes and dates. Second, we run the requested analyses. At a minimum, that includes descriptive summaries of demographics, comorbidities, treatment patterns, and healthcare utilisation and costs. Depending on the question, we might also do comparative analyses between drugs, persistence and switching analyses, or case control style comparisons between groups. Third, we extract curated data for downstream work, such as feeding economic models or further statistical analyses. On the modelling side, we often see logistic regression, time to event methods like survival analysis, or other regression frameworks to adjust for confounding. The key is to align each analysis with the original decision problem, whether that is estimating burden of disease, understanding current practice, or generating inputs for cost effectiveness or budget impact models.
Tom
From what you’re saying, it sounds like a lot of moving parts and stakeholders. How do sponsors typically work with HEOR programming teams, and what are some best practices to avoid common pitfalls when dealing with real world data and economic questions?
Jullia
As you’d imagine, collaboration is critical. A good HEOR study usually has a main programmer, a quality control programmer, a statistician, and sponsor representatives aligned from the start. Sponsors should come with a clear decision question but be open to refining it based on data realities. One best practice is to invest time upfront in understanding data limitations, such as missing variables, incomplete follow up, or coding changes over time. That prevents unrealistic promises and late surprises. Detailed cohort specifications, code lists, and flow diagrams help everyone see exactly who is included and why. From a programming standpoint, independent quality control is non-negotiable. A second programmer should replicate key steps to catch programming or logic errors. Documentation is also vital, because health technology assessment bodies and payers increasingly expect transparent, reproducible evidence. Finally, keep communication frequent and practical. Short, iterative reviews of interim outputs often surface issues earlier than a single, polished final delivery.
Tom
Thanks Jullia. Now before we close, could you give listeners a concise set of takeaways from what we've discussed?
Jullia
So for me, there are four practical takeaways. First, be clear on the decision you are trying to support. HEOR only adds value if the outcomes and economic questions are anchored to a real pricing, access, or policy decision. Second, respect the strengths and limits of real-world data. Large databases are powerful, but they are not randomised trials. Careful cohort definition, confounding control, and transparent assumptions are essential. Third, invest in the right infrastructure and skills. Efficient database environments, strong SQL and programming expertise, and robust quality control make the difference between exploratory analyses and decision grade evidence. Finally, think about stakeholders early. Payers, clinicians, and regulators often look for slightly different signals, so it helps to plan analyses and reporting with those audiences in mind from the outset. If teams keep those points in view, HEOR can move from a nice to have to a strategic tool across the product lifecycle.
Jullia
With that, we’ve come to the end of today’s episode on health economics and outcomes research. If you found this discussion useful, don’t forget to subscribe to QCast so you never miss an episode and share it with a colleague. And if you’d like to learn more about how Quanticate supports data-driven solutions in clinical trials, head to our website or get in touch.
Tom
Thanks for tuning in, and we’ll see you in the next episode.
About QCast
QCast by Quanticate is the podcast for biotech, pharma, and life science leaders looking to deepen their understanding of biometrics and modern drug development. Join co-hosts Tom and Jullia as they explore methodologies, case studies, regulatory shifts, and industry trends shaping the future of clinical research. Where biometric expertise meets data-driven dialogue, QCast delivers practical insights and thought leadership to inform your next breakthrough.
Subscribe to QCast on Apple Podcasts or Spotify to never miss an episode.
