Laura Webber
In many chronic diseases, the challenge is not simply estimating how many patients exist today, it is understanding how those patients will progress over time, and how that progression can be altered.
Diseases such as chronic kidney disease (CKD) and other chronic conditions unfold over many years. Patients move through different stages, face different risks, and respond to treatment in different ways. There isn’t a single, predictable path. For pharmaceutical, biotech companies and policy makers, this creates a central question:
How can evidence be generated that reflects the reality of patient variation over time, and its impact on long-term outcomes?
The real challenge: people are different, averages can be misleading
This variation – heterogeneity – is a fundamental driver of disease progression and long-term outcomes. Patients differ in ways that directly affect outcomes:
Two patients with the same diagnosis may have very different futures. One may remain stable for years, while another progresses rapidly and experiences complications much earlier. Some will respond well to treatment, others less so. Some will be diagnosed and treated early, others late.
And over time, these differences compound.
So outcomes are rarely driven by the “average patient”. Instead, much of the future burden is concentrated in sub-groups:
These groups may represent a minority, but they can account for a large share of complications, cost, and unmet need. This has important implications for companies developing therapies, as the value of a treatment may be concentrated in specific segments.
Understanding these dynamics requires insight into how outcomes are distributed across patients, and how those distributions evolve over time.
Heterogeneity over time: pathways, not just populations
Another defining feature of chronic diseases is that history matters. What has already happened to a patient – complications, treatments, or delays – influence what happens next. Risk is constantly evolving as patients move through different stages and experience different events.
At the same time, interventions are not applied uniformly:
This means that outcomes are shaped by pathways, not just starting points. Differences between patients interact with time, events, and interventions to create very different trajectories.
What kind of modelling is needed to understand long-term outcomes in chronic disease?
To reflect reality, models need to go beyond averages. They need to be able to:
Why microsimulation helps reflect how patient differences shape outcomes
Microsimulation is a particularly powerful technique for addressing this challenge. Instead of modelling an average patient, it simulates individuals each with their own risk, history, and path. Patients can experience events, receive treatments, and progress through disease in ways that reflect real-world variation. This makes it possible to:
Importantly, population-level results are still generated, but they emerge from real variation amongst individuals, rather than smoothed averages.
Making the “hidden” patients visible
A key advantage of this approach is the ability to understand the full distribution of outcomes.
In many chronic diseases, a relatively small group of patients accounts for a large share of future complications and costs. These may be fast progressors, high-risk individuals, or those who experience delays in care. These “tails” are often where the greatest unmet need – and the greatest opportunity for intervention – lies. Capturing them explicitly allows:
Why this matters for decision-making
For pharma and biotech companies, this means:
These are inherently questions about heterogeneity over time. Approaches that can reflect how patient differences shape trajectories – and how those trajectories respond to intervention – provide a stronger foundation for answering them.
Why this matters for policy-makers
Policymakers are rarely deciding: “What happens to the average person?”. They are deciding:
Moving beyond the average
In chronic diseases, there is no typical patient. Outcomes are shaped by differences and how that difference plays out over time. Recognising and modelling that heterogeneity is essential to understanding where burden is accumulating, where interventions can have the greatest impact, and how value is created in practice.
Microsimulation offers a uniquely powerful way to do this, by aligning the structure of the model with the way disease and care pathways actually operate.
For the drug development lifecycle and policy makers this means moving beyond simplified averages and towards a more realistic view of patients, pathways, and outcomes and ultimately, better-informed decisions.