01
Incidence & burden projection
Transmission models that project cases, hospitalizations, and deaths forward under explicit scenarios — accounting for seasonality, waning immunity, and the emergence of new variants.
Infectious disease modeling for decisions that matter.
EpiSight helps pharmaceutical, biotech, and public health organizations project how disease spreads — and translate those projections into decisions about products, trials, and populations.
What we do
Quantitative answers to epidemiological questions
Most of our work begins with a question that cannot be answered from historical data alone: how many people will need this, and when?
02
Viral kinetics & within-host dynamics
How viral load rises and falls over an infection, how much that varies between people, and whether a therapy shifts it. The framework we introduced for estimating viral trajectories is now widely used.
03
Trial design & simulation
Will the infection circulate where and when you enroll, and can your endpoint detect the effect you expect? We simulate the trial before you commit to a protocol.
04
Real-world evidence from claims data
How medicines and vaccines are actually used in practice, drawn from large administrative claims datasets — with particular depth in outpatient prescribing.
05
Model review & expert advisory
An independent read on a model built in-house or by another vendor — what it assumes, where it is fragile, and whether its conclusions survive those assumptions.
How we work
Research-grade methods, on a business timeline
- Fit for purpose
- The right model is the simplest one that answers your question. We would rather hand you a transparent model you can interrogate than an elaborate one you have to take on faith.
- Uncertainty, stated plainly
- Every projection carries uncertainty. We quantify it, show which assumptions drive it, and tell you where the model stops being informative — rather than burying the caveats in an appendix.
- Yours to keep
- Engagements conclude with documented code, data sources, and assumptions. Your team can re-run the analysis with new data after we are gone.
- Built to withstand review
- Our methods come out of peer-reviewed epidemiological research and are written to be defended — to a regulator, a reviewer, or a skeptical colleague.
Have a question that needs a number attached to it?
Tell us what decision you are trying to make. If we are not the right fit, we will say so and point you somewhere better.
Get in touch