EpiDetective
Open tools for reading cancer evidence

The evidence on what causes cancer is public. Reading it shouldn't require a PhD.

EpiDetective builds free tools that let you explore cancer research for yourself — for patients, doctors, journalists, and the people who write health policy.

What this is

Research you can question yourself.

Every tool here starts from research that is already published and already careful. The difficulty is that almost nobody outside the field can read it. Each tool rebuilds that research into something you can search, follow and question directly.

The aim is to make findings usable while keeping the uncertainty that makes them trustworthy. Where the science is still unsure, the tool says so plainly.

The tools

One is open now. More are being built.

Open now

Cancer Evidence Explorer

See which exposures have been linked to which cancers, and how strong the evidence is behind each link. It is built on the IARC Monographs, the World Health Organization's review of what can cause cancer. Start from a cancer, start from an exposure, or open the whole map.

186 exposures68 cancer types 466 links10 languages
Who builds this

Built by a cancer epidemiologist who works with this evidence every day.

I am Wenxin Wan, a postdoctoral scientist at the International Agency for Research on Cancer in Lyon, the World Health Organization's cancer agency. My work covers cancer epidemiology, causal inference, biostatistics and machine learning.

I did my PhD at Utrecht University's Institute for Risk Assessment Sciences, studying how exposures at work and in the environment affect cancer risk. That work helped establish the causal role of benzene in lung cancer, which had long been underestimated.

Day to day I work with large international cohort studies, following hundreds of thousands of people over many years to see which exposures are followed by disease. At IARC I lead the analysis for OPICO, which pools ten such cohorts across three continents to ask whether long-term opioid use carries a cancer risk, and I advise the Lung Cancer Cohort Consortium (LC3) on how evidence about exposures at work can feed into lung cancer screening. The methods come from epidemiology, causal inference, pharmacoepidemiology and machine learning.

I build these tools because the evidence I work with is public and carefully made, and still almost unreadable unless you already know the field. Closing that gap seemed worth the effort. Everything below is checkable, and the links go to the original sources.

PositionPostdoctoral scientist, IARC / WHO
DoctorateUtrecht University, Institute for Risk Assessment Sciences
FieldsCancer epidemiology · causal inference · biostatistics · machine learning
Research projectsOPICO · SYNERGY · EPHOR · ECRHS
Consortium rolesLead analyst, OPICO · scientific advisor, LC3
Notes

Working through the hard parts in public.