NeuroGEN
Neurological & Mental Health Global Epidemiology Network
so…
15 = 7 claims
+ 6 EHR
+ 2 biobank
≈ 320 million people
9 regions · 4 continents
View the 15 databases as a table
| Region | Database | Type |
|---|---|---|
| United States | Medicare (20% sample) | claims |
| United States | Medicaid | claims |
| Scotland | Public Health Scotland (ISD) | EHR |
| United Kingdom | Clinical Practice Research Datalink (CPRD) | EHR |
| United Kingdom | The Health Improvement Network (THIN) | EHR |
| United Kingdom | UK Biobank (UKB) | biobank |
| Netherlands | PHARMO Database Network | EHR |
| Finland | Finnish healthcare registers (FinReg) | claims |
| South Korea | National Health Insurance Service – National Health Information Database (NHIS-NHID) | claims |
| Taiwan | National Health Insurance Research Database (NHIRD) | claims |
| Taiwan | Chang Gung Research Database (CGRD) | EHR |
| Taiwan | Taiwan Biobank (TWB) | biobank |
| Hong Kong | Clinical Data Analysis and Reporting System (CDARS) | EHR |
| Australia | Pharmaceutical Benefits Scheme; 10% sample (PBS) | claims |
| Australia | Victorian Linked Health Data (VLHD) | claims |
Source: Tsai DHT et al. Cross-Regional Data Initiative for the Assessment and Development of Treatment for Neurological and Mental Disorders. Clin Epidemiol 2023. PMID 38146486.
Off the board
The proof is the easy part. What it means for patients is the point.
- 1
The burden is enormous
Cognitive disorders, including dementia, and stroke are leading causes of disability, and mental health disorders, including depression, account for up to a third of all years lived with disability.
PMID 32572794
- 2
Trials leave the hardest questions open
Randomised trials often do not give enough information on a medicine's benefits and risks for vulnerable groups. Pooling databases across countries gives the power to study rare outcomes and small groups, earlier safety answers after a medicine is approved, and results that hold beyond one health system.
PMID 32572794
- 3
Sometimes the answer overturns the question
Earlier studies had linked diabetes in pregnancy to ADHD in children. One NeuroGEN study tested that link with sibling comparisons, and for gestational diabetes it did not hold. The case is worked below.
case file · PMID 38589601
The case of the missing cause
diabetes in pregnancy → ADHD ?
everyone: HR 1.10 (1.04–1.17)
siblings, one exposed, one not:
HR 1.05 (0.94–1.17)
∴ the interval crosses 1 …
unlikely to be causal
(diabetes that began before pregnancy: HR 1.39 (1.25–1.55), still raised)
Hazard ratios for ADHD after gestational diabetes, 95% confidence intervals, log scale. Source: Chan AYL et al. Nature Medicine 2024, PMID 38589601. The verdict is about gestational diabetes only.
Sometimes the most useful thing the data can tell a mother is what was never her fault.
neurogen.hku.hk ↗