Skip to ContentSkip to Navigation
About us Faculty of Science and Engineering Data Science & Systems Complexity (DSSC) About Data Science & Systems Complexity (DSSC)

Research profile prof. dr. E. (Eelko) Hak

Groningen Research Institute of Pharmacy, unit Pharmacotherapy, -epidemiology & economics,

Research description:

Pharmaceutical care costs are exponentially increasing due to population ageing and valid information on individualized therapeutic effectiveness is urgently needed. Despite the availability of ‘Big Data’ from healthcare databases, suboptimal causal methodologies hinder finding the true therapy effects. Crude therapy effect estimates are flawed and ‘confounding by risk factors’ must and can be quantified, even in subgroups. Unconventional/novel designs which I have pioneered (c.f. J Clin Epi, Lancet Inf Dis) create exciting new opportunities to improve quantification of this ‘confounding bias’, notably unmeasured confounding.

Projects will investigate and validate unconventional and novel methodologies to control unmeasured confounding bias on the outcome of therapy effect studies, and to incorporate precision medicine methods as biomarkers and genetics. Our group has unique access to high-quality ‘Big Data’ from both the Groningen LifeLines-cohort data registry (ISO9000-certified), prescription registry as well as the UK Clinical Practice Research Database (CPRD) databases covering millions of patients and variables. Effects of therapies with different confounder scenarios will be investigated. We will study and further develop methodologies originating from different scientific disciplines as medicine, psychology, economics, informatics: for example (1) case-only, (2) two-stage sampling, (3) machine learning, and (4) instrumental variable designs. Cutting-edge statistical techniques allowing for missing data and time-varying confounding will be applied. Novel methods will be instrumental to direct mainstream causal health care research and methodologies maybe up-scaled and applied in large consortia such as Data4Lifesciences (NFU).

keywords: pharmacy, health care, epidemiology, causal methodology, prediction research, precision medicine, Big Data, statistics

Employee (corporate) webpage:
Google Scholar:

Last modified:06 March 2019 3.13 p.m.