David is co-lead of the Research Programme Healthcare Quality & Sustainability at the Department of Public Health. With this programme, they aim to personalize clinical decision making by developing and evaluating prediction models for progression and response to treatment, and methodology to support decision-making based on a shared responsibility of care provider and patient. They further aim to improve quality of care by evaluating the benefits and costs of health care interventions.
Motivated to be more meaningful for society, David changed his career in 2012 – from Head of the Pricing Depart of the insurance company Allianz Nederland – to researcher in medical decision making at the Department of Public Health. With his current research he empowers Personalized Medicine, by improving the derivation and validation of prediction models that guide treatment decisions. His research builds on a solid mathematical background, a strong drive to contribute to public health, ample experience in data sciences and clinical epidemiology, and an international network of collaborators. David is also affiliated, as a Research Associate, with the Predictive Analytics and Comparative Effectiveness (PACE) Center at Tufts Medical Center in Boston.
Erasmus Medical Center Rotterdam
Internal post address Na-2401
P.O. Box 2040
3000 CA Rotterdam
Visitor address:
Erasmus MC
The department is located at the 23th and 24th floor of the NA building.
Dr. Molenwaterplein 40
3015 GD Rotterdam
David started his academic career with methodological research on prediction modeling in clustered data. After obtaining a cum laude PhD degree in 2017, he developed and validated risk prediction models and counterfactual prediction models (to predict individualized treatment effects), techniques for developing counterfactual prediction models, and measures of the accuracy of both risk prediction and counterfactual prediction models. His work was funded through several research grants:
PCORI (Patient-Centered Outcomes Research Institute) grant: How Well Do Clinical Prediction Models (CPMs) Validate? A Large-Scale Evaluation of Cardiovascular Clinical Prediction Models (2017; Work-package leader)
PCORI (Patient-Centered Outcomes Research Institute) grant: Predictive Analytics Resource Center (2018; Co-PI)
ZonMW (Medical Sciences domain of the Dutch Research Council) grant: Clinical prediction models for COVID-19: development, international validation and use (2020; PI)
PCORI (Patient-Centered Outcomes Research Institute) grant: Generalizable prognostic models for patient-centered decisions in COVID-19 (2020; Co-PI)
NWO (Dutch Research Council) grant: Promising Start in the First 1000 Days of Life in the Netherlands (2021; Co-Investigator)
Horizon Europe grant: The 4D PICTURE Project: Design-based Data-Driven Decision-support Tools: Producing Improved Cancer Outcomes Through User-centered Research (2022; Work-package leader)
ZonMW (Medical Sciences domain of the Dutch Research Council) grant: Continuous updating of the ‘Covid Outcome Prediction in the Emergency department’ (COPE) model (2022; PI)
LEXCES (National Expertise Centre for Substance-Related Occupational Diseases) grant: Evaluation of the probability of causation approach and alternative measures (2024; Work–package leader)
Advanced Analysis of Prognosis Studies
The Netherlands Institute for Health Sciences (NIHES)
https://www.nihes.com/course/el014_advanced_analysis_of_prognosis_studies/
Clinical Epidemiology – Prognosis
The Netherlands Institute for Health Sciences (NIHES)
https://www.nihes.com/course/ck040_clinical_epidemiology/
Predictive Analytics course MOOC
Universiteit Leiden
https://www.coursera.org/learn/population-health-predictive-analytics/
Prediction Modeling
Erasmus MC Graduate School – PhD education and training
https://www.erasmusmc.nl/en/graduate-school/phd-trajectory/phd-courses#Core
Publications list
Most relevant publications
Weighted metrics are required when evaluating the performance of prediction models in nested case-control studies.
BMC Med Res Methodol
Derivation and Validation of the PRECISE-HBR Score to Predict Bleeding After Percutaneous Coronary Intervention
Circulation
Estimating individualized treatment effects from randomized controlled trials: a simulation study to compare risk-based approaches
BMC Med Res Methodol
Performance metrics for models designed to predict treatment effect
BMC Med Res Methodol
External Validation of the FREEDOM Score for Individualized Decision Making Between CABG and PCI
Journal of the American College of Cardiology
Redevelopment and validation of the SYNTAX score II to individualise decision making between percutaneous and surgical revascularisation in patients with complex coronary artery disease: secondary analysis of the multicentre randomised controlled SYNTAXES trial with external cohort validation
Lancet
Graphical calibration curves and the integrated calibration index (ICI) for survival models
Sat Med
Models with interactions overestimated heterogeneity of treatment effects and were prone to treatment mistargeting
J Clin Epidemiol
Personalized evidence based medicine: predictive approaches to heterogeneous treatment effects.
Bmj
Derivation and validation of the predicting bleeding complications in patients undergoing stent implantation and subsequent dual antiplatelet therapy (PRECISE-DAPT) score: a pooled analysis of individual-patient datasets from clinical trials
Lancet
https://calculator.syntaxscore2020.com
David developed and validated the SYNTAX Score II, a predictive score to guide decision making between coronary artery bypass graft surgery (CABG) and percutaneous coronary intervention (PCI) for individual patients with complex coronary artery disease. He updated the SYNTAX Score in 2020 and was senior author of the accompanying paper that was published in The Lancet.
http://www.precisedaptscore.com/predapt/webcalculator.html
David developed and validated the PRECISE-DAPT score, a prognostic score for out-of-hospital bleeding after stent implantation. The PRECISE-DAPT score improves decision making between short and long DAPT duration and is published in The Lancet.
David developed and validated a tool (COPE) for predicting death and need for intensive care unit (ICU) admission for patients who presented at the emergency department with suspected COVID-19. COPE is implemented as a publicly accessible web-based application and as independent Android and iPhone apps (‘COPE Decision Support’).
Predictive Analytics course MOOC
Universiteit Leiden https://www.coursera.org/learn/population-health-predictive-analytics/)