target menu
... / ... / Research and Education / Medical Decision Making

Medical Decision Making

The research group Medical Decision Making studies health-related decision problems. We aim to support patients, clinicians, health care policymakers and others in making the best decisions about diagnostic, therapeutic and other medical interventions.

To help improve patient care
Our work is often pragmatic in nature, with the aim to directly support actual decision making. We focus on methodological issues, especially on the development and application of clinical prediction models, and on end-of-life care and decision-making. Furthermore, we conduct discrete choice experiments aimed at clarifying the trade-offs people make in their choices about medical interventions such as preventive drug treatment or cancer screening. Another important part of our research focuses on quality of care. Examples are our studies of the success rates of in-vitro fertilisation clinics and of performance indicators for Dutch hospitals. The latter studies are of particular relevance to the current debate on accountability and transparency of health care.

Some of our recent contributions Medical decision making 1

  • Clinical prediction modelling textbook - A milestone was the presentation of a text book on prediction modelling which reflects work over the past 15 years. The book provides a sensible and modern strategy for model development, validation and updating, such that prediction models can better support medical practice (Steyerberg. Clinical prediction models. Springer, 2009).
  • National and international collaboration in end-of-life research - Together with researchers from the VU medical centre in Amsterdam, the UMC Utrecht and the AMC in Amsterdam we completed the fourth Medical decision making 2nationwide study on euthanasia and other end-of-life decision-making practices. This study was followed by the OPCARE-9 project, an EU funded international collaboration of nine countries aimed at finding clues for improvement of care for cancer patients in the last days of life.
  • Evaluation of Liverpool Care Pathway - We conducted a study on the Liverpool Care Pathway, a set of guidelines for the care of dying patients. Our findings contributed to this protocol being implemented by health care institutions and policymakers throughout the Netherlands.

Ongoing projects

  • Evaluation of markers to improve predictions
    New markers are studied in various fields to improve the prediction of diagnostic or prognostic outcomes. The Prostate Cancer Molecular Medicine (PCMM) project  will address two major clinical needs in prostate cancer:
    1. The reduction of overdiagnosis and overtreatment due to screening by biomarkers
    2. The improvement of therapy monitoring of advanced prostate cancer by targeting imaging
    Our group will focus on Health Technology Assessment issues, including methodological aspects of marker evaluations.
  • Performance indicators for health care
    There is a growing focus on quality and safety in healthcare. Outcome indicators are increasingly used to compare hospital performance. We study various methodological aspects of such indicators. See: Articles in Pubmed by E.W. Steyerberg and A.M. Dishoeck and the dissertation of H.F. Lingsma  
  • Hospital care for the elderly
    Frail elderly often have relatively poor outcomes after hospital admission. In close collaboration with health care practitioners and the Dept of Policy and Health, we evaluate a program of reactivation care to prevent functional loss in hospitalised elderly.
  • Safety of non-steroidal anti-inflammatory drugs
    The Safety Of non-Steroidal anti-inflammatory drugs project (SOS) aims at assessing the cardiovascular (CV) and gastrointestinal (GI) safety of individual non-steroidal anti-inflammatory drugs (NSAIDs). The ultimate goal of the SOS project is to be able to provide a decision analytic model to aid in the choice of treatment, taking into account both the CV (stroke, myocardial infarction, heart failure) and GI risks (upper gastrointestinal complications). We collaborate with the Dept of Medical Informatics to develop risk functions and decision models for this project (SOS website).
  • Hereditary cancers
    Lynch syndrome is caused primarily by mutations in the mismatch repair genes MLH1 and MSH2. We analyze MLH1/MSH2 mutation prevalence in several large cohorts of patients undergoing genetic testing as well as in population-based series of colorectal cancer patients. See: Articles in Pubmed by E.W. Steyerberg and M.G. van Lier. We have developed and validated prediction models for the likelihood of finding a mutation in at-risk patients. See: Articles in Pubmed by E.W. Steyerberg and S. Syngal
  • Euthanasia and other medical end-of-life decisions
    Medical decisions at the end of life include euthanasia, physician-assisted suicide, intensive pain and symptom management and decisions to forgo life-prolonging treatment. In follow up to previous research, the frequency and characteristics of medical end-of-life decisions in 2010 are studied. This study is conducted by the Erasmus MC, University Medical Center Rotterdam,  the VU medical center in Amsterdam and the University Medical Center Utrecht,  in collaboration with Statistics Netherlands. See Medical decisions in end-of-life care of the Knowledge Center Palliative Care.
  • Palliative sedation
    To better understand the practices of professional caregivers and to optimize care for the vulnerable group of suffering dying patients and their relatives, more insight is necessary in the experiences of physicians, nurses and relatives with palliative sedation, i.e. rendering patients unconscious to alleviate suffering in the terminal phase. The UNBIASED study is conducted in the UK (University of Nottingham, Lancaster University, Southampton University and University of London), Belgium (Vrije Universiteit Brussel and Ghent University) and the Netherlands (Erasmus MC and VUmc). See Care for the dying patient of the Knowledge Center Palliative Care.
  • Care for the dying
    The Liverpool Care Pathway for the dying supports caregivers in adequately addressing the needs of dying patients and their families. It has been shown to improve care in different settings, such as the hospital, the nursing home and the home setting. We are developing an electronic version of the LCP that facilitates the care process, enables quality control and benchmarking, and makes it feasible to systematically collect data on the characteristics of dying and care for the dying. We also study the quality of care for the dying in Erasmus MC and the potential contribution of key champion nurses specialized in palliative care. See: the website Het Zorgpad Stervensfase.