Methodological Expertise
EMGO+ has a strong focus on methodological issues. This especially concerns the topics of clinimetrics (Riekie de Vet, Caroline Terwee, Raymond Ostelo, Dirk Knol), Health Technology Assessment (HTA) (Maurits van Tulder, Raymond Ostelo), prognosis and prediction research (Martijn Heymans, Riekie de Vet) and longitudinal data analysis (Jos Twisk, Martijn Heymans).
Clinimetrics
Clinimetrics concerns the quality of measurements in medical and health science research as well as in clinical practice. Much attention is paid to patient reported outcomes.
The clinimetrics group consists of 15 investigators, including PhD students, postdocs, and senior researchers from the various EMGO+ research programs. The clinimetrics group convenes once a month to discuss clinimetric issues on the basis of own research, manuscripts in preparation, or methodological papers from the literature.
The clinimetric group aims to promote research on the quality of measurement instruments; to publish educational papers on clinimetric issues; and to advise colleagues on the choice of measurement instruments, the development of new ones, and the interpretation of results.
Topics of interest are methods to determine minimal important changes on patient reported outcomes, the application of item response theory methods to improve health status measurement instruments and the development of methods to perform clinimetric reviews. Recently, consensus-based standards for selection of health status measurement instruments have been established in the COSMIN study.
Health technology assessment
The ever-rising cost of healthcare demands questions about how limited resources can be allocated to optimize health within the population. Health technology assessment (HTA) is an important tool in answering those questions.
HTA is scientific research that systematically examines the short- and long-term consequences of the application of health-related technologies. It is characterized by its multidisciplinary and comprehensive nature. HTA’s goal is to disseminate objective, valid, and reliable information that informs both the daily practice of healthcare professionals and the far-reaching decisions of policymakers.
While it overlaps with other research sectors such as epidemiology, HTA at EMGO+ focuses on economic evaluation. At EMGO+, with its ample expertise in intervention studies, economic HTA evaluations are conducted alongside randomized controlled trials of diagnostic, preventive, and therapeutic interventions within the four research programs. In addition to economic evaluations, HTA researchers at EMGO+ also perform systematic reviews on interventions’ effectiveness (within the framework of the Cochrane Collaboration), develop evidence-based guidelines, and evaluate the implementation of those guidelines.
The HTA Unit’s main objective is to establish a high quality scientific research program, but it also consults, offers support and advice concerning economic evaluations to colleagues within the VUmc, and educates, training students in economic evaluation.
Prognosis and Prediction
Prognostic and prediction studies aim to distinguish between patients with a favorable and poor outcome. The aim of a prognostic model is to estimate (predict)the probability of a particular outcome as optimally as possible, and not just to explore the causality of the association between a specific factor and the outcome (explanatory). The results of these predictive studies are important to inform the patients about the probable course of their disease, to make adequate treatment decisions, or to plan health care facilities. Furthermore, evidence for causal prognostic factors may trigger the development of new interventions. The prediction working group of EMGO+ examines which methods are most adequate to design and analyze prognostic factors and prediction studies. Guidelines have been produced for EMGO+ researchers for the optimal performance of prognosis and prediction studies.
Longitudinal data analysis
Longitudinal data analysis aims to measure change in repeated measurements over time and the factors that influence this change. A typical feature of these measurements is that they are clustered. If measurements are taken from the same individual, within and between individual change can be assessed. The clusters than consist of repeated measurements over time obtained from a single individual. Because these measurements are obtained from the same individual, they will show positive correlation. Techniques have been developed that are able to account for this correlation. Most used techniques are generalized estimating equations (GEE) and Mixed models. When information is obtained from different measurement levels, i.e. time and individuals, these techniques are also called multilevel techniques.
EMGO+ is well known for their large longitudinal cohort studies that are conducted in the different research programs. Specific methodological expertise exists within EMGO+ to offer support and advice concerning longitudinal data analysis and to conduct high quality methodological longitudinal research.
Mixed methods
When research questions are leading for the choice of a research method, a mix of quantitative and qualitative methods will often be the most appropriate method. For instance, quantitative methods can give insight in the frequency of a phenomenon, while qualitative can shed light on the way this phenomenon is experienced and impacts the life of people who encounter this phenomenon.
Especially within the research program Quality of Care there is substantial experience with mixed methods of research. This not only requires knowledge on both quantitative and qualitative methods, but also on how to combine these types of methods within one research project. Mixed methods is more than combining results of separate quantitative and qualitative studies on the level of interpretation, but require integrating both methods in the design and analysis of a study. An example is nesting of a qualitative study in a quantitative framework: the information collected in for instance a quantitative survey is used to select a relevant selection of respondents for in-depth interviews.