Research Methodology

Traditionally, research within the domains of primary and long-term care have for a large part adopted more classical epidemiological approaches such as randomized controlled trials and observational cohort studies. Although these approaches still carry value, they do not suffice to answer all research questions related to the complex health (organization) problems we face. We have gained methodological expertise on tackling these research questions. We have built on our extensive expertise in prediction research, e.g. by innovative application of response prediction in randomized controlled trials. Additionally, we have gained expertise in working with registry data, in applying and further developing mixed methods designs and in meta-scientific research. In order to consolidate and further our knowledge and expertise we have identified three focus areas:

Advanced methods for handling and analyzing electronic health registry data

Over the past years we have gained extensive expertise in the methodology of working with electronic health registry data from e.g. general practice and midwifery care, varying from data management to development of text mining or predictive algorithms and other longitudinal data analyses within registry data. Project examples include:

• Homburg M et al. A Natural Language Processing Model for COVID-19 Detection Based on Dutch General Practice Electronic Health Records by Using Bidirectional Encoder Representations From Transformers: Development and Validation Study
• Blok G, et al. Development and validation of a clinical prediction rule for acute appendicitis in children in primary care
• Du J . An electronic health records based study on the longitudinal patterns of psychotropic drug prescriptions for subpopulations of community-dwelling older people with dementia. Thesis, 2024.
• Sanavro et al. Randomized stepped-wedge trial with routine care data to assess the impact of an interdisciplinary consultation platform on the number of appropriate referrals. Ongoing project.

Research designs for complex health (care) interventions

We have in depth expertise in finding and further developing methodology for complex health (care) interventions, including mixed methods designs applied in participatory action research, process evaluations and implementation research. Project examples include:

• Participatory action research to improve health and social care professionals' identification and referral of pregnant women with mental health problems and/or intellectual disabilities. Ongoing project.
• Implementation and evaluation of a daily physical activity intervention for patients with cancer in general practices. Ongoing project.
• Improving primary-secondary care collaboration for better shared decision making with and realizing appropriate care for older patients with cancer. Ongoing project

Meta-science

Studying research itself is one of the drivers of improving research methods and practice. Our experience in this field supports research in how to account for biases in certain studies, how interpretation of methodological parameters can best be done in science and practice and how research results in secondary care might translate to those in primary care. Project examples include:

• Holtman et al. Bringing diagnostic accuracy to primary care. Ongoing project.
• Hulshof et al. The quantitative appraisal of trial methods in reviews applied to studies on antipsychotic drug effects in dementia. Thesis, 2021.
• De Boer. Reproducibility through an epidemiological lens. Ongoing project