Nested Case-Control and Case-Cohort: Efficient Study Designs to Develop Biomarker-Based Prediction Models for Rare Outcomes

Prediction models that accurately predict patient prognosis and treatment response enable the development of personalized treatment plans in dermatology as well as outside dermatology. This can improve patient care and reduce the use of ineffective, potentially harmful treatments. Developing representative models for all patients can be a significant challenge, particularly in cases involving rare outcomes or expensive molecular biomarkers. The former requires large cohorts, whereas the latter requires a significant budget to measure these in a large volume of samples. Nested case-control and case-cohort designs are cost-effective designs that enable the development and validation of prediction models using only a proportion of samples of the source cohort, without compromising their applicability to the total population. These epidemiological designs are relatively unknown, and there is a lack of clear guidance on how to develop and validate models using these studies. We aim to inspire other researchers to apply these designs to their (skin) disease of interest and facilitate the development of prediction models that can have a high clinical impact on patient care.