Anita Berglund

We will adapt and optimize tools that combine advanced causal inference and AI methodology. We will integrate different implementations of the plug-in g-formula with deep neural networks (i.e., deep learning). We will then implement these tools for the generation of real world evidence about comparative effectiveness and safety in observational research. Specifically, our aims are:Adaptation of software tools for causal inference in register-based researchWe will extend and customize software for causal inference with time-varying treatments, confounders, and outcomes using Swedish longitudinal registers and other microdata. The deliverables will include guidelines, tutorials, and recommendations for the implementation of the methodology to register data and other sources of observational data.Development of use-cases of register-based causal inference researchWe will demonstrate the value of the methodology, and the software tools to implement it, for register-based research. Initially, we will quantify causal effects in cardiovascular research settings: beta blockers and myocardial infarction in people with preserved ejection fraction, and antiplatelets and mortality in people with acute coronary syndrome. We will then identify, conduct, and publish additional use cases in other health areasOur project will contribute to methodology development in causal inference from observational data, and therefore to increased use of registers for excellent and innovative research.

Swedish registers are a unique and powerful resource, enabling efficient and cost-effective investigation of a range of research questions. The Swedish Interdisciplinary Graduate School in Register-Based Research (SINGS) is a comprehensive two-year programme focusing on methodological, practical, ethical, and legal aspects of using registers in research. It aims to deepen knowledge and enhance skills on effective use of registers in research, making it relevant to all quantitative disciplines. Open to doctoral students at any Swedish university, it includes seven core and nine elective course weeks, with students required to earn at least 12 credits. To further create a stimulating and international environment, it offers seminars on cutting-edge methodological topics. Karolinska Institutet coordinates the school, with participation from six other Swedish universities and two international collaborators. Since 2010, SINGS has had a crucial role in fostering the next generation of researchers. Seven cohorts of students have been admitted. More than 700 students and over 100 teachers from different universities and research fields have taken part. Eighty courses have been held. This updated version of SINGS builds on its success and includes new collaborators and educational activities on emerging topics, thus enabling the further development of the highest academic standards with a strong pedagogical, methodological, and interdisciplinary profile in register-based research.

The association between weight, and cardiovascular disease and mortality is well established, however, the causal effect of weight-loss in midlife on these outcomes is less clear. Bariatric surgery results in substantial weight-loss and is an ideal candidate to study the causal effects of weight-loss. We propose a project that willcausal inference and machine learning methods to answer two important questions: 2) Is bariatric surgery effective for reducing cardiovascular disease and mortality, and if so, for who? 3) Which type of bariatric surgery (gastric bypass or sleeve gastrectomy) is most effective, and for who?We will use data from various Swedish registers to identify individuals with obesity who are eligible for bariatric surgery. We will then compare cardiovascular and mortality outcomes among those undergoing different types of bariatric surgery with those receiving non-surgical obesity management using causal inference methods. We will use causal forests and expert knowledge to estimate indiviual treatment effects, and identify the groups of patients who benefit the most from these surgeries.This 3-year project will be undertaken by the CAUSALab, at the Unit of Epidemiology, Karolinska Institutet. The team of co-applicants, have extensive experience in using observational data to obtain causal inferences, particularly in the field of cardiovascular disease. A postdoctoral researcher will be hired to work full-time on this project.