UC Berkeley

The Causal Roadmap is a formal framework for causal and statistical inference that supports clear specification of the causal question, interpretable and transparent statement of required causal assumptions, robust inference, and optimal precision. The Roadmap is thus particularly well-suited to evaluating longitudinal causal effects using large scale registries; however, application of the Roadmap to registry data also introduces particular challenges. In this work, Ioutline the use of simulations for estimation specification in real-world data problems. I further motivate this through two detailed case studies of the Causal Roadmap applied to longitudinal Danish National Registry data to evaluate the comparative effectiveness of second-line diabetes drugs on dementia and cardiovascular risk, respectively. I provide practical guidance on the implementation of the Roadmap using registry data, and highlight how rare exposures and outcomes over long-term follow up can raise challenges for flexible and robust estimators, even in the context of the large sample sizes provided by the registry. Moreover, I demonstrate how simulations can be used to help address these challenges by supporting careful estimator pre-specification. To protect research integrity, it is essential that the algorithm for statistical estimation and inference be pre-specified prior to conducting any effectiveness analyses. However, it is often unclear which algorithm will perform optimally for the real-data application. Here, I call for the use of simulations that realistically reflect the application, including key characteristics such as strong confounding and dependent or missing outcomes, to objectively compare candidate estimators and facilitate flexible yet full specification of the statistical analysis plan. Such simulations are informed by the Causal Roadmap and conducted after data collection but prior to effect estimation.

In Chapter 1, I outline the benefits simulations for pre-specification through two worked examples. In Chapter 2, I examine the effect of glucagon-like receptor-1 agonists (GLP1) on dementia onset, and utilize simulations to rigorously pre-specify our estimation selection. In Chapter 3, I walk through a guided simulation study using the example of GLP1 and adverse cardiac events, highlighting both strengths of this approach and areas of further research. Overall, this work shows how realistic simulations empower us to flexibly pre-specify an estimation approach and thus to improve the rigor and reproducibility of our research.