Enabling new types of questions in evidence synthesis

• Applicant: Dr. Adriani Nikolakopoulou
• Subject Area: Epidemiology and Medical Biometry/Statistics
• Term: since 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 499370385

Background: The PICO scheme has been used for over two decades to formalize the research question of a systematic review, by helping to specify the population, interventions, comparators, and outcomes of interest. Many research questions asked today, however, are broader and more complex, and thus do not fit into the PICO scheme. For example, recent literature on lowering low-density lipoprotein cholesterol (LDL-C) might in principle allow to answer complex, clinically relevant questions, such as on the benefit-risk profile of separate medications, how each medication works for different populations, and whether a combination of treatments works better than individual treatments do on their own. While some methods have been suggested that could potentially enable evidence synthesis for parts of such questions, available methodological developments are rather disjoint and not well aligned, hindering their combined use in real world examples. Aim: The aim of this project is to broaden the PICO scheme and corresponding statistical methods by creating, evaluating and establishing a holistic framework for synthesizing, assessing and presenting evidence from systematic reviews.Methods: Developments will be stimulated by a range of research questions in several clinical areas. Our starting point will be existing methods to synthesize evidence with broad definitions of populations, treatment comparisons and outcomes, such as network meta-analysis, population adjustment methods, multivariate meta-analysis, and multiple criteria decision analysis. We will integrate these methods, extend them and enrich them with novel approaches to create a framework that will allow researchers to ask new types of questions in evidence synthesis. As our new generalized framework will also uncover disagreements between different sources of evidence, we will also create tools to explore and account for such disagreements. We will develop guidance and tools for presenting results of the framework in a coherent and accessible way, from the assumptions underlying the synthesis to presenting the complex output of the analysis in a way relevant to decision-making. Finally, we will develop user-friendly open-source tools (R packages and web-application) for implementing the framework in practice, and we will promote it with an effective dissemination plan based on our collaborations with clinical partners and the methods community.Impact: This project will address an urgent need to extend the PICO scheme, and advance it to the new era of multiple populations, treatment comparisons, and outcomes of interest. The output of the project will extend the decision-making arsenal of evidence-based medicine with a much-needed methodological toolkit that will support clinicians, policy makers and patients to make better decisions for a given condition.

• DFG Programme: Independent Junior Research Groups