In a perspective article published in the Proceedings of the National Academy of Sciences on May 15, SFI External Professor Ross Hammond and Sharin Barkin argue that the health science research could benefit from combining traditional trials with computational models. According to the authors, the synergy between these two methods offers powerful advantages that could enhance the credibility and impact of research in the field.
Traditional randomized controlled trials (RCTs) have long been a standard approach in health sciences for establishing links between treatments and outcomes. While effective, RCTs have limitations such as difficulty in generalizing results, failure to identify mechanisms of action, and reliance on assumptions. However, recent case studies have shown promising results from integrating these methods with agent-based models (ABMs). ABMs are computational tools that simulate interventions in dynamic environments with diverse populations. These models require validation with real-world data from experiments or observations.
Although it is uncommon for health science research to combine RCTs and ABMs, by conducting iterative work that involves both RCTs and ABMs, researchers could gain a better understanding of complex diseases and optimize the use of limited resources to have a greater impact on health science research. The synergy between these two approaches allows for a more comprehensive analysis of treatment effects while providing insights into mechanisms of action that may be difficult to identify through traditional trials alone. Additionally, this approach allows researchers to test interventions under various conditions and populations, which can help improve their effectiveness and efficiency.
According to Hammond and Barkin, combining RCTs with ABMs has several advantages over traditional trial methods. Firstly, it allows for more flexibility in designing trials as researchers can adjust variables based on simulations conducted by ABMs. Secondly, it enables researchers to test interventions under different scenarios and conditions without relying solely on historical data or assumptions about future events.
Moreover, this approach provides an opportunity for collaboration between different disciplines as it requires expertise in both computer modeling and experimental design. This collaboration can lead to new discoveries and innovations in health sciences as researchers can combine their unique perspectives to develop novel approaches to complex problems.
In conclusion, Hammond and Barkin suggest that combining RCTs with ABMs could significantly enhance the credibility and impact of health science research. While this approach requires significant investment in technology and expertise, its potential benefits make it worth considering for future research projects.
Overall, this perspective article highlights an innovative approach that combines traditional randomized controlled trials with agent-based models to provide a more comprehensive analysis of treatment effects while offering insights into mechanisms of action that may be difficult to identify through traditional trial methods alone. By adopting this approach, we can improve our understanding of complex diseases while optimizing the use of limited resources towards achieving greater impact on health science research.
