The involvement of stakeholders in computational model development to support decision-making in public health: a...

The involvement of stakeholders in the development of computational models to support public health policy decisions has gained popularity, as it can enhance the quality of these models. This scoping review aims to synthesise the current body of literature on how stakeholders have been involved in the development of computational models in public health.

A literature search was performed in MEDLINE, APA PsycInfo and Scopus, from inception to October 2023. In addition, supplementary articles were identified through snowballing and a Google Scholar search. We searched for studies that developed computational models, involved stakeholders in the model development process, and focused on public health domains. Two reviewers independently screened abstracts and full-text articles.

A total of 3438 titles and abstracts were screened, 40 articles underwent full-text review, and 17 articles were included in this review, presenting information on 11 unique computational modelling studies. The studies spanned a diverse range of nine public health domains, including areas such as alcohol use and suicide prevention, with most studies developing system dynamics models. A key rationale for adopting a participatory approach was to ensure that the model accurately reflected the context-specific characteristics of the setting being modelled. Various stakeholder groups were involved, predominantly policy-makers, researchers and community representatives. During the development phases of problem mapping, model conceptualisation, and model validation, stakeholders were mostly engaged through participatory workshops. Studies reported that a major strength of the participatory approach was to facilitate transparent consensus-building processes during model development, while the management of stakeholder engagement timelines was experienced a major challenge.

Involving stakeholders in the development of computational models can improve the accurate representation of the model's context, thereby enhancing the quality and applicability of these models to support decision-making processes. However, there are limited computational modelling studies available in the literature that provide detailed elaboration on their methods of stakeholder engagement. There is a need for improved reporting and evaluation of the impact of participatory modelling processes, as well as more guidance on effective stakeholder involvement, to fully realise the potential of participatory modelling in public health.