Guidelines on reporting and assessing dynamic mathematical models of infectious diseases: a scoping review

Abstract

Background

Mathematical models are valuable tools for guiding public health policy decisions to combat the spread of infectious diseases. Nevertheless, a lack of appropriate quality assessment tools and reporting guidelines hinders the comprehensibility, transparency, and credibility of infectious disease modelling studies and the ability to assess their quality. In a first step towards addressing the need for reporting guidelines and quality assessment tools specific to infectious disease modelling, this scoping review identified common themes in existing reporting and quality assessment guidance for infectious disease modelling studies and adjacent fields. Methods

We conducted temporally-unrestricted searches on Medline (via Ovid), Web of Science, medRxiv, and bioRxiv in January 2024 to find articles that provide guidance on writing or assessing modelling studies within infectious disease modelling and adjacent fields including but not limited to healthcare and, more specifically, health economics. Articles were double-screened for eligibility via title-and-abstract screening and full-text screening. Recommendations made by eligible articles were classified into 31 subdimensions which were categorised into seven overarching dimensions (1. applicability; 2. model structure; 3. parameterisation and calibration; 4. validity; 5. uncertainty; 6. interpretation; 7. reproducibility, clarity, and transparency). We followed the PRISMA extension for reporting scoping reviews. Results

Our final review included 53 articles. All dimensions except for interpretation were covered by most articles (81%-98%). However, we found substantial heterogeneity in the frequency with which subdimensions were addressed (11%-96%). Subdimensions pertaining to parameter uncertainty and transparency about parameter values were mentioned in most articles (91%-96%); conversely, discussions about auxiliary publication details and software implementation were covered less frequently (11%-23%). Conclusions

This review shows that many recommendations made by reporting guidelines and quality assessment tools have thematic similarities and address common topics that are also relevant to infectious disease modelling. These identified themes and recommendations can be used as a starting point to inform the development of standardised guidelines for infectious disease modelling.