The aim of the Mt Hood challenges is to advance transparency, comparability, and methodological development across diabetes simulation models, while addressing emerging research and policy questions. A brief summary of each proposed challenge is provided below.
 

1. Reference Simulation Update 
 

This challenge builds on previous Mount Hood reference simulations and asks groups to replicate a standardised type 2 diabetes case using updated model versions. The objective is to assess how model outputs evolve over time due to structural changes, parameter updates, and methodological refinements, enabling comparison with prior registry results. We will again replicate the Type 2 diabetes reference case (hence, it is only necessary for modelling groups that have changed their models to redo it). We will also explore adding a pre-diabetes or obesity model challenge.

 
2. Comorbidity and Multimorbidity Challenge (Diabetes and Dementia)

This challenge focuses on incorporating comorbidities—particularly dementia—into diabetes models. Participants will simulate older populations with and without cognitive decline to evaluate impacts on outcomes such as QALYs, life expectancy, and costs. A key goal is to understand how models handle interactions between diabetes, aging, adherence, and quality of life.
 

This is likely to require many modelling groups to add a dementia model, which is now possible as a dementia risk equation has recently been published by the DOMUS model group. This challenge may also take advantage of the likely involvement of the International Pharmaco-Economic Collaboration on Alzheimer's Disease, which has undertaken very similar challenges in Alzheimer's disease: 

​

3. Synthetic Data Challenge 

​

This challenge is being undertaken in collaboration with the REDDIE project group.

Using synthetic datasets derived from the Swedish National Diabetes Register, this challenge explores how models perform under varying levels of data richness. Participants will simulate cohorts using:
• Aggregate (mean) characteristics
• Independent distributions
• Correlated risk factors
• Full pseudo-patient-level data
• A longitudinal component will compare model predictions against observed outcomes, providing insight into model validity and calibration
 

Although the synthetic datasets created in REDDIE do not contain any real patient data, they remain the property of the REDDIE collaboration and are subject to appropriate governance restrictions. Therefore, in order to share data, we will require participants to enter into a simple agreement restricting use of the data to the Mount Hood challenges.
 

4. Presymptomatic Screening in Type 1 Diabetes

This challenge evaluates the long-term clinical and economic value of screening for early-stage (presymptomatic) type 1 diabetes. Models will assess whether screening programs are cost-effective compared with no screening, incorporating disease progression, treatment effects, and downstream complications. This reflects growing policy interest in early detection and prevention strategies.
 

A group from RTI have developed the challenge, but we will need additional modelling groups to become involved for this to become a viable Mt Hood challenge.
 

If you would like more information or are interested in becoming involved in any of the challenges, please let us know via sending an email to mthood2016[at]gmail.com.

​

​