Simulation model of disease incidence driven by diagnostic activity

Area of Science:

• Oncology
• Epidemiology
• Health Services Research

Background:

• Early detection and treatment of chronic diseases like prostate cancer impact incidence, overtreatment, and mortality.
• Existing simulation models often rely on assumptions about disease natural history and require extensive data calibration.

Purpose of the Study:

• To develop a novel simulation model for chronic diseases that emulates real-life scenarios.
• To use a proxy for diagnostic activity, bypassing explicit modeling of disease natural history and clinical test properties.

Main Methods:

• Applied a new simulation model to Swedish nationwide population-based prostate cancer data.
• Evaluated the model's performance in reconstructing observed incidence and mortality rates.
• Predicted prostate cancer diagnoses under varying diagnostic activity levels from 2017 to 2060.

Main Results:

• The model accurately reconstructed observed prostate cancer incidence and mortality.
• High diagnostic activity is predicted to significantly increase diagnoses of lower-risk prostate cancer.
• Conversely, high diagnostic activity is associated with fewer metastatic cases and reduced overall prostate cancer mortality.

Conclusions:

• The developed model effectively emulates real-life chronic disease scenarios, specifically for prostate cancer.
• It can predict outcomes, inform decision-making regarding diagnostic strategies, and assess the impact of diagnostic activity on overdiagnosis and mortality.
• Long-term high diagnostic activity shows a favorable impact on prostate cancer outcomes.