Which Diseases Should Be Targeted for Prevention?: Flow Disruption Model Using the Disease Network

Abstract Background In public health, disease prevention targets are often decided based on expert opinion and short-term needs, rather than systematic, data-driven criteria. Few studies have proposed how to set disease prevention targets. Therefore, we propose the Flow Disruption Model to identify prevention target diseases using the disease progression network. Methods We first developed the statistically significant disease progression network using South Korea's National Health Insurance (NHIS) claims data (∼50M patients), where disease transitions are represented as directed edges weighted by attributable risk (AR). Using this network, we developed the Flow Disruption Model (FDM) to select prevention priority target diseases. The FDM quantifies how preventive interventions at specific disease-focused on primary and secondary/tertiary prevention by reducing in-/out-transitions-can reduce subsequent risk across the disease network. Priorities were assigned to diseases that most reduced cumulative subsequent risk. The model was applied as an example to identify priority prevention targets for reducing dementia incidence in women in their 70s. Results We identified prevention targets based on 208,869 disease pathways leading to dementia in women in their 70s. In the first round, secondary/tertiary prevention targeted degenerative nervous system diseases, menstrual disorders, and symptomatic mental disorders (50s). In the second round, cerebral palsy (70s) was targeted by secondary/tertiary prevention, and mood disorders (60s) by both. In the final round, polyneuropathies (70s) were targeted by both, and oedema (60s) by primary prevention. Conclusions The FDM identifies high-impact prevention targets for reducing the incidence of a specific disease. By quantifying the impact of prevention strategies, it provides insights to guide early prevention, optimize resource allocation and support evidence-based public health planning. Key messages • We developed the Flow Disruption Model to identify prevention targets where intervention can most effectively reduce the incidence of specific diseases within the population-level disease network. • Our model quantifies the impact of prevention strategies for reducing specific disease incidence, supporting data-driven selection of priority targets for early intervention/public health planning.