Table 3 Disease prevention policy models
From: Policy models for preventative interventions in cardiometabolic diseases: a systematic review
DYNAMO-HIA | CVD Policy Model | CHD Policy Model | Impact CHD | CVD-Predict | Scottish Policy model | SPHR Diabetes Model | |
|---|---|---|---|---|---|---|---|
Scope | NCDs (non-communicable diseases) including CVD, diabetes, and risk factors | CVD and related risk factors, focusing on prevention and treatment strategies | CVD and related risk factors, focusing on prevention and treatment strategies | CHD and CVD interventions, evaluating their effectiveness | CVD with a focus on prediction and risk stratification for better preventive measures | Public health with a specific focus on CVD and associated risk factors in Scotland | Diabetes and related risk factors, focusing on prevention, management, and health outcomes |
Applicability | Primarily European countries, but adaptable globally | Primarily used in the US | Primarily used in the US | Applicable globally with regional adaptations | Applicable globally, with a focus on predictive analytics | Primarily used in Scotland | Primarily used in the UK |
Data sources | European health surveys, epidemiological studies, and literature | National health surveys, clinical trials, epidemiological studies | National health surveys, clinical trials, epidemiological studies | National health surveys, clinical trials, epidemiological studies | National health surveys, clinical trials, epidemiological studies | Scottish health surveys, hospital records, national statistics | National health surveys, clinical trials, epidemiological studies |
Outcome of interests | Estimates incidence, prevalence, mortality, QALY health impact, under various policy scenarios | Estimates incidence, prevalence, mortality, and healthcare costs, cost -effectiveness | Estimates incidence, prevalence, mortality, QALY, health disparities healthcare costs of CHD and stroke | Estimates incidence, mortality, hospital admissions, cost-effectiveness | Estimates incidence, risk prediction, mortality, and health care costs, health outcomes, cost -effectiveness | Estimates incidence, mortality, hospital admissions, QALE, cost-effectiveness | Estimates incidence, prevalence, mortality, QALY, cost-effectiveness |
Key strengths | Comprehensive modelling of individual and population-level effects; integration of multiple risk factors and interventions for a nuanced analysis across health outcomes | Robust framework for evaluating interventions at a population level; flexible to include various types of interventions; extensive validation with US data | Extensive validation with US data; comprehensive risk factor integration | Comprehensive evaluation of interventions; focus on real-world applicability; extensive data sources | High granularity of individual risk prediction; ability to incorporate large datasets and update predictions with real-time data | Robust dataset specific to Scotland; focus on real-world applicability and policy impact; capable of addressing health inequalities and informing equitable policy decisions | Focus on diabetes-specific interventions and outcomes; ability to assess a wide range of potential interventions and their population-level impacts |
Key weaknesses | Complexity in adapting to non-European contexts– Requires extensive data input | Can be complex to adapt to new populations or to integrate novel interventions without substantial effort and data | Requires extensive and high-quality data for accurate projections; complexity of model may limit its accessibility for non-specialists | May not account for all complex interactions between risk factors and interventions; data limitations can affect accuracy | Requires access to high-quality, comprehensive health records; model accuracy can be affected by missing or inaccurate data | Limited to the Scottish population, which may limit generalisability to other regions; data limitations outside of Scotland may affect model accuracy | Complexity and data requirements can limit accessibility for some users; relies on accurate input data for precise predictions |