Background: The prevention and treatment of coronary heart disease (CHD) is complex. A variety of models have therefore been developed to try and explain past trends and predict future possibilities. The aim of this systematic review was to evaluate the strengths and limitations of existing CHD policy models.
Methods: A search strategy was developed, piloted and run in MEDLINE and EMBASE electronic databases, supplemented by manually searching reference lists of relevant articles and reviews. Two reviewers independently checked the papers for inclusion and appraisal. All CHD modelling studies were included which addressed a defined population and reported on one or more key outcomes (deaths prevented, life years gained, mortality, incidence, prevalence, disability or cost of treatment).
Results: In total, 75 articles describing 42 models were included; 12 (29%) of the 42 models were micro-simulation, 8 (19%) cell-based, and 8 (19%) life table analyses, while 14 (33%) used other modelling methods. Outcomes most commonly reported were cost-effectiveness (36%), numbers of deaths prevented (33%), life-years gained (23%) or CHD incidence (23%). Among the 42 models, 29 (69%) included one or more risk factors for primary prevention, while 8 (19%) just considered CHD treatments. Only 5 (12%) were comprehensive, considering both risk factors and treatments. The six best-developed models are summarised in this paper, all are considered in detail in the appendices.
Conclusion: Existing CHD policy models vary widely in their depth, breadth, quality, utility and versatility. Few models have been calibrated against observed data, replicated in different settings or adequately validated. Before being accepted as a policy aid, any CHD model should provide an explicit statement of its aims, assumptions, outputs, strengths and limitations.