The use of decision models in mental health economic evaluation: challenges and opportunities

A number of reasons have been suggested for the increased use of decision-analytic modelling in assessing the cost-effectiveness of mental health interventions: (1) the need to rapidly assess the value-for-money of new pharmaceuticals; (2) the short time horizon of clinical trials in mental health; (3) the lack of head-to-head comparisons between alternative treatment strategies; and (4) the necessity to generalise the study results. As simulation exercises become more widely accepted and used to inform policy making decisions, a trend towards more sophisticated modelling structures is starting to emerge. This paper undertakes a structured critical appraisal of cost-effectiveness modelling studies, using published examples relating to available drug treatments for schizophrenia and Alzheimer's disease (AD). The models identified display some methodological heterogeneity in terms of the choice of comparator, costing methodology, timeframe and perspective of the analysis, target population and model structure. The use of 'state-of-the-art' techniques, such as value-of-information analysis and probabilistic-sensitivity analysis has been somewhat limited. Health service researchers need to make sure that the best available techniques are employed to maximise the use of the information already available.