Outcome measurements in economic evaluations of drug misuse interventions

This thesis critically evaluates the measurement of outcomes in economic evaluations of drug misuse interventions. Three different aspects of measuring outcomes are examined: one focusing on non-monetary outcomes at the individual patient level; one focusing on monetary outcomes within studies using individual patient level data; and one focusing on long-term outcomes, both monetary and non-monetary. The many limitations of measuring these outcomes in existing economic evaluations of drug misuse interventions are exposed and the problems with conducting such studies are identified. The importance of this thesis is thus in providing an overview and methodological critique of the extant economic evaluations of drug misuse interventions. In addition a decision analytic model for a drug testing in schools programme is developed to illustrate how the limitations highlighted in the methodological critique might be addressed by future research. The findings of the thesis reveal the problems with using EQ-5D as a generic outcome measure for economic evaluations of drug misuse interventions, as is recommended by NICE in the UK. The nature of drug misuse problems requires that a wide range of different measures, including drug misuse specific measures, must be taken into account when evaluating drug misuse interventions. Similarly, the limitations with existing studies that attempt to estimate the monetary outcome of drug misuse interventions are exposed, as many studies fail to take into account all of the costs that will determine the monetary impact of an intervention for society. The thesis stresses the complexity of drug misuse and the need to measure the long-term outcomes of interventions, which may be best achieved by developing drug misuse modelling studies. However, these models are then revealed to be themselves limited, in part due to the lack of real-world data available to set their parameters.