Purpose: To develop a value of information model to assess the potential public health impact of any given clinical trial in neurology or neurosurgery
Method: A conceptual framework for using value of information (VOI) exercise to set research priorities was developed by engaging an expert advisory panel at the National Institutes of Neurological Disorders and Stroke. Elements of a simple model structure that would be applicable across many conditions and interventions were identified in a group exercise with the expert panel. Draft specifications organized around decision tables and analytical frameworks were developed for individual and population level models. A general VOI model applicable to any trial in neurology was developed using decision analysis software (Tree Age, Williamstown, MA). The model was designed to be deployed in a step wise process starting with validation of the model using test cases of previous clinical trials, with feedback from expert panel incorporated into the model at each stage.
Result: We have developed a mathematical model to estimate the potential public health impact of a clinical trial using a modified VOI approach. The impact of a particular trial may be evaluated by taking into account the burden of illness (incidence, prevalence, and the size of the population whose treatment might be effected by the trial results),expert estimates of the efficacy of the trial and the variety of possible results of the trial. Inputs to the model are gathered from data submitted by the trial applicant supplemented with data from literature as well as expert opinion. Health and cost outcomes associated with the decision to either perform the trial or not may be calculated in terms of a range of measures including number of person years affected by the results of the trial, QALYs, total cost, and incremental cost effectiveness ratio (ICER). The model is primarily facilitative to allow consideration of the impact of funding decision in terms of standard health and cost measures
Conclusion: Value of information analysis can serve as an important tool in estimating the public health impact of proposed clinical trials in neurology and may be an effective decision aid in setting research priorities.
Candidate for the Lee B. Lusted Student Prize Competition