A Forensic Evaluation of the National Emphysema Treatment Trial Using the Expected Value of Information Approach

Abstract

Expected value of information (EVI) analyses allow researchers to estimate the returns to conducting research. We used EVI techniques to estimate the value of the National Emphysema Treatment Trial (NETT), a multicenter randomized trial of lung-volume-reduction surgery (LVRS) versus medical therapy (MT) for patients with severe emphysema, then compared that result to the trial cost. We gathered information on costs and benefits of LVRS and MT before the trial and the costs of conducting the NETT, and compared these data with the results of the cost-effectiveness analysis conducted alongside the trial. We used 2 thresholds to represent the societal value of a quality-adjusted life year (QALY): USD 50,000 and USD100,000. The cost effectiveness of LVRS versus MT using historical (nontrial) information was USD 305,000/QALY. Based on these data and the threshold incremental cost-effectiveness ratio values, the expected value of perfect information was USD 46 million and USD 670 million for thresholds USD 50,000 and USD 100,000 per QALY, respectively. The NETT was powered for 1,250 patients in each arm; ultimately approximately 600 patients in each arm were recruited. With 1,250 patients per arm, the expected value of sample information was USD 660 million for the threshold of USD100,000. The actual cost of the NETT was approximately USD 60 million. The expected net benefit of sampling was USD 600 million. Given the difference between the cost of the trial and the economic benefits of the information, the EVI analyses suggest that federal investment in the NETT trial represented good value for money.