FIC-UAI Publication Database -- Query Results

Mondschein, S., Yankovic, N., & Matus, O. (2021). Age-dependent optimal policies for hepatitis C virus treatment. Int. Trans. Oper. Res., 28(6), 3303–3329. Abstract: In recent years, highly effective treatments for hepatitis C virus (HCV) have become available. However, high prices of new treatments call for a careful policy evaluation when considering economic constraints. Although the current medical advice is to administer the new therapies to all patients, economic and capacity constraints require an efficient allocation of these scarce resources. We use stochastic dynamic programming to determine the optimal policy for prescribing the new treatment based on the age and disease progression of the patient. We show that, in a simplified version of the model, new drugs should be administered to patients at a given level of fibrosis if they are within prespecified age limits; otherwise, a conservative approach of closely monitoring the evolution of the patient should be followed. We use a cohort of Spanish patients to study the optimal policy regarding costs and health indicators. For this purpose, we compare the performance of the optimal policy against a liberal policy of treating all sick patients. In this analysis, we achieve similar results in terms of the number of transplants, HCV-related deaths, and quality of adjusted life years, with a significant reduction in overall expenditure. Furthermore, the budget required during the first year of implementation when using the proposed methodology is only 12% of that when administering the treatment to all patients at once. Finally, we propose a method to prioritize patients when there is a shortage (surplus) in the annual budget constraint and, therefore, some recommended treatments must be postponed (added). Keywords: dynamic programming; public health; hepatitis C virus show.php?record=1041
Mondschein, S., Yankovic, N., & Matus, O. (2021). The Challenges of Administering a New Treatment: The Case of Direct -Acting Antivirals for Hepatitis C Virus. Public Health, 190, 116–122. Abstract: Objectives: We develop a patient prioritization scheme for treating patients infected with hepatitis C virus (HCV) and study under which scenarios it outperforms the current practices in Spain and Chile. Study design: We use simulation to evaluate the performance of prioritization rules under two HCV patient cohorts, constructed using secondary data of public records from Chile and Spain, during 2015-2016. Methods: We use the results of a mathematical model, which determines individual optimal HCV treatment policies as an input for constructing a patient prioritization rule, when limited resources are present. The prioritization is based on marginal analysis on cost increases and health-outcome gains. We construct the Chilean and Spanish case studies and used Monte Carlo simulation to evaluate the performance of our methodology in these two scenarios. Results: The resulting prioritizations for the Chilean and Spanish patients are similar, despite the significant differences of both countries, in terms of epidemiological profiles and cost structures. Furthermore, when resources are scarce compared with the number of patients in need of the new drug, our prioritization significantly outperforms current practices of treating sicker patients first, both in terms of cost and healthcare indicators: for the Chilean case, we have an increase in the quality-adjusted life years (QALYs) of 0.83 with a cost reduction of 8176 euros per patient, with a budget covering 2.5% of the patients in the cohort. This difference slowly decreases when increasing the available resources, converging to the performance indicators obtained when all patients are treated immediately: for the Spanish case, we have a decrease in the QALYs of 0.17 with a cost reduction of 1134 euros per patient, with a budget covering 20% of the patients in the cohort. Conclusion: Decision science can provide useful analytical tools for designing efficient public policies that can excel in terms of quantitative health performance indicators. Keywords: Hepatitis C virus; Patient prioritization; Heuristics show.php?record=1114

Abstract: In recent years, highly effective treatments for hepatitis C virus (HCV) have become available. However, high prices of new treatments call for a careful policy evaluation when considering economic constraints. Although the current medical advice is to administer the new therapies to all patients, economic and capacity constraints require an efficient allocation of these scarce resources. We use stochastic dynamic programming to determine the optimal policy for prescribing the new treatment based on the age and disease progression of the patient. We show that, in a simplified version of the model, new drugs should be administered to patients at a given level of fibrosis if they are within prespecified age limits; otherwise, a conservative approach of closely monitoring the evolution of the patient should be followed. We use a cohort of Spanish patients to study the optimal policy regarding costs and health indicators. For this purpose, we compare the performance of the optimal policy against a liberal policy of treating all sick patients. In this analysis, we achieve similar results in terms of the number of transplants, HCV-related deaths, and quality of adjusted life years, with a significant reduction in overall expenditure. Furthermore, the budget required during the first year of implementation when using the proposed methodology is only 12% of that when administering the treatment to all patients at once. Finally, we propose a method to prioritize patients when there is a shortage (surplus) in the annual budget constraint and, therefore, some recommended treatments must be postponed (added).

Abstract: Objectives: We develop a patient prioritization scheme for treating patients infected with hepatitis C virus (HCV) and study under which scenarios it outperforms the current practices in Spain and Chile.
Study design: We use simulation to evaluate the performance of prioritization rules under two HCV patient cohorts, constructed using secondary data of public records from Chile and Spain, during 2015-2016.
Methods: We use the results of a mathematical model, which determines individual optimal HCV treatment policies as an input for constructing a patient prioritization rule, when limited resources are present. The prioritization is based on marginal analysis on cost increases and health-outcome gains. We construct the Chilean and Spanish case studies and used Monte Carlo simulation to evaluate the performance of our methodology in these two scenarios.
Results: The resulting prioritizations for the Chilean and Spanish patients are similar, despite the significant differences of both countries, in terms of epidemiological profiles and cost structures. Furthermore, when resources are scarce compared with the number of patients in need of the new drug, our prioritization significantly outperforms current practices of treating sicker patients first, both in terms of cost and healthcare indicators: for the Chilean case, we have an increase in the quality-adjusted life years (QALYs) of 0.83 with a cost reduction of 8176 euros per patient, with a budget covering 2.5% of the patients in the cohort. This difference slowly decreases when increasing the available resources, converging to the performance indicators obtained when all patients are treated immediately: for the Spanish case, we have a decrease in the QALYs of 0.17 with a cost reduction of 1134 euros per patient, with a budget covering 20% of the patients in the cohort.
Conclusion: Decision science can provide useful analytical tools for designing efficient public policies that can excel in terms of quantitative health performance indicators.

Keywords: Hepatitis C virus; Patient prioritization; Heuristics

show.php?record=1114