Eva A Enns1, Lauren E Cipriano2, Cyrena T Simons3, Chung Yin Kong4,5. 1. University of Minnesota School of Public Health, Division of Health Policy and Management, Minneapolis, MN (EAE) 2. Ivey Business School, University of Western Ontario, London, ON, Canada (LEC) 3. Stanford University Medical School, Stanford, CA (CTS) 4. Institute for Technology Assessment, Massachusetts General Hospital, Boston, MA (CYK) 5. Harvard Medical School, Boston, MA (CYK)
Abstract
BACKGROUND: To identify best-fitting input sets using model calibration, individual calibration target fits are often combined into a single goodness-of-fit (GOF) measure using a set of weights. Decisions in the calibration process, such as which weights to use, influence which sets of model inputs are identified as best-fitting, potentially leading to different health economic conclusions. We present an alternative approach to identifying best-fitting input sets based on the concept of Pareto-optimality. A set of model inputs is on the Pareto frontier if no other input set simultaneously fits all calibration targets as well or better. METHODS: We demonstrate the Pareto frontier approach in the calibration of 2 models: a simple, illustrative Markov model and a previously published cost-effectiveness model of transcatheter aortic valve replacement (TAVR). For each model, we compare the input sets on the Pareto frontier to an equal number of best-fitting input sets according to 2 possible weighted-sum GOF scoring systems, and we compare the health economic conclusions arising from these different definitions of best-fitting. RESULTS: For the simple model, outcomes evaluated over the best-fitting input sets according to the 2 weighted-sum GOF schemes were virtually nonoverlapping on the cost-effectiveness plane and resulted in very different incremental cost-effectiveness ratios ($79,300 [95% CI 72,500-87,600] v. $139,700 [95% CI 79,900-182,800] per quality-adjusted life-year [QALY] gained). Input sets on the Pareto frontier spanned both regions ($79,000 [95% CI 64,900-156,200] per QALY gained). The TAVR model yielded similar results. CONCLUSIONS: Choices in generating a summary GOF score may result in different health economic conclusions. The Pareto frontier approach eliminates the need to make these choices by using an intuitive and transparent notion of optimality as the basis for identifying best-fitting input sets.
BACKGROUND: To identify best-fitting input sets using model calibration, individual calibration target fits are often combined into a single goodness-of-fit (GOF) measure using a set of weights. Decisions in the calibration process, such as which weights to use, influence which sets of model inputs are identified as best-fitting, potentially leading to different health economic conclusions. We present an alternative approach to identifying best-fitting input sets based on the concept of Pareto-optimality. A set of model inputs is on the Pareto frontier if no other input set simultaneously fits all calibration targets as well or better. METHODS: We demonstrate the Pareto frontier approach in the calibration of 2 models: a simple, illustrative Markov model and a previously published cost-effectiveness model of transcatheter aortic valve replacement (TAVR). For each model, we compare the input sets on the Pareto frontier to an equal number of best-fitting input sets according to 2 possible weighted-sum GOF scoring systems, and we compare the health economic conclusions arising from these different definitions of best-fitting. RESULTS: For the simple model, outcomes evaluated over the best-fitting input sets according to the 2 weighted-sum GOF schemes were virtually nonoverlapping on the cost-effectiveness plane and resulted in very different incremental cost-effectiveness ratios ($79,300 [95% CI 72,500-87,600] v. $139,700 [95% CI 79,900-182,800] per quality-adjusted life-year [QALY] gained). Input sets on the Pareto frontier spanned both regions ($79,000 [95% CI 64,900-156,200] per QALY gained). The TAVR model yielded similar results. CONCLUSIONS: Choices in generating a summary GOF score may result in different health economic conclusions. The Pareto frontier approach eliminates the need to make these choices by using an intuitive and transparent notion of optimality as the basis for identifying best-fitting input sets.
Authors: Martin B Leon; Craig R Smith; Michael Mack; D Craig Miller; Jeffrey W Moses; Lars G Svensson; E Murat Tuzcu; John G Webb; Gregory P Fontana; Raj R Makkar; David L Brown; Peter C Block; Robert A Guyton; Augusto D Pichard; Joseph E Bavaria; Howard C Herrmann; Pamela S Douglas; John L Petersen; Jodi J Akin; William N Anderson; Duolao Wang; Stuart Pocock Journal: N Engl J Med Date: 2010-09-22 Impact factor: 91.245
Authors: Tazio Vanni; Jonathan Karnon; Jason Madan; Richard G White; W John Edmunds; Anna M Foss; Rosa Legood Journal: Pharmacoeconomics Date: 2011-01 Impact factor: 4.981
Authors: Douglas C A Taylor; Vivek Pawar; Denise Kruzikas; Kristen E Gilmore; Ankur Pandya; Rowan Iskandar; Milton C Weinstein Journal: Pharmacoeconomics Date: 2010 Impact factor: 4.981
Authors: Cyrena T Simons; Lauren E Cipriano; Rashmee U Shah; Alan M Garber; Douglas K Owens; Mark A Hlatky Journal: Circ Cardiovasc Qual Outcomes Date: 2013-07-09
Authors: Jane J Kim; Karen M Kuntz; Natasha K Stout; Salaheddin Mahmud; Luisa L Villa; Eduardo L Franco; Sue J Goldie Journal: Am J Epidemiol Date: 2007-05-25 Impact factor: 4.897
Authors: Linwei Wang; Emanuel Krebs; Jeong E Min; W Christopher Mathews; Ank Nijhawan; Charurut Somboonwit; Judith A Aberg; Richard D Moore; Kelly A Gebo; Bohdan Nosyk Journal: Lancet HIV Date: 2019-07-11 Impact factor: 12.767
Authors: Fernando Alarid-Escudero; Eline M Krijkamp; Petros Pechlivanoglou; Hawre Jalal; Szu-Yu Zoe Kao; Alan Yang; Eva A Enns Journal: Pharmacoeconomics Date: 2019-11 Impact factor: 4.981
Authors: Fernando Alarid-Escudero; Richard F MacLehose; Yadira Peralta; Karen M Kuntz; Eva A Enns Journal: Med Decis Making Date: 2018-10 Impact factor: 2.583