Estimating health system opportunity costs: the role of non-linearities and inefficiency

Hernandez‑Villafuerte et al. Cost Effectiveness and Resource Allocation (2022) 20:56 https://doi.org/10.1186/s12962-022-00391-y Cost Effectiveness and Resource Allocation Open Access RESEARCH Estimating health system opportunity costs: the role of non‑linearities and inefficiency Karla Hernandez‑Villafuerte1, Bernarda Zamora2, Yan Feng3, David Parkin4, Nancy Devlin5 and Adrian Towse4* Abstract Background: Empirical estimates of health system opportunity costs have been suggested as a basis for the costeffectiveness threshold to use in Health Technology Assessment. Econometric methods have been used to estimate these in several countries based on data on spending and mortality. This study examines empirical evidence on four issues: non-linearity of the relationship between spending and mortality; the inclusion of outcomes other than mortality; variation in the efficiency with which expenditures generate health outcomes; and the relationship among efficiency, mortality rates and outcome elasticities. Methods: Quantile Regression is used to examine non-linearities in the relationship between mortality and health expenditures along the mortality distribution. Data Envelopment Analysis extends the approach, using multiple meas‑ ures of health outcomes to measure efficiency. These are applied to health expenditure data from 151 geographical units (Primary Care Trusts) of the National Health Service in England, across eight different clinical areas (Programme Budget Categories), for 3 fiscal years from 2010/11 to 2012/13. Results: The results suggest differences in efficiency levels across geographical units and clinical areas as to how health resources generate outcomes, which indicates the capacity to adjust to a decrease in health expenditure without affecting health outcomes. Moreover, efficient units have lower absolute levels of mortality elasticity to health expenditure than inefficient ones. Conclusions: The policy of adopting thresholds based on estimates of a single system-wide cost-effectiveness threshold assumes a relationship between expenditure and health outcomes that generates an opportunity cost esti‑ mate which applies to the whole system. Our evidence of variations in that relationship and therefore in opportunity costs suggests that adopting a single threshold may exacerbate the efficiency and equity concerns that such thresh‑ olds are designed to counter. In most health care systems, many decisions about provision are not made centrally. Our analytical approach to understanding variability in opportunity cost can help policy makers target efficiency improve‑ ments and set realistic targets for local and clinical area health improvements from increased expenditure. Keywords: Opportunity cost, Cost-effectiveness threshold, Quantile regression, DEA, Outcome elasticities, English NHS *Correspondence: 4 Office of Health Economics, London, UK Full list of author information is available at the end of the article Background Providing health care has an opportunity cost. In health care systems with a fixed budget this is the health benefits forgone from other health care that could have been provided with the resources used. This should be fundamental to many health policy considerations, including efficiency improvement and sociodemographic and geographical equity. Recent work quantifying health system © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Hernandez‑Villafuerte et al. Cost Effectiveness and Resource Allocation opportunity costs [1–10] has focussed on the adoption of new technologies and their displacement impact on other health care, usually expressed as the search for a ‘threshold’ against which Health Technology Assessment (HTA) agencies can judge cost-effectiveness. Martin et al. [1, 2] developed methods for examining the impact of health expenditure on health which were used by Claxton et al. [3] to estimate an opportunity cost based threshold for the NHS in England. This was updated by Lomas et al. [4] and has recently been revisited by Martin et al. [5]. Estimates have now been published for several countries [6–10]. Although these studies use slightly different approaches, most follow Claxton et al. in applying econometric methods to health system data to examine the relationship between health care expenditures and health outcomes from variations observed across health care ‘programmes’ and administrative units (health care payers or commissioners). They estimate the average relationship between spending and outcomes, based on mortality converted to Quality Adjusted Life Years (QALYs). The England NHS studies calculate QALYs as an adjustment to mortality figures, rather than measuring morbidity as a separately sourced category of health gain. This paper addresses three issues that have been raised about these methods [11]: first, linear regression models may not correctly specify the relationship between expenditure and mortality; secondly, using mortality (or QALYs) as the only health care outcome may not fully reflect health system priorities; and thirdly, variations in the efficiency with which health is produced may impact on the observed relationship of inputs and outcomes. We applied Quantile Regression (QR) and Data Envelopment Analysis (DEA) to English NHS data. QR permits estimation of non-linear relationships, examining point estimates of the expenditure/mortality relationship at different parts of the mortality distribution to show differences across PCTs with low to high mortality rates. DEA permits inclusion of non-mortality health outcomes aligned with health system priorities and enables measurement of the variations in efficiency. Use of these methods allowed us to address a fourth issue: the relationship among efficiency, mortality rates and outcome elasticities. Methods Quantile regression QR was used to explore differences across the expenditure/outcome relationship as between 151 local commissi (...truncated)