Different reasonable methodological choices can lead to vastly different estimates of the economic burden of diseases

Meaningful estimates of the burden of a disease or chronic condition (hereafter referred to as disease) are important for setting policy priorities by governments, non-governmental organisations, and other interested parties. These estimates are also crucial for evaluating the cost-effectiveness of policies and for enabling researchers and other stakeholders to identify the avenues for research with the highest potential impact. Unfortunately, obtaining such estimates is challenging. Moreover, comparable estimates across different diseases or across different populations for the same disease are desirable. However, estimates from different studies are often incomparable because of differences in how the burden is defined or because of different substantive and methodological choices. Landeiro and colleagues1Landeiro F, Harris C, Groves D, et al. The economic burden of cancer, coronary heart disease, dementia and stroke in England in 2018: an evaluation of two cohort studies. Lancet Healthy Longev 2024; published online July 25. https://doi.org/10.1016/S2666-7568(24)00108-9Google Scholar computed the economic burden of four diseases (cancer, coronary heart disease [CHD], dementia, and stroke) in England using consistent methodology and a broad definition of disease burden. This analysis is an important advance that will allow policy makers, researchers, and other stakeholders to assess the absolute and relative burden of these diseases in a meaningful way. The Global Burden of Disease2Murray CJL Findings from the Global Burden of Disease study 2021.Lancet. 2024; 403: 2259-2262Summary Full Text Full Text PDF PubMed Scopus (0) Google Scholar also uses a consistent methodology for estimating the burden of many diseases across countries. However, its methodology focuses only on mortality and morbidity, which are evaluated comprehensively, but does not account for many other costs included in the work by Landeiro and colleagues,1Landeiro F, Harris C, Groves D, et al. The economic burden of cancer, coronary heart disease, dementia and stroke in England in 2018: an evaluation of two cohort studies. Lancet Healthy Longev 2024; published online July 25. https://doi.org/10.1016/S2666-7568(24)00108-9Google Scholar most notably the costs of formal and informal care. Landeiro and colleagues' work illustrates the difficulties with obtaining suitable estimates and the effect of different reasonable methodological choices, which is most apparent when evaluating the cost of informal care. In this work, 1 h of informal care was valued at £7·85, which is the average wage of a (non-senior) home care worker, but 1 h of formal care was valued at £27·00, which is the average cost of hiring a caregiver (including overhead costs).3Curtis LA Burns A Unit costs of health and social care 2018. Project report. University of Kent, 2018Google Scholar Formal care has been valued at the cost that is actually paid (and the amount that would have been saved in the absence of the disease), whereas informal care has been valued as the amount that an informal caregiver would earn for the same task if they were a formal caregiver instead, making this choice of cost valuation a reasonable methodological choice. However, if a household needed to hire a formal caregiver instead of providing the care themselves, they would need to pay £27·00 per hour for caregiving. Therefore, one could also argue that this is the best measure of the cost of informal care. Using this valuation, the total cost of care would be £21·1 billion for cancer (vs the authors' estimate of £18·9 billion), £14·5 billion for coronary heart disease (vs £12·7 billion), £20·6 billion for dementia (vs £11·7 billion), and £10·1 billion for stroke (vs £8·6 billion). For cancer, CHD, and stroke, the differences in the total cost are non-negligible but modest, but for dementia, using the higher cost valuation of informal care almost doubles the cost, and this choice greatly affects the relative costs of the diseases. Landeiro and colleagues valued productivity losses in terms of average wage, which is approximately £16 per h (£129 per 8 h). Some authors prefer this method for valuing informal care,4Wimo A Seeher K Cataldi R et al.The worldwide costs of dementia in 2019.Alzheimers Dement. 2023; 19: 2865-2873Crossref PubMed Scopus (57) Google Scholar making it another reasonable methodology. This approach results in total costs falling in between the cost reported in the paper and the estimates that value informal care the same as formal care. As Landeiro and colleagues acknowledge, hours of supervision are likely to be severely undercounted, as these hours were not elicited in the English Longitudinal Study of Ageing survey.1Landeiro F, Harris C, Groves D, et al. The economic burden of cancer, coronary heart disease, dementia and stroke in England in 2018: an evaluation of two cohort studies. Lancet Healthy Longev 2024; published online July 25. https://doi.org/10.1016/S2666-7568(24)00108-9Google Scholar For dementia, the hours of supervision have been estimated to be 29% of the total hours of informal care in the European Region;5WHOGlobal status report on the public health response to dementia.https://www.who.int/publications/i/item/9789240033245Date: Sept 1, 2021Date accessed: June 14, 2024Google Scholar thus, including this proportion of hours of supervision would potentially increase the burden of informal care for dementia (and possibly stroke) by 41%, but likely much less for cancer and CHD. Alternative choices regarding productivity losses can substantially affect the absolute and relative cost estimates. The study by Landeiro and colleagues considered only the first 90 days of missed work as productivity loss due to morbidity, while all future productivity losses were attributed to mortality. If productivity losses due to morbidity encompass all future losses beyond the first 90 days of missed work, the costs would be much higher for cancer, CHD, and stroke, and only slightly higher for dementia, as the onset of dementia is mostly likely to occur at older age. The information provided in the paper is insufficient to estimate the magnitude of this effect. Similarly, women in their 50s who care for a parent often do not resume work after the parent dies. Including these potential losses of work hours as productivity losses could increase the estimated burden of informal care by seven times.6Skira MM Dynamic wage and employment effects of elder parent care.Int Economic Rev. 2015; 56: 63-93Crossref Scopus (0) Google Scholar These points should not be interpreted as a claim that the calculations in the paper by Landeiro and colleagues are wrong. Instead, they illustrate how different reasonable methodological choices can lead to vastly different estimates. The authors have contributed to the scientific community by explaining their methodological choices and providing detailed calculations of the components. This transparency will allow others to perform calculations similar to those that we have shown in this Comment, to understand the sensitivity of the results to different methodological choices. JL declares funding from NIH, Bright Focus, United Nations, World Bank, and Asian Development Bank; honoraria from NIH, University of California, Berkeley, Stanford University, Southern Illinois University of Medicine, University of Texas, Austin; support for attending meetings from University College London, University of Venice, University of California, Berkeley, Queens College Belfast, and Davos Alzheimer's Collaborative; and participation in the Advisory Board of China Health and Retirement Longitudinal Study, English Longitudinal Study of Ageing, Longitudinal Study of Health and Ageing in Kenya, Global Latino Sequencing Study for Alzheimer's Disease, Asian Development Bank, Cognitive Level Enhancement through Vision Exams and Refraction, and Northern Ireland Cohort for the Longitudinal Study of Ageing. EM declares funding and honorarium from NIH and funding from Bright Focus. The economic burden of cancer, coronary heart disease, dementia, and stroke in England in 2018, with projection to 2050: an evaluation of two cohort studiesThis study provides contemporary estimates of the wide-ranging impact of the most important chronic conditions on all aspects of the economy in England. The data will help to inform evidence-based polices to reduce the impact of chronic disease, promoting care access, better health outcomes, and economic sustainability.