Direct and Indirect Impacts of COVID-19 on Excess Deaths and Morbidity: Executive Summary
Total Page:16
File Type:pdf, Size:1020Kb
Direct and Indirect Impacts of COVID-19 on Excess Deaths and Morbidity: Executive Summary Department of Health and Social Care, Office for National Statistics, Government Actuary’s Department and Home Office 15 July 2020 Background The COVID-19 pandemic will impact the health of many people in England and unfortunately many people will lose their lives. This paper provides a summary of research and analysis, discussing and estimating the health impacts (both excess deaths1 and morbidity) from the pandemic. Impacts of the pandemic may be direct from COVID-19 or may be indirect from changes to the healthcare system or lockdown measures. We conceptualise harm to health using the following four categories: A. Health impacts from contracting COVID-19 (A) B. Health outcomes for COVID-19 worsened because of lack of NHS critical care capacity (B) C. Health impacts from changes to health and social care made in order to respond to COVID- 19, such as changes to emergency care (C1), changes to adult social care (C2), changes to elective care (C3) and changes to primary and community care (C4). D. Health impacts from factors affecting the wider population, both from social distancing measures (D1) and the economic impacts increasing deprivation (D2). The results are briefly discussed in the section below; summary tables of the mortality and morbidity impacts can also be found below. Methodology and scope It is important to note that the estimates presented are based on scenarios; they do not represent forecasts. This paper was written in the middle of the pandemic; the estimates represent a point in time, using evidence from the initial months of the pandemic to model scenarios going forwards. Estimates for the different categories of harm use different scenarios and assumptions; these are outlined below: • Category A: Estimates for direct COVID-19 deaths are based on a scenario that shows incidence of COVID-19, and hence number of deaths, is constant over time from 20th June to March 2021, at a level of 900 deaths per week. This scenario is one of five signed off by SAGE in May 2020 for planning purposes and is the closest of these scenarios to the latest available weekly data at the time of writing2. It should be noted this scenario is not a forecast or an official Government planning scenario. In this document it is referred to as the COVID-19 Static Scenario (or “CSS”). This scenario is compared to the other four SAGE scenarios in Annex H. • Category C: Harm from changes to emergency care (C1) and adult social care (C2) are also linked to the CSS. Estimates of harm from changes in elective care (C3) assume that 6- 1 Defined as any death due to the COVID-19 pandemic which would not have occurred otherwise within one year. 2 In the week ending 19th June 2020 there were 744 deaths registered with a mention of COVID-19 on the death certificate. 1 | Page months of non-urgent elective activity is delayed for up to 5 years; estimates from primary and community care (C4) are based on a 6-month reduction in activity. • Category D: For impacts from social distancing measures (D1), we assume a 2-month lockdown based on illustrative scenarios, which does not take into account the CSS or try to reflect the staged relaxation of restrictions that have occurred to date. We use percentage changes in disease specific mortality due to an OBR forecast increase in unemployment to model the short (1 year), medium (2-5 years) and long-term (5-45 years) impacts of a lockdown-induced recession (D2). For the long-term impacts we also take an alternative approach using the Index of Multiple Deprivation. Further discussion and the methodology underpinning the estimates cited here can be found in the full paper. The range of health impacts presented are not exhaustive; there are likely to be more impacts that are not discussed because of their complexity and indirect links to the pandemic. Summary of direct and indirect health impacts The impact on mortality and morbidity for each category of harm are discussed in more detail below. Figure 1 presents the total impact in terms of excess deaths. Figure 2 and Figure 3present the total Quality Adjusted Life Years (QALYs) for mortality and morbidity; in Figure 2 these are split across three different time periods, for Figure 3 these are split by mortality and morbidity. These comparisons show several interesting points: • The direct COVID-19 deaths account for the majority of all excess deaths • However, when morbidity is taken into account, the estimates for the health impacts from a lockdown and lockdown induced recession are greater in terms of QALYs than the direct COVID-19 deaths. • Much of the health impact, particularly in terms of morbidity, will be felt long after the pandemic is assumed to last (1 year for this exercise, though this is a scenario not a forecast). It should be noted that the health impacts modelled here represent a scenario with mitigations in place. Without mitigations, a far larger number of people would have died from COVID-19 such that the QALY impact from COVID-19 deaths would be more than three times the total QALY impact of all the categories (mortality and morbidity impacts) for the CSS mitigated scenario presented here. A comparison with an unmitigated scenario3 is provided in Annex G and shows that mitigation have prevented up to 1.5m direct COVID-19 deaths. 3 This is the “Unmitigated RWC” (31st March) scenario which is an illustrative scenario if social distancing measures were not introduced to prevent or delay the spread of disease. 2 | Page 80,000 60,000 40,000 20,000 - March 2020 to March 2021 Years 2-5 Lifetime -20,000 Category A - Direct COVID-19 deaths Cat. C (i) Emergency care Cat. C(ii) Adult social care Cat. C (iii) Delay non-urgent elective care Cat. C (iv) Primary and community care Cat. D lockdown, recession, inc. deprivation Figure 2 Estimated excess deaths from Categories A, C and D 1,200,000 1,000,000 800,000 600,000 400,000 200,000 - March 2020 to March 2021 Years 2-5 Lifetime Category A - Direct COVID-19 deaths Cat. C (i) Emergency care Cat. C(ii) Adult social care Cat. C (iii) Delay non-urgent elective care Cat. C (iv) Primary and community care Cat. D lockdown, recession, inc. deprivation Figure 1 Estimated total QALYs (morbidity and mortality) from Categories A, C and D 3 | Page 1,200,000 1,000,000 800,000 600,000 400,000 200,000 - Category A - Cat. C (i) Cat. C(ii) Adult Cat. C (iii) Delay Cat. C (iv) Cat. D lockdown, Direct COVID-19 Emergency care social care non-urgent Primary and recession, inc. deaths elective care community care deprivation QALY mortality QALY morbidity Figure 3 Estimated total lifetime QALYs for Categories A, C and D Description of health impacts Category A: Health impacts from contracting COVID-19 We estimate that from the 32,000 COVID-19 deaths registered between 21st March and 1st May, 25,000 were “excess deaths” in that they would not have occurred otherwise within 1-year4. Under the COVID-19 Static Scenario (CSS), it is estimated there would be an additional 53,000 COVID-19 deaths to March 2021, 42,000 of which would be “excess deaths”. In total this equates to 530,000 lost Quality Adjusted Life Years (QALYs) and 700,000 Years of Life Lost (YLL) over the 12 month period (21st March 2020 to 19th March 2021). For people who contract COVID-19 and survive, there are likely to be morbidity impacts particularly amongst those hospitalised and needing critical care, including cognitive, physical and mental health impairments. We estimate these equate to 40,000 lost QALYs within 1-year. The long-term health impacts are unknown. Category B: Health outcomes for COVID-19 worsened because of lack of NHS critical care capacity Following the experience of other countries ahead of England in the COVID-19 pandemic, there were concerns that the UK’s critical care capacity and ventilation provision would not be sufficient. Efforts were made to increase capacity through the creation of new Nightingale hospitals and extending existing provision. It is judged that there have been no cases of NHS critical care capacity being breached to date, and none are anticipated within the CSS. Analysis of an unmitigated scenario suggests there could have been up to 1 million additional COVID-19 deaths in patients unable to access the hospital care they require. This is on top of an estimated 400,000 COVID-19 deaths that would have occurred if mitigations to reduce infections were not put in place even with fatality rates that assume sufficient NHS critical care capacity. 4 Note, this is a different definition for “Excess deaths” to that used in other contexts, such as in the ONS weekly deaths statistics, which computes “Excess deaths” by comparing the weekly total to the previous 5- years average for the equivalent week. 4 | Page Category C: Health impacts from changes to health and social care made in order to respond to COVID-19 C1: Changes to emergency care Emergency attendance and admissions have decreased since the start of the pandemic; people may have been reluctant to attend accident and emergency departments because of fears or perceptions that they should remain at home, and some causes may have decreased due to lockdown measures. We estimate changes to emergency care may account for 6,000 existing excess deaths in March and April 2020. If emergency care in hospitals continues to be low for a full 12-months, this could result in an additional 10,000 excess deaths.