COVID-19 and Well-being

Life in the Pandemic

Report

25 November 2021

3. Quality of life in the first year of COVID-19

Abstract

As people contend with rising COVID cases and deaths, and cope with lockdowns, confinements and the economic fallout of the pandemic, their quality of life has been greatly altered. Both physical and mental health outcomes have declined. Preliminary evidence suggests that the resulting disruptions to schooling may be associated with serious long-term risks to children’s life chances. Average life satisfaction has fallen slightly in most countries, but early evidence in some cases also suggests a surprising level of resilience. Like many other outcomes, subjective measures of well-being are sensitive to the timing of data collection within 2020, reflecting evolving changes in COVID risks, lockdown measures and the overall government response. This underscores the need to strengthen rapid data collection systems in order to monitor and react to individuals’ changing circumstances in real time.

Quality of life has been dramatically impacted throughout the OECD by the COVID-19 pandemic. 30% of respondents from 15 OECD countries reported that their lives were greatly affected by the virus from June to December 2020; this percentage rose to 35% for the period January to June 2021 (Figure 3.1). Changes in material conditions (documented in Chapter 2) coupled with social distancing protocols and fears of the health impacts of the virus have meant that almost no aspect of everyday life has been untouched.

Figure 3.1. Over one-third of respondents report that COVID has greatly affected their lives

3.1. Physical and mental health

3.1.1. Excess mortality and life expectancy

Excess mortality data – rather than COVID-19 fatality rates – provide a more accurate picture of the direct and indirect impacts of the pandemic on the mortality of OECD residents

COVID has impacted longevity both directly – through increased mortality, especially of the elderly population – but also indirectly, as both the supply and demand of routine health services have been curtailed due to the virus. The surge of COVID-19 cases across the OECD meant that hospitals were forced to reallocate resources to treat victims of the pandemic, yet in many cases they still found themselves understaffed and underequipped to handle the influx of patients: both those suffering from COVID-19 complications, and those seeking treatment for other health problems. Half of European countries recruited retired or inactive healthcare workers, and many also hired students who were in their final years of medical studies. In Italy, at the height of the first wave of the pandemic, 80% of pre-crisis beds in intensive care units (ICU) were estimated to be housing COVID-19 patients: in Ireland, France and Belgium, the figure was closer to 65%. To address these problems, countries transformed other wards to ICUs, created field hospitals, transferred patients to different regions with more capacity and partnered with private hospitals (OECD/European Union, 2020[2]).

The abrupt shift to treating COVID-19 patients resulted in disruptions to other health services. A report from the World Health Organization found that in Europe, the five health services most disrupted during pandemic peaks were rehabilitation services (91% of countries surveyed reported disruptions), dental services (91%), non-communicable disease treatment and diagnosis (76%), family planning (74%) and outreach for immunisations (63%) (European Observatory on Health Systems and Policies & WHO, 2020[3]).

Limited capacity, coupled with the population’s fear of contracting the virus, resulted in fewer visits to health centres and hospitals. Across 15 OECD countries in the period April to December 2020, over 74% of people on average avoided going to hospitals or health centres to seek treatment; this share rose to 77% for the period January through June 2021 (Figure 3.2, Panel A). Monthly data for the OECD-12 average (Figure 3.2, Panel B) indicate that this fear dissipated somewhat from June to August 2020 but began rising steadily in September through the end of 2020. (The reasons for reduced visits to health centres are many. In addition to reductions in service provision and increased fear of contracting the virus, declines in visits may also be due to changes in activities and situations in which one might need to visit a health centre. For example, the observed reduction in traffic accidents and fatalities (Figure 4.7) and in seasonal flu cases (Jones, 2020[4]) both entail fewer visits to hospitals and/or health centres.)

Figure 3.2. On average in 2020, over 70% of people in 15 OECD countries avoided hospitals or health centres during the pandemic, with further increases in the early months of 2021

Excess mortality statistics are better able to capture the overall effects of the pandemic on mortality than data on deaths attributed to COVID-19. Depending on the way in which death certificates are filled out in different countries, combined with variable COVID-19 testing availability and practices, certain fatalities may be coded as COVID-related in some jurisdictions but not in others. Furthermore, the pandemic has led to disruptions to preventative care and continuity of care; the pandemic is clearly contributing to any resulting negative health impacts, even though the official cause of death is not recorded as COVID-19. For this reason, excess mortality statistics provide a more accurate picture of the ways in which the pandemic has affected mortality rates (Morgan et al., 2020[5]). The excess mortality statistics used in this report are defined as the increase in all-cause mortality over the expected mortality based on historical trends (here, compared to average values from 2015-19). As can be seen in Figure 3.3, the total number of deaths in 33 OECD countries with data increased by 16%, on average, over the first year of the pandemic: from 11 March 2020 – when the World Health Organization first declared COVID-19 a global pandemic – to early May 2021, compared to the 2015-19 average.1

The OECD countries with the highest rates of excess mortality between March 2020 and early May 2021 include Mexico, Colombia, Chile, the Czech Republic, Poland, the Slovak Republic and the United States, where deaths were more than 20% higher than the 2015-19 average for this period. Conversely, a handful of countries – including Norway, Iceland and New Zealand – experienced about the same or fewer deaths in 2020-21 as compared to the baseline period. This decline reflects the success of early confinement measures in ensuring relatively low exposure to the virus (aided by geographic isolation and/or low population density). In addition, with the population staying home, there were fewer fatalities from non-COVID causes of death, such as road traffic accidents (see Figure 4.7) and other communicable illnesses, such as the seasonal flu (Jones, 2020[4]). It is also important to keep in mind the dynamic nature of the data:2 different countries experience waves of the virus at different times. Panels B and C of Figure 3.3 depict excess death trends for G7 countries for which data are available. The virus peaked in Italy and France first, in early- to late-March 2020, whereas the first wave did not peak in the United Kingdom until mid-April. Some OECD countries had relatively low excess death rates in 2020 but saw an uptick in 2021 (OECD, forthcoming[6]). Early evidence suggests that excess mortality has, on average, been lower in the first half of 2021 (see Figure 3.3, Panels B and C) (Eurostat, 2021[7]), which may in part be due to the introduction of COVID-19 vaccines. However vaccination rates vary widely, both within and across OECD countries (OECD, 2021[8]).

There are large spatial inequalities of excess mortality within OECD countries: in many instances, within-country differences are larger than between-country disparities. On average, the hardest hit regions had excess mortality rates 17 percentage points higher than the least affected regions (Diaz Ramirez, Veneri and Lembcke, 2021[9]). More specifically, excess mortality in the worst affected regions of Mexico in 2020 was 60 percentage points higher than in the least affected areas of Mexico; the sub-national differences in Colombia were 90 percentage points. During the same time period, the gap between the most and least affected OECD countries was only 47 percentage points (Diaz Ramirez, Veneri and Lembcke, 2021[9]). In most OECD countries, metropolitan regions were hardest hit during the early months of the pandemic; however, the gap in excess mortality rates between metropolitan and remote regions diminished throughout 2020 and into 2021. While the virus was initially introduced to many countries through air travel in large metropolitan areas, as the pandemic has worn on it has spread to all areas within countries (Diaz Ramirez, Veneri and Lembcke, 2021[9]; OECD, 2020[10]).

Though it is too soon to say whether the pandemic has caused significant declines in life expectancy in OECD countries, preliminary evidence suggests that this is the case. As the pandemic lingers, and disproportionately affects certain groups more than others, the impact on life expectancy is gradually becoming apparent. Provisional estimates from 29 OECD countries show that life expectancy at birth has fallen by 0.6 years on average from 2019 to 2020 (Figure 3.4) (OECD, forthcoming[6]), declining in all but five countries.3 A report published by the US National Center for Health Statistics (NCHS) in early 2021 found that life expectancy in the United States fell on average by one year, from 78.8 in 2019 to 77.8 in 2020 (Arias, Tejada-Vera and Ahmad, 2021[11]), the largest year-on-year decline since World War II. In addition to the immediate excess mortality associated with the pandemic, a longstanding literature identifies higher GDP growth and lower unemployment as two determinants of life expectancy gains over time (OECD, 2017[12]). While the relationship is indirect, a prolonged economic recession may have negative impacts on life expectancy in the long term.

Figure 3.3. In the first year of the pandemic, deaths increased by 16% on average across the OECD

Figure 3.4. In 2020, life expectancy in 29 OECD countries fell by 0.6 years on average

3.1.2. Depression, anxiety, eating disorders and deaths from suicide

Symptoms of anxiety and depression spiked across the OECD in 2020, and as of mid-2021, have not begun to recover on average, though recent evidence suggests a recovery in some countries

Over the course of 2020, mental health deteriorated across the OECD area, with rates of anxiety and depression doubling in a number of countries (OECD, 2021[16]). Experience from previous pandemics illustrates how these events often have significant impacts on population-level mental health, leading to an increase in feelings of depression, anxiety, insomnia, post-traumatic stress disorder and even suicide (Cénat et al., 2020[17]; Tzeng et al., 2020[18]; Yip et al., 2010[19]). Symptoms of psychological distress are not solely confined to the patients themselves, but also affect family members (Tsang, Scudds and Chan, 2004[20]). The COVID-19 virus combines a number of threats to mental health: of contracting the virus, spreading the virus to loved ones, death, disruptions to daily routines, school closures, and losing one’s employment and livelihood, all while being forced to physically distance from friends and family (United Nations, 2020[21]).

Given that mental health conditions exist on a continuum, individuals who were previously able to cope may find that they are now struggling in the pandemic (Patel et al., 2018[22]). Prior to the pandemic, around 264 million people worldwide were affected by depression, and suicide was the second-largest cause of death for young people (United Nations, 2020[21]). The share of individuals feeling anxious, worried, depressed or taking little pleasure in everyday activities has been on the rise. From April through December 2020, 10.2% of respondents in 15 OECD countries had a Patient Health Questionnaire (PHQ‑4)4 score indicating severe mental distress, and an additional 13.6% were at risk of moderate mental distress. From January to June 2021, those rates rose slightly, on average, to 10.6% and 13.8%, respectively (Figure 3.5). Rates of severe mental distress rose in six countries and declined only in two (in the remaining seven countries with data, the change was insignificant) (Figure 3.5, Panel A), and rates of moderate mental distress rose in one country and improved in two (Figure 3.5, Panel B). In general, mental distress was highest during the height of lockdowns and stay-at-home restrictions, and subsequently declined from June to August 2020 when COVID-19 rates fell in the majority of OECD countries leading to a loosening of restrictions (OECD, 2021[16]) (see also Figure 3.8, Panel A).

Figure 3.5. The share of those experiencing severe or moderate mental distress in 15 OECD countries rose slightly, on average, between the second half of 2020 and the first half of 2021

Over one-quarter of respondents in 15 OECD countries were at risk of anxiety (25.0%) and depression (26.58%) in 2020; both rates increased slightly, but significantly, in the first half of 2021 (for anxiety to 26.6%, for depression to 26.64%, see Figure 3.6). For eight OECD countries with comparable baseline measures, the share of respondents at risk of depression by mid-2021 had risen significantly since 2014: by more than 20 percentage points in two European countries (Figure 3.7). While the source of the baseline data is different from the source of the pandemic-era data, implying that caution should be exercised in interpreting any individual country trajectory, both data sources use the same instrument (PHQ-2) to assess the risk of depression, therefore the overall trend of large deteriorations is likely to be true. See also OECD (2021[16]) for similar evidence of dramatic increases in the prevalence of mental health problems in 2020 compared to pre-pandemic baselines in OECD countries.

Data collected by national statistics offices across OECD countries corroborate these findings (see Box 3.1). The share of Canadians reporting fair or poor mental health during 24 April to 11 May 2020 was 24%, a significant increase from the 8% reported by the Canadian Community Health Survey (CCHS) for 2018 (Statistics Canada, 2020[23]). In Great Britain, data from the Opinions and Lifestyle Survey from June 2020 showed that 19% of Britons had some form of depression, as measured by the PHQ-8 questionnaire,5 and that depression rates had almost doubled from the pre-pandemic baseline. Furthermore, the pandemic resulted in declining mental health among those who previously had no, or only mild, symptoms: 13% of adults in Great Britain developed moderate to severe symptoms over the course of 2020 (ONS, 2020[24]). In March 2020, 50% of Britons reported high levels of anxiety; however, unlike depression, there is evidence suggesting that anxiety levels have fallen since early lockdown (ONS, 2020[25]). Survey data collected by the Japanese Ministry of Health, Labour and Welfare in August showed that 6% of respondents reported feeling depressed all day every day over the past two weeks, while 9% reported having no interest in things or having lost interest in things that used to be enjoyable to them (LINE Research Official Blog, 2020[26]).6

Figure 3.6. More than a quarter of the population in 15 OECD countries are at risk of anxiety and depression, 2020-21

Figure 3.7. The risk of depression has increased dramatically in eight European countries

Research in other OECD countries also points to higher risk for depression and anxiety as the pandemic progressed. A study by the Tel Aviv University and the Academic and Technology College of Tel-Hai found that during the first lockdown in May 2020, 23% of Israelis reported medium to high levels of anxiety and 14% high levels of depression, which had further increased to 29% (anxiety) and 20% (depression) by October 2020. By way of comparison, in 2018 only 12% of Israelis reported these levels of anxiety, and 9% for depression (Tel Aviv University, 2020[28]). The Israeli Central Bureau of Statistics (CBS) flash survey, the Civilian Resilience during the Coronavirus Survey, found similar results. The first round of data collection, from 26 April to 1 May 2020, showed that 16.2% of Israelis over the age of 21 felt depressed (“to a large extent”” or “to an extent”) and 34.4% felt stress and anxiety (CBS, 2020[29]); these rates rose to 21% and 42%, respectively, during a follow-up survey conducted from 12 July to 16 July 2020 (CBS, 2020[30]). A Korean study published in May 2020 showed that 47% of respondents reported feelings of anxiety and/or depression because of the pandemic (Park and Yu, 2020[31]). The COVID-19 Social Survey in Chile (refer to Box 5.3 for survey details) found that 21.4% of adults reported moderate or severe symptoms of anxiety and/or depression based on the PHQ-4 scale from 24 June to 7 August 2020 (Ministerio Desarrollo Social y Familia, 2020[32]). An online survey7 of mental health in Costa Rica found that risk for depression and anxiety (measured using the PHQ-4 scale) increased more than three-fold between March and October 2020 (UNED, 2021[33]).

Some country evidence suggests that although mental health declined precipitously in the early stages of the pandemic, it may have started to recover by mid-2021 in some instances. In Colombia, an online (non-representative) survey of more than 18 000 adults, running from 20 May to 20 June 2020 in the early days of the pandemic, found that 35% of respondents were at risk of depression (using PHQ‑2) and 29% at risk of depression (using GAD-2);8 the same survey also found that women and young adults were most at risk (Guzmán Mena and Tamayo, 2020[34]; Sanabria-Maxo et al., 2021[35]). Data collected by Colombia’s statistical office (DANE) as a part of its Social Pulse survey (refer to Box 4.1 for methodological information) around the same time, July 2020, found that 22.5% of Colombians reported feeling sad over the past seven days, 40.4% preoccupied or nervous, and 4.9% that it was impossible to feel positive feelings. However one year later, in July 2021, these rates had improved to 14.8%, 38.1%, 0.9%, respectively (DANE, n.d.[36]). Similarly, data from the Australian Bureau of Statistics’ Household Impacts of COVID-19 Survey collected in mid-April 2020 – when the country was experiencing peak levels of new COVID-19 cases – showed that 35.4% of respondents reported being nervous at least some of the time, up from 20.1% in 2017-18; and 41.8% felt restless or fidgety, compared to 23.7% pre-pandemic. Conversely, the rate of feeling “so depressed that nothing could cheer you up” remained more or less the same – 7.8% in 2017-18 vs. 7.5% in April 2020 (ABS, 2020[37]). However data from Australia National University’s COVID-19 impact monitoring survey (ANUPoll) showed that by April 2021, levels of psychological distress had reverted to pre-pandemic levels, with the only exception of young people, whose rates of psychological distress remained significantly higher in April 2021 than they were before COVID-19 (AIHW, 2021[38]).

Box 3.1. Innovation: National Statistics Offices have launched COVID-19-specific surveys that capture changes in mental distress

United States: Household Pulse Survey and COVID Impact Survey

The US Census Bureau is leading a multi-agency push to collect experimental survey data on the socio-economic and health effects of COVID-19 on American households. Unlike other Census Bureau surveys, the Household Pulse Survey is designed to mobilise data quickly, with results released in a matter of weeks rather than months or years. The Household Pulse Survey began in April 2020 and encompasses three phases of data collection (US Census Bureau, n.d.[39]).¹ The first phase collected data on a weekly basis, however subsequent phases moved to a biweekly approach.

A total of 14 million Americans were contacted via email and text, inviting them to complete a 20-minute online questionnaire. The survey receives around 100 000 responses weekly. Data were initially published within eight days of collection (though in later phases this moved to once every 14 days), to provide a real-time snapshot of how COVID is affecting the lives of Americans. Data are weighted to be representative at the national and state levels; in addition, representative statistics are available for the fifteen largest Metropolitan Statistical Areas. The survey is not longitudinal, in that it does not track unique respondents over time, nor does it include pre-pandemic outcomes for individuals. However, many of the topics covered in the Pulse Survey are covered in the annual National Health Interview Survey (NHIS), which allows for baseline comparisons (CDC, 2020[40]).

The Data Foundation, a private non-profit organisation, introduced the COVID Impact Survey in April 2020, designed to be complementary to the Household Pulse Survey. The survey is conducted by the National Opinion Research Center (NORC) at the University of Chicago and uses an address-based random sampling design to be representative of the population aged 18 years and over at the national and regional (ten states and eight metropolitan areas) levels. Around 2 000 respondents are interviewed during each week of data collection: the first wave of data collection took place in April 2020, the second in May 2020 and the third in June 2020. Survey modules cover physical, social and mental health as well as financial and economic health (Wozniak et al., 2020[41]). Data from both the Pulse and Impact surveys are used in this publication.

Data collected in the first phase of the Household Pulse (from 23 April to 21 July) showed that rates of anxiety (measured using GAD-2) among Americans reached 36% in mid-July, and 37% in early November. Rates of depression peaked at 29.6% in mid-July and rose to 30.2% in December 2020 (Figure 3.8, Panel A). The weekly averages from April through December point to a large increase from baseline values. Comparable data from January-June 2019 showed only 8% of respondents with symptoms of an anxiety disorder, and 7% of respondents with symptoms of depression (National Center for Health Statistics, 2020[42]). A separate study using nationally representative survey data in the United States came to similar conclusions. By comparing data during the pandemic (from the COVID-19 and Life Stressors Impact on Mental Health and Well-being study, 31 March-13 April) with data from earlier periods (National Health and Nutrition Examination Survey, 2017-18), Ettman et al. (2020[43]) found that depressive symptoms (measured using the PHQ-9) increased more than threefold during the COVID crisis. Rates of depression increased for all levels of severity, with the greatest rises for the share of those with mild to moderate depression. However there is some suggestive evidence that rates of depression and anxiety may be dropping over the course of 2021: as can be seen in Figure 3.8 (Panel A), both measures have been steadily falling since the early months of 2021.

Germany: Socio-Economic Factors in and Consequences of the Spread of the Coronavirus in Germany (SOEP-CoV)

The German Socio-Economic Panel (SOEP) conducted by the German Institute for Economic Research (DIW Berlin) is a longitudinal survey of over 30 000 individuals in 20 000 households that has been running for more than three decades (Goebel et al., 2019[44]). Households remain in the study, which means their outcomes can be tracked over a lifetime. The questionnaire covers topics relating to household composition and demographics, employment and earnings, along with health and life satisfaction indicators. During the early stages of the pandemic in April 2020, SOEP collaborated with researchers from the Universität Bielefeld to initiate a new project entitled, “Socio-Economic Factors in and Consequences of the Spread of the Coronavirus in Germany (SOEP-CoV)” (SOEP-CoV, 2021[45]), which capitalises on the existing survey infrastructure to provide panel data on a representative sample of Germans during the pandemic. By integrating with the existing SOEP, the new SOEP-CoV has comparable baseline (i.e. pre-pandemic) measures for a wide range of well-being outcomes.²

Two rounds of telephone interviews ran from 1 April to 28 June 2020, and from January to February 2021. Participating households were selected from the overall SOEP sample and randomly divided into nine tranches, in such a way that the complex design information of the existing SOEP subsamples was preserved (Kühne et al., 2020[46]). Tranches 1-5 were surveyed with a time difference of two weeks between each tranche. This difference was shortened to one week after tranche 5. 12 000 households were invited to participate in telephone interviews, and 6 700 households completed surveys. Reported data are weighted to account for non-responses. All individuals interviewed in the first wave were re-interviewed in 2021. Because all selected households participate in the longitudinal study, it is possible to track their outcomes before, during and after the pandemic (Kühne et al., 2020[46]).

The survey captures mental health, including anxiety and depression, through the PHQ-4 questionnaire. According to this survey, rates of severe and moderate depression remained relatively stable in Germany from 2016 through 2019, 2020 and 2021, however the share of those with mild depression increased significantly from 2019 to 2020 (Figure 3.8, Panel B). Yet by the subsequent year, this increased prevalence appears to have dissipated; in 2021 the share of the population reporting symptoms of mild depression, along with overall prevalence of anxiety and depression, were not significantly different from any pre-pandemic year (Figure 3.8, Panel B) (Entringer and Kröger, 2021[47]).

Figure 3.8. Anxiety and depression rates spiked in the United States in July and again in December 2020, while in Germany declines in mental health have been less marked

The pandemic has disrupted access to mental health services (Figure 3.9). Access to mental health services was already limited before the pandemic: 67% of working-age adults with mental distress who wanted mental health care reported having difficulties accessing it (OECD, 2021[48]). In 2020, hospitals in particularly hard-hit regions used beds normally reserved for mental health patients to care for those infected by COVID-19 (OECD, 2021[16]), while people also avoided in-person care due to fears of contracting the virus (United Nations, 2020[21]). According to a survey conducted by the World Health Organization from June to August 2020 (2020[49]), school and work-based mental health programmes were the services most likely to be completely disrupted due to the pandemic, followed closely by home or community outreach, and by interventions for caregivers (Figure 3.9); a second round survey, from January to March 2021, found that school-based mental health programmes continued to be the most disrupted service type (WHO, 2021[50]). Disruptions to school-based services have contributed to the decline in the mental health of young people (Chapter 6) (OECD, 2021[51]). In response, a number of OECD countries have introduced new modes of mental health service delivery, including telemedicine, online therapy and distress lines (OECD, 2021[16]; WHO, 2020[49]). A pre-pandemic OECD study found that telemedicine was an effective way of improving mental health; that cognitive behavioural therapy conducted remotely was equally effective as face-to-face treatment for conditions such as obsessive-compulsive disorder, insomnia and excessive consumption of alcohol; and that remote treatment is also effective in reducing the symptoms of depression and anxiety (Oliveira Hashiguchi, 2020[52]). However, this same study cautioned that barriers to access – such as lack of reimbursement for telehealth services, or lack of digital literacy – remain, especially for the elderly and those from low income households (Oliveira Hashiguchi, 2020[52]). It remains to be seen how the interplay of in-person service disruption, increased demand and new modes of service delivery, all in the context of a global pandemic, will impact population-level mental health.

Although suicides have not increased during the pandemic, evidence from some countries shows an uptick in self-harm, alcohol abuse and opioid overdoses

Data on suicide rates throughout 2020 and 2021 do not suggest any significant change from previous years (Pirkis et al., 2021[53]). In the early months of the pandemic there was widespread concern that suicides might increase dramatically because of declining population mental health, a deepening economic crisis and disruptions to mental health services. Indeed, evidence from the 2003 SARS epidemic showed that the areas hit by the virus experienced an increase in suicide rates, especially among elderly women (Chan et al., 2006[54]; Yip et al., 2010[19]). Additional research has shown the link between financial strain (encompassing debt, homelessness, unemployment) and suicides (Elbogen et al., 2020[55]). However, data from the COVID-19 pandemic do not provide evidence of an uptick in suicides. Data from regions of the United States, Australia, England and Germany showed no increase through the end of the second quarter of 2020 (a period encompassing both lockdowns and the immediate deconfinement period), while studies in Japan and Norway showed a decline during the early months of the crisis (John et al., 2020[56]; Knight, 2020[57]). Data from the United Kingdom Office for National Statistics (ONS) in fact found that suicides in England and Wales decreased during the first wave of the pandemic (Apr-Jul 2020), compared to previous years (ONS, 2021[58]). However, in the case of Japan, data from later in 2020 showed a subsequent increase (see Box 3.2). A study of 2 000 respondents in France in September 2020 showed that around 20% reported having suicidal thoughts. Although high, this was more or less the same level as 2016 (Gaubert, 2020[59]; Debout and Fourquet, 2016[60]). As the pandemic and its economic effects wear on, it will be important to monitor suicide risk (see Box 3.2): survey data from Belgium, France and the United Kingdom show that suicidal thoughts have increased among younger people, even if rates of actual suicides have not yet done so (OECD, 2021[51]). It is also important to measure suicide rates among sub-populations. In the United States, the overall number of suicides dropped by 5% between 2019 and 2020; however, suggestive evidence from some states, including Illinois, Maryland and Connecticut, suggests that suicides have risen for Black Americans (Rabin, 2021[61]).

Figure 3.9. School and work-related mental health programmes were the most likely to experience disruptions during the pandemic, across 21 OECD countries

In early April, the WHO issued a warning about the potential for excessive alcohol consumption during emergency situations such as the pandemic and ensuing lockdowns (WHO Regional Office for Europe, 2020[62]). Evidence from household surveys using online sampling suggests average measures of per capita alcohol consumption did not increase markedly, although more people reported drinking more since the start of the pandemic than those reporting drinking less (see Figures 9.3 and 9.4) (OECD, 2021[63]; OECD, 2021[64]). Increased alcohol consumption entails a number of health risks, including worse mental health outcomes, and it increases instances of domestic violence, along with a higher risk of disease or injury (OECD, 2021[63]).

Box 3.2. Spotlight: Suicides on the rise in Japan after falling in the first phase of the pandemic

Data from the Japanese National Police Agency show the dynamic nature of suicide rates in 2020 (Figure 3.10). Early on in the pandemic (April 2020) suicides in the country fell by 20% (Blair, 2020[65]); however, they subsequently rose by 7.7% in August (John et al., 2020[56]) and climbed to a five-year-seasonal-high by October 2020 (Wang, Wright and Wakatsuki, 2020[66]; Sakamoto et al., 2021[67]). Suicides then declined for the rest of 2020, and in 2021 they were no different from the five years prior. Suicides among Japanese women have been particularly high and drove the October 2020 increase (see Box 6.3). Tanaka and Okamoto (2021[68]) suggest that government subsidies, reduced working hours and school closures all contributed to lowering suicides in early 2020, implying that suicides may rise in the future as subsidies are rolled back while the adverse effects of the pandemic linger.

As a direct result of suicide increases, the Japanese government introduced a new cabinet-level position, the Minister of Loneliness, to diminish feelings of social isolation (Kodama, 2021[69]). The Japanese experience suggests that, even if suicides have not risen in most OECD countries to date, there may be cause for concern in the coming months and years. Indeed, this pattern in already evident in Korea: the number of people who engaged in self-harm was 36% higher in early 2020 than the year before, while rates of depression increased by almost 6% from 2019 (Ryall, 2020[70]).

Figure 3.10. Suicides in Japan fell in the early stages of the pandemic before rising in late 2020

Evidence shows that opioid overdoses are on the rise in some countries. In the United States the Centers for Disease Control and Prevention (CDC) reported 93 331 opioid overdoses in 2020, a 30% increase from the year prior and the highest annual rate ever recorded (Ahmad and Rossen, 2021[72]; Katz and Sanger-Katz, 2021[73]): in addition, mental health and overdose calls to first responders (those first on the scene in case of emergency, such as EMTs (emergency medical technicians), firefighters, police officers, etc.) doubled in 2020, compared to the two years prior (Graham, 2021[74]). In Canada, 2020 also proved to be the deadliest year for opioid overdoses. In the province of Alberta, overdose deaths reached a peak of 142 in July, the highest rate since data collection began in 2016 (Government of Alberta Ministry of Health, n.d.[75]). The increase coincided with a decrease in treatment availability because of the pandemic: as a result of the decline in treatment adherence9 in April (52.6%) and May (55.8%) 2020, the annual average treatment adherence decreased to 75.4% from 89.8% in 2019 (Government of Alberta Ministry of Health, 2020[76]).10 Data from England and Wales show that the number of drug-related deaths in 2020 were the highest since record keeping began in 1993, and 3.8% higher than the year prior; however the ONS cautions that, due to delays in reporting, a large number of these deaths may have occurred in 2019, i.e. before the start of the pandemic (ONS, 2021[77]).

3.2. Subjective well-being

Life satisfaction remained fairly resilient on average, but outcomes are heavily dependent on the timing of data collection in relation to the progression of the pandemic

While material conditions and mental health deteriorated in the wake of the pandemic, trends in subjective well-being measures – such as life satisfaction and negative affect balance – are currently less clear-cut. Data from the Gallup World Poll show that life satisfaction deteriorated by a very small amount on average across OECD countries (from 6.71 in 2019 to 6.66 in 2020), but the pattern across countries is inconsistent (Figure 3.11, Panel A). Similarly, the share of people within a country reporting low levels of life satisfaction (defined as answering less than or equal to 4, on a scale from 0 “least satisfied” to 10 “most satisfied”) has fallen in some countries but increased in others (Figure 3.12). Finally, the share of the population reporting a negative affect balance – the share of the population reporting more negative feelings (anger, sadness, worry) than positive feelings (enjoyment, laughing or smiling a lot, feeling well-rested) the day before being interviewed – has increased for OECD countries on average (from 12.9% in 2019 to 14% in 2020), although this increase has not been consistent across all OECD members (Figure 3.11, Panel B).

One factor explaining the lack of clear trends in life satisfaction is the sensitivity of the measure to the time when data are gathered. Context is always important for measurement, but when gauging the impacts of the pandemic on measures of subjective well-being, understanding the relationship between fieldwork timing and the progression of the pandemic in the national context is vital. Data from the Gallup World Poll were collected throughout 2020, with fieldwork taking place at different times in different countries (refer to Box 3.4 for full details). This, coupled with the fact that different countries experienced waves of the pandemic at different times, means that pooled national averages may mask substantial variations in life satisfaction over the course of 2020 (as shown in Box 3.3). For example, national data from France show that 2020 life satisfaction peaked in June/July as the country emerged from a strict lockdown, before falling dramatically as a second wave took hold in late 2020 (Figure 3.14, Panel C).

Figure 3.11. National averages show only small changes in life satisfaction and negative affect balance from 2019 to 2020

Figure 3.12. Deprivations in life satisfaction slightly increased on average in the OECD, with mixed patterns across countries

Countries that experienced worse than average pandemic outcomes, or severe COVID-19 situations that necessitated more stringent than average government responses, feature larger deteriorations in subjective well-being in 2020 than other countries (Figure 3.13). For example, countries that implemented more stringent than average lockdown policies had larger deteriorations in both life satisfaction and negative affect balance, than did countries with less stringent policies. Importantly, these measures are averages during the period in which Gallup fieldwork was underway, which differ across countries (refer to Box 3.4 for exact field dates). Similarly, countries with above-average excess death rates experienced a significant decline in life satisfaction, and countries with above-average stay-at-home measures experienced a significant deterioration in negative affect balance. The impact of economic support is somewhat more nuanced, with no significant differences in the subjective well-being outcomes for countries above the OECD average; however, negative affect balance significantly improved among countries with support below the OECD average. This is likely because the average outcomes for countries with low economic support includes both countries that were not, at the time, severely negatively impacted by the pandemic – thus not needing support – as well as countries that simply did not provide support. Similarly, respondents in countries that provided above-OECD average economic support likely experienced positive subjective well-being impacts from the support received, and negative subjective well-being impacts from necessitating the support in the first place (unemployment, job insecurity, etc.).

Figure 3.13. The pandemic context during fieldwork is key to understanding subjective well-being outcomes across OECD countries

Another important explanation for lack of clear trends is that averages in life satisfaction mask large inequalities within national populations: while the pandemic has caused great suffering in some parts of the population, its effects have been far from uniform. Women (especially mothers), parents of school-age children, young people, those with financial and employment difficulties, and racial and ethnic minority groups reported greater drops in both life satisfaction and negative affect balance than their peers (refer to Chapter 6, as well as Chapters 5 and 7).

Box 3.3. Spotlight: Evidence from individual OECD countries shows both declines in life satisfaction as well as resilience in the face of the pandemic

Evidence from the Office for National Statistics (ONS) Annual Population Survey (APS) in Great Britain shows that life satisfaction dropped in the second quarter of 2020, erasing more than five years’ worth of gains (Figure 3.14, Panel A) and ending a decade of small but steady improvements in subjective well-being (Hardoon, 2021[80]). As of September 2020, life satisfaction had not yet rebounded. Data from the German SOEP-CoV survey (see Box 3.1 for details) show a decline in life satisfaction in April-June 2020 relative to the previous year, but to levels not significantly different from recent years (i.e. 2017); however, a follow-up survey in 2021 showed that life satisfaction had fallen significantly, to the lowest levels since 2016 (Figure 3.14, Panel B).

Quarterly data from France show a different picture, at least in the early stages of the pandemic (Figure 3.14, Panel C). Subjective well-being data collected by the Observatoire du Bien-être of CEPREMAP beginning in 2016, based on a representative sample of 1 800 respondents, show that life satisfaction reached a four-year high in late May/early June 2020. This peak coincided with de-confinement measures introduced after several months of strict lockdown (CEPREMAP, n.d.[81]).This surprising result could reflect the mitigating effects of government policies to protect livelihoods during the lockdown, or to changing frames of reference: after months in confinement, the lifting of restrictions may have led to spikes in satisfaction. The relationship of the spike with lockdown measures seems likely, as data from March 2021 showed that life satisfaction had fallen to the lowest levels since data collection began: this coincided with the third national lockdown (Perona and Senik, 2021[82]).

Conversely, life satisfaction in New Zealand has remained more or less stable over the past few years (Figure 3.14, Panel D). Data indicate an average life satisfaction level of 7.9 in 2020, up from 7.7 in 2018. These increases were reasonably consistent across most population groups (gender, age, ethnicity, region) (New Zealand Government, 2021[83]). However, New Zealand was quick to react and contain the pandemic and as a result had fewer cases, fewer fatalities, and a shorter period of domestic restrictions overall in 2020, relative to many other OECD countries.

Life satisfaction trends in 2020 may also respond to non-pandemic events. Data from the ANU Poll found that life satisfaction in Australia fell from 6.9 in January 2020 to 6.5 in April 2020, but quickly rebounded to 6.8 by April and May; by November 2020, life satisfaction reached 7.1, which was not significantly different from levels in October 2019 (Biddle et al., 2020[84]; Biddle et al., 2020[85]). While ANU researchers suggested that increases in life satisfaction beginning in April and May could be the result of easing lockdown restrictions and declining COVID rates, they speculate that the low rates of life satisfaction in January 2020 (6.9, compared to 7.1 three months prior) are more likely to reflect the Black Summer Bushfire crisis than any pandemic-related event (Biddle et al., 2020[85]).

Figure 3.14. Changes to trends in life satisfaction in 2020 varied depending on OECD countries’ experiences of the pandemic

Box 3.4. Methods: Gallup World Poll Data in 2020

The Gallup World Poll has been collecting nationally representative data in over 150 countries, including all 38 OECD countries, since 2007. The questionnaire includes a variety of topics, including business and economics, government and politics, and health (including mental health). Four of the headline indicators in the OECD’s How’s Life? report are measured using Gallup World Poll data, due to the absence of official data harmonised across member countries.1 In general, Gallup World Poll data focus on people’s opinions and perceptions. The survey is a repeated cross-section, meaning that respondents are sampled during each survey wave and are not tracked over time. Sample sizes vary by country, year and indicator type (from as small as 400 to as large as 13 000 and more), but average around 1 000 respondents. Gallup World Poll data in 2020 provide an opportunity to study the impacts of COVID-19 as they unfolded, in that fieldwork took place at different times in different countries, each of which experienced peaks and troughs of the virus at different times (Figure 3.15). Because of varying on-the-ground circumstances, each figure in this publication that presents Gallup data includes the timing of field work and ranks countries in chronological order of data collection.

Figure 3.15. Fieldwork timelines for the Gallup World Poll, 2020

Gallup survey data are collected through a combination of face-to-face and telephone (mobile and/or landline) interviews, depending on the country. From around March 2020 onwards, all data collection transitioned to telephone interviewing. Prior to 2020, data in the majority of OECD countries were already collected entirely over the phone. However, in 14 OECD countries (Chile, Colombia, Costa Rica, the Czech Republic, Estonia, Greece, Hungary, Israel, Lithuania, Latvia, Mexico, Poland, the Slovak Republic and Turkey) and three partner countries (Brazil, the Russian Federation and South Africa), interviews previously conducted face-to-face were switched in 2020 to telephone-based, which may result in some mode effects. In this publication, all countries with data collection method switches are marked with † in figures.

1. These comprise negative affect balance, gender gap in feeling safe at night, perceived lack of social support, and trust in national government. In general, three-year pooled averages are used when reporting Gallup Data in official OECD publications. This increases the number of observations within each country to allow for sufficiently large samples to estimate inequalities in gender, age and educational attainment outcomes. Not all countries are surveyed annually, therefore pooled averages also reduce the gaps in country coverage. This procedure is relaxed in the current publication, since the main aim is to detect 2019 to 2020 changes in outcomes.

3.3. Knowledge and skills

Between March 2020 and June 2021, schools across the OECD were closed – either fully or partially – more than half the time, due to COVID-19

The pandemic has caused huge disruptions to learning, as schools closed and teaching switched to remote delivery. In some OECD countries, during the first year of the pandemic all schools of all levels were closed; in others, priority was placed on keeping primary schools open with social distancing in place (OECD, 2021[88]). In all, millions of children had their learning disrupted due to the pandemic (Figure 3.16, Panel A). Normal schooling will not return for many until late in the 2021-22 school year, at the earliest. Figure 3.16 (Panel B) below shows the share of instruction days schools were fully or partially closed due to COVID-19, closed for academic breaks (non-COVID related), or fully open from March 2020 through June 2021. The bulk of school closures occurred from March to May of 2020, when most OECD countries experienced the height of the first wave. The share of days during which schools were closed (either partially or completely) fell from June through August, as many OECD countries had summer breaks, but began to rise in September 2020. However, the decision of many OECD countries to keep primary and secondary schools open (OECD, 2021[89])12 means that overall closures increased by less than in the early days of the pandemic. The pandemic is likely to exacerbate learning inequalities across OECD countries: countries with lower education performance pre-COVID were more likely to suffer from longer periods of school closures in 2020 due to system capacity constraints (OECD, 2021[89]; OECD, 2021[88]).

Schools across the OECD used a combination of methods to deliver remote learning to students during school shutdowns. All 32 OECD countries that participated in the Special Survey on Joint National Responses to COVID-19 School Closures13 implemented some form of online learning in 2020 and 2021. Online platforms were used for primary and secondary institutions in all countries aside from Sweden and the Russian Federation, where online solutions were not used in primary schools (OECD, 2021[90]). Take-home packages and television were the second-most common form of instructional delivery, with the first method more commonly used (by 84% of countries surveyed) in primary and lower-secondary schools (OECD, 2021[90]). Given the focus on online instruction, most OECD countries reported introducing measures to ensure inclusion in distance learning, including the introduction of self-paced coursework, partnering with mobile networks to increase access to the Internet, distributing or subsidising digital infrastructure, and providing economic support to low-income households (OECD, 2021[90]).

Figure 3.16. Millions of children across OECD countries had their schooling interrupted

Even temporary school closures are likely to lead to significant learning losses. The World Bank estimates that learning disruptions caused by the pandemic could lead to a 25% increase in the share of secondary students performing below PISA level 2 (Azevedo et al., 2020[92]). Through a series of models using data from 157 countries, the report predicts the results of 3, 5, and 7 months of school closure, concluding that this may result in a loss equivalent to between 0.3 and 0.9 years of schooling, with an average of 0.6 years of schooling; up to 7 million students in primary or secondary education may drop out for financial reasons. Another study in the United States estimated that students who missed in-school instruction in spring 2020 due to the pandemic would return to school in the following autumn with only 70% of the learning gains in reading compared to a regular year, and with only 50% of the gains for mathematics (Soland et al., 2020[93]). Simulations conducted by the Department for Education (DfE) in England estimated that primary school pupils lost an average of 3.5 months in maths and 2.2 months in reading, as of March 2021 (Education Policy Institute, 2021[94]). Recent research from the OECD uses PISA assessment scores in mathematics, science and reading from 15-year-olds in Austria and Scotland to estimate the learning gains accrued by students in one year of schooling across different countries. The study showed that on average, students’ test scores increase by around a quarter of a standard deviation (the equivalent of 25 points) over the course of the year. These findings can provide an upper bound for estimates of how much learning was lost during COVID-related school closures; however, there is a great deal of variation across school systems, and the learning losses of 15-year-olds are not likely to be the same as for other age groups (Avvisati and Givord, 2021[95]).

The negative impacts of learning losses are not equally distributed. Those most affected worldwide are likely to be from low-income households, disadvantaged backgrounds, racial and ethnic minorities and students with learning disabilities (Chapter 6) (Azevedo et al., 2020[92]; Hanushek and Woessmann, 2020[96]). Furthermore, students without access to digital learning tools, without a suitable learning environment, or without support from parents are more at risk (Di Pietro et al., 2020[97]). Data from the 2018 Programme for International Student Assessment (PISA) highlight several potential risk areas: countries that experienced a high number of remote schooling days as a result of COVID closures, where a large student body does not have a quiet place at home to study (Figure 3.17) and/or lack Internet access to follow remote schooling (Figure 3.18), are at particular risk. Some OECD countries used alternate education delivery methods – such as take-home packages, television, mobile phones and/or radio – in places where Internet access was limited for some students (OECD, 2021[90]).

Because of the pandemic, there are fewer learning opportunities for vocational education and training (VET) students. Survey data of VET institutions across the OECD, collected in January-February 2021, show that over the course of 2020 all OECD countries at least partially closed VET institutions (OECD, 2021[98]). As in the case of primary and secondary education, VET institutions used online platforms during closures, however, distance learning is particularly difficult for the practice-focused components of VET education (OECD, 2021[98]). Opportunities for apprenticeships have fallen, limiting opportunities for VET students entering the workforce. Indeed, a survey of enterprises in 114 countries conducted by the International Labour Organization (ILO) from April to June 2020 found that 55% of firms reported that the training of apprentices had been completely disrupted by the pandemic, and another 31% indicated that this had been partially disrupted (ILO, 2021[99]). Shortages of work-based learning opportunities could mean fewer students able to graduate in the short term, but in the long term it may lead to fewer students choosing to enter VET programmes in the first place (OECD, 2021[98]). Given the difficulties in completing the work-based training components of VET, a number of programmes made adjustments to graduation requirements in 2020. Perhaps because of the relaxation of certain restrictions, in 8 of the 14 countries for which data are available, VET graduation rates increased in 2019-20 compared to the previous academic year (OECD, 2021[98]). However, given the changes to the programmes, there are some concerns as to how prepared these students will be in the long term. In general, these disruptions to VET training have a number of consequences, including potential long-term shortages of important skills in the labour market, along with an increase the share of youth not in education, employment or training (NEET) (OECD, 2021[100]).

Figure 3.17. OECD countries with high rates of school closures and large numbers of students without a suitable place to study at home are particularly at risk of falling behind in educational achievement

On-the-job training has also fallen during the pandemic. OECD data show that prior to COVID-19, workers across the OECD spent 4.9 hours per week, on average, in informal learning and 0.7 hours in non-formal learning.14 Simulations show that, during the pandemic, these rates would have dropped to 3.7 hours and 0.6 hours, respectively, i.e. a notable learning loss (OECD, 2021[102]). It is also estimated that low-skilled adults would have been disproportionately affected, as their jobs could not be conducted remotely, and on-the-job learning (non-formal and informal) would have been interrupted entirely. Disruptions to work-based training may not dissipate in the near term. Survey data from the ILO show that 52% of enterprises agreed with the statement that they envisaged a reduction in investment in training and development due to financial constraints “following the pandemic”: these rates were highest for micro, small and medium enterprises (61%) and for governmental and public organisations (52%) (ILO, 2021[99]).

Figure 3.18. Remote learning can be difficult if students lack access to digital tools

3.4. Environmental quality

3.4.1. Air pollution and access to green space

Lockdowns did not bring hoped-for reductions in air pollution, and evidence shows no significant improvements in air quality over the course of 2020

The pandemic has had profound impacts for population groups impacted by climate change and natural disasters as well as on long-term natural capital (Chapter 11); but it has also affected the day-to-day environmental conditions of OECD residents. Air pollution is a risk factor for COVID-19 infection and outcomes. Recent research conducted in the first half of 2020 estimated that long-term exposure to air pollution contributed to around 15% of global COVID-19 deaths (Pozzer et al., 2020[103]). A study in the United States showed that individuals with long-term exposure to high levels of fine particulate matter (PM_2.5) in the air had more severe COVID-19 symptoms, including increased mortality rates (Wu et al., 2020[104]). A study in Canada corroborates this finding, showing a positive association between the incidence of COVID-19 and long-term PM_2.5exposure (Stieb et al., 2020[105]). Conversely, a study by the Office for National Statistics in the United Kingdom found that once ethnicity was controlled for, long-term exposure to air pollution did not have a significant impact on COVID-19 deaths between 27 March and 12 June 2020 in England (ONS, 2020[106]). Even prior to the pandemic, research had demonstrated that exposure to elevated levels of PM_2.5increased the risk of heart disease, stroke, respiratory diseases and respiratory infections (OECD, 2020[107]). Nearly two-thirds of people in OECD countries were exposed to dangerous levels of PM_2.5in 2017 (OECD, 2020[108]).

The first lockdowns to curb the COVID-19 pandemic led to a temporary decrease in some forms of pollution and greenhouse gas emissions – primarily through lowered activity in the transport and industrial sectors (OECD, 2020[109]) (Chapter 11). While CO₂ and NO₂ emissions fell temporarily in most countries with strict business lockdown and confinement measures (Le Quéré et al., 2020[110]; Narain, 2020[111]; Berman and Ebisu, 2020[112]),15 the impacts of confinement on PM_2.5 levels are ambiguous. Studies in the United States, China, India and Australia reported a modest decline in PM_2.5, especially in areas that are more urban, colder and more industrialised (Berman and Ebisu, 2020[112]; He, Pan and Tanaka, 2020[113]; Wang et al., 2020[114]; Kumari and Toshniwal, 2020[115]; Duc et al., 2021[116]). However, other cross-country studies found no impact on PM_2.5 levels (Narain, 2020[111]; Air Quality Expert Group, 2020[117]). The sources of PM_2.5 are multifaceted and include both human-induced factors, such as industrial activity, as well as climate-induced factors.16

The ambiguous relationship between PM_2.5 levels and confinement measures is shown in Figure 3.19 below, which shows PM_2.5 levels from January 2020 to August of 2021 for the capital cities of France, Germany, Italy and the United Kingdom. As can be seen, there is little noticeable trend in the data; rates in 2020 and 2021 all fall within the range of previous values observed on that day from 2017‑19.

With people confined to their homes, concerns about the negative health impacts of indoor air pollution increased (Brunekreef, 2021[118]). According to the World Health Organization, around 4 million people die prematurely from illnesses stemming from household air pollution, including strokes, ischaemic heart disease and chronic obstructive pulmonary disease (COPD) (WHO, 2018[119]). Traditionally, indoor air pollution is associated with the use of cooking and heating fuels: those who have open flames in the home, or stoves fuelled by kerosene, biomass and/or coal, are most at risk. However, recent research suggests that the causes of indoor air pollution go well beyond the fuels themselves. Cooking large meals using modern stoves can cause spikes of indoor PM_2.5 concentrations as high as 250ugm-3 – a level comparable to the world’s most polluted cities (Patel et al., 2020[120]). Living near heavily polluted roadways (EPA, n.d.[121]), as well as the use of household cleaning supplies, tobacco and the presence of mould, can also contribute to indoor pollution, with household ventilation playing an important moderating role (Twilley, 2019[122]). Preliminary research from King’s College London suggests that levels of PM_2.5 exposure in the United Kingdom might have increased during lockdown periods, with the potential decline in outdoor PM_2.5 air pollution more than offset by an increase in exposure to indoor air pollution (Air Quality Expert Group, 2020[117]).

The mental health and well-being benefits of time spent in nature were all the more important when people were confined to their homes

Access to green space improves mental health and overall well-being. Research across a number of OECD countries has shown that those who spend more time in nature, and who have access to green spaces, are more likely to have higher levels of health (both self-reported and objective measures) (de Bell et al., 2020[123]; WHO Regional Office for Europe, 2016[124]), better psychological and emotional well-being (Mayer et al., 2008[125]; Engemann et al., 2019[126]; Crouse et al., 2021[127]; Astell-Burt and Feng, 2019[128]), and are more likely to live longer (Rojas-Rueda et al., 2019[129]). It is not just green space that provides well-being benefits: living and spending time near natural water features (“blue space”) is also associated with reduced risks of mortality, especially for women and older adults (Crouse et al., 2018[130]).

Evidence from COVID-19 corroborates the relationship between time spent in nature and general well-being. Data collected in the People and Nature Survey for England in April through June 2020 found that 85% of adult respondents reported that being in nature made them happy, and that those who spent time in a natural place within the past week had higher levels of reported happiness than those who did not (Natural England, 2020[131]). Data from the first week of April in France showed that those with access to green space (either public or private) reported significantly higher levels of subjective well-being (Recchi et al., 2020[132]).17 An online questionnaire of 3 000 Japanese respondents in June 2020 found that the frequency of greenspace use, and the existence of windows within the home overlooking greenspaces, was associated with higher levels of self-esteem, life satisfaction and happiness, and decreased levels of depression, anxiety and loneliness – even after controlling for socio-demographic characteristics and lifestyle variables (Soga et al., 2021[133]).

Figure 3.19. No clear link between pandemic lockdowns and PM2.5 levels in European cities throughout 2020 and 2021

Box 3.5. Further reading

OECD (forthcoming), Health at a Glance 2021: OECD Indicators, OECD Publishing, Paris
Morgan et al. (2020), “Excess mortality: Measuring the direct and indirect impact of COVID-19”, OECD Health Working Papers, No. 122, OECD Publishing, Paris, https://dx.doi.org/10.1787/c5dc0c50-en
OECD (2021), “Tackling the mental health impact of the COVID-19 crisis: An integrated, whole-of-society response,” OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/0ccafa0b-en
OECD (2021), The State of Global Education: 18 Months into the COVID Pandemic, OECD Publishing, Paris, https://doi.org/10.1787/1a23bb23-en
OECD (2021), Data Insights: Green Recovery, OECD, Paris, http://www.oecd.org/coronavirus/en/data-insights

References

[37] ABS (2020), Household Impacts of COVID-19 Survey, 24-29 June 2020, Australian Bureau of Statistics, https://www.abs.gov.au/statistics/people/people-and-communities/household-impacts-covid-19-survey/24-29-june-2020#emotional-and-mental-wellbeing.

[72] Ahmad, F. and L. Rossen (2021), Provisional drug overdose death counts, National Center for Health Statistics, https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm (accessed on 21 July 2021).

[38] AIHW (2021), The first year of COVID-19 in Australia: Direct and indirect health effects, Summary, Australian Institute of Health and Welfare, https://www.aihw.gov.au/reports/burden-of-disease/the-first-year-of-covid-19-in-australia/summary.

[117] Air Quality Expert Group (2020), Estimation of changes in air pollution emissions, concentrations and exposure during the COVID-19 outbreak in the UK, Department for Environment Food & Rural Affairs, https://uk-air.defra.gov.uk/library/reports.php?report_id=1005.

[11] Arias, E., B. Tejada-Vera and F. Ahmad (2021), “Provisional life expectancy estimates for January through June, 2020”, Vital Statistics Rapid Release, Vol. 10, https://www.cdc.gov/nchs/data/vsrr/VSRR10-508.pdf (accessed on 3 March 2021).

[128] Astell-Burt, T. and X. Feng (2019), “Association of urban green space with mental health and general health among adults in Australia”, JAMA Network Open, Vol. 2/7, pp. e198209-e198209, http://dx.doi.org/10.1001/jamanetworkopen.2019.8209.

[95] Avvisati, F. and P. Givord (2021), “The learning gain over one school year among 15-year-olds: An analysis of PISA data for Austria and Scotland (United Kingdom)”, OECD Education Working Papers, No. 249, OECD Publishing, Paris, https://doi.org/10.1787/d99e8c0a-en.

[92] Azevedo, J. et al. (2020), Simulating the potential impacts of the COVID-19 school closures on schooling and learning outcomes: A set of global estimates, The World Bank, https://www.worldbank.org/en/topic/education/publication/simulating-potential-impacts-of-covid-19-school-closures-learning-outcomes-a-set-of-global-estimates.

[112] Berman, J. and K. Ebisu (2020), “Changes in U.S. air pollution during the COVID-19 pandemic”, Science of the Total Environment, Vol. 739, p. 139864, http://dx.doi.org/10.1016/J.SCITOTENV.2020.139864.

[84] Biddle, N. et al. (2020), “Tracking outcomes during the COVID-19 pandemic (August 2020) – Divergence within Australia”, COVID-19 Briefing Paper, ANU Centre for Social Research and Methods, Australian National University, https://csrm.cass.anu.edu.au/research/publications/tracking-outcomes-during-covid-19-pandemic-august-2020-divergence-within.

[85] Biddle, N. et al. (2020), “Tracking outcomes during the COVID-19 pandemic (November 2020) – Counting the costs of the COVID-recession”, COVID-19 Briefing Paper, ANU Centre for Social Research and Methods, Australian National University, https://csrm.cass.anu.edu.au/research/publications/tracking-outcomes-during-covid-19-pandemic-november-2020-counting-costs-covid.

[65] Blair, G. (2020), “Japan suicides decline as COVID-19 lockdown causes shift in stress factors”, The Guardian, https://www.theguardian.com/world/2020/may/14/japan-suicides-fall-sharply-as-covid-19-lockdown-causes-shift-in-stress-factors.

[118] Brunekreef, B. (2021), Air pollution and COVID-19, Policy Department for Economic, Scientific and Quality of Life Policies, https://www.europarl.europa.eu/RegData/etudes/STUD/2021/658216/IPOL_STU(2021)658216_EN.pdf.

[134] CAMS (n.d.), European Air Quality Information in Support of the COVID-19 Crisis (database), Copernicus Atmosphere Monitoring Service, https://atmosphere.copernicus.eu/european-air-quality-information-support-covid-19-crisis (accessed on 18 August 2021).

[29] CBS (2020), Civil resilience during the Coronavirus crisis among persons aged 21 and over, Central Bureau of Statistics, https://www.cbs.gov.il/en/mediarelease/Pages/2020/Situation-Report-of-Civil-Resilience-during-the-Coronavirus-Crisis.aspx.

[30] CBS (2020), Civil resilience during the Coronavirus crisis, findings from a follow-up survey, Central Bureau of Statistics, https://www.cbs.gov.il/en/mediarelease/Pages/2020/Civil-Resilience-During-the-Coronavirus-Crisis-Findings-from-a-Follow-Up-Survey.aspx.

[40] CDC (2020), Pulse Survey | COVID-19, Centers for Disease Control and Prevention, https://www.cdc.gov/coronavirus/2019-ncov/communication/accomplishments/pulse-survey.html.

[17] Cénat, J. et al. (2020), “Prevalence of mental health problems in populations affected by the Ebola virus disease: A systematic review and meta-analysis”, Psychiatry Research, Vol. 289, p. 113033, http://dx.doi.org/10.1016/J.PSYCHRES.2020.113033.

[81] CEPREMAP (n.d.), Le Bien-être en France (database), http://www.cepremap.fr/Duree.html (accessed on 15 June 2021).

[54] Chan, S. et al. (2006), “Elderly suicide and the 2003 SARS epidemic in Hong Kong”, International Journal of Geriatric Psychiatry, Vol. 21/2, pp. 113-118, http://dx.doi.org/10.1002/GPS.1432.

[130] Crouse, D. et al. (2018), “Associations between living near water and risk of mortality among urban Canadians”, Environmental Health Perspectives, Vol. 126/7, http://dx.doi.org/10.1289/EHP3397.

[127] Crouse, D. et al. (2021), “Residential greenness and indicators of stress and mental well-being in a Canadian national-level survey”, Environmental Research, Vol. 192, http://dx.doi.org/10.1016/j.envres.2020.110267.

[36] DANE (n.d.), Encuesta Pulso Social (database), Departamento Administrativo Nacional de Estadística, https://www.dane.gov.co/index.php/estadisticas-por-tema/encuesta-pulso-social (accessed on 21 September 2021).

[123] de Bell, S. et al. (2020), “Spending time in the garden is positively associated with health and wellbeing: Results from a national survey in England”, Landscape and Urban Planning, Vol. 200, p. 103836, http://dx.doi.org/10.1016/J.LANDURBPLAN.2020.103836.

[60] Debout, M. and J. Fourquet (2016), Les Français et le suicide, La Fondation Jean Jaurès, https://www.jean-jaures.org/publication/les-francais-et-le-suicide/.

[97] Di Pietro, G. et al. (2020), The likely impact of COVID-19 on education: Reflections based on the existing literature and recent international datasets, Publications Office of the European Union, https://op.europa.eu/en/publication-detail/-/publication/b48d50f6-b753-11ea-bb7a-01aa75ed71a1/language-en.

[9] Diaz Ramirez, M., P. Veneri and A. Lembcke (2021), “Where did it hit harder?: The geography of excess mortality during the COVID-19 pandemic”, OECD Regional Development Papers, No. 21, OECD Publishing, Paris, https://doi.org/10.1787/ab4848a4-en.

[116] Duc, H. et al. (2021), “The effect of lockdown period during the COVID-19 pandemic on air quality in Sydney region, Australia”, International Journal of Environmental Research and Public Health 2021, Vol. 18, Page 3528, Vol. 18/7, p. 3528, http://dx.doi.org/10.3390/IJERPH18073528.

[94] Education Policy Institute (2021), EPI research for the Department for Education on pupil learning loss, Education Policy Institute, https://epi.org.uk/publications-and-research/department-for-education-publishes-new-epi-research-on-pupil-learning-loss/.

[55] Elbogen, E. et al. (2020), “Financial strain and suicide attempts in a nationally representative sample of US adults”, American Journal of Epidemiology, Vol. 189/11, pp. 1266-1274, http://dx.doi.org/10.1093/AJE/KWAA146.

[126] Engemann, K. et al. (2019), “Residential green space in childhood is associated with lower risk of psychiatric disorders from adolescence into adulthood”, Proceedings of the National Academy of Sciences of the United States of America, Vol. 116/11, pp. 5188-5193, http://dx.doi.org/10.1073/pnas.1807504116.

[47] Entringer, T. and H. Kröger (2021), “Psychische Gesundheit im zweiten Covid-19 Lockdown in Deutschland”, SOEP Papers on Multidisciplinary Panel Research, No. 1136, DIW SOEP, https://www.diw.de/documents/publikationen/73/diw_01.c.819601.de/diw_sp1136.pdf.

[121] EPA (n.d.), Research on near roadway and other near source air pollution, United States Environmental Protection Agency, https://www.epa.gov/air-research/research-near-roadway-and-other-near-source-air-pollution (accessed on 13 September 2021).

[43] Ettman, C. et al. (2020), “Prevalence of depression symptoms in US adults before and during the COVID-19 pandemic”, JAMA Network Open, Vol. 3/9, p. e2019686, http://dx.doi.org/10.1001/JAMANETWORKOPEN.2020.19686.

[3] European Observatory on Health Systems and Policies & WHO (2020), “Managing health services on a seesaw: Balancing the delivery of essential health services whilst responding to COVID-19”, Eurohealth, Vol. 26/2, https://apps.who.int/iris/handle/10665/336299.

[7] Eurostat (2021), Excess mortality starting to fall back in 2021, after a large death toll, Products Eurostat News, https://ec.europa.eu/eurostat/web/products-eurostat-news/-/ddn-20210416-2.

[27] Eurostat (n.d.), European Health Interview Survey (EHIS) (database), https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Glossary:European_health_interview_survey_(EHIS) (accessed on 16 June 2021).

[78] Gallup (n.d.), Gallup World Poll (database), https://www.gallup.com/178667/gallup-world-poll-work.aspx (accessed on 18 June 2021).

[59] Gaubert, C. (2020), Covid-19 : les suicides, la dernière vague que la crise pourrait déclencher, https://www.sciencesetavenir.fr/sante/cerveau-et-psy/covid-19-une-crise-qui-menace-d-augmenter-le-risque-de-suicides_149069.

[44] Goebel, J. et al. (2019), “The German socio-economic panel (SOEP)”, Jahrbücher für Nationalökonomie und Statistik, Vol. 239/2, pp. 345-360, http://dx.doi.org/10.1515/JBNST-2018-0022.

[76] Government of Alberta Ministry of Health (2020), Alberta COVID-19 opioid response surveillance report Q2 2020, http://open.alberta.ca/licence.

[75] Government of Alberta Ministry of Health (n.d.), Substance Use Surveillance Data, https://www.alberta.ca/substance-use-surveillance-data.aspx#jumplinks-0 (accessed on 17 June 2021).

[74] Graham, C. (2021), America’s crisis of despair: A federal task force for economic recovery and societal well-being, The Brookings Institution, https://www.brookings.edu/research/americas-crisis-of-despair-a-federal-task-force-for-economic-recovery-and-societal-well-being/?utm_campaign=Brookings%20Brief&utm_medium=email&utm_content=111998200&utm_source=hs_email.

[34] Guzmán Mena, J. and A. Tamayo (2020), Ansiedad y depresión, ’verdugos’ de los jóvenes durante la pandemia, Pontificia Universidad Javeriana Colombia, https://www.javeriana.edu.co/pesquisa/ansiedad-y-depresion-verdugos-de-los-jovenes-durante-la-pandemia/.

[79] Hale, T. et al. (2021), “A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker)”, Nature Human Behaviour, Vol. 5/4, pp. 529-538, http://dx.doi.org/10.1038/s41562-021-01079-8 (accessed on 11 June 2021).

[96] Hanushek, E. and L. Woessmann (2020), “The economic impacts of learning losses”, OECD Education Working Papers, No. 225, OECD Publishing, Paris, https://doi.org/10.1787/21908d74-en.

[80] Hardoon, D. (2021), A decade of subjective wellbeing data, What Works Wellbeing, https://whatworkswellbeing.org/blog/a-decade-of-subjective-wellbeing-data/?mc_cid=ca8e7d2523&mc_eid=19c16de127.

[113] He, G., Y. Pan and T. Tanaka (2020), “The short-term impacts of COVID-19 lockdown on urban air pollution in China”, Nature Sustainability 2020 3:12, Vol. 3/12, pp. 1005-1011, http://dx.doi.org/10.1038/s41893-020-0581-y.

[99] ILO (2021), Skilling, upskilling and reskilling of employees, apprentices & interns during the COVID-19 pandemic: Findings from a global survey of enterprises, International Labour Organization, http://www.ilo.org/skills/areas/work-based-learning/WCMS_794569/lang--en/index.htm.

[1] Imperial College London YouGov (2020), Covid 19 Behaviour Tracker Data Hub (database), https://github.com/YouGov-Data/covid-19-tracker (accessed on 13 July 2021).

[71] Japanese National Police Agency (2021), 令和２年中における自殺の状況, https://www.npa.go.jp/safetylife/seianki/jisatsu/R03/R02_jisatuno_joukyou.pdf (accessed on 12 May 2021).

[56] John, A. et al. (2020), “Trends in suicide during the COVID-19 pandemic: Prevention must be prioritised while we wait for a clearer picture”, BMJ, Vol. 371, p. m4352, http://dx.doi.org/10.1136/bmj.m4352.

[4] Jones, N. (2020), “How Coronavirus lockdowns stopped flu in its tracks”, Nature, http://dx.doi.org/10.1038/D41586-020-01538-8.

[73] Katz, J. and M. Sanger-Katz (2021), “‘It’s Huge, It’s Historic, It’s Unheard-of’: Drug Overdose Deaths Spike”, The New York Times, https://www.nytimes.com/interactive/2021/07/14/upshot/drug-overdose-deaths.html.

[57] Knight, B. (2020), Weniger Selbstmorde in Corona-Zeiten, DW, https://www.dw.com/de/weniger-selbstmorde-in-corona-zeiten/a-53620594.

[69] Kodama, S. (2021), “Japan appoints ’minister of loneliness’ to help people home alone”, Nikkei Asia, https://asia.nikkei.com/Spotlight/Coronavirus/Japan-appoints-minister-of-loneliness-to-help-people-home-alone.

[136] Kroenke, K., R. Spitzer and J. Williams (2001), “The PHQ-9: Validity of a brief depression severity measure”, Journal of General Internal Medicine, Vol. 16/9, pp. 606-613, http://dx.doi.org/10.1046/J.1525-1497.2001.016009606.X.

[137] Kroenke, K. et al. (2009), “An ultra-brief screening scale for anxiety and depression: The PHQ-4”, Psychosomatics, Vol. 50/6, pp. 613-621, http://dx.doi.org/10.1176/APPI.PSY.50.6.613.

[46] Kühne, S. et al. (2020), “The need for household panel surveys in times of crisis: The case of SOEP-CoV”, Survey Research Methods, Vol. 14/2, pp. 195-203, https://doi.org/10.18148/srm/2020.v14i2.7748 (accessed on 19 May 2021).

[115] Kumari, P. and D. Toshniwal (2020), “Impact of lockdown measures during COVID-19 on air quality– A case study of India”, International Journal of Environmental Health Research, http://dx.doi.org/10.1080/09603123.2020.1778646.

[110] Le Quéré, C. et al. (2020), “Temporary reduction in daily global CO2 emissions during the COVID-19 forced confinement”, Nature Climate Change 2020 10:7, Vol. 10/7, pp. 647-653, http://dx.doi.org/10.1038/s41558-020-0797-x.

[26] LINE Research Official Blog (2020), 第5回 新型コロナ対策のための全国調査, LINE Research Official Blog, https://research-platform.line.me/archives/35785304.html?utm_source=lg&utm_medium=link&utm_campaign=covid19_survey#3.

[125] Mayer, F. et al. (2008), “Why is nature beneficial?: The role of connectedness to nature”, Environment and Behavior, Vol. 41/5, pp. 607-643, http://dx.doi.org/10.1177/0013916508319745.

[32] Ministerio Desarrollo Social y Familia (2020), Encuesta social COVID-19 (COVID-19 social survey), http://observatorio.ministeriodesarrollosocial.gob.cl/encuesta-social-covid19 (accessed on 20 May 2021).

[5] Morgan, D. et al. (2020), “Excess mortality: Measuring the direct and indirect impact of COVID-19”, OECD Health Working Papers, No. 122, OECD Publishing, Paris, https://dx.doi.org/10.1787/c5dc0c50-en.

[111] Narain, U. (2020), Air pollution: Locked down by COVID-19 but not arrested, The World Bank, https://www.worldbank.org/en/news/immersive-story/2020/07/01/air-pollution-locked-down-by-covid-19-but-not-arrested.

[42] National Center for Health Statistics (2020), Anxiety and depression: Household Pulse Survey - COVID-19, https://www.cdc.gov/nchs/covid19/pulse/mental-health.htm.

[131] Natural England (2020), The people and nature survey for England: Adult data Y1Q1 (April - June 2020) (experimental statistics), Natural England, https://www.gov.uk/government/statistics/the-people-and-nature-survey-for-england-adult-data-y1q1-april-june-2020-experimental-statistics/the-people-and-nature-survey-for-england-adult-data-y1q1-april-june-2020-experimental-statistics (accessed on 24 February 2021).

[83] New Zealand Government (2021), Budget Policy Statement 2021, New Zealand Treasury, https://www.treasury.govt.nz/sites/default/files/2021-02/bps21.pdf.

[48] OECD (2021), A New Benchmark for Mental Health Systems: Tackling the Social and Economic Costs of Mental Ill-Health, OECD Health Policy Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/4ed890f6-en.

[135] OECD (2021), “Adult Learning and COVID-19: How much informal and non-formal learning are workers missing?”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/56a96569-en.

[89] OECD (2021), Education at a Glance 2021: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/b35a14e5-en.

[98] OECD (2021), Implications of the COVID-19 Pandemic for Vocational Education and Training, OECD Publishing, Paris, https://dx.doi.org/10.1787/55afea00-en.

[102] OECD (2021), OECD Skills Outlook 2021: Learning for Life, OECD Publishing, Paris, https://doi.org/10.1787/0ae365b4-en.

[63] OECD (2021), Preventing Harmful Alcohol Use, OECD Health Policy Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/6e4b4ffb-en.

[51] OECD (2021), “Supporting young people’s mental health through the COVID-19 crisis”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/84e143e5-en.

[16] OECD (2021), “Tackling the mental health impact of the COVID-19 crisis: An integrated, whole-of-society response”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/0ccafa0b-en.

[100] OECD (2021), “Teaching and learning in VET: Providing effective practical training in school-based settings”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/64f5f843-en.

[64] OECD (2021), “The effect of COVID-19 on alcohol consumption, and policy responses to prevent harmful alcohol consumption”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/53890024-en.

[8] OECD (2021), The race to vaccinate: Understanding the policy challenges around COVID-19 vaccines, OECD, Paris, https://www.oecd.org/coronavirus/en/vaccines.

[88] OECD (2021), The State of Global Education: 18 Months into the Pandemic, OECD Publishing, Paris, https://doi.org/10.1787/1a23bb23-en.

[90] OECD (2021), The State of School Education: One Year into the COVID Pandemic, OECD Publishing, Paris, https://doi.org/10.1787/201dde84-en.

[109] OECD (2020), “COVID-19 and the low-carbon transition Impacts and possible policy responses”, OECD Policy Responses to the Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/749738fc-en.

[140] OECD (2020), Data Insights: Despite temporary drop, global CO₂ emissions continue to grow, OECD, Paris, http://www.oecd.org/coronavirus/en/data-insights/despite-temporary-drop-global-co2-emissions-continue-to-grow.

[139] OECD (2020), Data Insights: Progress in reducing GHG emissions is insufficient to achieve the 2015 Paris Agreement, OECD, Paris, http://www.oecd.org/coronavirus/en/data-insights/progress-in-reducing-ghg-emissions-is-insufficient-to-achieve-the-2015-paris-agreement.

[107] OECD (2020), Environment at a Glance Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/ac4b8b89-en.

[108] OECD (2020), How’s Life? 2020: Measuring Well-being, OECD Publishing, Paris, https://dx.doi.org/10.1787/9870c393-en.

[10] OECD (2020), OECD Regions and Cities at a Glance 2020, OECD Publishing, Paris, https://dx.doi.org/10.1787/959d5ba0-en.

[101] OECD (2019), PISA 2018 Results (Volume II): Where All Students Can Succeed, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/b5fd1b8f-en.

[12] OECD (2017), Health at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/health_glance-2017-en.

[13] OECD (n.d.), COVID-19 Health Indicators (database), https://stats.oecd.org/index.aspx?queryid=104676# (accessed on 19 September 2021).

[6] OECD (forthcoming), Health at a Glance 2021: OECD Indicators, OECD Publishing, Paris.

[14] OECD (n.d.), Health Status (database), https://stats.oecd.org/Index.aspx?DataSetCode=HEALTH_STAT (accessed on 28 September 2021).

[2] OECD/European Union (2020), Health at a Glance: Europe 2020: State of Health in the EU Cycle, OECD Publishing, Paris, https://dx.doi.org/10.1787/82129230-en.

[52] Oliveira Hashiguchi, T. (2020), “Bringing health care to the patient: An overview of the use of telemedicine in OECD countries”, OECD Health Working Papers, No. 116, OECD Publishing, Paris, https://doi.org/10.1787/8e56ede7-en.

[58] ONS (2021), Deaths from suicide that occurred in England and Wales, Office for National Statistics, https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/articles/deathsfromsuicidethatoccurredinenglandandwales/aprilandjuly2020.

[77] ONS (2021), Deaths related to drug poisoning in England and Wales, Office for National Statistics, https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsrelatedtodrugpoisoninginenglandandwales/2020#measuring-the-data.

[25] ONS (2020), Coronavirus and anxiety, Great Britain, Office for National Statistics, https://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/articles/coronavirusandanxietygreatbritain/3april2020to10may2020.

[24] ONS (2020), Coronavirus and depression in adults, Great Britain: June 2020, Office for National Statistics, https://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/articles/coronavirusanddepressioninadultsgreatbritain/june2020.

[106] ONS (2020), Does exposure to air pollution increase the risk of dying from the coronavirus (COVID-19)?, Office for National Statistics, https://www.ons.gov.uk/economy/environmentalaccounts/articles/doesexposuretoairpollutionincreasetheriskofdyingfromthecoronaviruscovid19/2020-08-13.

[86] ONS (n.d.), Quarterly Personal Well-being Estimates – Seasonally Adjusted (database), https://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/datasets/quarterlypersonalwellbeingestimatesseasonallyadjusted (accessed on 26 February 2021).

[31] Park, H. and S. Yu (2020), “Mental healthcare policies in South Korea during the COVID-19 epidemic”, Health Policy and Technology, Vol. 9/3, pp. 279-280, http://dx.doi.org/10.1016/J.HLPT.2020.06.006.

[120] Patel, S. et al. (2020), “Indoor particulate matter during HOMEChem: Concentrations, size distributions, and exposures”, Environmental Science & Technology, Vol. 54/12, pp. 7107-7116, http://dx.doi.org/10.1021/ACS.EST.0C00740.

[22] Patel, V. et al. (2018), “The Lancet Commission on global mental health and sustainable development””, The Lancet Commissions, Vol. 392/10157, pp. 1553-1598, https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(18)31612-X/fulltext.

[82] Perona, M. and C. Senik (2021), Note de l’Observatoire du Bien-être n°2021-04 : Le Bien-être des Français – Mars 2021, CEPREMAP, http://www.cepremap.fr/en/2021/04/note-de-lobservatoire-du-bien-etre-n2021-04-le-bien-etre-des-francais-mars-2021/.

[53] Pirkis, J. et al. (2021), “Suicide trends in the early months of the COVID-19 pandemic: An interrupted time-series analysis of preliminary data from 21 countries”, The Lancet Psychiatry, Vol. 8/7, pp. 579-588, http://dx.doi.org/10.1016/S2215-0366(21)00091-2.

[103] Pozzer, A. et al. (2020), “Regional and global contributions of air pollution to risk of death from COVID-19”, Cardiovascular Research, Vol. 116/14, pp. 2247-2253, http://dx.doi.org/10.1093/CVR/CVAA288.

[15] Public Health England (n.d.), Wider Impacts of COVID-19 on Health (WICH) Monitoring Tool (database), https://analytics.phe.gov.uk/apps/covid-19-indirect-effects/ (accessed on 4 June 2021).

[61] Rabin, R. (2021), “U.S. suicides declined over all in 2020 but may have risen among people of color”, The New York Times, https://www.nytimes.com/2021/04/15/health/coronavirus-suicide-cdc.html.

[132] Recchi, E. et al. (2020), “Lockdown for all, hardship for some: Insights from the first wave of the CoCo project”, HAL, https://halshs.archives-ouvertes.fr/halshs-03070206/document.

[129] Rojas-Rueda, D. et al. (2019), “Green spaces and mortality: A systematic review and meta-analysis of cohort studies”, The Lancet Planetary Health, Vol. 3/11, pp. e469-e477, http://dx.doi.org/10.1016/S2542-5196(19)30215-3.

[70] Ryall, J. (2020), “South Korea′s soaring suicide, self-harm rates pinned on pandemic”, DW, https://www.dw.com/en/south-koreas-soaring-suicide-self-harm-rates-pinned-on-pandemic/a-54931167.

[67] Sakamoto, H. et al. (2021), “Assessment of suicide in Japan during the COVID-19 pandemic vs previous years”, JAMA Network Open, Vol. 4/2, p. e2037378, http://dx.doi.org/10.1001/JAMANETWORKOPEN.2020.37378.

[35] Sanabria-Maxo, J. et al. (2021), Psychometric properties of the online version of the Patient Health Questionnaire-4 in a large Colombian sample: Results from the PSY-COVID study, PSY-COVID, https://www.researchgate.net/publication/354610035_Psychometric_properties_of_the_online_version_of_the_Patient_Health_Questionnaire-4_in_a_large_Colombian_sample_Results_from_the_PSY-COVID_study.

[45] SOEP-CoV (2021), SOEP-CoV The spread of the coronavirus in Germany: Socio-economic factors and consequences, Funded by the German Federal Ministry of Education and Research (BMBF); the data can be accessed via the research data center of the SOEP, https://www.soep-cov.de/Startseite/.

[133] Soga, M. et al. (2021), “A room with a green view: The importance of nearby nature for mental health during the COVID-19 pandemic”, Ecological Applications, Vol. 31/2, p. e2248, http://dx.doi.org/10.1002/eap.2248.

[93] Soland, J. et al. (2020), The impact of COVID-19 on student achievement and what it may mean for educators, The Brookings Institution, https://www.brookings.edu/blog/brown-center-chalkboard/2020/05/27/the-impact-of-covid-19-on-student-achievement-and-what-it-may-mean-for-educators/.

[138] Spitzer, R. et al. (2006), “A brief measure for assessing generalized anxiety disorder: The GAD-7”, Archives of Internal Medicine, Vol. 166/10, pp. 1092-1097, http://dx.doi.org/10.1001/ARCHINTE.166.10.1092.

[23] Statistics Canada (2020), Canadians’ mental health during the COVID-19 pandemic, The Daily, https://www150.statcan.gc.ca/n1/daily-quotidien/200527/dq200527b-eng.htm.

[87] Stats NZ (n.d.), Wellbeing Statistics: March 2021 Quarter (database), https://www.stats.govt.nz/information-releases/wellbeing-statistics-march-2021-quarter (accessed on 15 June 2021).

[105] Stieb, D. et al. (2020), “An ecological analysis of long-term exposure to PM2.5 and incidence of COVID-19 in Canadian health regions”, Environmental Research, Vol. 191, p. 110052, http://dx.doi.org/10.1016/j.envres.2020.110052.

[68] Tanaka, T. and S. Okamoto (2021), “Increase in suicide following an initial decline during the COVID-19 pandemic in Japan”, Nature Human Behaviour 2021 5:2, Vol. 5/2, pp. 229-238, http://dx.doi.org/10.1038/s41562-020-01042-z.

[28] Tel Aviv University (2020), Israelis suffering from anxiety and depression at “disturbing” rates, Tel Aviv University, https://english.tau.ac.il/news/depression-covid19.

[20] Tsang, H., R. Scudds and E. Chan (2004), “Psychosocial impact of SARS”, Emerging infectious diseases, Vol. 10/7, pp. 1326-1327, http://dx.doi.org/10.3201/EID1007.040090.

[122] Twilley, N. (2019), “The hidden air pollution in our homes”, The New Yorker, https://www.newyorker.com/magazine/2019/04/08/the-hidden-air-pollution-in-our-homes.

[18] Tzeng, N. et al. (2020), “What could we learn from SARS when facing the mental health issues related to the COVID-19 outbreak? A nationwide cohort study in Taiwan”, Translational Psychiatry 2020 10:1, Vol. 10/1, pp. 1-9, http://dx.doi.org/10.1038/s41398-020-01021-y.

[33] UNED (2021), Salud mental y COVID-19 Costa Rica 2020, https://investiga.uned.ac.cr/2021/01/29/salud-mental-y-covid-19-costa-rica-2020/.

[91] UNESCO (n.d.), Global Monitoring of School Closures, COVID-19 (database), United Nations Educational, Scientific and Cultural Organization, https://en.unesco.org/covid19/educationresponse (accessed on 5 October 2021).

[21] United Nations (2020), COVID-19 and the Need for Action on Mental Health, https://unsdg.un.org/sites/default/files/2020-05/UN-Policy-Brief-COVID-19-and-mental-health.pdf.

[39] US Census Bureau (n.d.), Measuring Household Experiences during the Coronavirus Pandemic (database), https://www.census.gov/data/experimental-data-products/household-pulse-survey.html (accessed on 13 September 2021).

[66] Wang, S., R. Wright and Y. Wakatsuki (2020), “In Japan, more people died from suicide last month than from Covid in all of 2020”, CNN, https://edition.cnn.com/2020/11/28/asia/japan-suicide-women-covid-dst-intl-hnk/index.html.

[114] Wang, Y. et al. (2020), “Four-month changes in air quality during and after the COVID-19 lockdown in six megacities in China”, Environmental Science & Technology Letters, Vol. 7/11, pp. 802-808, http://dx.doi.org/10.1021/ACS.ESTLETT.0C00605.

[50] WHO (2021), Second round of the national pulse survey on continuity of essential health services during the COVID-19 pandemic, World Health Organization, https://www.who.int/publications/i/item/WHO-2019-nCoV-EHS-continuity-survey-2021.1.

[49] WHO (2020), The Impact of COVID-19 on Mental, Neurological and Substance Use Services: Results of a Rapid Assessment, World Health Organization, https://www.who.int/publications/i/item/978924012455 (accessed on 16 December 2020).

[119] WHO (2018), Household Air Pollution and Health, World Health Organization, https://www.who.int/news-room/fact-sheets/detail/household-air-pollution-and-health.

[62] WHO Regional Office for Europe (2020), Alcohol does not protect against COVID-19; Access should be restricted during lockdown, World Health Organization, https://www.euro.who.int/en/health-topics/disease-prevention/alcohol-use/news/news/2020/04/alcohol-does-not-protect-against-covid-19-access-should-be-restricted-during-lockdown.

[124] WHO Regional Office for Europe (2016), Urban green spaces and health - A review of evidence, World Health Organization, Copenhagen, https://www.euro.who.int/en/health-topics/environment-and-health/urban-health/publications/2016/urban-green-spaces-and-health-a-review-of-evidence-2016.

[41] Wozniak, A. et al. (2020), COVID Impact Survey, National Opinion Research Center, https://www.covid-impact.org/.

[104] Wu, X. et al. (2020), “Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study”, Science Advances, Vol. 6/45, http://dx.doi.org/10.1101/2020.04.05.20054502.

[19] Yip, P. et al. (2010), “The impact of epidemic outbreak: The case of severe acute respiratory syndrome (SARS) and suicide among older adults in Hong Kong”, Crisis, Vol. 31/2, pp. 86-92, http://dx.doi.org/10.1027/0227-5910/A000015.

Notes

← 1. The first quarter of 2020 was mostly unaffected by COVID-19. Due to a mild 2019-20 influenza season, excess mortality in most European countries was below the 2015-19 average for the period January to early March 2020 (Morgan et al., 2020[5]).

← 2. Data for some countries refer to the date a death is registered, rather then the date the death actually occurred. There may be variation across countries in delays in registration over holidays and weekends. In the longer term, data referring to date of registration can be retroactively amended to date of occurrence, however in the short-run weekly statistics may be influenced (Morgan et al., 2020[5]). Because all-cause mortality data are dynamic, changes (usually marginal) may be observed depending on when data were extracted, as data may be retroactively updated. All data for excess mortality figures in this report were extracted from the COVID-19 Health Indicators database (OECD, n.d.[13]) on 19 September 2021.

← 3. The drops in life expectancy reported in 2020 for many countries are calculated based on the age-specific mortality rates in 2020, which were highly affected by COVID-19. Therefore it is unclear whether the 2020 drops in life expectancy will last in the long term, or quickly return to 2019 levels.

← 4. The full Patient Health Questionnaire (PHQ) contains 59 questions, with modules focusing on mood, anxiety, alcohol, eating and somatoform disorders. The PHQ-4 screening tool is a short, four-question survey administered to respondents to gauge their mental condition, and to identify the presence and severity of depression and anxiety. PHQ-4 pulls two depression-related questions from the PHQ-9/8 (itself called the PHQ-2), and two anxiety-related questions from the Generalised Anxiety Disorder (GAD-7) questionnaire (itself called the GAD-2). Thus, the PHQ-4 is a combination of PHQ-2 and GAD-2. All question items are added together to provide a total score of mental distress: 0-2 normal, 3-5 moderate, 9-12 severe. A total score greater than or equal to 2 for the first two questions, pulled from the GAD-7, indicates that the respondent is at risk for anxiety. A total score greater than or equal to 2 for the final two questions, pulled from the PHQ-8, indicates that the respondent is at risk for depression (Kroenke et al., 2009[137]). The self-reported values from the PHQ surveys have been validated in separate studies comparing survey outcomes with actual diagnostic interviews with mental health professionals.

← 5. The PHQ-8 and PHQ-9 questionnaires are a common shortened version of the full PHQ survey (see the above endnote for more information). PHQ-9 is a nine-question survey designed to detect the presence and severity of depression disorders. The PHQ-8 questionnaire is the same but removes the final question regarding suicidal ideation. In the PHQ-8 survey, all items are added together to provide a total score of depression severity: 0-4 none, 5-9 mild depression, 10-14 moderate depression, 15-19 moderately severe depression, 20-24 severe depression (Kroenke, Spitzer and Williams, 2001[136]).

← 6. These rates are lower than those found in the YouGov Covid Data Hub survey; however, respondents were asked different questions. It is worth noting that the LINE data are preliminary and unweighted.

← 7. Data were weighted post-hoc to be nationally representative by sex, age, education, labour market status, labour sector and province of residence.

← 8. The Generalised Anxiety Disorder Questionnaire (GAD) identifies the risk of anxiety. Similar to the PHQ, data are self-reported. The validity of the tool has been backed by studies matching rates of clinical diagnoses of anxiety in respondents (Spitzer et al., 2006[138]).

← 9. Treatment adherence refers to the extent to which an individual’s behaviours are in compliance with the recommendations of their healthcare provider: taking medication, participating in counselling, instating lifestyle changes, etc.

← 10. 2020 data refer to Q1 and Q2 only.

← 11. Excess deaths are measured as the increase in the number of reported deaths from all causes in 2020 compared to the average from 2015-19 for the same period and are sourced from (OECD, n.d.[13]). Average values for “low excess deaths” refer to Australia, Austria, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Latvia, Lithuania, Norway, New Zealand, Poland, Portugal, the Slovak Republic, Slovenia, Sweden and Switzerland; those for “high excess deaths” include Belgium, Canada, Chile, Colombia, Italy, Israel, Mexico, the Netherlands, Spain, the United Kingdom and the United States. The remaining indicators are all sourced from the COVID-19 Government Response Tracker (Hale et al., 2021[79]). The stringency index runs from 0 “least stringent” to 100 “most stringent”, and combines data on school closures, workplace closures, cancellation of public events, restrictions on gatherings, closing of public transport, stay-at-home requirements, restrictions on internal movement and international travel controls. “Low stringency” includes Australia, Austria, Belgium, Denmark, Estonia, France, Germany, Greece, Hungary, Iceland, Italy, Japan, Korea, Latvia, Lithuania, New Zealand, Poland, the Slovak Republic and Switzerland; “high stringency” includes Canada, Chile, Colombia, Costa Rica, Finland, Ireland, Israel, Mexico, the Netherlands, Norway, Portugal, Slovenia, Spain, Sweden, Turkey, the United Kingdom, and the United States. “Low school closures” include Australia, Austria, Belgium, Denmark, Estonia, France, Germany, Greece, Hungary, Iceland, Japan, Korea, New Zealand, Poland, the Slovak Republic, Switzerland and the United Kingdom; “high school closures” include Canada, Chile, Colombia, Costa Rica, Latvia, Finland, Ireland, Israel, Italy, Lithuania, Mexico, Netherlands, Norway, Portugal, Slovenia, Spain, Sweden, Turkey and the United States. Stay at home policies refer to government policies on a scale from 0 (no restrictions) to 3 (most restrictions) that (1) recommend not leaving the house, (2) require not leaving the house aside from a range of exceptions, (3) or require not leaving the house aside for very minimal exceptions. “Low stay at home” includes Australia, Austria, Belgium, Estonia, France, Germany, Greece, Iceland, Ireland, Italy, Korea, Latvia, New Zealand, Norway, Poland, the Slovak Republic, Spain, Switzerland the United Kingdom; “high stay at home” includes Canada, Chile, Colombia, Costa Rica, Denmark, Finland, Hungary, Israel, Japan, Lithuania, Mexico, the Netherlands, Portugal, Slovenia, Sweden, Turkey and the United States. The economic support index also runs from 0 (least amount of support) to 100 (most amount of support) and encompasses financial support and debt relief to households. “Low economic support” includes Australia, Belgium, Canada, Costa Rica, Estonia, Finland, Germany, Hungary, Korea, Latvia, Lithuania, Mexico, New Zealand, Norway, Slovenia, Sweden, Switzerland and the United States; “high economic support” includes Austria, Chile, Colombia, Denmark, France, Greece, Iceland, Ireland, Israel, Italy, Japan, the Netherlands, Poland, Portugal, the Slovak Republic, Spain, Turkey and the United Kingdom.

← 12. Many countries that placed a priority on keeping primary and secondary schools open did so while maintaining strict measures to limit or prevent social mixing in other contexts, such as in hospitality, retail and cultural sectors.

← 13. The Survey on Joint National Responses to COVID-19 School Closures is a collaborative effort between the United Nations Educational, Scientific and Cultural Organization (UNESCO), the UNESCO Institute for Statistics (UIS), the United Nations Children's Fund (UNICEF), the World Bank (WB) and the OECD. The most recent round of data collection took place from January to February 2021.

← 14. The OECD defines “non-formal” learning as participating in activities such as workshops and employer-provided trainings, and “informal learning” as learning from others, learning by doing and learning new things at work (OECD, 2021[135]).

← 15. Although CO₂ emissions fell during lockdowns, the drop was only temporary, and global emissions are continuing to grow (OECD, 2020[140]). Similarly, lowered greenhouse gas emissions due to the pandemic will not be sufficient to reach the rates agreed in the 2015 Paris Agreement (OECD, 2020[139]).

← 16. In the United Kingdom, PM_2.5 levels during the spring 2020 were higher than in the same period in 2019, primarily due to meteorological conditions (Air Quality Expert Group, 2020[117]).

← 17. These findings come from the CoCo survey, a component of the “Coping with Covid-19: Social Distancing, Cohesion and Inequality in 2020 France” project. The CoCo survey is a part of the French Longitudinal Internet Studies for Social Sciences panel (ELIPSS), a panel survey maintained by the Centre de données socio-politiques (CDSP) at SciencesPo. ELIPSS is a panel of 1 400 French residents from 2012 through today. The sample is selected from census data. Responses are weighted to account for design effects, differential response refusal weights, and post-stratification weights for sex, age, education and region.

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Countries

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Publications

Publications

Featured publications

Data

Data

Featured data

News & Events

News & Events

Featured Events

About OECD

About

Engage with us

Work with us

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment