1. The impact of the COVID-19 pandemic on Latin American and Caribbean healthcare systems

Nicolas Larrain
OECD Health Division
Paris
Gabriel Di Paolantonio
OECD Health Division
Paris
Tom Raitzik Zonenschein
OECD Health Division
Paris
Rushay Naik
OECD Health Division
Paris
Frederico Guanais
OECD Health Division
Paris

Three years have passed since the COVID-19 crisis galvanised attention to health systems worldwide. Between December 2019 and October 2022, 631 million cases and 6.59 million deaths were reported. Moreover, the re-prioritisation of healthcare resources to deal with the immediate pandemic response has affected regular healthcare, generating significant backlogs of postponed services in every level of care. An estimated 2 273 620 excess deaths were experienced in the region in 2020 and 2021 compared to the expected deaths for these years as estimated by WHO (WHO, 2022[1]). At the same time, social and economic effects driven by the loss of life and the multidimensional effects of containment measures will have long lasting implications for health systems and, ultimately, the people they serve.

The effect of COVID-19 in Latin America and the Caribbean (LAC) is proportionally much larger when compared to other regions of the world. By July 2022, LAC accounted for only 8.5% of the world population but 13% of the world’s documented COVID-19 cases and 27% of the documented deaths (Herrera et al., 2022[2]). The health, social, and economic effects in the region have been catastrophic and have brought long-standing weaknesses and challenges of health systems to light (ECLAC, 2022[3]).

Previous literature has analysed extensively different aspects of the health and socio-economic impact of the pandemic (ECLAC, 2022[3]; Herrera et al., 2022[2]; OPS/PAHO, 2021[4]). However, there is still a significant knowledge gap in the understanding of the determinants of the pandemic impact and the characteristics of the most effective emergency responses.

In this chapter, we examine the mortality impact of the COVID-19 crisis and the factors that determined its severity across 33 countries in LAC1 as well as, whenever feasible, the OECD average. Given that four LAC countries are part of the OECD (Chile, Colombia, Costa Rica and Mexico), and two more have started the accession process (Brazil and Peru) readers should be aware of some overlap between groups. The impact of COVID-19 is evaluated from the perspective of the performance of LAC health systems at three levels: Outcomes, process, and structure (Donabedian, 1988[5]).

  • At the outcome level, the focus is on the mortality impact resulting from the COVID-19 virus itself and as extended consequences of the health emergency. Further, the chapter analyses disruptions to the routine provision of care as an intermediate outcome, in an effort to understand how the fragility of LAC health systems influenced the loss of life during the pandemic.

  • At the process level, the analysis considers and compares the actions taken by the authorities to manage and mitigate the effects of COVID-19, together with the public response and adherence to these measures.

  • At the structure level, the chapter builds on the analysis of variables defining care delivery, including health status, equipment, and human resources, explored in chapters three to nine of this publication. The analysis focuses on the main characteristics influencing preparedness to deal with the COVID-19 health emergency.

Assessing the pandemic impact in this order is essential for the objectives of the chapter. Starting at the outcome level allows for an analysis of the pandemic’s most devastating consequence in the LAC region, the loss of life. Later, this is compared to performance in intermediate outcomes and at process level to understand the effectiveness of the pandemic response; and at the structure level, to examine the importance of baseline health system capacities.

The selection of indicators at each assessment level prioritises the availability of information for a wider set of LAC countries (Box 1.1) and actionability of these measures towards health system improvement (Carinci et al., 2015[6]). Given that no formal causal analysis is intended, the report considers these criteria to be most important for an accurate and comprehensive understanding of the performance of LAC countries during the pandemic.

The chapter is organised as follows. LAC countries’ performance at the outcome level is presented in Section 1.2. The analysis of routine and essential care disruptions is presented in Section 1.3. Process-level performance is assessed in Section 1.4, focusing on countries’ pandemic response. Structural components and their effect on COVID-19 mortality impact is addressed in Section 1.5. Finally, Section 1.6 discusses the results of the assessments at the three levels in aggregate and provides policy recommendations for strengthening resilience of LAC health systems.

Excess mortality is calculated as the difference between expected and estimated deaths from all causes (Figure 1.1). In WHO’s calculation of excess deaths, estimated deaths are based on historic data from five to 20 years before the pandemic (WHO, 2022[1]). Excess mortality is considered a more accurate indicator of the direct health impact of COVID-19 due to limitations in the cross-country comparability of COVID-19 crude mortality data (Box 1.2). It is estimated that LAC had 2 273 620 excess deaths combining 2020 and 2021, the later year being by far the deadliest of the pandemic (Figure 1.2). This represents 15% of the total excess deaths in the world (WHO, 2022[7]). Considering LAC only accounted for close to 8.5% of the world population in this period (UN, 2022[19]), the disproportionate effect of COVID-19 in the region is undeniable.

Figure 1.1 presents the accumulated estimated excess mortality per 100 000 population for years 2020 and 2021. Only six of the 33 countries experienced particularly high excess deaths above the LAC average: Peru, Bolivia, Mexico, Ecuador, Saint Vincent and the Grenadines, and Guyana. The cases of Saint Vincent and Guyana need to be taken with caution, as their small population might exacerbate the perceived impact of the number of COVID-19 related deaths. Peru experienced the highest impact within the group (and region) with 873 excess deaths per 100 000, 150% more than the LAC average. The average of these six countries is 560 excess deaths per 100 000 (60% higher than the LAC average). Together, this group accounted for 48% of all excess deaths in LAC in the period 2020-21, but only 25% of the region’s population.

Comparing the performance of the LAC region to the OECD average can be used as the boundary for defining a second group of countries, with excess mortality higher than the OECD average but lower than LAC average. Colombia leads this group with 323 excess deaths per 100 000 population. This is 8% lower than the LAC average and 40% higher than the OECD average. The group (Colombia, Brazil, Paraguay, Guatemala, El Salvador and Bahamas) has an average of 285 estimated excess deaths per 100 000 population, 25% above OECD average and 18% below LAC average. These six countries accounted for 41% of all the excess deaths in the region while representing 45% of the region’s population.

The rest of the region can be interpreted as a third group of countries, with lower than OECD average excess deaths. This 21-strong group has an average of 81 estimated excess deaths per 100 000. This is 64% less than the OECD average. Together, this group accounted for 11% of the excess deaths and 30% of the region’s population.

However, the group of countries performing better than the OECD average is composed by several countries affected by the limitations of the excess mortality indicator (Box 1.2). For countries affected by these limitations, the excess mortality indicator does not accurately represent the health impact of the pandemic and, hence, should not be used for comparison. In this line, countries suitable for comparison in the group that experienced milder health effects can be identified using the following three criteria:

  • Excess mortality accumulated in 2020 and 2021 below the OECD average, given that these were the countries with the lowest excess mortality during the first two years of the pandemic.

  • Age-standardised mortality rate (all causes) below the LAC average in 2019. This criterion enables filtering out countries where high overall mortality disturbs the accuracy of the excess mortality indicator to signal a positive pandemic response. Age-standardised mortality rate is taken from the Global Burden of Disease Collaborative Network (IHME, 2021, 2021[22]).

  • Countries with more than 500 000 inhabitants. Finally, this criterion filters out countries where unintended consequences of the pandemic response, such as lower traffic and injury deaths are usually more volatile. Furthermore, estimations of excess mortality for countries with small population are considerably less precise (WHO, 2022[1]).

The resulting list of countries with positive outcomes in terms of the health effect of COVID-19 is composed (in order of population size) of Argentina, Venezuela, Chile, Cuba, Costa Rica, Panama, Uruguay, Jamaica, and Trinidad and Tobago. As presented by Herrera et al (2022[2]), a different estimation of excess mortality by the COVID-19 Excess Mortality Collaborators 2022 shows Venezuela as the only country in this group that considerably changes their relative position in the region and presents a higher than LAC average excess mortality. Consequently, conclusions regarding this country will be examined with care.

The stark differences between reported deaths and excess mortality (Figure 1.3), validates the use of excess mortality as a more accurate indicator of the pandemic effect on population health. Further, the age and sex stratification shows that males carried a heavier burden in terms of mortality during the pandemic, in line with previous studies (Huang et al., 2021[23]; Gebhard et al., 2020[24]; de Souza et al., 2021[25]; PAHO, 2021[26]). Finally, the difference between reported COVID-19 deaths and excess mortality during this period increased with age, either because of under-reporting COVID-19 deaths (OECD, 2021[27]) or due to deaths by other causes.

Around 60% of estimated excess deaths were in men and 75% of estimated excess deaths were attributed to people aged 60 years or more. When comparing estimated excess deaths by 100 000 population, males aged 70 years or more had an 80% higher risk of death than females from the same age group. The risk of death for females is on average 50% lower for every age group. These findings are consistent with previous literature (OPS/PAHO, 2021[4]).

Vulnerability and socio-economic inequities determined the burden of COVID-19 direct health impacts. This is clearly seen in the differences in mortality and hospitalisation of COVID-19 among vulnerable populations. As highlighted by ECLAC (ECLAC/PAHO, 2021[28]), socio-economic vulnerability is correlated with COVID-19 infection, severity of the disease, and death (de Souza et al., 2021[25]). Low-income areas had disproportionately high numbers of COVID-19 deaths in Brazil (Bermudi et al., 2021[29]) and Argentina (Macchia et al., 2021[30]). At the same time, higher education has shown a higher-than-expected protective effect during 2020 in Chile (Bilal, Alfaro and Vives, 2021[31]) and Brazil (SUS, 2022[32]). Studies exploring ethnic differences in COVID-19 deaths in 2020 in Brazil, Chile, Colombia, Mexico, and Peru, indicated that mortality affected non-white populations disproportionately (NU. CEPAL/German Agency for International Cooperation, 2021[33])). Further, a positive association has been found between countries with more COVID-19 cases and a higher migrant population (Migration data portal, 2022[34]). This is consistent with other sources, stating that migrants in Latin America are at a disadvantage in terms of pandemic preparation (Cabieses et al., 2020[35]).

The health workforce at the front line of the pandemic response is at higher risk of COVID-19 infection. During 2020, more than 1.3 million cases among health personnel were reported in the region and more than 6 000 died due to the disease (OPS/PAHO, 2021[4]). Women comprise a disproportionately greater share of frontline health personnel in the region: 86% of nurses in the region are female and 70% of the worldwide pandemic response front line are female (PAHO, 2021[26]; OECD, 2020[36]). This translates to 72% of COVID-19 cases (from March 2020 until January 2021) in health personnel being among women (OPS/PAHO, 2021[4]).

In 2020, COVID-19 disrupted all healthcare services worldwide at some level. During 2021, the disruption to essential health services remained in 90% of countries. Workforce shortages and supply chain disruptions were reported in more than 30% of countries responding to the WHO PULSE survey (WHO, 2022[8]). Primary and rehabilitative, palliative, and long-term care remained the most heavily affected services during this time (Kuehn, 2021[37]).

An important lesson from the pandemic is that health systems need to strengthen their resilience: the ability to prepare for shocks, absorb disruptions and recover as quickly as possible with minimal cost, and adapt by learning lessons to improve performance and manage future risks (OECD, 2023[38]). For this purpose, it is important to understand the extent to which COVID-19 altered the normal functioning of healthcare and how this reflected on health outcomes during the pandemic. Measuring the extent of disruption to healthcare services generates important information for minimising negative consequences and to recover as quickly as possible to respond to the next emergency.

In 14 LAC countries on average during 2020, 16% of urban households and 21% of rural household reported not being able to attend healthcare services when needed (Figure 1.4). In Ecuador, Peru and Bolivia (three countries in the highest excess deaths group) this percentage was higher than 27% of households in the country. While the relation between the severity of the pandemic and the disruption of in healthcare is undoubtedly endogenous, the fact that the disruption presented in the graph was one year before the highest point of excess deaths in the region suggests that the early disruption of healthcare was a key determinant in the later severity of excess deaths.

Most countries showed important recovery capacities during 2021, where on average only 2.7% of households didn’t receive medical care when needed. The biggest disruptions during the second year of the pandemic coincide with countries that had high excess deaths. On the contrary, in several countries with lower than OECD average excess deaths (Costa Rica, Argentina, Honduras and Chile) there were lower disruptions during 2020, with between 18% (rural Chile) and 3% (Rural Argentina) of households reporting disrupted care and almost none during 2021 (below 2%). The exception is Mexico that reported very low percentage of households perceiving disrupted care during the two years (less than 7.5%), but at the same time had the third highest estimated number of excess deaths in the region. These findings are congruent with those exposed by ECLAC (ECLAC, 2022[3]) and the World Bank (Herrera et al., 2022[2]).

Reasons for people not receiving care when needed can be classified in two ways. The first relates to the supply of services, when no care is available. Understanding the extent of this effect allows countries to provide the appropriate focus on strengthening care provision. The other option is on the demand side when people do not seek care. The varying reasons behind this choice affect policy recommendations: those who elect to not seek care as a personal decision account for different considerations than those who did not seek care because of a perceived mandate such as a stay-at-home restriction. In the latter case, the communication strategy of the authorities is a critical issue, and authorities must ensure that options to access care are well-communicated to hard-to-reach populations.

Figure 1.5 shows that out of the households that did not receive medical care when needed, 24% attributed this to no availability of medical personnel. This is directly related to the redistribution of healthcare resources because of the pandemic. In the same category, 0.2% of households could not receive medical care because the health facility was full. Furthermore, the “other reasons” (32%) category includes answers to the questions “hospital/clinic did not have enough supplies or tests” and “medical facility was closed” as well as “other reasons”. In this sense, it can also be considered a supply-side barrier.

On the demand side, 17% of households reported not receiving medical care because of being afraid of catching COVID-19, while 10% reported lack of money. These reasons resonate with the heavy health, social and economic impacts of the pandemic, and consequent high levels of mental distress. Finally, restrictions and a lack of transportation affected 15% of households not receiving healthcare when needed.

The impact of restrictions is particularly notable, as they are directly related to the planning of the authorities when imposing restrictions and remained fully within their control. Securing essential services such as transportation and correctly informing people of the nature of the restrictions is essential for the effectiveness of the pandemic response. Health systems also have space for action regarding the supply-related disturbances. The redistribution of resources during the pandemic response should secure services for important health needs. Finally, while health systems can only indirectly affect the disturbances related to the demand side, effective communication campaigns the distribution of protective equipment, and interventions for reducing the financial burden of healthcare, can be very effective in improving health systems responsiveness during a pandemic.

The pandemic disrupted several essential services. As exposed by the survey conducted by the University of Maryland in collaboration with Facebook (2020[14]), seasonal flu vaccination decreased in at least 10 countries (with available data2) between the flu season 2019 (June 2019 to Feb 2020) and the one in 2020 (after July 2020). The average decrease in the proportion of immunised population in these countries was to the order of 27%. On the other hand, Colombia and Honduras saw a small increase (2 percentile points) in the proportion of flu-immunised people. Furthermore, Diphtheria, tetanus toxoid and pertussis (DTP3) immunisation coverage3 amongst children aged 1 decreased from 90% to 82% on average in LAC (Chapter 7 – Childhood vaccination programmes).

Additionally, in 2020, medical consultations decreased in Brazil (30%), Mexico (9% – estimated), Costa Rica (17%), and Chile (32%) (OECD, 2022[18]). Furthermore, medical consultations of children decreased 67% in Chile and antenatal consultations decreased 56% in Peru (OECD, 2022[39]).

Hospitalizations for non-COVID-19 related reasons diminished considerably. Even though data in the region is scarce, between 2019 and 2020 Chile and Costa Rica decreased their non-COVID related hospitalisation discharges by 23% (OECD, 2022[18]). In 2020, Peru and Guatemala (out of 24 LAC countries reporting data) reported disruptions larger than 5% to 24-hour emergency room services. By the end of 2021, the disruption reached more than 50% of services in both countries. At the same time, ambulance services reached disruptions of more than 26% (WHO, 2022[8]). Excess deaths in these two countries were among the highest in the region and peaked in the second half of 2021.

While standardised and comparable data for the region is scarce, information can be obtained from individual cases. General care was drastically affected for people with cancer, heart attacks and strokes in Chile (Pacheco et al., 2021[40]). For example, from April to September 2020 diabetes consultations dropped 80% and hypertension consultations dropped 81% when compared to the previous year. A similar situation occurred in Mexico, where one source reports drops of 26% and 28% in diabetes and hypertension respectively (Arsenault et al., 2022[41]), while another reports drops of 32% for both types of consultations (Doubova et al., 2021[42]). In Peru, diabetes treatment dropped 8% during 2020, while no disruptions to hypertension treatment were recorded (INEI, 2021[43]). In Costa Rica, the decrease was of 32% for diabetes and 20% for hypertensive patients (CCSS, 2020[44]). Additionally, diabetes screening coverage rates decreased 32% and hypertension screening coverage rates decreased 23% in both Brazil and Chile (OECD, 2022[39]). In Mexico, a decline in diabetes and hypertension diagnoses was observed to the order of 17% and 22%, respectively (Doubova et al., 2021[42]).

In addition, large disparities in the level of disruption across socio-economic groups were observed in Peru, where patients with lower education (11.5%) presented a reduction in diabetes care almost twice as high as people with higher levels of education (6.3%) (INEI, 2021[43]).

In a research study surveying a group of 704 oncology physicians from 19 LAC countries, 70% of respondents reported a reduction of new patients, 72% noticed a decrease in follow-up consultations and 58% affirmed having changed the treatments offered to patients with cancer (Bernabe-Ramirez et al., 2022[45]). This is consistent with records showing cervical cancer screening decreases in Mexico (68%), Peru (49%), Brazil (45%) and Chile (9%) and breast cancer screening decreases in Mexico (79%), Peru (51%), Brazil (43%) and Chile (10%) (OECD, 2022[39]). Referrals for oncology patients were also affected: the study by Bernabe-Ramirez et al. (Bernabe-Ramirez et al., 2022[45]) found that 65% of surveyed doctors in LAC reported referral delays and 20% of them reported that surgeries had to be cancelled.

As a result of the disruption in cancer screening, registries management and cancer care services, recorded cancer cases decreased in Peru in 2020 by 50% when compared to the previous 4 year average (MINSA, 2021[46]). Furthermore, treatment for advanced-stage cancer decreased 37% between March 2020 and March 2021 (Chávez Amaya, 2021[47]). As a result, cancer mortality increased in the country by 5.3% when comparing 2020 with 2018 values (WHO, 2020[48]) (OECD, 2022[39]). In Chile, treatment for cervical cancer was reduced by 55% between 2019 and 2020 and a further 8% in 2021, with the reduction in services greater for patients using public health insurance (Superintendencia de Salud, 2022[49]). Brazil saw a decrease of 42% in oncological patient referral between 2019 and 2020 (Borges et al., 2020[50]).

Table 1.1 presents the proportion of countries by levels of disruption of essential healthcare services in 2021 (highest disruption reported during the year), disaggregated by excess deaths performance. Notably, countries experiencing lower than OECD average excess deaths had lower levels of disrupted services during 2021. The disruption in cancer treatment is particularly worrisome, given the immediate health impact it has on the population. Mental healthcare was also severely affected, where challenges included the limited capacity of health services to use virtual/telemedicine platforms to provide care to mental health patients (Antiporta and Bruni, 2020[51]).

During 2021, elective surgeries were disrupted in 59% of countries reporting data to the WHO PULSE survey (WHO, 2022[8]). In LAC, 19 out of 27 countries in the survey reported disruptions larger than 25% (Figure 1.9). While most countries in the region severely decreased elective surgeries, some countries with higher than LAC average excess deaths, such as Peru, Bolivia, Ecuador and Saint Vincent, experienced disruptions of less than 50%. Disrupting elective surgeries can have accumulating consequences on health and well-being, especially in the long run. However, given its non-emergency status, it is an area were authorities can re-distribute staff and resources to provide care for patients with COVID-19, as they did for most countries in the world (Nepogodiev et al., 2022[52]).

Population’s mental health was directly affected by the state of emergency and restrictions to meet family, friends, and other social communities. Research in the region reported that the most frequent reactions to both the pandemic itself and the public health measures that countries implemented were anxiety, stress, and fear. Because of the urgency to intervene the pandemic generated, anxiety was considered by mental health experts to be the number one indirect health effect concern. Given the overwhelming situation and the collapse of healthcare systems, community-based mental healthcare and strengthening the capacities of non-specialised primary health providers were highlighted to be effective interventions to minimise the effect on mental health (Antiporta and Bruni, 2020[51]).

Figure 1.7 presents the percentage of the population out of a representative sample of Facebook users that self-reported feeling nervous over the three years of the pandemic. Only in three countries the level of nervous people decreased between 2020 and 2021, and for all other countries the peak of self-reported nervousness was reached in 2021, coinciding with the peak of reported COVID-19 deaths. In addition, the chart shows how the sense of normality started to settle during 2022, with nervousness levels decreasing compared to 2021 in all the countries in the region. Other factors may also have affected this indicator. The OECD average shows a peak of nervousness in the population in 2022. This could be related to the ongoing conflict in Ukraine, as well as the surge in COVID-19 deaths during the first months of this year.

The percentage of people that reported wanting more information on how to remain mentally healthy during the pandemic is a good indicator of the responsiveness of the systems in this topic. Among the 19 LAC countries with available data, 41% of the population on average reported wanting more information. In contrast, only 29% of the population reported this along 29 OECD countries. In Peru and Chile, about 50% of people in 2021 would have liked to have more information on mental health. In contrast, 30% and 34% of people in Uruguay and Argentina were in this situation (The University of Maryland Social Data Science Center & Facebook, 2020[14]).

Countries’ pandemic response were in large part defined by the capabilities of their health systems, but also by structural factors that affected circulation of the virus in their countries. This analysis begins by comparing public health and social measures directed at mitigating the spread of the virus among the population. Second, a comparison of policies directed at strengthening the healthcare system is conducted, examining those put in place to treat COVID-19 patients as well as securing the provision of essential non-COVID-19 care during the pandemic.

The range of public health policies directed at mitigating the effects of the pandemic was summarised by the Oxford COVID-19 Government Response Tracker in four areas; containment and closure, economic, health system, and vaccination, and included 21 indicators (Oxford COVID-19 Government Response Tracker, 2022[15]). Most countries in LAC (26 out of 28 in the database) implemented all the containment and closure measures (eight indicators) at some point and at some level between 2020 and June 2022. However, the length and stringency of the measures varied between countries. Table 1.2 shows the length of time LAC countries implemented containment and closure measures at the strictest level. The most significant differences can be found in the lockdown indicator, where Chile leads the region with 426 days under orders of not leaving the house without a formal excuse. In contrast, nine countries did not implement the strictest level of lockdown for different reasons, one of them being complex legal procedures and congressional approval (Costa Rica). On average, countries that had lower than OECD average excess deaths had seven out of eight measures at the strictest level for longer than countries with higher excess mortality (Table 1.2). However, the average is heavily influenced by one or two countries in each group with exceptionally high or low values. Peru is a notable outlier in a counterintuitive direction, as the country implemented one of the strictest policies in the region but still had a low performance in terms of excess deaths.

Assessing the relationship between public health policies and excess mortality requires unravelling an important timing effect in pandemic response. On one hand, countries amid a COVID-19 surge of cases and deaths are pressured to strengthen their containment measures. On the other hand, if a country has stricter measures, it can minimise future surges in cases and deaths. Figure 1.8. presents the Containment Index for 27 countries in the region in the second quarter of three years of the COVID-19 pandemic (2020 to 2022). Six countries – Jamaica, Chile, Barbados, Suriname, Venezuela, Costa Rica, and Uruguay – had stricter containment measures in the second quarter of 2021 than the year prior in 2020. All of these countries had lower-than-OECD-average excess deaths. Furthermore, except for Peru and Guyana, countries in the graph that had higher-than-OECD-average excess deaths during 2020-21 decreased the stringency of their containment measures in 2021 by more than 11% compared to the same quarter in 2020 (Peru and Guyana decreased 5 and 6% in this period, respectively).

Considering that the surge in COVID-19 deaths happened during 2021 (Figure 1.2), Figure 1.8. shows how LAC adopted a strict approach early in 2020, with most countries in the region having a higher containment index than the OECD average at this stage. The objective of this strategy was to minimise the initial spread of the virus, in line with the world-wide “Flattening the curve” approach (The Economist, 2020[53]; OECD, 2020[54]) and was somewhat effective in the short term. However, following a mix of economic and political exhaustion, lowered public adherence with containment measures, and public health prospects brought forth by the COVID-19 vaccine, many countries were not able to maintain the stringency of their measures during 2021 which directly affected the death toll of the pandemic. This hypothesis is supported by Figure 1.9, where it is shown that the containment index in the second quarter of 2020 has a counterintuitive positive correlation with the cumulative excess mortality during the pandemic, but this correlation is negative (as it should be) when considering the same index one year later.

To facilitate adherence to containment measures, and to mitigate economic and social consequences of the pandemic, one of the most important policies implemented by countries worldwide were the direct transfers of resources to a large portion of the population (ECLAC, 2022[3]).

Figure 1.10 shows public spending per capita in social protection in 2019 and 2020 in the LAC region. Uruguay is the only country in the region (with available data) that considerably decreased (in nominal terms) the amount spent on social protection in 2020. This decrease is explained by the creation of the “Fondo Solidario COVID-19” an off-budget fund that centralised resources for the pandemic response (Restrepo, Palacios and Espinal, 2022[56]). Moreover, a significant portion of economic relief expenses came from unemployment insurance (Amarante, 2022[57]). Other countries also implemented off-budget resource transfers to mitigate the economic effect of the pandemic which are not reflected in the figure. For instance, in Chile, people were allowed to withdraw 10% of their individual social pension scheme four times between 2020 and 2022.

Moreover, eight countries (Brazil, Argentina, Chile, Costa Rica, Trinidad and Tobago, Mexico, Uruguay and Suriname) spent over the LAC average in social protection in 2020, the same countries which were below the LAC average of excess deaths except for Mexico. On the other hand, Ecuador and Peru are at the lower end of social protection spending while being over the LAC average of excess deaths. The relation between the variables is, however, not so evident. Honduras, Belize and Jamaica (among others) had lower than the OECD average excess deaths, while having some of the lowest levels of per capita spending in social protection during 2020.

The scope of the social protection policies had a clear effect on the ability of countries to maintain as many people as possible complying with pandemic restrictions. Economies in LAC are marked by high levels of inequalities, a large middle class with low levels of household savings and high levels of debt and important labour informality (ECLAC, 2022[3]). In this sense, if more households were supported and could secure their livelihood, more would have been financially able to stay at home and limit the spread of the virus. Figure 1.11 shows the percentage of households that received any form of financial assistance during 2020-21. While the data is limited in informing about the amounts of the assistance, there are some patterns that proxy the relationship between public support to maintain public health measures and overall excess deaths. Some countries with lower performance in terms of excess deaths appear at the lower end of the graph, such as Peru, Mexico and Ecuador. However, countries like Guyana, El Salvador, Brazil and Paraguay provided financial assistance to more than 45% of households and still had higher than OECD average excess deaths.

All countries in the region implemented policies to boost their budgets in health and to reorganise the system to better deal with the health emergency. While the average increase in public spending on health in the region in 2020 was nominally around 16% (12% when accounting for inflation) (ECLAC, 2021[16]), other important policies relate to enhancing the role of primary care, expanding the provision and incentives for healthcare workers, centralizing hospital care co-ordination and purchasing test kits, vaccines, respirators, protective equipment, and other supplies.

Public expenditure in health in 2019 and 2020 is showed in Figure . The largest increases (without accounting for inflation) are found in Guyana (99%), St. Vincent and the Grenadines (71%), Antigua and Barbuda(43%) and Argentina(42%). Argentina is a special case, as the high inflation registered in 2020 (42% (OECD, 2023[58])) neutralises the nominal increase to zero. The highest per capita expenditures are found in Chile (1 514 USD PPP), Costa Rica (1 264 USD PPP), and Bahamas (1 228 USD PPP), three countries below or near the OECD average of excess deaths (Figure ). Suriname and Uruguay are the only countries that significantly reduced their public health spending. For Suriname, the decrease responds to a general policy of fiscal austerity to deal with the economic crisis in the country (ECLAC/CEPAL, 2022[59]). Moreover, the creation of special funds for the pandemic (ECLAC/CEPAL, 2022[59]) might not be accurately reflected in the amount presented in the figure. In the case of Uruguay, the decrease is explained by the “Fondo Solidario COVID-19”, off-budget fund that centralised resources for the pandemic response (Restrepo, Palacios and Espinal, 2022[56]). Other mechanisms to finance pandemic health expenditure that are not reflected in the figure include the use of emergency and disaster funds, like in Costa Rica, where these funds were used to cover for the public use of private care service capacities.

In contrast to total public health spending, additional fiscal efforts to deal with the pandemic are better represented in the database of country fiscal measures between January 2020 and October 2021 in response to the COVID-19 pandemic by the International Monetary Fund (IMF, 2021[60]). However, there is no discernible correlation between the per capita amounts spent in COVID-19 additional fiscal efforts and levels of excess mortality, highlighting the importance of both the total spending, as a proxy of a stronger health system, and the early spending increase for a swift response.

The WHO highlighted that gatekeeping and care-co-ordination for treatment and follow-up of COVID-19 patients are effective measures to improve the efficiency of health systems in responding to the pandemic (WHO, 2021[61]). In this principle, primary healthcare is an essential actor in countries’ pandemic response. However, the overall response to the pandemic in LAC has been mainly centred on hospital care of severe COVID-19 patients. Despite countries having more emphasis in hospital care than primary care, the latter is likely to be an important determinant of better performance during the pandemic. In this sense, countries with stronger primary healthcare systems have shown signs of greater resilience during the pandemic (OECD, 2022[39]).

Primary healthcare (PHC) has been used in different capacities, summarised in prevention, early detection, and care management. Regarding prevention, PHC has been used to create awareness on COVID-19 risks and disseminating public health information, to identify and reach out to high-risk populations, and for contact tracing and quarantine supervision. In terms of early detection, PHC providers were a major actor in the countries’ testing capabilities. Finally, PHC providers had an important role in the care for mild COVID-19 cases and patients with long-lasting symptoms (“Long COVID-19”), either in the healthcare facilities of co-ordinating community care through digital health and home visits.

Table 1.3 summarises the role of PHC in the pandemic response in seven LAC countries. Costa Rica, Argentina and Chile relied on PHC for more testing capacities than other countries in the group. At the same time, they had fewer excess deaths (lower than OECD average) when combining figures from 2020 and 2021. All countries in the assessment relied on PHC to care for mild COVID-19 patients and to co-ordinate care in the community with either home visits or telehealth and telemonitoring. Contact tracing and supervising quarantine was the least common capacity assigned to PHC. However, caring for long COVID in PHC has been overlooked by most countries in the assessment, with only small initiatives in this regard in Mexico, Costa Rica, Peru and Colombia (OECD, 2022[39]).

The availability of COVID-19 testing in these seven countries was rather limited at the start of the pandemic, together with limited laboratory and technical capacities to test at the needed scale. In addition, centralised guidance from ministries of health for COVID-19 diagnosis often came at a later stage, requiring local authorities to develop their own protocols early on (OECD, 2022[39]). On the other hand, while COVID-19 vaccinations were also limited when first available at large scale, guidelines were fairly organised for its implementation. Some countries included explicit guidelines for managing indirect health effects of COVID-19 at PHC level. In Colombia and Chile for example, mental health guidelines included recommendations on how to deal with gender-based and domestic violence (OECD, 2022[39]).

To mitigate the sharp reduction in patient consultations and disruption of routine care, telemedicine, and other practices for delivering services remotely were implemented in many countries in the region. Mexico, Chile and Argentina implemented initiatives to maintain drug adherence of patients, enable online refills, external pick-ups, community drug deliveries and electronic prescriptions (Arsenault et al., 2022[41]; OIT/ILO, 2021[62]). Panama started an initiative to co-ordinate home visits with telemedicine for daily monitoring of patients, while Uruguay established a free phone medical consultation service (OIT/ILO, 2021[62]).

National programs to co-ordinate electronic consultations with general practitioners and specialists were implemented in Argentina, Peru, Colombia, Guatemala, Panama, Uruguay, Mexico, Costa Rica and Chile (LeRouge et al., 2019[63]). The new national strategies involved updating legislation and guidelines to address challenges limiting the implementation of tele-medicine. In Costa Rica, from January to May 2020, out of 680 644 consultations registered at the first level of care, 188 108 (28%) were conducted through alternative methods of care (phone or video calls) (CCSS, 2020[64]).

Countries in the region had different approaches to enhancing their hospital capacities, in line with their structural needs. Brazil focused on procuring ICU beds with the help of new legislation, allowing for the federal government to transfers funds for this purpose. Other countries opted for a centralised approach creating new intensive care in-patient facilities, either by revamping wards of existing hospitals, like the case of “Hospital Español” in Uruguay and the Hospital facilities of the “Caja de Seguro Social” (National Insurance Fund) in Panama, or by creating emergency modular hospitals in Argentina, Chile, Paraguay, Peru and Costa Rica. Additionally, some countries, like Costa Rica, temporarily converted whole hospitals and hospital wards for COVID-19 care. At the same time, countries worked to secure provision of essential supplies, medical equipment and personnel, and other basic utilities in their health facilities at all levels. Other infrastructure, such as stadiums, hotels and clubs have been converted to health centres to deal with mild or potential COVID-19 cases (OIT/ILO, 2021[62]).

The co-ordination between authorities and public-private health facilities was crucial for maximising countries’ capacities and efficiency to care for patients (COVID-19 and non-COVID-19) during the pandemic. Chile and Colombia adopted a centralised approach including management of public and private hospitals, in terms of hospital beds, mechanic ventilators and triage. Mexico secured a deal with a large private hospital network to treat for non-COVID-19 patients and decompress public hospitals to care for COVID-19 patients (OIT/ILO, 2021[62]). A similar situation happened in Costa Rica, where the institution managing healthcare (Caja Costaricence de Seguro Social”) directed private providers to care for non-COVID-19 patients to concentrate COVID-19 care in public hospitals.

The COVID-19 pandemic evidenced the weaknesses of healthcare systems in the region to appropriately supply their medical facilities with healthcare personnel, particularly in rural areas and at the primary healthcare level. Countries in the region adopted new legislation to increase the amount and efficiently distribute healthcare workers for the pandemic efforts. For example, several countries created new criteria to allow for health workforces-in-training to work alongside professional staff in limited capacities. At the same time there was an observed increase in the budget for remuneration of healthcare workers. However, only a few countries focused the budget increase to strengthen PHC or in policies to strengthen human resources management in the long term. It is not clear if the increase and efficient management of healthcare workers will remain in the future as an effort to improve healthcare quality. The gender perspective regarding health workforce during the pandemic was overlooked, an especially important issue because of the disproportionate burden in women to care for COVID-19 patients and other functions related to the pandemic response (Cho and Levin, 2022[65]). On a positive note, the pandemic allowed for collaborative efforts between countries in the region aimed at improving the management, training, and planning of healthcare human resources.

Out of ten LAC countries with available data, six increased the rate of doctors per 1 000 inhabitants, while three countries decreased and one maintained said rate (Figure 1.13). There is a strong association between excess deaths and the baseline rate of doctors before the pandemic. The five countries in LAC with more doctors per population before the pandemic had lower than OECD excess deaths. A sixth country, Chile, was ninth in the region. All these countries had more than 2.7 doctors per 1 000 people, in 2019, 26% above the LAC average. At the same time, the two countries with highest rates of excess deaths, Bolivia and Peru, are among the lowest in number of doctors before the pandemic, with 1.03 and 1.37 per 1 000 population, respectively.

In terms of the changes in number of doctors to deal with the pandemic in 2020, two countries which substantially increased their number of doctors, Honduras and Guatemala, had lower than LAC average excess deaths despite being at the lower end of the region in number of doctors before the pandemic, and still below the LAC average in 2020. In contrast, countries that reduced their rate of doctors had higher than OECD (Paraguay) or higher than LAC average excess deaths (Mexico and Guyana). As mentioned previously, it is not clear if the increase in healthcare workers will remain in the future. For some countries, like Costa Rica, this will most certainly not be the case given that new hires were transitory and aligned with the virus waves.

Health workforce shortages during the pandemic have motivated other OECD countries to re-evaluate the scope of practice attributed to healthcare staff such as nurses and community health workers. Supported by literature (Maier, Aiken and Busse, 2017[66]), countries have recognised that non-physician healthcare workers with enhanced education and training can improve access to services, reduce waiting times, and deliver the same quality of care as doctors for limited services.

In this regard, several OECD countries used fast-track licenses and provided exceptional training to mobilise healthcare providers to address the COVID-19 pandemic. Australia, Austria, Finland, Latvia, Spain, Switzerland, the United Kingdom and the United States, for instance, created new roles and rearranged tasks from both physician and non-physician health workers to maintain care continuity for non-acute COVID-19 cases in the community or at home. LAC countries could follow this example and expand the role of community health workers and nurses to care for certain conditions, such as routine control of chronic diseases, to maximise capacity during health emergencies and minimise care disruptions. Brazil, for example, uses community health workers for improving medication compliance (OECD, 2021[67]).

The COVID-19 vaccine first became available to the general population in the region in December 2020 in Chile, Costa Rica, Mexico and Argentina. Soon after, all the other countries in the region had access to the vaccine, where Haiti was the last country that started vaccinating people in July 2021. Consequently, especially during 2021, the focus of health authorities for containing the pandemic was given to quickly and efficiently rollout the COVID-19 vaccine. Figure 1.14 shows the percentage of the population fully vaccinated in two different moments of the pandemic. In Latin America and the Caribbean, the COVID-19 vaccine had, in general, a slower rollout compared to OECD countries. In part, this was due to procurement issues. Furthermore, the rollout in the Caribbean nations was lower than in the rest of LAC (ECLAC, 2022[3]). When comparing the vaccination rollout between excess deaths groups defined in  Section 1.3, several of the countries in the best performing group (Chile, Uruguay, Argentina, Cuba, Panama and Costa Rica) had higher than LAC average vaccination coverage by the 3rd quarter of 2021. On the other hand, most countries in the lower performance groups (Bolivia, Peru, Mexico, Colombia, Guyana and Brazil, among others) had around the average or lower than average vaccination coverage at the same period.

Primary healthcare facilitated the implementation of COVID-19 vaccine distribution in Argentina, Peru, Brazil, Chile and Costa Rica (Table 1.3). The involvement of PHC in these capacities was deemed an effective strategy to increase the efficiency of the pandemic response. By contrast, PHC did not play a central role in COVID-19 vaccination in Colombia and Mexico. The effectiveness of using PHC for the vaccination rollout is partially reflected in the share of the population fully vaccinated in the second quarter of 2022, a time where vaccines were widely available worldwide. While Mexico and Colombia had 63% and 69% of their population covered respectively, countries using PHC for vaccination rollout had more than 77% of their population covered by this time (Figure 1.14).

Being able to communicate the vaccination plan was crucial for a swift and efficient vaccine rollout. Figure 1.15 shows the percentage of respondents of a country representative sample that were aware of the procedure to get the COVID-9 vaccine in their country of residence during 2021. Uruguay, Chile, Argentina, Dominican Republic, and Antigua and Barbuda led the group with between 78% and 97% of people living in both urban and rural areas aware of how to receive the COVID-19 vaccine. Coincidentally, all these countries are in the best performing group in terms of excess deaths, and three of them capitalised on PHC for their vaccination efforts. On the other hand, countries that were less effective in communicating their vaccinations plans correlate with countries with levels of excess deaths over the LAC average. The difference between vaccination knowledge between rural and urban areas was more pronounced in Mexico, Guatemala, Bolivia and Colombia with a difference of up to 15 percentage points. At the same time, Mexico and Colombia did not rely on PHC for this purpose. On average, only 59% of the population knew how to get vaccinated in LAC in 2021, disaggregated to 61% of people in urban areas and 55% in rural areas. The community-based approach of primary healthcare lends a strong argument to capitalise on PHC for this purpose (OECD, 2022[39]).

Effective implementation of vaccination plans entailed both the appropriate dissemination of information and the vaccination schedule. For a successful implementation, the potential barriers preventing people from getting a vaccine needed to be acknowledged and addressed. Figure 1.16 presents the proportion of people (out of Facebook users) that encountered barriers preventing them from getting vaccinated. Barriers included: not possessing adequate documentation, not being able to take time off work or school, no available appointments, restricted appointment times, and travel to the vaccination site, among others. Barriers in the first quarter of 2021 affected 59% of responders, significantly more than the OECD average at the same time (34% of responders). There are no evident patterns between vaccine barriers and performance in terms of excess mortality. Nevertheless, several countries (Chile, Dom. Republic, Uruguay and Argentina) with less than OECD average excess deaths are at the lower end of the graph, where only about 30% of responders reported barriers by the third quarter of 2021. However, countries with more challenging outcomes in terms of excess deaths, such as Brazil, Mexico, El Salvador and Peru, also present lower than LAC average percentages of people encountering barriers. At the other end of the graph, in Bolivia, Ecuador and Paraguay more than 75% of responders reported barriers to get vaccinated in the second quarter of 2021. While this situation improved dramatically for Ecuador and Paraguay by the third quarter of the same year, in Bolivia 66% of responders were still reporting barriers to get vaccinated by the fourth quarter. A similar situation occurred in Nicaragua, with Venezuela and Honduras also having a high percentage of the population reporting barriers.

While regulations and public health mandates vary in the region and along the three (and counting) years of the pandemic, preventive recommendations were in place throughout 2020 and 2021 in all of LAC. In most countries, preventive measures were adopted by a significant majority of the population during this time. Among 14 LAC countries with available information (Figure 1.17), 98.6% of the population complied with physical distancing in 2020. Other preventive measures were not as popular. On average, 92.2% of the population complied with mandates to avoid gathering of more than ten people. In Chile this percentage increases to 95.7%, important considering the length of this mandate in the country (Table 1.2). In contrast, in Colombia adherence was 85%. A similar situation occurred with physical greetings (handshake, hugs) where on average 92.2% of the region complied with this measure.

Between May and July of 2020, there was substantial uncertainty about public health recommendations, restrictions and other mandates put in place to mitigate the effects of the pandemic. Campaigns were launched by governments to inform the population of actions to take. However, the knowledge of the population regarding actions taken by the government and authorities was, at this stage, quite low. Across the same 14 countries presented in Figure 1.17, on average 56.8% of the population knew that authorities were advising citizens to stay at home. However, only 6% on average knew about the temporary closing of business and schools, even though these measures were already being implemented at some level in each country. Imposing curfews or lockdowns was only known by 42% of the population on average, a percentage that diminished to 10.8% and 14.2% in Mexico and Bolivia, respectively. Perhaps the best indicators reflecting the lack of effectiveness in information campaigns in mid-2020 is the fact that, on average, only 11.1% of the population of these 14 countries (on average) knew that government or other authorities were disseminating knowledge about the COVID-19 pandemic.

Physical distancing was widely adopted by LAC countries. However, data for the year 2021 shows that there is a clear diminishing adherence over time (Figure 1.18). This deviates from the trend in excess deaths that registered its peak on the 3rd quarter of 2021 and only diminished by the end of the year. Compared to an average of 75% of people on average across OECD countries, 78% of people complied with physical distancing measures in the LAC region in the first quarter of 2021. Afterwards, LAC countries maintained adherence to physical distancing throughout the year, diminishing slowly until only complied by 65% of the population, compared with 53% on average in the OECD (Figure 1.18). Among LAC countries, there was high heterogeneity in the level of adherence to this particular measure. Even though the patterns are not straightforward, the best performers in terms of excess deaths had higher than average adherence during quarters two and three (green diamonds are consistently higher than red diamonds in Figure 1.18), along with the rise in excess deaths (except for Uruguay in the 3rd quarter). On the other hand, countries with higher than LAC average excess deaths had lower than LAC average percentage of the population adhering to physical isolation during these six months (except for Mexico). The high levels of physical isolation resonate with the high levels of mental distress observed previously, a factor which most likely influenced willingness to adhere to the measure.

The recommended or mandated use of face masks was also widely accepted in LAC countries since the start of the pandemic, where most countries report higher use than the OECD average. According to the survey by the University of Maryland in collaboration with Facebook (2020[14]), in the second quarter of 2020, 78% of the responders reported using a face mask all or most of the time when in public on average in LAC, compared to only 54% on average of OECD countries. One year later, during the second quarter of 2021, most countries reported their peak in face mask adherence, reaching 88% in LAC and 77% on the OECD average. Costa Rica presents an interesting case, as at the start of the pandemic it was the country with the lowest reported adherence (32%), but one year later it became the country that reported the highest adherence (92%), in line with the rise in cases and COVID-19 related deaths.

The COVID-19 vaccine is the single phenomenon that had the largest effect in the control of the pandemic. Vaccine acceptance was essential for an effective roll-out. Between June and July 2021, 80% of the population in 20 four LAC countries planned to get vaccinated against COVID-19 (The World Bank group, 2021[11]).

Based on a representative sample of Facebook users, Figure 1.19 shows how OECD countries, on average, reached all people that were willing to receive a COVID-19 vaccine in December 2020 after one year. In LAC, this level was reached three months after, in March of 2022. However, there is vast heterogeneity in the effectiveness of the vaccine rollout, and this is not explained by vaccine hesitancy. In 11 countries, a greater number of people were vaccinated with at least one dose than the proportion of the population reporting willingness to get vaccinated 12 to 15 months earlier. This speaks to the impact of effective communication and persuasion campaigns by health authorities. Two more countries (Nicaragua and Costa Rica) managed to reach the same level of coverage as people willing to receive a vaccine. On the other hand, Haiti, Paraguay, Bolivia, Honduras, Salvador, Guatemala and Mexico remained unable to reach the whole population that wanted to be vaccinated against COVID-19 and several of these countries had high levels of excess deaths.

Drivers of vaccine hesitancy are diverse (Figure 1.20). Based on the same sample of Facebook users (The University of Maryland Social Data Science Center & Facebook, 2020[14]), out of the Latin American and Caribbean responders that were hesitant about the COVID-19 vaccine, 55% reported concerns over potential side effects to be the reason for their hesitancy. While some effects of the COVID-19 vaccine have been found, research after more than one and a half years of mass implementation has established that the benefits outweigh the risks by a significant margin. Moreover, 46% of hesitant responders reported wanted to wait to see if the vaccine is safe and might get it later. Finally, 32% of respondents that reported being hesitant did not think the vaccine was effective. All three main reasons driving vaccination hesitancy can be substantially overcome by effective communication campaigns disseminating the latest research on the safety and effectiveness of the COVID-19 vaccine.

Other important drivers of vaccine hesitancy relate to trust in the government (14%), trust in vaccination (13%), influence by other people (13%), and not liking vaccination (10%) (Figure 1.20). People trust international (World Health Organization) and local health authorities the most when related to COVID-19 vaccines Figure 1.21. In LAC, the average portion of the population that trusts these institutions is around 70%, while in the OECD this percentage is only 53%. On the other hand, only 40% and 26% of respondents in LAC and the OECD trust politicians regarding vaccination. The data suggests expertise being an important factor driving trust in vaccination, which is positive for population health management in a crisis situation. However, closer social relationships and kinship, like family, have significantly more trustworthiness in LAC than compared with the OECD average, where 63% of LAC respondents trust their families’ opinions regarding the COVID-19 vaccine compared to only 40% in the OECD. High levels of trust in non-experts leaves space for disinformation to spread. For example, 8% of respondents in LAC in mid-2021 thought that masks were not necessary after being vaccinated. There was some variability between LAC countries in this respect, ranging from 5% in Bolivia, Brazil and Chile, to 12% in Haiti.

LAC countries baseline preparedness for the pandemic was at severe disadvantage when compared to other OECD countries. High levels of inequality and informality diminished the capacity of the population to adhere to public health measures, where those without access to social protection and health coverage faced additional barriers to receive health when needed. The region also has high percentage of people living in inadequate housing or informal settlements, making them more vulnerable to the virus. While these topics are not analysed in depth in this chapter, chapters three to nine in this publication provide a comprehensive characterisation of LAC health systems. Instead, we focus on the main baseline characteristics influencing the level of excess deaths during the COVID-19 pandemic.

Latin American and Caribbean countries had very different structural characteristics at the start of the pandemic. Some of these characteristics are highly associated with excess deaths during of the pandemic, while others are not. There are no clear patterns between the level of health in LAC countries (see Chapter 3) before the pandemic and the number of excess deaths between 2020 and 2021. Counterintuitively, there is strong association between countries with older population (see Chapter 9 – figure 9.1) and better performance in excess deaths. This is largely explained by income per capita, associated with both ageing and the ability to better respond to the pandemic. In contrast, the level of resources available in the health system behaves as expected, where best equipped countries perform better in terms of excess deaths, even though resource levels are lower than the OECD average.

Healthcare expenditure per capita in LAC is far below the OECD average of 3 999 USD PPP (see Chapter 6 – Figure 6.1). The LAC average was 1 155 USD PPP, 28% that of the OECD average. These low levels of total health expenditure could signal why health systems in the region were under-resourced and less prepared to face the challenges of the COVID-19 crisis. There is a strong association between higher health expenditure per capita and low excess deaths. Seven out of the nine countries in the best performing group had higher than 1 500 USD PPP expenditure per capita in health (except for Venezuela and Jamaica). On the other hand, the six countries with higher number of excess deaths in the region had an average health expenditure of 781 USD PPP, led by Mexico (1 117 USD PPP) and with Saint Vincent and the Grenadines being the lowest (620 USD PPP).

Health systems in LAC would benefit not only from more expenditure, but also from better expenditure. In fact, wasteful expenditure is a major issue in the region, especially in the areas of clinical care, operational, and governance waste, hindering the path towards universal health coverage (OECD/The World Bank, 2020[69]). During the COVID-19 crisis, the on-going wasteful expenditure has exacerbated the lack of health expenditure, hindering the capacity of health systems in the region to reallocate scarce resources to increase health workforce and resources.

LAC also lags the OECD average in terms of health workforce density. On average, LAC has 2.15 medical doctors per 1 000 population (latest data – 2017 to 2021) (see Chapter 8 – Figure 8.1). The OECD average in this period is 3.6 (68% more). As seen in 1.4.2, countries that performed best in terms of excess deaths had a higher density of doctors than countries that did not perform as well. The association with the number of nurses operates in the same direction (see Chapter 8 – Figure 8.4), but the correlation is weaker as the countries with the lower number of nurses per 1 000 population are situated in the middle group: with excess deaths above the OECD average but below the LAC average.

There are only three countries in LAC that have more hospital beds per 1 000 population than the OECD average (see Chapter 5 – figure 5.7). These are Barbados (5.8), Cuba (5.2), and Argentina (5.0), compared to the OECD average of 4.3. These three countries had strong performance in terms of excess deaths during 2020 and 2021. Likewise, countries with high excess deaths typically have a lower number of hospital beds than the LAC average (2.1 per 1 000 population), such as Peru (1.6), Bolivia (1.1), Mexico (1.4), Ecuador (1.5), Guyana (1.6), and Colombia (1.7).

Interestingly, some countries with similar numbers of hospital beds per 1 000 population had different performances in excess deaths. For example, Uruguay (2.8), Panama (2.3), Chile (2.1), Bahamas (2.9), and Brazil (2.3) had a similar rate of hospitals beds but major differences in the number of excess deaths, indicating that the efficiency of hospital bed management also played a crucial role. This trend is repeated when accounting for ICU beds capacity (OECD/The World Bank, 2020[69]). Among countries with low levels of excess deaths, Uruguay (19.9 beds) and Argentina (18.7 beds) have higher number of ICU beds per 100 000 population than the average of 22 OECD countries (12 beds) with available data. However, Chile (7.3 beds) and Costa Rica (2.7 beds), also countries with low number of excess deaths, have a considerably lower capacity. On the other hand, Brazil (20.6 beds) Colombia (10.5 beds) and Paraguay (10.3 beds) have more or similar levels to the OECD22 average but had higher levels of excess deaths. Finally, three countries with higher than LAC average excess deaths, Ecuador (6.9 beds), Mexico (3.3 beds) and Peru (2.9 beds), have low capacity of ICU beds (OECD/The World Bank, 2020[69]).

While there is substantial variation in LAC countries’ performance along the variables studied in the chapter, there are clear patterns among process and structure level performance and excess mortality. The assessment points towards fewer excess deaths being related to both the effective implementation and the comprehensiveness of the response, together with interventions to address identified weaknesses in response capacity and structural characteristics.

Table 1.4 puts together countries’ performance on excess mortality with key variables summarizing pandemic response, systems responsiveness and structural characteristics affecting response capacity. The predominance of green at the top of the figure shows that a comprehensive pandemic response, performing above average in most categories, was associated consistently with lower levels of excess deaths. At the other side of the spectrum, countries with the highest number of excess deaths exhibit several variables where they are among the worst performing in the region. For these countries, a lower performance in these areas rendered these factors in the health system sufficiently weak for COVID-19 to severely disrupt care provision and reduce effectiveness of public health measures. However, low performance in some areas was not necessarily a weak point in dealing with the pandemic. Countries that recognised their shortcomings and put interventions in place to address them showed lower levels of excess deaths. For example, Chile and Costa Rica had around or below the median number of hospitals beds but addressed this challenge with centralised management of hospital capacities, including private providers, and based on triage and risk assessment.

Moreover, comprehensiveness in the pandemic response enhanced effective implementation of public health measures. Countries with lower excess deaths integrated the needs of the population and put complementary policies in place to mitigate the health and socio-economic effects of the COVID-19 response. This minimised the social cost of the pandemic and allowed countries to link the stringency of their public health measures with the infection waves and the idle capacity of their health system. When in 2021 the focus of the pandemic response shifted towards mass vaccination, countries that complemented a quick vaccine rollout with information campaigns enhanced the effectiveness of the measure by both easing public uncertainty and improving vaccine acceptance. In the same line, countries that addressed healthcare disruptions with centralised co-ordination of healthcare resources and a strong reliance on primary healthcare systems for promotion, prevention (including vaccination), treatment and follow-up of COVID-19 were able to stop the vicious cycle of health system disruption and excess deaths.

Counterintuitive outliers in both high and low performance groups enlighten the discussion about the many other potential variables influencing the pandemic effect. Venezuela and Jamaica had a good performance in terms of excess deaths but performed lower than most countries in most of the selected variables in Table 1.4. However, these countries had relatively strict containment measures, and increased their stringency in 2021 compared to 2020. Combined with the isolation the economic and political context in Venezuela carried and Jamaica’s geography, these factors seemed to have been especially effective to mitigate the spread of the virus. Other island countries in the Caribbean with low levels of excess deaths might have benefited from the same effect. Venezuela’s case needs to be interpreted with caution, because in other counts of excess mortality (Wang et al., 2022[20]) the country appears with higher than LAC average excess deaths. This might indicate that the country was not as effectively isolated as initially thought.

Mexico, on the other hand, has the third highest number of excess deaths in the region (fourth in the excess deaths estimation by (Wang et al., 2022[20])), but it has average or better than average performance in most of the selected variables in Table 1.4. However, though not reviewed extensively in this chapter, Mexico has high prevalence of important risk factors for COVID-19, namely obesity and chronic conditions prevalence (see Chapters 3 and 4). In addition, there are still enhanced problems with care fragmentation and universal access to services arising from the transition that started in 2018, with the removal of “Seguro Popular” and the creation of the new universal health programme (OECD, 2022[39]). Health systems characteristics like the latter were not extensively reviewed in this chapter because of a lack of consistent and comparable indicators for the region but might have an important influence in the pandemic effect. In particular, system fragmentation, information systems for epidemiological control and care co-ordination, and medical supply chain management seem to have had an important effect and were a central part of countries’ pandemic response (OECD, 2023[38]). Moreover, while the emergency response of LAC countries was mainly managed at the central level of governance, the implementation of the response actions, together with the baseline preparedness of the health system, depended heavily in subnational governments. Following the shared responsibility central and subnational governments have on planning, financing, and delivering healthcare, future research studying the determinants of effective emergency response should explore the co-ordination and performance at and between these levels of governance.

Combining the lessons of this chapter and extensively studied recommendations for improving health system resilience (OECD, 2023[38]; Herrera et al., 2022[2]; OECD, 2022[39]), we highlight important considerations that are of particular interest for improving health system resilience in the countries and topics examined in this chapter as follows:

  • Promote emergency readiness: Improve co-ordination between providers (public and private) and enable centralised co-ordination in emergency situations. Introduce (or strengthen) mechanisms for swiftly releasing emergency funds. Prepare comprehensive response plans, considering the needs and perspectives of local communities, socio-economic costs of public health measures, mental health of the population, barriers and inequalities in care access and the importance of maintaining essential care services.

  • Promote workforce recruitment and strengthen primary care: Workforce limitations proved to be an important constraining factor in the pandemic. Empowering healthcare workforces to deliver quality and comprehensive care can maximise staff capacity. At the same time, a strong primary health system minimised healthcare disruptions and expedited the implementation of public health measures (vaccination).

  • Promote governance, trust, and data collection: Governance structures should integrate a whole-of-society approach for addressing shocks as severe as the COVID-19 pandemic. Trust is necessary for governance to be effective, and both communication campaigns and accurate decision making can improve confidence in government. Data collection and information systems are and will continue to be essential in informing accurate policy making and population health management during future shocks.

  • Promote population health and address health inequalities: Addressed extensively in Chapters 3 to 9 of this publication, health and healthcare in LAC present several weaknesses compared to other OECD countries. Addressing health risks such as obesity and wider determinants of poor health, like poverty, sanitation, and unemployment, are critical to improving preparedness for the next shock. Moreover, given the important disruptions to healthcare during the pandemic, management strategies should be established to prevent further forgone care and the consequent worsening of conditions.

References

[57] Amarante, V. (2022), “Fortalecimiento de los sistemas de protección social de la región: aprendizajes a partir de la pandemia de COVID-19”, Documentos de Proyectos, CEPAL, https://hdl.handle.net/11362/47830.

[51] Antiporta, D. and A. Bruni (2020), “Emerging mental health challenges, strategies, and opportunities in the context of the COVID-19 pandemic: Perspectives from South American decision-makers”, Revista Panamericana de Salud Pública, Vol. 44, p. 1, https://doi.org/10.26633/rpsp.2020.154.

[41] Arsenault, C. et al. (2022), “COVID-19 and resilience of healthcare systems in ten countries”, Nature Medicine 2022 5, pp. 1-11, https://doi.org/10.1038/s41591-022-01750-1.

[29] Bermudi, P. et al. (2021), “Spatiotemporal ecological study of COVID-19 mortality in the city of São Paulo, Brazil: Shifting of the high mortality risk from areas with the best to those with the worst socio-economic conditions”, Travel Medicine and Infectious Disease, Vol. 39, p. 101945, https://doi.org/10.1016/j.tmaid.2020.101945.

[45] Bernabe-Ramirez, C. et al. (2022), “HOLA COVID-19 Study: Evaluating the Impact of Caring for Patients With COVID-19 on Cancer Care Delivery in Latin America”, JCO Global Oncology 8, https://doi.org/10.1200/go.21.00251.

[31] Bilal, U., T. Alfaro and A. Vives (2021), “COVID-19 and the worsening of health inequities in Santiago, Chile”, International Journal of Epidemiology, Vol. 50/3, pp. 1038-1040, https://doi.org/10.1093/ije/dyab007.

[50] Borges, F. et al. (2020), “Impact of the COVID-19 pandemic in patient admission to a high-complexity cancer center in Southern Brasil”, Revista da Associação Médica Brasileira, Vol. 66/10, pp. 1361-1365, https://doi.org/10.1590/1806-9282.66.10.1361.

[35] Cabieses et al. (2020), “Migrantes venezolanos frente a la pandemia de COVID-19 en Chile: factores asociados a la percepción de sentirse preparado para enfrentarla”, Notas de Población, No. 111, CEPAL, https://hdl.handle.net/11362/46554.

[6] Carinci, F. et al. (2015), “Towards actionable international comparisons of health system performance: expert revision of the OECD framework and quality indicators”, International Journal for Quality in Health Care, https://doi.org/10.1093/intqhc/mzv004.

[44] CCSS (2020), Effects on health services, https://www.ccss.sa.cr/.

[64] CCSS (2020), Informe de resultados de la Evaluación de la Prestación de Servicios de Salud 2019 y monitoreo 2020, https://www.binasss.sa.cr/informeservicios2019.pdf.

[47] Chávez Amaya, C. (2021), Demanda asistencial de pacientes crónicos supera la capacidad de los servicios de salud | Ojo Público, Ojo Público, https://ojo-publico.com/3088/demanda-asistencial-de-pacientes-cronicos-supera-al-sistema-de-salud (accessed on 27 May 2022).

[65] Cho, M. and R. Levin (2022), “Implementación del plan de acción de recursos humanos en salud y la respuesta a la pandemia por la COVID-19”, Revista Panamericana de Salud Pública, Vol. 46, p. 1, https://doi.org/10.26633/rpsp.2022.52.

[25] Di Gennaro, F. (ed.) (2021), “On the analysis of mortality risk factors for hospitalized COVID-19 patients: A data-driven study using the major Brazilian database”, PLOS ONE, Vol. 16/3, p. e0248580, https://doi.org/10.1371/journal.pone.0248580.

[55] Di Paolantonio, G. (2020), “Fostering resilience in the post-COVID-19 health systems of Latin America and the Caribbean”, in Shaping the COVID-19 Recovery: Ideas from OECD’s Generation Y and Z, OECD, Paris, https://www.oecd.org/about/civil-society/youth/Shaping-the-Covid-19-Recovery-Ideas-from-OECD-s-Generation-Y-and-Z.pdf.

[5] Donabedian, A. (1988), “The quality of care. How can it be assessed?”, JAMA: The Journal of the American Medical Association, Vol. 260/12, pp. 1743-1748, https://doi.org/10.1001/jama.260.12.1743.

[13] Dong, E., H. Du and L. Gardner (2020), “An interactive web-based dashboard to track COVID-19 in real time”, The Lancet Infectious Diseases, Vol. 20/5, pp. 533-534, https://doi.org/10.1016/s1473-3099(20)30120-1.

[42] Doubova, S. et al. (2021), “Disruption in essential health services in Mexico during COVID-19: an interrupted time series analysis of health information system data”, BMJ global health, Vol. 6/9, https://doi.org/10.1136/BMJGH-2021-006204.

[3] ECLAC (2022), The sociodemographic impacts of the COVID-19 pandemic in Latin America and the Caribbean, Economic Commission for Latin America and the Caribbean, https://hdl.handle.net/11362/47923.

[16] ECLAC (2021), “Estadísticas e indicadores”, CEPALSTAT Bases de Datos y Publicaciones Estadísticas, https://statistics.cepal.org/portal/cepalstat/dashboard.html?indicator_id=4409&area_id=2313&lang=es.

[59] ECLAC/CEPAL (2022), “Suriname”, in Economic Survey of Latin America and the Caribbean 2022, CEPAL, https://repositorio.cepal.org/bitstream/handle/11362/48078/10/ES2022_Suriname_en.pdf.

[28] ECLAC/PAHO (2021), The prolongation of the health crisis and its impact on health, the economy and social development, Economic Commission for Latin America and the Caribbean, https://hdl.handle.net/11362/47302.

[24] Gebhard, C. et al. (2020), “Impact of sex and gender on COVID-19 outcomes in Europe”, Biology of Sex Differences, Vol. 11/1, https://doi.org/10.1186/s13293-020-00304-9.

[2] Herrera, C. et al. (2022), Building Resilient Health Systems in Latin American and the Caribbean: Lessons Learned from the COVID-19 Pandemic, World Bank, Washington DC, http://documents.worldbank.org/curated/en/099805001182361842/P1782990d657460cb0a2080ac0048f8b98f.

[23] Huang, B. et al. (2021), “Sex-based clinical and immunological differences in COVID-19”, BMC Infectious Diseases, Vol. 21/1, https://doi.org/10.1186/s12879-021-06313-2.

[22] IHME, 2021 (2021), Global Burden of Disease Study 2019, Institute for Health Metrics and Evaluation, Seattle, United States, https://vizhub.healthdata.org/gbd-results/.

[60] IMF (2021), Database of Country Fiscal Measures in Response to the COVID-19 Pandemic, International Monetary Fund, https://www.imf.org/en/Topics/imf-and-covid19/Fiscal-Policies-Database-in-Response-to-COVID-19.

[43] INEI (2021), Perú: Enfermedads No Transmisibles y Transmisibles, 2020, https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib1796/ (accessed on 31 May 2022).

[37] Kuehn, B. (2021), “Despite Improvements, COVID-19’s Health Care Disruptions Persist”, JAMA, Vol. 325/23, p. 2335, https://doi.org/10.1001/jama.2021.9134.

[63] LeRouge, C. et al. (2019), “Health System Approaches Are Needed To Expand Telemedicine Use Across Nine Latin American Nations”, Health Affairs, Vol. 38/2, pp. 212-221, https://doi.org/10.1377/hlthaff.2018.05274.

[30] Macchia, A. et al. (2021), “COVID-19 among the inhabitants of the slums in the city of Buenos Aires: a population-based study”, BMJ Open, Vol. 11/1, p. e044592, https://doi.org/10.1136/bmjopen-2020-044592.

[66] Maier, C., L. Aiken and R. Busse (2017), “Nurses in advanced roles in primary care: Policy levers for implementation”, OECD Health Working Papers, No. 98, OECD Publishing, Paris, https://doi.org/10.1787/a8756593-en.

[17] Mathieu, E. et al. (2021), “A global database of COVID-19 vaccinations”, Nature Human Behaviour, Vol. 5/7, pp. 947-953, https://doi.org/10.1038/s41562-021-01122-8.

[34] Migration data portal (2022), Datos sobre migración relevantes para la pandemia de COVID-19, https://www.migrationdataportal.org/es/themes/datos-sobre-migracion-relevantes-para-la-pandemia-de-covid-19 (accessed on 4.08.2022).

[46] MINSA (2021), Boletín Epidemiológico del Perú 2021 - Volume 30-SE24, https://www.dge.gob.pe/epipublic/uploads/boletin/boletin_202124_23_145452.pdf (accessed on 27 May 2022).

[21] 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://doi.org/10.1787/c5dc0c50-en.

[52] Nepogodiev, D. et al. (2022), “Projecting COVID-19 disruption to elective surgery”, The Lancet, Vol. 399/10321, pp. 233-234, https://doi.org/10.1016/s0140-6736(21)02836-1.

[33] NU. CEPAL/German Agency for International Cooperation (2021), The impact of COVID-19 on indigenous peoples in Latin America (Abya Yala): Between invisibility and collective resistance, Economic Commission for Latin America and the Caribbean, https://hdl.handle.net/11362/46698.

[58] OECD (2023), Inflation (CPI) (indicator), https://doi.org/10.1787/eee82e6e-en (accessed on 17 February 2023).

[38] OECD (2023), Ready for the Next Crisis? Investing in Resilient Health Systems, OECD Publishing, Paris, https://doi.org/10.1787/1e53cf80-en.

[18] OECD (2022), OECD Health Statistics, https://doi.org/10.1787/health-data-en.

[39] OECD (2022), Primary Health Care for Resilient Health Systems in Latin America, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/743e6228-en.

[27] OECD (2021), Health at a Glance 2021: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/ae3016b9-en.

[67] OECD (2021), Primary Health Care in Brazil, OECD Reviews of Health Systems, OECD Publishing, Paris, https://doi.org/10.1787/120e170e-en.

[54] OECD (2020), “Flattening the COVID-19 peak: Containment and mitigation policies”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/e96a4226-en.

[36] OECD (2020), SIGI 2020 Regional Report for Latin America and the Caribbean, Social Institutions and Gender Index, OECD Publishing, Paris, https://doi.org/10.1787/cb7d45d1-en.

[69] OECD/The World Bank (2020), Health at a Glance: Latin America and the Caribbean 2020, OECD Publishing, Paris, https://doi.org/10.1787/6089164f-en.

[62] OIT/ILO (2021), Respuestas de corto plazo a la COVID-19 y desafíos persistentes en los sistemas de salud de América Latina, Organización Internacional del Trabajo, https://www.ilo.org/lima/publicaciones/WCMS_768040/lang--es/index.htm.

[4] OPS/PAHO (2021), Resultados de salud desglosados por sexo en relación con la pandemia de COVID-19 en la Región de las Américas. De enero del 2020 a enero del 2021, Organización Panamericana de la Salud, https://iris.paho.org/handle/10665.2/53603.

[15] Oxford COVID-19 Government Response Tracker (2022), , https://ourworldindata.org/grapher/covid-containment-and-health-index (accessed on 1 August 2022).

[40] Pacheco, J. et al. (2021), “Gender disparities in access to care for time-sensitive conditions during COVID-19 pandemic in Chile”, BMC Public Health, Vol. 21/1, https://doi.org/10.1186/s12889-021-11838-x.

[26] PAHO (2021), Análisis de Género y Salud: COVID-19 en las americas, Organización Panamericana de la Salud, https://iris.paho.org/handle/10665.2/55557.

[56] Restrepo, J., D. Palacios and J. Espinal (2022), “Gasto en salud durante la pandemia por covid-19 en países de América Latina”, Grupo de Economía de la Salud, https://www.andi.com.co/Uploads/Documento%20de%20Trabajo%20Gasto%20en%20salud%20y%20covid-19.pdf.

[68] Ritchie, H. et al. (2020), Coronavirus Pandemic (COVID-19), https://ourworldindata.org/coronavirus.

[49] Superintendencia de Salud (2022), Casos GES (AUGE) acumulados a diciembre de 2021 - Biblioteca digital. Superintendencia de Salud. Gobierno de Chile., https://www.supersalud.gob.cl/documentacion/666/w3-article-20904.html (accessed on 7 July 2022).

[32] SUS (2022), Open Data SUS, Sistema Único de Saúde, Brazil, https://opendatasus.saude.gov.br/dataset/srag-2021-e-2022.

[53] The Economist (2020), “Covid-19 Flattening the curve”, The Economist, Vol. Vol. 434/No. 9183 - It's going global.

[14] The University of Maryland Social Data Science Center & Facebook (2020), The University of Maryland Social Data Science Center Global COVID-19 Trends and Impact Survey, in partnership with Facebook, https://covidmap.umd.edu/ (accessed on 23 August 2022).

[12] The World Bank (2022), World Development Indicators, https://databank.worldbank.org/home.

[11] The World Bank group (2021), Covid-19 High-Frequency Monitoring Dashboard, https://www.worldbank.org/en/data/interactive/2020/11/11/covid-19-high-frequency-monitoring-dashboard.

[19] UN (2022), World Population Prospects 2022: Summary of Results, https://www.un.org/development/desa/pd/content/World-Population-Prospects-2022.

[20] Wang, H. et al. (2022), “Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020–21”, The Lancet, Vol. 399/10334, pp. 1513-1536, https://doi.org/10.1016/s0140-6736(21)02796-3.

[7] WHO (2022), Estimates of Excess Mortality Associated With COVID-19 Pandemic (as of 25 March 2022)., World Health Organization, Geneva, https://www.who.int/data/sets/global-excess-deaths-associated-with-covid-19-modelled-estimates.

[1] WHO (2022), Methods for estimating the excess mortality associated with the COVID-19 pandemic, World Health Organization, Geneva, https://www.who.int/publications/m/item/methods-for-estimating-the-excess-mortality-associatedwith-the-covid-19-pandemic.

[8] WHO (2022), “Pulse survey on continuity of essential health services during the COVID-19 pandemic”, World Health Organization, Geneva, https://apps.who.int/iris/handle/10665/351527 (accessed on 4 July 2022).

[61] WHO (2021), Role of primary care in the COVID-19 response, World Health Organization Regional Office for the Western Pacific, https://apps.who.int/iris/handle/10665/331921.

[9] WHO (2021), WHO Coronavirus (COVID-19) detailed surveillance data dashboard, World Health Organization, https://app.powerbi.com/view?r=eyJrIjoiYWRiZWVkNWUtNmM0Ni00MDAwLTljYWMtN2EwNTM3YjQzYmRmIiwidCI6ImY2MTBjMGI3LWJkMjQtNGIzOS04MTBiLTNkYzI4MGFmYjU5MCIsImMiOjh9.

[48] WHO (2020), Peru, https://gco.iarc.fr/today/data/factsheets/populations/604-peru-fact-sheets.pdf (accessed on 27 May 2022).

[10] WHO (2020), WHO Coronavirus (COVID-19) Dashboard, World Health Organization, https://covid19.who.int/.

Notes

← 1. The full list of included countries can be found in the foreword.

← 2. Countries with available data: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Haiti, Honduras and Mexico.

← 3. Immunisation coverage reflects the percentage of children at age one that receive the final dose of the primary immunisation series in the respective vaccination programme in the recommended timeframe (Chapter 7).

Disclaimers

For the Organisation for Economic Co-operation and Development (OECD), the work is published under the responsibility of the Secretary-General of the OECD, and the opinions expressed and arguments employed herein do not necessarily reflect the official views of the Member countries of the OECD.

The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent.

The names of countries and territories used in this joint publication follow the practice of the OECD.

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

Photo credits: Cover © SALMONNEGRO-STOCK/Shutterstock.com.

Corrigenda to OECD publications may be found on line at: www.oecd.org/about/publishing/corrigenda.htm.

© OECD/The International Bank for Reconstruction and Development/The World Bank 2023

This Work is licensed under the Creative Commons Attribution Non-Commercial No Derivatives 3.0 IGO license (CC BY-NC-ND 3.0 IGO).