Impact of COVID-19 on mortality in Asia

ABSTRACT

The SARS-CoV-2 virus (termed COVID-19) has raised the mortality trends and diminished life expectancy globally. The impact of COVID-19 on mortality. however, differs across population groups, times, and locations. The current study examines some aspects of COVID-19’s impact on mortality in Asia using published estimates of excess mortality. The estimates of excess mortality all showed higher death rates than officially reported for Asian countries. Excess mortality in India and Indonesia was about 7 and 10 times higher than the reported mortality rates. In both these countries, COVID-19 was probably the leading cause of death. COVID-19 deaths were higher among men and older people. People with multiple comorbidities had a significantly higher risk of infection severity and fatality from COVID-19. There were significant inequalities in mortality rates across different socioeconomic groups, with poor households and low-paid workers disproportionately affected by COVID-19.

KEYWORDS:

• COVID 19
• morality
• Asia

This paper discusses a number of aspects of COVID-19’s impact on mortality in Asia (excluding Central and Western Asia). It sets the officially reported toll against estimates of excess mortality for Asian countries, while also comparing these with similar estimates for Western and Latin American countries. It examines differences between Asian countries, with a particular focus on the two countries with massive COVID-19 death tolls — India and Indonesia. Current vaccination levels in Asian countries are compared, and attention is also given to differential mortality among different population groups within countries.

The World Health Organization (WHO) estimate of global deaths directly and indirectly associated with the COVID-19 pandemic released in May 2022 amounts to 14.9 million. While this is a massive number, the global impact of mortality from COVID-19 remains considerably lower than that of the 1918–19 influenza epidemic, which caused an estimated 17 million to 50 million deaths in a world population only a quarter as large as today’s. But the death toll from COVID-19 is not over yet. It continues to climb, almost two and a half years after the pandemic began. The COVID-19 pandemic has induced the first decline in global life expectancy since World War II (Heuveline, 2022).

It should be noted that WHO’s estimate of global deaths is almost three times as high as the 5.42 million COVID-19 deaths reported by individual countries. There is considerable under-reporting of mortality from COVID-19, and the level of under-reporting is much greater in middle-income countries (especially lower middle-income countries) than in high income countries. But even in some high income countries, COVID-19 deaths are also considerably under-reported. There is great variation, for example, in the figures captured for non-hospital deaths.

Officially reported mortality from COVID-19 is much lower in Asia than in western countries

Table 1 shows officially recorded deaths per million population in a number of major Western and Latin American countries, compared with the three sub-regional groupings of East Asia, South Asia and Southeast Asia. Reported death rates are higher in Western and Latin American countries (with a few exceptions) than in any Asian country. If reported death rates are assumed to reflect reality, this is surprising, since one would expect the superior health infrastructure and resources in Western countries, compared with most Asian countries, to lead to lower COVID-19 mortality. Various reasons have been put forward to explain the differences, including generally younger populations in Asian countries, warmer climates, less travel, and genetic and immunological differences in the profiles of Caucasians compared with other ethnic groups (Landoni et al., 2021).

Table 1. Recorded COVID-19 Deaths per 100,000, Selected Countries and Territories.

The West and Latin America		East Asia		South Asia		Southeast Asia
7 May 2022
Hungary	481	Hong Kong	123	Sri Lanka	77	Malaysia	107
Brazil	308	Mongolia	65	Nepal	40	Indonesia	56
USA	306	S. Korea	45	India	37	Philippines	54
Greece	284	Japan	24	Afghanistan	19	Vietnam	44
Argentina	280	Taiwan	4	Bangladesh	17	Thailand	41
Italy	273	China	0.4	Pakistan	13	Myanmar	35
Russia	258					Singapore	23
UK	257					Cambodia	18
Mexico	247					Lao PDR	10
Germany	162
Netherlands	129
Denmark	107
Norway	55
Australia	29
14 July 2022
Hungary	486	Hong Kong	124	Sri Lanka	77	Malaysia	108
Brazil	312	Mongolia	64	Nepal	40	Indonesia	56
USA	312	S. Korea	48	India	37	Philippines	54
Greece	295	Japan	25	Afghanistan	19	Vietnam	44
Argentina	281	Taiwan	33	Bangladesh	17	Thailand	44
Italy	281	China	0.4	Pakistan	13	Myanmar	35
Russia	261			Iran	164	Singapore	24
UK	263					Cambodia	18
Mexico	248					Lao PDR	10
Germany	169
Netherlands	130
Denmark	112
Norway	64
Australia	40

Source: Johns Hopkins Coronavirus Resource Center (2022); Worldometers (2022)

The key issue here is whether the reported differences are real (perhaps reflecting the kinds of factors just cited), or simply reflect more complete recording in Western and Latin American countries. In seeking to answer this question, it is important to recognise that depending on official death tallies from COVID-19 for many countries will give a grossly inaccurate picture of the real impact of COVID-19 on deaths. This was well recognised early in the pandemic, leading to imaginative ways of modifying estimates of COVID-19 deaths. For example, in Jakarta, Indonesia’s capital city, from March through May 2020, the Indonesian government attributed 520 deaths to the coronavirus. But over 4,200 people, a more than normal figure, were buried in Jakarta cemeteries during those months, according to data from the city’s Department of Parks and Cemeteries (Wu et al., 2021). In the absence of other events that could have been expected to lead to a surge in burials, it is reasonable to attribute the additional burials to coronavirus deaths. In India, the press and civil organisations resorted to innovative sources to inflate the estimate, such as counting daily deaths in hospitals, the number of cremations and burials, corpses found floating in rivers, or obituaries published in local newspapers (Guilmoto, 2022). A different story emerged from Iran, where leaked data in 2020 indicated that the number of recorded deaths from coronavirus was nearly triple what the Iranian government was claiming (BBC, 2020).

Best method estimates of excess mortality from COVID-19

Clearly, methods need to be found to estimate the mortality from COVID-19 that are not reliant on official reports of deaths from COVID-19. The reliability of such reports is certain to vary considerably from country to country, depending partly on the completeness of vital registration systems and the extent to which deaths take place at home rather than in hospitals or health centres. But in assessing the impact of COVID-19 on mortality, the need goes beyond accurate recording of people who died because they had COVID-19. The concept of ‘excess mortality’ is relevant here.

Excess mortality associated with COVID-19 is defined as the difference between the total number of deaths that have occurred and the number of deaths that would have been expected in the absence of the pandemic. As explained by the WHO (2022a; see also Giattino et al., 2022),

• Excess mortality includes deaths attributable directly to COVID-19 that were counted and reported to WHO countries

• It includes deaths attributable directly to COVID-19 that were not counted or reported by countries

• It also includes deaths indirectly associated with COVID-19, due to other causes and diseases, resulting from the wider impact of the pandemic on health systems and society

• And it is minus any deaths that would have occurred under normal circumstances but were averted due to pandemic-related changes in social conditions and personal behaviours e.g., less traffic deaths or influenza deaths due to local lockdowns and less travel

Beginning fairly early in the pandemic, estimates of the ‘true’ COVID-19 toll, based on a range of approaches designed to come closer to the real number of COVID-related deaths (as noted above), have been undertaken for a wide range of countries (see Adam, 2022) as well as for individual countries such as Russia (Kobak, 2021) and India (Guilmoto, 2022). Major comparative studies of large numbers of countries have been undertaken by the New York Times, The Economist, Karlinsky & Kobak, 2021; and a team funded by the Bill and Melinda Gates Foundation (Wang et al., 2022). These were all based on comparing officially reported COVID-19 deaths in particular periods with estimates of overall excess mortality in the same period. They all showed much higher death rates than those officially recorded.

The two most recent and complete sets of estimates of excess mortality both cover the period between Jan. 1, 2020 and Dec. 31, 2021. The WHO estimates show that while for the world as a whole, 5.42 million COVID-19 deaths have been officially reported, 14.91 million people have died worldwide because of the COVID-19 pandemic, a figure almost three times as high. The estimated excess deaths were 4.5 million in 2020 and 10.4 million in 2021. The corresponding estimate from the Gates Foundation-supported team was 18.2 million.

Methodology of major comparative studies

Given the lack, in many regions (including much of Asia) of mechanisms for systematic mortality surveillance that would be required to monitor effectively the impact of COVID-19 on disease incidence and death, considerable effort has gone into improving the methodology of ‘excess mortality’ estimation. As noted by WHO, ‘calculating excess mortality is challenging’ (WHO, 2022c, p. 6). This section of the paper will summarise briefly the methodology used in two of the major comparative studies – the Gates Foundation-supported study (Wang et al., 2022) and the WHO study (WHO, 2022a). The procedures were quite complex.

Wang et al.’s (2022) study collected all-cause mortality reports for 74 countries and territories and 266 subnational locations (including 31 locations in low-income and middle-income countries) that had reported either weekly or monthly deaths from all causes during the pandemic over the course of 2020 and 2021 and for up to 11 years previously. In addition, excess mortality data were obtained for 12 states in India. Excess mortality over time was estimated as observed mortality minus expected mortality. An ensemble model was developed to predict expected deaths in the absence of the COVID-19 pandemic for the years 2020 and 2021. According to this study, the number of excess deaths due to COVID-19 was largest in the regions of South Asia, North Africa and the Middle East, and Eastern Europe.

The WHO (2022a) study was the product of a Technical Advisory Group for COVID-19 Mortality Assessment, convened jointly by the WHO and the United Nations Department of Economic and Social Affairs. Its methodology is reported in detail in WHO (2022b). The majority of Asian countries do not have all-cause mortality (ACM) data on an annual basis. For these countries, an ACM count had to be estimated. The WHO team followed a Bayesian approach to obtain a predictive distribution for the excess numbers of deaths.

Excess mortality estimates for world regions

Before focusing on excess mortality estimates for individual Asian countries, it is necessary to set Asian regions in context by examining the extent of excess mortality in major world regions. Table 2 is based on the data from Wang et al. (2022). While for the world as a whole, the excess mortality rate is three times the reported COVID-19 mortality rate, the excess is much lower for high income countries. This is excepting high income East Asian countries, where a higher estimated mortality rate for Japan raises the ratio. Sub-Saharan African countries have a very high ratio of excess mortality to reported COVID-19 mortality. But the Sub-Saharan African ratio is not much higher than South Asia, which on a world level is very high. Southeast Asia’s is considerably lower, though there is much inter-country variation.

Table 2. Relationship Between Reported COVID-19 Mortality Rate and Estimated Excess Mortality Rate, 2020-2021, Various Groupings and Regions.

Region or country	Reported covid-19 mortality rate (per 100,000)	Estimated excess mortality rate (per 100,000)	Ratio between excess mortality rate and reported covid-19 mortality rate
Global	39.2	120.3	3.07
High income	87.9	125.8	1.43
High income Asia Pacific	6.8	30.9	4.56
High income North America	121.5	167.1	1.38
Western Europe	94.7	140.0	1.48
Central Europe	157.8	315.7	2.00
Latin America	134.7	254.0	1.89
Eastern Europe	194.4	345.2	1.78
Sub-Saharan Africa	7.2	101.6	14.20
Central Asia	30.2	160.6	5.32
South Asia	16.0	151.7	9.46
Southeast Asia	24.0	90.8	3.78

Source: Wang et al. (2022).

Note: Western Europe includes southern Europe. Latin America includes Caribbean

Excess mortality estimates for Asian countries

The figures in Table 2 show that the different regions of Asia have a higher ratio of estimated excess mortality to reported COVID-19 mortality than high income North America, Europe or Latin America. In the case of South Asian countries, the ratio is particularly high. It is important to delve more closely into the data for some individual Asian countries (see Table 3). The ratio is extremely high in Afghanistan, Bangladesh, India, Nepal, Pakistan, and Lao PDR.

Table 3. Relationship Between Reported COVID-19 Mortality Rate and Estimated Excess Mortality Rate, 2020-2021, Asian Regions and Countries.

Region or country	Reported covid-19 mortality rate (per 100,000)	Estimated excess mortality rate (per 100,000)	Ratio between excess mortality rate and reported covid-19 mortality rate
High income Asia Pacific	6.8	30.9	4.56
South Asia	16.0	151.7	9.46
Afghanistan	8.1	211.9	26.06
Bangladesh	9.2	134.7	14.72
India	18.3	152.5	8.33
Kerala	84.3	165.1	1.96
Delhi	65.6	160.2	2.44
Maharashtra	59.6	259.5	4.36
Karnataka	28.7	212.9	7.41
West Bengal	10.5	116.7	11.14
Bihar	5.2	137.7	26.68
Uttar Pradesh	4.8	108.2	22.58
Nepal	19.1	202.5	10.61
Pakistan	6.6	152.6	22.99
Baluchistan	1.4	67.8	49.64
Punjab	5.9	141.1	23.98
Sindh	7.8	127.7	16.30
Southeast Asia	24.0	90.8	3.78
Cambodia	9.0	52.5	5.80
Indonesia	27.5	140.7	5.11
Lao PDR	2.8	34.7	12.47
Malaysia	53.3	81.4	1.53
Myanmar	19.2	100.9	5.25
Philippines	23.8	82.9	3.48
Sri Lanka	34.3	29.5	0.86
Thailand	15.3	24.9	1.62
Timor Leste	4.8	51.1	10.65
Vietnam	16.6	37.0	2.23

Source: Wang et al. (2022).

Sub-national estimates are available for India and Pakistan. The ratio between excess mortality and reported COVID-19 mortality in Table 3 varies dramatically between Indian states. It is low in Kerala, Delhi and Maharashtra, but much higher in West Bengal and extremely high in states such as Bihar and Uttar Pradesh, where the coverage of health services and recording of deaths is more limited. In Pakistan, the ratio is extremely high in Baluchistan, and quite high in Punjab.

Focusing on the most recent data from WHO, for the world as a whole, lower-middle-income countries have by far the largest number of excess deaths (7.87 million). It should be noted that five Asian countries (India, Indonesia, Pakistan, Bangladesh and the Philippines) make up a considerable proportion of the population of these lower middle-income countries.

Asian countries rank prominently among those where the ratio of excess deaths in 2020 and 2021 compared with reported COVID-19 deaths was the highest (see Table 4). The estimates of excess mortality in Asian countries from the two studies are roughly consistent in many cases, but diverge considerably in others. In India, Indonesia and the Philippines, excess mortality estimates are fairly close, but for Pakistan and Bangladesh, they diverge widely, with the WHO estimates giving a much lower multiple.

Table 4. Countries With Largest Multiples of Excess Deaths.

	WHO (2022a)	Wang et al. (2022)
Egypt	11.6 times	12.2 times
India	9.9 times	8.3 times
Pakistan	8 times	23.0 times
Indonesia	7.1 times	5.1 times
Bangladesh	5 times	14.7 times
Bolivia	4.5 times	8.2 times
Serbia	4.4 times	3.0 times
Kazakhstan	4.2 times	3.8 times
Philippines	3.6 times	3.5 times
Russia	3.5 times	1.6 times

Note: The table includes only countries with at least 10,000 officially recorded COVID deaths.

In Table 1, it is clear that Western countries tend to have much higher recorded COVID-19 deaths per million than most Asian countries. This raises the question: are these differences real, or do they reflect more complete recording in Western countries? We can now attempt to answer more generally the question on differences in COVID-19 death rates between Western and Asian countries, relying on data from the two sets of ‘excess mortality’ estimates already discussed.

Table 5 shows the excess in mean deaths per 100,000 population in Asian countries as compared with Western countries over the 2020–2021 period, based on the WHO estimates and those by Wang et al. (2022). These can be compared with the figures in Table 1, though it must be kept in mind that the figures in Table 1 refer to a somewhat longer period – up to May 2022, and would be even smaller if the cut-off was at end-2021.

Table 5. Excess Mean Deaths Per 100,000 Population, Western Countries and Three Groups of Asian Countries, 2020–2021.

Western countries		East Asia		South Asia		Southeast Asia
WHO (2022a)
USA	140	Mongolia	0	Sri Lanka	−21	Malaysia	12
Italy	133	S. Korea	6	Nepal	55	Indonesia	187
Spain	111	Japan	−8	India	171	Philippines	84
UK	109	China	−0.2	Afghanistan	57	Vietnam	−3
Russia	367			Bangladesh	43	Thailand	11
Germany	116			Pakistan	52	Myanmar	40
France	63					Singapore	13
Sweden	56					Cambodia	37
Canada	29					Lao PDR	12
Australia	−28
Wang et al. (2022)
USA	179	Mongolia	91	Sri Lanka	30	Malaysia	81
Italy	227	S. Korea	4	Nepal	203	Indonesia	141
Spain	187	Japan	44	India	153	Philippines	83
UK	127	China	0.6	Afghanistan	N.A.	Vietnam	37
Russia	375			Bangladesh	135	Thailand	25
Germany	121			Pakistan	153	Myanmar	101
France	124					Singapore	−15
Sweden	91					Cambodia	53
Canada	61					Lao PDR	35
Australia	−38

Source: WHO (2022a); Wang et al. (2022)

In interpreting the figures in Table 5, there appears to be two problems with the estimates:

1. For quite a number of wealthy Western countries, the reported COVID-19 death rates from the WHO database and the Johns Hopkins Coronavirus Resource Centre (see Table 1) are substantially higher than the excess mortality estimates from either the WHO or the Wang et al. (2022) studies. This is surprising. For countries such as the USA, Brazil, Italy, France, Netherlands, Norway, where vital registration systems are quite good (though there are a number of reasons to expect some under-recording of COVID-19 deaths), it would be expected that estimated excess mortality in these countries would be higher, not lower, than recorded COVID-19 mortality in most cases.

2. There are considerable discrepancies for particular countries between the two sets of estimates in Table 5. For example, Italy, France and Canada give widely differing estimates. Focusing on Asian countries, Mongolia, Japan, Nepal, Bangladesh, Pakistan, Malaysia, Vietnam and Myanmar show considerable discrepancies as well. However, it should also be noted that for some countries with very large populations – the USA, Russia, China, India, Indonesia and the Philippines – the two sets of estimates are quite close (although in the case of the USA, they are well below the officially reported COVID-19 deaths). These six countries account for a very substantial proportion of the total world population (47 per cent), and for about 40 per cent of estimated excess mortality due to COVID-19 (based on Wang et al., 2022).

Table 5 confirms two observations about Table 1. First, many Western countries had considerable excess death rates – Russia, USA, Italy, UK, etc., whereas East Asian countries did not. Many countries in South Asia and Southeast Asia also had lower excess death rates than the West, but this was not true of some major countries - India, Indonesia, the Philippines and, to a lesser extent, Nepal, Afghanistan and Pakistan. It also appears strange that two countries affected by major conflict situations – Afghanistan and Myanmar, where the official recording of COVID-19 deaths was likely to have been poor, did not feature more prominently among countries with high excess death rates. India and Indonesia had a higher excess death rate than Western countries such as USA, Italy, Spain and the UK, and because of their very large populations, their high excess death rates have a major impact on excess death rates for the world as a whole. India alone accounted for 22 per cent of the world’s total excess deaths.

Individual country effects of differential timing trends in emergence of sub-variants of COVID-19

The timeline of the emergence of different sub-variants of COVID-19 greatly affected trends in mortality rates. As noted by WHO (2022a), the impact of the pandemic occurs over several waves, with each characterised by unique regional distributions, mortality levels and drivers. In December 2020, the Alpha variant emerged, rapidly displacing the original strain of the virus; it was 50 per cent more transmissible in humans compared with the original virus. The Beta and Gamma variants then quickly appeared, succeeded by the Delta variant in April 2021, which was about 50 per cent more transmissible than the Alpha variant. In November 2021, the Omicron variant emerged, which is evasive of the antibodies that vaccination and prior infection confer, and can thereby breach our immune defences. Fortunately, protection against hospitalisation is not eroded to the same extent. Omicron is capable of re-infecting individuals who have been exposed to other forms of the virus. Though the Omicron variant has seen a massive upsurge in COVID-19 infections, it has also seen a much lower ratio of hospitalisations and deaths than earlier variants.

Comparing excess death rates between Western and Asian countries is likely to show considerably different relativities at particular points in time compared to other time periods. This is because of the earlier onset in Western and Latin American countries (and Iran) than in most of the Asian region, and the tendency for COVID-19 death rates to accelerate in waves (largely influenced by the emergence of new sub-variants – particularly Alpha, Delta and Omicron), with periods of greater quiescence in between. Despite very high COVID-19 caseloads in many parts of the world during 2020, some countries (e.g., Thailand, Vietnam, Cambodia, Malaysia, and Sri Lanka) appear to have largely avoided COVID-19 for the whole of 2020, and indeed most well into 2021, before experiencing sudden upsurges. Cambodia’s first reported COVID-related death, for example, did not occur until March 11, 2021. Notwithstanding its limited public health resources, it undoubtedly benefited from the low case numbers in neighbouring Thailand and Vietnam, its young and not very dense population, and international aid and government attention to the issue. India, Pakistan, Nepal, Indonesia, and Malaysia all experienced a strong decline in reported cases in some periods, despite no obvious change in reporting procedures.

Malaysia and Sri Lanka deserve special attention here, because of the remarkable similarity of their COVID-19 history. As former British colonies, they share similarities, such as a relatively small in area (excluding East Malaysia), and well-established public health systems with efficient surveillance. Both had very few COVID-19 deaths before 2021, but a wave of deaths began around May 2021, with a major upsurge in deaths between August and September 2021. In these two countries, public policy appears to have worked successfully in containing the pandemic in its early stages, but once the Delta variant became established, the death toll mounted rapidly (Arambepola et al., 2021).

How much has COVID-19 raised overall death rates in Asian countries?

It is possible, using the estimated excess mortality estimates already discussed, to provide an estimate of the percentage by which COVID-19 raised the overall death rates in individual Asian countries (see Table 6). This is a fairly rough exercise. For one, the numbers of expected deaths for these countries are derived from the United Nations Population Division’s World Population Prospects, which was published in 2019, so mortality estimates for individual countries for 2020 and 2021 are projections. However, the much more important element of uncertainty derives from the need to estimate excess mortality resulting from the COVID-19 pandemic. Such estimates for countries with good vital registration systems are not subject to much uncertainty, as reflected in the fairly close correspondence between the findings of the WHO and Wang et al. (2022) for most European countries and for Asian countries such as Republic of Korea, Japan and Singapore. In these three Asian countries, moreover, COVID-19 did not lead very much to an increase in mortality.

Table 6. Percentage by which Estimated Excess Mortality Exceeded Expected Mortality, Various Asian Countries, 2020–2021.

	WHO (2022a), %	Wang et al. (2022), %
Iran	14.2	33.6
Indonesia	14.0	20.0
India	11.6	19.8
Philippines	6.9	13.6
Afghanistan	4.6	39.7
Nepal	4.3	33.0
Bangladesh	3.9	22.5
Pakistan	3.8	28.2
Cambodia	3.0	8.5
Myanmar	2.3	11.0
Singapore	1.3	−3.0
Malaysia	1.1	13.7
Lao PDR	0.9	5.0
Thailand	0.7	3.1
Republic of Korea	0.5	0.7
Vietnam	0.2	5.7
Mongolia	0	13.2
China	−0.1	0.1
Japan	−0.4	3.9
Sri Lanka	−1.5	4.3

Source: Expected mortality derived from UN Population Division, World Population Prospects as assessed in 2019.

However, for some countries such as Iran, Afghanistan, Nepal, Bangladesh, and Pakistan, there are enormous discrepancies in the estimated excess mortality, and for other countries including Indonesia, India, Myanmar, Philippines and Malaysia, there are very substantial discrepancies. Given the high level of expertise in the teams which prepared each of the two sets of estimates, these discrepancies serve to highlight the fraught nature of estimating excess mortality from COVID-19 in situations where effective vital registration of deaths is severely lacking. It should be noted, however, that Malaysia has a fairly effective vital registration system, but still shows enormous differences between the two sets of excess mortality estimates.

Over the 2-year period, 2020-2021, COVID-19 was the leading cause of death in the Americas and the third leading cause of death in Europe (Troeger, 2022). In the USA, it was the third largest cause of death, only behind heart disease and cancer. In Italy, France, Spain, Brazil, Mexico and Peru, it was the leading cause of death (Troeger, 2022). Globally, based on the ‘excess mortality’ estimates already discussed, it was the second leading cause of death, responsible for one out of every 10 deaths.

What can we say about the situation in Asian countries? Not very much. But clearly, in economically advanced countries of East and Southeast Asia, COVID-19 was well down the list of causes of death. This was far from the case in many other countries of South and Southeast Asia. In Iran, Indonesia and India, both sources in Table 5 show excess mortality so high that it probably indicates that COVID-19 was the leading cause of death. For Afghanistan, Nepal, Bangladesh and Pakistan, based on the estimates by Wang et al. (2022), the same could be unequivocally said, but the WHO estimates are much lower, and would imply a much lower share of COVID-19 in total deaths.

Asia’s two COVID-19 Standouts: India and Indonesia

India

India has the third highest number of officially reported COVID-19 deaths in the world, but by far the highest number in reality. Reaching a consensus on the excess deaths in India relating to the pandemic has been complicated by the Indian government’s insistence on the veracity of its official pandemic death toll, despite considerable evidence that this is not the case.

In the early stages of the pandemic, India appeared to have suffered relatively few COVID-19 deaths. There were many speculations about the reason – one of course being that India had a smaller proportion of the population in old-age groups that were more susceptible to mortality from COVID-19. It was also argued that Indians have a different immune response – with some observers speculating that this could be related to widespread exposure to a tuberculosis vaccine that may stimulate the immune system.

The pandemic in India was expected to peak in June-July 2020, and indeed there was a slackening after that point, but India suffered a second wave of infections and deaths from February to September 2021 (most severe in the April-June period).

India is a big country, and the spread of the virus was not homogeneous. There are a wide range of situations throughout India. In particular, the majority of India’s population live in rural areas, where the spread of the virus may not be as rapid as in big cities. Outbreaks in one area could conceivably be dealt with locally, without affecting other areas. In the cities, the mass of low-income workers (including daily-waged workers) live in crowded conditions, unable to isolate or afford to stop work under any circumstances. The vulnerable elderly who live in multi-generational households are also unable to isolate.

Overall, in India about half the total deaths occur at home, especially in villages. Out of an estimated 10 million deaths annually, seven million do not have a medically certified cause of death and three million are simply not registered (Biswas, 2022). There are enormous differences in the quality of data. Registration is particularly low in the poorest states such as Uttar Pradesh and Bihar; by contrast, Kerala has excellent health indicators.

While India has sophisticated epidemiologists and public health planners, the Indian health system is not in good shape to deal with COVID-19. India spends just over 1 per cent of GDP on public health, ranking 170 out of 188 countries (Katiyar, 2021). Nearly 70 per cent of health costs are paid by citizens out of pocket, pushing many below the poverty line when health crises occur (Rao, 2018).

Test kits, protective equipment, hospital beds, ICU, and oxygen were in desperately short supply during the crisis months of mid-2020. Oxygen is particularly important for those catching the virus at younger ages – these are less likely to need ventilators.

When the Indian government (in its confrontation with the WHO in May 2022) claimed that 99.9 per cent of all COVID-19 deaths to date were registered, Shashi Tharoor – former Indian cabinet minister and former UN under-secretary-general – termed this a ‘preposterous claim’ (Tharoor, 2022). While completeness of registration of death has certainly improved considerably over time in India, in 2015 it was still only 44.2 per cent in Uttar Pradesh (Kumar et al., 2019).

Guilmoto’s (2022) study for India used four independent population samples (the official COVID-19 dataset for the state of Kerala, the employees of the Indian railways, Indian elected representatives, and Karnataka’s schoolteachers) to characterise the age and sex profile of India’s COVID-19 mortality and then assess its severity by fitting mortality patterns to the number of deaths reported in these samples. He estimated that India had experienced 3.2–3.7 million COVID-19 deaths up to November 2021. This estimate was roughly in line with the Indian estimates in the two major international studies of excess mortality. Guilmoto argued that there were two main reasons for the underestimation of COVID-19 deaths: first, many deaths go unrecorded as such for a lack of prior PCR testing; the cause of death is often selectively attributed to comorbidities (diabetes, asthma, etc.) and other apparent factors such as heart attacks. Second, COVID-19 is a source of shame and carries social stigma, both at the family level, and at the political level, thus providing no incentive to faithfully and exhaustively record cases of COVID-19.

Indonesia

Indonesia has experienced three waves of COVID-19 infections and deaths - the first around January-February 2021, the second (far larger) around July-September 2021, related to the more transmissible Delta variant, and the third around March-April 2022 (see IHME, 2022). More importantly, reported daily cases were greater in the third wave than in the second wave, but reported deaths were much lower. As in many other countries, this was due to the onset of the Omicron variant, which has proven highly transmissible, but with a much lower mortality rate than the Delta variant.

In Indonesia, there was extreme pressure on intensive care unit beds in the first and second waves. The pressures in Indonesia matched more or less those in India during their second waves, which occurred about the same time (see IHME, April 7, 2022, Figure 25.1). From June-August 2021, the official number of deaths from COVID-19 in Jakarta was 6,000. But in the same period, 18,000 excess deaths were recorded in the city, because there were thousands of people who died at home without being tested as they could not even get to a hospital (Al Jazeera, 2021). The wave ended in September 2021, when recorded case numbers fell dramatically.

Early in the mid-2021 s wave, the vaccination rate in Indonesia remained very low – around 6 per cent. Subsequently, the percentage of the population vaccinated increased considerably, reaching 81.7 per cent with a first dose and 56.2 per cent with a second dose in January 2022 (Azanella, 2022). This still leaves a substantial proportion of the population incompletely vaccinated. But it seems likely that the substantial drop-off in reported daily cases after September 2021 (while recognising that reported cases are still severely under-recorded) was related to the rapid increase in the proportion of the population that was vaccinated. As in India, the trend in new cases moved from the big cities (especially Jakarta) early in the pandemic to villages and from Java to the other islands where there is lesser testing capacity.

The Indonesian government’s response has been criticised for stressing the economic impacts of the crisis rather than its health dimension. The government was reluctant to take control measures that would harm the economy, continually lagging in restricting people’s movements and economic activities (Aspinall, 2021). It also restricted travel for only a 2-week mudik period in May 2021, when people return to their home areas for Lebaran (Eid) celebrations. Many still managed to get around the travel restrictions by leaving early or returning after the restricted period.

Levels of vaccination in Asian countries

It is generally agreed that vaccination is key to lowering death rates from COVID-19. Poorer countries – especially those in Africa – have been left behind in the race to vaccinate, compared to wealthier countries, which could afford to purchase greater supplies of vaccines. However, most Asian countries have high vaccination rates (see Table 7), many of them above vaccination rates in the USA, and almost all of them above Russia. However, Asian countries that are war-torn or suffering from civil strife (Afghanistan, Syria, Iraq, and Myanmar) have significantly lower vaccination rates.

Table 7. Asian Countries, Percentage of Population Vaccinated Against COVID-19, 18 May 2022.

	% Received one dose	% Fully vaccinated
Singapore	87.9	87.4
Rep. of Korea	87.1	86.2
Cambodia	90.6	86.1
Hong Kong	88.8	83.3
Malaysia	87.7	NA
Vietnam	85.9	81.6
Japan	81.9	80.7
Thailand	80.9	74.4
Lao PDR	80.8	69.4
Bangladesh	79.0	71.7
Nepal	78.1	69.0
Iran	77.7	69.6
Sri Lanka	78.3	66.3
India	73.8	64.1
Indonesia	73.8	61.4
Mongolia	70.5	67.5
Pakistan	62.2	56.6
Myanmar	57.3	45.0
Iraq	26.9	18.9
Afghanistan	14.1	12.4
Syria	14.7	9.4

Source: Reuters, 2022

Differential mortality among different groups within countries

Although cases of COVID-19 are spread rather evenly across different sex and age groups in the population, deaths are much more concentrated (Bauer et al., 2021). In general, international research shows that older age and being male were significantly associated with higher COVID-19 mortality rates (Guilmoto, 2020; Mulchandani et al., 2022; Nieves et al., 2021; Reeves & Deng, 2021). COVID-19 death rates increase with extreme regularity from age 15–90 + and follow a typical exponential Gompertz model, with rates doubling every six years (Guilmoto, 2020). This means that deaths are heavily concentrated among the elderly population. About one third of all deaths among both males and females occur among the population aged 80 years and older (WHO, 2022c, p. 7).

It is often noted that the main effect of COVID-19 is to hasten the deaths of people who are likely to die fairly soon anyway. For example, in Sweden, 90,000 people die each year. If 10,000 people in Sweden die from COVID-19 this year, it does not mean that 100,000 people will die this year, but rather something like 94,000; about 6,000 people will die prematurely from COVID-19, who would have died later from other causes.

Worldwide, male excess mortality from COVID-19 was about 50 per cent higher than women (Reeves & Deng, 2021), but the extent of excess male mortality also differed by age. In nine Western countries, for example, men aged 40–59 were almost 2.5 times more likely to die from COVID-19 than women of the same age, a rather extreme differential compared with typical sex differentials in mortality at such ages (Guilmoto, 2020). In South Asia, over 3 out of 4 COVID-related deaths were men (Nieves et al., 2021, p. 4). An earlier study (Dehingia & Raj, 2021) argued that while COVID-19 incidence is higher for men than women in India, the case fatality rate is higher for women than for men. These earlier findings are put into question by the later information. However, their point that India’s reliance on hospital-level information for the estimation of fatality rates could result in over-estimation of female case fatality rates because women are less likely to be admitted to hospital than men, and therefore more severe cases of COVID-19 are likely to be over-represented among those women who are admitted, should be recognised.

The main reasons for the over-representation of men in COVID-19 deaths are most likely the higher incidence of chronic diseases (e.g., cardiovascular disease) and comorbidities (e.g., hypertension), gender differences in risky behaviours such as smoking, and potentially, immunological differences between men and women. There is some evidence that men show a comparatively weaker or less effective immune response to COVID-19 infection (see Nieves et al., 2021).

The evidence on differentials according to socio-economic status is more widely available for Western countries than for Asian countries. Significant inequalities have been revealed across different socioeconomic profiles, disproportionately affecting the vulnerable and economically disadvantaged (WHO, 2022c, p. 7). Low-paid workers in cities in low income Asian countries tend to live and work in very crowded conditions, and find it impossible to isolate because of the need to keep working. In times of great stress in dealing with COVID-19, such as in India in 2020, decisions about where the next meal is coming from, whether the limited supply of water should be used for hand washing or washing of cooking and eating utensils, and how to pay for urgent medical attention are major issues for poor households.

Asian studies (mainly based on data for India, Indonesia, Bangladesh, Nepal and Thailand) show that the presence of multiple comorbidities, especially diabetes and cardiovascular conditions, had a strong significant association with infection severity and fatality from COVID-19 (Mulchandani et al., 2022, p. 56). Severe and critical patients were more likely to have multiple comorbidities (three or more) or have diabetes accompanied by other illnesses like hypertension and heart disease.

Conclusion

Over the 2020–2021 period, the COVID-19 pandemic has led to the first annual declines in life expectancy in major Asian countries since World War II: 3.5 years in the Philippines and 2.6 years in India (Heuveline, 2022, p. 41). Though Heuveline’s study does not give an estimate for Indonesia, a decline of more than 2 years in life expectancy could also be expected, based on available excess mortality estimates. The major reason for the decline in life expectancy in Asian countries was due to mortality changes at older ages. Feng and Gu (2023) estimate that nearly 70 per cent of life expectancy changes was due to higher mortality at older ages. On any reckoning, these are major impacts. Yet we continue to face an enormous degree of uncertainty in estimating the true death toll from COVID-19 in Asian countries.

As Wang et al. (2022, p. 1513) observed, ‘strengthening death registration systems around the world, long understood to be crucial to global public health strategy, is necessary for improved monitoring of this pandemic and future pandemics’. It is surely the case that without significant improvement in vital registration systems in Asian countries, tracking the progress of future pandemics will encounter the same kinds of problems that have bedevilled efforts to understand and deal with the COVID-19 pandemic in the region. In the absence of considerable improvement in vital registration systems, the ‘excess mortality’ methodology developed to derive better estimates of COVID-19 mortality can play a useful role, though it will certainly not provide clear answers to all questions.

Since the disadvantaged tend to suffer disproportionately from the effects of the pandemic, testing surveillance for COVID-19 should be equitable, and should cover population sub-groups from all socioeconomic strata. Not only this, but in line with the equity aims of the UN Sustainable Development Goals, countering the deleterious effects of COVID-19 should focus particularly on the disadvantaged sections of the community.

Even in countries where the increase in death rates from COVID-19 is relatively small (e.g., Japan, Republic of Korea and Australia), it should be recognised that COVID-19 puts greater pressure on health facilities, particularly intensive care units, than many other main causes of death. Most deaths from certain causes – e.g., suicide, road accidents and stroke – do not end up in ICUs at all. India, for example, has 150,000 people dying in road accidents each year.

There is much that we do not know. For example, we do not know how much damage to longer-term health is likely to result from cases of severe COVID-19 (Hampshire & Menon, 2022). While the downward trend in global excess deaths since March 2022 is an encouraging sign, it is still impossible to predict the future course of the pandemic, including the emergence of new sub-variants with greater or lesser potential for contagion, greater or lesser risks of breaching our immune defences, and the protection that vaccinations induce, at least in part, against hospitalisation and death, and hence lead to greater or lesser risks of mortality. The forecast that COVID-19 will gradually become endemic, like influenza, and thus more readily dealt with, remains more a hope than a well-supported expectation.

Disclosure statement

No potential conflict of interest was reported by the author(s).