How well is the global priority group of elderly above 60 vaccinated? We don’t know for sure. Primary vaccination data only cover 55% of countries and 29% of the global elderly count. Booster data are even more sparse. But the data that are available suggest elderly vaccination has been a mixed bag: it is highly regressive and appears not always to have been a strong priority. No surprise therefore that the world’s excess death tally is so high. Even when vaccines were available, vaccinations remained lacking. The global priority group of the elderly has been poorly prioritized.
The concentration of excess mortality among the elderly
The recent release by the WHO of excess death estimates by age distribution is a wake-up call for anyone interested in how the elderly have been affected during this age-discriminating pandemic. The estimates (see chart below) suggest a whopping 12.2 million elderly (60+) have died during 2020 and 2021. They made up 82% of the global excess death tally of 14.9 million.
This heavy toll is a multiple of what the officially published data suggest. Officially reported COVID-19 deaths end 2021 totaled 5.4 million, which differs by a factor of 3 from the WHO’s excess mortality estimate. Granted the official data are concerned only with COVID-19 fatalities, whereas excess mortality captures the broader effects of the pandemic – both positive and negative. Still, the sheer magnitude of the difference suggests massive undercounting.
The chart below is split up by World Bank income classification, which groups countries into high income, upper-middle income, lower-middle income and low income (hence, the acronyms HIC, UMIC, LMIC and LIC). Notice how much the UMICs and LMICs contribute to the global death toll. That’s not only because they count many elderly people in the absolute, but also because the elderly mortality rates have been high among these groups (check out this post for the details).
Given the high levels of effectiveness of vaccines, one obvious reason why so many elderly died is of course that they were not vaccinated. The underlying reasons for lack of vaccination or undervaccination may be plenty. Early on in the vaccination campaign, there was a sheer shortage of vaccines everywhere. Had vaccines been better prioritized at the global level (so that, for example, elderly diabetics in Nigeria had been vaccinated before healthy teenagers in the US), many preventable deaths could have been prevented.
Of course, elements other than cross-border vaccine inequity are at play. Vaccination efforts within countries may be poorly prioritized. That may be intentional if the belief is that younger cohorts are more exposed to the virus and transmission can be better curtailed by prioritizing them. But it may also be unintentional, reflecting factors such as poor planning, logistical challenges, vaccine hesitancy and consumer selectivity on the demand side.
Primary and booster vaccination of the global 60+ cohort
We should mention at the start that the WHO database on vaccine uptake by age has considerable gaps. Primary vaccination data covers only 55% of all countries and only 29% of the global headcount of the elderly. Booster data availability is slightly worse. HIC and, interestingly, LIC data are more available than UMIC and LMIC data. The absence of China, India and Indonesia means that the share of the elderly covered in the middle-income country groups is even lower. These data limitations are documented in the annex below as well as in the note to each chart.
Where do we stand on elderly vaccination today? We will start by showing the full vaccination rate as of the latest date for the elderly group of the 60+. In fact, this is the age cohort we will be focusing on here. Full vaccination refers to having received the final shot of the primary protocol (this could be the 1st, 2nd or 3rd depending on the vaccine). The country observations are grouped by World Bank income classification and scaled by the size of the 60+ cohort. The horizontal lines represent the average vaccination rate of the country income group.
The chart clearly shows that primary vaccination of the 60+ has been highly regressive. HICs and UMICs achieved high 60+ vaccination rates, but the rates are much lower for LMICs and LICs. The dispersion around the population-weighted average is large too especially for the non-HICs, i.e. the developing countries.
What about boosters? The situation for boosters is even more regressive: the poorer half of the world (the LMICs and LICs) have barely had a go at protecting their 60+ elderly. HICs and UMICs present a better picture at 75% and 50%, but in light of the contagiousness of the Omicron family one would ideally want to have these numbers close to 100%. Again the obligatory caveat: data availability. As the footnote mentions, the data gap for UMICs and LMICs is huge, but also for HICs and LICs we are well below 50% coverage for countries and elderly counts.
Vaccination of the 60+ cohort vs the 60- cohort
To add further insight, let us compare the 60+ vaccination rates with the 60- ones. This comparison gives us context about the undervaccination of the elderly 60+. We would like to see 60+ rates at high levels. Ideally, we would also want to see a substantial gap with the 60- rates especially.
The data are not perfect. For some countries, we have had to truncate the 60+ rate because it exceeded 100%. This could be explained by inclusion of non-residents in the elderly vaccination drive, undercounting of the elderly cohort, errors in the reported data or other data discrepancies related to the population numbers. Conversely, some countries showed 0% vaccination among the 60+ despite despite reporting substantial vaccination of the 60- population (e.g. Azerbaijan and Kazakhstan). This looks questionable. In sum, where we have data, we need to remain cautious. But let that not prevent us from having a look at what the data actually tell us.
The chart below plots the primary vaccination rate of the 60- (x axis) against the 60+ (y axis). It also shows the 45-degree line. The vast majority of country observations lie above this line. Thus, 60+ cohorts are generally better vaccinated than their 60- counterparts as per the primary protocol. For several countries though this is not the case. Notice for example Mozambique in the LIC panel.
The distance from the 45-degree line matters too. Among countries above the 45-degree line, we see huge differences in the gap with 60- vaccination. Look for example at Guinea (GIN) and Rwanda (RWA) among LICs or Côte d’Ivoire (CIV) and Lesotho (LSO) among LMICs and compare these countries with others closer to the 45-degree line. The dispersion that we observe suggests a huge variation in the how the elderly groups were effectively prioritized.
We face a new data issue with respect to boosters. Booster data for the total population (which we pull from Our World in Data) are not always available. (Yes, in some cases, we have elderly data on boosters but not on primary shots.) For several countries, information on boosters for the total population is not contemporaneously available with the elderly data (which we get from WHO directly). If we were to restrict the data to what is contemporaneously available, we would cut out the group of low-income countries completely.
For this reason, we show below a chart that fills the booster data “up and down” over time. In other words, we pick the closest date with total population booster data available and match that with the elderly booster data. This inevitably introduces an error especially in countries where the pace of vaccination has been rapid. Luckily (from the angle of tackling this problem), most countries that lack contemporaneous data are also the ones where booster progress has been slow. So we can be relatively assured that the visualization below is not too far off.
As the chart shows, the big picture on booster vaccination is quite different from the earlier one on primary vaccination. The levels are a lot lower for both variables, which conforms with the fact that 76% of the global population remains unboosted as of today. Several countries are below the 45-degree line. Many countries face positive gaps but much more progress is needed. That will require a better prioritization of the elderly group vaccination.
This post discussed only part of a bigger puzzle. We have looked at a subset of what one might construe as the priority group. We have excluded health workers and others in risky professions, those with co-morbidities and the immunocompromised. Even so, we can state with confidence that the global priority group has been poorly prioritized when it comes to the elderly population.
Three points emerge:
Countries vary enormously in their extent of prioritization of the elderly group. This reveals itself in countries where vaccination rates for the 60+ are lower than those of the 60- but also where 60+ vaccination is only marginally higher than 60- vaccination. Given the data availability issue, the true magnitude of the problem is probably considerably understated.
DATA ANNEX: What doesn’t get measured, doesn’t get managed
From a public policy point of view, availability of high-quality, timely and comprehensive data is and should be the starting point of any targeting effort to better protect the elderly. We are facing however a shortage of information on how well the global priority group is being protected. Data coverage on vaccination by age is far from comprehensive.
At the global level only 58% of countries publish some data on primary vaccination by elderly cohort and 50% for boosters, as shown in the first table below. However, major countries such as China, India and Indonesia are excluded, so in terms of coverage of the elderly priority group by those countries that do report data, we are only at 38% for primary vaccination and 33% for boosters.
HICs and LICs (high-income and low-income countries) report far better than the UMICs and LMICs (upper-middle-income and lower-middle-income countries). That’s reflected in both the count of countries and the number of elderly covered. It is likely due to the enhanced focus of the international community on LICs and the sheer size of giant non-reporters within the middle-income categories.
The above statistics refer to any cut of the elderly cohort. That means that we include countries that report, as their only source of information, the 65+ cohort (as Ethiopia, Uruguay and United States do), the 55+ cohort (Syria) or the 50+ cohort (Australia and Iraq). Most countries, however, provide data on the elderly population in the form of a 60+ bucket or a more granular combination of the 60-69, 70-79 and 80+ cohorts, which can be easily transformed into 60+.
To wrap up, the charts below the show share of countries that have age-differentiated data (first chart) and the share of the elderly population of such countries in the global elderly population (second chart). These two variables are shown for primary and booster vaccination, for a broad definition of the elderly cohort and a narrow 60+ one, by World Bank income classification and for the world as a whole.