Excess death estimates from multiverse analysis in 2009-2021

INTRODUCTION

Calculation of excess deaths is considered to be a very useful tool for estimating patterns of mortality changes over time in different countries and the impact of major events, such as pandemics (1-3). Excess deaths are meant to capture the composite sum of perturbations in disease incidence and other factors, including social, health care, lifestyle and natural catastrophes that may shape population fatalities in a given year. However, excess death calculations can lead to controversy with different teams of researchers generating markedly different estimates for the same country and year(s) (4-6). The reason is that the calculation of excess deaths requires making analytical choices for which there is no consensus. Specifically, one needs to select a reference baseline period (a time window in the past that will be used for extrapolating how many deaths would be expected in subsequent years) and a projected period (the time window for which an excess death estimate is made by comparing the observed versus expected number of deaths based on the past (reference) experience). Moreover, one should decide whether there are any time patterns and what is the form of these time patterns (e.g. whether overall mortality should be declining or increasing over time and, if so, in what form, e.g. linear or spline fit). Empirical work and simulations (4-10) have shown that these choices can make a substantial difference in the obtained excess death estimates.

When results depend on analytical choices, one methodological strategy is to explore the full range of results that can be obtained when a wide range of possible analytical choices and combinations thereof are considered (11-20). Analyses may range from a few dozen to several million different options (e.g. in selecting covariate sets in regressions) (15,17). Different terminology has been used for such approaches that generalize the concept of sensitivity analysis. Commonly used terms are “multiverse analysis” (11-14), “vibration of effects” (16-18) and “multi-analyst analysis” (19,20) (when multiple researchers are each asked to select independently their preferred analysis). Here, we propose a multiverse approach for excess deaths. Instead of making unavoidably arbitrary choices in selecting reference baseline and projected periods, we consider all possible reference baseline periods and projected periods in consecutive year time windows during a lengthy period of interest. Instead of prespecifying time patterns, this multiverse approach allows the data to demonstrate what might be the time patterns and how sensitive the results are to different analytical choices. All possible choices are considered for reference baseline periods. The multiverse approach also allows us to understand to what extent excess death estimates may shrink when longer projected periods are considered, in the range of 1-4 years. If perturbations lead to excess deaths increases due to the demise of individuals with limited life expectancy (21), then excess death peaks that are seen with short projected periods (e.g. 1 year) will diminish or even disappear when longer projected periods are considered. People who died at some point due to the perturbation would have died very soon anyhow. Conversely, if perturbations result in mortality peaks due to deaths of people who had long life expectancy, extending the projected period window will not have the same impact.

We applied this approach to 33 high-income countries that have the most reliable data for mortality according to age-stratified groups for the extended period 2009-2021. This allowed us to examine whether there are consistent time patterns in different countries, how sensitive the results are to analytical choices, and whether larger projected period windows, especially in the COVID-19 pandemic and the years preceding it, shrink substantially any observed mortality peaks.

RESULTS

Stability of relative ranking for the pandemic years’ excess deaths across 33 countries

The estimates of relative excess deaths (as percentage of expected deaths) can be used to compare different countries in a given time period. Despite large variabilities in the absolute estimates, the relative ranking of the 33 countries for a given period of interest was largely unperturbed, regardless of what reference baseline years were chosen. Table 2 shows the ranking of relative excess death estimates (as percentage of expected deaths) for the pandemic years 2020-2021 in all 66 analyses with different reference baseline windows. The USA had the highest estimates of relative excess deaths among all 33 countries in 49 of 66 analyses, the second highest in 16 analyses and the fourth highest in 1 analysis. Conversely, South Korea had the lowest estimates in 58 of 66 analyses, the second to lowest in 6, the third to lowest in 1, and the sixth to lowest in 1. Eastern European and Balkan countries closely followed the USA in the top excess death ranks consistently. Scandinavian countries, Australia, and New Zealand consistently were placed among the lowest excess death ranks next to South Korea. Other Western European countries typically occupied middle ranks. Tables S1A and S1B show that the distribution of country ranks for projected periods of 1 year and 3 years are similar to that shown in Figure 1 for 2020-2021; summing over more years does blur the ranking of middle-ranked countries.

• Download figure
• Open in new tab

Figure 1:

Variation with year from 2009 to 2021 of the excess death estimate expressed as a percentage of the expected deaths. The expected deaths are estimated from the average mortality values of each of the 66 different reference year-sets, which are all combinations of one or more consecutive years from 2009 to 2019. The average over all the reference years is shown at the black solid line, the maximum and minimum values are shown by the shaded area. The y-axis of every panel extends from -30% to 30%. The plots for different reference year sets are almost identical but shifted along the y-axis by different amounts. The sum of two years, which is particularly significant as the complete pandemic years are 2020 and 2021, is shown here.

View this table:

• View inline
• View popup
• Download powerpoint

Table 2:

Distribution of the country rank of the excess death estimates (from highest to lowest) in the pandemic 2-year period 2020-2021 expressed as a percentage of the expected deaths for the 33 countries as calculated for each of the 66 different reference baseline year sets. The countries are ordered by decreasing average rank (column 3); the standard deviation of the rank is also given in column 4. In most cases, the most common rank for a given year over the 66 reference years is on the diagonal. Exceptions are circled in green in the figure.

DISCUSSION

Our application of a multiverse approach to excess death data shows that consideration of different periods for reference baseline resulted in major variability in the absolute magnitude of the exact excess death estimates, but it did not affect substantially the relative ranking of different countries compared to others for specific years. Moreover, there have been distinct time patterns across different countries during 2009-2021. Countries differed markedly on whether they had a substantial decrease over time or not during 2009-2021, on whether they had a peak during the 2020-2021 pandemic years, and, if so, how high, and in the relative contribution of 2020 and of 2021 to this peak. With longer time windows for the projected period of interest (1 to 4 years), the range of excess deaths across different countries in the pandemic years and the 2 years preceding the pandemic shrank substantially and excess death estimates became less variable across countries. However, peaks did not disappear and for the USA in particular, excess deaths remained prominent even with long projected periods of interest.

In the multiverse literature from other fields, some analytical choices may be considered more meaningful or relevant than others. When researchers are asked to select independently what analysis mode they feel is most sensible, not all analytical choices are selected (19,20) and some types of choices may seem to make more sense. This may apply also for excess death calculations. E.g. it may seem not so appropriate to use a reference window of 2009 alone for projecting mortality in 2021. Indeed, one should definitely not use the excess deaths estimate of 2021 based on the 2009 reference in isolation as a reliable measure of excess deaths in 2021.However, previous studies on excess death calculations for the COVID-19 pandemic have used extremely different reference periods, ranging from a single year to over a decade (6). While in our own work (6) we used a 3-year reference period (2017-2019), it is difficult, if not impossible, to defend that shorter or longer reference periods would be less appropriate. The proposed multiverse approach removes the hurdle of this subjective choice and allows for consistency by considering all possible reference years and time windows. Thus, it allows one to get an objective comparative picture for the relative performance of each calendar year and an unbiased comparative picture for the presence and pattern of time trends.

The obvious heterogeneity of time patterns across different countries suggests that selection of specific time trends in modeling excess deaths may be a situation where one size does not fit all. Selection of specific anticipated time trend patterns may markedly affect the results in ways that are not verifiable for their appropriateness. E.g. selecting a model that anticipates a marked decrease in mortality over time makes it difficult for a country not to have excess deaths even if it does very well in a given year – but still falls short of an anticipated stellar improvement over time. There is no guarantee that mortality rates should continue declining, let alone markedly decline, over time with medical and other progress, even in the absence of major negative perturbation events such as pandemics or natural catastrophes (6). For advanced economies with aging populations, accumulating frailty and disease burden and restrictions or ceilings to progress and available resources, trends for decreased mortality may not be sustainable. The multiverse approach, when applied to multiple countries, allows a comparative assessment of the trajectory of different countries. This may be preferable and it may offer some genuine insights about which countries do well (short-term and long-term) and which do poorly – in comparison.

In this regard, some stark differences stand out for both long-term trends and for the pandemic years. The USA consistently performed very poorly with both stagnation in mortality during the pre-pandemic years and a sharp increase during the pandemic. Eastern European and Balkan countries showed sharp decreases during the pre-pandemic years and a sharp increase during the pandemic. Most western European countries had sharp decreasing trends with modest disruption during the pandemic. All Scandinavian countries, Australia, New Zealand, and South Korea have had largely unperturbed declining mortality patterns. The markedly different patterns may reflect a combination of social, health care, and pandemic factors. The USA has an ailing health system with approximately 30 million uninsured people (22), large inequalities (23), many people with poor access to care (24), and major ongoing non-infectious epidemics, including obesity (25), opioid abuse and overdose (26), and violent deaths (27). Eastern European and Balkan countries have limited resources for their healthcare systems and lower social welfare than other European countries (28) and some countries like Greece have long suffered from austerity (29). The best performers are excelling in social welfare and health system functionality and resources, even if there are differences across countries. Exceptions may occur within circumscribed populations and adverse settings even in countries with overall excellent trajectories. For example, the dysfunctional consequences of privatization in nursing homes in countries like Sweden or Canada (30,31) translated to peaks of excess deaths during circumscribed periods in the long-term care settings (32).

Consideration of longer projected period time windows diminished substantially the range of excess deaths in some countries, but not in others. Overall, when longer periods are considered, differences between most countries become less pronounced. However, larger windows had minimal effect in the USA, and this may reflect that the problems that lead to unfavorable mortality patterns in the USA reflect chronic dysfunctions that might have been accentuated by the pandemic but pre-existed and which affect also people with long life expectancy. Poverty, marginalization, homelessness, inequalities, drug overdoses, and violence affect indeed young and middle-aged populations. We have shown previously that the USA has had 40% of excess deaths contributed by the <65 age stratum, a higher percentage than all other highly developed countries (6). Conversely, in many other countries, large time windows for the projected period shrank substantially the excess death fluctuations. This suggests that in these countries excess deaths temporarily affect mostly people with relatively limited life expectancy (21).

Europe, while not a country, has historically aggregated excess death data in the EuroMOMO data base (https://www.euromomo.eu/) to include data from 21 countries in Europe plus Israel (33). If one were to aggregate data for the 19 of these 21 countries for which we have data (excluding Cyprus & Ireland), the fictional country composite that includes Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Israel, Italy, Luxembourg, Netherlands, Norway, Portugal, Slovenia, Sweden, Switzerland, and United Kingdom has a similar population to the USA (410 million versus 330 million) and the relative excess death of this European composite is only 2.46% for the pandemic years (2020+2021), which is in stark difference to the USA figures. It is also less than two year totals for 2009+2010 and 2010+2011, with values of 5.57% and 3.30% caused by elevated Influenza pandemics (34).

Some limitations should be acknowledged. First, there are some additional sources of analytical flexibility that can be considered in excess death calculations. These include the choice of age bins for age adjustment, and the use of additional adjustments for modeling the population profile over time. These would add additional variation with more multiverse options, but probably would not invalidate the major patterns that we observed. Second, we only modeled data from 33 countries that are the ones with the most reliable data. Extrapolations to other countries would be precarious, given the unreliability of the mortality information. Time patterns observed in the 33 countries may not necessarily apply to the remaining countries around the globe and local circumstances may make a difference. Third, we considered yearly interval increments so as to capture all 4 seasons in the unit of time, but in theory, the multiverse process can be applied for smaller units of time as well. Fourth, data on population and population structure in each country on a yearly basis are typically inferred from census data collected on more sparse timing, therefore they carry some uncertainty. Fifth, the pandemic impact and its consequences as well as the consequences of aggressive measures that were taken has continued more prominently in 2022 in some countries than others (35). It would be interesting to see whether differences across countries get further attenuated and/or some countries continue to stand out prominently when longer pandemic and post-pandemic periods (e.g. 2020-2022 and 2020-2023) are considered. Preliminarily results based on the first 8 months of 2022, it seems that several countries with death deficit in 2020-2021 (e.g., Australia, New Zealand and South Korea), had considerable excess deaths in 2022, while some others continued to have limited deaths (e.g. Sweden) and some hard-hit countries like USA and Greece continued to do very poorly (6,35).

A multiverse approach to excess death calculations may offer bird’s eye views on mortality patterns in comparative assessments of a large number of countries. These patterns may be more reliably informative than efforts to obtain isolated single-country estimates of excess deaths, which are subject to substantial uncertainty even in countries with the best-collected data. It may be best to avoid pre-specifying time patterns and to allow the data to show what time patterns may be emerging. Finally, observed time patterns may not necessarily continue into the future and multiverse analyses can be updated accordingly for additional years moving forward.

MATERIALS AND METHODS

Data availability

All data are in the manuscript, tables, and supplementary tables and in the publicly available databases listed in Supplementary Links to Data and deposited online at https://zenodo.org/record/7095753.

Supplementary Table and Figure Legends

Table S1: Distribution of the country rank of the excess death estimates in pandemic year 2021 expressed as a percentage of the expected deaths for the 33 countries as calculated for each of the 66 different reference baseline year sets. The countries are ordered by decreasing average rank (column 2); the standard deviation of the rank is also given in column 3. (A) Shows data for 2021 alone in Table S2A; (B) Shows summed data for 2019+2020+2021 in Table S2B; ; (C) Shows summed data for 2018+2019+2020+2021 in Table S2C. Such sums are labelled by their last year so all are marked as “2021”. In most cases, the most common rank for a given year over the 66 reference years is on the diagonal.

Table S2: Variation with year from 2009 to 2021 of the average excess death expressed as a percentage of the expected death for all 33 countries studied here. The expected death is estimated from the average mortality values of each of the 66 different reference year sets, which are all combinations of one or more consecutive years from 2009 to 2019. The average over all the reference years is shown in each table. The four tables are (A) for the sum of two adjacent years recorded at the later year (the value for 2021 is 2020+2021), (B) for individual years, (C) for the sum of three adjacent years recorded at the later year (the value for 2021 is 2019+2020+2021) and (D) for the sum of four adjacent years recorded at the later year (the value for 2021 is 2018+2019+2020+2021). The shading is calibrated by range of values in all tables considered together. Each table is sorted by descending using value in the year column “2021”.

Table S3: Comparing the percentage excess death in the pandemic years, 2020 and 2021, with that of other pairs of years between 2009 and 2019. In 9 of the 33 countries, the pandemic years, 2020+2021, marked as “2021” in red with yellow shading, were worse than any other pair of years.

• Download figure
• Open in new tab

Figure S1:

Distribution of the year rank of the of the excess death expressed as a percentage of the expected death as calculated for each of the 66 different reference baseline year sets for 33 countries and for 12 pairs of years (2009+2010 to 2020+2021). The years are ordered by decreasing average rank (in column 2); the standard deviation of the rank is given in column 3. If the ranking for all 66 calculations were the same, all counts would be on the diagonal. This is true for 16 of the 33 countries. In the other 19 countries, the largest deviations from identical ranking occur for AUT (Austria) where the three years ranked in position 2 to 3 are 2011, 2013, and 2021. The mean rank for 2013 of 4.24 is not much less than that of 2021 of 4.52. Clearly the ranking of years for a given country are largely independent of the reference years used to calculate the expected deaths.

• Download figure
• Open in new tab

• Download figure
• Open in new tab

• Download figure
• Open in new tab

Figure S2

Variation with year from 2009 to 2021 of the excess death expressed as a percentage of the expected deaths as shown in Figure 1. The plots are for different projected periods (A) for individual years, (B) for the sum of three adjacent years (recorded at the latest year so the value for 2021 is 2019+2020+2021), and (C) for the sum of four adjacent years (recorded at the latest year so the value for 2021 is 2018+2019+2020+2021).

Supplementary Links to Data