ABSTRACT
Background Past studies show a mixed relationship between the Global Health Security (GHS) Index and Covid-19 pandemic health outcomes. While some recent work suggests higher GHS Index scores are associated with better mortality outcomes there remains scope for additional analyses considering island nations and macroeconomic outcomes of the pandemic.
Methods Correlation and multiple linear regression analyses (controlling for per capita GDP) across GHS Index scores, age-standardised excess mortality for 2020–2021, and GDP per capita growth, for island and non-island jurisdictions separately.
Results The GHS Index moderately to strongly predicted better health outcomes in terms of age-standardised excess mortality through 2020–2021 in non-island jurisdictions (Pearson’s r = -0.53, p < 0.00001; Spearman’s rho = -0.61, p < 0.00001; β = -5.54 [95% CI = -3.68 to -7.40], p <0.00001), but not in island jurisdictions (r = -0.17, p = 0.25; rho = 0.00, p = 0.98; β = 0.65, p = 0.675). A 5-point rise in GHS Index was associated with a 26.75/100 000 population reduction in excess mortality. A higher GHS Index predicted higher year-on-year GDP growth through the pandemic for non-islands (2019–2020: β = 0.11, [95% CI = 0.04– 0.17, p = 0.00156]; 2020–2021: β = 0.09, [95% CI = 0.04–0.15, p = 0.00173]), but not for islands. For non-islands, a 5-point increase in GHS Index predicted increase in GDP per capita growth of 0.55% in 2019–2020, and 0.45% in 2020–2021.
Conclusion The GHS Index demonstrated clear associations with favourable health and macroeconomic outcomes of non-island jurisdictions through the Covid-19 pandemic, supporting its use to guide pandemic preparedness investments. Contrasting findings for islands suggest need to enhance how the Index measures border biosecurity capacities and capabilities, including the ability to support the exclusion/elimination strategies that successfully protected islands during the Covid-19 pandemic.
What is already known?The Global Health Security (GHS) Index has been shown to predict excess mortality through the Covid-19 pandemic when accounting for under-reporting and population age-structure.
However, it is debated whether the Index is equally applicable across different jurisdictions as a guide to pandemic preparedness.
What this study adds?When we analysed 49 island and 145 non-island jurisdictions separately, the GHS Index was much more strongly associated with good health outcomes in non-islands than islands. We also found higher Index scores were associated with better macroeconomic outcomes in non-islands.
How might this study affect research, practice, or policy?This study confirms the potential of the GHS Index as a starting point for pandemic readiness. Jurisdictions should look to enhance capacities and capabilities most associated with Covid-19 health and macroeconomic outcomes. The findings also suggests that border biosecurity, which island states have by virtue of their geography but other states need to generate by design, need much greater focus in the GHS.
INTRODUCTION
Pandemic threats
The Covid-19 pandemic caused enormous global health and economic impacts. It resulted in the first increase in global all-cause mortality in the 70-year post-war period (a 5.1% increase in the 2020-2021 period compared with a decline of 62.8% from 1950-2019).1 Excess mortality during the continuing pandemic period is estimated at 27.3 million deaths up to July 2024.2
There is a likelihood of equally or more severe pandemics in the future.3 4 The list of pathogens identified with pandemic potential is growing.5 Several are causing current concern, notably the influenza A(H5N1) panzootic now infecting cattle in the US. Its future pandemic potential is currently rated as ‘moderate risk’.6 There is also a more harmful Mpox clade currently spreading in Africa that has become the eighth public health emergency of international concern to be declared by WHO.7 These actual and potential harms highlight the benefits from implementing effective preparedness measures.
GHS Index and Covid-19 pandemic
The Global Health Security (GHS) Index is a comprehensive, criteria-based assessment of health security capabilities across 195 States Parties to the International Health Regulations. The Index was first published in 2019 by the Nuclear Threat Initiative (NTI), Johns Hopkins Center for Health Security, and the Economist Intelligence Unit.8 The metric encompasses six categories (that can be abbreviated as ‘Prevention’, ‘Detection and Reporting’, ‘Rapid Response’, ‘Health System’, ‘Compliance with International Norms’, and ‘Risk Environment’), which are composed of 37 indicators and various sub-indicators. It quantifies countries’ abilities or potential to carry out public health functions necessary for infectious disease outbreak prevention, detection and response, by giving an ‘Overall Score’ out of 100.
The GHS Index has face validity,9 and has been validated against communicable disease mortality globally with higher scores predicting a lower burden of communicable disease deaths by country.10 On this basis one would expect countries with higher GHS Index scores to have had better health outcomes through the Covid-19 pandemic.
However, studies of the relationship between Covid-19 outcomes and GHS Index scores early in the Covid-19 pandemic did not reveal the expected relationships. Paradoxically, high GHS Index scores (better prepared countries) looked to have worse Covid-19 outcomes in terms of Covid-19 deaths per capita and number of cases,11-14 as well as detection and response times,15 or the GHS Index had ‘no explanatory power’16 and pandemic preparedness indices generally were, ‘not meaningfully associated with standardised infection rates or infection fatality ratios.’17 Furthermore, Covid-19 outcomes were significantly associated with sociodemographic, political and governance variables not included in the 2019 GHS Index. Social cohesion, reduction in social polarisation and reduced perceptions of corruption were consistently correlated with lower excess mortality associated with the pandemic.18 Finally, the GHS Index did not sufficiently consider the importance of geography, with island nations such as Australia, New Zealand, and Pacific islands able to close their borders to prevent the virus from entering.19
On the other hand, some studies have shown the expected correlations between GHS Index scores and pandemic outcomes. For example, higher GHS Index scores were associated with low R0 values across 52 countries.20 African countries that fell in the ‘more prepared’ GHS Index category had fewer deaths and cases of Covid-19 than those in the ‘least prepared’ category.21 When deaths per capita later in the pandemic were analysed (July 2021), there was a strong negative correlation (r = -0.69, p < 0.01) between European countries’ GHS Index scores and their excess mortality rates, indicating that countries with higher GHS Index scores had lower excess mortality during the pandemic.22 Another study found a statistically significant slight negative correlation between GHS Index scores and excess mortality at the global scale.23
With respect to age, for each percentage point increase in the proportion of the population over age 65 years, excess mortality at 100 days in the Covid-19 pandemic increased by 3.35 deaths per 100 000 individuals (p=0.006).18 Indeed, when age-standardised excess mortality was analysed across all countries (n = 183) the expected correlation between higher GHS Index scores and fewer Covid-19 deaths emerged globally. Each 5-point increase in the GHS Index was associated with a 0.21 lower Comparative Mortality Ratio (CMR) for excess Covid-19 mortality, after adjusting for GDP per capita.24 This latter study highlighted several problems with previous analyses including with data quality and comparability, and the fact that different country population structures means that age-standardised mortality data is needed.
Islands and the Covid-19 pandemic
The inclusion of island nations likely skews many analyses of the relationship between GHS Index and Covid-19 pandemic outcomes. This is because island nations, especially small developing island nations in the Pacific, exhibit very low GHS Index scores.9 25 However, being an island country was associated with lower excess mortality during the pandemic.18 Islands singled out as performing particularly well include New Zealand, Australia, Singapore, Iceland, Taiwan, and the pseudo-island of South Korea.13 26 Also, the Oceania region, which is primarily composed of island nations, was noted to have the lowest excess mortality rate of any region.23
A likely explanation for the relative success of island jurisdictions is that they are surrounded by ocean, and many implemented effective border controls during the pandemic, potentially rendering irrelevant many internal factors that might contribute to GHS Index scores. Being an island also significantly increased the probability of a country pursuing a Covid-19 elimination strategy.16 Indeed ‘island status’ (reduced Covid-19 mortality) along with ‘aged population’ (increased mortality) were among the factors most correlated with excess mortality in one analysis.18
Additionally, countries with a higher score for ‘trade and travel restrictions’ (ie, countries who had not, pre-Covid-19, implemented restrictions due to infectious disease concerns, or had implemented restrictions only while considering International Health Regulations) had worse Covid-19 outcomes.18 This pattern hints that willingness to implement trade and/or travel restrictions might improve pandemic outcomes.
GDP and the Covid-19 pandemic
Potentially confounding any relationship between GHS Index scores and excess mortality is the fact that per capita gross domestic product (GDP) was a good predictor of excess mortality in 20 European countries (r = −0.71, p < 0.01).22 Additionally, GDP has been previously correlated with GHS Index scores,9 10 and excess mortality is broadly, though not consistently, associated with lower sociodemographic index (SDI) scores.1
A number of studies have found that countries using a strong proactive response to the Covid-19 pandemic or using an elimination strategy, resulted in more favourable GDP growth outcomes when compared to countries adopting a mitigation/suppression strategy.16 27-29 But none of these studies examined any relationship between pre-pandemic planning/GHS Index with Covid-19 pandemic macroeconomic outcomes.
Aims and hypotheses
All these findings suggest that a more definitive analysis of the relationship between GHS Index and Covid-19 outcomes should focus on excess mortality rather than number of deaths, age-standardise the data, explore outcomes other than just death, such as economic impacts, control for GDP per capita, and analyse islands and non-islands separately. We sought to explore the associations among these variables.
Hypothesis One was that the low average GHS Index scores of islands, and their relatively less severe Covid-19 outcomes, have confounded any relationship between GHS Index scores and Covid-19 outcomes.
Hypothesis Two was that higher GHS Index scores are associated with less reduction in GDP growth per capita during the Covid-19 pandemic (2020 and 2021).
METHODS
Study jurisdictions
The study population comprised all jurisdictions (n = 195) scored in the 2021 version of the GHS Index.9 We divided the study population into two sets, ‘island’ jurisdictions and ‘non-island’ jurisdictions. We defined island jurisdictions as including the following: individual sovereign islands, island archipelagos (eg, Indonesia, Philippines), island continents (ie, Australia), island jurisdictions that were non-sovereign but had GHS Index data eg, Cook Islands (New Zealand), and where the jurisdiction had a land border with another jurisdiction on the same island (eg, Ireland, UK, Timor-Leste, Papua New Guinea, Brunei, Dominican Republic, Haiti, Cyprus). Singapore was treated as an island. We excluded South Korea (although it can be considered a pseudo-island) and jurisdictions with mixed characteristics but where the capital city was on a continental land mass (eg, Malaysia, Corsica [France], Sardinia and Sicily [Italy]).
Data Sources
GHS Index
We obtained data on jurisdiction-level capacities and preparedness against biological threats from the latest (2021 version) of the GHS Index (as used by Ledesma et al).24 Data on six categories, and 37 indicators were extracted for analyses.9
Covid-19 outcomes
We obtained excess mortality and age-standardised excess mortality by year, for these jurisdictions, for 2020 and 2021 from the GBD Study Demographics Collaborators. The estimation methods for these data have previously been described in detail.1 Excess mortality is an important measure of the true mortality impact from the Covid-19 pandemic because it identifies the difference between observed all-cause mortality and mortality expected under normal conditions. This dataset demonstrates approximately 16 million global deaths due to Covid-19, with 20 nations having negative excess mortality in either 2020 or 2021.
GDP per capita and GDP growth
Due to the previous associations of GHS Index score with national GDP per capita and with SDI scores, we treated GDP as a covariate and obtained GDP data for these jurisdictions from the World Bank’s World Development Indicators, namely ‘GDP per capita (current US$)’,30 with annual data from 2015 up to and including 2021. We calculated GDP growth from 2019 to 2020, and from 2020 to 2021, as well as the geometric mean of GDP per capita for each 5-year period ending in 2019, 2020, 2021.
Statistical analysis
We ignored missing data and conducted both epidemiological and economic analyses. For the epidemiological analyses, we calculated the mean GHS Index overall score, and mean category scores for the six main categories for the groups of islands and non-islands separately. We analysed the relationship between GHS Index scores (down to the ‘indicator’ level) and both excess mortality (deaths per 1,000 population) and age-standardised excess mortality (deaths per 1,000 population), considering pre-pandemic GDP per capita (mean of five years to 2019) as a covariate.
For the economic outcome analyses, we analysed GHS Index scores versus national GDP growth. Specifically, we calculated the percentage change in GDP from 2019 to 2020 and from 2020 to 2021, as well as the geometric mean of GDP per capita across five years, from 2015 to 2019, and then the geometric mean across each five year period ending in 2020, 2021, and 2022.
We established both Pearson r and Spearman rho correlation coefficients to initially explore associations between GHS Index scores and Covid-19 related excess mortality. We then conducted multiple linear regressions. Because the GHS Index measures are highly correlated and to prevent unnecessary adjustment for variables that may potentially bias results, we followed Ledesma et al in using bivariate regressions to observe each relationship independent of the other indicators.24 We included GHS Index scores as the independent variable, excess mortality as the dependent variable and GDP per capita as an a priori covariate. In the economic outcome analysis, we treated the year-on-year change in 5-year geometric mean of GDP as the dependent variable. We calculated 95% confidence intervals and p values. A p value of < 0.05 was considered statistically significant, but we note the Bonferroni correction used by Ledesma et al and report important relationships significant at the p < 0.0009 level.
We conducted analyses for all jurisdictions in the study population combined, as well as for islands and non-islands separately. We repeated the same analyses using the overall GHS Index score as well as scores for the six categories of the GHS Index, and all 37 indicators up to the X.x level, ie, GHS Index scores for items 1.1, 1.2, etc.
In a sensitivity analysis we created two groups from the islands group. These were ‘jurisdictionally uncomplicated islands’ and ‘jurisdictionally complicated islands’. The latter being both those with land borders with other jurisdictions (eg, Dominican Republic/Haiti; Indonesia/Timor-Leste, Ireland/UK); plus those that are not fully independent eg, the various islands with constitutional links with New Zealand (Cook Islands, Niue).
RESULTS
Descriptive statistics across jurisdictional groupings (all jurisdictions, islands, non-islands) are presented in Table 1. Jurisdiction-specific age-standardised excess mortality for 2020– 2021 across all locations ranged from -59.6 to 897.4 per 100 000 population (n = 204). Islands demonstrated lower excess mortality (mean 58.79 vs 154.85). Mean GDP per capita growth was 0.7% from 2019 to 2020, and 3.1% from 2020 to 2021 across all jurisdictions (n = 197), with islands having a higher 5-year geometric mean growth rate than non-islands to the end of 2020 (1.03% vs 0.60%) and lower to the end of 2021 (2.51% vs 3.31%). The mean 2021 GHS Index score of non-islands was 38.9 (n = 145) and islands 35.0 (n = 49).
Descriptive values for jurisdiction groups included in the analysis
Correlations
Pearson’s r and Spearman’s rho exhibited substantially similar patterns. Table 2 (and an extended version Supplementary Table S1) displays the relationship between age-standardised excess mortality 2020–2021 and GHS Index scores. The overall GHS Index score is moderately to strongly negatively correlated with excess mortality in non-island jurisdictions (Pearson’s r = -0.53, p < 0.00001; Spearman’s rho = -0.61, p < 0.00001), but not in island jurisdictions (r = -0.17, p = 0.25, rho = 0.00, p = 0.98). In non-islands, all six GHS Index category scores were significantly negatively associated with excess mortality at the p < 0.0009 level, with r ranging from -0.30 to -0.57. Eleven (of 37) individual indicator scores had a statistically significant negative correlation where Pearson’s r indicated a stronger correlation than -0.40. Eight more individual indicator scores had a weaker negative correlation with excess mortality, that was nonetheless highly statistically significant, at the p < 0.0009 level (see Supplementary Material for full results tables).
Correlation of GHS Index overall score and subcategory scores with age-standardised excess mortality 2020–2021 and GDP per capita growth (%) 2020 to 2021 (Pearson’s r correlation by jurisdiction grouping: all, islands, non-islands; see Supplementary Material for full table including all 37 GHS Index indicators, and for GDP growth 2019 to 2020).
The relationship between GHS Index scores vs GDP per capita growth (%) is also shown in Table 2 (for 2020 to 2021) and Supplementary Table S2 (for 2019 to 2020). From 2019 to 2020, the GHS Index overall score (r = 0.26, p = 0.0022), and all category scores except ‘Rapid Response’ are positively associated with GDP per capita growth at the p < 0.05 level. In 2019-2020 ‘Early Detection’ (r = 0.32, p = 0.00009), had the strongest association. These associations were not found for islands.
From 2020 to 2021, the overall GHS Index score of non-islands shows a weak positive correlation with GDP per capita growth (r = 0.28, p = 0.00067). In this second year of the pandemic 17 individual indicators (of 37) had r > 0.20 and p < 0.05. Individual indicators with r > 0.30 and p < 0.0009, were health supply chain, and political and security risk. There were no statistically significant associations of GHS Index scores and GDP growth in islands.
Multivariable analyses
Table 3, and it’s extended version Supplementary Table S3, presents the results of the regression analysis examining the relationship between GHS Index scores and age-standardised excess mortality at the category and indicator level. After adjustment for mean GDP per capita 2015–2019, the GHS Index overall score was negatively associated with age-standardised excess mortality 2020-2021 for non-islands (β = -5.35, 95% CI = -3.59 to - 7.12). The results indicate that each 5-point increase in the GHS Index was associated with a 26.75 per 100 000 population reduction in excess mortality. All GHS Index categories remained associated with reduced 2020-2021 age-standardised excess mortality after adjustment for GDP. In total, 27 (of 37) different indicators were associated with reduced excess mortality at the p < 0.05 level. For islands, only the GHS Index category ‘Risk Environment’ was statistically significantly associated with reduced excess mortality. Figure 1 depicts the regression line and 95% CI for GHS Index overall score (2021) vs age-standardised excess mortality 2020–2021, for all jurisdictions, islands and non-islands.
Relationships between the 2021 Global Health Security Index overall score and age-standardised excess mortality 2020–2021 for all jurisdictions, non-islands, and islands. The black points represent jurisdictions while the coloured lines represent linear regression lines for the relationships with the shaded areas representing corresponding 95% CIs.
Relationship between GHS Index and Covid-19 pandemic health outcomes, as demonstrated by regression analysis of GHS Index overall, category, and 37 indicator-level scores vs age-standardised excess mortality per 100 000 population 2020–2021, controlling for mean GDP per capita 2015–2019.
Supplementary Table S4, shows the regression associations between GHS Index scores and GDP per capita growth, when controlling for mean GDP per capita 2015–2019. The GHS Index overall score was positively associated with GDP per capita growth 2019 to 2020 (β = 0.11, 95% CI = 0.04–0.17) and 2020 to 2021 (β = 0.09, 95% CI = 0.04–0.15) for non-islands, with no statistically significant association for islands. For non-islands, after adjustment, all GHS Index categories remained associated with increased GDP per capita growth across both periods, except for ‘Rapid Response’. The results indicate that each 5-point increase in the GHS Index predicts an increase in GDP per capita growth of 0.55% in 2019 to 2020, and 0.45% in 2020 to 2021 in non-islands. Figure 2 illustrates these relationships.
Relationships between the 2021 Global Health Security Index overall score and GDP per capita growth for all jurisdictions, islands, and non-islands across both 2019 to 2020 and 2020 to 2021. The black points represent jurisdictions while the coloured lines represent linear regression lines for the relationships with the shaded areas representing corresponding 95% CIs.
Sensitivity analysis
In a sensitivity analysis, ‘jurisdictionally uncomplicated’ islands and ‘jurisdictionally complicated’ islands (see Methods) were examined separately. There was no statistically significant association between GHS Index overall score and age-standardised excess mortality, or GDP per capita growth, for either grouping.
DISCUSSION
The Covid-19 pandemic has provided a strong empirical base for identifying the most important pandemic preparation and response capabilities. Our results support both of our initial hypotheses and these findings have important implications.
Firstly, after principled division of jurisdictions into ‘island’ and ‘non-island’ groups, we found the GHS Index moderately to strongly predicted age-standardised excess mortality in non-island jurisdictions, but found no such relationship for islands, with the exception of the ‘Risk Environment’ category (despite islands objectively better performance in terms of health outcomes). The inclusion of islands in previous analyses of the relationship between GHS Index scores and Covid-19 pandemic health outcomes has obscured important relationships. In addition to being moderately associated with improved health outcomes, the GHS Index is also weakly to moderately associated with improved economic performance through the first two years of the Covid-19 pandemic in non-island jurisdictions.
Contrary to much early-pandemic scepticism of composite preparedness indices, based on premature analysis of incomplete or inappropriate data, there appears to be real value in the GHS Index. This value had been anticipated by validation studies using pre-pandemic data on communicable diseases,31 and reiterated, though underestimated, by subsequent analysis of comparative mortality ratios.32
However, the absence of any meaningful relationship between GHS Index scores and island pandemic outcomes, along with the generally better pandemic outcomes of islands, also shows that future iterations of the Index will need to treat islands differently. The strong protective effect associated with being an island might not appear helpful as geography is not easily modified. However, this attribute can be seen as a proxy for the ability to implement strong border management which in turn supported an exclusion/elimination strategy for responding to Covid-19 pandemic. Consequently, we propose that the ability to implement an exclusion/elimination strategy in response to severe pandemics is added as a component of a future upgrade of the GHS Index design, along with an emphasis on border control components, while recognising the often smaller populations and low-resource settings of many island jurisdictions.
The global health security system operated by the World Health Organization (WHO) includes the International Health Regulations (IHR) which have recently been revised and there is a proposed global pandemic agreement.33 These regulations and agreements should also reflect the need for countries to manage their borders in the face of future severe pandemics and for the WHO to provide highly strategic global leadership in such situations.34
Secondly, the association between GHS Index scores and macroeconomic pandemic performance, as measured by GDP per capita growth, provides an economic rationale for health security through development of GHS Index capabilities and capacities. A 5-point increase in overall GHS Index scores was associated with a 26.75 per 100 000 population reduction in excess mortality and an average of 0.5% GDP growth across the two years analysed. For the median non-island jurisdiction analysed, that correlation equates to $US 24.67 per capita, per year of the pandemic, or nearly $25 million per year, for every 1 million population. Rough previous estimates have found that the $US 850 million investment in global health security by the United States through the Global Health Security Agenda, across 31 countries, was associated with GHS Index scores on average 6-points higher in recipient than matched non-recipient countries (ie, an average of $4.6 million invested per GHS Index point difference per country).31 Further economic analysis is seriously warranted, as it appears that developing GHS Index capabilities could be cost-effective on economic grounds alone, before potentially significant health benefits are calculated.
A number of other points warrant mention. In this and previous work, characteristics of politics and society show major correlations with pandemic outcomes. The Risk Environment category score had a stronger relationship to Covid-19 mortality than any of the others. This score includes government effectiveness, public confidence in governance, trust in medical and health advice, and related factors. This feature of the GHS Index has previously been noted, suggesting the GHS Index is a good measure of resilience, including preparedness and social capital.35 Trust in government and interpersonal trust, along with low levels of corruption have previously shown large statistically significant associations with lower standardised infection rates due to Covid-19, as well as higher vaccination coverage among middle-income and high-income countries.17 These factors are covered among the GHS Index indicators, and indeed exhibit some of the highest individual indicator to outcome correlations we obtained. Practically, this finding means jurisdictions should focus not just on the overall GHS Index score, but also assess how to improve areas tightly coupled to outcomes.
For non-islands, the GHS Index indicator-level scores most associated (β > -2.00) with fewer excess deaths (Table 3) included having: fewer public health vulnerabilities, socio-economic resilience, zoonotic disease capabilities, access to communications infrastructure, exercising response plans, adequate infrastructure, robust health supply chain, biosecurity systems, less political and security risk, health capacity in clinics, emergency response operation, and good immunization rates.
Of note, ‘trade and travel restrictions’ (β = 1.55 [95% CI = 0.61–2.48], p = 0.00131) and ‘JEE and PVS’ (β = 1.30 [95% CI = 0.09–2.50], p = 0.03490) were the only indicators positively associated with excess mortality (Table 3). Jurisdictions scored more highly on trade and travel restrictions if they did not implement any recent restrictions due to biothreats. It appears the indicator ‘rewards’ countries that do not have exclusion strategy related policies, yet Covid-19 data appears to show this worsens health outcomes. This is one area where the GHS Index assessment should be revised so that it aligns with actions that are likely to be protective against more severe pandemics, while also recognising the need to distinguish excessive border restrictions that may harm population health and economic functioning. ‘JEE and PVS’ means the jurisdiction has completed and published a Joint External Evaluation assessment and gap analysis or Performance of Veterinary Services assessment and gap analysis. It could be that the veterinary aspect of this is confounding human health measures. A closer examination is warranted.
With respect to GDP growth, for non-islands, we observed a number of expected relationships across time. For example, the ‘Prevention’ category exhibited a statistically significant association with GDP growth in the first year, but this lost statistical significance in the second year. Also, the ‘Detection and Reporting’ category had a stronger association with GDP growth in the first year of the pandemic than the second. In year one, indicators with the strongest correlations with good GDP outcomes were laboratory systems, health supply chains, and epidemiology workforce. In year two, the category ‘Health System’ strengthened its association with GDP growth.
The harms of the Covid-19 pandemic are ongoing in 2024 and it seems likely to become an endemic disease.36 But there is a serious risk of new biological threats. Pathogens such as Influenza A(H5N1) and Mpox have pandemic potential, there is the risk of novel coronaviruses, and the possibility of bioengineered organisms. Modelling suggests we can expect a ‘Covid-19 magnitude’ pandemic to have a return period of approximately 33–50 years and an 18–26% chance over the next decade,37 and recent forecasting work suggested a 1–4% chance of a biological catastrophe killing 10% of the human population by the year 2100.4
The world needs to continue to build resilience and mitigation capacities against biological threats. Ongoing work is needed in key areas such as metagenomic surveillance and diagnosis (eg, on aircraft wastewater), improved indoor air quality, and accelerated vaccine development and capacity for rapid mass production of new vaccines. Analysis of the GHS Index provides evidence pointing to the capabilities and capacities associated with good health and economic outcomes. Analysis of the Covid-19 pandemic has provided a wealth of information and now this information needs to inform future iterations of preparedness benchmarks.
The GHS Index needs to be reviewed and revised so that it is measuring the capabilities that are most strongly associated with effective pandemic preparedness and management. Doing that will make it a more valid tool for motivating donor funders and governments to invest in the most important areas and put fewer resources into those that appear less associated with good health and economic outcomes.
Study strengths and limitations
Key strengths of this study include use of the 2021 GHS Index, which better accounts for some capacities developed during the pandemic (some previous analyses used the 2019 version). Our study added to previous findings by demonstrating the significant explanatory benefit of distinguishing islands, as well as including macroeconomic, not merely health, impacts.
We used age-standardised excess mortality data (across a long 2-year window) avoiding the issue of undercounting Covid-19 deaths. We note that excess deaths fall into four categories that include: strictly non-Covid-19 deaths (eg, from other external events such as wars or natural disasters); indirect Covid-19 deaths (eg, deaths occurring from health system disruptions); direct Covid-19 deaths that were not reported; and direct Covid-19 deaths that were reported.3
We had to exclude some potentially relevant island jurisdictions from the analyses because they don’t have GHS Index scores, including Bermuda, Greenland, Taiwan, Tokelau, American Samoa, Guam, Northern Mariana Islands, Puerto Rico, and the Virgin Islands. Also, some islands have GHS Index data but lack World Bank GDP data eg, Cook Islands, and Niue. We further note that the definition of islands is complex. We included nations on two islands such as the UK; or part islands eg, Papua New Guinea; Australia is a Continental island; South Korea was categorised as not an island, but functionally resembles one.
GDP has some limitations as a metric. Changes in calculated annual growth rates are affected by the economic performance of the previous year. For example, a high growth rate does not necessarily reflect the strength of the economy over the past year, it might reflect poor economic performance of the previous year. For reasons such as this we used the pre-pandemic 5-year geometric mean of GDP when calculating growth rates. We also ignored potential covariates other than GDP. For example, body mass index (BMI) is a possible confounding variable, albeit accounting for only 1.1% (0·2–2·6) of variation in Covid-19 excess mortality.17
Ultimately, this was a correlation study and we can’t infer causation. However, there are clear and plausible links between the GHS Index indicators and both health and economic outcomes. The Index has previously demonstrated face and external validity, and the Covid-19 pandemic has provided prospective validation. Our findings are sufficiently strong that they support use of the GHS Index (with appropriate modifications for assessing border biosecurity capability) for guiding priorities for investment in future pandemic preparedness.
CONCLUSION
The GHS Index predicted better health outcomes in terms of age-standardised excess mortality through the first two years (2020–2021) of the Covid-19 pandemic. It also shows association with better macroeconomic outcomes as measured by GDP per capita growth. However, these findings apply to non-island jurisdictions only, with the GHS Index of island jurisdictions showing no similar correlations. Analysis of data from the Covid-19 pandemic revealed which GHS Index indicators are more and less strongly associated with pandemic outcomes, and should inform future iterations of preparedness indices.
The analysis also highlights a striking difference in the health security characteristics of islands and non-islands. This finding suggests that border biosecurity, which island states have by virtue of their geography but other states need to generate by design, need much greater focus in the GHS Index. These border control capacities and capabilities need to be able to support exclusion/elimination strategies, and anticipation of rapid vaccine roll-out.
The association with economic outcomes provides some evidence that health security investments might be justified on economic grounds alone and more research on this point would be valuable, along with prioritisation of the capacities and capabilities that are more strongly correlated with health and economic indicators.
Competing interests
The authors declare no conflicts of interest.
Funding
This study was unfunded.
Patient or public involvement
It was not appropriate or possible to involve patients or the public in the design, or conduct, or reporting, or dissemination plans of our research
Data availability
Data (csv): https://drive.google.com/file/d/1_dSfZoAJdBLG56wcVfyNVMdWlO_CEjVM/view?usp=drive_link
R Code: https://drive.google.com/file/d/1cZg3fWxyptwiV2taqdTCFVWH7czvyOqX/view?usp=drive_link
Acknowledgements
The authors thank Prof Austin Schumacher from the GBD Collaboration for sharing GBD data.
REFERENCES
