Determinants of COVID-19 incidence and mortality: A cross-country analysis
==========================================================================

* Gursimer Jeet
* Atul Sharma
* Aarti Goyal
* Shankar Prinja
* Arvind Pandey
* Sandip Mandal
* Mendu Vishnuvardhana Rao
* Sumit Aggarwal
* Narendra K Arora

## Abstract

**Objective** We undertook this study to explore the role of important determinants affecting global COVID-19 incidence and mortality taking multifactorial disease dynamics into consideration.

**Design** Secondary data as on March 28, 2020 were obtained for 97 countries. Association of COVID-19 cumulative incidence and mortality measures were assessed with ten indictors representing health system characteristics, climate, demography, promptness of international travel restriction and population movement using Generalized Linear Modelling.

**Main outcome measures** Country-specific COVID-19 cumulative incidence, cumulative cause-specific mortality and case fatality rate.

**Results** Significant inter-country variation in incidence and mortality rates were observed. Five variables were found to be associated with cumulative incidence: testing rate per 1000 population (β = 0.119, p < 0.01), UHC index (β = 0.043, p = 0.04), percentage elderly population (β = 0.122, p < 0.01), percentage below-poverty line population (β = -0.048, p < 0.01) and disability adjusted life years due to NCDs (β = -0.013, p < 0.01). Case fatality rate was observed to be associated with testing rate per 1000 population (β = -0.058, p = 0.03) and population density (β = 0.002, p = 0.02), while the cumulative cause-specific mortality was associated with only percentage elderly population (β = 0.096, p = 0.04) in the country.

**Conclusions** Health system response, population susceptibility and demography were the most important factors determining the progression. Policy response should focus towards increasing testing, primarily targeting high population density areas. Health system strengthening and reduction in population risk factors should be long term goals for a better response to such epidemics.

*Strengths and limitations of this study*

*   We undertook a large scale cross-country analysis to explore the impact of background factors in a country on COVID-19 disease dynamics.

*   Information from 97 countries as on March 28, 2020 were retrieved from public repositories and analysed.

*   Generalized linear models were employed to understand percentage change in county-specific cumulative incidence, cumulative cause-specific mortality and case fatality rate with selected explanatory variables representing health system characteristics, climate, demography, policy promptness and population movement.

*   Our main limitation was lack of data availability in the public domain, especially on the number of COVID-19 tests conducted by many countries, which forced us to restrict our sample size to 97 countries.

*   We could not include a quality parameter for community based mitigation strategies, hence the effectiveness of these in reducing COVID-19 incidence and mortality could not be assessed.

## Introduction

Novel Coronavirus disease (COVID-19) has emerged as one of the fastest growing epidemics of the 21st century. Despite having a lower case fatality rate than previous epidemics, its fast global spread, coupled with lack of effective treatment options forced World Health Organization (WHO) to declare it as a public health emergency of international concern on January 30, 2020, in less than a month of its first notification.[1, 2] The impact on lives across the 214 affected countries is so extensive that it is already being compared to mortality expenses in world war II.[3] As of 06 May 2020, more than three million cases (3,517,345) of COVID-19 have been reported, including 243,401 deaths across the world.[4] Countries such as United States of America (61906 deaths), Italy (29079 deaths), United Kingdom (28734 deaths) and Spain (25428 deaths) have been reeling from the shock, as the extensive spread of virus has exposed the health system preparedness in these countries. Most of the countries have also been suffering loss of life and economy, due to affected trade, transport, tourism and productivity in many other sectors.[5] Economists warn a major global recession and increase in unemployment rates as opposed to limited labor market damage initially suspected.[6] It is estimated that the global cost of the outbreak may be USD 4.1 trillion with recession as severe as global financial crisis experienced between 2007-2009.[7] Measures towards mitigation and control of COVID-19 have already forced many governments to pledge a substantial proportion of their Gross Domestic Product (GDP) to it, with Germany deciding to spend as high as 28.5%.[8]

As many of the disease characteristics of this novel outbreak are unknown, related scientific research has attracted an intense focus. The scale and pattern of the epidemic, basic reproduction number, case fatality rates and pre-symptomatic transmission rates are a few basic factors being studied and reported from different settings.[9-13] Another major set of research work included attempts to forecast the progression of the disease and ascertain the likely impact of proposed public health interventions using mathematical models and scenario analysis.[14-16] However, an equally important approach towards successful control of the disease lies in understanding the determinants defining the behavioral dynamics of this novel disease transmission. Excluding a few studies that have tried to relate the virus spread to changes in temperature, humidity or BCG vaccination induced innate immunity, most of the published research has left the relationships of the transmission with the background factors unexplored.[17-21].

Previous studies in relation to similar infectious conditions have identified numerous other social determinants affecting the outcomes, in addition to the pathogenic behavior of the microbe and environmental factors. The health system preparedness in general, and capacity of health system to respond in times of catastrophes in particular, can majorly influence the extent of spread as well as severity of the disease in the community. The proactive approach of the policy, evidenced with swiftness in decision making, in relation to imposition of travel restrictions, social distancing, timely isolation of infected populations, ban on mass gatherings and closure of workplaces & educational institutes, may play an important role in controlling the outbreak. The importance of both these factors was amply evident, when countries that had suffered the wrath of SARS epidemic demonstrated better ability to contain the spread of COVID-19 in its early stages.[5] Demographic characteristics may further influence the spread of infectious conditions, as susceptible populations such as elderly, patients with chronic non-communicable diseases (NCDs), patients with multiple co-morbidities and people living with HIV-AIDS may provide a more vulnerable environment. Finally, population density, ease of transportation and geographical proximity to the origin of outbreak may increase the likelihood of spread throughout, and hence need to be further explored.

Keeping this in mind, the current study was planned with an objective to explore the role of important socio demographic, economic, health system and environmental factors affecting the spread of COVID-19 epidemic globally. Concrete answers on these parameters might increase our understanding of the factors facilitating the spread of current epidemic and help in drafting solutions liable to be transferable as policy measures.

## Methods

### Model overview

We developed three statistical models employing Generalized Linear Modelling to quantify the e□ect of various determinants on the measures of incidence and mortality of COVID-19. The outcomes identified to represent the impact of the epidemic were country-specific cumulative incidence, case fatality rate and cumulative cause-specific mortality. Cumulative incidence was defined as the number of COVID-19 positive confirmed cases since 1st December 2019 till study end date, per million population, case-fatality rate as percentage of deaths reported among COVID-19 positive confirmed cases in the aforementioned period and cumulative cause-specific mortality as number of reported deaths due to COVID-19 in the same period, per million population in a country.

A number of probable determinants influencing these outcomes for communicable diseases were identified from previously published literature through a non-systematic review. These were divided into five broad domains: Health system characteristics, climate, demography, promptness of international travel restriction and population movement. A total of ten indicators were identified to represent these five domains. Universal health coverage (UHC) service coverage index provided by World Bank for each country was selected as an indicator of general health system preparedness. It is an index for essential health services (based on tracer interventions that include reproductive, maternal, newborn and child health, infectious diseases, NCDs and service capacity and access) presented on a scale of 0 to 100. Number of tests conducted in the country per 1000 population was utilized as an indicator of health system’s response to the epidemic. Details of these indicators and rationale for their use has been provided in Table 1.

View this table:
[Table 1:](http://medrxiv.org/content/early/2020/06/12/2020.06.11.20128561/T1)

Table 1: 
Indicators, their data sources, derivation and rational for use.

As different countries used different time frames to impose travel restrictions, we used the delay in decision making as an indicator to assess a country’s promptness in policy related matters. Number of days elapsed between confirmation of 1st COVID-19 case and date of imposition of international travel restrictions to China or other countries was computed and used as an explanatory variable in the models to explain the outcomes.

Two variables were selected to represent the country demography characteristics considered important in facilitating the disease transmission and severity. These were percentage population in the country living below international poverty line (BPL) and population density in the country per square km of surface area. Two indicators, percentage population above 65 years of age and disability-adjusted life years (DALYs) due to NCDs were included to represent burden of susceptible population in a country. DALYs was selected as it is a comprehensive summary measure to identify burden of a morbidity, that includes both years of life lost and years of life lived with disability.

Air transport passenger movement, which reflected the number of air passengers carried on both domestic and international aircraft carriers registered in the country, was selected as a proxy indicator for connectivity and population movement facilitating spread of the pathogen. Finally, in order to represent the general climate of the county around the time of 1st case identification, two indicators, average temperature in °Celsius and average percentage relative humidity, were utilized.

### Data collection

Data on all indicators was obtained from secondary sources in public domain. As the pandemic was (and is) still progressing, March 28, 2020 was considered as the end-point for data collection. Country-wise information on cumulative number of COVID-19 confirmed positive cases and associated deaths was collected from WHO Situation Report-68 for March 28, 2020.[22]

World Development Indicators (WDI) was used as the primary source of information for four explanatory variables: UHC service coverage index, percentage population above 65 years of age, percentage population living below international poverty line and population density per square km of surface area. WDI is a World Bank collection of development indicators, compiled from officially recognized international sources and considered providing accurate national, regional and global estimates.[23] UHC index values for different countries at the WDI online data base were available for years 2015 and 2017, from which the most recent available values for each country were selected. Data on below poverty line (BPL) population was obtained in terms of poverty headcount ratio at $1.90 a day using 2011 Purchasing Power Parity (PPP) rates.

Country-wise information on date of first COVID-19 positive case confirmation was derived from GitHub public repository.[24] Around 10% of this data was cross-verified with countries’ official health system/ COVID-19 response web portals to confirm accuracy. Data on Disability Adjusted Life Years (DALYs) due to non-communicable diseases (NCDs) per 100,000 population was derived from Global Burden of Disease study database.[25] The weather information for each country’s capital around the time of epidemic initiation was obtained from weather stations located at cities’ international airports. An established web portal called ‘Weather Underground’ was used for this purpose.[26] Mean temperature and humidity values were generated from hourly temperature and relative humidity values being reported by the stations on the day of first case identification in the country. For the largest 20 countries in the world, the information was collated from four selected cities at different latitudes to make the values representative at the country level.

Data on country-wise air transport passenger movement was obtained from International Air Transport Association’s World Air Transport Statistics Report for the year 2019.[27] Details of timeline, extent of travel restrictions imposed and number of Coronavirus tests conducted in various countries till March 28, 2020 were triangulated from multiple online sources. These included official federal Governments’ COVID-19 response information portals, press releases and newspaper reports. Some data was retrieved from *Worldometer* (A reference website providing real time information on number of tests conducted for COVID-19), which was cross-validated with information from other sources.[28]

Total number of countries (including territories and overseas dependencies) with at least one confirmed COVID-19 positive case till Mar 28, 2020 were 202.[22] Out of these, information on number of Coronavirus tests conducted was not available online for 95 countries. This included China, which provided information on number of tests conducted for a single province only. Additionally, data on UHC service coverage index, date of travel restrictions imposition or air passenger movement could not be located for ten other countries.

All data retrieval was conducted systematically by two researchers (AS and GJ), while a third researcher (SP) adjudicated any differences in interpretation between the two primary reviewers.

### Data analysis

The final dataset utilized for analysis included information from a total of 97 countries. Generalized Linear Models (GLM) were employed to assess the association of COVID-19 measures of incidence and mortality to the selected explanatory variables at country level.

This approach uses maximum likelihood estimation (MLEs) for parameters, and relaxes the assumptions of normality of residuals providing consistent estimates for further use. The relationship in the generalized linear model is assumed to be ![Formula][1]</img>  where *X**i* is the value of the i th predictor, *e* is the error, and *g*() is a function. Formally, the inverse function of *g*(), say *f*(), is called the link function; so that ![Formula][2]</img>  where *µ**y* stands for the expected value of *y*. Various link functions can be chosen, depending on the assumed distribution of the *y* variable e.g. identity, gamma, inverse normal, and Poisson distributions.[29, 30] In the current study, we employed the identity link function.

All three outcome measures were log-transformed to normalize their distribution before modelling. Omnibus test and deviance per degree of freedom were used to assess models’ goodness of fit. The deviance value per degree of freedom for goodness of fit ranged between 1.1 to 2.5, with significant omnibus tests in all three cases. The regression approaches employed were coupled with bootstrapping of data to estimate the 95% Confidence Interval for the estimates. All data were collected using MS Excel and analyzed using IBM SPSS ver. 22.

### Patient and public involvement

This research was done without any patient or public involvement. Any members of public were not invited to comment on the study design or consulted to develop outcomes or interpret the results. Members of the public were not invited to contribute to the writing or editing of this document for readability or accuracy.

## Results

A significant variation was observed between countries in terms of measures of incidence and mortality due to COVID-19. The number of confirmed cases ranged from 1 in some countries to as high as 86,498 in Italy, while the number of deaths attributable to COVID ranged from 0 to 9136 (Figure 1). The confirmed cases per million population were lowest in Angola (0.06 per million) and highest in Iceland (2625 per million), with a mean of 204 cases and a median of 61 cases per million population across the sample (Supplementary material S1). On the other hand, the case fatality rate ranged from 0% to 20%, with a mean case fatality of 2.1% and a median case fatality of 0.88% across the studied 97 countries till March 28, 2020 (Figure 1).

![Figure 1:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2020/06/12/2020.06.11.20128561/F1.medium.gif)

[Figure 1:](http://medrxiv.org/content/early/2020/06/12/2020.06.11.20128561/F1)

Figure 1: Summary measures of incidence and mortality due to COVID-19 across various countries till March 28, 2020*

Figure 2 presents the information on explanatory variables for the selected sample of countries. UHC service coverage index among these countries varied from 37 to 89. While the number of tests for COVID-19 that different countries conducted ranged from 0 to 31 (per thousand), the mean was only 2.50 per thousand population. Twenty-seven countries in the list closed their borders proactively before the first case was identified (Supplementary Material S2). Twelve countries had more than 20% of their population above 65 years of age. African and South-east Asian countries had highest proportion of BPL population, population density as well as DALYs due to NCDs (Supplementary Material S2). The mean temperature on the day of first case identification ranged from -4.56 °C (Norway) to 30.82 °C (Nigeria), whereas the humidity in the selected countries ranged from 37% to 96%. A significant variation was observed with respect to domestic and international air passenger movement in different countries, with a mean of 36 million and a median of 7.4 million travellers.

![Figure 2:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2020/06/12/2020.06.11.20128561/F2.medium.gif)

[Figure 2:](http://medrxiv.org/content/early/2020/06/12/2020.06.11.20128561/F2)

Figure 2: Summary country characteristics and policy response measures to COVID-19 across various countries till March 28, 2020*

Table 2 presents results of the three models analysing the associations among different variables of interest and outcome measures. Model 1 explored the relationships with cumulative incidence of COVID-19 in the population. In this model, five variables were found to be significantly associated with the outcome: Number of tests conducted per thousand population (β = 0.119, p < 0.01), UHC index (β = 0.043, p = 0.04), percentage population aged more than 65 (β = 0.122, p < 0.01), percentage population below poverty line (β = -0.048, p < 0.01) and DALYs due to NCDs (β = -0.013, p < 0.01). The results implied an increase of 12.2% in cumulative incidence with every 1 percent increase in elderly population aged above 65, 11.9% increase with per unit increase in number of tests conducted per thousand population and a 4% increase with UHC service coverage index. A decline in incidence is observed with increase in BPL population and DALYs due to NCDs (4.8% and 1.3% respectively).

View this table:
[Table 2:](http://medrxiv.org/content/early/2020/06/12/2020.06.11.20128561/T2)

Table 2: 
Factors associated with cumulative incidence, case-fatality and cause-specific mortality of COVID-19

Model 2 assessed the association between the predictor variables and the case fatality rate. Statistical significance was observed for number of tests conducted per thousand population (β = -0.058, p = 0.03) and population density per sq. km. (β = 0.002, p = 0.02) (Table 2). Hence, while case fatality rate increased by 0.2% with per unit increase in population density, it declined by 5.8% with every unit increase in number of tests conducted.

Model 3 showed only percentage population aged more than 65 (β = 0.096, p = 0.04) in the country to be associated with its cumulative cause-specific mortality. This implied a 9.6% increase in mortality due to COVID with every 1% increase in population above 65 years of age in the community.

## Discussion

The results of our study indicate the response of the health system, in terms of number of tests conducted, and the susceptibility of the population (higher elderly, BPL and DALYs due to NCDs) to be the most important factors determining the incidence of COVID-19 in the community. Susceptibility of the population also played an important role in determining cause-specific mortality. While general health system preparedness and population demography were also found to have a role in both incidence and case-fatality rates, we could not ascertain any association of the three outcome measures with general climate of the area, connectivity of the population or promptness of international travel restriction.

Health system performance and its capacity to deliver services has frequently been cited as an important determinant of morbidity rates in the community.[31] In line with the same, capacity of the system to bear the patient burden have been observed to be closely related to case-fatality rate for COVID-19.[32] Countries with a more evolved health system, in response to previous epidemics of SARS and MERS, have also been reported by some to counter the current epidemic better.[33] Increased testing for COVID-19 has been termed an extremely important control strategy, as it substantially reduces the risk of undetected positive cases in the community. It may also result in a higher cumulative incidence, as observed in our results, since countries with low testing rates may not be identifying cases in the community with mild or moderate symptoms leading to falsely low case counts.[34] In the same vein, the incremental effect of testing is higher towards detection of milder asymptomatic cases that do not result in mortality. This inflates the denominator for case-fatality rates, leading to a lower value. We also found negative association of testing with case fatality.[35] As a result, the case-fatality rate for COVID-19 appears to be a misleading indicator since it is significantly being affected by the testing rate. Hence, inter-country comparision of case fatality should be standardized for coverage of testing.

Previous research has established that COVID-19 affects individuals in older age groups more, as the mortality rates are disproportionately higher in this age group.[8, 36] Similar results were observed in our study, wherein we found positive associations between population above 65 years of age with both incidence as well as COVID-specific mortality. Research indicates that age-dependent decline in immune function, with defects in T-cell and B-cell function and excess production of type-2 cytokines, leads to deficient control of viral replication and more prolonged pro-inflammatory responses, potentially leading to poor outcomes in elderly population.[37] This may be coupled with psychological and physical barriers for this age group, such as more social isolation and poorer accessibility & affordability to health services, further aggravating the problem.[38]

We observed a negative association of COVID-19 cumulative incidence with below poverty line population in a country. There may be many plausible explanations for this, most of which had been adjusted for in our models. The poorer countries have, relatively, poorer performing health systems, lower testing rates and lower international passenger connectivity. Also, most of the developing countries have skewed population demography, with much higher proportion of younger population, unlike developed countries with higher proportion of elderly populations.[39] Our results indicated that higher population density was also significantly associated with higher case-fatality rates. This relationship has been validated earlier in context of MERS and SARS outbreaks, as well as in current COVID pandemic, where spatial associations have been well established as a key to spread.[40, 41] Because COVID-19 is primarily transmitted through respiratory droplets, and people in geographically compact neighborhoods areas tend to have more close interactions with each other, this might play a role in accelerating transmission of the virus.[34, 42]

Our study could not ascertain a significant association between incidence and promptness of decision making. This is in contrast with previous understanding which states that timing of community mitigation strategies help to decrease rates of infections as well as poor outcomes.[34] The reason for our results could be the inclusion of a single objective parameter (implementation of travel restrictions) to explain the models, not accounting for the extent and quality of other mitigation strategies put in place by various Governments. Our study results also could not establish any relationships between our measures of incidence and mortality, and the general climate variables. Previously published studies present contrasting evidence with respect to the role of temperature and humidity in COVID-19 transmission dynamics.[18, 43] While some of these state that there is decreased viability of the virus in higher temperatures, there is no evidence supporting that the case count will decline with increase in mean temperatures.[44] Again, we did not observe any association between the cumulative incidence and air passenger connectivity. This is in agreement with results of a previous study which stated that strategies to reduce global passenger volume will have negligible impact on the risk of importation of COVID-19 in different countries.[45]

Over the last few months, researchers have proposed numerous concepts to enhance understanding of origin, spread and control strategies of COVID-19. Several websites tracking virus spread and outcomes have been giving live updates on disease progression. Coupled with other readily available datasets on environmental conditions, population demographics, health system preparedness, several theories have been put forward and tested. However, ours is one of the first cross country analysis to take into consideration the impact of background factors in a country on COVID-19 disease dynamics. We examined a broad range of determinants from five major domains to understand their impact on incidence as well as fatality rates to find answers to key questions for better evidence-informed policymaking.

The study has a few limitations related to data availability and inferences which one can draw on causality. We excluded many countries, including China, from the analysis due to lack of authentic data related to number of COVID-19 tests conducted. This reduced the dataset from a total of 202 countries to 97 countries. Also, different countries employ different testing and reporting strategies which may bias the results. We could not include a quality parameter for community based mitigation strategies, such as cancellation of mass gatherings and school/ workplace closures, because, unlike travel ban which is easier to impose and examine, assessing the quality and effectiveness of other mitigation and control strategies is not entirely possible. Information related to mean temperature, mean relative humidity and impositions of travel restrictions for different countries was retrieved from different sources, however, the authors tried to alleviate the potential problems due to this by extensive triangulation and cross-verification of the data. Finally, the study is limited by retrospective nature of design for a continuously progressing epidemic. As more data on patterns and trends emerge, the role of health system, demographic and environmental variables may evolve to support or contradict the current findings.

## Conclusions and recommendations

While this analysis was being undertaken, control and mitigation strategies were in place in most parts of the world. Our analysis thus focused on studying effects of potential background determinants across the world, which may help in shaping the policy response towards this public health emergency. Keeping in mind the fact that it is not possible to plan quick interventions against non-modifiable risk factors, such as susceptibility of the population towards contracting COVID-19, the results of our study indicate that governments should focus the policy response towards increasing testing, with emphasis on areas with high population density. Identified positive active cases should be quarantined under strict containment measures to avert the effects of transmission to susceptible population with high risk factors due to NCDs. Slum areas with higher population density and BPL population, as well as old age homes have a higher risk of incidence and mortality, hence should be monitored diligently. Age-specific social distancing strategy should be actively pursued along with geographic social distancing. In addition to community mitigation strategies, health system strengthening, to improve UHC service coverage index, for a better response to the epidemic should be undertaken.[46, 47] This may be accomplished in short term by investing in material and human resources for health: increasing the numbers of trained work force as well as capacity building of existing staff.[48] Many countries have an existing Community Health Worker workforce operating in underserved areas, whose involvement in prevention and control activities may be explored.[49]

More research should be encouraged to understand the spatial associations of COVID-19 in various settings and explore other key determinants affecting the spread of virus. Since test conduction rates proved to be an important determinant of both identification of cases (or incidence) and mortality due to this disease, the inter-state or inter-national comparisons of positive cases or deaths should be adjusted for the level of testing in the corresponding regions. Similarly, model based analysis for forecasting the outcomes should be calibrated for the extent of under-reporting due to insufficient number of tests conducted for COVID-19 case detection. The case-fatality rate should also be standardized as per the coverage of testing prior to comparison among different areas. All these changes will significantly improve the ability of the current models to successfully predict the future scenarios and recommend potential solutions to the policymakers. Though global community is working towards development of vaccines, enhanced therapeutics and diagnostic methods for this new disease, the policy response should also keep in consideration the associations established in this study to improve the global as well as local response.

## Data Availability

All data collected and utilized for study analysis has been provided as supplementary material 1 and 2.

## Footnotes

### Contributor and guarantor information

GJ, AS, and SP were responsible for the initial plan, study design, data interpretation and manuscript drafting. GJ, AS and AG were responsible for data collection and statistical analysis. SP, SM, MVR, AP, SA and NKA were responsible for interpretation of the data and critical revision of the paper. SP is the guarantor and had full access to all of the data, including statistical reports and tables, and take full responsibility for the integrity of the data and the accuracy of the data analysis. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

### Copyright

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats and media (whether known now or created in the future), to i) publish, reproduce, distribute, display and store the Contribution, ii) translate the Contribution into other languages, create adaptations, reprints, include within collections and create summaries, extracts and/or, abstracts of the Contribution, iii) create any other derivative work(s) based on the Contribution,

iv) to exploit all subsidiary rights in the Contribution, v) the inclusion of electronic links from the Contribution to third party material where-ever it may be located; and, vi) licence any third party to do any or all of the above.

### Competing interests

All authors have completed the ICMJE uniform disclosure form at [www.icmje.org/coi\_disclosure.pdf](http://www.icmje.org/coi_disclosure.pdf) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

### Ethics approval

Not required.

### Data sharing

All data collected and utilized for study analysis has been provided as supplementary material 1 and 2.

### Transparency statement

The corresponding author (AS) affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted and that any discrepancies from the study as planned have been explained.

### Role of the funding source

None of the authors received any specific funding for this study from their affiliated institutions or external funders.

### Dissemination to public communities

The results from the present study will be disseminated to appropriate audiences such as academia, policy makers, and the general public through various channels, including press release, social media, e-newsletter, websites of collaborators’ institutions, and monthly bulletins.

*   Received June 11, 2020.
*   Revision received June 11, 2020.
*   Accepted June 12, 2020.


*   © 2020, Posted by Cold Spring Harbor Laboratory

This pre-print is available under a Creative Commons License (Attribution-NonCommercial-NoDerivs 4.0 International), CC BY-NC-ND 4.0, as described at [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/)

## References

1.  1.WHO Director-General’s opening remarks at the media briefing on COVID-19 - 11 March 2020. 2020.
    
    

2.  2.World Health Organization. Rolling updates on coronavirus disease (COVID-19) [Internet]. [cited 2020 March 21]. Available from: [https://www.who.int/emergencies/diseases/novel-coronavirus-2019/events-as-they-happen](https://www.who.int/emergencies/diseases/novel-coronavirus-2019/events-as-they-happen).
    
    

3.  3.Ferguson NM, Laydon D, Nedjati-Gilani G, et al. Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand. London: Imperial College London, 2020 16.03.2020. Report No.: 9.
    
    

4.  4.World health Organisation. Coronavirus disease 2019 (COVID-19) Situation Report –106. Geneva: World Health Organisation, 2020.
    
    

5.  5.Wilder-Smith A, Chiew CJ, Lee VJ. Can we contain the COVID-19 outbreak with the same measures as for SARS? The Lancet Infectious Diseases.
    
    

6.  6.Hutt R. The economic effects of COVID-19 around the world. World Economimc Forum, 2020.
    
    

7.  7.Abiad A, Arao RM, Platitas R, et al. The impact of COVID-19 on developing Asia—an update. Asian Development Outlook 2020. 2020.
    
    

8.  8.Wang D, Hu B, Hu C, et al. Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus–Infected Pneumonia in Wuhan, China. JAMA. 2020;323(11):1061–9.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jama.2020.1585&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom) 

9.  9.Zhang J, Litvinova M, Wang W, et al. Evolving epidemiology and transmission dynamics of coronavirus disease 2019 outside Hubei province, China: a descriptive and modelling study. The Lancet Infectious diseases. 2020.
    
    

10. 10.Jung SM, Akhmetzhanov AR, Hayashi K, et al. Real-Time Estimation of the Risk of Death from Novel Coronavirus (COVID-19) Infection: Inference Using Exported Cases. Journal of clinical medicine. 2020;9(2).
    
    

11. 11.Wei WE, Li Z, Chiew CJ, et al. Presymptomatic Transmission of SARS-CoV-2 — Singapore, January 23–March 16, 2020. Morbidity and Mortality Weekly Report (MMWR). 2020;69.
    
    

12. 12.Choi SC, Ki M. Estimating the reproductive number and the outbreak size of Novel Coronavirus disease (COVID-19) using mathematical model in Republic of Korea. Epidemiology and health. 2020:e2020011.
    
    

13. 13.Zhao S, Chen H. Modeling the epidemic dynamics and control of COVID-19 outbreak in China. Quantitative biology (Beijing, China). 2020:1–9.
    
    

14. 14.Fox GJ, Trauer JM, McBryde E. Modelling the impact of COVID-19 upon intensive care services in New South Wales. The Medical Journal of Australia. 2020.
    
    

15. 15.Chang SL, Harding N, Zachreson C, et al. Modelling transmission and control of the COVID-19 pandemic in Australia. 2020.
    
    

16. 16.Moghadas SM, Shoukat A, Fitzpatrick MC, et al. Projecting hospital utilization during the COVID-19 outbreaks in the United States. Proc Natl Acad Sci USA. 2020;117(16):9122–6.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoicG5hcyI7czo1OiJyZXNpZCI7czoxMToiMTE3LzE2LzkxMjIiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMC8wNi8xMi8yMDIwLjA2LjExLjIwMTI4NTYxLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

17. 17.Sajadi MM, Habibzadeh P, Vintzileos A, et al. Temperature, Humidity and Latitude Analysis to Predict Potential Spread and Seasonality for COVID-19. SSRN. 2020.
    
    

18. 18.Luo W, Majumder MS, Liu D, et al. The role of absolute humidity on transmission rates of the COVID-19 outbreak. medRxiv. 2020:2020.02.12.20022467.
    
    

19. 19.Bukhari Q, Jameel Y. Will Coronavirus Pandemic Diminish by Summer? SSRN. 2020:15.
    
    

20. 20.Miller A, Reandelar MJ, Fasciglione K, et al. Correlation between universal BCG vaccination policy and reduced morbidity and mortality for COVID-19: an epidemiological study. medRxiv. 2020:2020.03.24.20042937.
    
    

21. 21.Qiu Y, Chen X, Shi W. Impacts of social and economic factors on the transmission of coronavirus disease (COVID-19) in China. medRxiv. 2020:2020.03.13.20035238.
    
    

22. 22.World health Organisation. Coronavirus disease 2019 (COVID-19) Situation Report –68. Geneva: World Health Organisation, 2020.
    
    

23. 23.The World Bank Group. World Development Indicators. UHC Service Coverage Index [Internet]. The World Bank. 2017. Available from: [https://data.worldbank.org/indicator/SH.UHC.SRVS.CV.XD](https://data.worldbank.org/indicator/SH.UHC.SRVS.CV.XD).
    
    

24. 24.Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE). Novel Coronavirus (COVID-19) Cases: CSSE COVID-19 Dataset [Internet]. 2020. Available from: [https://github.com/CSSEGISandData/COVID-19](https://github.com/CSSEGISandData/COVID-19).
    
    

25. 25.Institute of Health Metrics and Evaluation (IHME). GBD Data Visualizations [Internet]. IHME, University of Washington 2016 [cited 12.4.2020]. Available from: [http://www.healthdata.org/gbd/data-visualizations](http://www.healthdata.org/gbd/data-visualizations).
    
    

26. 26.Weather Underground [Internet]. 2020 [cited 2020 April 04]. Available from: [https://www.wunderground.com/history](https://www.wunderground.com/history).
    
    

27. 27.IATA. WATS World Air Transport Statistics 2019. Geneva: International Air Transport Association; 2019.
    
    

28. 28.Worldometers.info. COVID-19 Coronavirus Pandemic Dover, Delaware, U.S.A.2020 [updated April 12, 2020April 12, 2020]. Available from: [https://www.worldometers.info/coronavirus/#page-top](https://www.worldometers.info/coronavirus/#page-top).
    
    

29. 29.McCullagh P. Generalized Linear Models. 2nd ed. New York: Routledge; 1989.
    
    

30. 30.Wood SN. Generalized additive models: an introduction with R: Chapman and Hall/CRC; 2017.
    
    

31. 31.Kruk ME, Gage AD, Joseph NT, et al. Mortality due to low-quality health systems in the universal health coverage era: a systematic analysis of amenable deaths in 137 countries. Lancet (London, England). 2018;392(10160):2203–12.
    
    

32. 32.Son K-B, Lee T-J, Hwang S-s. Does the health system’s response matter to cope with the COVID-19 outbreak? Recent evidence from South Korea. Bulletin of the World Health Organization.
    
    

33. 33.Bouey J. Strengthening China’s Public Health Response System: From SARS to COVID-19. American Journal of Public Health. 2020:e1–e2.
    
    

34. 34.Geographic Differences in COVID-19 Cases, Deaths, and Incidence — United States MMWR Morb Mortal Wkly Rep. 2020;69:465–71.
    
    

35. 35.Hellewell J, Abbott S, Gimma A, et al. Feasibility of controlling COVID-19 outbreaks by isolation of cases and contacts. The Lancet Global Health. 2020;8(4):e488–e96.
    
    

36. 36.Du R-H, Liang L-R, Yang C-Q, et al. Predictors of Mortality for Patients with COVID-19 Pneumonia Caused by SARS-CoV-2: A Prospective Cohort Study. European Respiratory Journal. 2020:2000524.
    
    

37. 37.Opal SM, Girard TD, Ely EW. The Immunopathogenesis of Sepsis in Elderly Patients. Clinical Infectious Diseases. 2005;41:S504–12.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1086/432007&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16237654&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom) 

38. 38.Fitzpatrick AL, Powe NR, Cooper LS, et al. Barriers to Health Care Access Among the Elderly and Who Perceives Them. American Journal of Public Health. 2004;94(10):1788–94.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.2105/AJPH.94.10.1788&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=15451751&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000224309900031&link_type=ISI) 

39. 39.Shrestha LB. Population Aging In Developing Countries. Health Affairs. 2000;19(3): doi: [https://doi.org/10.1377/hlthaff.19.3.204](https://doi.org/10.1377/hlthaff.19.3.204).
    
    

40. 40.Kang D, Choi H, Kim J-H, et al. Spatial epidemic dynamics of the COVID-19 outbreak in China. International Journal of Infectious Diseases. 2020;94:96–102.
    
    

41. 41.Wong G, Liu W, Liu Y, et al. MERS, SARS, and Ebola: The Role of Super-Spreaders in Infectious Disease. Cell Host & Microbe. 2015;18(4):398–401.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.chom.2015.09.013&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26468744&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom) 

42. 42.Malta M, Rimoin AW, Strathdee SA. The coronavirus 2019-nCoV epidemic: Is hindsight 20/20? E Clinical Medicine. 2020: doi: 10.1016/j.eclinm.2020.100289.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.eclinm.2020.100289&link_type=DOI) 

43. 43.Chan KH, Peiris JSM, Lam SY, et al. The Effects of Temperature and Relative Humidity on the Viability of the SARS Coronavirus. Advances in Virology. 2011:doi:10.1155/2011/734690.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1155/2011/734690&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22312351&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom) 

44. 44.Xie J, Zhu Y. Association between ambient temperature and COVID-19 infection in 122 cities from China. Sci Total Environ. 2020;724:138201-.
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom) 

45. 45.Shi S, Tanaka S, Ueno R, et al. Impact of travel restrictions on importation of novel coronavirus infection: An effective distance approach. Bull World Health Organ. 2020:doi: [http://dx.doi.org/10.2471/BLT.20.255679](http://dx.doi.org/10.2471/BLT.20.255679).
    
    

46. 46.Chetterje P. Gaps in India’s preparedness for COVID-19 control. The Lancet Infectious Diseases. 2020;20(5): doi: 10.1016/S473-3099(20)30300-5.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S473-3099(20)30300-5&link_type=DOI) 

47. 47.Malpass D. Coronavirus highlights the need to strengthen health systems World Bank Blogs [Internet] 2020. [cited 2020]. Available from: [https://blogs.worldbank.org/voices/coronavirus-covid19-highlights-need-strengthen-health-systems](https://blogs.worldbank.org/voices/coronavirus-covid19-highlights-need-strengthen-health-systems).
    
    

48. 48.Chatterjee P, Nagi N, Agarwal A, et al. The 2019 Novel Coronavirus Disease (COVID-19) Pandemic: A Review of the Current Evidence. Indian J Med Res. 2020:doi: 10.4103/ijmr.IJMR\_519\_20.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access\_num=10.4103/ijmr.IJMR_519_20&link_type=DOI) 

49. 49.Cash R, Patel V. Has COVID-19 subverted global health? The Lancet. 2020: doi: 10.1016/S0140-6736(20)31089-8.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(20)31089-8&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F06%2F12%2F2020.06.11.20128561.atom)

 [1]: /embed/graphic-3.gif
 [2]: /embed/graphic-4.gif