See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/343180976

Early trends of socio-economic and health indicators influencing case fatality

rate of COVID-19 pandemic

Article  in  Monaldi archives for chest disease = Archivio Monaldi per le malattie del torace / Fondazione clinica del lavoro, IRCCS [and] Istituto di clinica tisiologica e malattie apparato
respiratorio, Università di Napoli, Secondo ateneo · July 2020
DOI: 10.4081/monaldi2020.1388

CITATIONS

6 authors, including:

Shahir Asfahan Gopal Chawla

All India Institute of Medical Sciences Jodhpur All India Institute of Medical Sciences Jodhpur
12 PUBLICATIONS   1 CITATION    44 PUBLICATIONS   3 CITATIONS   

SEE PROFILE SEE PROFILE

Naveen Dutt Neeraj Gupta

All India Institute of Medical Sciences Jodhpur All India Institute of Medical Sciences Jodhpur
64 PUBLICATIONS   88 CITATIONS    57 PUBLICATIONS   197 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

COVID-19 View project

Smoking Cessation in Obstructive Airway Disease View project

All content following this page was uploaded by Gopal Chawla on 24 July 2020.

The user has requested enhancement of the downloaded file.

 Monaldi Archives for Chest Disease 2020; volume 90:1388

Early trends of socio-economic and health indicators influencing case

fatality rate of COVID-19 pandemic
Shahir Asfahan,1 Aneesa Shahul,1 Gopal Chawla,1 Naveen Dutt,1 Ram Niwas,1 Neeraj Gupta2
1Department of Pulmonary Medicine; 2Department of Paediatrics, All India Institute of Medical Sciences, Jodhpur, India

health indicators were accessed from the World Health

Abstract Organisation (WHO) database. Various socioeconomic indicators
and health indicators were selected for this analysis. After select-
Coronavirus disease 2019, i.e. COVID-19, started as an out- ing from univariate analysis, the indicators with the maximum
break in a district of China and has engulfed the world in a matter correlation were used to build a model using multiple variable lin-
of 3 months. It is posing a serious health and economic challenge ear regression with a forward selection of variables and using
worldwide. However, case fatality rates (CFRs) have varied adjusted R-squared score as the metric. We found univariate
amongst various countries ranging from 0 to 8.91%. We have regression results were significant for GDP (gross domestic prod-
evaluated the effect of selected socio-economic and health indica- uct) per capita, POD 30/70 (probability of dying between age 30

ly
tors to explain this variation in CFR. Countries reporting a mini- and exact age 70 from any of cardiovascular disease, cancer, dia-
mum of 50 cases as on 14th March 2020, were selected for this betes or chronic respiratory disease), HCI (human capital index),

on
analysis. Data about the socio-economic indicators of each coun- GNI (gross national income) per capita, life expectancy, medical
try was accessed from the World bank database and data about the doctors per 10000 population, as these parameters negatively
corelated with CFR (rho = -0.48 to -0.38, p<0.05). Case fatality

e
rate was regressed using ordinary least squares (OLS) against the
us
socio-economic and health indicators. The indicators in the final
Correspondence: Dr. Gopal Chawla, Department of Pulmonary
model were GDP per capita, POD 30/70, HCI, life expectancy,
Medicine, All India Institute of Medical Sciences, Jodhpur, India.
medical doctors per 10,000, median age, current health expendi-
al
Tel. +91.9999883667.
E-mail: [email protected] ture per capita, number of confirmed cases and population in mil-
ci

lions. The adjusted R-squared score was 0.306. Developing coun-

Key words: COVID-19; pandemic; case fatality rate. tries with a poor economy are especially vulnerable in terms of
er

COVID-19 mortality and underscore the need to have a global

Funding: None. policy to deal with this on-going pandemic. These trends largely
m

confirm that the toll from COVID-19 will be worse in countries

Conflict of interest: The authors declare no conflict of interest.
ill-equipped to deal with it. These analyses of epidemiological
om

Contributions: SA, AS, GC, ND, study concepts; SA, GC, ND, RN, data are need of time as apart from increasing situational aware-
study design; SA, GC, ND, definition of intellectual content; SA, GC, ness, it guides us in taking informed interventions and helps poli-
literature search; SA, GC, clinical studies; SA, GC, experimental stud- cy-making to tackle this pandemic.
-c

ies; SA, GC, ND, data acquisition; SA, GC, data analysis; SA,GC,
NG, statistical analysis; SA, GC, ND, manuscript preparation; SA,
on

AS, GC, ND, manuscript editing; SA, AS, GC, RN, NG, manuscript
review; SA, GC, Guarantors. All the authors have read and approved
the final version of the manuscript and agreed to be accountable for
Introduction
N

all aspects of the work.

Corona virus (CoV) comprises of a large family of viruses that
Availability of data and materials: All data generated or analyzed dur- are common in human beings as well as animals. Other viruses of
ing this study are included in this published article. this family are known to cause severe illnesses like MERS
(Middle East respiratory syndrome) and SARS (severe acute res-
Ethics approval and consent to participate: Not applicable.
piratory syndrome). SARS-CoV-2 is a novel virus which was first
Received for publication: 18 May 2020. identified in December 2019 as a cause of upper and lower respi-
Accepted for publication: 15 July 2020. ratory tract infection in Wuhan, a city in the Hubei Province of
China. This infection was labelled as Coronavirus disease 2019,
©Copyright: the Author(s), 2020
i.e. COVID-19. It rapidly spread through person-to-person trans-
Licensee PAGEPress, Italy mission via droplets and fomites, resulting in an epidemic
Monaldi Archives for Chest Disease 2020; 90:1388 throughout China, and then gradually spreading to other parts of
doi: 10.4081/monaldi.2020.1388
the world [1]. A pandemic by definition involves a wide spectrum
This article is distributed under the terms of the Creative Commons of societies and nations [2]. Historically, the impact of pandemics
Attribution Noncommercial License (by-nc 4.0) which permits any have varied between countries. Socio-economic and health indica-
noncommercial use, distribution, and reproduction in any medium, tors of a country may reflect the levels of preparedness and ability
provided the original author(s) and source are credited. to handle a pandemic. The current pandemic of COVID-19 is not
different. It continues to exact a heavy toll in terms of mortality

[Monaldi Archives for Chest Disease 2020; 90:1388] [page 451]

Original Article

and different countries have adopted different strategies to tackle reported in patients confirmed with COVID-19 divided by the
this. Although the disease is of recent origin, the case fatality rates number of confirmed cases reported. Socio-economic and health
(CFR) are different for different countries with some countries indicators were considered to prepare an explanatory model to
bearing the brunt of it while others have tried to limit the impact chart the variation in CFR of different countries so far (Table 1).
for now [3]. Previous experience with the H1N1 pandemic has The following socioeconomic indicators were taken from the
shown that levels of socio-economic and health indicators do influ- World Bank database [6]:
ence case fatality rates. In previous H1N1 pandemic an economi- (i) Gross domestic product (GDP)
cally stronger region like Europe performed better with a pooled (ii) Population
hospital fatality rate while in some developing countries it was as (iii) Life expectancy
high as 52% [4]. This study aims to evaluate the various factors (iv) Gross national income (GNI) Per Capita
that are involved in the determination of CFR associated with (v) GDP per capita
COVID-19. We have selected a wide spectrum of socio-economic (vi) Human capital index (HCI)
and health indicators of countries dealing with COVID-19 and GDP is the sum of gross value added by all resident producers in
attempted to explain some of the variability in CFR so that this the economy plus any product taxes and minus any subsidies not
data might be useful in planning strategies in countries yet to bear included in the value of the products. GDP is an indicator of the
the full impact of COVID-19. overall level of economic development and influences the level of
healthcare accessible to its citizens. Population of the country was
included because it can impact natural resources and social infras-
tructure. This can place pressure on a country’s sustainability and
Materials and Methods imposes a burden on the healthcare systems. Life expectancy at birth
indicates the number of years a new-born infant would live if pre-

ly
Data on the total numbers of confirmed cases and the number vailing patterns of mortality at the time of its birth were to stay the
of deaths as of 14th March 2020 were accessed from the WHO same throughout its life. It is an important indicator of the health sta-

on
database. CFR of all countries reporting at least 50 confirmed tus of the country. GNI per capita is the gross national income, con-
cases of COVID-19 were considered for analysis (Figure 1) [5].
verted to U.S. dollars using the World Bank Atlas method, divided
Although the cut-off date of 14th March, 2020 appears quite early
by the midyear population. GNI in comparison to the GDP may bet-

e
in this pandemic, however this study was envisaged to understand
ter reflect the economic status of countries with large foreign remit-
us
the initial dynamics of the spread and impact of pandemic.
tances/external aid. GDP per capita is a measure of a country’s eco-
Case fatality rate has been defined as the number of deaths
nomic output divided by the population. It is a good measure of a
country’s standard of living which in turn affects access to affordable
al
healthcare [6]. Human capital index measures the amount of human
ci

capital that a child born today can expect to attain by age 18, given
the risks of poor health and poor education that prevail in the country
er

where she lives. It is designed to highlight the impact of improve-

ments in current health and education which shapes the productivity
m

of the next generation of workers.

om
-c

Table 1. Mean and standard deviations of the indicators selected

for analysis.
on

Variables Mean ± SD (n=45)

GDP in billions (US$) 1629.96±3614.9
N

Population in millions 99.08±283.1

Life expectancy 79.06±4
GNI per capita (US$) 33561.33 ±22588.75
GDP per capita (US$) 38532.31±21334.42
HCI 0.7±0.12
Median age (years) 36.23±6.66
CHE per capita (US$) 2901.33±2590.95
CHE as % GDP 7.72±3.04
Prevalence (%) of tobacco smoking 23.41±7.23
POD 30/70 14.05±5.21
Hospital bed/10,000 38.62±26.67
Medical doctors/10,000 28.18 ±12.91
Confirmed cases 3185.84+12307.92
Confirmed cases per million population 70.51±118.13
Population density/km2 389.39±1248.28
Figure 1. Flowchart of the study. Case fatality rate 1.55±2.25

[page 452] [Monaldi Archives for Chest Disease 2020; 90:1388]

 Original Article

The following health indicators were incorporated into the the Pearson correlation (Table 2). Factors with a p-value <0.2 were
model and accessed from the WHO database [7]: considered to build a model. Features with the lowest p-values
(i) Median age were selected in ascending order as part of the forward selection
(ii) Current health expenditure (CHE) per capita and multiple variable linear regression was applied against case
(iii) Current health expenditure (CHE) as a percentage of GDP fatality rate ordinary least squares (OLS). Forward selection of fea-
(iv) Prevalence (%) of tobacco smoking tures was done based on the adjusted R-squared score.
(v) Probability (%) of dying between age 30 and exact age 70 Heat map of all the features was constructed to assess the cor-
from any of cardiovascular disease, cancer, diabetes or relations between different features (Figure 2). All features were
chronic respiratory disease (POD 30/70) standardized to unit variance based on the following formula. The
(vi) Hospital beds per 10000 population standard score of sample x is calculated as Z = (X - U) / S where
(vii) Medical doctors per 10000 population U is the mean of the training samples, and S is the standard devia-
(viii) Confirmed cases of COVID-19 tion of the training samples. Standardization was done to maintain
(ix) Confirmed cases per million population uniformity.
(x) Population density/Km2
Median age is a determinant of age distribution in the society.
Age distribution could be a factor in the case fatality rates as
COVID-19 is preferentially more severe in the older age group. Results
Current Health Expenditure (CHE) per capita is the health budget
divided by the population. CHE as a percentage of GDP is the per- A total of 47 countries were selected for analysis based on the
centage of GDP spent on health resources for the country. CHE per inclusion criteria of reporting at least 50 cases of confirmed
capita and CHE as a percentage of GDP are a measure of spending COVID-19 (Figure 1). Taiwan and San Marino were removed

ly
on public healthcare and this may influence the response of a coun- from analysis due to the paucity of data from the above-mentioned
try to a pandemic. The prevalence of tobacco smoking mirrors the sources.

on
respiratory ailments in a country which might predispose to respira- A total of 143,363 were confirmed and 5379 deaths were
tory complications of COVID-19. POD 30/70 may reflect the back- reported in total from our dataset with a mean CFR of 3.75%. CFR
ground health co-morbidities in a country which might get compli- exhibited a wide range (0 to 8.91%) with a standard deviation of

e
cated by COVID-19 and may impact mortality as it has been shown 2.25% (Figure 3). In the univariate analysis, the parameters corre-
us
that case fatality rises in patients with co-morbidities. Hospital beds lating positively with CFR were POD 30/70, number of confirmed
and doctors per 10000 population are indicative of the robustness of cases, population in millions, GDP, prevalence of tobacco smok-
the healthcare system in the country. We evaluated confirmed cases ing, and confirmed cases per million population. Parameters asso-
al
per million population and population density to assess whether ciated with a negative correlation in univariate analysis were GDP
these determinants have a role in determining CFR [7]. per capita, HCI, GNI per capita, life expectancy, medical doctors
ci

per 10000 population, median age, CHE per capita, hospital beds
er

per 10000, CHE as a percentage of GDP and population

density/km2. Some of these correlations may appear counter-intu-
Statistical analyses
m

itive at first, however, we believe these changes reflect the differ-

ential impact of the pandemic on the developed and developing
om

Statistical analysis was performed using Python 3.6. countries, particularly in its initial course.
Continuous variables are presented as mean ± SD. Univariate cor- In the multiple variable linear regression, a total of nine indi-
relation of each feature was done against the case fatality rate using cators were selected based on the forward selection which gave the
-c
on

Table 2. Univariate regression with Pearson correlation coefficient between indicators and case fatality rates.
N

Pearson Correlation Coefficient p

GDP per capita(US$) -0.482 0.001
POD 30/70 0.459 0.002
HCI -0.454 0.002
GNI per capita(US$) -0.439 0.003
Life expectancy -0.438 0.003
Medical doctors/10,000 -0.435 0.003
Median age -0.389 0.008
CHE per capita(US$) -0.291 0.053
Confirmed cases 0.259 0.086
Population in millions 0.24 0.113
Hospital bed/10,000 -0.188 0.217
CHE as % GDP -0.167 0.274
GDP in billions (US$) 0.159 0.295
Prevalence(%) of tobacco smoking 0.123 0.42
Population density/km2 -0.105 0.494
Confirmed cases per million population 0.035 0.817

[Monaldi Archives for Chest Disease 2020; 90:1388] [page 453]

Original Article

best adjusted R-squared score of 0.306 (Table 3). The F-statistic Resources that influence pandemic response can be categorized
(3.152) for the model was significant (p=0.00691). Individually in into financial, human and physical resources. Apart from these
the model, GDP per capita and the number of confirmed cases had resources leadership, intergovernmental relationships, onsite
a p<0.05 (Table 3). response and information sharing have also been identified as fac-
tors that influence response [8]. Here in our study we have tried to
study financial and human response to the current pandemic. In the
previous studies on the H1N1 pandemic, evaluating the effects of
Discussion economic indicators on CFR and the effect of CFR on economic
indicators were influential in tailoring the responses to future pan-
We conducted a study to assess the impact of various social demics. Some of the measures like school closures and quarantines
and health indicators on case fatality rate of COVID-19. In 45 have been re-instated in the present pandemic as learned from that
countries included for the study according to inclusion and exclu- experience [9]. As in the previous study that explored determinants
sion criteria, we observed that various indicators had a significant of H1N1 hospitalization, populations at the margins of society with
effect on case fatality. These factors may be one of the reasons for poor social security were at higher risk even within an ecosystem
varying CFR in various countries. We propose the model of select- of an individual first world country [10].
ed indicators that may help in identifying the countries where CFR Univariate regression results were significant for GDP per
in the coming days of an evolving pandemic can be very high. capita, POD 30/70, HCI, GNI per capita, life expectancy, medical
COVID-19 is a disease that presents in a variety of ways rang- doctors per 10000 population and median age. Predictably all the
ing from being asymptomatic to being severely ill. While the above-mentioned factors were negatively correlated with CFR
severity of illness contributes to the fatality of this new disease, the except POD 30/70. However, GNI per capita was removed from
asymptomatic and mild cases play a major role in transmission [1]. the analysis as it correlated strongly with GDP per capita. These

ly
on
e
us
al
ci
er
m
om
-c
on
N

Figure 2. Heat map of the correlation between all the features under study.

[page 454] [Monaldi Archives for Chest Disease 2020; 90:1388]

 Original Article

factors are strongly related to the overall development status of a tion of the country helped improve the explanatory model for CFR.
country and reflect the strength of the medical systems in place. In The overall model was significant as determined by the F-statistic,
multivariate analysis, in addition to the significant univariate fac- however, in the adjusted model, GDP per capita and numbers of
tors, CHE per capita, the number of confirmed cases and popula- confirmed cases were significant variables underlying the impor-

Table 3. Multiple variable linear regression results of our model OLS (ordinary least squares) regression results.
Dependent variable Case fatality rate R squared 0.448
No. of observations 45 Adj R squared 0.306
Df residuals 35 F- statistic 3.152
Df model 9 Prob (F statistic) 0.00691
Coefficient Standard error p [0.025 0.975]
0 0.126 1.000 -0.255 0.255
GDP per capita (US$) -0.4010 0.164 0.020 -0.735 -0.067
POD 30/70 0.4755 0.357 0.191 -0.248 1.199
HCI -0.2728 0.331 0.415 -0.945 0.399

ly
Life expectancy 0.5541 0.505 0.280 -0.471 1.579

on
Medical doctors per 10,000 -0.2548 0.222 0.260 -0.706 0.197
Median age -0.2165 0.283 0.449 -0.790 0.357
CHE per capita (US$) 0.2922 0.186 0.125 -0.086 0.670

e
Confirmed cases 0.3871 0.183 0.042 0.015 0.759
Population in millions -0.2318 0.206
us 0.268 -0.650 0.187
al
ci
er
m
om
-c
on
N

Figure 3. Case fatality rates by countries reporting at least 50 confirmed cases of COVID-19.

[Monaldi Archives for Chest Disease 2020; 90:1388] [page 455]

Original Article

tance of these factors in the overall explanation of CFR in these number of medical doctors per 10000 population, HCI, health care
countries. Our model of selected socio-economic and health indi- budget spent per capita and instituting population control mea-
cators could explain 30.6% of the variability in CFR. sures which will enable us to effectively tackle future pandemics.
GDP per capita is an overall metric of a nation and it can be It is pertinent to note that our model despite an extensive reper-
extrapolated to the kind of health care systems which exists in the toire of indicators could only explain up to 30.6% of the variability
country. It comes as no surprise that increasing GDP per capita is in the case fatality rates. This indicates that the COVID-19 pan-
associated with decreasing mortality. The total number of con- demic is difficult to predict and model at present. Either more and
firmed cases was directly related to an increase in CFR. It may be different indicators need to be evaluated or more data is needed or
possible that a higher number of confirmed cases may result in more time is needed to understand the dynamics of this disease.
overwhelming healthcare systems leading to higher mortality. The limitation of this study is that with approximately 3
POD 30/70 reflects the background co-morbidities in the popula- months of data so far into this pandemic we were only able to catch
tion and an increase in this metric was associated with an increase the early trends and we believe with the passage of time and more
in the CFR. This implies that increased protection and care need to data at our disposal, we may be able to refine the model.
put in place for people with significant co-morbidities. An increase
in HCI was associated with lower CFR indicating that long term
development correlates with better management of pandemics.
Increasing life expectancy was associated with higher CFR which Conclusions
may reflect the higher burden of the older population in the coun-
try. An increase in medical doctors per 10,000 population led to a Each pandemic brings its own set of problems, but analysing
decrease in CFR which is understandable as it is a metric of better early trends during the pandemic may help in shaping the
health care systems. This measure can be improved to tackle future response to the pandemic itself. COVID-19 appears to be difficult

ly
pandemics and in the short term, doctors could be deputed to to control because of this virus’ high infectivity. The globalisation
underserved areas to tackle this disease. of economy and integration of various industrial sectors across the

on
Counter-intuitively, we can see that increasing median age led globe has made this pandemic spread fast, unlike any other pan-
to lower CFR and higher CHE per capita led to higher CFR. It demic. Countries with better economic infrastructure in place
could be possible that lower median age is a characteristic of might fare well in this pandemic compared to lesser privileged

e
developing countries which due to the nature of their relatively countries. Policymakers at international, regional and governmen-
under-developed healthcare systems could have led to higher CFR
seen so far [11]. As for higher CHE per capita leading to higher
us
tal levels need to formulate policies to address this imbalance as a
breakdown of the healthcare system in one place will put all con-
CFR, we believe that higher CHE per capita is a characteristic of nected countries at risk. A rapidly increasing number of confirmed
al
a developed country which implies a better testing for COVID-19, cases may overwhelm any healthcare systems; hence all possible
better reporting of data and lesser number of missed deaths measures should be instituted to keep it down so that effective
ci

attributable to other causes when in fact it could have been caused care can be provided to affected patients. Measures like quaran-
er

by COVID-19 [4]. tine, lockdowns, travel restrictions may help in keeping the mor-
Although countries with a fledgling economy like India have tality rates low. Overall, this pandemic like any pandemic exposes
m

been relatively unaffected so far with fewer cases compared to oth- the lacunae in existing systems and also provides an opportunity
ers in the region and beyond, however, going by past experience of to learn what measures work best and how to be prepared for
om

handling H1N1 pandemic, it portends an ominous outcome if newer challenges and what we learn today, may shape a better
unchecked and un-helped. Glaring health in-equalities in a devel- tomorrow.
oping country may allow for unchecked local transmission with
-c

potentially disastrous results [12]. Digging early into indicators

will help look into other parts of response management. Because
References
on

model which can be useful in China or United states may not be

feasible in countries like India or Pakistan. Since economic and
health care response could prove to be deficient in these countries 1. Guo YR, Cao QD, Hong ZS, et al. The origin, transmission and
N

we may need to look at other parts of response management which clinical therapies on coronavirus disease 2019 (COVID-19)
are leadership, intergovernmental relationships and innovative outbreak - an update on the status. Mil Med Res 2020;7:11.
measures. 2. WHO. What is a pandemic? Accessed on: 19 March 19 2020.
As we go forward in this yet evolving pandemic there are some Available from: http://www.who.int/csr/disease/swineflu/fre-
short-term and long-term measures that can be taken. Short term quently_asked_questions/pandemic/en/
measures are those where intervention is anticipated for 12 weeks. 3. Kucharski AJ, Russell TW, Diamond C, et al. Early dynamics
Countries with low GDP per capita like those in sub-Saharan of transmission and control of COVID-19: a mathematical
Africa, Southeast Asia, and Latin America may need help with modelling study. Lancet Infect Dis 2020;20:553-8.
resources to augment their healthcare systems to tide over this cri- 4. Wong JY, Kelly H, Cheung C-MM, et al. Hospitalization fatal-
sis. Measures like improving hygiene, encouraging handwashing, ity risk of influenza A(H1N1)pdm09: A systematic review and
better droplet etiquettes, social distancing, closing of schools post- meta-analysis. Am J Epidemiol 2015;182:294–301.
poning public gatherings and partial to complete lockdown includ- 5. WHO. Novel Coronavirus (2019-nCoV) situation reports.
ing public transport ban can prove helpful in halting the spread of Accessed on: 19 March 19 2020. Available from: https://www.
infection. All effective measures need to be put in place to limit the who.int/emergencies/diseases/novel-coronavirus-2019/situa-
total number of confirmed cases so that the health systems can tion-reports
respond better and do not impact the mortality of the other back- 6. World Bank. Open Data. Accessed on: 19 March 2020.
ground diseases [13]. Indeed, long term measures should be Available from: https://data.worldbank.org/
focussed towards improving healthcare systems including the 7. WHO. MoNITOR: Mother and Newborn information for

[page 456] [Monaldi Archives for Chest Disease 2020; 90:1388]

 Original Article

tracking and results. Accessed on: 19 March 2020. Available nants of health and pandemic H1N1 2009 influenza severity.
from: https://www.who.int/data/maternal-newborn-child-ado- Am J Public Health 2012;102:e51–8.
lescent/monitor 11. Our World in Data [Internet]. Age structure. 20 September
8. Lee K-M, Jung K. Factors influencing the response to infectious 2019. Accessed on: 19 March 2020. Available from:
diseases: Focusing on the case of SARS and MERS in South https://ourworldindata.org/age-structure
Korea. Int J Environ Res Public Health 2019 22;16:1432. 12. Kumar S, Quinn SC. Existing health inequalities in India:
9. Pasquini-Descomps H, Brender N, Maradan D. Value for informing preparedness planning for an influenza pandemic.
money in H1N1 influenza: A systematic review of the cost- Health Policy Plan 2012;27:516-26.
effectiveness of pandemic interventions. Value Health 13. Class RJ, Glass LM, Bayeler WE, Min HJ. Targeted social dis-
2017;20:819–27. tancing designs for pandemic influenza. Emerg Infect Dis
10. Lowcock EC, Rosella LC, Foisy J, et al. The social determi- 2006;12:1671-81.

ly
on
e
us
al
ci
er
m
om
-c
on
N

[Monaldi Archives for Chest Disease 2020; 90:1388] [page 457]

View publication stats