Epidemiological Determinants of Acute Respiratory Syndrome Coronavirus-2 Disease Pandemic and The Role of the Bacille-Calmette-Guerin Vaccine in Reducing Morbidity and Mortality

1Division of Epidemiology, ICAR-Indian Veterinary Research Institute, Izatnagar-243122, India. 2Legal Advisor on International Trade and Investment Laws, 23A Vaibhav, Suncity Vistaar, Bareilly, 243122, India. 3Division of Bacteriology and Mycology, ICAR-Indian Veterinary Research Institute, Izatnagar-243122, India. 4Avian Diseases, Division of Pathology, ICAR-Indian Veterinary Research Institute, Izatnagar-243122, India. 5Department of Biotechnology, DCR University of Science & Technology, Murthal-131039 Sonepat Haryana India. 6Department of Engineering, Boise State University, Idaho, USA.


iNTRODuCTiON
Novel Coronavirus or severe acute respiratory syndrome Coronavirus-2 (SARS-CoV-2) infection causes the Coronavirus disease known as COVID-19, which is presently a devastating pandemic. It has spread throughout most of the world (210 countries and territories) and has affected more than 2.52 million people. With a high case fatality rate (CFR, 21%) COVID-19 has killed more than 174,000 people 1 . Besides morbidity and mortality, the pandemic may lead to a global economic depression, which will be evident in the coming months and years, leaving millions of people jobless, homeless, and foodless. Despite the fact that lockdown and social distancing in response to such contagious diseases have negative socioeconomic effects 2,3 , almost half of the world is in lockdown today from the widespread fear of COVID-19. An understanding of epidemiology and the determinants of disease spread, morbidity, and mortality, in addition to socioeconomic cost-benefit analyses are prerequisites for overcoming this problem. Although monkeys have been shown to be susceptible 4 and some cats, dogs, and tigers tested positive for COVID-19 infection ,5-7 , the disease is mainly spreading through human-to-human contact. However, the epidemiology of COVID-19 remains unclear with respect to its origin and the reservoir and intermediate hosts (if any). The aim of this study was to analyze the association of different socioeconomic and population demographic factors with morbidity, mortality, CFR, and percent positivity among people tested for COVID-19, using data available in the public domain.

Data Sources
Data from different countries related to the total number of COVID-19 tests, the positivity rate in tested samples, morbidity, mortality, CFR, and population density were retrieved from Worldometer 8 . The gross domestic product per capita (using the purchase power parity [PPP] metric) 9,10 and data pertaining to the use of Bacille-Calmette-Guerin (BCG) vaccination 11,12 ; literacy rate; median age 13 ; heath care expenditure per capita; life expectancy at birth; healthy life expectancy; the burden of tuberculosis (TB), acquired immunodeficiency disease caused by human immunodeficiency virus (HIV-AIDS), malaria, cardiovascular disease (CVD), neoplasm, and diabetes; deaths due to energy-protein (food) deficiency (EPD); and per capita government spending on safe water and sanitation 14,15 were downloaded from various websites in the public domain.
These data were entered into a Microsoft Excel2007 spreadsheet for tabulation, analysis (Chi squared test and calculations of correlation coefficients and odds ratios [ORs]), ranking, and other studies. The significance of each correlation was determined using standard reference tables.

ReSulTS AND DiSCuSSiON
Factors showing a significant correlation with morbidity, mortality, CFR, and percent positivity for COViD- 19 The factors showing a significant positive or negative correlation with COVID-19 mortality ( Fig. 1) and morbidity ( Fig. 2) rates, CFR, and percent positivity among tested individuals are listed in Table 1 & 2, which show odds ratios and statistical significance.

BCG vaccination, TB burden, and COViD-19
In most developed countries, BCG vaccination ceased many years ago, after the countries were almost free of TB or achieved a low TB burden (less than 100 cases per million people) 14, 16 . The BCG vaccine is controversial due to its questionable efficacy in protecting from Mycobacterium tuberculosis infection 11,12 . The analysis of COVID-19 data from 190 countries with known BCG vaccination status revealed that morbidity (r = -0.62, p<0.0001) and mortality (r = -0.58, p<0.0001) rates and percent positivity among individuals tested for SARS-CoV-2 infection (r = -0.25, p = 0.01) showed significant negative correlations with the mandatory use of the BCG vaccine ( Table 2). This observation indicated the protective efficacy of childhood BCG vaccination in protecting from COVID-19 morbidity and mortality. A plausible reason may be the induction of nonspecific and cross-protective immunity by BCG vaccination 17,18 , even against viral infections.
Up to April 15, 2020, COVID-19 morbidity (Fig. 3) and mortality ( Fig. 4) rates were lower in countries with mandatory BCG vaccination than in those with optional or no BCG vaccination. The odds of having >500 COVID-19 cases per million people was 133 times higher in countries with non-mandatory BCG vaccination than in those with mandatory BCG vaccination (99% CI, 17.5-1011.37; Table 2). The odds of having >50 COVID-19-related deaths per million people was 42.96 times higher in countries without mandatory BCG vaccination The burden of TB was also negatively correlated (p = 0.05) with COVID-19 mortality (r = -0.17) and morbidity (r = -0.19) rates. This may be owing to the association of Mycobacterium spp. with the non-specific induction of immunity against viral diseases 17,18 . Although COVID-19 morbidity (OR 24.46; 99%CI, 1.72-347.8) and mortality (OR 10.83; 95% CI, 1.57-92.2) rates were significantly higher in countries with a low TB burden (<100 deaths/million/year) than in countries with a high TB burden, COVID-19 CFR was positively correlated with TB burden (r = 0.17, p = 0.05). This observation indicated that TB, which may result in injured or stressed lungs, may make a patient more prone to the respiratory failures seen with COVID-19. Underlying disorders that cause respiratory distress have been identified as important co-morbidities leading to lethality in COVID-19 patients 24 . However, further studies are required to test this hypothesis.

The median age of the population and COViD-19 morbidity, mortality, and CFR
Neither CFR nor percent positivity among individuals tested for COVID-19 showed a significant correlation with the median age of the population. COVID-19-related morbidity (Fig. 3) and mortality (Fig. 4) rates were higher in countries with a higher median age (>30years) than in those with a lower median age (≤30 years). However, countries with a high (>40 years) median age had a greater chance of having higher morbidity (OR 11.2; 99%CI, 2.68-47.9) and mortality (OR 8.18; 99%CI, 1.57-42.49) rates than those with a low (≤30 years) or medium  Table 2). The median age of the population showed a strong positive correlation with COVID-19 mortality (r = 0.34, p = 0.0001) and morbidity (r =0.40, p = 0.0001) rates. High COVID-19 morbidity and mortality rates have been reported to be associated with other co-morbidities of old age 24 . However, the positive correlation between high median age and morbidity (r = 0.44, p= 0.0001) and mortality (r = 0.35, p = 0.0001) rates in countries with a greater healthy life expectancy, indicated that the role of co-morbidities may be modulated by some currently unknown confounders of COVID-19 and median age associations.
After including the effects of both median age and BCG vaccination in the analysis, mandatory BCG vaccination was found to be an important negative modulator of SARS-CoV-2 infection, irrespective of median age ( Table 2).

Disease burden and COViD-19 morbidity, mortality, and CFR
Besides TB, as discussed above, the burden of HIV-AIDS and CVD and EPD death rates also showed significant negative correlations with either COVID-19 morbidity, mortality, or both (Table 1). However, these factors did not show a significant correlation with COVID-19 CFR. Neither the prevalence of malaria nor diabetes showed a significant correlation with COVID-19 CFR, morbidity, or mortality. However, in earlier case-to-case analyses, diabetics have been shown to be more prone to COVID-19-related mortality 25 . A higher burden of neoplasia was strongly associated with increased COVID-19 morbidity (r = 0.47, p = 0.0001) and mortality (r = 0.39, p = 0.0001) rates. Neoplasia has been reported to be an important co-morbidity associated with worse outcomes of SARS-CoV-2 infection in different countries [24][25][26][27] . Though several co-morbidities have been shown [24][25][26][27] to modulate the outcome of COVID-19, further studies and targeted data analysis are needed to better understand the relationship between these diseases.

COViD-19 and population density
Population density showed no significant correlation with COVID-19 CFR, morbidity, or mortality. This is contrary to the popular belief that contagious diseases spread faster in densely populated countries 28 because of the greater chance of interpersonal contact. This contradiction

Fig. 2. Determinants of COVID-19 morbidity (per million population) patterns
Journal of Pure and Applied Microbiology may be due to several factors involved in the spread of COVID-19, including national and international travel and social or religious congregations 19,29 . The similar instances of person-to-person contact and person-to-contaminated-surface contact in closed areas with virus-containing aerosols can result in an increase in disease burden, even in places with a low population density. literacy rate and COViD- 19 Literacy is always an advantage when dealing with pandemics. The present analysis indicated that literacy was positively correlated (r=  21, p = 0.01). However, there was no correlation between literacy and mortality or CFR, probably because novel viruses like SARS-CoV-2 are not likely to discriminate between literate and illiterate people, as they all may be at the same plane of immunity in a given country.
Role of per capita GDP, per capita health expenditure, availability of safe water, food sufficiency, and sanitation in COViD- 19 In this study, countries were classified into either high-, low-, or medium-income groups 30 based on per capita GDP (using PPP statistics), to evaluate the effect of per capita GDP on COVID-19related morbidity and mortality. High PPP showed a strong positive correlation with morbidity (r = 0.47; p<0.0001) and mortality (r = 0.25, p = 0.01; Table 1). Most wealthy countries with a high PPP have populations with a comparatively higher median age 8,12 and most have also discontinued the use of BCG 10,11 . However, correlating and making conclusions regarding these factors independently may introduce bias. When comparing the countries with a high PPP (> $12,476), with and without a mandatory BCG vaccination policy, BCG vaccination showed a strong (p<0.001) negative correlation with COVID-19 morbidity and mortality. In high-income countries with an optional BCG vaccination policy, the odds of having > 500 COVID-19 cases per million people was 134.6 times higher (99% CI, 8.66-2093.23) and the odds of having > 50 COVID-19 deaths per million people was 12.4 times higher (99%CI, 3.14-48.9) than in high-income countries with a mandatory BCG vaccination policy. A similar correlation was evident between per capita health expenditure and COVID-19 morbidity (r = 0.47, p<0.0001) and mortality (r = 0.25, p = 0.01). However, the number of deaths owing to EPD was negatively correlated with COVID-19 morbidity (r = 0.19, p<0.05) and mortality (r = 0.17, p = 0.05). This indicated that the countries with insufficient resources may be failing to do the required number of tests to diagnose silent COVID-19 infections or alternatively, they may be protected because of the effect of mandatory BCG vaccination in all such countries because of their higher burden of TB 12, 15,16 . is COViD-19 misdiagnosed?
Our analysis revealed that increased testing (number of tests per million people) was correlated with a lower COVID-19 CFR (r = -0.17, p = 0.05), but higher morbidity (r = 0.57, p = 0.001) and mortality (r = 0.19, p = 0.05) rates, suggesting that COVID-19 might be over-diagnosed. The present diagnosis is expected to overestimate SARS-CoV-2 infectivity and COVID-19 morbidity and mortality; since it is targeted to detect SARS-CoV-2 in all suspected flu cases and all deaths in the pandemic regions. A higher number of cases have been registered in countries with high literacy rates and stronger economies, which also have higher testing capacity, further indicating that it may have been over-diagnosed.
Presently, testing for COVID-9 is aimed at identifying cases, without any goal of disease mapping and thus, until testing is performed with a sound sampling plan, the real incidence and prevalence of the disease cannot be accurately determined. Moreover, in the absence of real mapping of the disease, its determinants may not be identified correctly 31 .
Further, the number of individuals that tested positive for COVID-19 may be dissuasive in understanding the disease, owing to compromised positive and negative predictive values 32,33 . Falsepositivity in COVID-19 testing 32 may be proving costly, but it can be tolerated for effective epidemic control 19 . However, the current RT-PCR-based test has a high negative predictive value 33 and may result in missing a large number of positive cases, which may be contributing to the propagation of the ongoing pandemic. Negative RT-PCR results for COVID-19 necessitate repeated testing to confirm the negative result. Thus, a better test with higher positive and negative predictive values is required to determine the true number of positive and negative cases. The requirement for repeated and parallel testing may force many poverty-ridden countries to completely abandon testing owing to its high cost. It becomes difficult to eradicate a disease when the necessities of life are becoming more difficult to maintain, as in most of lower-middle-income countries, because of high economical cost of testing, quarantine, and isolation of COVID-19 patients. A recent study on the prevalence of anti-SARS-CoV-2 antibodies in Santa Clara County 34 revealed that the RT-PCRbased test can diagnose only a small fraction (1/50 to 1/80) of SARS-CoV-2-infected people, indicating that the real infection fatality rate (IFR) may be just 0.1-0.9%. The IFR for Santa Clara County was found to be0.1-0.2%, which represents one death in 480-820 infections. These observations indicate that there is much more research required to reach valid conclusions regarding the optimal strategy to map and control COVID-19.
COVID-19 mainly spreads through person-to-person contact during official and unofficial meetings, travel via public transport, and social and religious congregations. The contagiousness of COVID-19 has transformed it into a pandemic within a very short period of 2 months. Although social distancing and lockdown are considered effective strategies to contain the disease and are in place around the world, many of the affected countries may fail to contain the pandemic. This failure may be owing to frequent violations 29 and misunderstanding of the practice of social distancing and the fact that these are impracticable and costly means to control an epidemic caused by agents with high contagiousness (R>2) and low morbidity and mortality rates 2,3 . At present, in most of the worst-affected countries, the COVID-19 morbidity rate is 1% and the mortality rate is 0.1%, making it a fit case for an upcoming endemic. Eventually, to prevent further economic downturn, the lockdown will be lifted, which will allow people to interact freely and propagate the disease at low levels. Although the WHO has denied any beneficial correlation between BCG vaccination andCOVID-19 morbidity and mortality, the current analysis has revealed the possible utility of the BCG vaccine in reducing COVID-19 morbidity and mortality rates and the higher susceptibility of aged people to COVID-19.