COVID-19 Pandemic: Analysis and Statistics of Confirmed Cases

*Vicente Enrique Machaca Arceda, Miguel Angel Machaca Arceda and Pamela Claudia Laguna Laura*

## **Abstract**

Coronavirus COVID-19 started in December 2019, and it has spread rapidly across China and the whole world. In this chapter, we analyzed the number of confirmed cases in US, India, France, Russia and Brazil. Additionally, we took into account Latin American countries like Argentina, Colombia, Peru, Chile and Mexico. We noticed, how some countries got a low death rate, despite its high number of confirmed cases (US). Additionally, it is interesting, how some countries with a high percentage of obesity got the highest death rate (Mexico). Also, we noticed a decreasing number in confirmed cases after a intensive vaccination plan (US). Finally, we evaluated Weibull Long Short-Term Memory (W-LSTM) and Multiplicative Trend Exponential Smoothing (MTES) to predict confirmed cases, in this case, W-LSTM showed a more realistic forecasting.

**Keywords:** Coronavirus, COVID-19, Analysis, Forecasting, LSTM, MTES

#### **1. Introduction**

Coronavirus COVID-19 pandemic started in December 2019 in Wuhan, China. This virus has high viral infectivity [1], so it has spread rapidly across China and other countries. Furthermore, 140,849,925 confirmed cases and 3,013,217 deaths were reported in the whole world until the last April 20th [2].

This new coronavirus made huge strain on the health system around the world forcing to establish decisions like quarantines and social distances in a effort to contain the spread of the virus [3]. Some countries with high incomes like United Kingdom, Italy, Spain and United States of America had to take measures such as hiring retired health personnel to assist battle infections.

Also, countries like United States agreed with car and weapon manufacturers to provide ventilators to help in the pandemic fight. The situation in countries with low and middle incomes were challenged, because they already have poor and weak health systems before COVID-19. They had limited financial resources, unavailable medications and inadequate health personnel, also in these countries exist a gap on the socio-economic. A person of higher socio-economic standing are more likely to have access to quality health services and medications [4].

Since the identification of SARS-CoV-2 virus, the scientific community was starting to develop over 300 vaccines projects, 40 of them are now on undergoing clinical evaluation, 10 of these are in Phase III and 3 of them have passed the phase III with effective outcomes. The existing data propound that the vaccine candidates can reduce the spread of the pandemic protecting individuals. On the other hand, the fast development of vaccines candidates carries with some unresolved issues (only time could clarify). Moreover, technical and ethical problems were added with the production of billions of doses [5]. Despite, there are dozens of potential vaccine candidates [6], the herd immunity has not achieved yet.

Nowadays, the scientist communities are publishing several papers of studies about COVID-19. For instance, a research team had published an analysis of confirmed cases with Multiplicative Trend Exponential Smoothing (MTES) and Long Short-Term Memory (LSTM) [7]. Nonetheless, other researchers made a comparison with Auto-Regressive Integrated Moving Average (ARIMA), Nonlinear Autoregression Neural Network (NARNN) and LSTM to predict the confirmed cases of Denmark, Belgium, Germany, France, United Kingdom, Finland, Switzerland and Turkey, they concluded that LSTM was the most accurate model [8]. In addition, LSTM had been used to predict the trends and possible ended time of COVID-19 [9]. Also, other research used LSTM to predict the cumulative recovered, fatalities and confirmed cases [10].

In this Chapter, we analyzed the evolution of COVID-19. We took into account countries with the major number of confirmed cases like US, India, France, Russia and Brazil. Additionally, we took into account Latin American countries like Argentina, Colombia, Peru, Chile and Mexico. We analyzed the evolution of confirmed cases, deaths, the effects of vaccination and finally, we evaluated some models to forecast the number of confirmed cases.

### **2. Analysis of confirmed cases and deaths**

In this section, we analyzed the number of confirmed cases and deaths in some countries. We focused on countries like US, India, France, Russia and Brazil, because, they got the major number of confirmed cases around the world. Additionally, we focused in the some Latin American countries like Argentina, Colombia, Peru, Chile and Mexico. They also, got the highest number of confirmed cases in Latin America.

#### **2.1 American countries**

The number of confirmed cases, varies from each country to another. For example in **Figure 1**, we showed the evolution of confirmed cases in top countries with the major impact. In this case, United States (US), shows an increasing curve

**Figure 1.** *Evolution of confirmed and deaths in US, India, France, Russia and Brazil. (a) Confirmed cases. (b) Deaths.*

#### *COVID-19 Pandemic: Analysis and Statistics of Confirmed Cases DOI: http://dx.doi.org/10.5772/intechopen.98891*

with 31,786,856 confirmed cases until April-20th. Furthermore, in **Figure 2**, we plotted the confirmed cases and deaths per million inhabitants. This shows a more realistic overview.

An interesting point is related to the differences between, confirmed cases and deaths in some countries. For instance, despite US got the major number of infected people, it has 1.77 of mortality rate (see **Table 1**), meanwhile France and Russia got 2.23 and 2.00 respectively. This difference between countries, could be related to the vaccinations, medical system, population's social behavior, etc. For example, meanwhile US has 5262 hospitals, Peru has 390, there is a huge difference. So, it is not adequate to use just one metric to measure the pandemic impact, we need to evaluate other metrics in order to understand this COVID-19 pandemic.

Brazil represents another interesting case. Brazil, has the highest public cost of health services in Latin America but it has 2.48% of death rate (the highest). The Brazil's president played a key role in the severity of the virus, at the beginning of the pandemic, he overestimated the virus.

#### **2.2 Latin American countries**

Additionally, in **Figures 3** and **4**, we shows the confirmed cases and deaths in Latin American countries. In this case, Mexico has an interesting behavior, this country got a death rate of 8.85 (see **Table 1**). According to some researches, the severity of COVID-19 is positively correlated with several factors, such as age and coexisting diseases. Moreover, obesity is considered as the main risk factor [11–13].

Obesity, is the key problem in Mexico. According to some surveys in 2000, 2006, 2012, and 2018, the adult obesity increased 42.2%. Moreover, the latest national survey (2018), concluded that 40.2% female adults and 36.1% male adults

#### **Figure 2.**

*Evolution of confirmed and deaths per million habitants in US, India, France, Russia and Brazil. (a) Confirmed cases. (b) Deaths.*


#### **Table 1.**

*Death rate for US, India, France, Russia, Brazil, Argentina, Colombia, Peru, Chile and Mexico.*

**Figure 3.**

*Evolution of confirmed and deaths in Argentina, Colombia, Peru, Chile and Mexico. (a) Confirmed cases. (b) Deaths.*

#### **Figure 4.**

*Evolution of confirmed and deaths per million habitants in Argentina, Colombia, Peru, Chile and Mexico. (a) Confirmed cases. (b) Deaths.*

suffer from obesity. More alarming, only 23.5% of the adult population had a healthy weight (*BMI* <sup>&</sup>lt;<sup>¼</sup> <sup>25</sup>*kg=m*2) [14].

#### **3. Vaccination against COVID-19**

In this section, we review the main COVID-19 vaccination projects. The effects of virus variants and the impact of vaccination in US, India, France, Russia, Brazil, Argentina, Colombia, Peru, Chile and Mexico.

#### **3.1 COVID-19 variants**

Unfortunately, like other viruses, COVID-19 virus evolves over time. Normally, when the virus replicates, it makes copies of itself with little changes (mutations), a virus with one or more mutations is call a"variant" of the original virus. Moreover, the US government inter agency group developed a Variant Classification scheme: Variant of Interest (VOI), Variant of Concern (VOC) and Variant of High Consequence (VOHC) [15]. In **Table 2**, we describe each variant.

In **Table 3**, we resumed the VOI variant of COVID-19, some of them present a reduced neutralization by antibody treatments and convalescent and postvaccination sera [16–19]. In **Table 4**, VOC variant are presented, for instance B.1.1.7 and B.1.351 have approximately 50% increased transmission [20, 21].


#### **Table 2.**

*COVID-19 variant classification proposed by US government inter agency group.*


#### **Table 3.**

*COVID-19 VOI variants detected.*


#### **Table 4.**

*COVID-19 VOC variants detected.*

Variants B.1.427 and B.1.429 have 20% increased transmissibility [22]. Moreover, all of VOC variants presents a reduction in neutralization by convalescent and post-vaccination sera. In order to see a detailed description of each variant, visit: SARS-CoV-2 Variant Classifications and Definitions [15].
