Introduction

Corona Virus disease (COVID-19) is a disaster, speedily dispersion, and affected death worldwide. It is a breathing infection that can extent from individual to the individual causing virus-like indicators and in severe cases death. It was first reported in China and the World Health Organization affirmed the occurrence of an epidemic but it has now spread throughout the universe. ^[1,2]

COVID-19 impurity is rumored to an extent through breathing dewdrops, contact with bodily fluids, and with contaminated surfaces. ^[3] COVID-19 epidemic spreads from individual to individual from through infected air precipitations, people contact with fingers or faces that contain the virus and touch their eyes, nose, or mouth with the unclean hands.

Africa recorded 23,209 Corona Virus deaths and the confirmed cases also rise to 1,045,234 (Africa CDC, 10 August 2020). Ethiopia is currently one of the most impacted nations relative to the proportion of COVID-19 case development and contagion. A state of emergency lasting five-months has been imposed by the Prime Minister of government. All land borders have since been shut as a virus control measure with a number of restrictions and implementation of physical distancing and mask-wearing measures. But the pandemic is rising due to temperature change (summer season) and ignoring the ministry of health rules about controlling the COVID-19 pandemic. Ethiopia recorded 407 Corona Virus Deaths since the epidemic began, according to Ethiopia ministry of health (9 August 2020). In addition, Ethiopia reported 22,818 Corona Virus Cases.

The main restriction to speechless the matter is the deficiency of potential injections to treat or control the disease. The aim of this investigation was to assess the risk factors of COVID-19 disease in Ethiopia using multiple linear regression models.

Subjects and Methods

Source of the Data

The source of the data for this study was the 2020 Ethiopia health ministry from March up to July.

Variables in the study

The response variable for this study is the number of deaths on COVID-19 per month. The predictor variables are the number of new confirmed cases, recovered, critical, total tests, and average temperature.

Methods of data analysis

Multiple linear regression model is a regression model with one response or regressed variable and two or more than two explanatory or predictor variable. The general form of multiple linear regression models is as follows:

Y= ${\beta }_0$ + ${\beta }_1{x}_1$ + ${\beta }_2{x}_2+$ … + ${\beta }_p{x}_p$ + $\epsilon$ ,

Where, Y is response variable, X₁, X₂, …, X_p are the explanatory variables, p is the number of explanatory variable and $\epsilon$ is the stochastic disturbance term.

Model assumptions

The technique of least squares can be used to estimate the regression coefficients. Analysis of variance (ANOVA) is the technique of decomposing the total sum of squares into its components. i.e., the technique decomposes the total variation in the dependent variable into the explained and unexplained variations. A test of model adequacy is accomplished by testing the null hypothesis:

H₀: β1 = β2 = ……= βp= 0

H_A: H₀ is not true

The test statistic is given by:

F_cal= $\frac{\frac{RSS}{P}}{\frac{ESS}{n-P-1}}$

Where, n is the sample size, RSS is regression sum square and ESS is error sum of square. We say that the linear model is adequate in explaining the relationship between the response variable and one or more of the predictor variables if:

F_cal > F_α (P, n-P-1)

Tests on the regression coefficients

To test whether each of the coefficients is significant or not, the null and alternative hypotheses are given by:

H₀: β_j =0

H_A: β_j ≠ 0 for j=1,2,….,p.

The test statistic is:

Decision rule

If | t_j | > t_α/2 (n-pi1), we reject H₀ and conclude that β_j is significant, that is, the regressor variable X_j , j = 1,2,….,p, significantly affects the dependent variable Y.

Results

As shown in [Table 1], among 468,814 total tests, 9,027 recovered, 20,900 confirmed cases, and 365 deaths from Coronavirus diseases in Ethiopia.

By using the backward selection method; Temperature, total tests, and critical or serious cases were the most important variables.

Based on [Table 2], p-value (0.002) < 0.05 (5% level of significance). Thus, the linear regression model is adequate.

As shown below in [Table 3], Critical cases and average temperature per month significantly affect the number of deaths at the 5% level of significance. Critical or serious cases significantly and positively affect the number of deaths per month, while average temperature significantly and negatively affects the number of deaths in COVID-19. The estimated coefficient of the critical cases is 0.570. Holding average temperature and total tests constant, a one-person increase in critical cases results in a 0.57 units increase in the number of deaths. The estimated value of deaths decreases by 35.1 units for each increase of 1 degree Celsius of average temperature, holding critical cases, and total tests constant.

Discussion

The principal purpose of this research was to identify the risk factors of COVID-19 in Ethiopia based on the 2020 Ministry of health data. As the results revealed, average temperature per month was found to be an important predictor for the deaths of COVID-19 in Ethiopia. Based on the data, if we went from dry season to summer season, it increased the Covid-19 deaths. That means the average temperature in June and July was cold (low) as compared to other months and occurred high death rates in the summer season. Researchers revealed that the low average temperature was safe (not harmful) to the virus.^{[4,5,6,7,8,9]}

According to the results, the critical case was a significant determinant of COVID-19 deaths. Anyone who has entered the serious cases was old age and living other infected diseases such as hypertension, HIV aids, sugar diseases, people who have breathing problems, and so on. The finding of this study indicated that as the number of critical cases increases the number of passing away in the COVID-19 virus also rises. A high critical case was elevated the COVID-19 deaths. This result is consistent with the findings by. ^[10,11,12]

Conclusion

The vital of this investigation was to assess the factors of COVID-19 deaths. The study was based on secondary data obtained from the ministry of health (Ethiopia). The current study indicated that critical cases and the average temperature were the risk factors that influence COVID-19 deaths in Ethiopia. In order to minimize the death of COVID, We should be able to clean your hand regularly with soap and water, wearing mouth-mask, following instructions from public and global health institutions especially keep the social distance from critical cases, and stay at home mainly in the summer season.