Tuberculosis Prevalence Correlation to COVID-19 Mortality in Malaria Free Countries

Background: Both malaria and latent tuberculosis ( LTB) are possible factors related to decreased COVID-19 mortality. The malaria endemicity variable is a possible confounder when conducting a study on the correlation of LTB prevalence to COVID-19 mortality. Studies regarding LTB prevalence" according to different studies" did not adjust malaria endemicity as a possible confounder. Many malaria-endemic countries are high TB prevalent. Malaria-free countries could be: high, moderate, or low in TB prevalence. The main aim of this study is to look for the influence of TB prevalence on COVID-19 mortality. TB prevalence reflects LTB prevalence in the absence of malaria endemicity as a possible confounding factor in TB studies. Material: and methods: The total chosen countries were 69 non-malaria endemic countries. Countries were classified according to TB prevalence groups into low, moderate, and high prevalent groups. Covid-19 deaths/Million(M) inhabitants were taken as reported on September 2, 2020. "Kendall's-τ Correlation Coefficient", "Kruskal-Wallis test, and Mann-Whitney test were used in statistical analyses. Results: : We found inverse relationships between TB prevalence and COVID-19 deaths/ (M) inhabitants and a highly positive significant correlation coefficient was reported (0.008) in Kendall's-τ correlation coefficient test. Kruskal-Wallis test showed a significant relationship within studied groups. Furthermore, the low TB prevalent group had significant reverse associations with both high and moderate TB prevalent groups in the Mann-Whitney test. Conclusion: In the absence of possible malaria confounding, TB prevalence in malaria-free countries is inversely related to COVID-19 mortality in a highly significant association.


Introduction
Microorganisms may modulate hosts' protective heterologous cross-immunological reactions, which can last for a long time. This cross-protection can be achieved by inducing the training of innate immune cells. This promotes host resistance against a wide spectrum of pathogens. TB prevalence in countries re ects the LTB prevalence as far as TB prevalence constitutes approximately one-tenth or one-twentieth of TB prevalence. , . The incubation period for TB infection is very rarely more than 2 years after infection.. , On the other hand, if a person crosses these 2 years without showing manifestation of the disease, immunity "generated by LTB " shows long-time positivity making a person with previous LTB immune from getting the disease . This immunity can be tested by speci c immunologic tests that did not differentiate past infections from recent ones since it did not indicate the presence of live bacteria. This TB immunoreactivity can persist whether a patient is treated by curative treatment or not. 8 Studies show that LTB infection prevalence is associated with a decrease in COVID-19 mortality. These studies suggested a heterogeneous immune response to COX-COV2 viral infection. ,,, . Despite this signi cant association, some countries are hugely disparate case fatality rates among low % LTBI countries. Also, disparity exists in case fatality rates among high % LTBI countries. This disparity existed despite that these studies have suggested a role of BCG vaccination as far as BCG is a form of mycobacterial species, has heterogeneous immune effects, and was suggested to have an added effect to LTB effect.
These studies did not consider the possible role of malaria in explaining these disparities. We consider malaria's role in this study and restrict the study of TB prevalence to malaria-free countries.
Malaria: Early in this pandemic, malaria-endemic areas were suggested to have a low risk for catching COVID-19. 13,,, Past malaria exposure was suggested to attenuate monocyte-associated immunopathology induced by SARS-CoV-2 through inducing epigenetic reprogramming of monocytes/macrophages toward a regulatory phenotype that mitigated in ammatory responses. In the same context, another 2 studies founde a signi cant correlation between malaria endemicity and COVID-19 mortality. ,

Material and methods
Patients or the public were not involved in this work, given that we used related publically published morbidity and mortality statistics. Countries that have achieved at least 3 consecutive years of zero indigenous cases of malaria up to 2019 were selected. The total chosen countries were 69, as shown in (Appendix A) attached to this manuscript. These countries may also eligible to apply for a WHO We used the following methods: "Kendall's-τ Correlation Coe cient", "Kruskal-Wallis test, and Mann-Whitney test". Also we used "Stem-Leaf" plot for exploring data behavior concerning COVID-19 mortality/ (M) inhabitant in relation to different TB prevalence groups.
All statistical operations were performed through using the ready-made statistical package SPSS, ver. 22.

Results And Findings
Table (1) represents "Kendall's-τ Correlation Coe cient" and P-values for studied the amount and the direction concerning relationships among the studied markers.
Results showed there was a strong relationship among studied markers (Covid-19 deaths/M and TB prevalence) as the number of deaths increased with lowering the prevalence of tuberculosis and too highly positive signi cant correlation coe cient was reported ( p-value = 0.008). This indicates that the research hypothesis is ful lled in general with a highly signi cant p-value. Table (2) represents summary statistics of (COVID-19 deaths/M) among studied groups. These statistics include 5% trimmed mean, median, minimum, maximum readings, range, and interquartile range).
Maximum gures of mortalities were among the low TB prevalence group and minimum within the high TB prevalence group. Furthermore, the 5% trimmed mean is shown to be higher within low TB prevalence group compared to moderate and high TB prevalence groups. Kruskal-Wallis Test showed a p-value of 0.021 which is less than the signi cance level of 0.05, hence we rejected the null hypothesis and conclude that the medians are not all equal. Signi cant comparisons among TB prevalence categories using the Mann-Whitney test were signi cant regarding the relations between a low TB prevalence group with both groups high TB prevalence and moderate TB prevalence. The shape of the distribution. of the recorded COVID-19 deaths within TB prevalence groups is well shown in stem-leaf graphical plots ( gure 1). This illustration shows clearly how COVID-19 deaths were highest within the low TB prevalence group and lowest within the highest TB prevalence group. In the moderate TB prevalence group. The mortality was between the other two groups.

Discussion
A confounding factor (third variable) may mask an actual association or falsely demonstrate an apparent association between the study variables where no real association exists. If confounding factors are not measured and considered, bias may result in the conclusion of the stud. Confounding occurs when a measure designed to assess a particular construct inadvertently measures something else as well. [22] The confounding factor can interfere with the real effect. For this reason, the etiological importance of a variable needs to be prevented or removed as much as possible. Confounding may be prevented by the use of randomization, restriction, or matching. [24] The term restriction is used when a researcher chooses only one variation of a participant to restrict the possibility of including a confounding variable. As far as malaria is a possible confounding factor suggested to create a cross heterogeneous effect towards COX-Cov2, We designed this study to exclude this factor by restricting the study sample to malaria-free countries. Literature raised the question of unknown contribution factors regarding COVID-19 mortality variances. [25] Some of the literature raised the possibility of malaria as one factor in explaining the variances in COVID-19 mortalities. 13,15,16,17 ,18 , 1 9,20 . One of these studies found that malaria has an added effect on TB prevalence effect in decreasing the COVID-19 deaths in malaria-endemic areas. 1 9 Countries are classi ed as either malaria-free or malaria-endemic countries. All malaria-free countries have a certain TB prevalence. In general, countries might be classi ed as low, intermediate, and high TB incidence countries. Malaria endemic countries are usually highly endemic in TB. [25] , [26] This might lead to confounding when testing relation of TB or malaria against COVID-19 mortality.
Previous studies testing TB in uence on COVID-19 mortality fail to control this factor.
In this study by using Kendall's-τ correlation coe cient test, there was a high reverse-directional signi cant correlation between COVID-19 mortality rates and TB prevalence (in absence of suggested confounded factor of malaria endemicity) with a reported p-value (0.008) (table 1).
Kendall's τ has been classically used to test the signi cance of cross-correlation between two variables when their distributions signi cantly deviate from the normal distribution. In that case, a signi cance test based on the distribution-free τ, which is a function of the ranks of the variates rather than their actual values, offers more power than other parametric tests. [27] The nding in this study of TB prevalence in uence on COVID-19 mortality in malaria-free countries is strongly in agreement with a previous study in malaria-endemic countries. 19 Furthermore, it consolidates other studies on the relation of LTB to covid-19 mortality. 10,11,12,,13 ,14,[28] .
A second test conducted in this study was the Kruskal-Wallis Test. It is a non-parametric method used for testing whether samples originate from the same distribution or not by determining whether the medians of two or more groups are different. [29] , [30] It was conducted among different TB prevalence groups. It showed that the Chi-Square result equals 7.740 with P-value equals 0.021 (S)These results signify that there are differences among the groups, but as Kruskal-Wallis Test is an omnibus test statistic and doesn't tell which group is different from other groups. It is used for comparing two or more independent samples of equal or different sample sizes. The Kruskal-Wallis test is a rank-based test that is similar to the Mann-Whitney U test but can be applied to one-way data with more than two groups. It is a nonparametric alternative to the one-way ANOVA test, which extends the two-samples Wilcoxon test.
However, like most non-parametric tests, the Kruskal-Wallis test is not as powerful as the ANOVA but, assumptions of one-way ANOVA are not met in our sample.
A Mann-Whitney U test (another non-parametric test) was used to compare the differences between two independent samples as far as the sample distributions are not normally distributed as shown before.
It showed that the low TB prevalent group when tested against groups moderately TB prevalent and highly prevalent group show signi cant association with decreased mortality from covid-19 (table 2).
The non-signi cant association between moderate and high TB prevalent groups (table 2) needs further consideration. The test fails to nd a signi cant association due to possible existing confounding factors. A possible one is the malaria elimination date since faster progress was achieved in malaria elimination recently, compared to TB control in many countries. According to WHO, the malaria elimination net is widening. Furthermore, more countries are moving towards zero indigenous cases: The number of countries with fewer than 100 indigenous cases was 17, 25, and 27 in 2010, 2017, and,2018 respectively, which is a strong indicator that elimination is within reach. . [31] Despite other signi cant associations, some countries are disparate fatality rates / M inhabitants among low TB prevalence countries. Also, a disparity exists in fatality rates among high TB prevalence countries. We suggest other possible confounding factors in addition to malaria residual immunity already mentioned which include but are not restricted to BCG policy of country and BCG coverage, other mycobacterial cross-reaction or effects by other vaccines population size measures taken, habits, some LAVs which have induced a broad, nonspeci c, protection against unrelated pathogens and decreased mortality from conditions other than the targeted infectious diseases. [32]. Furthermore, this study was done without any regard for income, healthcare facilities.
Controlled clinical studies need to be conducted before further considering reviewing global strategies for the prevention and treatment of TB and malaria. According to global strategy in the treatment of TB, the main goal is to treat the active cases in areas with a high incidence of TB, but in areas with a low incidence of TB, the goal also includes prophylactic treatment for LTBI [33] Whoever, in recent years, studies have gradually narrowed down to the preventive treatment of LTBI for high-risk target groups. Targeted TB testing and treatment programs in USA and many European countries conducted among high-risk groups. [34] , [35] . It was questionable that chasing after LTB infection treatment a time before [36].If prophylaxis is provided for all LTBI patients, it will result in an enormous waste of resources and increase the likelihood of anti-TB drug resistance. 33 Added to that we raised the possible role of LTB on decreasing COVID-19 mortality.
TB prevalence in this study was taken as the highest available gure for up to 10 years ago (since 2011 outward) as far as the immunity created by latent or active TB infection last for a long time. The major dilemma is that there is no test to assure that every person diagnosed with an immune reaction to LTB treatment can be guaranteed free from the active form of infection, although it has been agreed that reactivation is unlikely after 2 years.
We considered 10 years the least time for immune-reaction to wane as far as it is well known that a related Mycobacterium which is Mycobacterium Bovis ( BCG ), waned by at least by 10 years.. [37] A longer time for immune reaction time after natural infection is possible, but the exact time for waning such immunity is unknown yet.

Conclusions
TB prevalence is inversely related to COVID-19 mortality in high signi cance association.
Low TB prevalence countries have signi cantly different statistical COVID-19 mortalities in relation to both moderate and high TB prevalent countries.
Recommendations: clinical trials are recommended for any updating global strategies for the prevention and treatment of TB and malaria.

Declarations
• Ethics approval and consent to participate: 'Not applicable • Consent for publication: I certify that this study has not been previously published. The publisher has my permission to publish this study. With the consent, I give the publisher copyright license • Availability of data and materials: We used publically available data. Patients were not involved.
All data generated or analyzed during this study are included in this published article [and its supplementary information les].
• Competing interests: There is no con ict of interest to be declared.
• Funding: No funding source to be declared.
• Author contributions: The author wrote the manuscript and provided data, conducted the statistical analyses, and reviewed the nal manuscript.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. suppnonmalarious.docx