key: cord-0723377-86gsai03
authors: Maiga, Kulwa; Hugo, Alfred
title: Modelling the impact of health care providers in transmission dynamics of COVID-19
date: 2022-04-29
journal: Results Phys
DOI: 10.1016/j.rinp.2022.105552
sha: b21271c952fe46c5a4fa8a995b1c0aa14bf12abe
doc_id: 723377
cord_uid: 86gsai03

In this paper, a mathematical model is proposed and analysed to assess the impacts of health care providers in transmission dynamics of COVID-19. The stability theory of differential equations is used to examine a mathematical model. The results of both local and global stability of disease-free equilibrium points were determined by using Routh–Hurwitz criteria and Metzler matrix method which verified that was locally and globally asymptotically stable. Also, the endemic equilibrium point was determined by the Lyapunov function which showed that [Formula: see text] was globally asymptotically stable under strict conditions. The findings revealed that non-diagnosed and undetected health care providers seems to contribute to high spread of COVID-19 in a community. Also, it illustrates that an increase in the number of non-diagnostic testing rates of health care providers may result in high infection rates in the community and contaminations of hospitals’ equipment. Therefore, the particular study recommend that there is a necessity of applying early diagnostic testing to curtail the COVID-19 transmission in the health care providers’ community and reduce contaminations of hospital’s equipment.

The ongoing Corona virus disease-2019 (COVID- 19) was reported in the mid-December-2019

in Wuhan City at Hubei Province, China causing a severe acute respiratory syndrome corona virus-2 (SARS-COV-2), where the first infected patient was identified after exposure into the seafood market [1, 2] . The existence of this pandemic disease in the world affected many people without regarding their economic and development status.

The World Health Organisation in its 6 th September, 2021 report showed that more than Furthermore, the corona virus spreads through hospital's equipment due to insufficient precautions taken by health care providers' population resulted from the unavailability of equipment and poor facilities.

The infected individuals with COVID-19 expressing several signs of the disease. The foremost clinical symptoms are high fever, coughing, sneezing, headache, hemoptysis, diarrhoea, dysphonia, heart failure, and respiratory failure [6, 7] . Since COVID-19 affecting many countries in the World, then East Africa as among the victims struggled to prevent further disease transmission by insisting people wearing face masks, using sanitizers, practising social distancing, introducing lock-down, and suspending all activities involving gathering such as closing learning institutions Mumbu & Hugo [1] .

Various mathematical epidemiologists have been proposed different studies targeting the transmission dynamics and control the spread of COVID-19 in a population by introducing the best control measures such as personal protective equipments, face masks, disinfectants, sanitizers, and applying physical distancing but still the disease affects many people around the world.

Some of the mathematical model includes:

The SEIRD model proposed by Viguerie et al [8] applied partial differential equations to simulate the spread of COVID-19. The results inform the health authority in Italy of the best way of allocating medical resources and effective control measures regarding geographical location.

However, Mumbu & Hugo [1] developed the SMUEIHR model for the human population, by comparing the use and disuse of masks and observed that before and after introducing wearing face masks and hospitalization the basic reproduction number was R 0 = 3.8 and R 0 = 0.698 respectively. Hence from the values of R 0 , the researchers recommended that wearing face J o u r n a l P r e -p r o o f Journal Pre-proof masks is mandatory whenever contacted with the crowded population as preventive measures for control spreads of COVID-19 transmission in Tanzania.

Besides, Baek et al [9] developed SEIR containing compartments of Doctors, Nurses, Patients, and Caregivers model for assessing the effectiveness of early screening and quarantined in tertiary hospitals to identify infected patients. The results showed that early screening and quarantined in South Korea reduced the number of cases from 81.3% to 42% within 60 days after implementation. The study insists on implements early screening to all visitors and early stage of exposure to infections.

Furthermore, Mekonen et al [5] studied SEIRDM model for self protection behaviour changes from asymptotic individuals with COVID-19. The outcome showed that basic reproduction number depends on the contamination of environments and population. The study suggested that for preventing further COVID-19 transmission, the population should be aware with the environments.

Thus, the particular study needs to apply mathematical techniques to analyse the preventive measures required for minimizing the transmission dynamics of COVID-19 in health care providers' population.

Therefore, this study intends to analyse the impacts of health care providers' population in transmission dynamics of COVID-19.

In this paper, a deterministic mathematical model on transmission dynamics of COVID-19 was formulated and analysed. The biological model consists of seven sub-compartments namely: a susceptible health care providers' population whose at high risk to be infected Alfred et al [10] when contacted with infectious denoted by S H , exposed health care providers E H are expected to be infectious with the disease but do not express clinical symptoms Zhang et al [11] . The infected class is further subdivided into two classes that are diagnosed and detected I D whose clinically identified having SARS-COV-2 infections while non-diagnosed and undetected I U health care providers' population are unidentified clinically and does not show disease symptoms due to either low virus concentrations or strong body immunity Kasali et al [12] , but treatment class It is also assumed that hospital's equipment acquired infections through interaction of nondiagnosed and undetected health care providers' population only since they are difficult to identify infected individuals. The total health care providers' population at any time t is denoted by;

Hence, the following parameters were constructed in formulating a mathematical model:

(i) Health care providers' population get exposed either directly or indirect contacts to infectious through infection rate λ = β (I D +ζ I U )

where G represents the virus concentrations and capacity of hospital's equipment to hold SARS-COV-2 [5] , β is an effective probability of infection rate, β w is probability of infection rate from C H class, ζ is a modification of parameter in reduction of disease transmission from non-diagnosed and undetected health care providers [5, 12] . (iii) However, recruitment rate Π represents individuals visited at hospitals for medical treatment and consultations while µ is a natural mortality rate occurred for each class of health care providers' population.

(iv) The susceptible health care providers' population increases at the rate of individual waned their body immunity γ, recovered after using medications σ and diagnosed but not having infections rate θ . Furthermore, some of non-diagnosed and undetected having infection recovered at the rate α and the rest who undergoes severe clinical symptoms seeks medications at the rate η while all severe infected individuals are treated at the rate ρ.

(v) The COVID-19 disease induced death rates τ and ν at diagnosed and detected, nondiagnosed and undetected classes respectively.

(vi) The proportion rates of diagnosed and non-diagnosed with respect to exposed health care providers' population E H is ε and 1 − ε respectively, while for those detected with COVID-19 infections seek treatments at the rate ω 

For non-negative initial conditions we have:

In this subsection, the model system (2) is biologically well defined if all its solutions exists and is bounded within the positive invariant region such that;

The solution of model system (2) lying within R 7 + are uniformly bounded in the invariant region Γ for ∀t 0.

Γ={S

for ∀t 0 Then, by differentiating equation (1) with respect to time t and substituting model system (2) of the first six equation, we obtain;

Simplifying equation (3) and assuming that there is a well protections and control measures observed by health care providers' population, then no infections induced by COVID-19 disease.

J o u r n a l P r e -p r o o f

It follows that solution of equation (6) is given as;

Hence as t −→ ∞ in 5, we obtained;

From equation (6), the health care providers' population implies that Γ is positively invariant bounded region of solutions in R 6 + such that;

However, for the virus concentrations on hospital's equipment class, we have;

By using the concept of Gronwall inequality in 9 as applied by [5] , for 0 C H (0) ≤ Πδ µq , we get;

Hence, the virus concentrations in the hospital's equipment are also bounded in the invariant region positively such as:

Therefore, combining the solutions of 7 and 11, the model system (2) is epidemiologically well defined in the bounded invariant region Γ

In this section, we determine the equilibria points of the model equation, reproduction number, and analysis of the model.

The model 2 posses the following equilibria points:

(i) Disease free equilibrium point; (2) is expressed in terms of I * D at steady state, Then we have;

where:

is the endemic equilibrium point of the system (2).

The basic reproduction number is used by the researchers to determine whether the disease exists or cleared out in the entirely community. The basic reproduction number is used by the researchers determine whether the disease it will either continue existing or dying out in the community. In this, model system (2), to compute the basic reproduction number, we used the approach as applied by [2, 13, 14] . Hence, generation of new infections is denoted by matrix F and matrix V represents the disease transfer among compartments determined at disease free equilibrium that are given as follows; 

Thus, basic reproduction number is the largest eigenvalue of the next generation matrix, in which after computation and further simplifications, we get:

Biologically, the basic reproduction number R 0 shows that new number of infected health care providers' population with COVID-19, it depends on β while keeping constants other parameter values:

If β < (θ + µ + τ + ω) (µ + η + α + δ ) , then R 0 < 1 and health care providers' population and hospital equipments will be free from COVID-19 infections.

If β > (θ + µ + τ + ω) (µ + η + α + δ ), then R 0 > 1 and the disease will invade and spreads in health care providers' population which results to virus concentration on the hospital's equipment.

To determine the local stability of disease free equilibrium point of the model system (2), we applied the linearisation techniques to formulate a Jacobian matrix with corresponding to disease free equilibrium point, E 0 . However, in this study, we used the concept of Routh-Hurwitz criteria to prove local stability of the disease free equilibrium as applied in the work of [12, 16] .

Theorem 2 If R 0 < 1, then the disease free equilibrium point E 0 of the model system (2) is locally asymptotically stable, otherwise unstable.

We need to attest that the real part of the eigenvalues of the Jacobian Matrix J at disease free equilibrium point E 0 has negative sings.

Hence, we have a Jacobian matrix J at disease free equilibrium point E 0 given as:

From matrix (11), we obtained the real parts of eigenvalues with negative sign as follows:

Also, the remaining eigenvalues are determined by using matrix block reductions as applied in [10, 15] .

However, after cancellations of the first, fifth and sixth rows and columns respectively, the matrix (14), reduced to a 4 × 4 matrix. Thus, the reduced Jacobian matrix J E 1 0 is given by;

Now to obtain the eigenvalues of matrix (14), we determine its determinant as follows:

Hence, characteristic equation J E 1 0 is a fourth order of degree polynomial which is written by;

Journal Pre-proof where, A = a 1 + a 2 + a 3 + a 4 , B = β ε φ ζ − β φ (ζ + ε ) + a 2 a 1 + a 3 a 1 + a 4 a 1 + a 3 a 2 + a 4 a 2 + a 4 a 3 ,

Then B > 0 if; β ε φ ζ + a 2 a 1 + a 3 a 1 + a 4 a 1 + a 3 a 2 + a 4 a 2 + a 4 a 3 > β φ (ζ + ε ) .

C = a 1 a 2 a 3 − a 1 a 2 a 3 R 0 − β φ ζ a 4 + a 1 a 2 a 4 + a 1 a 3 a 4 + a 2 a 3 a 4 − Π δ φ β w a 2 (1 − ε ) µG , a 1 a 2 a 4 + a 1 a 3 a 4 + a 2 a 3 a 4 > a 1 a 2 

Since A, B,C, D > 0; then we need to determine the necessary conditions for AB − C and ABC −C 2 − A 2 D to be greater than zero.

Consider AB −C, then; a 3 ) ,

β εφ ζ (a 1 + a 3 + a 4 ) + a 2 1 a 2 + a 2 1 a 3 + a 2 1 a 4 + a 1 a 2 2 + 2a 1 a 4 (a 2 + a 3 ) + 2a 2 a 3 (a 1 + a 4 ) + a 1 a 2 3 + a 1 a 2 4 + a 2 2 a 3 + a 2 2 a 4 + a 2 a 2 3 + a 2 a 2 4 + a 2 3 a 4 + a 3 a 2 4 > β εφ (a 1 + a 2 + a 4 ) + β φ ζ (a 1 + a 3 ) .

; 1 a 2 a 3 + a 1 a 2 a 4 + a 1 a 3 a 4 + a 2 a 3 a 4 ) + (β εφ (a 1 + a 2 + a 4 ) 1 a 2 a 3 + a 1 a 2 a 4 + a 1 a 3 a 4 + a 2 a 3 a 4 ) − a 1 a 2 a 3 a 4 (a 1 + a 2 + a 3 + a 4 ) 2 ,

So ABC −C 2 − A 2 D is to be greater than zero if; m (a 1 a 2 a 3 + a 1 a 2 a 4 + a 1 a 3 a 4 + a 2 a 3 a 4 ) + (β εφ (a 1 + a 2 + a 4 ) 1 a 2 a 3 + a 1 a 2 a 4 + a 1 a 3 a 4 + a 2 a 3 a 4 ) + a 1 a 2 a 3 a 4 (a 1 + a 2 + a 3 + a 4 ) 

where, a 1 = φ + µ, a 2 = θ + µ + τ + ω, a 3 = µ + η + α + δ , a 4 = q, m = β εφ ζ (a 1 + a 3 + a 4 ) + a 2 1 a 2 + a 2 1 a 3 + a 2 1 a 4 + a 1 a 2 2 + 2a 1 a 4 (a 2 + a 3 ) + 2a 2 a 3 (a 1 + a 4 ) + a 1 a 2 3 + a 1 a 2 4 + a 2 2 a 3 + a 2 2 a 4 + a 2 a 2 3 + a 2 a 2 4 + a 2 3 a 4 + a 3 a 2 4 .

Since, eigenvalues of matrix (14) have negatives real parts, and the necessary and sufficient conditions for Routh-Hurwitz criteria are met.

Therefore, the stability of disease-free equilibrium point, E 0 is locally asymptotically stable whenever, R 0 < 1, otherwise unstable. (2) is said to be globally asymptotically stable whenever R 0 < 1 at disease free equilibrium point E 0 , otherwise it is unstable within invariant region Γ.

To examine the global stability of disease free equilibrium point E 0 , we use the method done by [17, 18] . Hence, we rewrite our model system (2) as:

where, Y 1 = uninfected classes of the model equation (2) Y 2 = infected classes of the model equation (2) Y * 1 = represents disease free equilibrium point E 0 of uninfected classes of model equation (2) We define, J o u r n a l P r e -p r o o f Journal Pre-proof

Thus, theorem 2 is epidemiologically well posed if the following conditions holds the global stability of disease free equilibrium points E 0 :

i. A 1 must be a matrix with real negative eigenvalues on the main diagonal.

ii. A 2 must be a Metzler matrix (off diagonal matrix must have non-negative element).

Using equation (17)and (18), we obtained the following results:

J o u r n a l P r e -p r o o f Journal Pre-proof

Since matrix A 1 and A 2 satisfies the necessary conditions (i) and (ii) respectively, we conclude that model system (2) at disease free equilibrium points E 0 is globally asymptotically stable which makes epidemiologically meaningful in the invariant region Γ.

Global stability analysis of endemic equilibrium point.

In this paper, the global stability of the endemic equilibrium point E * is determined by using the Lyapunov function as applied by Ngalya and Kuznetsov [19] . Therefore, to obtain global stability E * , we need to prove that Lyapunov function L is globally asymptotically stable contained in the invariant region Γ.

If R 0 ≥ 1, then endemic equilibrium point E * of the model system (2) is globally asymptotically stable defined in the invariant region Γ otherwise unstable.

Using the Lyapunov function L, we prove that endemic equilibrium points E * is globally asymptotically stable if R 0 ≥ 1.

Thus, we define the following Lyapunov function L as:

where, 

To examine the impact of health care providers' population in transmission dynamics of COVID-19 disease. We simulate the mathematical model (2) using the Matlab programming language and the set of parameter values obtained from recently published articles while unknown parameters were assumed under epidemiological reasoning as indicated in Table 1 . Since, the researchers did not go to the data collections, then the assumed initial data were used in the model simulations basing on the real environments of health care providers' population in transmission dynamics of COVID-19. 

The implementation of mass diagnostic testing to susceptible health care providers' population is an important approach that can help health authorities and stakeholders in general to control the transmission of COVID-19 disease in our community. On the other hand, results obtained by Mumbu & Hugo 2020 [1] showed that non-wearing face masks of infected individuals increase the human-to-human transmission of COVID-19

infections in a population. However, these findings concurred with the study done by [5, 9, 22, 24, 25] which showed that implementing of early screening helped to detect all infected individuals and reducing further transmission of COVID-19.

Thus, from the results obtained we can recommend that all hospitals should introduce prediagnosis testing for all health care providers before and after medical consultation of any individuals.

Basing on the analytical and simulation results obtained from this study, it concluded that nondiagnosed and undetected health care providers' population seems to contribute to high spread of COVID-19 pandemic in community and contamination of hospitals' equipment; since, most health care providers have indirect interactions with contaminated hospitals' equipment at the time of implementing medical consultations.

Hence, the study proved the necessity of implementing early diagnostic testing and taking med- 

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. 

The authors declares that, there are no conflicts of interest regarding the publication of this paper.

The secondary data supporting this research article are obtained from the previously reported studies and included within the article.

Mathematical modelling on COVID-19 transmission impacts with preventive measures: a case study of Tanzania

Modelling the effective reproduction number of vector-borne diseases: The yellow fever outbreak in Luanda

Healthcare worker infections and deaths due to COVID-19: A survey from 37 nations and a call for WHO to post national data on their website

World Health Organisation (WHO) (2021) Corona virus disease-2019 (COVID-19)

Weekly Operational Update on COVID-19

Modeling the effect of contaminated objects for the transmission dynamics of COVID-19 pandemic with self protection behavior changes

The epidemiology and pathogenesis of coronavirus disease (COVID-19) outbreak

World Health Organization declares global emergency: A review of the 2019 novel coronavirus (COVID-19)

Simulating the spread of COVID-19 via a spatially-resolved susceptible-exposed-infected-recovered-deceased (SEIRD) model with heterogeneous diffusion

A mathematical model of COVID-19 transmission in a tertiary hospital and assessment of the effects of different intervention strategies

Modelling transmission dynamics of coronavirus with infected immigrants in Tanzania

Crowding effects on the dynamics of COVID-19 mathematical model

Modelling the Dynamics of COVID-19 in Nigeria

Reproduction numbers and sub threshold endemic equilibria for compartmental models of disease transmission

Reproduction numbers of infectious disease models

Bifurcation and Chaos Analysis of Vector-Borne Disease Model with Application to Rift Valley Fever

An eco-epidemiological mathematical model with treatment and disease infection in both prey and predator population

Modeling the Dynamics of Rabies Transmission with Vaccination and Stability Analysis

Computation of threshold conditions for epidemiological models and global stability of the disease-free equilibrium (DFE)

Modelling the impact of Bemisia tabaci in dynamics of tomato yellow leaf curl virus

The stability of dynamical systems, CBMS-NSF regional conference series in applied mathematics 25

Design of a non-linear model for the propagation of COVID-19 and its efficient non standard computational implementation

Mathematical modeling of the spread of the corona virus disease 2019 (COVID-19) taking into account the undetected infections. The case of China

Mathematical model of COVID-19 transmission dynamics in South Korea: The impacts of travel restrictions, social distancing, and early detection

Modeling behavioral change and COVID-19 containment in Mexico: A tradeoff between lockdown and compliance

Computational simulation of the COVID-19 epidemic with the SEIR stochastic model. Computational and Mathematical Organization Theory

The authors would like to acknowledge the materials used to prepare this manuscript by citing accordingly.

Differentiating equation (23) with respect to time t, we get: (24), gives the following results:Moreover, using the endemic equilibrium point, E * we get:Again further simplifications, and collecting negative and positive terms together, we obtained the following results:which can be written as;where, Therefore, by considering the La Salle's invariant principle [20] , equation (28) is strictlyLyapunov function if and only if dL dt < 0 which signifies the endemic equilibrium point, E * is globally asymptotically stable whenever R 0 ≥ 1 and otherwise unstable within the invariant region Γ. Epidemiologically, this informs that COVID-19 infections will be endemicity in health

The authors declare no potential conflicts of interest regarding this work.