key: cord-327544-7ws2kleo
authors: Hammoumi, Aayah; Qesmi, Redouane
title: Impact assessment of containment measure against COVID-19 spread in Morocco
date: 2020-08-22
journal: Chaos Solitons Fractals
DOI: 10.1016/j.chaos.2020.110231
sha: 
doc_id: 327544
cord_uid: 7ws2kleo

Since the appearance of the first case of COVID-19 in Morocco on March, 02,2020, the cumulative number of reported infectious cases continues to increase and, up to date, the peak-time of infection is not reached yet. In this study, we propose a Susceptible-Asymptomatic-Infectious deterministic model to evaluate the impact of compulsory containment imposed in Morocco on March, 21 on the spread of COVID-19 epidemic across the country. The model takes account of the unconfined individuals that continue to work or to leave their home for urgent needs and the existence of infectious asymptomatic and unreported individuals within susceptible population. Furthermore, the model is able to predict the peak-size, peak-time, final size and epidemic duration according to different rates of containment. Advanced knowledge of these details will be of great interest to establish an optimal plan-of-action to control or eradicate the epidemic. Indeed, mitigating and delaying the epidemic peak allow the official health authorities to anticipate and control the spread of COVID-19. Moreover, prediction of the epidemic duration can help the government to predict the end time of containment to avoid consequent social-economic damages as well. Using our model, the basic reproduction number R(0) is estimated to be 2.9949, with [Formula: see text] reflecting a high speed of spread of the epidemic. The model shows that the compulsory containment can be efficient if more than 73% of population are confined. In the absence of other efficient measure of control, even with 90% of containment, the end-time is estimated to happen on July, 4,2020 with 7558 final cumulative cases. Furthermore, a threshold value of containment rate, below which the epidemic duration is postponed, has been determined. Finally, the sensitivity analysis is performed and showed that the COVID-19 dynamics strongly depends on the asymptomatic duration as well as the contact and containment rates. Our previsions can help the government to adjust its plan-of-action to fight the disease and to face the social-economic shock induced by the containment.

• The model is able to predict the peak-size, peak-time, final size and epidemic duration.

• We estimated the basic and control reproduction numbers.

• The model shows that the compulsory containment can be efficient if more than 73% of population are confined.

• We performed sensitivity analysis which shows that COVID-19 depends strongly on the asymptomatic duration as well as the contact and containment rates.

Morocco on March, 21 on the spread of COVID-19 epidemic across the country. The model takes account of the unconfined individuals that continue to work or to leave their home for urgent needs and the existence of infectious asymptomatic and unreported individuals within susceptible population. Furthermore, the model is able to predict the peak-size, peaktime, final size and epidemic duration according to different rates of containment. Advanced knowledge of these details will be of great interest to establish an optimal plan-of-action to control or eradicate the epidemic. Indeed, mitigating and delaying the epidemic peak allow the official health authorities to anticipate and control the spread of COVID-19. Moreover, prediction of the epidemic duration can help the government to predict the end time of containment to avoid consequent social-economic damages as well. Using our model, the basic reproduction number R 0 is estimated to be 2.9949, with CI (2.6729 − 3.1485) , reflecting a high speed of spread of the epidemic. The model shows that the compulsory containment can be efficient if more than 73% of population are confined. In the absence of other efficient measure of control, even with 90% of containment, the end-time is estimated to happen on July, 4, 2020 with 7558 final cumulative cases. Furthermore, a threshold value of containment rate, below which the epidemic duration is postponed, has been determined. Finally, the sensitivity analysis is performed and showed that the COVID-19 dynamics strongly depends on the asymptomatic duration as well as the contact and containment rates. Our previsions

COVID-19 is an infectious disease caused by a novel betacoronavirus that primarily targets the human respiratory system [12] . Severity of COVID-19 symptoms can range from very mild to severe. Indeed, some people generally have mild to moderate respiratory illness and recover without requiring special treatment [32] . Others, like older adults and people with existing chronic medical conditions, are more likely to develop serious illness and have a high risk of death [32] . Human-to-human transmission of disease occurs via direct contact with the droplets saliva, discharge from the nose of an infectious person's coughs, sneezes or through contaminated objects and surfaces [32, 12] . The novel coronavirus was first emerged in the Chinese city of Wuhan in December 2019. In a few months, the virus rapidly spread throughout China and was exported to 114 countries around the world [12, 35] .

The worldwide number of humans diagnosed with COVID-19 has surpassed 118,000 and nearly 4,300 people have died [1] . On March 11, 2020, the World Health Organization (WHO) has officially declared the novel coronavirus outbreak a pandemic [2] . To the best of our knowledge, no specific vaccine or antiviral exist for COVID-19 up to date, but there are many ongoing clinical trials evaluating potential treatments [14] . Meanwhile, classical public health measures such as isolation and quarantine, social distancing and community containment were adopted by many countries to prevent dissemination of the disease within their populations and to curb the epidemic [1, 35] . The containment measure, defined as an intervention applied to an entire community in order to lower intermixing of unreported infectious individuals with susceptibles as well as the spread of the virus [35] , ranges from social distancing to community-use of face masks, including locking entire cities or areas.

Morocco is one of the countries affected by COVID-19. As of April 09, 2020, the total number of confirmed cases in Morocco was reported to be 1374 infectious and 97 deaths [26] . becoming ill are identified and tested [17, 18, 19, 20, 21, 22] . Since seven imported cases have emerged, Moroccan government decided to suspend flights to neighboring countries affected by COVID-19 such as Italy, Spain, Algeria and France [29] . However, reported cases continued to grow with eight new imported cases and a first confirmed local case appeared on March 14 [23] . The epidemic spreads to other Moroccan cities such as Tetouan,

Rabat and Khouribga and, therefore, Morocco rapidly responded to the COVID-19 infection and the Moroccan authorities implemented a series of control measures to limit person-toperson transmission. The Ministry of interior has decided to close all schools of the country and promotes online interactive learning as an alternative [28] . Movements of Moroccan people are restricted to their living space and all gathering of more than fifty persons has become prohibited [29] . Mosques and non-essential common areas are closed except for pharmacies and stores selling necessary goods to citizen. Moreover, all international flights are suspended and all public events are canceled [29] . On March 19, the government decreed a state of health emergency and the compulsory containment of population has been declared [27] . However, a part of the population continued to work to supply the basic necessities to the confined population. Every day, infectious individuals are reported by the Ministry of Health and all individuals that have a close contact with them are identified and quarantined.

Unfortunately, the unreported and asymptomatic infectious cases may exist within confined or working people and constitute a source of contamination of susceptible individuals like family members, co-workers, sellers, etc. Despite being the most affected country in the world by the disease, China is the first country that has succeeded in mitigating the progression of COVID-19 by a stringent containment of the population [33, 36] . Others countries such as Morocco still continue their fight to curb the growth in the number of daily infectious cases. Up to date, the number of confirmed cases of COVID-19 continues to increase and changes on a daily basis. The efficacy of containment and how long it must be maintained is still unknown. The longer the period of containment, the more serious consequences on the socioeconomic situation of the country will be. In addition, the short peak period of the epidemic and large notified cases prevent health care teams from adequately preparing for and responding to the huge load of patients. Thus, lowering the peak size and the postponing the peak time of the reported cases are of major interest.

In this study, using the reported cumulative confirmed cases in Morocco from March 2nd, 2020 to April 9, 2020 (See Table 1 

The data of reported symptomatic infectious cases is collected each day at 11 pm from the official Coronavirus Portal of Morocco [26] . 

The model without containment

The population considered in our basic model , as shown in Fig Here, we assume that reported symptomatic infectious individuals are hospitalized and

can not contact susceptibles anymore. We assume also, as confirmed by Rothe et al. [13] , that asymptomatic individuals can infect susceptible individuals through an effective contact. Furthermore, MacIntyre in [15] proved that asymptomatic and symptomatic infectious individuals share the same infection probability. Taking account of the previous assumptions, the dynamics of COVID-19 can be described as follows: Susceptibles (S) contacted with either unreported symptomatic (Ĩ u ) or asymptomatic infectious individuals (Ã), at an effective contact rate, c, are infected with infection probability, β, and move to the asymptomatic infectious class (Ã). After an average period 1/δ days the asymptomatic infectious individuals (Ã) become symptomatic and proceed either to the unreported symptomatic infectious (Ĩ u ), at rate δ 1 , or to the reported symptomatic infectious (Ĩ r ) at rate δ 2 with δ = δ 1 + δ 2 . Once becoming symptomatic, individuals of classĨ u andĨ r remain asymptomatic for 1/µ days on average before they are recovered.

Since the containment measure started 19 days since the first reported case then the model equations without containment is defined for 0 ≤ t < t 0 := 19 as follows Here, we assume that confined asymptomatic and confined unreported individuals can still spread the virus to their families. Furthermore, since the Moroccan government starts to impose public major measures, and taking account the fact that Moroccan individuals gradually began to reduce their contact with the people nearby, we assume that the contact rate changes from a constant rate to an exponentially decreasing rate, cl(t), with time. It is meaningful to assume that the only subpopulation that will not be confined is the reported symptomatic infectious subpopulation.

The model with containment control will given, for t ≥ t 0 days, by the following equations

is the total asymptomatic infectious population and I u (t) = I N u (t) + I c u (t) is the total unreported symptomatic infectious population. Moreover, if the containment rate of susceptible, asymptomatic infectious and unreported infectious subpopulations is denoted by p, then the new initial data for system (2.2) shall be given by

The basic reproduction number, R 0 , is the average number of secondary infections produced when one infectious individual is introduced into a host susceptible population. This quantity determines whether a given disease may spread, or die out in a population. To compute this number, we apply the next generation matrix method in [31] . Computation method of R 0 is presented in the Appendix.

Here, R 0 can be explained as follows: Assume that one asymptomatic infectious individual is introduced into the susceptible population. This asymptomatic individual produces, on average, βcS 0 1 δ asymptomatic individuals during his average lifespan 1/δ. These asymptomatic individuals then become unreported symptomatic infectious individuals over their lifespan 1/δ at a rate δ 1 and then each infectious symptomatic produces, on average, βcS 0 1 µ asymptomatic individuals during his lifespan 1/µ.

The control reproduction number, R c , is an important value, used to determine whether a control policy, such as containment in our case, will be efficient to decrease the number of secondary infections to be less than one. Here, we compute the control reproduction number related to the early stage of the containment (During the first 20 days starting from the first day of containment). Computation method of R c is presented in the Appendix.

The parameter estimation is a crucial step of our study since the epidemiologically relevant choice of the parameters must establish and confirm the observed dynamics of the infection during the onset of the epidemic. Table 2 gives the values of the parameters used in the model. In our simulations, the the total population of the Moroccan Kingdom,S(0), is chosen based on estimates from [5] , the asymptomatic duration, 1/δ, based on estimates from [26] and the symptomatic duration, 1/µ, based on estimates from [34] . Note that the data of reported cases used to estimate the model parameters was carried out before containment and during the epidemic period. Thus, we assume that the contact rate, c, since the first reported case and before the first day of containment is around 10 contacts per person in average [9] . Furthermore, we assume that a likely confined individual can only go out once every 2 weeks on average to do the necessary shopping while an individual who cannot respect containment, due to a job of paramount importance, must go 5 days a week to work.

Thus, it is meaningful to assume that, during the short time from March 21 to April 9, the average contact rate satisfy the relation 1 39−20´3 9 20 cl(t)dt ≈ c/10. Thus, we assume that α ≈ 0.078. The other parameters and initial data are estimated as follows:

Since the first and the only infectious symptomatic individual is reported on March 2nd, 2020, which corresponds to t = 0, thenĨ r (0) = 1. For the estimation of β,Ã(0) andĨ(0) we will use the data of cumulative reported cases collected from March 2nd to March 20 (before the start of containment) in Table ( 1) and we follow the procedure by [8] . The cumulative reported infectious population is given, for t ≥ 0, by CR(t) = δ 2´t 0Ã (s)ds + 1. It is obvious that cumulative reported infectious population increases slowly and then accelerates rapidly with time. Hence, we will use exponential regression with 95% of confidence level to find an exponential function that best fits the data, from March 2nd to March 20th, in Table ( From the second equation of system (2.1) and using (2.3) we obtain 

The government have deployed a series of severe control measures to limit transmission of COVID-19 across Morocco. The Moroccan people entered compulsory containment 19

days since the first infectious case has been confirmed. In this study we adopted a deterministic mathematical model of COVID-19 dynamics which took into account the Moroccan containment strategy used to control and eradicate the disease in the country. We used reported infectious case data, from March 2nd to April 9th, 2020, provided by the Health Ministry of Morocco to parameterize the model. On the one hand, as shown in Fig. 3 .1, our simulations shows that our model fit well the cumulative data of reported infectious cases giving in Table 1 , under the absence of compulsory containment conditions. On the other hand, to be as close as possible to reality, we used the cumulative reported case data from March 21th to April 09th and we applied the same method in Section 2.4

to estimate the containment rate p during these first 20 days of containment (See appendix 4). The Fig. 3 .2 showed that our model is able to predict the provided real reported data during the first 20 days past after the beginning of containment. These results are horrifying and no country in the world could control a major wave of infectious cases. For practical reasons, it is imperative to reduce the peak-value so as not to exceed the limit capacity of hospital beds and lead the country towards an uncontrollable dramatic situation. It is also better to delay the peak appearance time to allow official health structures to prepare for such a high level of emergency. Increasing the containment rate p from 0.73 to 0.90 lead to a rapid saturation of cumulative number of reported infectious cases, ranging from 16950 to 7558 (Fig. 3.4) . Furthermore, as shown in Fig. 3 .4, the peak times of reported infectious cases decreased slightly while the epidemic duration decreased significantly from 180 to 125 days and the peak-sizes decreased from 338 to 188 cases (See Table 3 ). If Moroccan government succeeded to contain more than 90% of population, then the peak size of infectious cases would be less than 188 reported cases on April, 27th, 2020. However, these results could be overestimated or underestimated since many epidemiological factors are still unknown and are subject of study by researchers. Up to date, Morocco still did not reach the epidemic peak yet. Furthermore, if the containment period will last so long, Morocco must combine its actual strategy with the mass-testing or the contact-tracing strategies which allow finding and controlling the suspected cases through the susceptible population [10, 6] . For example, to the best of our knowledge, South Korea is the only country that has succeeded to greatly slow its epidemic without resorting to drastic containment [3] . The increase of the containment rate does not significantly affect the peak-time but acted differently on epidemic duration (see Fig. 3.4) . Indeed, we found that, by increasing the containment rate, there exists a threshold control reproduction number R c = 1 corresponding to a threshold containment rate p * = 0.73, below which the epidemic end-time is postponed ( Fig. 3 .5) and above which the epidemic end-time is advanced (Fig. 3.4) . This phenomenon can be explained as follows: In the absence of containment, the whole population becomes infected within a short time. However, by increasing the rate of containment gradually to the threshold p * , the daily number of infectious individuals becomes lower and continues to significantly infect the susceptible individuals, which increases the duration of the epidemic.

Moreover, if the containment rate passe the threshold, the containment becomes efficient since the daily number of infectious individuals significantly diminish, which in turn decrease the number of new infected and, consequently, shorten the duration of the epidemic. Finally, the study of sensitivity analysis supported the results above and found that the containment rate has significant impact with correlation coefficient PRCC=−0.8. In general, the only three parameters playing a major role on the model dynamic are the contact rate, the containment rate and the asymptomatic duration ( Fig. 3.6 ). However, our sensitivity analysis study show that the asymptomatic duration have a greater impact, with correlation coefficient PRCC=0.863, and consequently, can play a major role in COVID-19 spread across Morocco. Indeed, it has been shown that younger age individuals are the most contributor of silent transmission of COVID-19 to older family members [15] , which could be the case in Morocco since 26% of the Moroccan community are less then 15 years old [4] . Fortunately, the Moroccan government have decided to close schools earlier to avoid such a critical epidemic consequences.

The COVID-19 infection was primary imported to Morocco from European countries. and the associated Jacobian matrix is given by

 and R 0 is its spectral radius which is given

Following the remark on the contact behaviour and the estimation given in Section 2.4

we will assume that, in average, cl(t) = c/10. 

Consequently, the control reproduction number is given by R c = (βcpS(t 0 )/10 + βc(1 − p)S(t 0 )) 1 δ + δ 1 δµ .

Estimation of p between March 21 and April 9

The cumulative reported cases for t ≥ t 0 is given by CR(t) = δ 2´t t 0 A(s)ds+I r (t 0 ) = be at .

Here we will use once again the exponential regression with 95% of confidence level to find an exponential function that best fits the data, from March 21 to April 9, in Table (1) .

We found that exponential model given by be at with a = 0.139 with confidence interval CI (0.125 − 0.153) and b = 7.8303 with CI (5.176 − 11.858) fits well the data with a correlation coefficient given by R = 0.981. Differentiating the both terms of CR(t) leads to

Since the initial susceptible population is not dramatically affected in the early phase of the epidemic, we will assume that S(t) ≈S(t 0 ). Adding the third and fourth equation of system and

Solving equations (4.5) and (4.6) for p lead to p = 1 9βcS(t 0 )/10 βcS(t 0 ) − (a + µ) (a + δ) a + µ + δ 1 .

Evaluation and Treatment Coronavirus (COVID-19) [update

WHO Declares COVID-19 a Pandemic

Strategies to control COVID-19 and future pandemics in Africa and around the globe

Population et développement au Maroc : vingt-cinq ans après la conférence du Caire de 1994

Dynamic Modeling to Identify Mitigation Strategies for Covid-19 Pandemic 2020

The reproductive number of COVID-19 is higher compared to SARS coronavirus

Understanding unreported cases in the 2019-nCov epidemic outbreak in Wuhan, China, and the importance of major public health interventions

Social contacts and mixing patterns 366 relevant to the spread of infectious diseases

Covid-19 mass testing facilities could end the epidemic rapidly

Covid-19 and Italy: what next? Health Policy

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

Transmission of 2019-nCoV infection from an asymptomatic contact in Germany

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

Global spread of COVID-19 and pandemic potential

The Ministry of Health of Morocco, Communiqué N°10: Morocco announces 1st COVID-19 case, www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°11: Morocco registers 2nd COVID-19 case, www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°12: Morocco registers third COVID-19 case. www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°14: Morocco announces two new cases infected with COVID-19

The Ministry of Health of Morocco, Communiqué N°15: Morocco registers sixth COVID-19 case. www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°16: Morocco registers seventh COVID-19 case. www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°18: Morocco registers eighth COVID-19 case. www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°19: Morocco announces 9 new cases infected with COVID-19. www.covidmaroc

The Ministry of Health of Morocco, Communiqué N°20: Morocco announces 11 new cases infected with COVID-19. www.covidmaroc

2020) Communiqué N°22: The epidemiological situation of COVID-19. www.covidmaroc

The Ministry of Health of Morocco, The Official Coronavirus Portal of Morocco. www.covidmaroc

The Ministry of interior

The Ministry of interior

The Ministry of interior, MAP-ANTI-CORONA mapanticorona.map

COVID-19 R0: Magic number or conundrum

Reproduction numbers and subthreshold endemic equilibria for compartmental models of disease transmission

Report of the WHO-China joint mission on Coronavirus disease (COVID-19)

WHO, News briefing on

Isolation, quarantine, social distancing and community containment: pivotal role for old-style public health measures in the novel coronavirus (2019-nCoV) outbreak

A systematic approach is needed to contain COVID-19 Science Bulletin

The authors thank the anonymous referees, whose careful reading, insights, valuable comments, and suggestions significantly enabled us to improve the quality of the paper.

The linearized equation related to infectious individuals of system (2.1) is given by

The author declare that there is no conflict of interest 25