A Regional Model for projecting COVID-19 in Northern Mindanao, Philippines Jayrold P. Arcedea,∗, Randy L. Caga-ananb, May Anne E. Matae, Youcef Mammeric, Rhoda A. Namocod, Ian Christian A. Gonzalesf, Zython Paul Lachicae aDepartment of Mathematics, Caraga State University, Butuan City, Philippines bDepartment of Mathematics and Statistics, MSU-Iligan Institute of Technology, Iligan City, Philippines cLaboratoire Ami´enoisde Math´ematiqueFondamentale et Appliqu´ee,CNRS UMR 7352, Universit´ede Picardie Jules Verne, 80069 Amiens, France dDepartment of Applied Mathematics, University of Science and Technology of Southern Philippines, Cagayan De Oro, Philippines eDepartment of Mathematics, Physics and Computer Science, University of the Philippines Mindanao, Davao City, Philippines fCenter for Health Development Northern Mindanao, Department of Health, Cagayan de Oro City, Philippines Abstract To provide the most accurate and timely advice possible for health policy mak- ers about a reliable projections of COVID 19 in Northern Mindanao, and the possible impact of measures designed for control, we proposed a compartmental model for COVID 19 which essentially capture the dynamics at a local level with parameters suited for the region. Closed-form formulas for the basic and effective reproduction numbers of the model are obtained. Sensitivity analysis is done via Latin Hypercube Sampling - Partial Rank Correlation Coefficient methodology. Results show a strong positive correlation between basic repro- duction number and the transmission rate from susceptible to exposed. Further- more, we also provided numerical experiments for each province and the whole region. Among the results, a decaying reproductive number which approaches or below unity. Lastly, four intervention strategies were considered. Without much testing capacity and no vaccine, simulations show that the best strategy is the combination of containment, lockdown, and social distancing, but very good ∗Corresponding author:[email protected] Preprint submitted to ?? November 10, 2020 improvements can also be expected from amplifying the efforts to quarantine exposed individuals. Keywords: COVID-19, Philippines, probable, regional, quarantine 1 1. Introduction 2 There are more than a million viruses which still out there underscovered 3 and every decade some appeared and create havoc to the populace. One of 4 which are the Coronaviruses (CoV). They are a large family of viruses which 5 causes illness ranging from the common cold to more severe diseases, examples 6 of which was the Severe Acute Respiratory Syndrome (SARS-CoV) in 2003, and 7 the Middle East Respiratory Syndrome (MERS-CoV) in 2013. A recent strain 8 of this family of viruses known as novel coronavirus disease 2019 or COVID-19 9 [23], first emerged on late November 2019 in Wuhan City, Hubei Province of 10 the People's Republic of China. However, it was only identified the first week 11 of this year as the number of cases with respiratory disease grew in the city. 12 On January 11, 2020, the first COVID-19-induced death in China was reported. 13 Three days after, the first COVID-19 case outside China was reported, and 14 ever since, the disease has continued to spread worldwide at a faster rate and 15 becomes pandemic. As of April 18, the total global number of COVID-19 cases 16 exceeds 2,243,512, with over 154,209 deaths [2]. The emergence of COVID-19 17 poses serious threats to global health and economy despite the aggressive efforts 18 to prevent it. 19 As a novel pathogen, there is no specific treatment for COVID-19. Hospital- 20 ized cases are managed with supportive care. The most effective intervention to 21 prevent mortality and to reduce the number of cases is through population-wide 22 interventions, also known as Non-Pharmaceutical Interventions (NPI). These in- 23 clude efficient testing, isolation, contact tracing, social distancing through the 24 reduction of mobility and cancellation of mass gatherings. Because SARS-CoV-2 25 is transmitted mainly through droplets and contact, reduced interaction within 2 26 the population will slow down disease transmission. 27 In the Philippines, the COVID-19 status was raised to Code RED, level 2, 28 last March 7, 2020. The President Rodrigo Roa Duterte (PPRD) then placed 29 the entire Luzon on a lockdown banning land, domestic air, and domestic sea 30 travel for 14 days. After its lapse, PPRD further imposed an extended lock- 31 down for two weeks more following experts' opinions. However, despite one 32 of Southeast Asia's strictest and earliest lockdown measures implementation, 33 COVID-19 casualties continue to soar up topping Southeast Asian countries 34 identifying Metro Manila being the epicenter of the outbreak [20]. However, 35 the government believes that its swift move to close borders and put half its 36 population under home quarantine had averted a far greater toll and a health- 37 care disaster [19]. As of May 23, the Philippines records 13,597 cases, with 857 38 deaths [1]. 39 The Philippines is an archipelago that is divided into 17 regions. Given the 40 country's meager resources, fragile healthcare system, limited testing capacity, 41 and bureaucratic culture, there is no way that all regions will be afforded much 42 of its resources. Recently the government approved the amelioration fund to be 43 given among the poorest in the country as a way to buy time while waiting for a 44 vaccine. Moreover, the move of Congress to grant a temporary special power to 45 the president dubbed as "Bayanihan to heal as one Act" [21] although was not 46 well received by some, had accomplished its purposes: (1) effecting draconian 47 measures in the form of lockdown/containment, (2) continued quarantine (3) 48 implementing social distancing with the help of military and police force (4) 49 increasing testing capacity, and finally, (5) bolstering the number of accredited 50 hospitals and laboratories for COVID-19 testing. Furthermore, hospitals have 51 put under the lens of criticisms for not accepting patients due to full wards, 52 insufficient number of test kits and protective equipment, and a depleted work 53 force[7]. 54 This study focuses on COVID 19 implications at the regional level. In par- 55 ticular, we want to provide insights to health officials which will aid in the imple- 56 mentation of NPI for prevention and mitigation while anticipating budgetary, 3 57 logistics, and human resources need in planning for the eventual surge of cases. 58 In doing so, we develop a compartmental model as popularized by Brauer et al. 59 [6] to describe COVID 19 dynamics at a regional level. We explore some public 60 concerns regarding the implementation of the following strategies: 1) social dis- 61 tancing 2.) containment period/quarantine duration, 3) testing/detection, 4) 62 hospitalization availability, i.e., beds capacity. 63 Finally, we estimate unknown epidemiological rates that will accurately de- 64 scribe the trends in the number of PUIs and PUMs. These estimated rates 65 were used as baseline parameter values to provide projections for the number of 66 hospitalized PUIs in the next few weeks in Region X as recommendations for 67 health officials to prepare them ahead. 68 Our place of concern is Region X, which has the second-highest occurrence 69 of COVID-19 cases in Mindanao next to Davao City. We note that the peak of 70 COVID-19 infection may be asynchronous at the regional level, which may be 71 hampered due to localized lockdown and archipelagic territory. Region X also 72 called Northern Mindanao is composed of 5 provinces: Bukidnon, Camiguin, 73 Lanao del Norte, Misamis Occidental, and Misamis Oriental. The region is 74 also a home of Cagayan de Oro, a highly urbanized city that serves as the 75 regional center and business hub of the region. Recently, the city mayor created 76 noise from observers for resisting pressures from medical practitioners to impose 77 lockdown on the city citing economic collapse [12]. As of May 23 Region X 78 recorded 22 confirmed positive COVID cases with 8 deaths [3]. 79 2. Materials and methods 80 2.1. PUM, PUI/Hospitalized, Discharge, Quarantine datasets 81 In this study, we used the publicly available dataset of COVID-19 cases 82 posted by Department of Health (DOH) in bit.ly/dohcovid19data. The data 83 posted is for research purposes or public consumption. Henceforth, ethical ap- 84 proval is not required. 4 85 We have collected and cleaned the data for Region X and its provinces 86 namely Misamis Occidental, Misamis Oriental, Bukidnon, and Lanao Del Norte. 87 However, we do not include Camiguin since the data is not sufficient for our pur- 88 poses. The dataset include daily count of PUM, PUI/Hospitalized, Discharge, 89 and Quarantine individuals. Further, dates reflected in this report are the ar- 90 rival dates of the individuals within the area of the region. 91 2.2. Mathematical model 92 The model developed in this study was an extension of the compartmental 93 model [6, 17] to describe the dynamics of COVID-19 infection in the North- 94 ern Mindanao with the available data sets. In the model, the individuals are 95 classified into five types: 96 1. Susceptible (S) - person at risk of contracting the disease, 97 2. Exposed (E) - person who have been exposed to the disease but not nec- 98 essarily infected, 99 3. Hospitalized (H) - Person Under Investigation (PUI), who is positive with 100 COVID-19 symptoms, that is hospitalized, 101 4. Quarantined (Q) - a PUI who is negative with COVID-19 symptoms or a 102 Person Under Monitoring (PUM) that is being quarantined, 103 5. Removed (R) - a person who recovers from the disease hence removed in 104 the system assuming they gained immunity for a time of interest. 105 A schematic diagram of the model is shown in Figure 1.