case study of covid 19 in the philippines

Log in using your username and password

• Search More Search for this keyword Advanced search
• Latest content
• For authors
• Browse by collection
• BMJ Journals More You are viewing from: Google Indexer

You are here

• Volume 10, Issue 11
• The Philippine COVID-19 Outcomes: a Retrospective study Of Neurological manifestations and Associated symptoms (The Philippine CORONA study): a protocol study
• Article Text
• Article info
• Citation Tools
• Rapid Responses
• Article metrics

• http://orcid.org/0000-0001-5621-1833 Adrian I Espiritu 1 , 2 ,
• http://orcid.org/0000-0003-1135-6400 Marie Charmaine C Sy 1 ,
• http://orcid.org/0000-0002-1241-8805 Veeda Michelle M Anlacan 1 ,
• http://orcid.org/0000-0001-5317-7369 Roland Dominic G Jamora 1
• 1 Department of Neurosciences , College of Medicine and Philippine General Hospital, University of the Philippines Manila , Manila , Philippines
• 2 Department of Clinical Epidemiology, College of Medicine , University of the Philippines Manila , Manila , Philippines
• Correspondence to Dr Adrian I Espiritu; aiespiritu{at}up.edu.ph

Introduction The SARS-CoV-2, virus that caused the COVID-19 global pandemic, possesses a neuroinvasive potential. Patients with COVID-19 infection present with neurological signs and symptoms aside from the usual respiratory affectation. Moreover, COVID-19 is associated with several neurological diseases and complications, which may eventually affect clinical outcomes.

Objectives The Philippine COVID-19 Outcomes: a Retrospective study Of Neurological manifestations and Associated symptoms (The Philippine CORONA) study investigators will conduct a nationwide, multicentre study involving 37 institutions that aims to determine the neurological manifestations and factors associated with clinical outcomes in COVID-19 infection.

Methodology and analysis This is a retrospective cohort study (comparative between patients with and without neurological manifestations) via medical chart review involving adult patients with COVID-19 infection. Sample size was determined at 1342 patients. Demographic, clinical and neurological profiles will be obtained and summarised using descriptive statistics. Student’s t-test for two independent samples and χ 2 test will be used to determine differences between distributions. HRs and 95% CI will be used as an outcome measure. Kaplan-Meier curves will be constructed to plot the time to onset of mortality (survival), respiratory failure, intensive care unit (ICU) admission, duration of ventilator dependence, length of ICU stay and length of hospital stay. The log-rank test will be employed to compare the Kaplan-Meier curves. Stratified analysis will be performed to identify confounders and effects modifiers. To compute for adjusted HR with 95% CI, crude HR of outcomes will be adjusted according to the prespecified possible confounders. Cox proportional regression models will be used to determine significant factors of outcomes. Testing for goodness of fit will also be done using Hosmer-Lemeshow test. Subgroup analysis will be performed for proven prespecified effect modifiers. The effects of missing data and outliers will also be evaluated in this study.

Ethics and dissemination This protocol was approved by the Single Joint Research Ethics Board of the Philippine Department of Health (SJREB-2020–24) and the institutional review board of the different study sites. The dissemination of results will be conducted through scientific/medical conferences and through journal publication. The lay versions of the results may be provided on request.

Trial registration number NCT04386083 .

• adult neurology
• epidemiology

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:  http://creativecommons.org/licenses/by-nc/4.0/ .

https://doi.org/10.1136/bmjopen-2020-040944

Statistics from Altmetric.com

Request permissions.

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

The Philippine COVID-19 Outcomes: a Retrospective study Of Neurological manifestations and Associated symptoms Study is a nationwide, multicentre, retrospective, cohort study with 37 Philippine sites.

Full spectrum of neurological manifestations of COVID-19 will be collected.

Retrospective gathering of data offers virtually no risk of COVID-19 infection to data collectors.

Data from COVID-19 patients who did not go to the hospital are unobtainable.

Recoding bias is inherent due to the retrospective nature of the study.

Introduction

The COVID-19 has been identified as the cause of an outbreak of respiratory illness in Wuhan, Hubei Province, China, in December 2019. 1 The COVID-19 pandemic has reached the Philippines with most of its cases found in the National Capital Region (NCR). 2 The major clinical features of COVID-19 include fever, cough, shortness of breath, myalgia, headache and diarrhoea. 3 The outcomes of this disease lead to prolonged hospital stay, intensive care unit (ICU) admission, dependence on invasive mechanical ventilation, respiratory failure and mortality. 4 The specific pathogen that causes this clinical syndrome has been named SARS-CoV-2, which is phylogenetically similar to SARS-CoV. 4 Like the SARS-CoV strain, SARS-CoV-2 may possess a similar neuroinvasive potential. 5

A study on cases with COVID-19 found that about 36.4% of patients displayed neurological manifestations of the central nervous system (CNS) and peripheral nervous system (PNS). 6 The associated spectrum of symptoms and signs were substantially broad such as altered mental status, headache, cognitive impairment, agitation, dysexecutive syndrome, seizures, corticospinal tract signs, dysgeusia, extraocular movement abnormalities and myalgia. 7–12 Several reports were published on neurological disorders associated with patients with COVID-19, including cerebrovascular disorders, encephalopathy, hypoxic brain injury, frequent convulsive seizures and inflammatory CNS syndromes like encephalitis, meningitis, acute disseminated encephalomyelitis and Guillain-Barre syndrome. 7–16 However, the estimates of the occurrences of these manifestations were based on studies with a relatively small sample size. Furthermore, the current description of COVID-19 neurological features are hampered to some extent by exceedingly variable reporting; thus, defining causality between this infection and certain neurological manifestations is crucial since this may lead to considerable complications. 17 An Italian observational study protocol on neurological manifestations has also been published to further document and corroborate these findings. 18

Epidemiological data on the proportions and spectrum of non-respiratory symptoms and complications may be essential to increase the recognition of clinicians of the possibility of COVID-19 infection in the presence of other symptoms, particularly neurological manifestations. With this information, the probabilities of diagnosing COVID-19 disease may be strengthened depending on the presence of certain neurological manifestations. Furthermore, knowledge of other unrecognised symptoms and complications may allow early diagnosis that may permit early institution of personal protective equipment and proper contact precautions. Lastly, the presence of neurological manifestations may be used for estimating the risk of certain important clinical outcomes for better and well-informed clinical decisions in patients with COVID-19 disease.

To address this lack of important information in the overall management of patients with COVID-19, we organised a research study entitled ‘The Philippine COVID-19 Outcomes: a Retrospective study Of Neurological manifestations and Associated symptoms (The Philippine CORONA Study)’.

This quantitative, retrospective cohort, multicentre study aims: (1) to determine the demographic, clinical and neurological profile of patients with COVID-19 disease in the Philippines; (2) to determine the frequency of neurological symptoms and new-onset neurological disorders/complications in patients with COVID-19 disease; (3) to determine the neurological manifestations that are significant factors of mortality, respiratory failure, duration of ventilator dependence, ICU admission, length of ICU stay and length of hospital stay among patients with COVID-19 disease; (4) to determine if there is significant difference between COVID-19 patients with neurological manifestations compared with those COVID-19 patients without neurological manifestations in terms of mortality, respiratory failure, duration of ventilator dependence, ICU admission, length of ICU stay and length of hospital stay; and (5) to determine the likelihood of mortality, respiratory failure and ICU admission, including the likelihood of longer duration of ventilator dependence and length of ICU and hospital stay in COVID-19 patients with neurological manifestations compared with those without neurological manifestations.

Scope, limitations and delimitations

The study will include confirmed cases of COVID-19 from the 37 participating institutions in the Philippines. Every country has its own healthcare system, whose level of development and strategies ultimately affect patient outcomes. Thus, the results of this study cannot be accurately generalised to other settings. In addition, patients with ages ≤18 years will be excluded in from this study. These younger patients may have different characteristics and outcomes; therefore, yielded estimates for adults in this study may not be applicable to this population subgroup. Moreover, this study will collect data from the patient records of patients with COVID-19; thus, data from patients with mild symptoms who did not go to the hospital and those who had spontaneous resolution of symptoms despite true infection with COVID-19 are unobtainable.

Methodology

To improve the quality of reporting of this study, the guidelines issued by the Strengthening the Reporting of Observational Studies in Epidemiology Initiative will be followed. 19

Study design

The study will be conducted using a retrospective cohort (comparative) design (see figure 1 ).

• Download figure
• Open in new tab
• Download powerpoint

Schematic diagram of the study flow.

Study sites and duration

We will conduct a nationwide, multicentre study involving 37 institutions in the Philippines (see figure 2 ). Most of these study sites can be found in the NCR, which remains to be the epicentre of the COVID-19 pandemic. 2 We will collect data for 6 months after institutional review board approval for every site.

Location of 37 study sites of the Philippine CORONA study.

Patient selection and cohort description

The cases will be identified using the designated COVID-19 censuses of all the participating centres. A total enumeration of patients with confirmed COVID-19 disease will be done in this study.

The cases identified should satisfy the following inclusion criteria: (A) adult patients at least 19 years of age; (B) cases confirmed by testing approved patient samples (ie, nasal swab, sputum and bronchoalveolar lavage fluid) employing real-time reverse transcription PCR (rRT-PCR) 20 from COVID-19 testing centres accredited by the Department of Health (DOH) of the Philippines, with clinical symptoms and signs attributable to COVID-19 disease (ie, respiratory as well as non-respiratory clinical signs and symptoms) 21 ; and (C) cases with disposition (ie, discharged stable/recovered, home/discharged against medical advice, transferred to other hospital or died) at the end of the study period. Cases with conditions or diseases caused by other organisms (ie, bacteria, other viruses, fungi and so on) or caused by other pathologies unrelated to COVID-19 disease (ie, trauma) will be excluded.

The first cohort will involve patients with confirmed COVID-19 infection who presented with any neurological manifestation/s (ie, symptoms or complications/disorder). The comparator cohort will compose of patients with confirmed COVID-19 infection without neurological manifestation/s.

Sample size calculation

We looked into the mortality outcome measure for the purposes of sample size computation. Following the cohort study of Khaledifar et al , 22 the sample size was calculated using the following parameters: two-sided 95% significance level (1 – α); 80% power (1 – β); unexposed/exposed ratio of 1; 5% of unexposed with outcome (case fatality rate from COVID19-Philippines Dashboard Tracker (PH) 23 as of 8 April 2020); and assumed risk ratio 2 (to see a two-fold increase in risk of mortality when neurological symptoms are present).

When these values were plugged in to the formula for cohort studies, 24 a minimum sample size of 1118 is required. To account for possible incomplete data, the sample was adjusted for 20% more. This means that the total sample size required is 1342 patients, which will be gathered from the participating centres.

Data collection

We formulated an electronic data collection form using Epi Info Software (V.7.2.2.16). The forms will be pilot-tested, and a formal data collection workshop will be conducted to ensure collection accuracy. The data will be obtained from the review of the medical records.

The following pertinent data will be obtained: (A) demographic data; (B) other clinical profile data/comorbidities; (C) neurological history; (D) date of illness onset; (E) respiratory and constitutional symptoms associated with COVID-19; (F) COVID-19 disease severity 25 at nadir; (G) data if neurological manifestation/s were present at onset prior to respiratory symptoms and the specific neurological manifestation/s present at onset; (H) neurological symptoms; (i) date of neurological symptom onset; (J) new-onset neurological disorders or complications; (K) date of new neurological disorder or complication onset; (L) imaging done; (M) cerebrospinal fluid analysis; (N) electrophysiological studies; (O) treatment given; (P) antibiotics given; (Q) neurological interventions given; (R) date of mortality and cause/s of mortality; (S) date of respiratory failure onset, date of mechanical ventilator cessation and cause/s of respiratory failure; (T) date of first day of ICU admission, date of discharge from ICU and indication/s for ICU admission; (U) other neurological outcomes at discharge; (V) date of hospital discharge; and (W) final disposition. See table 1 for the summary of the data to be collected for this study.

• View inline

Data to be collected in this study

Main outcomes considered

The following patient outcomes will be considered for this study:

Mortality (binary outcome): defined as the patients with confirmed COVID-19 who died.

Respiratory failure (binary outcome): defined as the patients with confirmed COVID-19 who experienced clinical symptoms and signs of respiratory insufficiency. Clinically, this condition may manifest as tachypnoea/sign of increased work of breathing (ie, respiratory rate of ≥22), abnormal blood gases (ie, hypoxaemia as evidenced by partial pressure of oxygen (PaO 2 ) <60 or hypercapnia by partial pressure of carbon dioxide of >45), or requiring oxygen supplementation (ie, PaO 2 <60 or ratio of PaO 2 /fraction of inspired oxygen (P/F ratio)) <300).

Duration of ventilator dependence (continuous outcome): defined as the number of days from initiation of assisted ventilation to cessation of mechanical ventilator use.

ICU admission (binary outcome): defined as the patients with confirmed COVID-19 admitted to an ICU or ICU-comparable setting.

Length of ICU stay (continuous outcome): defined as the number of days admitted in the ICU or ICU-comparable setting.

Length of hospital stay (continuous outcome): defined as the number of days from admission to discharge.

Data analysis plan

Statistical analysis will be performed using Stata V.7.2.2.16.

Demographic, clinical and neurological profiles will be summarised using descriptive statistics, in which categorical variables will be expressed as frequencies with corresponding percentages, and continuous variables will be pooled using means (SD).

Student’s t-test for two independent samples and χ 2 test will be used to determine differences between distributions.

HRs and 95% CI will be used as an outcome measure. Kaplan-Meier curves will be constructed to plot the time to onset of mortality (survival), respiratory failure, ICU admission, duration of ventilator dependence (recategorised binary form), length of ICU stay (recategorised binary form) and length of hospital stay (recategorised binary form). Log-rank test will be employed to compare the Kaplan-Meier curves. Stratified analysis will be performed to identify confounders and effects modifiers. To compute for adjusted HR with 95% CI, crude HR of outcomes at discrete time points will be adjusted for prespecified possible confounders such as age, history of cardiovascular or cerebrovascular disease, hypertension, diabetes mellitus, and respiratory disease, COVID-19 disease severity at nadir, and other significant confounding factors.

Cox proportional regression models will be used to determine significant factors of outcomes. Testing for goodness of fit will be done using Hosmer-Lemeshow test. Likelihood ratio tests and other information criteria (Akaike Information Criterion or Bayesian Information Criterion) will be used to refine the final model. Statistical significance will be considered if the 95% CI of HR or adjusted HR did not include the number one. A p value <0.05 (two tailed) is set for other analyses.

Subgroup analyses will be performed for proven prespecified effect modifiers. The following variables will be considered for subgroup analyses: age (19–64 years vs ≥65 years), sex, body mass index (<18.5 vs 18.5–22.9 vs ≥23 kg/m 2 ), with history of cardiovascular or cerebrovascular disease (presence or absence), hypertension (presence or absence), diabetes mellitus (presence or absence), respiratory disease (presence or absence), smoking status (smoker or non-smoker) and COVID-19 disease severity (mild, severe or critical disease).

The effects of missing data will be explored. All efforts will be exerted to minimise missing and spurious data. Validity of the submitted electronic data collection will be monitored and reviewed weekly to prevent missing or inaccurate input of data. Multiple imputations will be performed for missing data when possible. To check for robustness of results, analysis done for patients with complete data will be compared with the analysis with the imputed data.

The effects of outliers will also be assessed. Outliers will be assessed by z-score or boxplot. A cut-off of 3 SD from the mean can also be used. To check for robustness of results, analysis done with outliers will be compared with the analysis without the outliers.

Study organisational structure

A steering committee (AIE, MCCS, VMMA and RDGJ) was formed to direct and provide appropriate scientific, technical and methodological assistance to study site investigators and collaborators (see figure 3 ). Central administrative coordination, data management, administrative support, documentation of progress reports, data analyses and interpretation and journal publication are the main responsibilities of the steering committee. Study site investigators and collaborators are responsible for the proper collection and recording of data including the duty to maintain the confidentiality of information and the privacy of all identified patients for all the phases of the research processes.

Organisational structure of oversight of the Philippine CORONA Study.

This section is highlighted as part of the required formatting amendments by the Journal.

Ethics and dissemination

This research will adhere to the Philippine National Ethical Guidelines for Health and Health-related Research 2017. 26 This study is an observational, cohort study and will not allocate any type of intervention. The medical records of the identified patients will be reviewed retrospectively. To protect the privacy of the participant, the data collection forms will not contain any information (ie, names and institutional patient number) that could determine the identity of the patients. A sequential code will be recorded for each patient in the following format: AAA-BBB where AAA will pertain to the three-digit code randomly assigned to each study site; BBB will pertain to the sequential case number assigned by each study site. Each participating centre will designate a password-protected laptop for data collection; the password is known only to the study site.

This protocol was approved by the following institutional review boards: Single Joint Research Ethics Board of the DOH, Philippines (SJREB-2020-24); Asian Hospital and Medical Center, Muntinlupa City (2020- 010-A); Baguio General Hospital and Medical Center (BGHMC), Baguio City (BGHMC-ERC-2020-13); Cagayan Valley Medical Center (CVMC), Tuguegarao City; Capitol Medical Center, Quezon City; Cardinal Santos Medical Center (CSMC), San Juan City (CSMC REC 2020-020); Chong Hua Hospital, Cebu City (IRB 2420–04); De La Salle Medical and Health Sciences Institute (DLSMHSI), Cavite (2020-23-02-A); East Avenue Medical Center (EAMC), Quezon City (EAMC IERB 2020-38); Jose R. Reyes Memorial Medical Center, Manila; Jose B. Lingad Memorial Regional Hospital, San Fernando, Pampanga; Dr. Jose N. Rodriguez Memorial Hospital, Caloocan City; Lung Center of the Philippines (LCP), Quezon City (LCP-CT-010–2020); Manila Doctors Hospital, Manila (MDH IRB 2020-006); Makati Medical Center, Makati City (MMC IRB 2020–054); Manila Medical Center, Manila (MMERC 2020-09); Northern Mindanao Medical Center, Cagayan de Oro City (025-2020); Quirino Memorial Medical Center (QMMC), Quezon City (QMMC REB GCS 2020-28); Ospital ng Makati, Makati City; University of the Philippines – Philippine General Hospital (UP-PGH), Manila (2020-314-01 SJREB); Philippine Heart Center, Quezon City; Research Institute for Tropical Medicine, Muntinlupa City (RITM IRB 2020-16); San Lazaro Hospital, Manila; San Juan De Dios Educational Foundation Inc – Hospital, Pasay City (SJRIB 2020-0006); Southern Isabela Medical Center, Santiago City (2020-03); Southern Philippines Medical Center (SPMC), Davao City (P20062001); St. Luke’s Medical Center, Quezon City (SL-20116); St. Luke’s Medical Center, Bonifacio Global City, Taguig City (SL-20116); Southern Philippines Medical Center, Davao City; The Medical City, Pasig City; University of Santo Tomas Hospital, Manila (UST-REC-2020-04-071-MD); University of the East Ramon Magsaysay Memorial Medical Center, Inc, Quezon City (0835/E/2020/063); Veterans Memorial Medical Center (VMMC), Quezon City (VMMC-2020-025) and Vicente Sotto Memorial Medical Center, Cebu City (VSMMC-REC-O-2020–048).

The dissemination of results will be conducted through scientific/medical conferences and through journal publication. Only the aggregate results of the study shall be disseminated. The lay versions of the results may be provided on request.

Protocol registration and technical review approval

This protocol was registered in the ClinicalTrials.gov website. It has received technical review board approvals from the Department of Neurosciences, Philippine General Hospital and College of Medicine, University of the Philippines Manila, from the Cardinal Santos Medical Center (San Juan City) and from the Research Center for Clinical Epidemiology and Biostatistics, De La Salle Medical and Health Sciences Institute (Dasmariñas, Cavite).

Acknowledgments

We would like to thank Almira Abigail Doreen O Apor, MD, of the Department of Neurosciences, Philippine General Hospital, Philippines, for illustrating figure 2 for this publication.

• Adhikari SP ,
• Wu Y-J , et al
• Department of Health
• Philippine Society for Microbiology and Infectious Diseases
• Hu Y , et al
• Li Yan‐Chao ,
• Bai Wan‐Zhu ,
• Hashikawa T ,
• Wang M , et al
• Paterson RW ,
• Benjamin L , et al
• Hall JP , et al
• Varatharaj A ,
• Ellul MA , et al
• Mahammedi A ,
• Vagal A , et al
• Collantes MEV ,
• Espiritu AI ,
• Sy MCC , et al
• Merdji H , et al
• Sharifi Razavi A ,
• Poyiadji N ,
• Noujaim D , et al
• Zhou H , et al
• Moriguchi T ,
• Goto J , et al
• Nicholson TR , et al
• Ferrarese C ,
• Priori A , et al
• von Elm E ,
• Altman DG ,
• Egger M , et al
• Li J , et al
• Centers for Disease Control and Prevention
• Khaledifar A ,
• Hashemzadeh M ,
• Solati K , et al
• McGoogan JM
• Philippine Research Ethics Board

VMMA and RDGJ are joint senior authors.

AIE and MCCS are joint first authors.

Twitter @neuroaidz, @JamoraRoland

Collaborators The Philippine CORONA Study Group Collaborators: Maritoni C Abbariao, Joshua Emmanuel E Abejero, Ryndell G Alava, Robert A Barja, Dante P Bornales, Maria Teresa A Cañete, Ma. Alma E Carandang-Concepcion, Joseree-Ann S Catindig, Maria Epifania V Collantes, Evram V Corral, Ma. Lourdes P Corrales-Joson, Romulus Emmanuel H Cruz, Marita B Dantes, Ma. Caridad V Desquitado, Cid Czarina E Diesta, Carissa Paz C Dioquino, Maritzie R Eribal, Romulo U Esagunde, Rosalina B Espiritu-Picar, Valmarie S Estrada, Manolo Kristoffer C Flores, Dan Neftalie A Juangco, Muktader A Kalbi, Annabelle Y Lao-Reyes, Lina C Laxamana, Corina Maria Socorro A Macalintal, Maria Victoria G Manuel, Jennifer Justice F Manzano, Ma. Socorro C Martinez, Generaldo D Maylem, Marc Conrad C Molina, Marietta C Olaivar, Marissa T Ong, Arnold Angelo M Pineda, Joanne B Robles, Artemio A Roxas Jr, Jo Ann R Soliven, Arturo F Surdilla, Noreen Jhoanna C Tangcuangco-Trinidad, Rosalia A Teleg, Jarungchai Anton S Vatanagul and Maricar P Yumul.

Contributors All authors conceived the idea and wrote the initial drafts and revisions of the protocol. All authors made substantial contributions in this protocol for intellectual content.

Funding Philippine Neurological Association (Grant/Award Number: N/A). Expanded Hospital Research Office, Philippine General Hospital (Grant/Award Number: N/A).

Disclaimer Our funding sources had no role in the design of the protocol, and will not be involved during the methodological execution, data analyses and interpretation and decision to submit or to publish the study results.

Map disclaimer The depiction of boundaries on the map(s) in this article does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. The map(s) are provided without any warranty of any kind, either express or implied.

Competing interests None declared.

Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Patient consent for publication Not required.

Provenance and peer review Not commissioned; externally peer reviewed.

Read the full text or download the PDF:

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Study Protocol

Assessing the effect of the COVID-19 pandemic, shift to online learning, and social media use on the mental health of college students in the Philippines: A mixed-method study protocol

Roles Funding acquisition, Writing – original draft

Affiliation College of Medicine, University of the Philippines, Manila, Philippines

Roles Methodology, Supervision, Visualization, Writing – original draft, Writing – review & editing

Affiliations Department of Clinical Epidemiology, College of Medicine, University of the Philippines, Manila, Philippines, Institute of Clinical Epidemiology, National Institutes of Health, University of the Philippines, Manila, Philippines

Roles Methodology

Affiliation Department of Psychiatry, College of Medicine, University of the Philippines, Manila, Philippines

Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

• Leonard Thomas S. Lim, 
• Zypher Jude G. Regencia, 
• J. Rem C. Dela Cruz, 
• Frances Dominique V. Ho, 
• Marcela S. Rodolfo, 
• Josefina Ly-Uson, 
• Emmanuel S. Baja

• Published: May 3, 2022
• https://doi.org/10.1371/journal.pone.0267555
• Peer Review
• Reader Comments

Introduction

The COVID-19 pandemic declared by the WHO has affected many countries rendering everyday lives halted. In the Philippines, the lockdown quarantine protocols have shifted the traditional college classes to online. The abrupt transition to online classes may bring psychological effects to college students due to continuous isolation and lack of interaction with fellow students and teachers. Our study aims to assess Filipino college students’ mental health status and to estimate the effect of the COVID-19 pandemic, the shift to online learning, and social media use on mental health. In addition, facilitators or stressors that modified the mental health status of the college students during the COVID-19 pandemic, quarantine, and subsequent shift to online learning will be investigated.

Methods and analysis

Mixed-method study design will be used, which will involve: (1) an online survey to 2,100 college students across the Philippines; and (2) randomly selected 20–40 key informant interviews (KIIs). Online self-administered questionnaire (SAQ) including Depression, Anxiety, and Stress Scale (DASS-21) and Brief-COPE will be used. Moreover, socio-demographic factors, social media usage, shift to online learning factors, family history of mental health and COVID-19, and other factors that could affect mental health will also be included in the SAQ. KIIs will explore factors affecting the student’s mental health, behaviors, coping mechanism, current stressors, and other emotional reactions to these stressors. Associations between mental health outcomes and possible risk factors will be estimated using generalized linear models, while a thematic approach will be made for the findings from the KIIs. Results of the study will then be triangulated and summarized.

Ethics and dissemination

Our study has been approved by the University of the Philippines Manila Research Ethics Board (UPMREB 2021-099-01). The results will be actively disseminated through conference presentations, peer-reviewed journals, social media, print and broadcast media, and various stakeholder activities.

Citation: Lim LTS, Regencia ZJG, Dela Cruz JRC, Ho FDV, Rodolfo MS, Ly-Uson J, et al. (2022) Assessing the effect of the COVID-19 pandemic, shift to online learning, and social media use on the mental health of college students in the Philippines: A mixed-method study protocol. PLoS ONE 17(5): e0267555. https://doi.org/10.1371/journal.pone.0267555

Editor: Elisa Panada, UNITED KINGDOM

Received: June 9, 2021; Accepted: April 11, 2022; Published: May 3, 2022

Copyright: © 2022 Lim et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This project is being supported by the American Red Cross through the Philippine Red Cross and Red Cross Youth. The funder will not have a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

The World Health Organization (WHO) declared the Coronavirus 2019 (COVID-19) outbreak as a global pandemic, and the Philippines is one of the 213 countries affected by the disease [ 1 ]. To reduce the virus’s transmission, the President imposed an enhanced community quarantine in Luzon, the country’s northern and most populous island, on March 16, 2020. This lockdown manifested as curfews, checkpoints, travel restrictions, and suspension of business and school activities [ 2 ]. However, as the virus is yet to be curbed, varying quarantine restrictions are implemented across the country. In addition, schools have shifted to online learning, despite financial and psychological concerns [ 3 ].

Previous outbreaks such as the swine flu crisis adversely influenced the well-being of affected populations, causing them to develop emotional problems and raising the importance of integrating mental health into medical preparedness for similar disasters [ 4 ]. In one study conducted on university students during the swine flu pandemic in 2009, 45% were worried about personally or a family member contracting swine flu, while 10.7% were panicking, feeling depressed, or emotionally disturbed. This study suggests that preventive measures to alleviate distress through health education and promotion are warranted [ 5 ].

During the COVID-19 pandemic, researchers worldwide have been churning out studies on its psychological effects on different populations [ 6 – 9 ]. The indirect effects of COVID-19, such as quarantine measures, the infection of family and friends, and the death of loved ones, could worsen the overall mental wellbeing of individuals [ 6 ]. Studies from 2020 to 2021 link the pandemic to emotional disturbances among those in quarantine, even going as far as giving vulnerable populations the inclination to commit suicide [ 7 , 8 ], persistent effect on mood and wellness [ 9 ], and depression and anxiety [ 10 ].

In the Philippines, a survey of 1,879 respondents measuring the psychological effects of COVID-19 during its early phase in 2020 was released. Results showed that one-fourth of respondents reported moderate-to-severe anxiety, while one-sixth reported moderate-to-severe depression [ 11 ]. In addition, other local studies in 2020 examined the mental health of frontline workers such as nurses and physicians—placing emphasis on the importance of psychological support in minimizing anxiety [ 12 , 13 ].

Since the first wave of the pandemic in 2020, risk factors that could affect specific populations’ psychological well-being have been studied [ 14 , 15 ]. A cohort study on 1,773 COVID-19 hospitalized patients in 2021 found that survivors were mainly troubled with fatigue, muscle weakness, sleep difficulties, and depression or anxiety [ 16 ]. Their results usually associate the crisis with fear, anxiety, depression, reduced sleep quality, and distress among the general population.

Moreover, the pandemic also exacerbated the condition of people with pre-existing psychiatric disorders, especially patients that live in high COVID-19 prevalence areas [ 17 ]. People suffering from mood and substance use disorders that have been infected with COVID-19 showed higher suicide risks [ 7 , 18 ]. Furthermore, a study in 2020 cited the following factors contributing to increased suicide risk: social isolation, fear of contagion, anxiety, uncertainty, chronic stress, and economic difficulties [ 19 ].

Globally, multiple studies have shown that mental health disorders among university student populations are prevalent [ 13 , 20 – 22 ]. In a 2007 survey of 2,843 undergraduate and graduate students at a large midwestern public university in the United States, the estimated prevalence of any depressive or anxiety disorder was 15.6% and 13.0% for undergraduate and graduate students, respectively [ 20 ]. Meanwhile, in a 2013 study of 506 students from 4 public universities in Malaysia, 27.5% and 9.7% had moderate and severe or extremely severe depression, respectively; 34% and 29% had moderate and severe or extremely severe anxiety, respectively [ 21 ]. In China, a 2016 meta-analysis aiming to establish the national prevalence of depression among university students analyzed 39 studies from 1995 to 2015; the meta-analysis found that the overall prevalence of depression was 23.8% across all studies that included 32,694 Chinese university students [ 23 ].

A college student’s mental status may be significantly affected by the successful fulfillment of a student’s role. A 2013 study found that acceptable teaching methods can enhance students’ satisfaction and academic performance, both linked to their mental health [ 24 ]. However, online learning poses multiple challenges to these methods [ 3 ]. Furthermore, a 2020 study found that students’ mental status is affected by their social support systems, which, in turn, may be jeopardized by the COVID-19 pandemic and the physical limitations it has imposed. Support accessible to a student through social ties to other individuals, groups, and the greater community is a form of social support; university students may draw social support from family, friends, classmates, teachers, and a significant other [ 25 , 26 ]. Among individuals undergoing social isolation and distancing during the COVID-19 pandemic in 2020, social support has been found to be inversely related to depression, anxiety, irritability, sleep quality, and loneliness, with higher levels of social support reducing the risk of depression and improving sleep quality [ 27 ]. Lastly, it has been shown in a 2020 study that social support builds resilience, a protective factor against depression, anxiety, and stress [ 28 ]. Therefore, given the protective effects of social support on psychological health, a supportive environment should be maintained in the classroom. Online learning must be perceived as an inclusive community and a safe space for peer-to-peer interactions [ 29 ]. This is echoed in another study in 2019 on depressed students who narrated their need to see themselves reflected on others [ 30 ]. Whether or not online learning currently implemented has successfully transitioned remains to be seen.

The effect of social media on students’ mental health has been a topic of interest even before the pandemic [ 31 , 32 ]. A systematic review published in 2020 found that social media use is responsible for aggravating mental health problems and that prominent risk factors for depression and anxiety include time spent, activity, and addiction to social media [ 31 ]. Another systematic review published in 2016 argues that the nature of online social networking use may be more important in influencing the symptoms of depression than the duration or frequency of the engagement—suggesting that social rumination and comparison are likely to be candidate mediators in the relationship between depression and social media [ 33 ]. However, their findings also suggest that the relationship between depression and online social networking is complex and necessitates further research to determine the impact of moderators and mediators that underly the positive and negative impact of online social networking on wellbeing [ 33 ].

Despite existing studies already painting a picture of the psychological effects of COVID-19 in the Philippines, to our knowledge, there are still no local studies contextualized to college students living in different regions of the country. Therefore, it is crucial to elicit the reasons and risk factors for depression, stress, and anxiety and determine the potential impact that online learning and social media use may have on the mental health of the said population. In turn, the findings would allow the creation of more context-specific and regionalized interventions that can promote mental wellness during the COVID-19 pandemic.

Materials and methods

The study’s general objective is to assess the mental health status of college students and determine the different factors that influenced them during the COVID-19 pandemic. Specifically, it aims:

• To describe the study population’s characteristics, categorized by their mental health status, which includes depression, anxiety, and stress.
• To determine the prevalence and risk factors of depression, anxiety, and stress among college students during the COVID-19 pandemic, quarantine, and subsequent shift to online learning.
• To estimate the effect of social media use on depression, anxiety, stress, and coping strategies towards stress among college students and examine whether participant characteristics modified these associations.
• To estimate the effect of online learning shift on depression, anxiety, stress, and coping strategies towards stress among college students and examine whether participant characteristics modified these associations.
• To determine the facilitators or stressors among college students that modified their mental health status during the COVID-19 pandemic, quarantine, and subsequent shift to online learning.

Study design

A mixed-method study design will be used to address the study’s objectives, which will include Key Informant Interviews (KIIs) and an online survey. During the quarantine period of the COVID-19 pandemic in the Philippines from April to November 2021, the study shall occur with the population amid community quarantine and an abrupt transition to online classes. Since this is the Philippines’ first study that will look at the prevalence of depression, anxiety, and stress among college students during the COVID-19 pandemic, quarantine, and subsequent shift to online learning, the online survey will be utilized for the quantitative part of the study design. For the qualitative component of the study design, KIIs will determine facilitators or stressors among college students that modified their mental health status during the quarantine period.

Study population

The Red Cross Youth (RCY), one of the Philippine Red Cross’s significant services, is a network of youth volunteers that spans the entire country, having active members in Luzon, Visayas, and Mindanao. The group is clustered into different age ranges, with the College Red Cross Youth (18–25 years old) being the study’s population of interest. The RCY has over 26,060 students spread across 20 chapters located all over the country’s three major island groups. The RCY is heterogeneously composed, with some members classified as college students and some as out-of-school youth. Given their nationwide scope, disseminating information from the national to the local level is already in place; this is done primarily through email, social media platforms, and text blasts. The research team will leverage these platforms to distribute the online survey questionnaire.

In addition, the online survey will also be open to non-members of the RCY. It will be disseminated through social media and engagements with different university administrators in the country. Stratified random sampling will be done for the KIIs. The KII participants will be equally coming from the country’s four (4) primary areas: 5–10 each from the national capital region (NCR), Luzon, Visayas, and Mindanao, including members and non-members of the RCY.

Inclusion and exclusion criteria

The inclusion criteria for the online survey will include those who are 18–25 years old, currently enrolled in a university, can provide consent for the study, and are proficient in English or Filipino. The exclusion criteria will consist of those enrolled in graduate-level programs (e.g., MD, JD, Master’s, Doctorate), out-of-school youth, and those whose current curricula involve going on duty (e.g., MDs, nursing students, allied medical professions, etc.). The inclusion criteria for the KIIs will include online survey participants who are 18–25 years old, can provide consent for the study, are proficient in English or Filipino, and have access to the internet.

Sample size

A continuity correction method developed by Fleiss et al. (2013) was used to calculate the sample size needed [ 34 ]. For a two-sided confidence level of 95%, with 80% power and the least extreme odds ratio to be detected at 1.4, the computed sample size was 1890. With an adjustment for an estimated response rate of 90%, the total sample size needed for the study was 2,100. To achieve saturation for the qualitative part of the study, 20 to 40 participants will be randomly sampled for the KIIs using the respondents who participated in the online survey [ 35 ].

Study procedure

Self-administered questionnaire..

The study will involve creating, testing, and distributing a self-administered questionnaire (SAQ). All eligible study participants will answer the SAQ on socio-demographic factors such as age, sex, gender, sexual orientation, residence, household income, socioeconomic status, smoking status, family history of mental health, and COVID-19 sickness of immediate family members or friends. The two validated survey tools, Depression, Anxiety, and Stress Scale (DASS-21) and Brief-COPE, will be used for the mental health outcome assessment [ 36 – 39 ]. The DASS-21 will measure the negative emotional states of depression, anxiety, and stress [ 40 ], while the Brief-COPE will measure the students’ coping strategies [ 41 ].

For the exposure assessment of the students to social media and shift to online learning, the total time spent on social media (TSSM) per day will be ascertained by querying the participants to provide an estimated time spent daily on social media during and after their online classes. In addition, students will be asked to report their use of the eight commonly used social media sites identified at the start of the study. These sites include Facebook, Twitter, Instagram, LinkedIn, Pinterest, TikTok, YouTube, and social messaging sites Viber/WhatsApp and Facebook Messenger with response choices coded as "(1) never," "(2) less often," "(3) every few weeks," "(4) a few times a week," and “(5) daily” [ 42 – 44 ]. Furthermore, a global frequency score will be calculated by adding the response scores from the eight social media sites. The global frequency score will be used as an additional exposure marker of students to social media [ 45 ]. The shift to online learning will be assessed using questions that will determine the participants’ satisfaction with online learning. This assessment is comprised of 8 items in which participants will be asked to respond on a 5-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree.’

The online survey will be virtually distributed in English using the Qualtrics XM™ platform. Informed consent detailing the purpose, risks, benefits, methods, psychological referrals, and other ethical considerations will be included before the participants are allowed to answer the survey. Before administering the online survey, the SAQ shall undergo pilot testing among twenty (20) college students not involved with the study. It aims to measure total test-taking time, respondent satisfaction, and understandability of questions. The survey shall be edited according to the pilot test participant’s responses. Moreover, according to the Philippines’ Data Privacy Act, all the answers will be accessible and used only for research purposes.

Key informant interviews.

The research team shall develop the KII concept note, focusing on the extraneous factors affecting the student’s mental health, behaviors, and coping mechanism. Some salient topics will include current stressors (e.g., personal, academic, social), emotional reactions to these stressors, and how they wish to receive support in response to these stressors. The KII will be facilitated by a certified psychologist/psychiatrist/social scientist and research assistants using various online video conferencing software such as Google Meet, Skype, or Zoom. All the KIIs will be recorded and transcribed for analysis. Furthermore, there will be a debriefing session post-KII to address the psychological needs of the participants. Fig 1 presents the diagrammatic flowchart of the study.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0267555.g001

Data analyses

Quantitative data..

Descriptive statistics will be calculated, including the prevalence of mental health outcomes such as depression, anxiety, stress, and coping strategies. In addition, correlation coefficients will be estimated to assess the relations among the different mental health outcomes, covariates, and possible risk factors.

Several study characteristics as effect modifiers will also be assessed, including sex, gender, sexual orientation, family income, smoking status, family history of mental health, and Covid-19. We will include interaction terms between the dichotomized modifier variable and markers of social media use (total TSSM and global frequency score) and shift to online learning in the models. The significance of the interaction terms will be evaluated using the likelihood ratio test. All the regression analyses will be done in R ( http://www.r-project.org ). P values ≤ 0.05 will be considered statistically significant.

Qualitative data.

After transcribing the interviews, the data transcripts will be analyzed using NVivo 1.4.1 software [ 50 ] by three research team members independently using the inductive logic approach in thematic analysis: familiarizing with the data, generating initial codes, searching for themes, reviewing the themes, defining and naming the themes, and producing the report [ 51 ]. Data familiarization will consist of reading and re-reading the data while noting initial ideas. Additionally, coding interesting features of the data will follow systematically across the entire dataset while collating data relevant to each code. Moreover, the open coding of the data will be performed to describe the data into concepts and themes, which will be further categorized to identify distinct concepts and themes [ 52 ].

The three researchers will discuss the results of their thematic analyses. They will compare and contrast the three analyses in order to come up with a thematic map. The final thematic map of the analysis will be generated after checking if the identified themes work in relation to the extracts and the entire dataset. In addition, the selection of clear, persuasive extract examples that will connect the analysis to the research question and literature will be reviewed before producing a scholarly report of the analysis. Additionally, the themes and sub-themes generated will be assessed and discussed in relevance to the study’s objectives. Furthermore, the gathering and analyzing of the data will continue until saturation is reached. Finally, pseudonyms will be used to present quotes from qualitative data.

Data triangulation.

Data triangulation using the two different data sources will be conducted to examine the various aspects of the research and will be compared for convergence. This part of the analysis will require listing all the relevant topics or findings from each component of the study and considering where each method’s results converge, offer complementary information on the same issue, or appear to contradict each other. It is crucial to explicitly look for disagreements between findings from different data collection methods because exploration of any apparent inter-method discrepancy may lead to a better understanding of the research question [ 53 , 54 ].

Data management plan.

The Project Leader will be responsible for overall quality assurance, with research associates and assistants undertaking specific activities to ensure quality control. Quality will be assured through routine monitoring by the Project Leader and periodic cross-checks against the protocols by the research assistants. Transcribed KIIs and the online survey questionnaire will be used for recording data for each participant in the study. The project leader will be responsible for ensuring the accuracy, completeness, legibility, and timeliness of the data captured in all the forms. Data captured from the online survey or KIIs should be consistent, clarified, and corrected. Each participant will have complete source documentation of records. Study staff will prepare appropriate source documents and make them available to the Project Leader upon request for review. In addition, study staff will extract all data collected in the KII notes or survey forms. These data will be secured and kept in a place accessible to the Project Leader. Data entry and cleaning will be conducted, and final data cleaning, data freezing, and data analysis will be performed. Key informant interviews will always involve two researchers. Where appropriate, quality control for the qualitative data collection will be assured through refresher KII training during research design workshops. The Project Leader will check through each transcript for consistency with agreed standards. Where translations are undertaken, the quality will be assured by one other researcher fluent in that language checking against the original recording or notes.

Ethics approval.

The study shall abide by the Principles of the Declaration of Helsinki (2013). It will be conducted along with the Guidelines of the International Conference on Harmonization-Good Clinical Practice (ICH-GCP), E6 (R2), and other ICH-GCP 6 (as amended); National Ethical Guidelines for Health and Health-Related Research (NEGHHRR) of 2017. This protocol has been approved by the University of the Philippines Manila Research Ethics Board (UPMREB 2021-099-01 dated March 25, 2021).

The main concerns for ethics were consent, data privacy, and subject confidentiality. The risks, benefits, and conflicts of interest are discussed in this section from an ethical standpoint.

Recruitment.

The participants will be recruited to answer the online SAQ voluntarily. The recruitment of participants for the KIIs will be chosen through stratified random sampling using a list of those who answered the online SAQ; this will minimize the risk of sampling bias. In addition, none of the participants in the study will have prior contact or association with the researchers. Moreover, power dynamics will not be contacted to recruit respondents. The research objectives, methods, risks, benefits, voluntary participation, withdrawal, and respondents’ rights will be discussed with the respondents in the consent form before KII.

Informed consent will be signified by the potential respondent ticking a box in the online informed consent form and the voluntary participation of the potential respondent to the study after a thorough discussion of the research details. The participant’s consent is voluntary and may be recanted by the participant any time s/he chooses.

Data privacy.

All digital data will be stored in a cloud drive accessible only to the researchers. Subject confidentiality will be upheld through the assignment of control numbers and not requiring participants to divulge the name, address, and other identifying factors not necessary for analysis.

Compensation.

No monetary compensation will be given to the participants, but several tokens will be raffled to all the participants who answered the online survey and did the KIIs.

This research will pose risks to data privacy, as discussed and addressed above. In addition, there will be a risk of social exclusion should data leaks arise due to the stigma against mental health. This risk will be mitigated by properly executing the data collection and analysis plan, excluding personal details and tight data privacy measures. Moreover, there is a risk of psychological distress among the participants due to the sensitive information. This risk will be addressed by subjecting the SAQ and the KII guidelines to the project team’s psychiatrist’s approval, ensuring proper communication with the participants. The KII will also be facilitated by registered clinical psychologists/psychiatrists/social scientists to ensure the participants’ appropriate handling; there will be a briefing and debriefing of the participants before and after the KII proper.

Participation in this study will entail health education and a voluntary referral to a study-affiliated psychiatrist, discussed in previous sections. Moreover, this would contribute to modifications in targeted mental-health campaigns for the 18–25 age group. Summarized findings and recommendations will be channeled to stakeholders for their perusal.

Dissemination.

The results will be actively disseminated through conference presentations, peer-reviewed journals, social media, print and broadcast media, and various stakeholder activities.

This study protocol rationalizes the examination of the mental health of the college students in the Philippines during the COVID-19 pandemic as the traditional face-to-face classes transitioned to online and modular classes. The pandemic that started in March 2020 is now stretching for more than a year in which prolonged lockdown brings people to experience social isolation and disruption of everyday lifestyle. There is an urgent need to study the psychosocial aspects, particularly those populations that are vulnerable to mental health instability. In the Philippines, where community quarantine is still being imposed across the country, college students face several challenges amidst this pandemic. The pandemic continues to escalate, which may lead to fear and a spectrum of psychological consequences. Universities and colleges play an essential role in supporting college students in their academic, safety, and social needs. The courses of activities implemented by the different universities and colleges may significantly affect their mental well-being status. Our study is particularly interested in the effect of online classes on college students nationwide during the pandemic. The study will estimate this effect on their mental wellbeing since this abrupt transition can lead to depression, stress, or anxiety for some students due to insufficient time to adjust to the new learning environment. The role of social media is also an important exposure to some college students [ 55 , 56 ]. Social media exposure to COVID-19 may be considered a contributing factor to college students’ mental well-being, particularly their stress, depression, and anxiety [ 57 , 58 ]. Despite these known facts, little is known about the effect of transitioning to online learning and social media exposure on the mental health of college students during the COVID-19 pandemic in the Philippines. To our knowledge, this is the first study in the Philippines that will use a mixed-method study design to examine the mental health of college students in the entire country. The online survey is a powerful platform to employ our methods.

Additionally, our study will also utilize a qualitative assessment of the college students, which may give significant insights or findings of the experiences of the college students during these trying times that cannot be captured on our online survey. The thematic findings or narratives from the qualitative part of our study will be triangulated with the quantitative analysis for a more robust synthesis. The results will be used to draw conclusions about the mental health status among college students during the pandemic in the country, which will eventually be used to implement key interventions if deemed necessary. A cross-sectional study design for the online survey is one of our study’s limitations in which contrasts will be mainly between participants at a given point of time. In addition, bias arising from residual or unmeasured confounding factors cannot be ruled out.

The COVID-19 pandemic and its accompanying effects will persistently affect the mental wellbeing of college students. Mental health services must be delivered to combat mental instability. In addition, universities and colleges should create an environment that will foster mental health awareness among Filipino college students. The results of our study will tailor the possible coping strategies to meet the specific needs of college students nationwide, thereby promoting psychological resilience.

Understanding COVID-19 dynamics and the effects of interventions in the Philippines: A mathematical modelling study

Affiliations.

• 1 Department of Biology, University of Hawaii at Manoa, Hawaii, USA.
• 2 Department of Mathematics, Ateneo de Manila University, Quezon City, Philippines.
• 3 Department of Information Systems and Computer Science, Ateneo de Manila University, Quezon City, Philippines.
• 4 National Telehealth Center, National Institutes of Health, University of the Philippines, Manila, Philippines.
• 5 School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia.
• 6 School of Population Health and Community Medicine, University of New South Wales, Sydney, Australia.
• 7 The Kirby Institute, University of New South Wales Sydney, Sydney, Australia.
• 8 World Health Organization Regional Office for the Western Pacific, Manila, Philippines.
• 9 Australian Institute of Tropical Health and Medicine, James Cook University, Queensland, Australia.
• PMID: 34308400
• PMCID: PMC8279002
• DOI: 10.1016/j.lanwpc.2021.100211

Background: COVID-19 initially caused less severe outbreaks in many low- and middle-income countries (LMIC) compared with many high-income countries, possibly because of differing demographics, socioeconomics, surveillance, and policy responses. Here, we investigate the role of multiple factors on COVID-19 dynamics in the Philippines, a LMIC that has had a relatively severe COVID-19 outbreak.

Methods: We applied an age-structured compartmental model that incorporated time-varying mobility, testing, and personal protective behaviors (through a "Minimum Health Standards" policy, MHS) to represent the first wave of the Philippines COVID-19 epidemic nationally and for three highly affected regions (Calabarzon, Central Visayas, and the National Capital Region). We estimated effects of control measures, key epidemiological parameters, and interventions.

Findings: Population age structure, contact rates, mobility, testing, and MHS were sufficient to explain the Philippines epidemic based on the good fit between modelled and reported cases, hospitalisations, and deaths. The model indicated that MHS reduced the probability of transmission per contact by 13-27%. The February 2021 case detection rate was estimated at ~8%, population recovered at ~9%, and scenario projections indicated high sensitivity to MHS adherence.

Interpretation: COVID-19 dynamics in the Philippines are driven by age, contact structure, mobility, and MHS adherence. Continued compliance with low-cost MHS should help the Philippines control the epidemic until vaccines are widely distributed, but disease resurgence may be occurring due to a combination of low population immunity and detection rates and new variants of concern.

Keywords: CDR, Case detection rate; COVID-19; COVID-19, Coronavirus disease 2019; HIC, High-income countries; ICU, Intensive care unit; LMIC; LMIC, Low- and middle-income countries; MHS, Minimum Health Standards; Minimum Health Standards policy; NPI, Non-pharmaceutical intervention; Philippines; SEIR.

© 2021 Published by Elsevier Ltd.

Grants and funding

• 001/WHO_/World Health Organization/International

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 04 April 2022

Economic losses from COVID-19 cases in the Philippines: a dynamic model of health and economic policy trade-offs

• Elvira P. de Lara-Tuprio 1 ,
• Maria Regina Justina E. Estuar 2 ,
• Joselito T. Sescon 3 ,
• Cymon Kayle Lubangco   ORCID: orcid.org/0000-0002-1292-4687 3 ,
• Rolly Czar Joseph T. Castillo 3 ,
• Timothy Robin Y. Teng 1 ,
• Lenard Paulo V. Tamayo 2 ,
• Jay Michael R. Macalalag 4 &
• Gerome M. Vedeja 3  

Humanities and Social Sciences Communications volume  9 , Article number:  111 ( 2022 ) Cite this article

41k Accesses

8 Citations

19 Altmetric

Metrics details

• Science, technology and society

The COVID-19 pandemic forced governments globally to impose lockdown measures and mobility restrictions to curb the transmission of the virus. As economies slowly reopen, governments face a trade-off between implementing economic recovery and health policy measures to control the spread of the virus and to ensure it will not overwhelm the health system. We developed a mathematical model that measures the economic losses due to the spread of the disease and due to different lockdown policies. This is done by extending the subnational SEIR model to include two differential equations that capture economic losses due to COVID-19 infection and due to the lockdown measures imposed by the Philippine government. We then proceed to assess the trade-off policy space between health and economic measures faced by the Philippine government. The study simulates the cumulative economic losses for 3 months in 8 scenarios across 5 regions in the country, including the National Capital Region (NCR), to capture the trade-off mechanism. These scenarios present the various combinations of either retaining or easing lockdown policies in these regions. Per region, the trade-off policy space was assessed through minimising the 3-month cumulative economic losses subject to the constraint that the average health-care utilisation rate (HCUR) consistently falls below 70%, which is the threshold set by the government before declaring that the health system capacity is at high risk. The study finds that in NCR, a policy trade-off exists where the minimum cumulative economic losses comprise 10.66% of its Gross Regional Domestic Product. Meanwhile, for regions that are non-adjacent to NCR, a policy that hinges on trade-off analysis does not apply. Nevertheless, for all simulated regions, it is recommended to improve and expand the capacity of the health system to broaden the policy space for the government in easing lockdown measures.

Similar content being viewed by others

Modelling COVID-19 pandemic control strategies in metropolitan and rural health districts in New South Wales, Australia

Mathematical modeling of COVID-19 in 14.8 million individuals in Bahia, Brazil

Cost-effectiveness analysis of COVID-19 intervention policies using a mathematical model: an optimal control approach

Introduction.

The Philippine population of 110 million comprises a relatively young population. On May 22, 2021, the number of confirmed COVID-19 cases reported in the country is 1,171,403 with 55,531 active cases, 1,096,109 who recovered, and 19,763 who died. As a consequence of the pandemic, the real gross domestic product (GDP) contracted by 9.6% year-on-year in 2020—the sharpest decline since the Philippine Statistical Agency (PSA) started collecting data on annual growth rates in 1946 (Bangko Sentral ng Pilipinas, 2021 ). The strictest lockdown imposed from March to April 2020 had the most severe repercussions to the economy, but restrictions soon after have generally eased on economic activities all over the country. However, schools at all levels remain closed and minimum restrictions are still imposed in business operations particularly in customer accommodation capacity in service establishments.

The government is poised for a calibrated reopening of business, mass transportation, and the relaxation of age group restrictions. The government expects a strong recovery before the end of 2021, when enough vaccines have been rolled out against COVID-19. However, the economic recovery plan and growth targets at the end of the year are put in doubt with the first quarter of 2021 growth rate of GDP at -4.2%. This is exacerbated by the surge of cases in March 2021 that took the National Capital Region (NCR) and contiguous provinces by surprise, straining the hospital bed capacity of the region beyond its limits. The government had to reinforce stricter lockdown measures and curfew hours to stem the rapid spread of the virus. The country’s economic development authority proposes to ensure hospitals have enough capacity to allow the resumption of social and economic activities (National Economic and Development Authority, 2020 ). This is justified by pointing out that the majority of COVID-19 cases are mild and asymptomatic.

Efforts in monitoring and mitigating the spread of COVID-19 requires understanding the behaviour of the disease through the development of localised disease models operationalized as an ICT tool accessible to policymakers. FASSSTER is a scenario-based disease surveillance and modelling platform designed to accommodate multiple sources of data as input allowing for a variety of disease models and analytics to generate meaningful information to its stakeholders (FASSSTER, 2020 ). FASSSTER’s module on COVID-19 currently provides information and forecasts from national down to city/municipality level that are used for decision-making by individual local government units (LGUs) and also by key government agencies in charge of the pandemic response.

In this paper, we develop a mathematical model that measures the economic losses due to the spread of the disease and due to different lockdown policies to contain the disease. This is done by extending the FASSSTER subnational Susceptible-Exposed-Infectious-Recovered (SEIR) model to include two differential equations that capture economic losses due to COVID-19 infection and due to the lockdown measures imposed by the Philippine government. We then proceed to assess the trade-off policy space faced by the Philippine government given the policy that health-care utilisation rate must not be more than 70%, which is the threshold set by the government before declaring that the health system capacity is at high risk.

We simulate the cumulative economic losses for 3 months in 8 scenarios across 5 regions in the country, including the National Capital Region (NCR) to capture the trade-off mechanism. These 8 scenarios present the various combinations of either retaining or easing lockdown policies in these regions. Per region, the trade-off policy space was assessed through minimising the 3-month cumulative economic losses subject to the constraint that the average health-care utilisation rate (HCUR) consistently falls below 70%. The study finds that in NCR, a policy trade-off exists where the minimum economic losses below the 70% average HCUR comprise 10.66% of its Gross Regional Domestic Product. Meanwhile, for regions that are non-adjacent to NCR, a policy that hinges on trade-off analysis does not apply. Nevertheless, for all simulated regions, it is recommended to improve and expand the capacity of the health system to broaden the policy space for the government in easing lockdown measures.

The sections of the paper proceed as follows: the first section reviews the literature, the second section explains the FASSSTER SEIR model, the third section discusses the economic dynamic model, the fourth section specifically explains the parameters used in the economic model, the fifth section briefly lays out the policy trade-off model, the sixth discusses the methods used in implementing the model, the seventh section presents the results of the simulations, the eighth section discusses and interprets the results, and the final section presents the conclusion.

Review of related literature

Overview of the economic shocks of pandemics.

The onslaught of the Coronavirus Disease 2019 (COVID-19) pandemic since 2020 has disrupted lifestyles and livelihoods as governments restrict mobility and economic activity in their respective countries. Unfortunately, this caused a –3.36% decline in the 2020 global economy (World Bank, 2022 ), which will have pushed 71 million people into extreme poverty (World Bank, 2020 ; 2021 ).

As an economic phenomenon, pandemics may be classified under the typologies of disaster economics. Particularly, a pandemic’s impacts may be classified according to the following (Benson and Clay, 2004 ; Noy et al., 2020 ; Keogh-Brown et al., 2010 ; 2020 ; McKibbin and Fernando, 2020 ; Verikios et al., 2012 ): (a) direct impacts, where pandemics cause direct labour supply shocks due to mortality and infection; (b) indirect impacts on productivity, firm revenue, household income, and other welfare effects, and; (c) macroeconomic impacts of a pandemic.

For most pandemic scenarios, social distancing and various forms of lockdowns imposed by countries around the world had led to substantial disruptions in the supply-side of the economy with mandatory business closures (Maital and Barzani, 2020 ; Keogh-Brown et al., 2010 ). Social distancing will have contracted labour supply as well, thus contributing to contractions in the macroeconomy (Geard et al., 2020 ; Keogh-Brown et al., 2010 ). Thus, in general, the literature points to a pandemic’s impacts on the supply- and demand-side, as well as the displacement of labour supply; thus, resulting in lower incomes (Genoni et al., 2020 ; Hupkau et al., 2020 ; United Nations Development Programme, 2021 ). Often, these shocks result from the lockdown measures; thus, a case of a trade-off condition between economic losses and the number of COVID-19 casualties.

Static simulations for the economic impacts of a pandemic

The typologies above are evident in the analyses and simulations on welfare and macroeconomic losses related to a pandemic. For instance, computable general equilibrium (CGE) and microsimulation analyses for the 2009 H1N1 pandemic and the COVID-19 pandemic showed increases in inequities, welfare losses, and macroeconomic losses due to lockdown and public prevention strategies (Cereda et al., 2020 ; Keogh-Brown et al., 2020 ; Keogh-Brown et al., 2010 ). Public prevention-related labour losses also comprised at most 25% of the losses in GDP in contrast with health-related losses, which comprised only at most 17% of the losses in GDP.

Amidst the COVID-19 pandemic in Ghana, Amewu et al. ( 2020 ) find in a social accounting matrix-based analysis that the industry and services sectors will have declined by 26.8% and 33.1%, respectively. Other studies investigate the effects of the pandemic on other severely hit sectors such as the tourism sector. Pham et al. ( 2021 ) note that a reduction in tourism demand in Australia will have caused a reduction in income of tourism labourers. Meanwhile, in a static CGE-microsimulation model by Laborde, Martin, and Vos ( 2021 ), they show that as the global GDP will have contracted by 5% following the reduction in labour supply, this will have increased global poverty by 20%, global rural poverty by 15%, poverty in sub-Saharan Africa by 23%, and in South Asia by 15%.

However, due to the static nature of these analyses, the clear trade-off between economic and health costs under various lockdown scenarios is a policy message that remains unexplored, as the simulations above only explicitly tackle a pandemic’s macroeconomic effects. This gap is mostly due to these studies’ usage of static SAM- and CGE-based analyses.

Dynamic simulations for the economic impacts of a pandemic

An obvious advantage of dynamic models over static approaches in estimating the economic losses from the pandemic is the capacity to provide forward-looking insights that have practical use in policymaking. Epidemiological models based on systems of differential equations explicitly model disease spread and recovery as movements of population across different compartments. These compartmental models are useful in forecasting the number of infected individuals, critically ill patients, death toll, among others, and thus are valuable in determining the appropriate intervention to control epidemics.

To date, the Susceptible-Infectious-Recovered (SIR) and SEIR models are among the most popular compartmental models used to study the spread of diseases. In recent years, COVID-19 has become an important subject of more recent mathematical modelling studies. Many of these studies deal with both application and refinement of both SIR and SEIR to allow scenario-building, conduct evaluation of containment measures, and improve forecasts. These include the integration of geographical heterogeneities, the differentiation between isolated and non-isolated cases, and the integration of interventions such as reducing contact rate and isolation of active cases (Anand et al., 2020 ; Chen et al., 2020 ; Hou et al., 2020 ; Peng et al., 2020 ; Reno et al., 2020 ).

Typical epidemiological models may provide insight on the optimal lockdown measure to reduce the transmissibility of a virus. However, there is a need to derive calculations on economic impacts from the COVID-19 case projections to arrive at a conclusion on the optimal frontier from the trade-off between health and economic losses. In Goldsztejn, Schwartzman and Nehorai ( 2020 ), an economic model that measures lost economic productivity due to the pandemic, disease containment measures and economic policies is integrated into an SEIR model. The hybrid model generates important insight on the trade-offs between short-term economic gains in terms of productivity, and the continuous spread of the disease, which in turn informs policymakers on the appropriate containment policies to be implemented.

This approach was further improved by solving an optimal control of multiple group SIR model to find the best way to implement a lockdown (Acemoglu et al., 2020 ). Noting the trade-offs between economic outcomes and spread of disease implied in lockdown policies, Acemoglu et al. ( 2020 ) find that targeted lockdown yields the best result in terms of economic losses and saving lives. However, Acemoglu et al. ( 2020 ) only determine the optimal lockdown policy and their trade-off analysis through COVID-associated fatalities. Kashyap et al. ( 2020 ) note that hospitalisations may be better indicators for lockdown and, as a corollary, reopening policies.

Gaps in the literature

With the recency of the pandemic, there is an increasing but limited scholarship in terms of jointly analysing the losses brought about by the pandemic on health and the economy. On top of this, the literature clearly has gaps in terms of having a trade-off model that captures the context of low- and middle-income countries. Devising a trade-off model for said countries is an imperative given the structural and capability differences of these countries from developed ones in terms of responding to the pandemic. Furthermore, the literature has not explicitly looked into the trade-off between economic losses and health-care system capacities, both at a national and a subnational level.

With this, the paper aims to fill these gaps with the following. Firstly, we extend FASSSTER’s subnational SEIR model to capture the associated economic losses given various lockdown scenarios at a regional level. Then, we construct an optimal policy decision trade-off between the health system and the economy in the Philippines’ case at a regional level. From there, we analyse the policy implications across the different regions given the results of the simulations.

The FASSSTER SEIR model

The FASSSTER model for COVID-19 uses a compartmental model to describe the dynamics of disease transmission in a community, and it is expressed as a system of ordinary differential equations (Estadilla et al., 2021 ):

where β  =  β 0 (1– λ ), \(\alpha _a = \frac{c}{\tau }\) , \(\alpha _s = \frac{{1 - c}}{\tau }\) , and N ( t ) =  S ( t ) =  E ( t ) +  I a ( t ) +  I s ( t ) +  C ( t ) +  R ( t ).

The six compartments used to divide the entire population, namely, susceptible ( S ), exposed ( E ), infectious but asymptomatic ( I a ), infectious and symptomatic ( I s ), confirmed ( C ), and recovered ( R ), indicate the status of the individuals in relation to the disease. Compartment S consists of individuals who have not been infected with COVID-19 but may acquire the disease once exposed to infectious individuals. Compartment E consists of individuals who have been infected, but not yet capable of transmitting the disease to others. The infectious members of the population are split into two compartments, I a and I s , based on the presence of disease symptoms. These individuals may eventually transition to compartment C once they have been detected, in which case they will be quarantined and receive treatment. The individuals in the C compartment are commonly referred to as active cases. Finally, recovered individuals who have tested negative or have undergone the required number of days in isolation will move out to the R compartment. Given that there had only been rare instances of reinfection (Gousseff et al., 2020 ), the FASSSTER model assumes that recovered individuals have developed immunity from the disease. A description of the model parameters can be found in Supplementary Table S1 .

The model has several nonnegative parameters that govern the movement of individuals along the different compartments. The parameter β represents the effective transmission rate, and it is expressed as a product of the disease transmission rate β 0 and reduction factor 1 −  λ . The rate β 0 is derived from an assumed reproduction number R 0 , which varies depending on the region. The parameter λ reflects the effect of mobility restrictions such as lockdowns and compliance of the members of the population to minimum health standards (such as social distancing, wearing of face masks etc.). In addition, the parameter ψ captures the relative infectiousness of asymptomatic individuals in relation to those who exhibit symptoms.

The incubation period τ and fraction of asymptomatic cases c are used to derive the transfer rates α α and α s from the exposed compartment to I a and I s compartments, respectively. Among those who are infectious and asymptomatic, a portion of them is considered pre-symptomatic, and hence will eventually develop symptoms of the disease; this is reflected in the parameter ω. The respective detection rates δ a and δ s of asymptomatic and symptomatic infectious individuals indicate the movement from the undetected infectious compartment to the confirmed compartment. These parameters capture the entire health system capacity to prevent-detect-isolate-treat-reintegrate (PDITR) COVID-19 cases; hence, they will henceforth be referred to as HSC parameters. The recoveries of infectious asymptomatic individuals and among the active cases occur at the corresponding rates θ and r . Death rates due to the disease, on the other hand, are given by ∈ I and ∈ T for the infectious symptomatic and confirmed cases, respectively.

Aside from the aforementioned parameters, the model also utilises parameters not associated with the COVID-19 disease, such as the recruitment rate A into the susceptible population. This parameter represents the birth rate of the population and is assumed to be constant. In addition, a natural death rate per unit of time is applied to all compartments in the model, incorporating the effect of non-COVID-19 related deaths in the entire population.

Economic dynamic model

The trade-off model aims to account for the incurred economic losses following the rise and fall of the number of COVID-19 cases in the country and the implementation of various lockdown measures. The model variables are estimated per day based on the SEIR model estimate of daily cases and are defined as follows. Let Y E ( t ) be the economic loss due to COVID-19 infections (hospitalisation, isolation, and death of infected individuals) and Y E ( t ) be the economic loss due to the implemented lockdown at time t . The dynamics of each economic variable through time is described by an ordinary differential equation. Since each equation depends only on the values of the state variables of the epidemiological model, then it is possible to obtain a closed form solution.

Economic loss due to COVID-19 infections (hospitalisation, isolation, and health)

The economic loss due to hospitalisation, isolation, and death Y E is described by the following differential equation:

where z  = annual gross value added of each worker (assumed constant for all future years and for all ages), w  = daily gross value added, ι i  = % population with ages 0–14 ( i  = 1), and labour force with ages 15–34 ( i  = 2), 35–49 ( i  = 3) and 50–64 ( i  = 4), s r  = social discount rate, κ  = employed to population ratio, T i  = average remaining productive years for people in age bracket i , i  = 1, 2, 3, 4, and T 5  = average age of deaths from 0–14 years old age group. Note that the above formulation assumes that the young population 0–14 years old will start working at age 15, and that they will work for T 1 −15 years.

Solving Eq. ( 7 ), we obtain for t  ≥ 0,

In this equation, the terms on the right-hand side are labelled as (A), (B), and (C). Term (A) is the present value of all future gross value added of 0–14 years old who died due to COVID-19 at time t . Similarly, term (B) is the present value of all future gross value added of people in the labour force who died due to COVID-19 at time t . Term (C) represents the total gross value added lost at time t due to sickness and isolation.

The discounting factors and the population age group shares in (A) and (B) can be simplified further into K 1 and K 2 , where \(K_1 = \iota _1\left( {\frac{{\left( {s_r + 1} \right)^{T_1 + T_5 - 13} - \left( {s_r + 1} \right)}}{{s_r\left( {s_r + 1} \right)^{T_1 + 1}}}} \right)\) and \(K_2 = \mathop {\sum}\nolimits_{i = 2}^4 {\iota _i\left( {\frac{{\left( {s_r + 1} \right)^{T_i + 2} - \left( {s_r + 1} \right)}}{{s_r\left( {s_r + 1} \right)^{T_i + 1}}}} \right)}\) . By letting L 1  = z( K 1  +  K 2 ) ∈ I  +  κw (1 –  ∈ I ) and L 2  = z( K 1  +  K 2 ) ∈ T  +  κw (1 –  ∈ T ), we have:

Economic losses due to lockdown policies

Equation ( 7 ) measures the losses due mainly to sickness and death from COVID-19. The values depend on the number of detected and undetected infected individuals, C and I s . The other losses sustained by the other part of the population are due to their inability to earn because of lockdown policies. This is what the next variable Y L represents, whose dynamics is given by the differential equation

where φ  = the displacement rate, and κ and w are as defined previously.

Solving the differential equation, then

Note that [ S ( t ) +  E ( t ) +  I a ( t ) +  R ( t )] is the rest of the population at time t , i.e., other than the active and infectious symptomatic cases. Multiplying this by κ and the displacement rate φ yields the number of employed people in this population who are displaced due to the lockdown policy. Thus, κwφ [ S ( t ) +  E ( t ) +  I a ( t ) +  R ( t )] is the total foregone income due to the lockdown policy.

Economic model parameters

The values of the parameters were derived from a variety of sources. The parameters for employment and gross value added were computed based on the data from the Philippine Statistics Authority ( 2021 , 2020 , 2019a , 2019b ), the Department of Health’s Epidemiology Bureau (DOH-EB) ( 2020 ), the Department of Trade and Industry (DTI) ( 2020a , 2020b ) and the National Economic Development Authority (NEDA) ( 2016 ) (See Supplementary Tables S2 and S3 for the summary of economic parameters).

Parameters determined from related literature

We used the number of deaths from the data of the DOH-EB ( 2020 ) to disaggregate the long-term economic costs of the COVID-related deaths into age groups. Specifically, the COVID-related deaths were divided according to the following age groups: (a) below 15 years old, (b) 15 to 34 years old, (c) 35 to 49 years old, and (d) 50 to 64 years old. The average remaining years for these groups were computed directly from the average age of death of the respective cluster. Finally, we used the social discount rate as determined by NEDA ( 2016 ) to get the present value of the stream of foregone incomes of those who died from the disease.

Parameters estimated from local data

The foregone value added due to labour displacement was estimated as the amount due to workers in a geographic area who were unable to work as a result of strict lockdown measures. It was expected to contribute to the total value added in a given year if the area they reside or work in has not been locked down.

The employed to population ratio κ i for each region i was computed as

where e i was total employment in region i , and Pi was the total population in the region. Both e i and Pi were obtained from the quarterly labour force survey and the census, respectively (Philippine Statistics Authority, 2020 , 2019a , 2019b ).

The annual gross value added per worker z i for region i was computed as

where g ji was the share of sector j in total gross value added of region i , GVA ji was the gross value added of sector j in region i (Philippine Statistics Authority, 2021 ), and e ji was the number of employed persons in sector j of region i . If individuals worked for an average of 22.5 days for each month for 12 months in a year, then the daily gross value added per worker in region i was given by

Apart from this, labour displacement rates were calculated at regional level. The rates are differentiated by economic reopening scenarios from March 2020 to September 2020, from October 2020 to February 2021, and from March 2021 onwards (Department of Trade and Industry, 2020a , 2020b , 2021 ). These were used to simulate the graduate reopening of the economy. From the country’s labour force survey, each representative observation j in a region i is designated with a numerical value in accordance with the percentage operating capacity of the sector where j works in. Given the probability weights p ji , the displacement rate φ i for region i was calculated by

where x ji served as the variable representing the maximum operating capacity designated for j ’s sector of work.

Policy trade-off model

The trade-off between economic losses and health measures gives the optimal policy subject to a socially determined constraint. From the literature, it was pointed out that the optimal policy option would be what minimises total economic losses subject to the number of deaths at a given time (Acemoglu et al., 2020 ). However, for the Philippines’ case, lockdown restrictions are decided based on the intensive care unit and health-care utilisation rate (HCUR). The health system is said to reach its critical levels if the HCUR breaches 70% of the total available bed capacity in intensive care units. Once breached, policymakers would opt to implement stricter quarantine measures.

Given these, a policy mix of various quarantine restrictions may be chosen for as long as it provides the lowest amount of economic losses subject to the constraint that the HCUR threshold is not breached. Since economic losses are adequately captured by the sum of infection-related and lockdown-related losses, Y E ( t ) +  Y L ( t ), then policy option must satisfy the constrained minimisation below:

where the objective function is evaluated from the initial time value t 0 to T .

The COVID-19 case information data including the date, location transformed into the Philippine Standard Geographic Code (PSGC), case count, and date reported were used as input to the model. Imputation using predictive mean matching uses the mice package in the R programming language. It was performed to address data gaps including the date of onset, date of specimen collection, date of admission, date of result, and date of recovery. Population data was obtained from the country’s Census of Population and Housing of 2015. The scripts to implement the FASSSTER SEIR model were developed using core packages in R including optimParallel for parameter estimation and deSolve for solving the ordinary differential equations. The output of the model is fitted to historical data by finding the best value of the parameter lambda using the L-BFGS-B method under the optim function and the MSE as measure of fitness (Byrd et al., 1995 ). The best value of lambda is obtained by performing parameter fitting with several bootstraps for each region, having at least 50 iterations until a correlation threshold of at least 90% is achieved. The output generated from the code execution contains values of the different compartments at each point in time. From these, the economic variables Y E ( t ) and Y L ( t ) were evaluated using the formulas in Eq. ( 7 ) and ( 8 ) in their simplified forms, and the parameter and displacement rate values corresponding to the implemented lockdown scenario (Fig. 1 ).

The different population states are represented by the compartments labelled as susceptible (S), exposed (E), infectious but asymptomatic ( I a ), infectious and symptomatic ( I s ), confirmed (C), and recovered (R).

We simulate the economic losses and health-care utilisation capacity (HCUR) for the National Capital Region (NCR), Ilocos Region, Western Visayas, Soccsksargen, and for the Davao Region by implementing various combinations of lockdown restrictions for three months to capture one quarter of economic losses for these regions. The National Capital Region accounts for about half of the Philippines’ gross domestic product, while the inclusion of other regions aim to represent the various areas of the country. The policy easing simulations use the four lockdown policies that the Philippines uses, as seen in Table 1 .

Simulations for the National Capital Region

Table 2 shows the sequence of lockdown measures implemented for the NCR. Each lockdown measure is assumed to be implemented for one month. Two sets of simulations are implemented for the region. The first set assumes a health systems capacity (HSC) for the region at 17.99% (A), while the second is at 21.93% (B). A higher HSC means an improvement in testing and isolation strategies for the regions of concern.

From the sequence of lockdown measures in Table 2 , Fig. 2 shows the plot of the average HCUR as well as the corresponding total economic losses for the two sets of simulations for one quarter. For the scenario at 17.99% HSC (A), the highest loss is recorded at 16.58% of the annual gross regional domestic product (GRDP) while the lowest loss is at 12.19% of its GRDP. Lower average HCUR corresponds to more stringent scenarios starting with Scenario 1. Furthermore, under the scenarios with 21.93% HSC (B), losses and average HCUR are generally lower. Scenarios 1 to 4 from this set lie below the 70% threshold of the HCUR, with the lowest economic loss simulated to be at 9.11% of the GRDP.

These include the set of trade-off decisions under a health system capacity equal to 17.99%, and another set equal to 21.93% (Source of basic data: Authors’ calculations).

Overall, the trend below shows a parabolic shape. The trend begins with an initial decrease in economic losses as restrictions loosen, but this comes at the expense of increasing HCUR. This is then followed by an increasing trend in losses as restrictions are further loosened. Notably, the subsequent marginal increases in losses in the simulation with 21.93% HSC are smaller relative to the marginal increases under the 17.99% HSC.

Simulations for the Regions Outside of NCR

Table 2 also shows the lockdown sequence for the Ilocos, Western Visayas, Soccsksargen, and Davao regions. The sequence begins with Level III only. Meanwhile, the lowest lockdown measure simulated for the regions is Level I. Two sets of simulations with differing health system capacities for each scenario are done as well.

With this lockdown sequence, Fig. 3 shows the panel of scatter plot between the average HCUR and total economic losses as percentage of the respective GRDP, with both parameters covering one quarter. Similar to the case of the NCR, the average HCUR for the simulations with higher health system capacity (B) is lower than the simulations with lower health system capacity (A). However, unlike in NCR, the regions’ simulations do not exhibit a parabolic shape.

These include trade-offs for a Ilocos Region, b Western Visayas Region, c Soccsksargen Region, and d Davao Region (Source of basic data: Authors’ calculations).

Discussion and interpretation

The hypothetical simulations above clearly capture the losses associated with the pandemic and the corresponding lockdown interventions by the Philippine government. The trend of the simulations clearly shows the differences in the policy considerations for the National Capital Region (NCR) and the four other regions outside of NCR. Specifically, the parabolic trend of the former suggests an optimal strategy that can be attained through a trade-off policy even with the absence of any constraint in finding the said optimal strategy. This trend is borne from the countervailing effects between the economic losses due to COVID-19 infection ( Y E ) and the losses from the lockdown measures ( Y L ) implemented for the region. Specifically, Fig. 4(a), (b) show the composition of economic losses across all scenarios for the NCR simulation under a lower and higher health system capacity (HSC), respectively.

These include losses under a HSC = 17.99% and b HSC = 21.93% in the National Capital Region (Source of basic data: Authors’ calculations).

In both panels of Fig. 4 , as quarantine measures loosen, economic losses from infections ( Y E ) tend to increase while the converse holds for economic losses due to quarantine restrictions ( Y L ). The results are intuitive as loosening restrictions may lead to increased mobility, and therefore increased exposure and infections from the virus. In fact, economic losses from infections ( Y E ) take up about half of the economic losses for the region in Scenario 7A, Fig. 4(a) .

While the same trends can be observed for the scenarios with higher HSC at 21.93%, the economic losses from infections ( Y E ) do not overtake the losses simulated from lockdown restrictions ( Y L ) as seen in Fig. 4(b) . This may explain the slower upward trend of economic losses in Fig. 2 at HSC = 21.93%.

The output of the simulation for the Davao region shows that the economic losses from COVID-19 infections ( Y E ) remain low even as the lockdown restrictions ease down. At the same time, economic losses from lockdown restrictions ( Y L ) show a steady decline with less stringent lockdown measures. Overall, the region experiences a decreasing trend in total economic losses even as the least stringent lockdown measure is implemented for a longer period. This pattern is similar with the regions of Ilocos, Western Visayas, and Soccsksarkgen.

The results of the simulations from Figs. 2 and 3 also demonstrate differing levels of economic losses and health-care utilisation between the two sets of scenarios for NCR and the four other regions. Clearly, lower economic losses and health-care utilisation rates were recorded for the scenarios with higher HSC. Specifically, lower total economic losses can be attributed to a slower marginal increase in losses from infections ( Y E ) as seen in Fig. 4(b) . Thus, even while easing restrictions, economic losses may be tempered with an improvement in the health system.

With the above analysis, the policy trade-off as a constrained minimisation problem of economic losses subject to HCUR above appears to apply in NCR but not in regions outside of NCR. The latter is better off in enhancing prevention, detection, isolation, treatment, and reintegration (PDITR) strategy combined with targeted small area lockdowns, if necessary, without risking any increases in economic losses. But, in all scenarios and anywhere, the enhancement of the HSC through improved PDITR strategies remains vital to avoid having to deal with local infection surges and outbreaks. This also avoids forcing local authorities in a policy bind between health and economic measures to implement. Enhancing PDITR in congested urban centres (i.e., NCR) is difficult especially with the surge in new daily cases. People are forced to defy social distance rules and other minimum health standards in public transportation and in their workplaces that help spread the virus.

We extended the FASSSTER subnational SEIR model to include two differential equations that capture economic losses due to COVID-19 infection and due to the lockdown measures, respectively. The extended model aims to account for the incurred economic losses following the rise and fall of the number of active COVID-19 cases in the country and the implementation of various lockdown measures. In simulating eight different scenarios in each of the five selected regions in the country, we found a tight policy choice in the case of the National Capital Region (NCR) but not in the cases of four other regions far from NCR. This clearly demonstrates the difficult policy decision in the case of NCR in minimising economic losses given the constraint of its intensive care unit (ICU) bed capacity.

On the other hand, the regions far from the NCR have wider policy space towards economic reopening and recovery. However, in all scenarios, the primary significance of improving the health system capacity (HSC) to detect and control the spread of the disease remains in order to widen the trade-off policy space between public health and economic measures.

The policy trade-off simulation results imply different policy approaches in each region. This is also to consider the archipelagic nature of the country and the simultaneous concentration of economic output and COVID-19 cases in NCR and contiguous provinces compared to the rest of the country. Each local region in the country merits exploration of different policy combinations in economic and health measures depending on the number of active COVID-19 cases, strategic importance of economic activities and output specific in the area, the geographic spread of the local population and their places of work, and considering local health system capacities. However, we would like to caution that the actual number of cases could diverge from the results of our simulations. This is because the parameters of the model must be updated regularly driven generally by the behaviour of the population and the likely presence of variants of COVID-19. Given the constant variability of COVID-19 data, we recommend a shorter period of model projections from one to two months at the most.

In summary, this paper showed how mathematical modelling can be used to inform policymakers on the economic impact of lockdown policies and make decisions among the available policy options, taking into consideration the economic and health trade-offs of these policies. The proposed methodology provides a tool for enhanced policy decisions in other countries during the COVID-19 pandemic or similar circumstances in the future.

Data availability

The raw datasets used in this study are publicly available at the Department of Health COVID-19 Tracker Website: https://doh.gov.ph/covid19tracker . Datasets will be made available upon request after completing request form and signing non-disclosure agreement. Code and scripts will be made available upon request after completing request form and signing non-disclosure agreement.

Acemoglu D, Chernozhukov V, Werning I, Whinston M (2020) Optimal targeted lockdowns of a multi-group SIR model. In: National Bureau of Economic Research Working Papers. National Bureau of Economic Research (NBER). https://www.nber.org/system/files/working_papers/w27102/w27102.pdf . Accessed 16 Jun 2021

Amewu S, Asante S, Pauw K, Thurlow J (2020) The economic costs of COVID-19 in Sub-Saharan Africa: insights from a simulation exercise for Ghana. Eur J Dev Res 32(5):1353–1378. https://doi.org/10.1057/s41287-020-00332-6

Article   PubMed   PubMed Central   Google Scholar  

Anand N, Sabarinath A, Geetha S, Somanath S (2020) Predicting the spread of COVID-19 using SIR model augmented to incorporate quarantine and testing. Trans Indian Natl Acad Eng 5:141–148. https://doi.org/10.1007/s41403-020-00151-5

Article   Google Scholar  

Bangko Sentral ng Pilipinas (2021) 2021 Inflation Report First Quarter. https://www.bsp.gov.ph/Lists/Inflation%20Report/Attachments/22/IR1qtr_2021.pdf . Accessed 16 Jun 2021

Benson C, Clay E (2004) Understanding the economic and financial impacts of natural disasters. In: World Bank Disaster Risk Management Paper. World Bank. https://elibrary.worldbank.org/doi/abs/10.1596/0-8213-5685-2 . Accessed Jan 2022

Byrd R, Lu P, Nocedal J, Zhu C (1995) A limited memory algorithm for bound constrained optimization. SIAM J Sci Comput 16:1190–1208. https://doi.org/10.1137/0916069

Article   MathSciNet   MATH   Google Scholar  

Cereda F, Rubião R, Sousa L (2020) COVID-19, Labor market shocks, and poverty in brazil: a microsimulation analysis. In: poverty and equity global practice. World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/34372/COVID-19-Labor-Market-Shocks-and-Poverty-in-Brazil-A-Microsimulation-Analysis.pdf?sequence=1&isAllowed=y . Accessed 21 Feb 2021

Chen D, Lee S, Sang J (2020) The role of state-wide stay-at-home policies on confirmed COVID-19 cases in the United States: a deterministic SIR model. Health Informatics Int J 9(2/3):1–20. https://doi.org/10.5121/hiij.2020.9301

Article   CAS   Google Scholar  

Department of Health-Epidemiology Bureau (2020) COVID-19 tracker Philippines. https://doh.gov.ph/covid19tracker . Accessed 12 Feb 2021

Department of Trade and Industry (2020a) Revised category I-IV business establishments or activities pursuant to the revised omnibus guidelines on community quarantine dated 22 May 2020 Amending for the purpose of memorandum circular 20-22s. https://dtiwebfiles.s3-ap-southeast-1.amazonaws.com/COVID19Resources/COVID-19+Advisories/090620_MC2033.pdf . Accessed 09 Feb 2021

Department of Trade and Industry (2020b) Increasing the allowable operational capacity of certain business establishments of activities under categories II and III under general community quarantine. https://dtiwebfiles.s3-ap-southeast-1.amazonaws.com/COVID19Resources/COVID-19+Advisories/031020_MC2052.pdf . Accessed 09 Feb 2021

Department of Trade and Industry (2021) Prescribing the recategorization of certain business activities from category IV to category III. https://www.dti.gov.ph/sdm_downloads/memorandum-circular-no-21-08-s-2021/ . Accessed 15 Mar 2021

Estadilla C, Uyheng J, de Lara-Tuprio E, Teng T, Macalalag J, Estuar M (2021) Impact of vaccine supplies and delays on optimal control of the COVID-19 pandemic: mapping interventions for the Philippines. Infect Dis Poverty 10(107). https://doi.org/10.1186/s40249-021-00886-5

FASSSTER (2020) COVID-19 Philippines LGU Monitoring Platform. https://fassster.ehealth.ph/covid19/ . Accessed Dec 2020

Geard N, Giesecke J, Madden J, McBryde E, Moss R, Tran N (2020) Modelling the economic impacts of epidemics in developing countries under alternative intervention strategies. In: Madden J, Shibusawa H, Higano Y (eds) Environmental economics and computable general equilibrium analysis. Springer Nature Singapore Pte Ltd., Singapore, pp. 193–214

Chapter   Google Scholar  

Genoni M, Khan A, Krishnan N, Palaniswamy N, Raza W (2020) Losing livelihoods: the labor market impacts of COVID-19 in Bangladesh. In: Poverty and equity global practice. World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/34449/Losing-Livelihoods-The-Labor-Market-Impacts-of-COVID-19-in-Bangladesh.pdf?sequence=1&isAllowed=y . Accessed 21 Feb 2021

Goldsztejn U, Schwartzman D, Nehorai A (2020) Public policy and economic dynamics of COVID-19 spread: a mathematical modeling study. PLoS ONE 15(12):e0244174. https://doi.org/10.1371/journal.pone.0244174

Article   CAS   PubMed   PubMed Central   Google Scholar  

Gousseff M, Penot P, Gallay L, Batisse D, Benech N, Bouiller K, Collarino R, Conrad A, Slama D, Joseph C, Lemaignen A, Lescure F, Levy B, Mahevas M, Pozzetto B, Vignier N, Wyplosz B, Salmon D, Goehringer F, Botelho-Nevers E (2020) Clinical recurrences of COVID-19 symptoms after recovery: viral relapse, reinfection or inflammatory rebound? J Infect 81(5):816–846. https://doi.org/10.1016/j.jinf.2020.06.073

Hou C, Chen J, Zhou Y, Hua L, Yuan J, He S, Guo Y, Zhang S, Jia Q, Zhang J, Xu G, Jia E (2020) The effectiveness of quarantine in Wuhan city against the Corona Virus Disease 2019 (COVID-19): A well-mixed SEIR model analysis. J Med Virol 92(7):841–848. https://doi.org/10.1002/jmv.25827

Article   CAS   PubMed   Google Scholar  

Hupkau C, Isphording I, Machin S, Ruiz-Valenzuela J (2020) Labour market shocks during the Covid-19 pandemic: inequalities and child outcomes. In: Covid-19 analysis series. Center for Economic Performance. https://cep.lse.ac.uk/pubs/download/cepcovid-19-015.pdf . Accessed 21 Feb 2021

Inter-Agency Task Force for the Management of Emerging Infectious Diseases (2020) Omnibus Guidelines on the Implementation of Community Quarantine in the Philippines with Amendments as of June 3, 2020. https://www.officialgazette.gov.ph/downloads/2020/06jun/20200603-omnibus-guidelines-on-the-implementation-of-community-quarantine-in-the-philippines.pdf . Accessed 09 Feb 2021

Kashyap S, Gombar S, Yadlowsky S, Callahan A, Fries J, Pinsky B, Shah N (2020) Measure what matters: counts of hospitalized patients are a better metric for health system capacity planning for a reopening. J Am Med Inform Assoc 27(7):1026–1131. https://doi.org/10.1093/jamia/ocaa076

Keogh-Brown M, Jensen H, Edmunds J, Smith R (2020) The impact of Covid-19, associated behaviours and policies on the UK economy: a computable general equilibrium model. SSM Popul Health 12:100651. https://doi.org/10.1016/j.ssmph.2020.100651

Keogh-Brown M, Smith R, Edmunds J, Beutels P (2010) The macroeconomic impact of pandemic influenza: estimates from models of the United Kingdom, France, Belgium and The Netherlands. Eur J Health Econ 11:543–554. https://doi.org/10.1007/S10198-009-0210-1

Article   PubMed   Google Scholar  

Laborde D, Martin W, Vos R (2021) Impacts of COVID-19 on global poverty, food security, and diets: insights from global model scenario analysis. Agri Econ (United Kingdom) 52(3):375–390. https://doi.org/10.1111/agec.12624

Maital S, Barzani E (2020) The global economic impact of COVID-19: a summary of research. https://www.neaman.org.il/en/Files/Global%20Economic%20Impact%20of%20COVID19.pdf . Accessed Jan 2022

McKibbin W, Fernando R (2020) The global macroeconomic impacts of COVID-19: seven scenarios. https://www.brookings.edu/research/the-global-macroeconomic-impacts-of-covid-19-seven-scenarios/ . Accessed Jan 2022

National Economic and Development Authority (2020) Impact of COVID-19 on the economy and the people, and the need to manage risk. https://www.sec.gov.ph/wp-content/uploads/2020/12/2020CG-Forum_Keynote_NEDA-Sec.-Chua_Impact-of-COVID19-on-the-Economy.pdf . Accessed Feb 2021

National Economic Development Authority-Investment Coordination Committee (2016) Revisions on ICC Guidelines and Procedures (Updated Social Discount Rate for the Philippines). http://www.neda.gov.ph/wp-content/uploads/2017/01/Revisions-on-ICC-Guidelines-and-Procedures-Updated-Social-Discount-Rate-for-the-Philippines.pdf . Accessed 12 Feb 2021

Noy I, Doan N, Ferrarini B, Park D (2020) The economic risk of COVID-19 in developing countries: where is it highest? In: Djankov S, Panizza U (eds) COVID-19 in developing economies. Center for Economic Policy Research Press, London, pp. 38–52

Google Scholar  

Peng T, Liu X, Ni H, Cui Z, Du L (2020) City lockdown and nationwide intensive community screening are effective in controlling the COVID-19 epidemic: analysis based on a modified SIR model. PLoS ONE 15(8):e0238411. https://doi.org/10.1371/journal.pone.0238411

Pham T, Dwyer L, Su J, Ngo T (2021) COVID-19 impacts of inbound tourism on australian economy. Ann. Tourism Res. 88:103179. https://doi.org/10.1016/j.annals.2021.103179

Philippine Statistics Authority (2019a) 2018 Labor Force Survey (Microdata). https://psa.gov.ph/content/2018-annual-estimates-tables . Accessed 12 Feb 2021

Philippine Statistics Authority (2019b) Updated population projections based on the results of 2015 POPCEN. https://psa.gov.ph/content/updated-population-projections-based-results-2015-popcen . Accessed 12 Feb 2021

Philippine Statistics Authority (2020) Census of population and housing. https://psa.gov.ph/population-and-housing . Accessed May 2020

Philippine Statistics Authority (2021) National Accounts Data Series. https://psa.gov.ph/national-accounts/base-2018/data-series . Accessed 05 Apr 2021

Reno C, Lenzi J, Navarra A, Barelli E, Gori D, Lanza A, Valentini R, Tang B, Fantini MP (2020) Forecasting COVID-19-associated hospitalizations under different levels of social distancing in Lombardy and Emilia-Romagna, Northern Italy: results from an extended SEIR compartmental model. J Clin Med 9(5):1492. https://doi.org/10.3390/jcm9051492

Article   CAS   PubMed Central   Google Scholar  

United Nations Development Programme (2021) The Socioeconomic Impact Assessment of COVID-19 in the Bangsamoro Autonomous Region in Muslim Mindanao. UNDP in the Philippines. https://www.ph.undp.org/content/philippines/en/home/library/the-socioeconomic-impact-assessment-of-covid-19-on-the-bangsamor.html . Accessed Nov 2021.

Verikios G, McCaw J, McVernon J, Harris A (2012) H1N1 influenza and the Australian macroeconomy. J Asia Pac Econ 17(1):22–51. https://doi.org/10.1080/13547860.2012.639999

World Bank (2020) Projected Poverty Impacts of COVID-19 (Coronavirus). https://www.worldbank.org/en/topic/poverty/brief/projected-poverty-impacts-of-COVID-19 . Accessed Jan 2022

World Bank (2021) Global economic perspectives. World Bank, Washington DC

World Bank (2022) GDP growth (annual %). https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG . Accessed Jan 2022

Download references

Acknowledgements

We thank Dr. Geoffrey M. Ducanes, Associate Professor, Ateneo de Manila University Department of Economics, for giving us valuable comments in the course of developing the economic model, and Mr. Jerome Patrick D. Cruz, current Ph.D. student, Massachusetts Institute of Technology, for initiating and leading the economic team in FASSSTER in the beginning of the project for their pitches in improving the model. We also thank Mr. John Carlo Pangyarihan for typesetting the manuscript. The project is supported by the Philippine Council for Health Research and Development, United Nations Development Programme and the Epidemiology Bureau of the Department of Health.

Author information

Authors and affiliations.

Department of Mathematics, Ateneo de Manila University, Quezon City, Philippines

Elvira P. de Lara-Tuprio & Timothy Robin Y. Teng

Department of Information Systems and Computer Science, Ateneo de Manila University, Quezon City, Philippines

Maria Regina Justina E. Estuar & Lenard Paulo V. Tamayo

Department of Economics, Ateneo de Manila University, Quezon City, Philippines

Joselito T. Sescon, Cymon Kayle Lubangco, Rolly Czar Joseph T. Castillo & Gerome M. Vedeja

Department of Mathematics, Caraga State University, Butuan City, Philippines

Jay Michael R. Macalalag

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the study conception and design. Model conceptualization, data collection and analysis were performed by EPdL-T, MRJEE, JTS, CKL, CJTC, TRYT, LPT, JMRM, and GMV. All authors commented on previous versions of the manuscript, and read and approved the final manuscript.

Corresponding author

Correspondence to Cymon Kayle Lubangco .

Ethics declarations

Competing interests.

Proponents of the project are corresponding authors in the study. Research and corresponding publications are integral to the development of the models and systems. The authors declare no conflict nor competing interest.

Ethical approval

Not applicable

Informed consent

Additional information.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

de Lara-Tuprio, E.P., Estuar, M.R.J.E., Sescon, J.T. et al. Economic losses from COVID-19 cases in the Philippines: a dynamic model of health and economic policy trade-offs. Humanit Soc Sci Commun 9 , 111 (2022). https://doi.org/10.1057/s41599-022-01125-4

Download citation

Received : 06 August 2021

Accepted : 09 March 2022

Published : 04 April 2022

DOI : https://doi.org/10.1057/s41599-022-01125-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Content Search

Philippines

Philippines: a primary health care case study in the context of the COVID-19 pandemic

Attachments.

Executive summary

Through the implementation of the Universal Health Care (UHC) Act (1), the Philippines’ health system, especially its chief health agency the Department of Health (DOH), has sought to address a triple disease burden and the COVID-19 pandemic. The aim of this case study is to examine key aspects of primary health care (PHC) in the Philippines to inform future policy and practice, incorporating lessons learned during the COVID-19 pandemic between January 2020 and July 2022.

The devolution of the country’s health system places management and implementation of health care under local government units (LGUs). The DOH steers national PHC directives and programmes. Although devolution has allowed LGUs to innovate around models of care to better reach marginalized communities, the health system remains fragmented. This is exemplified by the limited referral and coordination channels among levels of governance and service delivery. The non-profit portion of the private sector helps close service delivery gaps for PHC through partnerships with nongovernmental organizations (NGOs), technical assistance from the academic community, and community-owned projects and patient groups, but these mechanisms often limit individual participation. This is separate from the for-profit portion of the private sector, which functions as a parallel health system not directly under the DOH’s management.

Exacerbated by the COVID-19 pandemic, challenges to the full implementation of UHC include a scarcity of health care workers, especially in rural areas, and variable health financing schemes resulting in increased out-of-pocket (OOP) expenditure. Efforts to strengthen PHC could address health workforce and financing gaps and seek to harness empowered local structures.

Despite the fragmentation of the health system and limited resources, PHC service delivery is enabled through strong local mechanisms, many of which were created during the COVID-19 pandemic. These mechanisms include ordinances for the implementation of national health programmes, increased buy-in from local leaders for PHC, multisectoral collaboration for health, continual grassroots feedback from patients, and innovations around monitoring and quality assurance of service delivery. PHC-oriented research could enable further innovation at national and local levels, including to support utilization of digital technologies. For example, there may be opportunities to scale PHC innovations such as remote consultations and diversified models of care.

Related Content

Philippines, koica, un agencies launch 2 mobile health clinics to reduce adolescent pregnancies in samar, southern leyte, save the children philippines wraps up transformative initiatives in cotabato, save the children philippines and advocates clamor for urgent passage of teen pregnancy prevention bill.

Philippines + 4 more

Asia and the Pacific: Weekly Regional Humanitarian Snapshot (23 - 29 April 2024)

• Case Report
• Open access
• Published: 14 April 2020

First COVID-19 infections in the Philippines: a case report

• Edna M. Edrada 1 ,
• Edmundo B. Lopez 1 ,
• Jose Benito Villarama 1 ,
• Eumelia P. Salva Villarama 1 ,
• Bren F. Dagoc 1 ,
• Chris Smith 2 , 3 ,
• Ana Ria Sayo 1 ,
• Jeffrey A. Verona 1 ,
• Jamie Trifalgar-Arches 1 ,
• Jezreel Lazaro 1 ,
• Ellen Grace M. Balinas 1 ,
• Elizabeth Freda O. Telan 1 ,
• Lynsil Roy 1 ,
• Myvie Galon 1 ,
• Carl Hill N. Florida 1 ,
• Tatsuya Ukawa 2 ,
• Annavi Marie G. Villanueva 2 ,
• Nobuo Saito 4 ,
• Jean Raphael Nepomuceno 2 ,
• Koya Ariyoshi 5 ,
• Celia Carlos 6 ,
• Amalea Dulcene Nicolasora 6 &
• Rontgene M. Solante 1  

Tropical Medicine and Health volume  48 , Article number:  21 ( 2020 ) Cite this article

93k Accesses

42 Citations

59 Altmetric

Metrics details

A Correction to this article was published on 07 May 2020

This article has been updated

The novel coronavirus (COVID-19) is responsible for more fatalities than the SARS coronavirus, despite being in the initial stage of a global pandemic. The first suspected case in the Philippines was investigated on January 22, 2020, and 633 suspected cases were reported as of March 1. We describe the clinical and epidemiological aspects of the first two confirmed COVID-19 cases in the Philippines, both admitted to the national infectious disease referral hospital in Manila.

Case presentation

Both patients were previously healthy Chinese nationals on vacation in the Philippines travelling as a couple during January 2020. Patient 1, a 39-year-old female, had symptoms of cough and sore throat and was admitted to San Lazaro Hospital in Manila on January 25. Physical examination was unremarkable. Influenza B , human coronavirus 229E, Staphylococcus aureus and Klebsiella pneumoniae were detected by PCR on initial nasopharyngeal/oropharyngeal (NPS/OPS) swabs. On January 30, SARS-CoV-2 viral RNA was reported to be detected by PCR on the initial swabs and she was identified as the first confirmed COVID-19 case in the Philippines. Her symptoms resolved, and she was discharged. Patient 2, a 44-year-old male, had symptoms of fever, cough, and chills. Influenza B and Streptococcus pneumoniae were detected by PCR on initial NPS/OPS swabs. He was treated for community-acquired pneumonia with intravenous antibiotics, but his condition deteriorated and he required intubation. On January 31, SARS-CoV-2 viral RNA was reported to be detected by PCR on the initial swabs, and he was identified as the 2nd confirmed COVID-19 infection in the Philippines. On February 1, the patient’s condition deteriorated, and following a cardiac arrest, it was not possible to revive him. He was thus confirmed as the first COVID-19 death outside of China.

Conclusions

This case report highlights several important clinical and public health issues. Despite both patients being young adults with no significant past medical history, they had very different clinical courses, illustrating how COVID-19 can present with a wide spectrum of disease. As of March 1, there have been three confirmed COVID-19 cases in the Philippines. Continued vigilance is required to identify new cases.

The novel coronavirus 2019 (COVID-19) is responsible for more fatalities than the severe acute respiratory syndrome (SARS) coronavirus, despite being in the initial stage of a global pandemic. It is thought that the index case occurred on December 8, 2019, in Wuhan, China [ 1 ]. Since then, cases have been exported to other Chinese cities, as well as internationally, highlighting concern of a global outbreak [ 2 ]. The first suspected case in the Philippines was investigated on January 22, 2020, and 633 suspected cases have been reported as of March 1. Of them, 183 were in the National Capital Region of Manila, of whom many were admitted to San Lazaro Hospital (SLH) in Manila, the national infectious disease referral hospital [ 3 , 4 ]. We describe the epidemiologic and clinical characteristics of the first two confirmed COVID-19 cases in the Philippines, including the first death outside China.

In this case report, we describe two cases: patient 1, the first confirmed COVID-19 case, and patient 2, the second confirmed case, even though the symptoms of patient 2 started first. The cases are presented based on reports from the clinicians involved in patient care and results of investigations available to them at the time. Figure 1 shows a timeline of symptoms for both patients according to the day of illness and day of hospitalisation.

Timeline of symptoms according to day of illness and day of hospitalisation

History prior to hospitalisation

Both patients were Chinese nationals on vacation in the Philippines travelling as a couple. They had no known comorbidities and reported no history of smoking. Patient 2, a 44-year-old male, reported fever on January 18, 2020, whilst the couple were residing in Wuhan, China. It was reported that he was in contact with someone that was unwell in Wuhan, but not that he had visited the seafood market. During January 20 to 25, they travelled from Wuhan via Hong Kong to several locations in the Philippines (Fig. 2 ). Patient 1, a 39-year-old female, developed cough and sore throat on January 21. Due to persistence of symptoms of patient 2, they travelled to Manila on January 25. In Manila, patient 2 was denied entry to a hotel because he was febrile and both patients were transferred to San Lazaro Hospital (SLH), the national referral hospital for infectious diseases [ 4 ]. On admission, patient 2 was classified as a COVID-19 person under investigation (PUI) based on his travel history and fever [ 2 ] and was transferred to a designated isolation area with negative pressure rooms. Patient 1 did not fit the PUI criteria due to absence of fever, but was also isolated because of possible exposure.

Travels of patient 1 and 2

Clinical course of patient 1

On admission to the ward on January 25 (illness day 5), patient 1 complained of a dry cough, but the sore throat had improved. She was awake and conversant with a blood pressure of 110/80, HR 84, RR 18 and temperature 36.8 °C. Her chest was clear. The remainder of the physical examination was unremarkable. Nasopharyngeal and oropharyngeal swab (NPS/ORS) specimens were collected and sent to the Research Institute for Tropical Medicine (RITM) in Muntinlupa City [ 5 ]. A chest radiograph was reported as unremarkable (Fig. 3 ).

Posteroanterior chest radiograph of patient 1, 27 January 2020 (illness day 7). Unremarkable

On January 27, the results were released of a commercially available respiratory pathogen multiplex real-time PCR for detection of pathogen genes on the NPS/OPS samples (FTD Respiratory pathogens 33, Fast Track Diagnostics) at the RITM Molecular Biology Laboratory. These assays reported detection of Influenza B viral RNA, human coronavirus 229E viral RNA, Staphylococcus aureus DNA and Klebsiella pneumoniae DNA. A 10-day course of oseltamivir 75 mg BID was given on the basis of the influenza result. The NPS/OPS specimen was then sent by RITM to the Victorian Infectious Disease Reference Laboratory (VIDRL) in Melbourne, Australia, for COVID-19 testing [ 6 ].

On January 29, further NPS/ORS specimens were collected and sent to the RITM. On January 30, the result of the initial NPS/OPS sent to VIDRL reported detection of 2019-nCoV (subsequently termed SARS-CoV-2) viral RNA by real-time PCR. The patient was thus identified by the Department of Health as the first confirmed COVID-19 case in the Philippines [ 6 ].

On illness days 6 to 10, she remained afebrile with minimal cough and clear breath sounds. During this time, real-time PCR for detecting SARS-CoV-2 was established at the RITM using the Corman et al. protocol [ 7 ]. Further NPS/OPS specimens collected on January 29 (reported on January 31) and January 31 (reported on February 2) also reported detection of SARS-CoV-2 viral RNA. On illness day 11, the patient reported resolution of symptoms. She remained afebrile and clinically stable apart from two episodes of loose watery stool on illness day 12. Further samples were collected on February 2 and 4. On February 8 (illness day 19), she was discharged when SARS-CoV-2 was no longer detected on an NPS/OPS sample.

Clinical course of patient 2

In contrast, patient 2 experienced a more severe clinical course. On admission (illness day 8), he reported fever, cough and chills. On examination, he was awake and conversant with a temperature of 38.3 °C, blood pressure of 110/80, HR 84, RR 18, and SpO 2 of 96% on room air. His chest was clear. The remainder of the physical examination was unremarkable.

A working diagnosis of community-acquired pneumonia and COVID-19 suspect was made. He was started on ceftriaxone 2 g intravenously (IV) once daily (OD) and azithromycin 500 mg OD. NPS/ORS specimens were collected and sent to the RITM. On January 27, the results of a respiratory pathogen real-time PCR detection panel performed at RITM on the NPS/OPS samples were released, reporting detection of Influenza B viral RNA and Streptococcus pneumoniae DNA. The NPS/OPS samples were sent to the VIDRL for additional testing. Oseltamivir 75 mg BID was commenced on the basis of the influenza result.

During illness days 9 and 10, his fever continued with occasional non-productive cough. He remained clinically stable apart from intermittent SpO 2 desaturations of 93–97% on 2–3 L/min of oxygen. On illness day 11, he developed increasing dyspnoea with reduced SpO 2 at 88% despite 8 L/min of oxygen via a face mask and haemoptysis and was noted to have bilateral chest crepitations. A chest radiograph was reported as showing hazy infiltrates in both lung fields consistent with pneumonia (Fig. 4 ). Meropenem 2 g IV three times a day (TDS) was commenced.

Posteroanterior chest radiograph of patient 2, 27 January 2020 (illness day 10). Hazy infiltrates in both lung fields consistent with pneumonia

On illness day 12, he became increasingly dyspnoeic, hypoxic and agitated and was intubated and sedated with a midazolam drip. An endotracheal aspirate (ETA) and a further NPS/OPS were collected and sent to the RITM. Vancomycin, 30 mg/kg loading dose followed by 25 mg/kg BD, was commenced with a working diagnosis of severe community-acquired pneumonia due to Streptococcus pneumoniae secondary to Influenza B infection, plus consideration of COVID-19 pending the ETA result. A complete blood count showed values within the normal range (Table 1 ). On illness day 13, he continued to be febrile (38.5–40.0 °C) with bibasal crackles. Vital signs were stable with adequate urine output. A chest radiograph was reported as showing worsening of pneumonia (Fig. 5 ).

Posteroanterior chest radiograph of patient 2, 30 January 2020 (illness day 13). Endotracheal tube in situ approximately 2 cm above the carina. There is worsening of the previously noted pneumonia

On illness day 14, increased crepitations in both lung fields were noted. Blood cultures showed no growth after 24 h of incubation. An HIV test was non-reactive. On this day, the RITM reported detection of SARS-CoV-2 viral RNA by real-time PCR from the NPS/OPS taken on illness day 12 and hence the 2nd confirmed COVID-19 infection in the Philippines. This result was later confirmed on February 4 on the initial admission sample sent to VIDRL.

On the morning of illness day 15, the patient remained febrile at 40 °C, with BP 110/70, HR 95, RR 30, SpO 2 99% with 80% FiO2, and adequate urine output. However, the patient’s condition deteriorated with the formation of thick sputum and blood clots in the ET tube. Despite frequent suctioning, the patient’s condition deteriorated. He was noted to have laboured breathing followed by a cardiac arrest. Despite several rounds of cardiopulmonary resuscitation, it was not possible to revive the patient. He was thus confirmed as the first COVID-19 death outside of China.

Discussion and conclusion

This case report describes the first two confirmed cases of COVID-10 in the Philippines and highlights several important clinical and public health issues. Despite both patients being young adults with no significant past medical history, they had very different clinical courses, illustrating how COVID-19 can present with a wide spectrum of disease [ 8 ]. Whilst patient 1 had a mild uncomplicated illness consistent with an upper respiratory tract infection and recovery, patient 2 developed a severe pneumonia and died.

One possible explanation for the differing clinical courses is the presence of co-infection. In both patients, the real-time PCR detection panel was reported to be positive for multiple pathogens. The Staphylococcus aureus and Klebsiella pneumoniae detected in patient 1 most likely represent bacterial colonisation, and it is unclear to what extent her presentation was due to influenza or COVID-19 or both. Patient 2 tested positive for COVID-19, Influenza B , and Streptococcus pneumoniae , all of which can cause respiratory infection and severe pneumonia. Unfortunately, sputum culture was not possible due to biosafety concerns. It is unclear which pathogen was the leading cause of death, but previous research has shown that outcomes of acute viral respiratory infection are worse if multiple pathogens are present [ 9 ]. This highlights the importance of testing for other respiratory pathogens in addition to COVID-19 in order to optimise antimicrobial therapy.

Patient 2 developed increasing dyspnoea on day 11 of illness, similar to the first COVID-19 case in the USA, where mild symptoms were initially reported with progression to pneumonia on day 9 of illness [ 10 ]. The median time from illness onset to dyspnoea in a case series in Wuhan was 8 days (range 5–13) [ 11 ]. The explanation for patient 2’s worsening condition and development of haemoptysis was progression of pneumonia rather than acute respiratory distress syndrome or pulmonary embolism, but it was not possible to perform a CT scan, additional laboratory tests or an autopsy to further assess this. Although he was treated with broad-spectrum antimicrobials, it is not clear if the outcome would have been better in a high-resource setting. Both patients were treated with oseltamivir in view of positive results for Influenza B . Further studies are required to establish the optimal treatment and role of antiviral medication for patients with suspected or confirmed COVID-19 infection.

Our cases contrast with the US case in terms of the relative paucity of lab data and time to receive results. Limited in-house testing was undertaken due to biosafety concerns. In the case of patient 2, the diagnosis of COVID-19 was not made until a day before the patient died. This was because SARS-2-CoV testing was being established in the Philippines at the time that the patients were admitted, and initial samples had to be sent to Australia. Although the delay of diagnosis is unlikely to have altered management, expansion of COVID-19 diagnostics including multiplex panels for other respiratory pathogens is urgently needed for prompt diagnosis of patients for screening of hospital personnel or other contacts.

Three SLH hospital staff who were caring for the patients developed symptoms and themselves became PUIs, but were later discharged following negative SARS-CoV-2 testing and symptom resolution. This highlights the risk of an outbreak in the hospital, or a ‘super-spreader’ scenario, as was observed in other settings during the early stages of the SARS coronavirus infections in 2003 [ 12 ]. In the case of SARS, as with COVID-19, SLH managed two cases and was able to contain the infection without further spread [ 13 ].

The third confirmed COVID-19 case was announced on February 3 from a sample taken on January 23, also a Chinese national who had travelled from Wuhan. She recovered and returned to China on January 31. Contact tracing has been undertaken of all three patients [ 14 ]. Despite travel to several locations in the Philippines whilst experiencing symptoms, as of March 1, there has not been any confirmed local transmission arising from these cases and the number of PUIs has decreased [ 3 ]. However, as infection can be mild or subclinical, local transmission cannot be excluded. Increasing the number of laboratories able to perform SARS-CoV-2 testing would allow better surveillance and improve detection of COVID-19 cases.

In conclusion, as of March 1, there have been three confirmed COVID-19 cases in the Philippines including the first death outside of China. No local transmission has been confirmed. Continued vigilance is required to identify new cases.

Availability of data and materials

Change history, 07 may 2020.

An amendment to this paper has been published and can be accessed via the original article.

Abbreviations

Coronavirus disease 2019

Novel coronavirus

Nasopharyngeal swab/oropharyngeal swab

Polymerase chain reaction

Person under observation

Research Institute for Tropical Medicine

Severe acute respiratory syndrome

San Lazaro Hospital

Zhu N, Zhang D, Wang W, Li X, Yang B, Song J, et al. A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med. 2020; NEJMoa2001017. Available from: http://www.nejm.org/doi/10.1056/NEJMoa2001017 .

Wu JT, Leung K, Leung GM. Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study. Lancet. 2020;6736:20.

Google Scholar  

Republic of the Philippines Department of Health. 2019-NCOV CASE TRACKER. [cited 2020 Feb 6]. Available from: https://www.doh.gov.ph/node/19197 .

Republic of the Philippines Department of Health. San Lazaro Hospital. 2020 [cited 2020 Feb 6]. Available from: http://slh.doh.gov.ph/ .

Republic of the Philippines Department of Health. Research Institute for Tropical Medicine. [cited 2020 Feb 7]. Available from: http://ritm.gov.ph/ .

Peter Doherty Institute for Infection and Immunity. Victorian Infectious Disease Reference Laboratory (VIDRL) [Internet]. [cited 2020 Feb 7]. Available from: https://www.vidrl.org.au/ .

Corman VM, Landt O, Kaiser M, Molenkamp R, Meijer A, Chu DK, et al. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Euro Surveill. 2020;25(3):1–8.

Article   Google Scholar  

Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020;6736(20):1–10 Available from: https://doi.org/10.1016/S0140-6736(20)30183-5 .

Yoshida L, Suzuki M, Nguyen HA, Le MN, Vu TD, Yoshino H, et al. Respiratory syncytial virus: co-infection and paediatric lower respiratory tract infections. Eur Respir J. 2013;42:461–9.

Rothe C, Schunk M, Sothmann P, Bretzel G, Froeschl G, Wallrauch C, et al. Transmission of 2019-nCoV infection from an asymptomatic contact in Germany. N Engl J Med. 2020:2019–20.

Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet (London, England). 2020;6736(20):1–7 Available from: http://www.ncbi.nlm.nih.gov/pubmed/32007143 .

Munster V, Koopmans M, van Doremalen N, van Riel D, de Wit E. A novel coronavirus emerging in China — key questions for impact assessment. NEJM. 2020:4–6.

Lopez J. Severe Acute Respiratory Syndrome (SARS) control and surveillance :The Philippine experience. In: 4th Health Rsearch For Action National Forum. Manila; 2003.

Republic of the Philippines Department of Health. DOH CONFIRMS 3RD 2019-NCOV ARD CASE IN PH. 2020 [cited 2020 Feb 5]. Available from: https://www.doh.gov.ph/doh-press-release/doh-confirms-3rd-2019-nCoV-ARD-case-in-PH .

Download references

Acknowledgements

We are very grateful to the patients for allowing us to prepare and publish this case report.

Not applicable

Author information

Authors and affiliations.

San Lazaro Hospital, Manila, Philippines

Edna M. Edrada, Edmundo B. Lopez, Jose Benito Villarama, Eumelia P. Salva Villarama, Bren F. Dagoc, Ana Ria Sayo, Jeffrey A. Verona, Jamie Trifalgar-Arches, Jezreel Lazaro, Ellen Grace M. Balinas, Elizabeth Freda O. Telan, Lynsil Roy, Myvie Galon, Carl Hill N. Florida & Rontgene M. Solante

School of Tropical Medicine and Global Health, Nagasaki University, Nagasaki, Japan

Chris Smith, Tatsuya Ukawa, Annavi Marie G. Villanueva & Jean Raphael Nepomuceno

Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, UK

Chris Smith

Department of Microbiology, Faculty of Medicine, Oita University, Oita, Japan

Nobuo Saito

Institute of Tropical Medicine, Nagasaki University, Nagasaki, Japan

Koya Ariyoshi

Research Institute for Tropical Medicine, Alabang, Philippines

Celia Carlos & Amalea Dulcene Nicolasora

You can also search for this author in PubMed   Google Scholar

Contributions

All of the authors contributed to the writing of this case report. The authors read and approved the final manuscript.

Corresponding author

Correspondence to Chris Smith .

Ethics declarations

Ethics approval and consent to participate.

This case report was not part of a research study, and hence, ethical approval was not sought. Written consent was obtained from patient 1 and on behalf of patient 2.

Consent for publication

Written consent for the preparation and publication of a case report was provided by patient 1 and on behalf of patient 2, following discussion with his brother.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The original version of this article was revised: One of the authors’ name is revised.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Edrada, E.M., Lopez, E.B., Villarama, J.B. et al. First COVID-19 infections in the Philippines: a case report. Trop Med Health 48 , 21 (2020). https://doi.org/10.1186/s41182-020-00203-0

Download citation

Received : 02 March 2020

Accepted : 22 March 2020

Published : 14 April 2020

DOI : https://doi.org/10.1186/s41182-020-00203-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Case report
• Coronavirus
• Philippines

Tropical Medicine and Health

ISSN: 1349-4147

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Western Pac Surveill Response J
• v.12(1); Jan-Mar 2021

Early response to COVID-19 in the Philippines

Arianna maever l. amit.

a College of Medicine, University of the Philippines Manila, Manila, Philippines.

b School of Medicine and Public Health, Ateneo de Manila University, Pasig City, Philippines.

Veincent Christian F. Pepito

Manuel m. dayrit.

Low- and middle-income countries (LMICs) with weak health systems are especially vulnerable during the COVID-19 pandemic. In this paper, we describe the challenges and early response of the Philippine Government, focusing on travel restrictions, community interventions, risk communication and testing, from 30 January 2020 when the first case was reported, to 21 March 2020. Our narrative provides a better understanding of the specific limitations of the Philippines and other LMICs, which could serve as basis for future action to improve national strategies for current and future public health outbreaks and emergencies.

The Philippine health system and the threat of public health emergencies

Despite improvements during the past decade, the Philippines continues to face challenges in responding to public health emergencies because of poorly distributed resources and capacity. The Philippines has 10 hospital beds and six physicians per 10 000 people. ( 1 , 2 ) and only about 2335 critical care beds nationwide. ( 3 ) The available resources are concentrated in urban areas, and rural areas have only one physician for populations up to 20 000 people and only one bed for a population of 1000. ( 4 ) Disease surveillance capacity is also unevenly distributed among regions and provinces. The primary care system comprises health centres and community health workers, but these are generally ill-equipped and poorly resourced, with limited surge capacity, as evidenced by lack of laboratory testing capacity, limited equipment and medical supplies, and lack of personal protective equipment for health workers in both primary care units and hospitals. ( 5 ) Local government disaster preparedness plans are designed for natural disasters and not for epidemics.

Inadequate, poorly distributed resources and capacity nationally and subnationally have made it difficult to respond adequately to public health emergencies in the past, as in the case of typhoon Haiyan in 2013. ( 6 ) The typhoon affected 13.3 million people, overwhelming the Government’s capacity to mobilize human and financial resources rapidly to affected areas. ( 7 ) Failure to deliver basic needs and health services resulted in disease outbreaks, including a community outbreak of gastroenteritis. ( 8 ) Access to care has improved in recent years due to an increase in the number of private hospital beds; ( 5 ) however, improvements in private sector facilities mainly benefit people who can afford them, in both urban and rural areas.

In this paper, we describe the challenges and early response of the Philippine Government, focusing on travel restrictions, community interventions, risk communication and testing, from 30 January 2020 when the first case was reported, to 21 March 2020.

Early response to COVID-19

Travel restrictions.

Travel restrictions in the Philippines were imposed as early as 28 January, before the first confirmed case was reported on 30 January ( Fig. 1a ). ( 9 ) After the first few COVID-19 cases and deaths, the Government conducted contact tracing and imposed additional travel restrictions, ( 10 ) with arrivals from restricted countries subject to 14-day quarantine and testing. While travel restrictions in the early phase of the COVID-19 response prevented spread of the disease by potentially infected people, travellers from countries not on the list of restricted countries were not subject to the same screening and quarantine protocols. The restrictions were successful in delaying the spread of the disease only briefly, as the number of confirmed cases increased in the weeks that followed. ( 11 ) Fig. 1b shows all interventions, including travel restrictions undertaken before 6 March, when the Government declared the occurrence of community spread, and after 11 March, when WHO declared COVID-19 a pandemic.

New cases of COVID-19 in the Philippines, 30 January–21 March 2020

[ insert Figure 1a ]

Timeline of key events and developments in the Philippines, 30 January–21 March 2020

[ insert Figure 1b ]

Community interventions

The Government declared “enhanced community quarantine” (ECQ) for Metro Manila between 15 March and 14 April ( Fig. 2a ), which was subsequently extended to the whole island of Luzon ( Fig. 2b ). The quarantine consisted of: strict home quarantine in all households, physical distancing, suspension of classes and introduction of work from home, closure of public transport and non-essential business establishments, prohibition of mass gatherings and non-essential public events, regulation of the provision of food and essential health services, curfews and bans on sale of liquor and a heightened presence of uniformed personnel to enforce the quarantine procedures. ( 12 ) ECQ – an unprecedented move in the country’s history – was modelled on the lockdown in Hubei, China, which was reported to have slowed disease transmission. ( 13 ) Region-wide disease control interventions, such as quarantining of the entire Luzon island, were challenging to implement because of their scale and social and economic impacts, but they were deemed necessary to “flatten the curve” so that health systems were not overwhelmed. ( 14 ) While the lockdown implemented by the Government applied only to the island of Luzon, local governments in other parts of the country followed this example and also locked down. The ECQ gave the country the opportunity to mobilize resources and organize its pandemic response, which was especially important in a country with poorly distributed, scarce resources and capacity.

Provinces placed under enhanced community quarantine (ECQ). (2a) The Government declared ECQ in Metro Manila effective 15 March 2020; (2b) The Government declared ECQ on the entire island of Luzon effective 17 March 2020.

[ insert Figure 2 ]

Risk communication

The Government strengthened and implemented national risk communication plans to provide information on the new disease. The Government conducted daily press briefings, sponsored health-related television and Internet advertisements and circulated infographics on social media. Misinformation and conspiracy theories about COVID-19 were nevertheless a challenge for a population that spends more than 10 hours a day on the Internet. ( 15 , 16 ) These spread quickly and became increasingly difficult to correct. Furthermore, the Government’s messages did not reach all households, despite access to health services and information, resulting in limited knowledge of preventive practices, except for hand-washing. ( 17 )

Testing is key to controlling the pandemic but was done on a small scale in the Philippines. As of 19 March, fewer than 1200 individuals had been tested, ( 11 ) as only the Research Institute for Tropical Medicine located in Metro Manila performed tests and assisted subnational reference laboratories in testing. ( 18 ) No positivity rates for RT–PCR tests were reported until early April 2020. Because of the limited capacity for testing at the start of the pandemic, the Department of Health imposed strict protocols to ration testing resources while ramping up testing capacity. Most tests were conducted for individuals in urban areas, where the incidence was highest. ( 19 )

Conclusions

At the start of the COVID-19 pandemic, the country’s initial response lacked organizational preparedness to counter the public health threat. The Philippines’ disease surveillance system could conduct contact tracing, but this was overwhelmed in the early phases of outbreak response. Similarly, in February, only one laboratory could conduct reverse transcriptase polymerase chain reaction (RT–PCR) testing, so the country could not rapidly deploy extensive laboratory testing for infected cases. In addition, the primary care system of the Philippines did not serve as a primary line of defence, as people went straight to hospitals in urban areas, overwhelming critical care capacity in the early stages of the COVID-19 pandemic.

In response to the early phase of the pandemic, the Government of the Philippines implemented travel restrictions, community quarantine, risk communication and testing; however, the slow ramping up of capacities particularly on testing contributed to unbridled disease transmission. By 15 October, the number of confirmed cases had exponentially grown to 340 000 of which 13.8% were deemed active. ( 11 ) The lack of pandemic preparedness had left the country poorly defended against the new virus and its devastating effects. Investing diligently and consistently in pandemic preparedness, surveillance and testing capacity in particular is a lesson that the Philippines and other LMICs should learn from COVID-19.

Acknowledgements

Conflict of interests.

None reported

Not applicable

• Open access
• Published: 14 May 2024

Social media influence on COVID-19 vaccine perceptions among University students: a Malawi case study

• Mervis Folotiya 1 &
• Chimwemwe Ngoma   ORCID: orcid.org/0000-0001-8648-1244 1 , 2  

BMC Public Health volume  24 , Article number:  1312 ( 2024 ) Cite this article

27 Accesses

Metrics details

Introduction

The global fight against the COVID-19 pandemic relies significantly on vaccination. The collective international effort has been massive, but the pace of vaccination finds hindrance due to supply and vaccine hesitancy factors. Understanding public perceptions, especially through the lens of social media, is important. This study investigates the influence of social media on COVID-19 vaccine perceptions among university students in Malawi.

The study utilized a quantitative methodology and employed a cross-sectional study design to explore the relationship between social media dynamics and COVID-19 vaccine perceptions among 382 randomly sampled students at MUBAS. Data, collected by use of a Likert-scale questionnaire, was analyzed using IBM SPSS 20 for descriptive statistics and Pearson correlation tests.

The findings reveal crucial correlations. Specifically, trust in online vaccine information shows a positive correlation ( r  = 0.296, p  < 0.01) with active engagement in social media discussions. Conversely, a negative correlation surfaces concerning individuals’ reactions to vaccine availability in Malawi ( r = -0.026, p  > 0.05). The demographic overview highlights the prevalence of the 16 to 30 age group, representing 92.9% of respondents.

Conclusions

The identified correlations emphasize the need for careful communication strategies tailored to combat misinformation and enhance vaccine acceptance among the younger demographic in Malawi. The positive correlation between trust in online vaccine information and social media engagement underscores digital platforms’ potential for disseminating accurate information. Conversely, the negative correlation with vaccine availability reactions suggest the presence of complex factors shaping public perceptions.

Peer Review reports

Vaccination has become an important weapon in the global fight against the Coronavirus Disease 2019 (COVID-19), helping to slow down the virus’s spread [ 1 ]. Global efforts to implement COVID-19 vaccination campaigns have been unprecedented in scale and scope [ 2 ], with governments, pharmaceutical firms, and organizations collaborating to develop, manufacture, and distribute vaccines at a faster pace [ 3 ]. However, the difficulties have been complex, involving issues like vaccine hesitancy and unequal vaccine supply and distribution [ 4 ].

The pandemic has had an impact on Malawi’s healthcare delivery and public health interventions, primarily due to the country’s limited healthcare infrastructure, socio-economic disparities, and cultural beliefs that influence perceptions of healthcare practices [ 5 , 6 ]. This context has also contributed to vaccine hesitancy, resulting in a lower uptake of COVID-19 vaccines in the country. As of May 2023, the uptake rate stands at 40%, which is lower than the country’s 60% target [ 7 , 8 ].

A recent scoping review on COVID-19 vaccination hesitancy among Malawians reveals that COVID-19 vaccine reluctance is primarily the result of misinformation, with vaccines perceived as harmful or dangerous. Myths such as infertility, severe disability, or even death have contributed to vaccine hesitancy [ 9 ]. The review also reveals that some people refuse vaccinations because of their religious convictions and beliefs [ 9 ]. The challenges posed by vaccine hesitancy in Malawi highlight the need for targeted communication strategies and public health initiatives that consider the country’s unique socio-cultural context, aimed at achieving widespread vaccination coverage and understanding public opinions about COVID-19 vaccinations [ 8 , 10 , 11 ]. These strategies must address factors such as misinformation, lack of trust in healthcare institutions, fear of side effects, and cultural beliefs surrounding vaccination, which contribute to reluctance to accept COVID-19 vaccines.

Vaccine perception plays an important role in determining the success of vaccination efforts, and these perceptions are shaped by exposure to (mis)information amplified by the media, the community, and the health system. Notably, social networks may either positively or negatively impact vaccination uptake, depending on their views on vaccines [ 12 ]. Given these challenges, the success of vaccination campaigns relies not only on the development and distribution of vaccines but also on how these interventions are perceived by the public. Public attitudes and beliefs surrounding vaccine safety, efficacy, and necessity significantly impact vaccine uptake [ 11 ].

Social media has emerged a powerful tool in the current information-dissemination landscape for influencing public opinion. Its role in health communication has expanded significantly, providing a dynamic platform for sharing information, influencing attitudes, and shaping behavior [ 13 ]. During the COVID-19 pandemic, social media platforms played an important role in amplifying public health messages [ 14 ]. However, social media’s very nature, which is marked by a rapid flow of information and a variety of sources, also poses numerous challenges. The spread of false and scientifically inaccurate information are some of the issues that health communication must deal with in the digital age [ 15 ].

Within the unique university settings, the dynamics of vaccine perceptions take on a distinctive dimension. The convergence of diverse backgrounds, cultures, and perspectives among university students creates a rich tapestry of attitudes towards health-related issues. Understanding the specific distinctions within this demographic is crucial for tailoring effective public health interventions. Factors such as lack of access, affordability, health disparities, educational background, peer pressure, political views, and lack of trust in institutions may be influenced in unique ways [ 16 ].

Despite the growing body of literature exploring the influence of social media on vaccine perceptions, a research gap exists concerning its specific impact on COVID-19 vaccine perceptions among university students in Malawi. While other studies have examined aspects of social media influence on COVID-19 perceptions [ 17 , 18 , 19 , 20 ], such as the role of trust in online vaccine information, and engagement in social media discussions, there remains a need for more research within this specific demographic. Therefore, this study seeks to contribute in bridging this gap by providing valuable insights that not only enhance the academic understanding of the subject but also provide practical implications for public health communication strategies tailored to the Malawian university setting.

The study employed a quantitative methodology, and utilized a cross-sectional study design to investigate the relationship between social media dynamics and COVID-19 vaccine perceptions among university students at Malawi University of Business and Applied Sciences (MUBAS), which had a student population of 7,619 during the 2022/2023 academic year.

To ensure unbiased participant selection and equitable representation, a simple random sampling technique was employed. Using the population of students at MUBAS, a sampling frame was created in Excel, including student IDs as the unique identifier number. Subsequently, a random number generator was integrated in Excel to randomly select participants in the study to ensure that each participant had an equal chance of being selected. Using the Taro Yamane method, a sample size of 380 respondents was determined from the total student population of 7619, at precision level of 5%. However, for practical considerations, the sample size was adjusted upwards to 388. Potential participants were approached through an invitation process, and were informed about the purpose of the study. A total of 382 complete questionnaires were collected.

Data collection

A self-administered questionnaire, organized into three sections (demographics, access to COVID-19 vaccine information on social media, and awareness of COVID-19 vaccine information), was used for data collection. Respondents’ attitudes and perspectives were recorded using the Likert scale. The data collection process involved a comprehensive exploration of variables, covering aspects such as social media usage, access to COVID-19 vaccine information, engagement levels, trust, and demographic details. The survey yielded a response rate of 99.5%, indicating robust participant engagement and contributing to the reliability of the gathered data. Ethical considerations, including confidentiality, consent, and voluntary participation, were observed to safeguard the privacy of the participants. Additionally, the study was evaluated and approved by the MUBAS Postgraduate Research Evaluation Committee, study number MMS/20/PG/004.

Data analysis

Rigorous editing procedures were applied to ensure data completeness and consistency. IBM SPSS 20 software was used for data coding, cleaning, and analysis. Descriptive statistics and Pearson correlation tests were also performed to derive insights from the dataset. In-depth analyses involved the interpretation of demographic data, an assessment of social media dynamics, and exploration of relationships through Pearson correlation. Results were interpreted based on statistical significance, effect size, and alignment with existing research, contributing to a better understanding of the interaction between social media, COVID-19 vaccine misinformation, and hesitancy among university students.

The study established that all participants in the study utilized social media and internet for various purposes. A noteworthy positive correlation ( r  = 0.296, p  < 0.01) emerged, indicating a strong association between trust in online vaccine information and active engagement in social media discussions. Conversely, a significant negative correlation ( r = -0.610, p  < 0.01) was identified, shedding light on the relationship between individuals’ reactions to vaccine availability in Malawi and their trust in online information.

Demographic overview

This section provides an overview of the respondents’ demographics using descriptive statistics. Four key characteristic namely age, education year, gender, and nationality, were analyzed to unveil insights into the composition of the study participants.

The study participants exclusively comprised Malawian students, aligning with the university’s predominantly undergraduate student population, which constitutes a significant percentage of the study sample. International students registered at the university are mostly postgraduate students and represented by less than 5% of the study population. Respondents covered various age groups, with the majority falling within the 21 to 30 years range ( n  = 230). This age bracket represents 60.2% of the total respondents. 26.7% of the respondents were in their fourth year, 25.4% in the first year, and 21.7% and 21.2% in the second and third years, respectively. Postgraduate students represented the smallest group at 5%. Regarding gender, 50.3% of respondents were female, while 49.7% were male.

Correlation matrix of key variables

Correlation is deemed significant at the 0.01 level (two-tailed).

N  = 382 for all correlations.

Table  1 presents an analysis of key variables related to COVID-19 vaccine perceptions among university students. The Pearson correlation coefficient shows the relationship between trust in online COVID-19 vaccine information, COVID-19 vaccine effectiveness, participation in COVID-19 discussions on social media, and response to the COVID-19 vaccine.

A noteworthy positive correlation ( r  = 0.296, p  < 0.01) between trust in online vaccine information and participation in social media discussions is observed, indicating that people who trust online vaccine information are more likely to participate in digital health discussions. On the other hand, a significant correlation ( r = -0.610, p  < 0.01) between trust in internet-based vaccine information and individuals’ attitudes toward vaccination, highlighting the influence of trust on vaccination perceptions. Additionally, a modest negative correlation ( r = -0.087, p  > 0.05) between the COVID-19 vaccine and participating in Covid-19 discussions on social media is noted. Although not statistically significant, this relationship suggests a possible trend by which individuals protected from COVID-19 may exhibit lower levels of social media engagement.

The demographic overview in Table  2 reveals that a majority of respondents ( n  = 230), fall within the 21 to 30 age range. Notably, the 16 to 30 age range constitutes an accumulative 92.9% of the total respondents. This aligns with predominant internet usage patterns globally [ 21 ], and emphasizes the importance of understanding the perspectives of this demographic in shaping COVID-19 vaccine perceptions. Customizing information dissemination to this group’s preferences and habits becomes important, since they make up a significant portion of social media and internet users.

The gender distribution in Table  2 , shows a near-balanced representation, with 50.3% female and 49.7% male respondents. Previous research on the connection between gender and social media use reveal broader trends indicating the popularity of social media among females than males [ 22 , 23 , 24 ]. The study benefits from the diverse perspectives contributed by both genders in social media discussions, which is highlighted by this balance.

The study examined the relationships among the variables influencing the perceptions of COVID-19 vaccines. The analysis in Table  1 reveals a positive correlation between trust in online vaccine information and active engagement in COVID-19 discussions on social media ( r  = 0.296, p  < 0.01). This aligns with existing literature, which suggests that people who place trust in online health information are more likely to actively participate in digital health dialogues [ 25 ]. Furthermore, the relationship between trust and engagement extends beyond mere participation, it influences the dissemination of accurate information and the formation of informed opinions within online communities [ 26 ].

One effective way to promote trust in online vaccine information is through transparent communication practices, including providing information from reputable sources, ensuring data accuracy and reliability, engaging with credible health experts, and conducting educational campaigns on health literacy and critical thinking [ 27 ].

Moreover, the positive correlation emphasizes the potential of digital platforms, particularly social media, in promoting health literacy and influencing public health behaviors [ 28 ]. As individuals trust the information they encounter online, they are more likely to share it with their social networks, leading to broader awareness and understanding of vaccination-related issues [ 26 ]. This phenomenon has implications for public health communication strategies, indicating that efforts to build trust in online vaccine information can have cascading effects on community engagement and knowledge dissemination, ultimately contributing to improved vaccination rates and public health outcomes [ 29 ].

The negative correlation ( r = -0.026, p  > 0.05) found between people’s reactions to vaccine availability in Malawi and their trust in online vaccine information is an interesting finding. On the surface, one might expect that higher trust in online vaccine information would correspond to more positive reactions toward vaccine availability. However, this result is consistent with studies that highlight the complexity of public attitudes toward vaccination, often influenced by contextual and socio-cultural factors [ 30 , 31 ]. This unexpected finding prompts an exploration into the factors influencing public sentiment in the context of vaccine availability. Factors such as misinformation, fear of side effects, and cultural beliefs surrounding vaccination play a crucial role in shaping vaccine perceptions among Malawians, contributing to vaccine hesitancy and lower uptake rates compared to the set targets [ 5 , 6 ].

Additionally, a recent study on the impact of social media news on COVID-19 vaccine hesitancy and vaccination behavior suggests that individuals are more sensitive to vaccine risk news than safety news on social media, indicating a relationship between the type of information and its impact on perception [ 17 ]. This resonates with the findings of this study, highlighting the complex nature of public sentiment, shaped by the interaction of trust, engagement, and the specific content of vaccine-related information on social media.

Moreover, although the negative correlation between COVID-19 vaccination and participation in COVID-19 discussions on social media ( r = -0.087, p  > 0.05) was not statistically significant, it indicates a possible a potential trend worth exploring further. This finding suggests that people who have received the COVID-19 vaccine may show slightly lower levels of participation in COVID-19 discussions on social media. This observation raises questions about the factors influencing online engagement among vaccinated individuals within this population group. One of the possible factors contributing to this trend could be that vaccinated individuals may feel a reduced sense of urgency or concern about COVID-19 compared to unvaccinated individuals, leading to less active participation in discussions about the virus on social media.

This study underscores the interaction between trust in online vaccine information, social media engagement, and public perception regarding COVID-19 vaccination. The positive correlation identified between trust and active participation in social media discussions highlights the role of reliable online sources in shaping public discourse. The negative correlation between trust and individuals’ reactions to vaccine availability prompts a deeper exploration into the factors influencing public perception. By acknowledging and addressing these factors, policymakers and healthcare providers can enhance vaccine acceptance and uptake rates.

Limitations and future directions

While the study provides valuable insights, certain limitations warrant acknowledgment. The reliance on self-reported data and the cross-sectional design inherent in the methodology limit the extent to which causal inferences can be drawn [ 32 ]. Additionally, the exclusive focus on students from MUBAS may not fully capture the broader spectrum of the population. To address these constraints, future research could employ longitudinal designs and incorporate diverse demographic groups for a more comprehensive understanding of the dynamics shaping COVID-19 vaccine perceptions.

The findings of this study underscore the importance of promoting trust in online vaccine information and leveraging digital platforms, particularly social media, to enhance health literacy and influence public health behaviors. Addressing vaccine hesitancy requires tailored communication strategies that are responsive to widespread concerns. By actively promoting trust in the veracity of online vaccine information and recognizing the influence of contextual and socio-cultural factors on public sentiment, public health campaigns can effectively utilize social media platforms to promote positive attitudes and perceptions regarding COVID-19 vaccination.

Data availability

The data obtained from the project is accessible and can be provided by the first author upon reasonable request.

Abbreviations

19–Coronavirus Disease 2019

Malawi University of Business and Applied Sciences

Tang B, Zhang X, Li Q, Bragazzi NL, Golemi-Kotra D, Wu J. The minimal COVID-19 vaccination coverage and efficacy to compensate for a potential increase of transmission contacts, and increased transmission probability of the emerging strains. BMC Public Health [Internet]. 2022;22(1). https://doi.org/10.1186/s12889-022-13429-w .

Machado BAS, Hodel KVS, Fonseca LMDS, Pires VC, Mascarenhas LAB, Da Silva Andrade LPC et al. The importance of vaccination in the context of the COVID-19 pandemic: A brief update regarding the use of vaccines. Vaccines [Internet]. 2022;10(4):591. https://doi.org/10.3390/vaccines10040591 .

Druedahl LC, Minssen T, Price WN. Collaboration in times of crisis: A study on COVID-19 vaccine R&D partnerships. Vaccine [Internet]. 2021;39(42):6291–5. https://doi.org/10.1016/j.vaccine.2021.08.101 .

Blasioli E, Mansouri B, Tamvada SS, Hassini E. Vaccine Allocation and Distribution: A Review with a Focus on Quantitative Methodologies and Application to Equity, Hesitancy, and COVID-19 Pandemic. Operations Research Forum [Internet]. 2023;4(2). https://doi.org/10.1007/s43069-023-00194-8 .

Phiri M, MacPherson E, Panulo M, Chidziwisano K, Kalua K, Chirambo CM et al. Preparedness for and impact of COVID-19 on primary health care delivery in urban and rural Malawi: a mixed methods study. BMJ Open [Internet]. 2022;12(6):e051125. https://doi.org/10.1136/bmjopen-2021-051125 .

Chawinga WD, Singini W, Phuka J, Chimbatata N, Mitambo C, Sambani C et al. Combating coronavirus disease (COVID-19) in rural areas of Malawi: Factors affecting the fight. African Journal of Primary Health Care & Family Medicine [Internet]. 2023;15(1). https://doi.org/10.4102/phcfm.v15i1.3464 .

TRADING ECONOMICS. Malawi Coronavirus COVID-19 vaccination rate [Internet]. TRADING ECONOMICS. 2024 [cited 2024 Mar 23]. https://tradingeconomics.com/malawi/coronavirus-vaccination-rate .

Bwanali AN, Lubanga A, Mphepo M, Munthali L, Chumbi GD, Kangoma M. Vaccine hesitancy in Malawi: a threat to already-made health gains. Annals of Medicine and Surgery [Internet]. 2023;85(10):5291–3. https://doi.org/10.1097/ms9.0000000000001198 .

Nkambule E, Mbakaya BC. COVID-19 vaccination hesitancy among Malawians: a scoping review. Systematic Reviews [Internet]. 2024;13(1). https://doi.org/10.1186/s13643-024-02499-z .

West R, Hurst NB, Sharma S, Henry B, Vitale-Rogers S, Mutahi W et al. Communication strategies to promote vaccination behaviours in sub-Saharan Africa. BMC Global and Public Health [Internet]. 2023;1(1). https://doi.org/10.1186/s44263-023-00004-7 .

Paul E, Steptoe A, Fancourt D. Attitudes towards vaccines and intention to vaccinate against COVID-19: Implications for public health communications. The Lancet Regional Health - Europe [Internet]. 2021;1:100012. https://doi.org/10.1016/j.lanepe.2020.100012 .

Loreche AM, Pepito VCF, Sumpaico-Tanchanco LB, Dayrit MM. COVID-19 vaccine brand hesitancy and other challenges to vaccination in the Philippines. PLOS Global Public Health [Internet]. 2022;2(1):e0000165. https://doi.org/10.1371/journal.pgph.0000165 .

Benetoli A, Chen T, Aslani P. How patients’ use of social media impacts their interactions with healthcare professionals. Patient Education and Counseling [Internet]. 2018;101(3):439–44. https://doi.org/10.1016/j.pec.2017.08.015 .

Obi-Ani NA, Anikwenze C, Isiani MC. Social media and the Covid-19 pandemic: Observations from Nigeria. Cogent Arts & Humanities [Internet]. 2020;7(1):1799483. https://doi.org/10.1080/23311983.2020.1799483 .

Kadam A, Atre S. Negative impact of social media panic during the COVID-19 outbreak in India. Journal of Travel Medicine [Internet]. 2020;27(3). https://doi.org/10.1093/jtm/taaa057 .

Gilbert-Esparza E, Brady A, Haas S, Wittstruck H, Miller J, Kang Q et al. Vaccine hesitancy in college students. Vaccines [Internet]. 2023;11(7):1243. https://doi.org/10.3390/vaccines11071243 .

Zhang Q, Zhang R, Wu WC, Liu Y, Yu Z. Impact of social media news on COVID-19 vaccine hesitancy and vaccination behavior. Telematics and Informatics [Internet]. 2023;80:101983. https://doi.org/10.1016/j.tele.2023.101983 .

Cascini F, Pantović A, Al-Ajlouni YA, Failla G, Puleo V, Melnyk A et al. Social media and attitudes towards a COVID-19 vaccination: A systematic review of the literature. EClinicalMedicine [Internet]. 2022;48:101454. https://doi.org/10.1016/j.eclinm.2022.101454 .

Gudi SK, George SM, José J. Influence of social media on the public perspectives of the safety of COVID-19 vaccines. Expert Review of Vaccines [Internet]. 2022;21(12):1697–9. https://doi.org/10.1080/14760584.2022.2061951 .

Wilson SL, Wiysonge CS. Social media and vaccine hesitancy. BMJ Global Health [Internet]. 2020;5(10):e004206. https://doi.org/10.1136/bmjgh-2020-004206 .

Statista. Age distribution of internet users worldwide 2021 [Internet]. Statista. 2023. https://www.statista.com/statistics/272365/age-distribution-of-internet-users-worldwide/ .

Booker C, Kelly Y, Sacker A. Gender differences in the associations between age trends of social media interaction and well-being among 10–15 year olds in the UK. BMC Public Health [Internet]. 2018;18(1). https://doi.org/10.1186/s12889-018-5220-4 .

Karatsoli M, Nathanail E. Examining gender differences of social media use for activity planning and travel choices. European Transport Research Review [Internet]. 2020;12(1). https://doi.org/10.1186/s12544-020-00436-4 .

Chidiac M, Ross C, Marston HR, Freeman S. Age and Gender Perspectives on Social Media and Technology Practices during the COVID-19 Pandemic. International Journal of Environmental Research and Public Health [Internet]. 2022;19(21):13969. https://doi.org/10.3390/ijerph192113969 .

Impact of internet use on health-related behaviors and the patient-physician relationship: a survey-based study and review [Internet]. PubMed. 2008. https://pubmed.ncbi.nlm.nih.gov/19075034 .

Westney ZV, Hur I, Wang L, Sun J. Examining the effects of disinformation and trust on social media users’ COVID-19 vaccine decision-making. Information Technology & People [Internet]. 2023; https://doi.org/10.1108/itp-05-2022-0410 .

Fan J, Wang X, Du S, Mao A, Du H, Qiu W. Discussion of the Trust in Vaccination against COVID-19. Vaccines [Internet]. 2022;10(8):1214. https://doi.org/10.3390/vaccines10081214 .

Al-Dmour H, Masa’deh R, Salman A, Abuhashesh M, Al-Dmour R. Influence of social media platforms on public health protection against the COVID-19 pandemic via the mediating effects of public health awareness and behavioral changes: Integrated model. Journal of Medical Internet Research [Internet]. 2020;22(8):e19996. https://doi.org/10.2196/19996 .

De Freitas L, Basdeo D, Wang H. Public trust, information sources and vaccine willingness related to the COVID-19 pandemic in Trinidad and Tobago: an online cross-sectional survey. The Lancet Regional Health - Americas [Internet]. 2021;3:100051. https://doi.org/10.1016/j.lana.2021.100051 .

AlShurman BA, Khan AF, Mac C, Majeed M, Butt ZA. What demographic, social, and contextual factors influence the intention to use COVID-19 vaccines: a scoping review. International Journal of Environmental Research and Public Health [Internet]. 2021;18(17):9342. https://doi.org/10.3390/ijerph18179342 .

Larson HJ, Jarrett C, Eckersberger E, Smith D, Paterson P. Understanding vaccine hesitancy around vaccines and vaccination from a global perspective: A systematic review of published literature, 2007–2012. Vaccine [Internet]. 2014;32(19):2150–9. https://doi.org/10.1016/j.vaccine.2014.01.081 .

Levy JT, Maroney J, Kashem MA. Introduction to clinical research. In: Elsevier eBooks [Internet]. 2023. pp. 105–10. https://doi.org/10.1016/b978-0-323-90300-4.00040-9 .

Download references

Acknowledgements

We extend our heartfelt gratitude to Dr. Jolly Ntaba, the Supervisor of this study, for his invaluable guidance, and insightful feedback during the research project. We also appreciate the peer review and feedback provided by Mr. Andrew Kaponya and Mr. Ronald Udedi from the Department of Journalism and Media Studies, which greatly contributed to the improvement of this study.

The authors did not receive any funding to undertake the research and develop this paper.

Author information

Authors and affiliations.

Malawi University of Business and Applied Sciences, Blantyre, Malawi

Mervis Folotiya & Chimwemwe Ngoma

Department of Research and Innovation, ThinkSmart Consulting, Lilongwe, Malawi

Chimwemwe Ngoma

You can also search for this author in PubMed   Google Scholar

Contributions

The original concept for this study was conceived by MF, who also designed the study and undertook data collection. Data analysis was conducted by MF and CN. CN took the lead in composing the initial draft of this paper. Subsequently, MF, and CN engaged in a critical revision process to enhance its intellectual depth. All authors participated in reviewing, reading, and endorsing the final version of the paper.

Corresponding author

Correspondence to Chimwemwe Ngoma .

Ethics declarations

Ethics approval and consent to participate.

The study was evaluated and proved by the Malawi University of Business and Applied Sciences (MUBAS) Postgraduate Research Evaluation Committee, study number MMS/20/PG/004. In addition, informed consent was obtained from all participants prior to their involvement in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Folotiya, M., Ngoma, C. Social media influence on COVID-19 vaccine perceptions among University students: a Malawi case study. BMC Public Health 24 , 1312 (2024). https://doi.org/10.1186/s12889-024-18764-8

Download citation

Received : 07 December 2023

Accepted : 02 May 2024

Published : 14 May 2024

DOI : https://doi.org/10.1186/s12889-024-18764-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Vaccination
• Social media
• University students
• Vaccine perceptions
• Communication strategies

BMC Public Health

ISSN: 1471-2458

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Help | Advanced Search

Quantitative Biology > Populations and Evolution

Important: e-prints posted on arXiv are not peer-reviewed by arXiv; they should not be relied upon without context to guide clinical practice or health-related behavior and should not be reported in news media as established information without consulting multiple experts in the field.

Title: Optimal epidemic control by social distancing and vaccination of an infection structured by time since infection: the covid-19 case study

Abstract: Motivated by the issue of COVID-19 mitigation, in this work we tackle the general problem of optimally controlling an epidemic outbreak of a communicable disease structured by time since exposure, by the aid of two types of control instruments namely, social distancing and vaccination by a vaccine at least partly effective in protecting from infection. Effective vaccines are assumed to be made available only in a subsequent period of the epidemic so that - in the first period - epidemic control only relies on social distancing, as it happened for the COVID-19 pandemic. By our analyses, we could prove the existence of (at least) one optimal control pair, we derived first-order necessary conditions for optimality, and proved some useful properties of such optimal solutions. A worked example provides a number of further insights on the relationships between key control and epidemic parameters.

Submission history

Access paper:.

• HTML (experimental)
• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Back to the future? Airline sector poised for change post-COVID-19

It’s difficult to overstate just how much the COVID-19 pandemic has devastated airlines. In 2020, industry revenues totaled $328 billion, around 40 percent of the previous year’s. In nominal terms, that’s the same as in 2000. The sector is expected to be smaller for years to come; we project traffic won’t return to 2019 levels before 2024.

Financial woes aside, the pandemic’s longer-term effects on aviation are emerging. Some of these are obvious: hygiene and safety standards will be more stringent, and digitalization will continue to transform the travel experience. Mobile apps will be used to store travelers’ vaccine certificates and COVID-19 test results.

Other effects, though, are more profound. Unlike the 2008 global financial crisis, which was purely economic and weakened spending power, COVID-19 has changed consumer behavior—and the airline sector—irrevocably.

This article will explore five fundamental shifts in the aviation industry that have arisen from the pandemic. For each of these shifts, we also issue a call to action. By responding to these shifts decisively now, carriers should be able to look beyond the pandemic and adapt to the long-term realities of COVID-19.

1. Leisure trips will fuel the recovery

Business travel will take longer to recover, and even then, we estimate it will only likely recover to around 80 percent of prepandemic levels by 2024. Remote work  and other flexible working arrangements are likely to remain in some form postpandemic and people will take fewer corporate trips.

In previous crises, leisure trips or visits to friends and relatives tended to rebound first, as was the case in the United Kingdom following 9/11 and the global financial crisis (Exhibit 1). Not only did business trips take four years to return to precrisis levels after the attacks on the World Trade Center but they also had not yet recovered to pre-financial-crisis levels when COVID-19 broke out in 2020. Therefore, we expect that as the pandemic subsides, the rise in leisure trips will outpace the recovery of business travel.

Some carriers are highly dependent on business travelers—both those traveling in business class and those who book economy-class seats right before they need to travel. While leisure passengers fill up most of the seats on flights and help cover a portion of fixed costs, their overall financial contributions in net marginal terms are negligible, if not negative. Most of the profits earned on a long-haul flight are generated by a small group of high-yielding passengers, often traveling for business. But this pool of profit-generating passengers has shrunk because of the pandemic.

A McKinsey Live event on 'Returning to corporate travel: How do we get it right?'

The call: Revisit flight economics

Airlines should reevaluate the economics of their operations, especially long-haul flights. First, a smaller contribution from business traffic could necessitate a different pricing logic. For example, today most carriers price point-to-point nonstop flights at a premium. Travelers who value time over price—mostly business travelers—book these nonstop flights. Leisure travelers, even those traveling in premium classes, are more price sensitive and may choose an indirect routing. This large gap between nonstop pricing and connect pricing may need to narrow.

Second, lower business traffic may require network changes. Airlines added many flights over the past few years between hubs and smaller cities, using small-size widebodies such as the Boeing 787. These flights work because of the high-yielding business demand. With business demand subdued, economics favor larger aircraft flying less frequently. Airlines may find that larger aircraft such as Airbus A350s or Boeing 777s—which have lower unit costs—become the base of the long-haul network.

Third, airlines may also look at reconfiguring the layout of their cabins to address the increased share of leisure traffic. At the simplest level, lower business-class demand may warrant smaller business-class cabins. Taking this further, products may shift to better cater to premium-leisure passengers, such as growth of premium-economy cabins or development of business-class seats more suitable for traveling as couples or groups.

2. Staggering debt levels will lead to ticket price increases and a larger role for government in the sector

Many airlines have had to borrow huge sums of money to stay afloat and cope with high daily cash burn rates. Tapping into state-provided aid, credit lines, and bond issuances, the industry collectively amassed more than $180 billion worth of debt in 2020, 1 “COVID-19 lowers airline credit ratings and raises the cost of debt,” International Air Transport Association, August 21, 2020, iata.org. a figure equivalent to more than half of total annual revenues that year. And debt levels are still rising (Exhibit 2). Repaying these loans is made even harder by worsening credit ratings and higher financing costs.

These costs will need to be recouped. Therefore, we’ll likely see ticket prices rise. By our estimates, this could amount to a rise in ticket prices of about 3 percent, assuming a ten-year repayment window for only the additional debt taken on.

Furthermore, when demand for air travel returns, it will likely outpace supply initially. We see a glut of latent demand of people eager to travel. It will take time for airlines to restore capacity, and bottlenecks such as delays in bringing aircraft back to service and crew retraining could lead to a supply–demand gap, resulting in higher short-term prices.

In many cases, airline rescue efforts come in the form of government bailouts—with strings attached. We’re seeing a reemergence of, or increase in, the level of state ownership and influence. In Europe alone, TAP Air Portugal, Lufthansa Group, and Air Baltic all received state aid combined with an increase or reintroduction of government shareholdings.

The call: Be a constructive collaborator

As the state becomes a more active player—whether as a creditor, a direct shareholder, or as part of the board—airlines will find themselves having to deal more closely with the authorities. Instead of seeing this as a necessary restriction to access much-needed funds, airlines can treat it as an opportunity to shape how the sector evolves with a key stakeholder.

Airlines can work with regulators to set standards across a gamut of issues. These could include committing to reductions in greenhouse-gas emissions in return for more labor flexibility; increasing the cash-on-hand requirements to make airlines more resilient against future shocks; more balanced value sharing between airlines and other sectors such as airports; or changes in the ownership caps to allow greater inflows of foreign capital, reducing the reliance on state capital further down the road.

3. We will see a greater disparity of performance among airlines in the future

Some airlines have responded to the pandemic by restructuring for greater efficiency; others are merely muddling through. Occasionally, this is linked to state-aid programs, which may reduce the incentive for much-needed measures such as cost, organizational, and operational restructuring. Airlines that are not proactively transforming risk failing to set the business up for longer-term structural value creation.

As such, we’re seeing some airlines pull ahead. Before COVID-19, an airline boasted an ROIC well ahead of the overall industry’s rate of 5.8 percent. Not only did its stronger position pre-COVID-19 enable it to navigate the crisis thus far without taking on government loans of the scale relative to other airlines, it also made it possible for it to restructure to emerge with an even more competitive cost base.

Another group of carriers that have an opportunity to transform their business are airlines that have access to a restructuring process, such as Chapter 11 in the United States. These carriers can renegotiate midlife leases, shed excess debt, and emerge leaner. They will be fierce competitors going forward.

The call: Aim higher when it comes to IT and digital investment

Becoming better can necessitate investment. Even though many airlines find themselves in financial straits, we recommend investing more in IT and digitalization, not less. Before the pandemic, airlines spent roughly 5 percent of their revenue on IT. This is relatively low compared with other sectors. By means of comparison, the retail industry spends around 6 percent on average, and financial services 10 percent.

Airlines could consider stepping up IT and automation investment now. For example, airlines can respond to the quicker recovery of domestic and short-haul flights by investing in direct sales and owning the customer relationship. Relationships with IT and distribution providers could be reexplored. Carriers can also invest in the customer experience—such as making check-in and boarding processes more seamless—and support services—from revenue accounting to invoicing—to drive the next level of efficiency. Beyond this, the next horizon is analytics, which involves, among other efforts, using data  in smarter ways to enhance decision making, requiring some investment but yielding significant payoffs .

4. Aircraft markets may be oversupplied for some time to come

In the years before COVID-19, aircraft OEMs ramped up production in the anticipation of continued growth. This has led to a glut in aircraft availability. Furthermore, some carriers have returned relatively new aircraft to lessors, such as Norwegian Air Shuttle when it exited the long-haul market. Prices for used-aircraft leases have plummeted and are likely to remain lower. For instance, the monthly lease rate of a 2016 vintage Boeing 777-300ER aircraft was around $1.2 million in 2019. In 2020, the rate fell to less than $800,000. New aircraft are rumored to be available at even deeper discounts.

The call: Act countercyclically now, if you can

If finances permit, carriers can consider acting countercyclically: locking in orders for new aircraft or confirming operating leases now when demand is low. Aircraft are a significant expense for an airline, making up 10 to 15 percent of a carrier’s cost base. As lease rates and OEM pricing fluctuate with supply and demand levels, inking deals during a crisis could allow carriers to enjoy a cost advantage for years to come.

5. Air freight will see undersupply for some time

Over the past ten years, low cargo rates and the unprofitability of the cargo business have led many airlines to relinquish or scale back their dedicated cargo freighter fleets. However, cargo has been a lifeline for the aviation industry during COVID-19. Before the pandemic, cargo typically made up around 12 percent of the sector’s total revenue; that percentage tripled last year. Based on data from the Airline Analyst, only 21 (down from 77 in 2019) of the airlines around the world that disclosed their operating performance achieved positive operating profits for the third quarter of 2020, traditionally the industry’s most profitable quarter. Among these 21 airlines, cargo revenue accounted for 49 percent of total revenues on average.

During the pandemic, e-commerce sales soared while many passenger flights—which are responsible for delivering around half of total air cargo—were grounded. As a result, cargo yields increased by about 30 percent last year. As commercial flights gradually return, belly supply will increase, although not to pre-COVID-19 levels for at least a few years, as the industry is expected stay smaller than before the pandemic for several years.

The call: Bring back freighters, carefully

In response to the high demand and low supply of air freight right now, carriers could investigate short- to medium-term opportunities to boost their cargo services. Airlines can enhance their flexibility through measures such as increasing the deployment of so-called preighters, or passenger airplanes that are used to transport cargo. Airlines may look at freighter conversions, especially as their passenger fleets reduce in number.

Airlines need to be agile. Rushing headlong into developing and maintaining a large freighter fleet again comes with risk. Airlines need to grow cargo in an agile way that allows for quick adjustments; pursuing such a play should be seen as part of a wider theme of establishing a more flexible production setup. High fixed costs combined with unpredictable demand levels outside an airline’s control increase the need for airlines to be able to scale down supply nimbly.

The impact of the COVID-19 pandemic is far from over. There is some relief to be found in various parts of the world now that vaccinations have begun, but the road to recovery for air traffic will take several years. The shape of the post-COVID-19 airline sector is becoming clearer and holds lessons for airlines today. Multiple longer-running trends have been accelerated, such as digitization and the phasing out of less efficient aircraft. Burdened by debt, many carriers have depleted their cash reserves. But the forecast is not without bright spots. Travel will become greener and more efficient, and people are itching to travel again for holidays. Taking steps now will help airlines thrive in this transformed sector.

Jaap Bouwer is a senior expert in McKinsey’s Amsterdam office, Steve Saxon is a partner in the Shenzhen office, and Nina Wittkamp is a partner in the Munich office.

The authors wish to thank Alex Dichter and Vik Krishnan for their contributions to this article.

This article was edited by Jason Li, a senior editor in the Shanghai office.

Explore a career with us

Related articles.

Airline data: What next beyond crisis response?

Will airline hubs recover from COVID-19?

‘Prepare for the marathon and be ready for the course to change’: An interview with the Boston Logan Airport CEO