The aim of the proposed project is to provide a comprehensive assessment of the mental health impact of the ongoing COVID-19 outbreak in Spain. A prospective cohort study of three groups will be conducted: (1) COVID-19 cases or close contacts recently isolated or quarantined; (2) healthcare workers; (3) adult general population.

Recruitment through epidemiologic surveillance services, hospitals, primary care centers in 6 Spanish Autonomous Communities (groups 1 & 2) and a random sample of individuals using telephone numbers as the sampling frames(group 3). Web-based surveys or telephone interviews at baseline and 6-month follow-up assess: current living and employment status; COVID-19 infection status; mental health (post-traumatic stress, panic, depression, anxiety, alcohol and substance use, suicidality; services use); psychological functioning; general health status and quality of life. Specific modules focus on isolation or quarantine (group 1) and care for COVID-19 patients (group 2). Short bi-weekly brief follow-up surveys among groups 1 & 2 will provide additional assessment of main outcomes in the two months following baseline assessment.

Background

Previous research on infectious disease outbreaks found substantial mental health impact among cases placed under isolation or quarantine, among those working in the various healthcare systems managing the outbreaks, and in the general population. Mental disorders expected to rise during and after viral outbreaks include anxiety disorders, especially post-traumatic stress disorder, as well as depression, and suicidality.

Justification

The magnitude of the ongoing 2019 novel coronavirus disease (COVID-19) pandemic points to the urgent need to quantify mental health impact of the outbreak. Epidemiological research enables rapid health needs assessment to benchmark and monitor the ongoing onset and persistence of adverse mental health outcomes in the population, and to provide situational awareness, i.e., essential information to understand current and midterm mental healthcare needs, plan adequate responses, and allocate appropriate resources. The project is part of an ongoing international effort to collect cross-nationally comparable data on the mental health impact of the COVID-19 pandemic (World Mental Health Surveys).

Hypotheses:

1. Adverse mental health outcomes, such as symptoms of depression and anxiety(1), post-traumatic stress disorder, and suicide, will increase from the onset of the COVID-19 outbreak, compared to the previous mental health status. The most pronounced increase will be in vulnerable groups, i.e., among recently isolated or quarantined individuals and healthcare workers.
2. Working as healthcare worker, having been quarantined, having family members dead by COVID19 will be some of the substantial risk factors to develop sub-sequent high levels of adverse mental health. Also, some preoutbreak individual characteristics will be predictors of the mental and psychological health impact, such as the level of formal education or having a history of psychiatric illness.
3. A substantial part of the participants would require mental health services and some type of psychological support. Most of the subjects reporting symptoms will not be using the already available resources.

OBJECTIVES

The primary objectives of the proposed project are to:

1. assess the mental and psychological health impact of the COVID-19 outbreak among cases or close contacts recently isolated or quarantined, healthcare workers and the general population
2. identify risk and protective factors for adverse mental health onset and persistence;
3. quantify the use of available mental health resources;
4. quantify and characterize unmet mental healthcare needs. These objectives will be achieved both at the short-term (onset and bi-weekly surveys) and the medium term (6 months).

The secondary objectives of the proposed project are to:

1. assess the potential benefits of using available mental health resources;
2. provide a reference point (benchmark) and evolution over time for future long-term research on COVID-19 related mental and psychological health impact in the population.

Study population and recruitment

Three separate population groups will be assessed: (1) recently isolated or quarantined COVID-19 cases and close contacts; (2) healthcare workers, including medical personnel directly and indirectly involved with patients as well as other personnel (e.g. administrative workers) and (3) a representative sample of the general adult population (including those tele-working and working outside of home).

Even though the terms quarantine and isolation are often used interchangeably, here we differentiate them following Brooks et al. According to this, Quarantine is the separation and restriction of movement of people who have potentially been exposed to a contagious disease to ascertain if they become unwell, so reducing the risk of them infecting others, and Isolation is the separation of people who have been diagnosed with a contagious disease from people who are not sick. Finally, we use the term confinement to define the situation of the population in Spain who, in order to prevent further spread of the virus, has been restricted to stay at home, and is only allowed to get out for specific tasks that need to be justified.

Calculation and justification of the sample size

General population: a fixed sample size of n= 3,500 ensures a two-sided 95%CI of 0.025 for a sample proportion of 0.20. Health professionals: up to 85,000 professionals work in the participating institutions (30,000 at primary health care centers, 55,000 at hospitals). Assuming a 15% participation rate, we expect a sample size of >10,000. COVID-19 cases: Given high numbers of (hospitalized) cases in Barcelona and Community of Madrid (the main recruitment areas) and the population coverage of the participating institutions, we expect to recruit n=2,000 cases and close contacts. Within each population subgroup, a sample of at least 2,000 individuals achieves power > 0.80, with alpha=0.05, to detect an odds ratio of 1.6 on a binary independent variable (of which 30% are in the group X=1) from a multivariable logistic regression (with R2 < 0.3 between X and remaining independent variables). An overall sample of 16,000 achieves power > 0.90 to detect an odds ratio of 1.2 under the same conditions.

Study methodology/Procedures

Adult population groups 1 (cases and close contacts) and 2 (healthcare workers) will be assessed at baseline using web-based self-report surveys (with response time ~15 minutes) including both common and specific modules (specific for groups 1 & 2). People with unavoidable limitations to answer the webbased questionnaire will also be eligible to participate in the study. In this case, the interview will be conducted by telephone. The telephone interview will be carried out from the same center or may be outsourced, subject to the patient's consent to share his/her contact details. Additionally, upon completion of the baseline survey, group 1 (cases and close contacts) and group 2 (healthcare workers) will be invited to respond very brief bi-weekly surveys (response time 3 - 5 minutes) on a maximum of 4 occasions to prospectively assess adverse mental health and the use of available resources to counter mental health impact. General population participants will be assessed through telephone interviews. All baseline participants of the 3 adult population groups will be assessed at 6 months follow-up using web-based self-report surveys or telephone interviews similar to their baseline assessment. At follow-up, groups 1 and 2 will also receive more extensive clinical screeners that allow to accurately assess common psychiatric disorders.

Analyses

The data will be pseudo-anonymized through encrypted identifiers, separating the personal information from the rest of the study data, to guarantee privacy and ensure the anonymized treatment of the data in the analysis.

Descriptive statistics of mental health status, mental health service use, and use of available resources at baseline and 6-month follow-up, and bivariate associations of these outcomes with potential risk and protective factors are estimated, stratified by population groups.

Bi-weekly brief survey data will be analyzed using time-lagged multilevel models and network analysis to estimate the associations of short-term dynamic risk and protective factors with mental health status and use of resources.

Multivariable generalized linear models will be used to identify baseline risk and protective factors for onset and persistence/deterioration of adverse mental health outcomes at follow-up. Potential benefits of use of available resources on preventing onset or persistence of adverse mental health outcomes will be analyzed using propensity scores methods to address the lack of baseline randomization and likely imbalance between treatment groups (natural experiment research design). All analyses will adjust for time-of-survey.

Multiple imputation methods will be used to deal with missing values.

LIMITATIONS AND RISK AND CONTINGENCY PLANS

The established network of collaborators we already have in place guarantees adequate recruitment of otherwise hard-to-reach groups 1 and 2 (cases and close contacts, healthcare workers). The lack of randomized sampling of surveillance agencies and healthcare facilities may affect representativeness of findings; this limitation will be countered by statistical weighting techniques based on all information available a posteriori in centralized national case register and healthcare facility data. Response rates will be maximized by repeated invitations and reminders.

ETHICAL CONSIDERATIONS AND CONFIDENTIALITY

The study is in line with the principles established by national and international regulations, including the Declaration of Helsinki (64th WMA General Assembly, Fortaleza, Brasil, Oct 2013) and the Code of Ethics. The project will start upon approval by the independent PSMAR Clinical Research Ethics Committee (PSMAR-CEIm). Recruitment in collaborating institutions will be initiated only after project approval by its corresponding ethics committee, whenever additional approval is required.

All personal data will be handled following Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016, and the national Organic Law 3/2018, of December 5. All participants will provide explicit informed consent. When participants access the link to the online questionnaire platform, the participant's information form specific for the subpopulation they belong to will appear in the first screen, which will be written in understandable language, and will describe in clear language the aims, methods, and implications of the research. At the end of the information about the study, the different points of the informed consent will be presented in the form of check options that will have to be clicked. Access to the questionnaire will only be allowed if all check options have been activated.

The answers to the different consent questions will be stored in the database, together with the other responses, and may be used as evidence that the participant did, indeed, consent. Exemption from obtaining a signed informed consent has been requested, given the observational, online data collection nature of the study and given the fact that, to be able to access the questionnaire, respondents will have to actively check all consent questions

The Qualtrics survey platform (web-based surveys; qualtrics.com) will be used for data collection in subpopulations of cases and close contacts (1) and healthcare center workers (2). This platform has three of the most important security certifications: FedRAMP, ISO 27001 and HITRUST Common Security Framework (CSF), to meet the security requirements of the most regulated industries and organizations and those of the RGPD (European General Data Protection Regulation). In addition, the platform has additional technology that allows the user to also comply with this regulation (GDPR OneTouch Data Deletion). The data collected through the platform is stored in servers located in the EU. All data stored in the EU data center is encrypted via the AES256 cypher standard. All data transmitted to the Qualtrics platform is encrypted via the industry standard protocol TLS 1.2 and higher. The only personal information that will be requested to participants in the study are contact details with the solely objective to be able to invite participants to the follow up surveys and to send reminders within the framework of the project. This personal data will be collected within the questionnaire, once the participant has given his or her consent to participate in the study. The data will be pseudo-anonymized and neither the study investigators nor the persons in charge of the analysis will have access to the personal data. Only the data manager designated for the study will be able to relate the personal data of the participants to the questionnaire responses. All data will be collected in a research file under the responsibility of the institution, will be considered confidential and will only be used by researchers for the purposes of the project. A Data Protection Officer involved in all stages of data processing has been appointed, DPO: protecciodedades@imim.es).

With regard to the Adult General Population sample (subpopulation 3), the external company IPSOS survey company (web-based surveys; ipsos.com), will be in charge of selecting the sample, inviting participants to the study and carrying out the interviews. A verbal informed consent to participate will be obtained: at the beginning of the contact call, a presentation text with information on the study will be read, and explicit consent will be asked to the participant to respond a questionnaire in two different occasions, baseline and after 6 months. This part of the call where explicit informed consent is requested will be recorded and stored. . At the end of the 6-month follow-up interview, participants will be asked again for verbal recorded consent asking permission for the company to provide contact details of the participant to the study investigators, so that investigators can recontact them for eventual subsequent follow up assessments of the study. IPSOS will ensure that personal data of the participants will be treated with the maximum confidentiality measures according to the current regulations in force (Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016, and the national Organic Law 3/2018, of December 5), and that personal data will be exclusively used within the framework of this study. IPSOS data protection policy is detailed in the following link: https://www.ipsos.com/sites/default/files/201804/Global\_Data\_Protection\_and\_Privacy\_Policy.pdf.

A specific survey platform developed by IPSOS will be used in this group. IPSOS complies with all relevant regulations for the industry, including the new EU directive: GDPR, but also with the ethical, privacy and anti-spam precepts recommended by European Society for Opinion and Marketing Research (ESOMAR). The extensive privacy policy is displayed during the registration process and must be accepted by all the panelists. To this end, Ipsos applies strict data protection and security protocols for both client and respondent data. Physical security measures include strict control of access to data centers, firewalls, antivirus procedures, data retention/destruction policies, or disaster recovery plans.

Given the observational approach of the study, without therapeutic intervention, it is considered that it will pose minimum risk to the respondents. Participants with psychological distress will be provided with a list of effective care resources at the end of the questionnaire, including coordinates to nearby emergency care in case of suicidal symptoms.

Inclusion Criteria:

• Recently isolated or quarantined cases and close contacts:

  a) Individuals aged 18 or older; and b) laboratory confirmed or suspected COVID-19 cases or quarantined close contacts.

• Healthcare workers:

  a) health center employees currently working in the participating institutions, including medical and nursing personnel directly and indirectly involved with patients as well as all other personnel (administrative workers, logistic workers, etc.); and b) aged 18 years or older.

• Spanish general adult population:

  1. residents in Spain; b) aged 18 years or older; c) with a land-line or mobile telephone

Exclusion Criteria:

• Recently isolated or quarantined cases and close contacts:

  a) individuals too sick to participate (but they will become eligible after clinical improvement); b) unable to understand the survey language; and c) do not provide explicit consent to participate.

• Healthcare workers:

  a) individuals too sick to participate (but they will become eligible after clinical improvement); b) unable to understand the survey language; and c) do not provide explicit consent to participate.

• Spanish general adult population:

  1. unable to understand the survey language; and b) do not provide explicit consent to participate.

Three separate population groups will be assessed: (1) recently isolated or quarantined COVID-19 cases and close contacts; (2) healthcare workers, including medical personnel directly and indirectly involved with patients as well as other personnel (e.g. administrative workers); (3) a representative sample of the general adult population (including those tele-working and working outside of home).