key: cord-0731165-i1c6dcmg
authors: Midorikawa, Haruhiko; Aiba, Miyuki; Lebowitz, Adam; Taguchi, Takaya; Shiratori, Yuki; Ogawa, Takafumi; Takahashi, Asumi; Takahashi, Sho; Nemoto, Kiyotaka; Arai, Tetsuaki; Tachikawa, Hirokazu
title: Confirming validity of The Fear of COVID-19 Scale in Japanese with a nationwide large-scale sample
date: 2021-02-10
journal: PLoS One
DOI: 10.1371/journal.pone.0246840
sha: 4e07ff1b244afe3b8be87d28732811128d58658f
doc_id: 731165
cord_uid: i1c6dcmg

Assessing fear and anxiety regarding COVID-19 viral infection is essential for investigating mental health during this epidemic. We have developed and validated a Japanese-language version of The Fear of COVID-19 Scale (FCV-19S) based on a large, nationwide residential sample (n = 6,750) recruited through news and social media responding to an online version of the questionnaire. Data was collected from August 4–25, 2020. Results correlated with K6, GAD-7 and IES-R psychological scales, and T-tests and analysis of variance identified associated factors. All indices indicated the two-factor model emotional fear reactions and symptomatic expressions of fear a better fit for our data than a single-factor model in Confirmatory Factor Analysis (χ(2) = 164.16, p<0.001, CFI 0.991, TLI = 0.985, RMSEA = 0.043). Socio-demographic factors identified as disaster vulnerabilities such as female sex, sexual minority, elderly, unemployment, and present psychiatric history associated with higher scores. However, respondent or family member experience of infection risk, or work/school interference from confinement, had greatest impact. Results suggest necessity of mental health support during this pandemic similar to other disasters.

Corona virus disease 2019 (COVID-19) has become an important and urgent threat to global health. Since the cluster of cases of pneumonia of unknown etiology were reported in Wuhan, China in December 2019 [1] , COVID-19 transmission continued spreading, and on 30 January 2020 WHO declared the outbreak a Public Health Emergency of International Concern [2] . Subsequently, despite various public health responses aimed at slowing the spread of COVID-19, many countries have faced a critical health crisis [3] . As of early September 2020, that facilitates sharing surveys via email, smartphone applications, and social media platforms such as Facebook and Twitter. Exclusion criteria for participants were: second or later responses with a duplicate IP address, and no response. The number of participants was confirmed to be sufficient by referring to the previous study on FCV-19S [21] and literature that indicated the minimum sample size required for CFA [34] .

FCV-19S is a seven-item self-administered scale developed by Ahorsu et al. [21] . Answers included "strongly disagree," "disagree," "neither agree nor disagree," "agree," and "strongly agree". The minimum score possible for each question is 1 (strongly disagree), and the maximum is 5 (strongly agree). A total score is calculated by adding up each item score (ranging from 7 to 35). The higher the score, the greater the fear of COVID-19. We created the Japanese version of the FCV-19S by referring to the guidelines of the International Society for Pharmacoeconomics and Outcomes Research Task Force [35] . Two psychiatrists independently translated the original version of the FCV-19S from English to Japanese, with the permission of the original author. Both translations were integrated into a single version back-translated into English by a native speaker literate in Japanese and reviewed by the research team. The questionnaire also included socio-demographic items (gender, age group, occupation, residence, and history of psychiatric treatment), and psychological scales: FCV-19S-J, Kessler Screening Scale for Psychological Distress (K6), Generalized Anxiety Disorder -7 (GAD-7), and Impact of Event Scale-Revised (IES-R). The K6 is a short, six-item questionnaire developed to screen for mood and anxiety disorders. The total score of the K6 ranges from 0-24. We used the reliable and validated Japanese version of the K6 in the survey [36] . The GAD-7 is a self-administered questionnaire developed to assess the severity of Generalized Anxiety Disorder (GAD) by extracting questions related to anxiety disorders from the PHQ [37] . It consists of seven items, and symptom intensity during the past two weeks is rated on a 4-point scale (0 to 3) with a total score ranging from 0-21. We used the reliable and validated Japanese version of the GAD-7 in the survey [38] . The IES-R (Impact of Event Scale-Revised), a revision of the former IES [39] , is a self-administered questionnaire developed to investigate traumatic distress [40] . The IES-R consists of 22 items, including 7 intrusion items, 8 avoidance items, and 7 hyperarousal items, and evaluates the intensity of symptoms in the past week on a 5-point scale (0 to 4). The total score ranges from 0-88. We used the reliable and validated Japanese version of the IES-R for the survey [41] .

Finally, question items tapped experience during the COVID-19 epidemic. A stress level question "How stressful have you felt over the past month in relation to the COVID-19?" had Likert scale responses: "not at all (1)," "not much (2)," "neither (3)," "a little (4)," "very much (5)" and "other (6)". Other times were: "I was at risk of infection," "My family members were at risk of infection," "I was home alone because I was at risk of infection," "Self-confinement interfered with work or school," and "I or my family members have been discriminated against or bullied." Multiple selections were possible, and the first three experiences were combined into one item "I or my family was at risk of infection" in the analysis. Participants were also asked a multiple-choice question about effective coping methods during the self-confinement period.

We first examined the distribution of respondents' socio-demographic characteristics and experience during COVID-19 epidemic. Next, we conducted confirmatory factor analyses (CFA) of single-factor and two-factor model with maximum likelihood estimation to verify structural validity of Japanese translated FCV-19S-J items and compared goodness of fit:: comparative fit index (CFI � .95), Tucker-Lewis index (TLI � .95), root mean square error of approximation (RMSEA � .06), and Akaike's information criterion (AIC) [42, 43] . Reliability was evaluated using Cronbach's alpha. To examine construct validity, we also calculated Pearson correlation coefficients between the FCV-19S-J and self-assessment mental health scales K6, GAD-7, IES-R. Strength of correlation coefficient values were: r < 0.30 "weak," r = 0.30-0.59 "moderate", and r � 0.60 "strong" [44] . T-tests and analysis of variance (ANOVA) examined association between FCV-19S-J scores, and socio-demographic and epidemic experience data during COVID-19. P-values were adjusted by Bonferroni correction for multiple comparisons (p < 0.05/27 = 0.00185). The percentage of missing values for each item in the sociodemographic characteristics was shown in the characteristics of respondents. Since the number of missing values was very small, we did not consider the presence of missing values in the socio-demographic factor in the factor analysis. In the correlation analysis, T-test, and ANOVA, missing values were excluded for each statistical testing. CFA was performed with the statistical package Amos version 26 for Windows and the other statistical analyses were performed using IBM SPSS version 22 for Windows.

Before starting the survey, we explained the purpose of the survey and that participation was voluntary and that survey was anonymous. Participants gave their consent by ticking a box to confirm that they understood the information provided to them and voluntarily agreed to participate in the survey. Only those who agreed to cooperate in the survey would be able to proceed to the questionnaire. This study was approved by the medical ethics committee of the University of Tsukuba (Registration No.1546-1).

A total of 7912 responses were obtained, 7389 after duplicate responses were removed based on IP addresses. The attrition rate was 19.2% (1,420/7,389). Of these, n = 6750 were included in the analysis (no response to FCV-19S-J = 639). 4283 (63.5%) of the respondents were female. Age of respondents ranged from teenagers to over 60 years old, but only 4.3% of the respondents were 60y<. All socio-demographic variables are shown in Table 1 . Table 2 reports the factor loadings of single-factor and two-factor models. In the two-factor model, Factor 1 emotional fear reactions was composed of psychological dimensions such as anxiety and fear, and Factor 2 symptomatic expressions of fear with physiological dimensions such as sweating, palpitations, and insomnia.

The results of the CFA model fit are reported in Table 3 . All indices indicated better fit for the two-factor model than a single-factor model. Reliability analysis assessing both models indicated a Cronbach's alpha of 0.83 for the single-factor model, and 0.77 for Factor 1 and 0.83 for Factor 2.

The correlations between FCV-19S-J and other mental health self-assessment measures are shown in Table 4 . Total FCV-19S-J � Factor 1 and Total FCV-19S-J � Factor 2 were highly correlated (r > 0.60). Total FCV-19S � K6, � GAD-7, and � IES-R were moderately correlated (0.30< r <0. 60) , as were correlations between individual factors and the other scales. Because the association levels between FCV-19S-J items and other scales are not disparately dissimilar, this indicates construct validity; in particular, concurrent validity for the instrument and convergent validity for the factors being measured.

The mean score of FCV-19S was 16.67±4.851 and the mode was 15, which accounted for 8.3% of the total score. The mean score of Factor1 was 12.18±3.490 and the mode was 14, which accounted for 11.6% of the total score. The mean score of Factor 2 was 4.48±1.983 and 50.0% of the respondents answered no to all questions.

The association between the FCV-19S-J and socio-demographic factors and experience during COVID-19 epidemic is shown in Tables 5 and 6 . FCV-19S-J total score, Factor 1 score and Factor 2 scores were significantly different by gender, age group, occupation, and psychiatric 

Confirming validity of FCV-19S in Japanese with a nationwide large-scale sample history. However, all of these effect sizes were small, and may be significant due to large sample size. Females were higher in total and Factor 1 scores, non-identified in Factor 2. Respondents in their 50s and older tended to have higher scores than other age groups. The unemployed scored consistently higher even compared to health care workers. With regard to residency, respondents in a prefecture with a population of 5 million< had higher scores only in total score. Psychiatric history was also significant in total and factored scores, as was infection risk to self or family, and experiencing work/school interruption. Whether respondents or family members had been discriminated or bullied against was not associated with FCV-19S-J scores, nor was effective coping during self-confinement.

We examined validity of the Japanese version of FCV-19S and clarified the factors related to the fear of COVID-19. To our knowledge, this is one of the largest nationwide surveys in Japan. The results of the CFA revealed that two-factor model had a better model fit than onefactor model. In a Japanese sample, Masuyama et al. concluded the translated FCV-19S was composed of two factors [30] and the goodness of fit of the one-factor model was not high in Wakashima's study [31] . Therefore, a two-factor model composed of emotional fear reactions and symptomatic expressions of fear appears plausible. Similar results have been confirmed in the Spanish version using a sample in Peru and Lima [45] , the Russian version of a sample in Russia and Belarus [46] , and the Hebrew version among an Israeli population [27] . The names of these two factors were taken from the study by Tzur Bitan et al. [27] . As a result of validation, the two-factor model showed sufficient values in all of the factor loadings, the goodness of fit indices, and Cronbach's alpha. These results, plus correlation levels with other self-assessment scales, confirm the FCV-19S-J has sufficient validity. The results of this study showed 80% of respondents felt major stress during the COVID-19 epidemic. The average score for FCV-19S was 16.67. The US and New Zealand, Spain, Russia, Israel and Pakistan were distributed between 15-19 points [22, 25-27, 46, 47] , and Asian countries such as Iran, Bangladesh and Turkey had scores above 20 [21, 24, 48] . Compared to other countries, the FCV-19S-J score in Japan was comparatively lower, perhaps due to lower disease 

Confirming validity of FCV-19S in Japanese with a nationwide large-scale sample Arabic, Italian and Spanish versions, average scores for Item 3, Item 6 and Item 7 tended to be lower [23] [24] [25] 49] . These tendencies were consistent with our data. However, the original version of Ahorsu et al.'s data did not show this trend, suggesting the degree to which fear of COVID-19 causes physical responses varies across countries and regions. The FCV-19S-J scored higher in females and older adults, similar to Bangladesh, Greece and India [24, [50] [51] [52] , and a People's Republic of China study using a different instrument [53] In addition, a new finding from our study is respondents who selected a gender other than male or female had higher scores, especially in Factor 2 (physical symptoms). Transgender, non-conforming people and gender dysphoria often have mental health problems due to discrimination, prejudice and social inequalities [54, 55] . Social distance policies associated with COVID-19 may worsen these problems by severing their relationships with supportive and affirmative people and organizations [56] . More attention should be paid to this point.

With regard to age, some reports show higher FCV-19S scores in older adults [50] , while others do not [24] . Older adults are known to be at higher risk after COVID-19 infection because they have lower immunity and often have chronic diseases [57] [58] [59] . For this reason, it is understandable that their fear of COVID-19 is higher than that of other generations. The high scores of the unemployed are consistent with the study in Pakistan [22] .

With the report of low socioeconomic status being associated with fear of COVID-19 [27] and the report of suicide due to the economic slump associated with lockdown [60] , the mental health of the socially vulnerable people is an important issue during the COVID-19 epidemic. It is also important to note that FCV-19S-J scores among health care workers did not differ significantly from other occupations. Doshi D et al. reported higher FCV-19S scores among health care workers [52] . As the extent of the risk of infection in the site where each health care worker works is not clear, it would be desirable for future research to clarify this point.

Another new finding is the FCV-19S-J score was higher in psychiatric patients. Chang et al. reported that the FCV-19S is useful for measuring fear of COVID-19 even in patients with psychiatric disorders [51] : however, this study did not compare results with those of patients without psychiatric disorders. Some reports suggest that people with mental illness are more likely to have mental health problems during the COVID-19 epidemic [61, 62] . It should be noted that fear of COVID-19 may be heightened in individuals with pre-existing mental health problems.

Compared to factors mentioned so far, presence of experience, such as whether or not respondents or their family were at risk of infection and whether or not self-confinement interfered with work or school showed larger effects. This is reasonable as these are considered to be more direct factors for fear of COVID-19. As for the result that experiences of discrimination and bullying were not related to FCV-19S-J scores, this might be due to only a small number in the current study reporting such experiences.

Socio-demographic factors found to be associated with fear of COVID-19 in this study, such as females, sexual minorities, the elderly, and the unemployed, were consistent with the characteristics of vulnerable populations during disasters. Females, children, adolescents, the poor, the elderly, and those with pre-existing health problems have been identified as vulnerable populations that often experience psychological morbidity as a result of disasters [63, 64] . This association suggests that COVID-19 is a disaster and that utilizing the findings of disaster psychiatry can be useful in the COVID-19 epidemic. In addition, among the factors identified in this study, attention needs to be paid to older age and unemployment in particular, from the perspective of the digital divide [65] . This is because, during in the midst of the COVID-19 epidemic use of online services is growing rapidly [66] ; however, some from these groups may not be able to use such services. In addressing mental health issues associated with COVID-19, greater attention may need to be paid to those who are unable to benefit from digital healthcare technologies.

The limitations of the study are as follows. First, sampling bias needs to be considered because anyone was free to participate in this survey. Socio-demographic factors such as gender, age and living place are not representative of the general population of Japan. Also, the possibility that a population with a strong interest in COVID-19 was selected cannot be ruled out. The reason for the large number of respondents with present psychiatric history may be partially due to psychiatrists' use of social media and snowball sampling in recruiting. Second, the situation of infection changed during the study period, and trends of infection varied by region. For this reason, individual responses cannot be assured to be responses to the same situation. Third, because of the cross-sectional design of this study, it was difficult to assess causality.

Despite these limitations, this study is significant as it confirms the validity of FCV-19 using large-scale data and investigates fear of COVID-19 and related factors at the nationwide level. Besides mental health care of those actually affected by COVID-19, support may be considered necessary for those with vulnerable factors our study identified. Further research is needed to determine which populations are more likely to have a heightened fear of COVID-19 and which are more likely to cause physical and mental problems. To this end, it is desirable to conduct longitudinal study by using representatives of the general population.

(PDF) S1 Dataset. Anonymized data set. (XLSX) 

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Novel Coronavirus(2019-nCoV) Situation Report-10

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

World Health Organization. WHO Coronavirus Disease (COVID-19) Dashboard. 2020

COVID-19: Transmission, prevention, and potential therapeutic opportunities

Projecting the transmission dynamics of SARS-CoV-2 through the postpandemic period. Science (80-)

COVID-19 Special Site

The Psychological Impact of 'Mild Lockdown' in Japan during the COVID-19 Pandemic: A Nationwide Survey under a Declared State of Emergency

Trends in suicide in Japan by gender during the COVID-19 pandemic, up to

Immediate psychological responses and associated factors during the initial stage of the 2019 coronavirus disease (COVID-19) epidemic among the general population in China

Multidisciplinary research priorities for the COVID-19 pandemic: a call for action for mental health science. The Lancet Psychiatry

The psychological impact of COVID-19 on the mental health in the general population

The psychological impact of quarantine and how to reduce it: rapid review of the evidence

Mental health problems and social media exposure during COVID-19 outbreak

How mental health care should change as a consequence of the COVID-19 pandemic

Public responses to the novel 2019 coronavirus (2019-nCoV) in Japan: Mental health consequences and target populations. Psychiatry

Fear and Stigma: The Epidemic within the SARS Outbreak

Functional Fear Predicts Public Health Compliance in the COVID-19 Pandemic

The fear of COVID-19 and its role in preventive behaviors

The Fear of COVID-19 Scale: Development and Initial Validation

The Psychometric Validation of FCV19S in Urdu and Socio-Demographic Association with Fear in the People of the Khyber Pakhtunkhwa (KPK) Province in Pakistan

Validation and Psychometric Evaluation of the Italian Version of the Fear of COVID-19 Scale

Psychometric Validation of the Bangla Fear of COVID-19 Scale: Confirmatory Factor Analysis and Rasch Analysis

The fear of COVID-19 scale: Validation in spanish university students

Validation of the Fear of COVID-19 Scale in a US College Sample

Fear of COVID-19 scale: Psychometric characteristics, reliability and validity in the Israeli population

Validation of the Spanish version of Fear of COVID-19 Scale: Its association with acute stress and coping

The Fear of COVID-19 Scale: Psychometric Properties of the Ethiopian Amharic Version

Validation and Psychometric Properties of the Japanese Version of the Fear of COVID-19 Scale Among Adolescents

The Japanese version of the Fear of COVID-19 scale: Reliability, validity, and relation to coping behavior

Elderly people and responses to COVID-19 in 27 Countries

SurveyMonkey User Manual Customer Guide for Account Navigation, Survey Creation, Distribution & Analysis

Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety. Educational and psychological measurement

Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and Cultural Adaptation

The performance of the Japanese version of the K6 and K10 in the World Mental Health Survey Japan

A brief measure for assessing generalized anxiety disorder: The GAD-7

Validation and utility of a Japanese version of the GAD-7

Impact of event scale: A measure of subjective stress

editor. Assessing psychological trauma and PTSD

Reliability and validity of the Japanese-language version of the Impact of Event Scale-Revised (IES-R-J): Four studies of different traumatic events

Confirmatory factor analysis for applied research

Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives

Criteria for assessing the tools of disability outcomes research

Psychometric Properties of a Spanish Version of the Fear of COVID-19 Scale in General Population of

COVID-19 Fear in Eastern Europe: Validation of the Fear of COVID-19 Scale

Evaluation of the English Version of the Fear of COVID-19 Scale and Its Relationship with Behavior Change and Political Beliefs

Intolerance of Uncertainty and Mental Wellbeing: Serial Mediation by Rumination and Fear of COVID-19

Psychometric Evaluation of the Arabic Version of the Fear of COVID-19 Scale

Psychometric Properties of the Greek Version of FCV-19S

Factors related to preventive COVID-19 infection behaviors among people with mental illness

Assessing Coronavirus Fear in Indian Population Using the Fear of COVID-19 Scale

A nationwide survey of psychological distress among Chinese people in the COVID-19 epidemic: Implications and policy recommendations

A systematic review of social stress and mental health among transgender and gender non-conforming people in the United States

Psychiatric disorders in individuals diagnosed with gender dysphoria: A systematic review

Sexual and Gender Minority Stress Amid the COVID-19 Pandemic: Implications for LGBTQ Young Persons' Mental Health and Well-Being. Public Health Reports

COVID-19 Health Worries and Anxiety Symptoms among Older Adults: The Moderating Role of Ageism

Analyze the psychological impact of COVID-19 among the elderly population in China and make corresponding suggestions

Mental Health and the Covid-19 Pandemic

COVID-19 suicides in Pakistan, dying off not COVID-19 fear but poverty?-The forthcoming economic challenges for a developing country

Psychiatric symptoms related to the COVID-19 pandemic

Do psychiatric patients experience more psychiatric symptoms during COVID-19 pandemic and lockdown? A case-control study with service and research implications for immunopsychiatry

Disaster victims speak: Part II. Summary and implications of the disaster mental health research

Management of trauma in special populations after a disaster

Determinants of broadband access and affordability: An analysis of a community survey on the digital divide

Digital Mental Health and COVID-19: Using Technology Today to Accelerate the Curve on Access and Quality Tomorrow

Arai, Hirokazu Tachikawa.