“What If We Get Sick?”: Spanish Adaptation and Validation of the Fear of Illness and Virus Evaluation Scale in a Non-clinical Sample Exposed to the COVID-19 Pandemic

Abstract

Distinct sources of stress have emerged during the COVID-19 pandemic. Particularly, fear is expected to generate significant psychological burden on individuals and influence on either unsafe behavior that may hinder recovery efforts or virus-mitigating behaviors. However, little is known about the properties of measures to capture them in research and clinical settings. To resolve this gap, we evaluated the psychometric properties of a novel measure of fear of illness and viruses and tested its predictive value for future development of distress. We extracted a random sample of 450 Chilean adult participants from a large cross-sectional survey panel and invited to participate in this intensive longitudinal study for 35 days. Of these, 163 ended up enrolling in the study after the demanding nature of the measurement schedule was clearly explained to them. For this final sample, we calculated different Confirmatory Factor Analyses (CFA) to evaluate the preliminary proposed structure for the instrument. Complementarily, we conducted a content analysis of the items to qualitatively extract its latent structure, which was also subject to empirical test via CFA. Results indicated that the original structure did not fit the data well; however, the new proposed structure based on the content analysis did. Overall, the modified instrument showed good reliability through all subscales both by its internal consistency with Cronbach’s alphas ranging from 0.814 to 0.913, and with test–retest correlations ranging from 0.715 to 0.804. Regarding its convergent validity, individuals who scored higher in fears tended to also score higher in depressive and posttraumatic stress symptoms at baseline. Furthermore, higher fears at baseline predicted a higher score in posttraumatic stress symptomatology 7 days later. These results provide evidence for the validity, reliability, and predictive performance of the scale. As the scale is free and multidimensional potentially not circumscribed to COVID-19, it might work as a step toward understanding the psychological impact of current and future pandemics, or further life-threatening health situations of similar characteristics. Limitations, practical implications, and future directions for research are discussed.

Psychometric Validation of the Arabic Fear of Illness and Virus Evaluation

“…The suggested structure for FIVE in the present study is based on its factor analysis and model fit indices. As a previous study [24] suggested four domains for items 1-19 (Fears of Getting Sick, Fears that Others May Get Sick, Fears of Concrete Limitations and Fears of not being able to meet the Basic Needs), our findings that were based on a sufficiently larger and sociodemographically different sample obtained a structure (Fears of Getting Sick; items 1-5, Fears that Others Get Sick; items 6-9 and Fears of the Impact on Social Life; items 10-19) that is somewhat different from that proposed in the Spanish version. On the other hand, when items 1-35 were analyzed, the data showed an improved fit in four domains that was slightly different from the one originally set by Ehrenreich-May.…”

“…The conformity between subscales' items in the current Arabic version of FIVE examined in the present study is comparable to the Arabic Fear of COVID-19 Scale (FCV-19S) [28]. For instance, the internal consistency of Fears about Contamination and Illness subscale (items 1-9) was 0.89 and Fears about Social Distancing (items 10-19) was 0.91, which are relatively higher than that found in the Arabic FCV-19S [28] and Spanish FIVE [24]. The internal consistency of proposed structure of FIVE in Spanish language was α = 0.88 for Fears of Getting Sick from an illness or virus (items 1-4); α = 0.74 for Fears that Others May Get Sick from an illness or virus (items 5-9); α = 0.85 for Fears of Concrete Limitations due to an illness or virus (items 10,11,13,[15][16][17]; while the subscale Fears of not being able to meet Basic Needs of subsistence and work due to an illness or virus (items 12,14,18,19) had α = 0.79.…”

“…The correlation coefficient of FIVE with Anxiety subscale was high (r = 0.83) and moderate with depression subscale (r = 0.66) in a previous study [38]. FIVE was also correlated with both depressive symptomatology (PHQ-9) and posttraumatic stress (ITQ) [24].…”

“…As discussed by a previous study [24], items 1-19 are the main questions in FIVE that help in understanding the extent of fear among individuals, more specifically items 1-9 concerned with fear of self or others getting infected. In addition, the developer did not include items 20-33 when scoring the general fear because they only provide additional information about the change in people's behaviors due to exposure to an illness or virus [24]. Additionally, items 34-35 are related to functionality but not measuring fear itself.…”

“…A number of tools were developed to assess the psychological influence of COVID-19 outbreak [22,23]. Dr. Jill Ehrenreich-May had developed Fear of Illness and Virus Evaluation scale in English (FIVE), a 35-item self-report measure consisting of four subscales; Fears about Contamination and Illness, Fears about Social Distancing, Behaviors Related to Illness and Virus Fears and Impact of Illness and Virus Fears [24]. Before using the scale in different demographic populations, it is preferred to assess the psychometric properties of the scale in the target language [25].…”