Fears in preschoolers: translation, reliability, and validity of the Fear Survey Schedule for Infant-Preschoolers (FSSIP) - Brazilian version

Abstract Objective To translate the Fear Survey Schedule for Infants-Preschoolers (FSSIP) into Brazilian Portuguese and to examine its reliability and validity for assessing fears among Brazilian preschoolers. Methods Two independent bilingual professionals conducted translation and back-translation of the original survey. The translated version was used to assess 152 preschool children divided in two groups: Clinical - 71 children referred for treatment for nighttime fear, and Control - 81 children enrolled at kindergarten who had not been referred for any mental health service in the previous 6 months. All parents filled out the FSSIP, the Child Behavior Checklist (CBCL/1.5-5), and a sociodemographic questionnaire. Results Cronbach’s alpha coefficients were 0.949 (95% CI: 0.94-0.96) for the entire sample; 0.948 (95% CI: 0.93-0.96) for the Clinical Group, and 0.95 (95% CI: 0.93-0.96) for the Control Group. The mean score for fears was higher in the clinical group (60.19 vs. 51.53, t = -2.056; p = 0.042), indicating acceptable discriminate validity. We also found positive, moderate, and statistically significant correlations between FSSIP and most CBCL scores, indicating good convergent validity. Conclusion The Brazilian Portuguese version of the FSSIP showed good psychometric properties, and hence may be used in research and clinical settings to evaluate fears in preschoolers.


Introduction
Fear is part of normal development and comprises a basic human emotion. It is an inborn emotional reaction produced by the perception of present or impending danger, leading to avoidance of threat, thereby having clear survival value. [1][2][3] As part of normal development, child fears have been extensively investigated. [4][5][6][7] However, for many individuals, fear is a significant problem and may impair normal development. 7,8 Research indicates associations between fears in early childhood and the development of specific phobias and other anxiety disorders in early adolescence and adulthood. 9,10 Therefore, development of instruments to assess fear in the pediatric age group is important and useful for clinical and scientific applications.
The Fear Survey Schedule for Children-Revised (FSSC-R) 11 is an 80-item self-report questionnaire frequently used to assess fears in 7-to-18-year-old children and adolescents. It allows diverse fears to be http://dx.doi.org/10.47626/2237-6089-2020-0170 investigated in boys and girls, encompassing different cultural backgrounds, and provides useful information on the relationships between fears and anxiety disorders.
There is considerable evidence of its validity in clinical and community samples from various countries. 2,[12][13][14][15][16][17] Based on the FSSC-R, 11 the Fear Survey Schedule for Infants-Preschoolers (FSSIP) was developed by Warren, Ollendick, and Simmens 18 to assess fears in children from 18 months to 5 years of age (mean = 26 months, standard deviation [SD] = 9 months, interquartile range: 23-27 months). This measure differs from the FSSC-R because it includes 29 additional items pertinent to the lives of infant, toddlers and preschoolers (e.g., "clowns," "having to go to daycare or preschool") and excludes 17 items that are less relevant for children under 5 years old (e.g., "terrorists," "taking a test").
The FSSIP is composed of 92 items referring to content of young children's fears, for which parents respond on a scale ranging from 0 (none) to 2 (a lot). The higher the score, the greater the intensity of the fears. The Fear Score (FS) is the sum of the ratings based on all items and the High Intensity Fear Score (HIFS) is the total number of fears rated "a lot." The FSSIP was tested for its reliability and validity in the United States by Warren, Ollendick, and Simmens. 18 Several methods were used to examine its convergent validity. Parents and other caregivers were interviewed and completed questionnaires concerning children's fears and child fearful behavior was directly observed in a structured play setting. Measures (Diagnostic Interview for Infants to Preschoolers for Anxiety, Infant Toddler

Social and Emotional Assessment, Child Behavior
Checklist, and Caregiver-Teacher Report Form) used to test convergent validity were significantly correlated with the FS and HIFS. The authors found both scoring methods showed good internal consistency, low-tomoderately high test-retest correlations and good convergent and discriminant validity for an Englishspeaking sample. 18 The two scoring methods correlated highly, and there was no compelling evidence that one was better than the other.
The different versions of the FSSC-R for children and adolescents 2,11 and infants and preschoolers 18 have been used in studies with international samples and are considered evidence-based instruments for identification and evaluation of children's fears. 2,[12][13][14][15][16][17]19,20 The presence of frequent and intense fears in preschool children is related to difficulties in developing autonomy, independence, and other socio-emotional skills necessary in adolescence and adulthood. In addition, if not identified and treated early, childhood fears may evolve into specific phobias and/or other anxiety disorders, conditions associated with impairments in adaptive behavior and socioemotional functioning. 8,9,21 Thus, early identification and treatment of fears in preschool children constitute low-cost interventions that may prevent mental health problems in the future.
However, to date, no instruments for assessing fears in preschoolers have been developed in Brazil and no studies reporting the translation and validation of international instruments could be found. The aim of this study was to translate the FSSIP into Brazilian Portuguese and to examine its reliability and validity for assessing fears among Brazilian preschoolers.

Methods
The study was carried out in two phases. First, translation and back-translation were performed. After that, the new Brazilian Portuguese version of the FSSIP was filled out by the parents of two groups of children: the first comprised children referred for treatment due to nighttime fears and the second was a control group (non-referred children).

Phase 1
The Brazilian version of the FSSIP was prepared in two steps. First, two bilingual professionals translated the questionnaire into Brazilian Portuguese independently.
In the second step, the two translations were randomly distributed to two bilingual professionals. Also independently, these two professionals back-translated the two Portuguese translations developed in the first step back into English. The most accurate translation was chosen by consensus, considering conceptual and semantic equivalence and cultural appropriateness.

Phase 2
Parents of children aged between 2 and 6 years were recruited from two sources to select a mix of anxious and non-anxious children: (1) a kindergarten; and (2) a psychology school clinic that offered treatment for nighttime fears. After conclusion of the translation process, the Brazilian version of the FSSIP was used in a randomized clinical trial that evaluated the effectiveness of a bibliotherapy treatment program for nighttime fears in young children 22

Sociodemographic questionnaire
Elaborated by the researchers to gather information regarding: age, gender, parental educational level, marital status, and social status of the family.

Fear Survey Schedule for Infant-Preschoolers (FSSIP)
Developed and validated in the United States, 18 this instrument is for screening and identifying the content

Phase 1
The final version of the Brazilian Portuguese FSSIP was obtained after completion of the translation and cultural adaptation procedures. It is presented in Table 1.

Phase 2
The FSSIP -Brazilian Portuguese version was used to assess two groups of children: a group referred for psychological treatment and a control group (not referred for any mental health service in the previous six-months). Overall, the sample presented a balanced proportion considering sex and age, with more girls in both groups, and a mean age of 4 years. Approximately three out of four parents were married, and most of them had at least a high school level diploma. However, it is important to note that the frequency of parents with university education was higher in the clinical group and the median family income was also higher in this group. Table 2 contains the descriptive analysis of the sample's sociodemographic characteristics.
Considering that the original FSSIP has a single factor structure, the internal consistency of the FSSIP -Brazilian Portuguese was calculated considering all items. Cronbach's alpha coefficients    To verify the FSSIP's discriminative capacity, the mean total scores obtained by the children in both groups were compared. Mean differences were significant: non-  From their parents' perspective, the Brazilian preschool children evaluated in this study had more fears than American children. 18 In the present study, we found mean total FSSIP scores that were significantly higher than the mean reported in the American study, both in the clinical and control group. The number of feared situations, objects, or animals was also much higher. One should analyze this result with caution. It may indicate that this sample of Brazilian children is actually more fearful or that Brazilian children are more exposed to situations with the potential to produce fear, such as urban violence and dangerous animals. Finally, the higher mean fears scores found among Brazilian children, regardless of whether they had been referred to mental health services, could be due to cultural differences in the interpretation of fear and its intensity.
In our study, we found moderate, positive, and statistically significant correlations between the fear scores, assessed by the FSSIP, and the CBCL emotional