Psychometric Evidence of the Online Version of the Coping Scale of Hospitalization, Illness and Treatment − Parents Version (COPHAT-P)

Hospitalizations can be stressful for children, adolescents, and their parents, thus requires assessment and proper management. This study’s objective was to provide psychometric validation of the online version of the Escala Coping da Hospitalização, Adoecimento e Tratamento − Versão para Pais (COPHAT-P) [Coping Scale of Hospitalization, Illness and Treatment – parents version (COPHAT-P)]. A total of 98 Brazilian parents/legal guardians of hospitalized or previously hospitalized children participated. Participants completed a sociodemographic questionnaire along with the COPHAT-P and the Escala de Coping da Hospitalização − versão para cuidadores (COPE-H-Cuidador) [Hospitalization Coping Scale – caregiver version (COPE-H-Caregiver)]. Confirmatory Factor Analysis did not indicate a good fit. Three factors were extracted from the Exploratory Factor Analysis, which were further supported by network analyses. Convergent validity of the online version of the COPHAT-P was supported by its associations with the COPE-H-Caregiver, except for the “maladaptive coping” dimension. The online version of the COPHAT-P had high internal consistency (α=0.94). In sum, the online version of the COPHAT-P demonstrates satisfactory validity evidence. validación psicométrica de la versión online de la Escala Coping da Hospitalização, Adoecimento e Tratamento − versão para pais (COPHAT-P) [Escala de Afrontamiento de la Hospitalización, Enfermedad y Tratamiento − versión para padres (COPHAT-P)]. Participaron 98 padres/tutores legales brasileños de niños hospitalizados o previamente hospitalizados. Los participantes completaron un cuestionario sociodemográfico junto con el COPHAT-P y la Escala de Coping da Hospitalização − versão para cuidadores [Escala de Afrontamiento de la Hospitalización − versión para cuidadores (COPE-H-Cuidador)]. El análisis factorial confirmatorio no indicó un buen ajuste del modelo. Se extrajeron tres factores del Análisis Factorial Exploratorio, que fueron confirmados por los análisis de red. La validez convergente de la versión online del COPHAT-P fue apoyada por sus asociaciones con el COPE-H- Cuidador, con la excepción de la dimensión “afrontamiento inadaptado”. La versión online del COPHAT-P tuvo una alta consistencia interna (α=0,94). En resumen, la versión online del COPHAT-P demuestra una evidencia de validez satisfactoria.


Introduction
Pediatric hospitalizations are often a source of anxiety, depression, and stress for parents (Compas et al., 2014;Doupnik et al., 2017;Craig et al., 2019;Barreto & Boeckel, 2019;Bedford & Bench, 2019). These responses are heightened when hospitalizations occur during childhood. In addition to changing the family's routine, childhood hospitalizations may also affect child development (Gomes et al., 2016). The parents' emotional health is also associated with their children's emotional well-being, and a child's recovery may be negatively or positively influenced by the parents' stress level (Compas et al., 2012). Crepaldi (2006) and Simões et al. (2020) state that an illness harms a child because, in addition to disrupting their lives in general, it possibly affects their development, formal education, as well as social and familial relations. Additionally, a child's psychological condition may be harmed due to anguish, fear, and pain elicited by the medical procedures performed during the hospitalization. Depending on how a child experiences a hospitalization, their emotional well-being may worsen due to being separated from their home and family members (Crepaldi, 2006;Freitas et al., 2021).
According to the Coping Motivational Theory (CMT), coping with stress is a regulatory response that involves behavior, emotions, and motivational orientation. The objective of the regulatory response is to deal with stressors, which can be seen as a threat or challenge imposed on three basic psychological needs -autonomy, competence, and relationship (Compas et al., 2012;Doupnik et al., 2017;Craig et al., 2019). The CMT explains that coping is a developmental process. Hence, a coping repertoire increases as individuals develop, whether through trial and error or by mimicking their peers and caregivers, adaptive and maladaptive coping strategies (Skinner & Zimmer-Gembeck, 2016). Thus, psychological instruments are needed to assess coping strategies according to age groups and specific stressors.
Identifying coping strategies used by parents/caregivers in a hospitalization context may significantly contribute to developing individualized assessments and interventions focusing on this population's main difficulties. As such, a better understanding of coping styles is essential to designing more effective interventions (Compas et al., 2017;Enumo et al., 2019). To this end, it has been proposed that observing the parents' emotional responses to a child's hospitalization may enable health workers to alleviate the stress caused by such an experience on the child by promoting adaptive coping strategies by the parents of hospitalized children (Craig et al., 2019;Barreto & Boeckel, 2019).
Several instruments address coping strategies among children within a hospital setting. In Brazil, several instruments assess coping strategies, including the Escala de Coping da Hospitalização (COPE-H) [Hospitalization Coping Scale (COPE-H)], which can be used in children and caregivers (Garioli, 2016). More recently, an instrument called the Coping da Hospitalização, Adoecimento e Tratamento [Coping Hospitalization Scale (COPHAT)] was developed to assess the overall context of treatment and illness including post-discharge among children, adolescents, and their parents and caregivers (Amaral, 2019). The latest version of the scale is the COPHAT-P, which consists of 35 items and was developed using the traditional paper-and-pencil format.
The present study aimed to validate an online version of the COPHAT-P to be used outside of a hospital setting. We hypothesize that: (i) A four-factor solution, like the original COPHAT-P, will be found for the online version; (ii) The items of the online version of the COPHAT-P will present high internal consistency, (iii) the online version of the COPHAT-P total score will be positively correlated with the COPE-H; (iv) Emotional problems scores would be directly proportional to COPHAT-P scores.

Participants
Participants for the current study consisted of parents of currently hospitalized children or children hospitalized in the past 12 months in addition to meeting the following inclusion criteria: (i) being the biological mother/father or legal guardian; (ii) being the child's caregiver (e.g., grandparents, aunt, uncle); and (iii) the child should be hospitalized (for any reason) at the time of data collection or some point in the past 12 months.
A total of 593 individuals completed the questionnaire, but only 98 met the inclusion criteria (N=98; M age =39.4; SD=5.41). All participants answered the questionnaire online, which was available for one month. The study was disseminated on different social media platforms.

Instruments
A sociodemographic questionnaire consisting of eight general questions was used to characterize the sample (i.e., sex, age, marital status, education, religion, Brazilian region of residence, number of children, and occupation -employed, unemployed, or self-employed). Four questions were used to specifically address the child's hospitalization (i.e., number of hospitalizations, indication, child's age in the last hospitalization, and length of hospitalization).
The parent's emotional problems (depression, anxiety, and stress) during the child's last hospitalization were also rated on a scale from 0 (I experienced nothing) to 5 (I experienced it very intensively). These data were used to assess the external validity (i.e., convergent validity) of the online version of the COPHAT-P. Participants were classified into three groups according to their scores on depression, anxiety, and stress: Mild (0 to 1 point); Moderate (2 to 3 points), or Severe (4 to 5 points). Significant differences were found using Analysis of Variance (ANOVA) for all the symptoms with large effect sizes: depression [(F(2;95)  Escala Coping da Hospitalização, Adoecimento e Tratamento versão para pais (COPHAT-P) [Coping Hospitalization Scale, Illness and Treatment -Parent version (COPHAT-P)]. The COPHAT-P was developed to assess the general context of therapy and disease, including post-discharge. It is a 35-item instrument rated on a Likert scale ranging from 0 (never) to 4 (always). The items are distributed into four factors: (i) Disease and treatment understanding; (ii) Hospitalization; (iii) Adherence to therapy and treatment success; and (iv) Expectations for resuming school. Thus far, only the COPHAT-P paper-and-pencil version has been used with parents whose children were hospitalized. The authors reported reliability of α=0.85 in the original paper.
Escala de Coping da Hospitalização versão para cuidadores (COPE-H-Cuidador) [Hospitalization Coping Scale -Caregiver version (COPE-H-Caregiver)] was developed by Garioli (2016) to assess the coping strategies adopted by caregivers during hospitalizations. It is a 66-item instrument rated on a Likert scale ranging from 0 (never) to 4 (always). The items measure four different categories of coping: (i) Maladaptive coping; (ii) Adaptive coping; (iii) voluntary; (iv) involuntary disengagement coping. The authors reported a satisfactory internal consistency (α=0.70) of the COPE-H in the original submission.

Procedures
This study complies with Resolutions No. 466 and 516, Brazilian National Council, the National Ethics Committee's guidelines, and the Normative Resolution PUC-Campinas No. 009/19, the guidelines for research conducted by the University's members. The Institutional Review Board approved this study (No. 3.663.091, CAAE: 23248019.1.0000.5481). The author of the COPHAT-P original version (Amaral, 2019) provided permission to conduct the present validation study.
All instruments were included in the SurveyMonkey® platform, and a pilot study was conducted to test the online version of the COPHAT-P instructions regarding its online completion and assess participants' potential concerns regarding the online version. Potential errors in the questionnaire or problems in the collection system (database) were also checked. The questionnaire was then disseminated on various social media (Instagram, WhatsApp, Facebook, Linked In) and through e-mail.

Data analysis
The continuous variables were standardized (Z-scores), and values equal to or higher than three standard deviations were excluded. Kolmogorov-Smirnov test and Levene's test were used to assess the normality and homogeneity of data, respectively. One-Way ANOVA was used to identify differences in the instrument's scores according to different sociodemographic characteristics. The effect size was calculated using Eta Squared Test (η²). This procedure was adopted as per previous studies (Cruz et al., 2018;Cunha et al., 2018;Taurisano et al., 2020).

Factorial Structure and Internal Consistency
The data presented good adequacy (KMO=0.836; Bartlett's sphericity test p<0.001). The instrument's factorial structure was initially assessed using Confirmatory Factor Analysis The CFA did not present a good fit; thus, Exploratory Factor Analysis (EFA) was performed using the Principal Axis and Varimax rotation. The COPHAT-P internal consistency was verified using Cronbach's alpha coefficient.

Network Analysis
The clustering of items was also verified using Network Analysis (NA) based on the correlation between the items. Graphs were generated using LASSO's regularization technique (Least Absolute Shrinkage and Selection Operator), estimating a network between the items based on their partial correlations. In the graphic representation models, the items (nodes) are linked through green edges (positive correlations) or red edges (negative correlations). The edges' density indicates strong correlations: thick edges indicate strong correlations, and thin edges indicate weak correlations. We conducted the NA based on previous studies (Andrade et al., 2021;Andrade et al., 2020a, Andrade et al., 2020b, Andrade et al., 2020cOliveira-Pinheiro et al., 2020).

Convergent validity
Convergent validity was examined using Spearman's correlation between the online version of the COPHAT-P factors and total score with the COPE-H three factors and total score. Sociodemographic and emotional variables were also included in the correlation matrix. Table 1 shows the mean score obtained from the online version of the COPHAT-P compared to different sociodemographic characteristics. Most participants were married women, employed, and from southeast Brazil. Additionally, most reported a religious affiliation and higher education.

Results
Regarding emotional problems, most participants reported moderate levels of stress and depression and high anxiety levels during their child's hospitalization. An ANOVA suggested differences between the score obtained in the online version of the COPHAT-P for only the emotional aspects related to their child's hospitalization. Participants reporting high levels of anxiety, depression, or stress scored significantly higher in the COPHAT-P than those reporting few symptoms (p<0.001). Regarding internal consistency (Table 2), results showed the sample's good adequacy (KMO=0.836) and a high level of reliability (α=0.94). Item 2 obtained the highest mean score (2.827), while item 16 obtained the lowest mean (0.602) among all 35 items. We detected that the CFA indicated unsatisfactory goodness of fit: CFI=0.662; TLI=0.634; RMSEA=0.111 (95%CI=0.103 -0.119). The EFA indicated a three-factor solution ( Table 3). All items presented factor loading above 0.3 and were considered in the COPHAT-P final version. Factor 1 (Hospitalization) was composed of 16 items and explained 21.7% of the variance. Factor 2 (Expectations for returning to school) was composed of 10 items and explained 14.7% of the variance. Factor 3 (Understanding the disease and treatment) was composed of 9 items and explained 10% of the variance. Thus, the final model explained 46.4% of the variance. Note: The extract method was performed using exploratory factor analysis (extraction method -Principal Axis; Rotation: Varimax). A cutoff of .30 was used to include items. The eigenvalue was 3.51 and explained 46.5% of the COPHAT-P total variance. Table 4 presents Spearman's correlations with the COPE-H and the emotional and sociodemographic variables regarding convergent validity. Only the COPE-H dimension 2 (Adaptive Coping) was not correlated with the online version of the COPHAT-P total score. Regarding emotional aspects, the online version of the COPHAT-P was significantly correlated with all variables, especially with the perception of depression (rho coefficient ρ=0.520). Length of hospitalization (in days) and the number of hospitalizations were the only two sociodemographic variables that significantly correlated with the online version of the COPHAT-P total score.  Figure 1A represents the clustering of items (nodes) based on the three factors identified in the EFA. Factor 1 (Hospitalization) presented the highest frequency of strong partial correlations among the items. Items 13-18 presented the strongest correlation in the entire network (r=0.479), followed by items 17-27 (r=0.351). Strong partial correlations were identified in Factor 2 (Expectations for returning to school) between items 30-31 (r=0.337) and 29-30 (r=0.298). Factor 3 (Understanding the disease and treatment) presented the lowest strong partial correlations between the items. Regarding centrality levels ( Figure 1B), items 5 and 25 had the highest values in the three measures considered (i.e., Betweenness, Closeness, and Degree). centrality measures. This suggests that parents become very concerned with their children's academic studies during hospitalizations. Children and adolescents are entitled to receive psycho-pedagogical support when in hospital, and the activities should be adapted to their developmental stage (Smerdel & Murgo, 2018). However, attending school also provides social-related factors beyond formal education, which may be associated with the parents' uncertainties concerning their children's education.
Additionally, the strong connection found between item 9 -"It is hard for my child to accept the changes taking place in his/her body because of the treatment." with items 8 -"My child hinders the treatment" and 24 -"My child has difficulty depending on others" is noteworthy because these items belong to different factors. The same results are also found with item 16 -"The staff hides from my child the reason for his/her treatment," which was positioned close to "Hospitalization." However, this item belongs to the "Understanding of the disease and treatment" factor. These results indicate that knowledge concerning the disease and therapy may support treatment adherence and help children accept the necessary medical procedures for their illness. A psychologist may support this process by providing health education and psychoeducation (Joo et al., 2017;Stein et al., 2019) regarding disease illness and therapy.
Assisting the parents/caregivers is also vital and can help them relate with the hospitalized child or adolescent. A systematic literature review reports that psychological interventions conducted among parents can improve psychological control, which benefits their children's hospitalization (Law et al., 2019).
Note that basic psychological needs are involved in this process. The items highlighted in the network analysis show that there was a relevant impact on autonomy and competence. When these aspects are not adequately addressed, individuals may perceive stressors as threats and adopt maladaptive coping strategies (Skinner & Zimmer-Gembeck, 2016). Therefore, assessing this process is crucial to intervene in the context of hospitalizations appropriately.
Regarding the convergent validity of the online version of the COPHAT-P, based on the COPE-H-Caregiver, the results showed the COPE-H-Caregiver Factor 2 (Maladaptive Coping) was not correlated to the online version of the COPHAT-P. This finding was expected because these scale items were designed to identify the intensity of maladaptive coping. The parents' perceptions regarding depression, length of hospitalization, and the number of times their children were hospitalized were positively correlated with the online version of the COPHAT-P. These results are in line with other studies reporting that a child's hospitalization can be a source of anxiety, depression, and stress for parents (Compas et al., 2014;Doupnik et al., 2017;Craig et al., 2019;Barreto & Boeckel, 2019;Bedford & Bench, 2019).
Studies addressing coping strategies report relevant information for psychological assessment and intervention. In methodological terms, this study advances toward what Skinner and Zimmer-Gembeck (2016) proposed to conduct studies addressing coping from a developmental perspective. In this sense, the online version of the COPHAT-P can support understanding regarding the coping process and the development of interventions to decrease maladaptive coping strategies and improve parents' quality of life facing a child's hospitalization.
This study's limitations include the fact that a convenience sample composed of a small number of participants was used, restricting the instrument's external validity. Additionally, Programa de Mestrado e Doutorado em Psicologia, UCDB -Campo Grande, MS ISSN: 2177-093X Revista Psicologia e Saúde, v. 14, n. 2, abr./jun. 2022, p. 21-37 due to a lack of valid instruments in Brazil, the online version of the COPHAT-P convergent validity was verified based on a single instrument (COPE-H). Lastly, a test-retest was not conducted to identify the temporal stability of data distribution.

Conclusion
In summary, this study's results showed that the online version of COPHAT-P demonstrated good validity indexes and is a useful tool that can be used in the future to assess the coping strategies adopted by parents facing the hospitalization of a child. It is quick and easy to use and can be used by health workers to design therapeutic interventions or preventive measures directed to this population.