Abstract
PURPOSE:
Family-centered rehabilitative care optimizes outcomes for children with significant developmental disabilities. Family-centered services involve assessing family resources that promote positive developmental outcomes for children. Little is known regarding family resources in the context of caring for a child with developmental disabilities in Brazil due to an absence of validated measures. This study describes the translation and cultural adaptation of the Family Resource Scale and explored the measurement quality of the resulting measure (the Brazilian-Family Resource Scale, or B-FRS).
METHODS:
A rigorous serial translation process that emphasized linguistic accuracy as well as cultural adaptation was utilized. The resulting 27-item B-FRS was theoretically related and reflected the contextual intent of the original measure.
RESULTS:
A four-factor scoring approach yielded acceptable internal consistency estimates for the subscales and total scale score. Overall, low levels of family resources were reported by caregivers of children with Congenital Zika Syndrome. Low family resources were associated with parental depressive and stress-related symptoms.
CONCLUSION:
Confirmatory factor analysis of the B-FRS in a larger sample is recommended. Practitioners in Brazil should broadly consider family needs and resources to provide family-centered care that is effective for the child and engages the family in a way that highlights their strengths and promotes positive developmental trajectories.
Introduction
Zika virus infection emerged as a global public health concern in 2015 and continues to have a lasting impact on those who were affected by the epidemic. Although infection was typically associated with mild illness, the 2015-2016 Zika outbreak in Brazil occurred concurrently with an unusual increase in the number of infants born with microcephaly. Exposure to Zika infection during pregnancy is associated with congenital microcephaly and a constellation of other serious birth defects affecting the central nervous system, called Congenital Zika Syndrome (CZS) [1, 2]. From 2015–2019, 3,332 confirmed cases of CZS were reported in Brazil with nearly 70% of these cases occurring in the northeast region of the country [3]. Most children with CZS have significant functional impairments and neurodevelopmental skill deficits, but variability within developmental profiles regarding the specific areas of functioning impacted and severity of impact is observed [4].
Zika virus impacts not only children’s development, but indirectly their caregivers’ mental health and the overall wellbeing of the family [5]. Accumulating evidence suggests that parents of children with CZS experience reduced quality of life, fatigue, increased stress, lower life satisfaction, and psychological difficulties [6–8]. For example, in a study based in Rio de Janeiro and Recife, mothers of children with CZS reported more stress, depressive symptoms, and anxiety compared to caregivers of children without microcephaly and developmental delays [9]. In another recent study, de Souza and colleagues assessed general mental health, positive and negative affect, fatigue, and life satisfaction in 86 parents of children ages 1–20 months with CZS (M age = 9.67 months) [6]. Findings showed that 18.6% of parents indicated a negative evaluation of their mental health, and nearly 7% indicated poor mental health and probable emotional disorder. In multivariate models, mental health was predicted by less positive and more negative affect, lower life satisfaction, and more fatigue.
Recognizing the complex needs of children and their families, published treatment guidelines specify a family-centered, individualized approach to providing services for children with CZS and highlight the necessity of identifying supports for caregivers [10, 11]. Hallmarks of family-centered care in services for children with disabilities include presenting families with developmental information, helping to facilitate adaptations in daily care and family routines, and engaging families in long-term planning to help establish and maintain positive developmental trajectories for children [12, 13]. Despite attention to the needs of vulnerable children and families in Brazil’s guiding principles for health care and established federal policies [14–17], two recent systematic reviews suggest that family-centered practices are not routinely implemented in early intervention services [18, 19].
Marini and colleagues [19] suggest that a barrier to the provision of family-centered early childhood intervention in Brazil is the paucity of scientific literature to help guide practice. Specifically, they argue that there are large conceptual and empirical gaps related to the provision of family-centered practices in Brazil that must be addressed to align service provision for children with disabilities with recommended approaches to early intervention. The ability to quantify levels of family capacity and need is foundational for practitioners to effectively provide empirically supported family-centered treatments. Similarly, an ongoing challenge for researchers seeking to address these gaps is a dearth of validated measurement tools that provide insight into the needs and experiences of families caring for children with disabilities. A critical first step in supporting children and families in the early intervention context is identifying the presence (or absence) of family resources in intrapersonal, interpersonal, and environmental domains that could help families to reduce stress and experience an increased capacity to support their children’s learning and developmental outcomes [20]. Although measures of family resources exist in the extant literature, none have been translated and validated in Brazilian Portuguese. Consequently, practitioners in Brazil do not have a comprehensive, standardized way of assessing family resources, including both family needs and strengths.
The most successful outcomes in early intervention are demonstrated by young children who receive services early in development and for a longer period, compared to older children receiving shorter amounts of services [21]. Thus, it is critical that early intervention services are engaging and impactful for families of young children with special needs, including those with CZS. Incorporating a focus on family resources is one way to promote family engagement that, in turn, increases achievement of treatment goals. Resource-based interventions involve identifying a family’s unique needs and using this information to help them acquire useful resources that promote both child and family wellbeing [22]. A first step in implementing tailored resource-based interventions for children with disabilities including CZS involves gathering accurate knowledge regarding the availability and adequacy of family resources. Although anecdotal evidence from Brazilian health care professionals suggests that inadequate family resources are a barrier to effective treatment, the availability of resources for families of children with CZS in Brazil has not been described adequately in part due to the lack of validated measures that are translated into Brazilian Portuguese and culturally adapted for this population [23].
Current study
The purpose of the present study was to advance the provision of family-centered early childhood intervention in Brazil by translating and providing preliminary validation of a questionnaire to help guide family-centered care-oriented professionals in developing effective family resource-based interventions. Simply translating an existing measure into a different language does not ensure cultural appropriateness or that the translated measure retains the same psychometric characteristics as the original version [24, 25]. This is because a measure that has been validated in a certain context is bound to the cultural nuances of the sample, as well as time. Additionally, there could be varying dialects of the translated language that could lead to errors in the transformations of the words that alter the semantics [26]. For example, Borsa and colleagues [27] argue that a translation into Brazilian Portuguese focused merely on linguistic equivalency may result in an overly complex translation that the target population would not understand, or an overly simplistic translation in which the content in the instrument is too condensed. Thus, a cultural adaptation process was engaged that, in addition to translation, included outside reviewers and an iterative process with members of the target population to ensure the validity of the translated measure. After the culturally-grounded translation process was completed, resulting data were tested through psychometric evaluation to further establish validity.
Methods and results
Questionnaire identification
The Family Resource Scale (FRS) is an English-language self-report rating scale that measures resources available to families with young children [28]. The FRS reflects a social systems perspective with items measuring both individual and contextual aspects of family functioning. The original 30-item measure was validated in a US sample of 45 mothers of preschool-aged children with developmental or cognitive delays ranging from low to middle socioeconomic status and was found to have six factors: Growth and Support, Necessities and Health, Physical Necessities and Shelter, Intrafamily Support, Childcare, and Personal Resources [28]. The authors reported internal consistency for the total score of 0.92, but internal consistency estimates were not reported for the subscales.
Subsequent efforts to establish the measurement quality of the FRS have produced varying results in different samples, with debate surrounding factor structure and the appropriate number of subscales. In a sample of 162 Australian families of children with behavioral or developmental concerns, factor analysis of the FRS suggested a three-factor model (Basic Needs, Additional Financial Needs, and Time for Self or Family) [29]. More recently, in a sample of 300 families with concerns about child behavioral problems, a four-factor structure (Basic Needs, Essential Care, Time for Family, and Extra Money and Time for Yourself) was found to provide the best fit [30]. Samples of primarily low-income families have shown support for both a shortened 20-item four-factor structure (Basic Needs, Money, Time for Self, and Time for Family) [31] and a 30-item six-factor structure (Basic Needs, Housing and Utilities, Benefits, Social Needs/Self-care, Child Care, and Extra Resources) [32].
To the authors’ knowledge, only one study has translated and culturally adapted the FRS for a non-English speaking population. The Arabic-Family Resource Scale (A-FRS) underwent forward translation, expert panel back-translation, pre-testing, and cognitive interviewing before being administered to 115 rural and urban Jordanian families with children diagnosed with or at risk for cerebral palsy and receiving rehabilitation services [33]. Nearly half of this sample was low-income with a mean child age of 4.6 years. All 30 items were retained in a six-factor model (Physical and Health Necessities, Intra-family Support, Family Entertainment, Personal Support, Basic Necessities, and Childcare). Given mixed findings in studies exploring the psychometric qualities of the FRS, researchers are cautioned that the scale should be tested for validity and reliability when used with other populations [30].
Translation and cultural adaptation
The methodology used to guide the translation and cultural adaptation process was patterned after a serial approach to translation [34] and was further informed by Brazilian specific methodology [27] to increase cultural utility of the finished assessment. There were six steps in this process: 1) individual translation, 2) translations merged by a committee who discussed until consensus was reached for the best translation, 3) assessment for clarity and understanding in target language, 4) linguistic equivalence determination, 5) instrument field testing, and 6) psychometric analysis. During stages one through three, the focus was not solely on language determination but more importantly on cultural adaptation, with the goal of ensuring that the construct being evaluated mirrored the intended construct of the original instrument. The overarching objective was to create an instrument useful to practitioners and researchers for evaluating family resources in Brazil.
The translation team consisted of four primary translators. Three of the four individuals were native Portuguese speakers, and one was a native English speaker. All were fluent in both English and Portuguese, and all had spent time in both the United States and Brazil. Three translators held doctoral degrees in mental health and/or education-related professional fields, and the third had a master’s degree in neuropsychology and was a doctoral student. First, each member of the team worked independently to create what they believed to be the most accurate Brazilian Portuguese translation of the FRS, hereafter called the B-FRS, paying particular attention to both language and Brazilian culture. Second, they met as a group to compare translations and collectively develop what they agreed was the most accurate and culturally appropriate version of each item in the instrument.
Discrepancies in the individual translations reflected minor variations in wording and were resolved through discussion until group consensus was reached. All 30 items from the original FRS were retained. During stage three, two native Portuguese speakers who had not been involved in the translation nor were familiar with the English version of the FRS independently reviewed the translated measure for overall clarity and their perception of cultural appropriateness. The first reviewer, a colleague of one of the Brazilian research team members, reviewed the translated measure with respect to face validity. That reviewer had experience with early childhood as well as CZS and advised minor changes to the questionnaire which the research team reviewed and agreed upon. After these changes were made, a second native Portuguese speaker was asked to inspect the revised measure for grammatical and language purposes. Again, minor changes were suggested to make the instrument more accessible to a wider range of education levels which the research team then reviewed together and adopted, establishing the first draft of the B-FRS.
Linguistic equivalency
Three bilingual doctoral students who were enrolled at the principal investigator’s home university in the United States but who were not studying in the social sciences nor previously involved in the study translated the first draft of the B-FRS back into English. One of these students was a native English speaker and two were native Portuguese speakers. Comparison of the original FRS and the back-translated versions produced by the three students showed consistency in item meaning and only minor differences in wording. This outcome demonstrated accuracy between content in both versions and indicated that no additional changes were needed to the translated questionnaire. This version was then used for field testing (described below).
Field testing
The purpose of field testing was to administer the instrument to the target population and collect data to allow for evaluation of the psychometric qualities of the B-FRS. The sample for field testing of the B-FRS included 50 caregivers of young children with CZS who were receiving medical care at the Instituto de Medicina Integral Professor Fernando Figueira (IMIP) and participating in a larger study exploring the impact of CZS on families. IMIP is located in Recife, the capital of Pernambuco state, a city that experienced a large volume of cases associated with the Zika outbreak. IMIP is part of the Sistema Único de Saúde and serves low-income individuals and families. Institutional Review Board approval was obtained both from the United States-based university overseeing the research and the IMIP hospital Ethics Committee.
Eligibility criteria for the larger study specified that participants were the primary caregiver of a child between the ages of birth to three years who had been diagnosed with CZS and had been attending weekly rehabilitation services at the hospital. The study was open to all caregivers regardless of race/ethnicity, gender, and age. There were no language proficiency criteria to participate in the study. If a participant could not read, they were provided the option of having the research assistant read the items to them. All participants opted to read the items themselves. Among the 50 participants, the majority (92%) were mothers and their mean age was 31.1 years (Standard Deviation [SD] = 9.0). The sample included one father and three grandparents. The children were between seven and 37 months of age (M = 25.9, SD = 6.2) and 66% were female. Participants reported their race/ethnicity as follows: White (28%), Black (14%), Indigenous (4%), and Mixed race/ethnicity (54%). All participants reported family income between one to three minimum Brazilian salaries (each minimum salary is equivalent to approximately $300 USD per month). Most (70%) had at least one other child within four years of age of the child with CZS.
Graduate research assistants approached and recruited caregivers for the study during their child’s weekly therapy appointments at the hospital. If a participant was eligible and expressed interest in participating, the research assistant obtained written informed consent after a discussion of the anticipated risks and potential benefits of study participation, as well as the voluntary nature of research participation. Enrolled participants then had the option to complete the study during the current visit and/or the following week during their child’s appointment. Research assistants administered questionnaires that included the B-FRS and took approximately 60–90 minutes to complete. If a participant was not able to complete the study in the time they had available that day, they were permitted to finish during the next week when they returned to the hospital for subsequent appointments. Responses were recorded on iPads with Qualtrics for data management and confidentiality. Data were collected between January 2018 and April 2018.
In lieu of a direct honorarium to participants, which was not permitted by hospital policy, a monetary donation was made to the clinic where participants’ children were receiving care.
Psychometric testing
Item analysis
Evaluation of the psychometric qualities of the B-FRS utilized data collected during field testing. The performance of each item was examined to determine the need to eliminate any items from the 30-item translated version. First, examination of the frequencies of each item revealed that there was adequate distribution of responses across the scale choices, indicating variability in the sample. However, on three items, most respondents selected “Not Applicable,” suggesting that these items lacked content validity in the target population. For these items, the percent selecting “Not Applicable” was as follows: Heat for house/apartment = 76%, Childcare/daycare = 96%, and Money for special equipment = 84%. These three items were dropped in all subsequent analyses of the B-FRS.
Means, SDs, and ranges were calculated for each of the remaining 27 items (Table 1). The range of scores for 25 items varied from 1 to 5. Responses for Question 19 ranged from 2 to 5, and responses for Question 15 ranged from 1 to 4. Responses on most of the items (48.1%) indicated that respondents’ access to resources was inadequate as indicated by their ratings of “seldom adequate” to “not at all adequate.” The lowest average rating was for “family to be together” (Q15), and the highest average rating was for “food for two meals a day” (Q1). Only 14.8% of the item means had optimal ratings in terms of adequacy of resources.
Descriptive statistics for Family Resource Scale (FRS) items
Descriptive statistics for Family Resource Scale (FRS) items
The original FRS was designed to yield a total scale score as well as six subscales. For the current study, a total score for the B-FRS was calculated by taking the mean of all items except the three that were dropped because they were not applicable in this sample (Q5, Q20, and Q21). A total scale score was calculated for participants who were missing less than 20% of the remaining items. In the current sample, the B-FRS mean total score was 3.18 (SD = 0.81). 1 The Cronbach’s alpha coefficient, which is a proxy for the degree to which a set of items measures a single unidimensional latent construct [35], was used to calculate an internal reliability estimate for the total scale. Results revealed excellent internal consistency for the B-FRS total scale score (α= 0.93).
Subscale scores for the B-FRS were calculated following the published scoring conventions for the original FRS and two other studies that described their scoring protocols and reliability estimates at the subscale level [30, 31]. Cronbach’s alpha was calculated for each subscale, and the obtained internal consistency estimates were compared with those reported in prior studies. Results are summarized in Table 2. The original FRS scoring protocol, which included six subscales, had acceptable alphas for the subscales of Growth and Support and Necessities and Health (α= 0.85 and 0.88); however, the Physical Necessities and Shelter, Intrafamily Support, and Personal Resources subscales had low internal consistency estimates (α= 0.08 to 0.67). Additionally, a score could not be calculated for the Childcare subscale because the items loading on the scale in the original FRS were not included in the 27-item B-FRS. Van Horn and colleagues’ (2001) scoring protocol for the Family Resources Scale-Revised also did not appear to provide an ideal scoring approach for the B-FRS. In this case, only two of the four subscales (Basic Needs and Money) had acceptable internal consistency estimates.
Alphas for subscale protocols in a Brazilian Portuguese sample
Alphas for subscale protocols in a Brazilian Portuguese sample
Note. aSubscale items were dropped in development of Brazilian-FRS. FRS = Family Resources Scale.
Patwardhan et al.’s [30] four-factor structure provided a reasonable approach for scoring the B-FRS. In this approach, subscale scores are created by taking the mean of the items with the caveat that no more than 20% to 25% of the items should be missing for that subscale for scales with three or more items. Among the four subscales yielded with this scoring approach, the Time for Family subscale had the only unacceptable reliability estimate (α= 0.45). However, examination of the correlations between the two items comprising this subscale, “Time for family to be together” (Q15) and “Time to be with child” (Q16), indicated that these items were significantly related r = 0.33 (p < 0.05). As such, the overall low reliability for the scale was likely due to having only two items rather than a true lack of association between the constituent items that comprised the subscale. This scoring approach was adopted for all subsequent analyses. Means and SDs for the B-FRS subscales are presented in Table 2.
Criterion validity for the B-FRS was assessed by examining the Pearson correlations between the B-FRS subscale scores and the total score, as well as whether the B-FRS total and four subscale scores were associated with other relevant variables in the assessment battery in the expected directions, including parental symptoms of depression and anxiety, parenting stress, and parental coping strategy use. Measurement of these variables was obtained as follows.
Parental depressive symptoms were assessed using the Brazilian Portuguese version of the Beck Depression Inventory-II (BDI-II), a 21-item inventory that measures symptoms of depression [36]. Respondents rate how much they are bothered by each symptom using a 0–3 rating scale, yielding total scores ranging from 0 to 63. The BDI-II has high internal consistency and factorial validity. Cronbach’s alpha was 0.89.
Symptoms of parental anxiety were measured using the Brazilian Portuguese version of the Beck Anxiety Inventory [37, 38]. Caregivers rated how much they were bothered by 21 symptoms over the past week using a four-point scale ranging from 0 to 3. Items were summed to obtain a total score ranging from 0 to 63. The Beck Anxiety Inventory has high internal consistency, item-total correlations from 0.30 to 0.71 (median = 0.60), and demonstrated high test-retest reliability and convergent validity with other scales. Cronbach’s alpha in the current sample was 0.90.
Parenting stress was assessed using the Parenting Stress Index, Fourth Edition-Short Form (PSI-4-SF) [39]. Parents responded to 36 statements using a five-point scale (1 = “I strongly agree” to 5 = “I strongly disagree”). Items are combined to yield a Total Stress score. The PSI-4-SF provides t scores and percentile scores as normative metrics. The normal range for scores is within the 16th and 84th percentiles. Scores within the 85th and 89th percentiles are considered high, and scores above the 90th percentile are considered clinically significant. The present study utilized the European Portuguese version of the PSI available from the publisher. As was done previously, some expressions were culturally adapted for use with a Brazilian Portuguese sample [40]. Cronbach’s alpha for the Total Stress score was 0.89.
The 30-item Family Crisis Oriented Personal Scales (FCOPES) assesses problem-solving, coping attitudes, and behavioral strategies used by families of children with disabilities in difficult situations [41]. Items are measured on a five-point scale ranging from “1 = strongly disagree” to “5 = strongly agree.” A total Coping score is obtained by summing responses for each item, after reverse-scoring four items. A higher Coping score indicates more use of coping strategies. The FCOPES has good internal reliability (α= 0.87). Cronbach’s alpha in this sample was 0.84.
After screening for outliers on all variables following established procedures [42], values for skewness and kurtosis as well as tests for normality were examined to ensure that items adhered to a normal distribution. Finding that all variables were normally distributed, Pearson correlations between B-FRS total score and subscale scores and the continuous variables were then calculated. Correlations between each of the four subscales and the total score of the B-FRS items were as follows: Basic Needs (r = 0.95, p < 0.001), Extra Money and Time (r = 0.95, p < 0.001), Time for Family (r = 0.57, p < 0.001), and Essential Care (r = 0.76, p < 0.001). Table 4 shows the correlations between parent psychosocial variables and B-FRS subscales and total score. The B-FRS total score was significantly related (p < 0.001) to all variables in expected ways: higher resources were associated with lower depression (r = –0.60) and anxiety scores (r = –0.45), lower parenting stress (r = –0.49), and higher coping strategy use (r = 0.54).
Discussion
The absence of a translated and validated instrument for measuring family resources in the context of early childhood intervention in Brazil is a significant barrier for research and effective service provision. The purpose of this study was to translate and provide initial validation of the FRS to create a Brazilian Portuguese version of the measure that has both empirical integrity and cultural relevance for Brazilian families caring for children with special health care and developmental needs. A rigorous translation process that emphasized linguistics as well as cultural appropriateness was utilized following published guidelines for the cross-cultural translation of questionnaires [26, 43]. Face validity, content validity, internal consistency reliability, and construct validity were considered throughout the translation and evaluation of the questionnaire.
This process resulted in the B-FRS, which was determined to be both theoretically related and reflective of the contextual intent of the developers of the FRS [28] and was deemed appropriate for use with a Brazilian Portuguese-speaking population. The B-FRS includes 27 of the 30 items that comprised the original FRS. Dropped items asked respondents to rate the adequacy of the following resources: 1) Heat for your house/apartment, 2) Childcare/day care for your child(ren), and 3) Money to buy special equipment/supplies for your child. Most participants selected the “not applicable” response option for these resources, suggesting that the items were confusing or had low content validity for the target population. Question 1 (heat) may have been deemed not applicable by the participants due to the tropical climate in which they live. Heat may rarely be needed or simply not viewed as a resource by Brazilians living in Recife because it is a humid, tropical coastal city [44].
The perceived non-applicability of childcare availability for caregivers in this sample may reflect that care for very young children with disabilities is often mother-centered [45]. Previous research on families and mothers of children with disabilities has found that many mothers in varying cultural contexts end or interrupt their career to take care of their children [46, 47]. This may be especially true in the Brazilian context, as evidenced in phenomenological research with Brazilian mothers of children with disabilities [48]. Rodrigues et al.’s [49] qualitative investigation of Brazilian caregivers of children with intellectual disabilities found a common theme among the mothers interviewed in that they recognized the child’s care as “complex and difficult to share with other people” (p. 418). Brazilian caregivers may feel other people, including childcare providers, are not as equipped to meet their child’s needs, especially given the young age of the children. Additionally, Rodrigues also found that mothers in their sample relied on sisters, parents, or cousins to help with their child’s care sporadically, so that formal, consistent childcare or daycare was not seen as necessary [49]. Since the caregivers in this sample reported having optimal adequacy regarding access to babysitting, it may be that caregivers were not relying on formalized, paid childcare and instead relying on family members and friends for more informal sources of help with their child as needed.
Money to buy special equipment or supplies for the child may not have been relevant in this context because the children were too young to need any specialized equipment. It is possible that caregivers could find specialized equipment like wheelchairs or glasses more important as their child ages. Alternatively, caregivers may not have been aware of any additional supplies or equipment their child might need at present or in the future. Other published studies reveal that parents of young children with CZS receive little guidance, support, and education related to caring for their child, so it is possible they did not find this questionnaire item relevant to their experiences [50, 51]. Some caregivers have reported that health care workers and providers have sometimes been reluctant to help them with their child, so it is also possible that the caregivers were not made aware of any additional services from which their child could benefit [49, 52]. Additionally, the hospitals where the children were receiving services may not have been equipped with specialized materials themselves, so they may not have had any to give. Overall, however, examination of the remaining 27 items suggested that the B-FRS was sensitive to detecting differences among families in the adequacy of various resources, evidenced by ratings for most individual items spanning the possible range of response options.
Questionnaires that are adapted for use with samples that are culturally, linguistically, or otherwise different from the sample used in instrument development often do not retain the same psychometric characteristics as the original versions [25]. With respect to the FRS, mixed findings have been reported in every study exploring the psychometric qualities of the questionnaire, leading researchers to caution that the scale should be tested for validity and reliability when used in other settings and with other populations [30]. The small sample size used in field testing prohibited factor analysis of the B-FRS. To identify an appropriate strategy for scoring the B-FRS, the completed instrument was scored following each of the other published scoring conventions [28–33], and the obtained internal consistency estimates for the total scale and subscales were compared with those reported in prior studies.
The results revealed that the scoring approach identified by Patwardhan and colleagues [30] provided the best fit for the B-FRS. The aforementioned study was conducted in a sample of US families seeking assistance for their children with behavioral difficulties. Using exploratory factor analysis, the authors determined that a four-factor solution provided the best fit for their data, although this solution did not replicate any of the existing factor solutions found in prior psychometric investigations of the FRS. The same four-factor scoring approach was adopted for the current sample because it yielded internal consistency estimates for the four subscales that were the most acceptable, compared with other published scoring approaches. Moreover, the scoring rubric utilized in this approach [30] was the most robust in terms of dealing with missing data and using a mean scoring approach versus a sum score approach.
Comparison of the means and SDs for the four subscales in this sample and the Patwardhan sample revealed that the current sample reported a lower average rating of resource availability for three of the four subscales as follows: Basic Needs, 3.42 versus 4.45; Time for Family, 2.72 versus 3.93; and Essential Care, 2.85 versus 4.01. These results indicated that family resources were much lower for Brazilian families compared with Patwardhan and colleagues’ American sample and were inadequate in this sample of families of children with CZS. This may have important implications for the effectiveness of interventions and long-term developmental outcomes for children with CZS in Brazil. Consistent with a hierarchy needs perspective, unmet basic needs take precedent in terms of influencing and directing human behavior [53, 54]. Consequently, families of children who report low basic needs may be more likely to devote their time and energy to trying to get these critical family needs met, versus spending time addressing children’s therapeutic needs, which may be perceived as an important but less immediate need for the family [28]. Further, to the extent that a family is expending time and energy into getting basic needs met, adding additional demands such as professionally-prescribed treatments may have the unintended consequence of increasing family stress, which in turn may worsen both child and parent health and wellbeing.
Bivariate associations of Brazilian-Family Resources Scale (B-FRS) subscales and total score with parent psychosocial variables
Bivariate associations of Brazilian-Family Resources Scale (B-FRS) subscales and total score with parent psychosocial variables
Note. *p < 0.05, **p < 0.01, ***p < 0.001.
The findings revealed significant associations between low family resources and indices of parental psychosocial wellbeing. Specifically, the strongest relation was observed between basic needs and depressive symptoms. Considering that parents and caregivers of children with developmental delays and disabilities have been shown to be at an elevated risk for symptoms of stress and depression [55, 56], it was not surprising to see evidence of this in the sample. Having lower resources to meet the family’s basic needs was associated with higher depressive symptoms. Lower basic needs were also associated with lower levels of coping strategy usage. Since healthy coping strategy use in caregivers of children with disabilities has been shown to mitigate stress, families that have sufficient resources to meet their most basic needs may be buffered from experiencing negative mental health outcomes [57]. It may be helpful for service providers to consider whether a family is able to meet its most basic needs and, if not, address this in the child’s intervention or therapy. Ensuring that families have their basic needs met may improve the child’s therapy outcomes if it reduces caregiver stress and risk for mental health problems. Another strong, negative relationship was observed between extra time for family and depression and stress. This finding is consistent with a cross-sectional survey of mothers of children with disabilities in the US, in which perceived social support was found to be a protective factor between child behavior and maternal depression [58]. Family-centered interventions and therapies that build services into the family’s everyday lives, thereby giving the caregivers more time to spend with their family and less time isolated or traveling with their child, may be beneficial.
Overall, these results were broadly consistent with the social determinants theoretical framework which focuses on understanding how the circumstances in which people live shape their mental health [59]. This framework hypothesizes that populations with economic and social vulnerabilities (e.g., low socioeconomic status, unreliable employment, inadequate social support) are disproportionately affected by poor mental health and that cumulative stress and physical health are important factors through which the adverse impacts of social determinants intensify [60]. The construct of social determinants in relation to Brazilian caregivers’ mental health may encompass a wide range of factors, including societal influences such as economic opportunities, individual characteristics such as age and gender, and malleable factors such as educational attainment, financial strain, living conditions, health behaviors, and child disability and/or health-related needs [61].
Successfully addressing family resources and the social determinants that impact them in the context of early intervention will require changes to the ways that early intervention services in Brazil are conceptualized and delivered, as well as increased federal funding that can help to support programs seeking to align their service approach with best practices in family-centered early intervention. At present, children and families receive early intervention services primarily through the Sistema Único de Saúde (SUS), Brazil’s government-run public health care system. The SUS is the largest public health care system in the world and provides health care services free of charge to nearly 100% of the Brazilian population. However, despite clear recognition of the importance of early intervention for children with special needs in both the SUS principles and Brazil’s established federal health care policies (e.g., National Policy for Comprehensive Child Health Care, the National Health Policy for Persons with Disabilities and the Disability Health Care Network, and the National Primary Care Policy), existing laws in Brazil do not translate into all children in need receiving early intervention services because these policies are not adequately supported by federal funds. Thus, many children cannot access and benefit from intervention services. Moreover, there are indications that critical family resource needs are not being addressed, even among children and families who do receive services. Specifically, two recent systematic reviews suggest that family-centered care principles, which acknowledge the key role of family resources, are not routinely implemented in the early intervention approaches and specific practices used with young children who receive services [18, 19]. This may be because early intervention services are embedded within the SUS health sector and therefore guided by a medically-oriented rehabilitative care model. Intervention practices are aimed at stimulating skill development in specific areas of identified developmental deviation (i.e., motor, sensory, proprioceptive, speech-language, and social-emotional skills), with service providers including physical therapists, speech therapists, and psychologists. Ecological approaches that extend assessment and intervention into family and community contexts and consider influences such as the adequacy of family resources are not currently standard care in early intervention in Brazil. Rather, family support for children with CZS has been provided primarily via community-based programs that are not integrated into more formal government-funded early intervention services or through grassroots informal support networks that vary in structure and aims [18, 63].
The findings of this study should be considered in light of several methodological limitations, particularly in terms of generalizing the findings and applying the results to clinical practice for families caring for children with special needs. First, the sample may not be representative of all of Brazil. The study participants were receiving care at a hospital in Recife an urban city located in the northeast region of Brazil. Families who reside in rural regions or the southern part of the country may have different resources and needs. Testing the B-FRS in other parts of the country is needed to further verify the cultural appropriateness of the translated measure. Additionally, this study was conducted with very young children with CZS. These results may not generalize to older children in this context as caregivers may prioritize different resources and needs as their child continues to develop. Second, the participants were caring for babies and toddlers with CZS and thus reported on resources that were available for this age group with a particular condition. Families caring for older children, or children with other types of special needs, may require different resources or have access to a range of other services that support their wellbeing. Additional research with the B-FRS is needed to delineate family resources in more diverse samples, including older children, families of children with other health or developmental problems, and those residing in different regions of Brazil. Research with fathers and other non-maternal caregivers is also needed, as the sample was heavily biased towards mothers as they were typically the primary caregiver present at the hospital when recruitment took place. Another limitation is that this was a cross-sectional study that relied on self-report data, which is subject to self-report bias. A fourth limitation of this study is the small sample size. With only 50 participants, it was not possible to conduct a factor analysis to confirm the factor structure of the B-FRS. Moreover, a small sample size can limit the ability to detect significant associations among study variables. Finally, a general limitation of the B-FRS is that some constructs that may influence family resources and wellbeing among families, such as religious spiritual needs, adult educational needs, and financial planning or legal needs, were not included in the original FRS. Large scale future validation efforts of the B-FRS with more diverse samples should incorporate additional items addressing these needs, as well as include the three items that were eliminated for further item analysis, to yield the most psychometrically and culturally sound instrument possible.
Conclusions
This study has important implications for researchers and practitioners focusing on children with special health needs and their families. Having a reliable and valid measure of family resources can help researchers to more rigorously investigate how family resources influence service utilization patterns in the field of early intervention, modifiable factors related to treatment adherence, and the links between family resources and children’s ultimate developmental outcomes [32]. For early intervention service providers, the B-FRS appears to be a promising tool for assessing the adequacy of resources in families of children with complex developmental needs in Brazil. As an assessment measure, the B-FRS can be used to better understand the adequacy of a wide range of resources that are relevant to intervention planning and resources-based practices. For example, interventions may focus on helping families acquire resources to satisfy critical basic needs to improve intervention effectiveness. This is a crucial first step in exploring how family needs and resources can shape early intervention in a Brazilian context, promoting positive developmental trajectories for the child and family.
Footnotes
Acknowledgments
The authors have no acknowledgements.
Conflicts of interest
The authors have no conflicts of interest to report.
Ethical considerations
This study was reviewed and has been continuously approved by the Institutional Review Board located at the United States-based university where the American co-authors are employed. The approval date was 16 January 2018 and the approval number is 2018017157EP.
In addition, the Ethics Committee of the Brazilian rehabilitation hospital where the study participants were recruited and the data were collected reviewed and approved the research. The approval date was 13 December 2017 and the approval number is 2.436.221.
Informed consent was obtained for all human subjects and is appropriately described in the manuscript (p. 10).
Funding
The authors have no financial interests and did not receive any benefits from the direct applications of this research. Funding for this project was provided by the University of Nebraska-Lincoln and the Fundação Maria Cecilia Souto Vidigal (FMCSV) through its collaborative Pilot Impact Program. Opinions expressed herein are those of the authors and do not reflect the position of the University of Nebraska or FMCSV.
Direct comparisons with the original 30-item FRS were not made because the measure development paper provides only a total sum score for the scale rather than reporting the means of all items and also did not account for missing item responses.