This study showed that the associated factors of family impact in children with EA are the child’s level of generic HRQOL, school absence, tracheomalacia, as well as the parental use of carer’s allowance and parent educational level. Furthermore, the study findings suggest that an increased number of feeding, digestive and respiratory problems in children with EA are negatively associated with family impact. Hence, this illustrates multifactorial explanations of the family impact that expand beyond a medical-surgical model.
In the present study, parents’ ratings of their EA child’s generic HRQOL were highly associated with the level of family impact. Already in the univariable analysis, the child’s generic HRQOL explained 50% of the total family impact scores. Following stepwise regression, which is a method of regressing multiple variables and simultaneously removing the weakest correlated variable, parents’ ratings of their child’s generic HRQOL remained a part of the multifactorial model to explain the total family impact scores. The PedsQLTM Family Impact Module Total Scale Score is built up by two subscales measuring parents’ HRQOL and family functioning, with most items targeting parents’ HRQOL. Therefore, the results could mean that children’s and their parents’ HRQOL are interlinked, as reported by parents. In the PedsQLTM 4.0 generic core scales, the parent-proxy should rate how he/she believes the child experiences their HRQOL [29]. Parents of children with chronic conditions tend to rate their child’s generic HRQOL lower than their children [32], including parents of children with EA[33, 34]. However, previous studies have also shown that the child-parent agreement in ratings of the EA child’s HRQOL is mainly good and any discrepancy is not explained by the level of family functioning[34].
Moreover, we found that EA children’s school situation impacted the family. This finding agrees with a small German study, which showed that frequent school absence was associated with impaired mental HRQOL in their parents[23]. In pediatric cyclic vomiting syndrome, the authors also made similar observations. They suggested that children missing school create parental worry, parents missing work, decreased family income, and limits parents’ capacity to attend to other responsibilities[35]. In a Swedish study, school absence ≥1/month in children with EA was associated with the child’s use of school support, including educational support, support with nutritional intake, or both[36]. Educational support may be provided to children with EA with emotional and behavioral problems, a group of children previously shown to be at risk for having worse family functioning[25]. Although regulations of children’s right to school support vary between Sweden and Germany, it may imply underlying health care needs of the children, which could explain our findings.
This study showed that the independent clinical factor in the multifactorial model explaining family impact was the presence of severe tracheomalacia in children with EA. The relation between severe tracheomalacia (verified through bronchoscopy) and family impact is a new finding to the authors’ knowledge. In our study sample, bronchoscopy was performed preoperatively in theatre or follow-up care of children with severe airway symptomatology. As detailed in Table 1, we standardized criteria to define severe tracheomalacia[37], with the most recent bronchoscopy valid for inclusion. EA children with severe tracheomalacia are a disease subgroup with a risk for respiratory and gastroesophageal dysmotility problems[8]. Our subsequent analysis showed that several feeding and airway problems were significantly more prevalent in EA children with severe tracheomalacia than those without. Similarly, this group used more antireflux medication, bronchodilators or inhaled steroids. Other clinical variables indicating disease severity, such as associated anomalies, premature birth, and a child with a history of gastrostomy feeding, were significantly associated with lower family impact scores in the univariable analysis. However, after the multivariable regression analysis, these clinical variables were not retained as a part of the multifactorial model. Previous research has shown that associated anomalies in children with EA negatively impact family functioning[24] while parents of children with VACTERL association (i.e., with at least three of the following: vertebral defects, anal atresia, cardiac defects, tracheoesophageal fistula, renal anomalies, limb abnormalities) self-report levels of anxiety and depression comparable to non-clinical samples [38].
In this study, 31% of parents received carer’s allowance, which corresponded to poor family functioning, probably because carer’s allowance is provided to parents of children with severe healthcare needs. Furthermore, 56% of parents had no college/university degree, a factor that is consistent with previous studies in corresponding to lower parental HRQOL[39]. In contrast to other studies of parents’ mental health and HRQOL[14], including studies of rare diseases [23, 40], being a mother was not a risk factor for negative impact on the family, but similar to previous studies[41], mothers represented the majority of proxy-reports of children. Perhaps the caregiver burden and limitation in career development add to the explanation of a participant’s perceived family impact. Moreover, child age was significant in the univariable regression analysis but was not an independent factor of family impact in children with EA. In previous studies, the parents’ psychological burden is less in older children with EA than younger children[19, 23, 26], which is reasonable as the child’s health may improve with increased age[42]. A reason for our results may be the broader set of variables included in the multivariable stepwise regression analysis.
Lastly, we investigated the impact of the child’s symptoms on the family. While feeding disorders have been shown to negatively impact families of children with EA[24], this study adds that the child’s feeding and digestive problems are explained substantially more in the variation of the total family impact scores than the child’s respiratory problems. This is congruent with factors influencing EA children’s condition-specific HRQOL[28]. Healthy newborn infants learn to suck competently within days after birth and develop a feeding-sleeping cycle in the context of a caregiver relationship[43]. Following the repair of EA, children may need a prolonged time to establish peroral feeding compared to healthy infants, with a need for gavage feeding or even gastrostomy feeding for a time. In a previous study of maternal-infant social interaction, the most significant area of concern for mothers of children with EA was during feedings [44]. The child’s experiences of vomiting, choking, and food impaction may interact with the child’s ability and willingness to eat. From a parent’s perspective, the frequent daily attempts at trying to feed a child with feeding difficulties can be tiring and impact family mealtimes[43]. This context could help to explain our study results.
Implications of the study findings
Knowledge of risk factors for poor family functioning in children with EA provides essential information for health care professionals and patient advocacy groups encountering these families in order to understand which families may need targeted support. Some families of children with EA can adjust and develop a fulfilling life after encountering a medically challenging life event. Others are likely to develop worry, stress, or depressive symptoms in the long term[22, 45]. Our study findings would stress the importance of holistic caring support to optimize EA child and family health, especially since it has been shown that there is ample room for improvement to facilitate the pathways to psychosocial care for children with rare diseases and their families[46]. First, since the level of generic HRQOL in children with EA was related to family impact, communication of the child's HRQOL in a clinician-family encounter during follow-up care with subsequent efforts to strengthen the children's HRQOL could be of importance to the child[47], but also to the whole family. Then, because a higher symptom burden in children with EA was associated with worse family impact, this emphasizes the need for multidisciplinary follow-up care managed by doctors, pediatric nurses, dieticians, speech therapists, psychologists, and social workers. This team can help monitor and treat the child's symptoms properly, but also provide practical and educational support to families of children with EA and facilitate a good adaptational process. This would be in line with patient-driven health-care recommendations for adults with EA and their families[48]. Moreover, since our study results imply that family impact in children with digestive symptoms and feeding difficulties is high (i.e worse), these families may benefit from having targeted support early after the child is born with EA. In this perspective it has been pointed out that early intervention on feeding issues in infancy may reduce later problems[49], and that mothers of children with EA may mourn the loss of their initial fundamental role as feeders of their child[43]. A possible support to these families may therefore be a health care professional and/or a patient advocacy group to listen to the parents, allow them to verbalize their feelings about feedings of their child and reinforce adaptation to some of their caregiver roles.
Study strengths and weaknesses
Although EA is rare, numerous children worldwide are affected, and specialized knowledge is desirable to help them and their families. To the authors’ knowledge, our study represents the only international study to focus on the family impact in children with EA. We had a high number of respondents compared to previous literature within the field that investigated parent och family impact [19, 23–25]. Congenital characteristics (prematurity, low birth weight, anatomical subtype of EA, associated anomalies) and frequency of digestive, feeding, and respiratory problems agree with the previous literature[2]. This suggests the generalizability of the study findings. However, as the PedsQLTM Family Impact Module[30] asks questions regarding problems due to the child’s chronic condition, this study focused on negative impacts. Comparing family impact with healthy peers was inappropriate, and a sibling- or child/self-report was not applicable. We did not use a standardized, validated score assessment of the parents’ socioeconomic standards or clinical symptoms since appropriate ones were not available in Swedish and German at the time of the study. Nevertheless, this limits the study findings' generalizability. We did not detail what additional school support included, and the cognitive ability of children EA may vary since cognitive dysfunction was not an exclusion criterion for our study. We measured the proportion of parents with a doctor-diagnosed disease and parents on sick leave the last year. However, details are not further presented in this article because of personal integrity reasons. Of note, we did not systematically study the educational level of the non-responding parent in the family, nor his/her perception of family impact. In pediatric research, the child’s self-report of HRQOL is of primary importance, but we used the parent-proxy-report because it enables score calculations from child age 2-18 years. The study setting was two North European countries with similar and different societal conditions for families of children with EA. Although the overall response rate was acceptable, missing data may weaken the regression analysis. The relationship of many possible factors influencing the total family impact score was tested, increasing the risk of Type 1 error. However, stepwise regression was used to remove the weakest correlated variable.