We found that several parent-reported feeding difficulties in children aged from 2 to 7 and from 8 to 17 years were associated with lower generic HRQOL, and that an increasing number of feeding difficulties influenced generic HRQOL negatively.
The presence of four feeding difficulties in children from 2–7 and 8–17 years (“eats small portions of food to facilitate eating”, “supplementary nutrition to increase energy content of diet”, “using a gastrostomy tube” and “nutritional intake via a food infusion pump”), were associated with significantly lower total generic HRQOL scores in the parental report. However, the association was not significant in the child report. It has previously been shown that these four feeding difficulties are independently predicted by surgical and/or congenital factors in children with EA.[22] To “take more than 30 minutes to finish a main meal,” and “eat with extra support from an adult,” were significantly associated with low HRQOL in the child and parent reports in the older age group (8 to 17).
The six feeding difficulties listed above were associated with lower scores in the physical functioning domain of the parental report for both age groups. Amin et al previously observed that the need for long-term follow up for nutritional support of children with EA is associated with lower physical functioning scores[34] and Legrand et al observed an association between lower scores in the physical functioning domain and gastroesophageal reflux disease.[35] The physical functioning domain of PedsQL 4.0 revolves around the child’s ability to perform daily physical activities (playing, running, managing personal hygiene) and physical discomfort. Our results could indicate that some feeding difficulties captures a subgroup of EA patients with a more severe condition, which is interesting and warrants further research.
We found that the emotional functioning domain was the least impacted by the feeding difficulties mentioned above. This could perhaps be explained by the feeding difficulties assessed in this study, some of which may also be viewed as behaviour/actions taken by the child to try and eliminate dysphagia or choking (for example eating slowly or eating small portions). In fact, some of these actions could fit the description of problem-solving focused coping strategies in nutritional intake situations,[26] which are associated with good HRQOL.[36] Dysphagia has been shown to be associated with lower emotional functioning scores in PedsQL.[35] The weak association with the emotional functioning indicates that parent-observed items in this study might not be enough to capture children’s subjective experiences of dysphagia.
To “take more than 30 minutes to finish main meals,” and “eat with extra support from an adult,” were associated with lower total HRQOL for both children and their parents in the older group. A hypothesis regarding these results is that children aged 8–17 years might be bothered by feeding difficulties that affect social interactions with their peers and family during mealtimes. This is supported by the association of these two feeding difficulties with low scores in the social and school functioning domains in child and parent reports. The social functioning domain of PedsQL asks about the child’s ability to make friends, social exclusion, and his/her ability to agree with and socialize with peers. The school functioning domain focuses on the child’s ability to keep up with schoolwork, to concentrate in school and whether he/she misses out on school due to health issues. Previous studies have shown that when using a condition-specific approach in the older EA group, both social aspects of eating and social isolation are prominent HRQOL issues from both a child and parent’s perspective.[26] All but one (“supplementary nutrition to increase energy content of diet”) of the six feeding difficulties associated with low HRQOL were associated with low scores in the social functioning domain.
Our study showed that for children with EA, an increase in the number of feeding difficulties is associated with a decrease in generic HRQOL. The cumulative negative relationship is important for young children (2 to 7 years) with EA, who commonly have one or several feeding difficulties (> 60%).[22] The linear association between the number of feeding difficulties and HRQOL implies that feeding difficulties are an important influence on generic HRQOL among children with EA. Since only 25% of the negative association between feeding difficulties and the 2–7 year old child’s lower generic HRQOL could be explained by our model, there are evidently other factors along with parent-reported feeding difficulties that impact HRQOL. The explanation rate for the 8-17year-olds was higher in the parental report (42%) and lower in the child report (17%). Other than feeding difficulties, concurrent anomalies, GERD, barky cough and prematurity have been associated with lower generic HRQOL in children with EA.[37, 38] When taking a condition-specific approach, HRQOL is impaired by several congenital and surgical factors, but throughout childhood it is digestive symptoms which impact EA-specific HRQOL the most.[18] Some of the parent-reported feeding difficulties might be connected to specific EA-related morbidities which in their turn cause low HRQOL. Further research is warranted to determine how feeding difficulties relate to other EA-related symptomatology.
The relationship between feeding difficulties and generic HRQOL in our study was analyzed through a widely used validated generic HRQOL instrument. The nine questions assessing feeding difficulties may not fully comprise EA-specific eating morbidity and some of the feeding difficulties (such as to have a gastrostomy and nutritional intake via food infusion pump) may be interrelated.. Still, of importance, our results indicate that these feeding difficulties have impact on HRQOL. The prevalence and character of feeding difficulties may differ in different child ages. The age range of the groups was chosen because of the cut-off ages in the PedsQL-instrument (self-report from age 8) and as Swedish children start school at age seven. Therefore the patient material was divided into two age groups. The older group (8 to 17 years) was twice as big as the younger (2 to 7 years) group due to its wider age range. A further division of subgroups was not carried out because even as our material comes from one of the largest cohorts in the field,[16] the low prevalence of some feeding difficulties would make statistical analysis less feasible. In order to further understand the relationship between feeding difficulties and HRQOL, including and how to target treatment and follow up, a standardized feeding difficulty score for children with EA should be developed to serve in a multicenter study and enable assessment of a multifactorial model of HRQOL, identifying the specific contribution of feeding difficulties on HRQOL scores in children with EA .