Background
Childhood communication disorders can include speech and/or language problems to the extent that some children require special education intervention. Speech and/or language impairments that stem from no clear cause are considered to be primary, whereas impairments resulting from a developmental disability such as hearing impairment, are referred to as secondary. Approximately 7 % of school-aged children will unexpectedly fail to develop language, confirmed by performing poorly on tests of receptive and/or expressive language [
1,
2]. Further investigations confirm that the language problems occur in the absence of not only cognitive impairment, but also any other condition that could explain the deficit. This presentation represents a primary language disorder called specific language impairment (SLI) [
1,
2]. Emotional/behavioral issues [
3], along with gross and fine motor comorbidities are not unusual in this cohort of children [
4]. Consequently, SLI has both immediate and long-term influences on children’s lives, with many of these influences extending beyond the language domain often leading to academic underachievement [
5], along with problems in developing and maintaining interpersonal relationships [
6].
Health-related quality of life (HRQOL) is the perception of the impact an illness/injury, medical treatment and/or health care policy has on one’s life [
7]. HRQOL is multidimensional and includes physical functioning, along with psychosocial dimensions of emotional, social and any related constructs [
8]. HRQOL has been explored in many child populations, using either generic or condition-specific questionnaires. It is agreed that a child reporting his/her own perceptions is optimal, however in paediatric populations it may only be possible to explore a child’s HRQOL indirectly, via parents/guardians or health professional’s completion of a proxy-report [
9]. Considering the array of challenges faced by children with SLI, this diagnosis has the potential to negatively affect every day functioning across various domains and consequently the HRQOL of these children. Yet little attention has been dedicated to the impact SLI has on a child’s HRQOL. Children 8–11 years old (
n = 55) diagnosed with receptive-SLI (confirmed by language scores falling 1standard deviation (SD) below the mean) had different profiles when compared to norms, and scored significantly lower in speech and sleep domains of the 17D questionnaire [
10]. Although these differences were not large enough to impact the overall HRQOL score, the results do provide preliminary evidence that even a mild presentation of SLI can influence HRQOL domains [
10]. To date, no further studies have investigated the HRQOL perceptions from children with SLI and their parents.
Extending beyond a diagnosis of SLI, Markham et al. [
11] presented the only other paper that reported the child’s perception of the impact speech and/or language problems have on his/her HRQOL. Markham et al. [
11] used focus group interviews to discuss the perceptions of 29 children and young people aged 6.6–18.8 years with a broad range of speech and/or language problems, enrolled fulltime in either mainstream (
n = 22), or special schools for children with communication needs (
n = 7). This study reported that the speech and/or language difficulties of these children where perceived by the children to impact negatively on their HRQOL [
11]. In addition, Markham et al. [
11] reported credible evidence that children with speech and/or language difficulties can provide insight into their own HRQOL, and that common themes found in these discussions closely reflected domains of generic HRQOL measures [
11]. These findings complemented a previous study conducted by Markham & Dean [
12] where parents (
n = 11) of children 2–9 years of age with speech and/or language difficulties, along with speech language pathologists (
n = 12), and other professionals (
n = 12) (such as educational staff), came together in focus group interviews to discuss their perceptions of the impact speech and language difficulties have on the HRQOL of predominately mainstreamed children [
12].
A few additional studies have used proxy-reporting to explore the HRQOL of children with communication disorders. Parents of 3-year-old children with a language disorder completed the TNO-A21 Pre-school Children Quality of Life Questionnaire, which revealed that all children with language disorders, irrespective of the criteria used for identifying a language problem, had a lower HRQOL, discriminating 3-year-old children with language problems from those without [
13]. In a follow-up study the parents who reported a persistent language disorder in their 8-year-old child, completed the Child Health Quality of Life Questionnaire - parent form 28 and rated their child’s general behavior, self-esteem, mental health, attitude towards schoolwork, emotional stability and psychosocial summary score all significantly lower than children without language disorders [
14]. Therefore, emerging evidence suggests that mild speech and/or language disorders have the potential to impact negatively on a child’s HRQOL as perceived by the child or his/her proxy.
An explanation for the lack of attention devoted to this cohort of children, may be related to the challenges associated with selecting a suitable HRQOL measure.
A recent review explored the literature to identify the measures used for children with speech and language disorders [
15]. Out of the four measures identified, three of these would not have been useful for the current study cohort, as two were designed for adult use, while another was for children younger than 6 years. The final measure, the Pediatric Speech and Language Quality of Life Scale, may be suitable for use with children with severe SLI, particularly when a condition-specific HRQOL measure for use with children with SLI is non-existent. However, this measure was not available at the time of data collection.
It is becoming accepted that SLI is not simply a language impairment, but is a heterogenic disorder where children may present with problems impacting cognitive and motor function. Indeed, condition-specific measures need to be specifically attuned to the particular ways in which communication disorders impact the HRQOL of children, while being sensitive and specific enough to identify small and important differences relevant to children with speech and language disorders [
15]. However, it is equally important to gain a holistic understanding of the child’s HRQOL, particularly with conditions that have the potential to impact various HRQOL domains, and ideally allow for comparison against a range of paediatric conditions. Therefore, in order to explore the extent that the diverse problems experienced by children with severe SLI are impacting on the child’s perceived HRQOL a generic measure was sought. Most generic paediatric HRQOL measures have not yet been proven valid and/or reliable for use with children/adolescents with speech and language disorders [
15], including children with SLI.
The Pediatric Quality of Life Inventory™ Version 4.0 Generic Core Scales (PedsQL™) [
9] is one of the most commonly used generic measures for exploration of HRQOL in both healthy children and different cohorts of children with various diagnoses. However, this measure has yet to be explored for use with children with SLI [
15]. The validity and reliability of the PedsQL™ is well established [
8,
9] for both child self-report (from 5 years of age) [
16] and proxy-report (from 2 years of age) [
17]. This measure is particularly appealing for use with children with severe SLI due to the use of visual cues and developmentally appropriate language (consistent with the expected reading level of a child below the youngest target age of the particular questionnaire version) [
9,
18]. It also allows comparison to be made amongst children with a variety of disabilities or impairments. Varni et al. [
8,
16,
17] performs a consistent series of statistical analyses to explore initial feasibility, reliability and interpretation of results on the PedsQL™ to determine the suitability of the PedsQL™ for use with a particular cohort, for example children with developmental disorder [
19]. These analyses completed by Varni et al. [
8,
16,
17] have been published and supported for use in studies exploring HRQOL measures prior to Varni et al. [
8,
16,
17]. The PedsQL™ has not been validated for use with children with speech and language disorders despite demonstrating sensitivity and responsiveness in other populations [
15]. This study is a first step in addressing this gap by exploring the suitability of the PedsQL™ for use with children with severe SLI.
The HRQOL of children with severe SLI needs to be explored, as severe presentations require more understanding and support. Reporting children’s own perceived HRQOL will assist parents/guardians, professionals and policy makers to gain insight into the needs of this group. Further, it would assist service planning to holistically address identified issues to achieve optimal outcomes in children with a severe presentation. Therefore, this study aimed to examine the initial feasibility and reliability of the PedsQL™ in school-aged children with severe SLI and their parents. The method employed by Varni et al. [
8,
16,
17] and colleagues [
19] for confirming suitability of the PedsQL™ will be used in this study. If the measure is suitable, we further aim to explore and compare the HRQOL perceptions of children with severe SLI and their parents, to the normative population, to identify if these children, or gender and age subgroups of them, are at risk for impaired HRQOL. Given the previous findings, it is hypothesized that the PedsQL™ will be an appropriate measure for use with this cohort and will identify a high risk of impaired HRQOL compared to the healthy population mean, with the largest differences being in social and school functioning, regardless if reported by the child or his/her proxy.
Discussion
The results of this study support the initial feasibility and reliability of using the PedsQL™ child self-report and proxy-report with school-aged children with severe SLI. The fact that there were no missing item responses, suggests that children with severe SLI and their parents are willing and able to provide good-quality data. In addition, the lack of ceiling or floor effects suggests that children with severe SLI were not navigating to either end of the scale. Children with severe SLI and their parents, at no time perceived themselves to score the worst possible score on any item. This was reflected in a consistently 0 % floor effect. This could reflect the fact that the children’s main challenges encompass speech and/or language development which none of the items in this measure specifically target. It can be anticipated that a measure specific for language impairment would produce some floor effects. In addition, failure to produce any significant ceiling effects, suggest that this measure is ideal as it captures challenges children with severe SLI perceive to be difficult across various domains of HRQOL, highlighting the complexity of this developmental disorder. Therefore, it would appear that the PedsQL™ is a promising measure for use with this cohort of children. However, it is worth highlighting that the researchers were dependent on signing the words while reading the questions to most of the children. Without the use of sign language, it is questionable if the children could have answered all the questions. In addition, it is worth noting that the simplicity of the vocabulary that Varni et al. (1999) have used in developing this scale facilitated both the comprehension of questions by the child with language impairment (in the absence of cognitive impairment), and the researchers’ capacity to precisely sign the sentence to avoid any bias. Although the measure is quick to administer, some children became distracted and appeared to focus less on the work at hand. When this occurred, the child was given a break to complete some movement tasks, such as jumping activities. One small break seemed to be enough to redirect the child’s attention to the questionnaire allowing for completion. Future researchers may need to adopt the above-mentioned strategies in order to optimize data collection from children with severe SLI.
Consistent with previous literature [
11], this study confirmed the hypothesis that children with severe SLI and their parents both perceived the child to be at-risk for impaired overall HRQOL relative to the healthy population sample, reflective of scores obtained by children with physician-diagnosed severe chronic health conditions [
8]. The total HRQOL summary score fell below the cut-off score for 60 % of children with severe SLI, with a similar 56 % of parents rating their children below the cut-off score.
It has been reported that parents more often score their children worse than their child on measures of HRQOL [
26]. Consistent with this literature, the parents of children with severe SLI rated their child’s HRQOL worse than their children themselves across all domains except physical functioning [
26]. However, only the social functioning domain reached significance, with parents (similar to their children) reporting low social functioning but at a significantly lower rate. It has been reported that the primary caregiver is the most valid proxy [
24]. In this study, 100 % of mothers, of whom a large percentage were stay at home mothers, completed the questionnaire. Considering children and parents did not complete their questionnaire simultaneously, it would appear that the responses from the children seem reasonable in comparison to the mother’s. While the authors are not suggesting a proxy-report is entirely adequate to replace the child’s report, the marked similarities between parent and child reports suggest that these children can reflect and provide information pertaining to their own HRQOL.
Further investigation into the child and parent responses reveals that out of the four scale scores, children rated both physical and social functioning below the cut-off point. Interestingly, contrary to our hypothesis, the children in this study rated physical functioning the lowest. It may be expected that a group of children with severe communication challenges would identify their own poor social functioning. However, it may not have been anticipated that these children would rate their physical functioning low as well, considering their most evident challenge is communication, and not physical. However, this outcome complements emerging evidence that up to 90 % of children with various speech-language delays/disorders present with motor comorbidity to some extent [
4]. Until now it was not clear that the motor comorbidity also impacted their HRQOL, and that together the perceived impairment in social and motor performance was severe enough to impact the total summary score for 60 % of the children in this study. These results contribute to the on-going debate within the literature supporting the premise that SLI is not limited to language impairments, but encompass problems in other developmental domains. Parents did not perceive an impact on physical functioning to the same extent as their children. In fact, compared to the cut-off scores, parents’ perceptions of physical functioning had the largest difference, however, unlike the children the parents mean score fell above the healthy population cut-off score. This difference may be because parents see their child’s language impairment as the primary impairment, and therefore may not identify physical challenges as a problem. This is consistent with the literature where parents often overestimated their child’s gross motor skills in children with vocabulary difficulties [
27,
28]. This is concerning as it is parents’ unique perceptions of their child’s HRQOL that guide and direct clinical decision making and utilization of health care services [
17,
29,
30]. Parents failing to identify their child’s physical challenges will likely not seek services to improve physical capacity, potentially missing opportunistic intervention. Although the parents did not perceive physical functioning problems to the extent that their child did, the results from this study suggest that HRQOL measures used with children with severe SLI should embrace questions pertaining to physical functioning if the researchers or clinicians wish to gain an understanding of the extent of difficulties perceived by the child and his/her parent. Further, this will ensure children are referred to inter-professional management accordingly, to enhance management. Unlike the children, out of the four scale scores, parents reported both emotional and social functioning below the cut-off score. For emotional functioning, 50 % of parents scored their child below the cut-off score, with the mean score only just falling below the at-risk score. In comparison only 28 % of children rated themselves below the cut-off score. Although the pre-defined 15 % threshold was not met, 12 % of children had a ceiling effect in emotional functioning suggesting these children perceived themselves as not feeling scared, sad, angry, and worried or as having troubles sleeping. It could equally be argued that the children drifted towards the ceiling because they had difficulties understanding these emotions, however, the responses were not consistent enough to create a significant ceiling effect, so this measure remains useful for this cohort. A significant difference in the social functioning scores may reflect the parents’ increased awareness of disparities and broader basis for comparison of their child’s functioning, while the children may have rated this based on their main socializing experiences amongst like individuals, which was at the dedicated school. Similarly, the pre-defined 15 % threshold was not met, yet 12 % of children had a ceiling effect in social functioning supporting the premise that children perceived themselves able to get along with other children and make friends, likely within the protected environment of the school.
Considering the children in this study are schooled in a setting catering to children with similar diagnoses, it is understandable that both children and their parents rated school functioning above the cut-off score. However, the child school functioning mean score was just above the cut-off score, and 42 % of parents along with 51 % of children, perceived the child’s school functioning to be “at-risk.” This suggests that even in a protected setting, many children with severe SLI and their parents identify difficulties in school functioning.
Although females consistently scored themselves lower than males, caution is needed when interpreting this due to the small numbers of females impacting data analysis. However, the large percentage of males in our study reflects the higher prevalence of males diagnosed with SLI [
1]. Similarly, there were differences between the age groups, however, limited numbers impacted data analysis. Importantly, the age group results suggest that children with severe SLI do not grow out of their impairment and continue to perceive an impact on their HRQOL well into adolescence. Scale score differences between the age groups likely reflect the perceptions of importance that different domains have at different ages. For example, younger children who often engage in more movement-based play perceived their physical functioning to be “at-risk” for impairment, whereas, older children who likely spend more time socializing and studying identified social and school difficulties. Further, any real differences could be due to an increased awareness of disparities between themselves and typically developing peers due to unique life experiences. Consequently, further research is required to explore the HRQOL of children with SLI from both larger samples, longitudinally, and across subgroups not only age and gender, but also receptive compared to expressive language impairments. In order to further guide the management of children with SLI, studies should qualitatively explore reasons for disparities, or analyze assessment findings in light of HRQOL perspectives. This study has confirmed the suitability of the PedsQL™ for use with children with severe SLI, yet there may be another HRQOL measure that is equally suitable or even optimal for use with this cohort. Gomersall et al. [
15] proposed the comparison of multiple generic HRQOL measures in one study. The PedsQL™ compared to other measures will be an ideal approach to identifying the optimal measure for clinicians to use with children with severe SLI.
Limitations of this study include data collection from one school, with a small sample size, particularly the number of females and older children limiting analysis. However, this study is the first to explore the HRQOL of children with severe SLI and provides preliminary evidence that SLI impacts multiple HRQOL domains and that these children should not be excluded from future studies.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
KN: contributed to conception and design of the study, supervision of data acquisition, contributed to data analysis and interpretation, drafted the manuscript, and approved the final version of the manuscript. PW: contributed to planning and designing the study, conducted data analysis, contributed to interpreting the results, drafted the manuscript, and approved the final version of the manuscript.