Abstract
Introduction: Children with enlarged tonsils and/or adenoids are at risk for sleep disordered breathing (SDB), a common disorder in childhood characterised by snoring, choking, and pauses in breathing during sleep. SDB has been associated with a number of adverse effects on aspects of daytime functioning, including quality of life, behaviour, general cognitive functioning and memory. It has been documented that these difficulties can have a negative effect on school performance, but it is not known how children with SDB may differ from day to day in their school performance, or how their academic growth from week to week may be different from children without SDB. Furthermore, treatment of SDB by removing obstructive tonsillar and adenoid tissues can result in improvements in SDB symptoms, quality of life, and behaviour, but it is unclear whether this extends to school achievement. Previous research has used measures of academic performance such as parent and teacher report or standardised achievement tests. These kinds of measurements are not able to provide information about growth in academic achievement, and are not designed to be sensitive to treatment effects.
The current research is a case series examining children on the waiting list for adenotonsillectomy (AT), a high risk group for SDB, using a multimethod-multi-informant approach to assessment. These children were assessed in domains where changes have previously been demonstrated after AT, i.e., SDB symptoms, quality of life and behaviour, as well as in academic performance, where results of previous research have been more equivocal. Methods previously used in SDB research to assess academic performance were used alongside alternative indicators of progress in literacy and numeracy. These measures were informed by evidence-based measurement approaches used to evaluate academic intervention, and were selected to probe skills specific to each child’s school year. This approach has previously demonstrated sensitivity to treatment effects in interventions for neurodevelopmental disorders.
Methods: 20 case children aged 3-11 years were recruited from the waiting list for adenotonsillectomy at Dunedin Public Hospital. 19 of these children were matched to controls for age, sex, and school year, including academic performance where possible. 44% of participants were male, and 46% were preschool aged. All children were described in terms of their symptoms of SDB using parent ratings on the Pediatric Sleep Questionnaire, quality of life using the OSA-18 and Pediatric Quality of Life Inventory, behaviour using the SWAN rating scale and BASC-2 BESS , general cognition using the WPPSI-III and WISC-IV, and academic performance at matched time points both before surgery and at a follow up 7 months after surgery. In addition, an AHI was derived from home-based sleep studies conducted both before surgery and at a 3 month follow up. Academic performance was measured using parent and teacher judgements in relation to National Standards, performance on a standardised achievement test (WIAT-II Australian), and using the developmentally appropriate literacy and numeracy measures that were expected to be sensitive to changes across the study. The literacy and numeracy indicators were administered twice, one week apart, at each time point: twice before surgery (pre-‘surgery’), twice at the 3 month follow up (post-‘surgery’), and twice again at the 7 month follow up. These assessment points are similar to the timeline of previous SDB research. The current research takes both visual and statistical approaches to data analysis, allowing for examination of individual case-control pairs and within-subjects changes as well as group-level statistics.
Results: At pre-‘surgery’, case children on average had more parent-reported SDB symptoms than matched controls (p = .001). This pattern was confirmed by AHI from overnight sleep studies indicating that the majority of case participants had clinically significant levels of SDB. As well as increased SDB symptoms, case children on average had poorer quality of life (range p = .001 - .032), behavioural and emotional functioning (p = .020), and memory (p = .001) than matched controls according to parent ratings. Teacher ratings did not suggest elevated levels of behavioural or emotional difficulties for either cases or controls and did not differ between groups, and mean general cognitive functioning scores were also similar between the groups. Typically used measures of academic performance, parent and teacher report and standardised achievement test scores, did not differ on average between case and control children, although case participants’ scores on the WIAT-II mathematics composite score were in the ‘low-average’ range compared to norms. Comparisons of literacy and numeracy indicators between the two pre-‘surgery’ sessions suggested that in general, case children appeared to have more consistent performance than controls from week-to-week, but less growth in these skills. When looking at mean performance on the tasks at pre-‘surgery’, several control children were outperforming their matched case participant, but there was a lot of individual variation in scores.
After surgery, case children had improved parent ratings of SDB symptoms (p = .001), quality of life (range p = .001 - .023), behavioural and emotional functioning (p = .008), and memory (p = .014). With the exception of SDB symptoms and memory difficulties, which remained elevated for case children in comparison to matched controls, case and control children no longer differed statistically for mean scores in any other domain of daytime functioning. Teacher ratings of reading in relation to National Standards improved for case children (p = .011), but no other typically used measures of school performance changed notably following surgery. Improvements in scores on the literacy and numeracy indicators suggested different patterns of growth for different academic domains and age groups. For preschoolers, case children showed evidence of linear growth in an early literacy composite that was steeper than for control children on average (p = .008). However, improvement in early numeracy scores suggesting growth in these skills did not differ between case and control participants on average. Year 1 and 2 case participants showed considerable growth in a phoneme segmentation fluency task that was not observed for their matched controls, but there were no clear patterns for any other literacy or numeracy tasks. Children in Years 3 to 5 had improvements in literacy tasks that for some participants were greater than their matched controls, but did not appear to improve in numeracy. Year 6 and 7 students, in contrast, demonstrated evidence of growth in numeracy tasks, but most did not improve in literacy tasks.
Conclusions: This research is consistent with previous studies reporting that children with SDB are at risk for poor academic performance or growth. The literacy and numeracy indicators used in the current research provided information about academic performance in these children beyond what could be ascertained from more conventionally used measures. This finding suggests that these more sensitive measures of academic growth have potential for evaluating effects of SDB treatment on academic performance in a larger group of children. The current study has clearly demonstrated that children who are undergoing surgery for removal of tonsils and adenoids in Dunedin are likely to have more symptoms of SDB and related negative effects on daytime functioning when compared to similar children in the community. It also provides preliminary evidence that treatment of SDB in early childhood may help to improve learning trajectories in literacy skills, supporting existing evidence for early identification and intervention for SDB. Many children are affected by health problems other than SDB or behavioural difficulties that have been associated with poorer school performance. The literacy and numeracy indicators used in the current research show promise for assessing the effects of surgical intervention for SDB, and therefore have the potential to explore the effects that treatment of other disorders that can affect school performance, such as otitis media (glue ear) or ADHD, could have on academic achievement.