The relationship between auditory behaviours and speech and language development in children with cleft lip and palate

Abstract

Purpose of review 

Children born with a cleft lip and palate (CLP) are at risk of speech problems. There is a high prevalence of conductive hearing loss due to otitis media with effusion and recent research has highlighted broader difficulties with language and educational attainment. There has been increased interest in the impact of auditory processing on a child's development, but little study into links between auditory skills and speech and language development.

Recent findings 

Children with CLP consistently struggle with auditory tasks. Difficulties processing words in a noisy background, auditory attention and temporal processing are frequently reported. A recent study in the United Kingdom investigated the auditory behaviours of 95 children aged 5–8 years with nonsyndromic CLP and compared these with speech and language profiles. Auditory difficulties correlated with speech development and most auditory behaviours correlated with language outcomes. There was also a correlation of language with speech outcomes, suggesting a more complex and interdependent relationship.

Summary 

Children with CLP present with high levels of difficulties with auditory skills. The potential impact of this on broader development needs further research, but clinicians should ensure that auditory skills beyond the hearing test are monitored, and advice given to parents to minimize the impact.

INTRODUCTION

Cleft lip and palate (CLP) is the most common craniofacial anomaly affecting over 1000 live births in the United Kingdom annually ^[1]. Clefts can broadly be categorized into unilateral cleft lip and palate (UCLP), bilateral cleft lip and palate (BCLP) and cleft palate only (CPO). The severity in each of these categories can vary. The umbrella term CLP will be used in this review. CLP can further be described as nonsyndromic and syndromic; about 30% of babies born with CLP will have an associated syndrome ^[2]. A child growing up with a CLP may experience differences in their speech development, facial growth, and hearing.

Children with CLP are at increased risk of conductive hearing loss due to eustachian tube dysfunction. There have been nine abnormalities of the structure of the eustachian tube reported in cleft patients, including a shorter eustachian tube and problems with the insertion of the tensor veli palatini muscle, which shares a role in the function of the palate and the eustachian tube ^[3]. Reported rates of conductive hearing loss in cleft vary but it is commonly accepted as being higher compared with the noncleft population ^[4,5]. Compared with noncleft peers, middle-ear disease is prolonged and often results in later complications ^[6].

Auditory processing difficulties reflect a deficit in the way the brain recognizes and processes sounds ^[7]. They can occur in the presence of normal hearing but are frequently a result of persistent hearing difficulties in early childhood ^[8]. A diagnosis of auditory processing disorder is controversial due to the extensive overlap of problems with dyslexia, speech and language disorder and attention deficit disorder ^[9–11]. It is therefore preferable to describe observable auditory behaviours within given contexts. 

Box 1: no caption available

BACKGROUND

The literature on auditory behaviours in children with CLP has been growing since the late 1990s. Studies consistently show deficits in processing speech in noisy environments and temporal processing of rapid stimuli, as well as attention difficulties ^[12–15]. The use of parent reporting is common in studies. Some studies have also used electrophysiological methods to measure neural processing of auditory stimuli ^[16,17]. Further evidence of auditory dysfunction has come from brain imaging studies ^[18–20].

RECENT FINDINGS

Three articles investigating auditory behaviours have been published within the last 2 years. All have reported similar findings to earlier studies. Zarei et al.^[21] studied 23 children with CLP and normal hearing aged 8–12 years in Iran. They tested temporal processing using a Gaps-in-Noise test, processing speech-in-noise, and binaural integration through a dichotic digit task. Results were compared with an age-matched control group (n = 30). The CLP group performed significantly worse in tests of temporal processing with a mean percentage correct of 64–65% compared with 74–75% in the control group. Although gaps were detected at between 6 and 7 ms by the CLP group, which is within normal limits, this was significantly different from the control group who could detect gaps as short as 3 ms. The CLP group processed only 71–72% of words in the test of speech-in-noise when the target word was at the same decibel level as the background noise. This compared with 93–94% in the control group. By contrast the CLP group performed better than the control group in tests of binaural integration.

Hofer-Martini et al.^[22▪] studied 48 children aged 5–16 years in Germany. All children had normal hearing at the time of assessment. Assessment included speech-in-noise and binaural integration and auditory memory. Binaural integration and auditory memory were a problem for only 17% of the group whereas 69% struggled with speech-in-noise. The authors noted that difficulties were frequently seen in the younger ages and that skills improved, except for processing of speech-in-noise which remained difficult even for the adolescents.

Maximino et al.^[23▪] studied 22 children aged 7–9 years with UCLP in Brazil. They investigated binaural integration, auditory attention and temporal processing, and included memory and language measures. This cohort showed a high level of history of conductive hearing loss (82%). Only one child out of 22 presented with adequate scores on all tests of auditory behaviour. Five children failed all tests. The most difficult task was temporal processing with 81% of the group failing the gap detection text. Results were inconclusive for dichotic listening and attention, with 50% of the group passing both. Expressive language was measured using an expressive vocabulary test and receptive language using a token test where children follow instructions of increasing difficulty by pointing to shapes and colours. Only 9% presented with a difficulty with expressive vocabulary but 50% had difficulties with receptive language. Twenty seven percent presented with difficulties with auditory memory. Although the authors found no correlation between auditory and language measures, nor hearing and auditory or language measures, they do report that 82% had concomitant language and auditory difficulties. They argue that the lack of correlation suggests that the language and auditory tests measure different things and therefore clinicians need to carry out comprehensive assessment of this patient group.

There is consistent evidence from across the world to suggest difficulties with auditory behaviours in people with CLP. Only one has considered this in relation to language outcomes ^[22▪], although the language measures were not comprehensive. No studies up to now have considered auditory difficulties in relation to speech and language development and the clinical implications of this, which is surprising given the substantial risk of these difficulties in children with CLP ^[24–27].

The Cleft Linguistic and Auditory Skills (CLAS) study is a current study which tries to address this gap in the research ^[28▪▪]. The CLAS study was carried out between 2019 to 2021. Its aim was to investigate the relationship between auditory behaviours, language skills and speech outcome.

Participants were recruited from regional cleft centres in England. They were aged between 5 and 8 years old (mean average age 6 years 4 months). Participants had a diagnosis of nonsyndromic UCLP, BCLP or CPO. Any children with a known syndrome, learning difficulty, sensorineural hearing loss or English as a second language were excluded. There was a total of 95 participants recruited to the study.

Participants were assessed using a battery of measures of speech, language and auditory behaviours. All assessments were readily available to be used in a clinic or school setting. The auditory assessments included a hearing screen using the SHOEBOX QuickTest (v.5.6.5, Ottawa, Ontario, Canada) ^[29], a test of temporal processing using the Random Gap Detection Test ^[30], auditory attention measures using the Test of Variables of Attention (TOVA v9.0, Langley, Washington, USA) ^[31], processing speech-in-noise measures using the Auditory Figure Ground (AFG) 8+ and AFG 0 tests from the SCAN-3:C assessment battery ^[32] and a parent report of behaviours using the Evaluation of Children's Listening and Processing Skills (ECLiPS) ^[33].

Data were analysed using descriptive statistics of average mean and median results for the group. A cut-off point of one SD below the mean average was used to describe the percentage of children clinically at risk of ongoing difficulties. Spearman's rho correlations were explored to investigate the relationship between measures of hearing, auditory behaviours and speech and language. Not all participants were able to comply with all assessments.

On the day of assessment, 50 children (58.8%) had normal hearing, 27 (31.8%) had a mild hearing loss and eight (9.4%) had a moderate loss. A total of 61% were reported to have had a history of hearing loss.

Forty-eight percentage (n = 41) of participants were unable to pass the screen for the Random Gap Detection Test indicating a difficulty with temporal processing. The results of the other measures of auditory behaviours are shown in Table 1.

Table 1 - Average standard and scaled scores for measures of auditory behaviours

Measure	n	Median	Mean (SD)	95% CI
AFG8+	91	9.0	9.4 (2.4)	8.9–9.9
AFG0	90	8.0	8.2 (2.6)	7.7–8.8
Attention – distractibility	83	91.0	84.9 (23.1)	79.8–89.9
Attention – focus on target	83	83.0	80.0 (24.2)	74.7–85.3
ECLiPS auditory processing	91	7.0	7.1 (3.8)	6.3–7.9
ECLiPS impact of environment	91	7.0	7.9 (3.7)	7.1–8.7

AFG, Auditory Figure Ground; CI, confidence interval; ECLiPS, Evaluation of Children's Listening and Processing Skills.

For measures of speech-in-noise 22% (n = 20) fell below one SD of the mean average when the target word was 8 dB louder than the background noise. This rose to 35.6% (n = 32) when the target word and the background noise were at the same level. For measures of auditory attention 32.5% (n = 27) fell below one SD for levels of distractibility and 41% (n = 34) were unable to focus on the auditory stimuli enough to react appropriately. The ECLiPS questionnaire showed 45.1% (n = 41) of parents reporting poor levels of auditory processing of sounds and words, and 42.9% (n = 39) reporting a negative impact of the auditory environment on their child's ability to process sounds and words.

There was a weak significant correlation between hearing on the day of assessment and ability to process speech-in-noise (AFG8+: r_s = −0.219, P = 0.049; AFG0: r_s = −0.229, P = 0.040), but no correlation with a history of hearing loss. Parent report of the impact of auditory environment correlated with scores on the AFG0 test (r_s = 0.253, P = 0.019). Speech outcomes correlated with both of these measures (AFG0: r_s = 0.229, P = 0.030; ECLiPS environment: r_s = 0.271, P = 0.009). Auditory attention (focus) was also correlated with speech outcome (r_s = 0.262, P = 0.017).

Almost all measures of auditory behaviours correlated with language measures. Results of these correlations are shown in Table 2.

Table 2 - Correlations between measures of language and auditory behaviours

	RGDT	AFG8+	AFG0	Attention – distractibility	Attention – focus	ECLiPS auditory processing	ECLiPS environment
Receptive language	−0.223∗ (P = 0.041)	0.217∗ (P = 0.039)	0.257∗ (P = 0.017)	0.204 (P = 0.065)	0.227∗ (P = 0.039)	0.239∗ (P = 0.022)	0.249∗ (P = 0.018)
Expressive language	−0.073 (P = 0.507)	0.290∗ (P = 0.005)	0.321∗ (P = 0.002)	0.197 (P = 0.074)	0.347∗∗ (P = 0.001)	0.395∗∗ (P < 0.001)	0.401∗∗ (P < 0.001)

AFG, Auditory Figure Ground; ECLiPS, Evaluation of Children's Listening and Processing Skills; RGDT, Random Gap Detection Test.

^∗P < 0.05.

^∗∗P < 0.001.

Correlations were also seen between language measures and speech outcomes (receptive language: r_s = 0.228, P = 0.026; expressive language: r_s = 0.239, P = 0.020).

Logistic regression was used to analyse auditory behaviour measures as predictors of speech and language outcomes. When measures were entered into the model and adjusted for hearing on the day and history of hearing difficulties, environmental impact as reported by parents was found to be a significant predictor, with those struggling with noisy environments almost seven times more likely to have speech development below the 5th centile [odds ratio (OR): 6.88 (95% confidence interval (CI): 1.11–42.50); P = 0.038]. Similar results were seen for language outcomes, with environmental impact making it 12 times more likely that a child will have receptive language difficulties below one SD of the mean average [OR: 12.16 (95% CI: 1.25–118.48); P = 0.031].

The CLAS study aimed to replicate findings from previous research of auditory behaviours in people with CLP using assessments that are readily available clinically, and to examine the relationship of auditory behaviours to speech and language outcomes. This study showed consistently lower than average scores for measures of auditory behaviour in children with CLP. Group average scores were lower than the normative mean but within the normal range. However, up to 40% of children were seen to have clinically significant difficulties on at least one measure of auditory behaviour.

A test of temporal processing using a gap detection test showed 48% of children failing this screen. This task is difficult for children of the age in this study, but the findings here do replicate other studies. In fact, in some studies much higher levels of difficulty have been reported. For example, Maximino et al.^[23▪] reported an 81% failure rate in their study as discussed earlier and Boscariol et al.^[12] reported a ‘poor performance’ rate as high as 95%. In a large study with over 200 participants, Ma et al.^[14] found a maturational delay of 2 years in temporal processing in children with CLP. Temporal processing skills are thought to be crucial for the development of language skills as the ability to process rapid auditory stimuli is required to understand speech. Indeed, in this study we saw a positive correlation with the ability to process rapid auditory stimuli and receptive language measures. If there is a maturational delay to these skills, we would expect there to be concomitant language delays in the early years with some catch up in older children. This is exactly as reported in a recent review ^[26].

Auditory attention skills were poor in this group with around 40% of children showing some signs of difficulty focusing on auditory stimuli. No other articles consider auditory attention through assessment of pure auditory stimuli. However, previous studies have found comparable results to this study using attention measures with a linguistic basis and parent report of attention. Maximino et al.^[23▪] reported 50% of their cohort failing an attention measure which used words as the target. Lemos and Feniman ^[34] used a similar measure and found children with CLP made more errors than their noncleft peers and were significantly more distracted (P = 0.008). In studies using parent report, parents of children with CLP consistently report difficulties with attention. Ma et al.'s ^[15] study showed 40.7% of parents of children with CPO reported difficulties with attention span compared with 27.7% of those with CLP and only 10% of those with children in the control group of typically developing children. Auditory attention is key to developing language. Simply put, if a child struggles to attend to sounds and filter out distractions they will find it more difficult to learn novel words. This study found that there was a relationship between ability to attend to auditory material and both speech and language development.

Processing speech in noisy environments is a skill closely related to auditory attention. The ability to filter out distracting noise to attend to a stimulus is the first step being able to process speech sounds and words ^[35]. This is an auditory behaviour consistently shown to be of difficulty for children with CLP. In this study, children found it more difficult to process the words when the background noise was increased, with up to 35% performing at a clinically low level. There was a correlation with hearing on the day, which is not surprising, but this serves as a reminder to those working with children with CLP to be vigilant about the auditory environment due to frequent episodes of conductive hearing loss that are likely to occur. As discussed above, Hofer-Martini et al.^[22▪] reported a failure rate on speech-in-noise tests of 69% across all ages. Similarly, Yang ^[36] reported highly significant differences between children with CLP and noncleft peers (P < 0.001) in measures of speech-in-noise.

Parent report in this study using the ECLiPS questionnaire ^[33] was found to correlate with a child's performance on the speech-in-noise tests and the attention tests. Parent report of difficulties processing and attending in noisy environments was also found to be a significant predictor of speech and language outcomes.

CONCLUSION

Studies from across the world using both behavioural experimental and electrophysiological measures have consistently found poor auditory behaviours in children with CLP. The most recent study reviewed here found that children aged 5–8 years with CLP showed that these difficulties were detectable using parent report and readily available clinical tools. The study further found relationships between auditory skills and speech and language outcomes. The potential impact of this on broader development and educational outcomes needs further research, but clinicians should ensure that auditory skills beyond the hearing test are monitored, and advice given to parents and teachers to minimize the impact. Speech and Language Therapists working in this field need to work closely with colleagues in audiology and ENT in the early years to optimize outcomes for children with CLP and to work closely with families to optimize listening environments in the crucial language acquisition years.

Acknowledgements

I would like to thank Dr Helen Stringer and Professor Cristina McKean at Newcastle University for their support and supervision of this study.

Financial support and sponsorship

The current study/project was carried out in part fulfilment of a clinical doctoral fellowship funded by the NIHR (National Institute for Health Research) [ICA-CDRF-2017-03-002]. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.

Conflicts of interest

There are no conflicts of interest.