The prevalence of and potential risk factors for Developmental Language Disorder at 10 years in the Raine Study

Abstract

Aim This study sought to determine the prevalence of Developmental Language Disorder (DLD) in Australian school‐aged children and associated potential risk factors for DLD at 10 years. Methods This study used a cross‐sectional design to estimate the prevalence of DLD in Generation 2 of the prospective Raine Study. Participants included 1626 children aged 10 years with available language data. Primary outcomes included variables matching diagnostic criteria for DLD. Associations of other potential prenatal and environmental variables were analysed as secondary outcomes. Results The prevalence of DLD in this sample was 6.4% (n = 104) at 10 years. This sub‐cohort comprised 33.7% (n = 35) with expressive language deficits, 20.2% (n = 21) with receptive language deficits, and 46.2% (n = 48) with receptive‐expressive deficits. No significant difference in sex distribution was observed (52.9% male, p = 0.799). Children who were exposed to smoke in utero at 18 weeks gestation were at increased risk of DLD at 10 years (OR = 2.56, CI = 1.23–5.35, p = 0.012). Conclusions DLD is a relatively prevalent condition in Australian children, even when assessed in middle childhood years. These findings can inform future research priorities, and public health and educational policy which account for the associations with potential risk factors.

Low language capacity in childhood: A systematic review of prevalence estimates

“…It is likely that the identification of being male as a risk factor is the result of retrospective reporting bias and referral bias in clinical samples, which has sex ratios ranging from 2:1 to 6:1 (WHO, 2019). Although there was some variability in sex distribution within the included studies, researchers who were stricter in their adherence to to established diagnostic criteria consistently reported that 53% of their low language sample were male (Calder et al, 2022;Norbury et al, 2016aNorbury et al, , 2016bOryadi-Zanjani et al, 2015). Therefore, it is important to reinforce the use of updated diagnostic criteria to mitigate the referral bias observed in clinical samples.…”

“…In their review, Rudolph (2017) identified that children were equally as likely to meet criteria for Specific Language Impairment (SLI, now Developmental Language Disorder or DLD) if they were a ‘later talker’ as they would if their mother had low maternal education, they had a low 5‐min APGAR score, were born later or were male. Additionally, prospective cohort studies have identified risk factors for low language capacity beyond the preschool years such as parental education, socioeconomic disadvantage, non‐English speaking background, a family history of speech/language difficulties (Christensen et al., 2014; Reilly et al., 2018) and mothers smoking while pregnant (Armstrong et al., 2017; Calder et al., 2022). Most epidemiological studies have also failed to reveal sex‐related differences in language capacity, suggesting being a male may not be identified as a risk factor in unselected population samples (cf.…”

Do parent‐reported early indicators predict later developmental language disorder? A Raine Study investigation

“…However, Tomblin et al found that once parental education was controlled for, the association was washed out, leading the authors to conclude that smoking is more likely a determinant of disadvantaged parenting environment than a causal risk factor for language disorder. Indeed, Calder et al (2022) highlighted that there was a significantly higher proportion of families who had household incomes < A$27,000 per year who did not provide language data, which made it difficult to rule out social determinants of DLD at 10 years. Another limitation noted in this study was that two important criteria for determining DLD diagnosis could not be obtained from the Raine Study dataset: (1) early onset of language difficulties; and (2) the functional impact of the condition.…”

“…Analysis of existing data was approved by the Raine Study and Curtin Human Research Ethics Committee (HREC approval number HRE2021-0117). Demographic information and the frequencies of children with (n = 104) and without (n = 1522) DLD are presented in Table 1, which is rereported from Calder et al (2022Calder et al ( , tab. 2, p. 2047.…”

“…There are alarming patterns of inequity to service access for children with DLD. Although there does not appear to be a significant difference in the sex distribution of children with DLD in population-based studies (Calder et al, 2022;Norbury et al, 2016;Tomblin et al, 1997), Whitehouse (2010) conducted a meta-analysis of sex ratio differences in family aggregation studies of DLD (then referred to as specific language impairment-SLI), and found that more males than females were identified to have DLD via direct assessment methods; however, there were no sex differences observed through indirect assessment of language. There also appears to be a referral bias favouring males in receipt of clinical services for language difficulties.…”