View in EDS HTML Full Text PDF Full Text

Subgroups in Language Trajectories from 4 to 11 Years: The Nature and Predictors of Stable, Improving and Decreasing Language Trajectory Groups

Saved in:
Bibliographic Details
Title: Subgroups in Language Trajectories from 4 to 11 Years: The Nature and Predictors of Stable, Improving and Decreasing Language Trajectory Groups
Language: English
Authors: McKean, Cristina (ORCID 0000-0001-9058-9813), Wraith, Darren, Eadie, Patricia, Cook, Fallon, Mensah, Fiona, Reilly, Sheena
Source: Journal of Child Psychology and Psychiatry. Oct 2017 58(10):1081-1091.
Availability: Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: cs-journals@wiley.com; Web site: http://www.wiley.com/WileyCDA
Peer Reviewed: Y
Page Count: 11
Publication Date: 2017
Document Type: Journal Articles
Reports - Research
Descriptors: Language Acquisition, Child Development, Language Aptitude, Longitudinal Studies, Cohort Analysis, Regression (Statistics), Predictor Variables, Environmental Influences, Biological Influences, Non English Speaking, Disadvantaged, Family Environment, Childrens Literature, Emotional Problems, Family Literacy, Learning Disabilities, Intervention, Progress Monitoring, Foreign Countries, Language Tests, Intelligence Tests, Behavior Problems, Child Behavior, Screening Tests, Questionnaires, Achievement Tests, Statistical Analysis
Geographic Terms: Australia
Assessment and Survey Identifiers: Clinical Evaluation of Language Fundamentals, Kaufman Brief Intelligence Test, Strengths and Difficulties Questionnaire, Wide Range Achievement Test
DOI: 10.1111/jcpp.12790
ISSN: 0021-9630
Abstract: Background: Little is known about the nature, range and prevalence of different subgroups in language trajectories extant in a population from 4 to 11 years. This hinders strategic targeting and design of interventions, particularly targeting those whose difficulties will likely persist. Methods: Children's language abilities from 4 to 11 years were investigated in a specialist language longitudinal community cohort (N = 1,910). Longitudinal trajectory latent class modelling was used to characterise trajectories and identify subgroups. Multinomial logistic regression was used to identify predictors associated with the language trajectories children followed. Results: Three language trajectory groups were identified: "stable" (94% of participants), "low-decreasing" (4%) and "low-improving" (2%). A range of child and family factors were identified that were associated with following either the low-improving or low-increasing language trajectory; many of them shared. The low-improving group was associated with mostly environmental risks: non-English-speaking background, social disadvantage and few children's books in the home. The low-decreasing group was associated with mainly biological risks: low birth weight, socioemotional problems, lower family literacy and learning disability. Conclusions: By 4 years, services can be confident that most children with low language will remain low to 11 years. Using rigid cut-points in language ability to target interventions is not recommended due to continued individual variability in language development. Service delivery models should incorporate monitoring over time, targeting according to language abilities and associated risks and delivery of a continuum of interventions across the continuum of need.
Abstractor: As Provided
Number of References: 43
Entry Date: 2017
Accession Number: EJ1154693
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGDDplVsUMkxCT3Hk-GIn_jAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDHpDfVczq99POBdPBAIBEICBm8xc-6qHS5OqXFFtE0YzC3F_L1f-wcFdYE9r-29c1_KDNcF2n9jzfbRMGbp19-6w2XQAciIxkT3Jvjm_DgwyR6iMf-OeCp2zYUOj3_xol5rOvsl6mZF0l7K_9dULEpmuba3hRTFgcdLwV9Fdd7r_1kZgi0mxvQF2z_DKK7Vs8LPxat57yPM91rW07ksLxXXuQzgPG4XB19yHoyC1
Text:
  Availability: 1
  Value: <anid>AN0125199443;jyy01oct.17;2024Jun04.08:15;v2.2.500</anid> <title id="AN0125199443-1">Subgroups in language trajectories from 4 to 11 years: the nature and predictors of stable, improving and decreasing language trajectory groups. </title> <p>Background: Little is known about the nature, range and prevalence of different subgroups in language trajectories extant in a population from 4 to 11 years. This hinders strategic targeting and design of interventions, particularly targeting those whose difficulties will likely persist. Methods: Children's language abilities from 4 to 11 years were investigated in a specialist language longitudinal community cohort (N = 1,910). Longitudinal trajectory latent class modelling was used to characterise trajectories and identify subgroups. Multinomial logistic regression was used to identify predictors associated with the language trajectories children followed. Results: Three language trajectory groups were identified: ‘stable’ (94% of participants), ‘low ‐ decreasing’ (4%) and ‘low ‐ improving’ (2%). A range of child and family factors were identified that were associated with following either the low ‐ improving or low ‐ increasing language trajectory; many of them shared. The low ‐ improving group was associated with mostly environmental risks: non ‐ English ‐ speaking background, social disadvantage and few children's books in the home. The low ‐ decreasing group was associated with mainly biological risks: low birth weight, socioemotional problems, lower family literacy and learning disability. Conclusions: By 4 years, services can be confident that most children with low language will remain low to 11 years. Using rigid cut ‐ points in language ability to target interventions is not recommended due to continued individual variability in language development. Service delivery models should incorporate monitoring over time, targeting according to language abilities and associated risks and delivery of a continuum of interventions across the continuum of need.</p> <p>Language development; language disorder; longitudinal trajectory; latent class</p> <p>There is growing recognition that limited language abilities in childhood can have lifelong implications. The associated difficulties with forming and maintaining peer relationships (Conti ‐ Ramsden & Botting, [<reflink idref="bib8" id="ref1">8</reflink>] ), and with literacy, and educational attainment (Snowling, Adams, Bishop, & Stothard, [<reflink idref="bib35" id="ref2">35</reflink>] ) have measurable downstream consequences for adult mental health, social inclusion and employment (Law, Rush, Schoon, & Parsons, [<reflink idref="bib17" id="ref3">17</reflink>] ). The promotion of robust child language development is therefore recognised as a global priority in many educational and social policies. This paper characterises the prevalence and natural history of developmental trajectory subgroups in child language development between 4 and 11 years in a specialist language longitudinal community cohort of children in Victoria, Australia.</p> <p>Instability in language profiles in the preschool years is well recognised (Bornstein, Hahn, & Putnick, [<reflink idref="bib7" id="ref4">7</reflink>] ). Longitudinal population samples have demonstrated most children who experience early ‘delays’ catch up with their peers (Ghassabian et al., [<reflink idref="bib10" id="ref5">10</reflink>] ; Henrichs et al., [<reflink idref="bib14" id="ref6">14</reflink>] ; Reilly et al., [<reflink idref="bib29" id="ref7">29</reflink>] ; Zambrana, Pons, Eadie, & Ystrom, [<reflink idref="bib42" id="ref8">42</reflink>] ). Conversely, these studies also reveal that after a positive start some children develop later language difficulties (Ghassabian et al., [<reflink idref="bib10" id="ref9">10</reflink>] ; Zambrana et al., [<reflink idref="bib42" id="ref10">42</reflink>] ). Approximately, 7% of 4 ‐ to 5 ‐ year ‐ old children are estimated to have language problems (Norbury et al., [<reflink idref="bib23" id="ref11">23</reflink>] ) and although instability is more pronounced in the preschool than school years (Bornstein et al., [<reflink idref="bib7" id="ref12">7</reflink>] ), a significant proportion of children continue to move between impaired and nonimpaired categories after school entry (McKean et al., [<reflink idref="bib21" id="ref13">21</reflink>] ; Zubrick, Taylor, & Christensen, [<reflink idref="bib43" id="ref14">43</reflink>] ).</p> <p>Current knowledge regarding children's language trajectories means intervention services are likely to both over ‐ and underservice some children. Identifying and understanding the differences between children who are likely to have persisting long ‐ term difficulties, those whose difficulties may resolve and those for whom language difficulties emerge later in development is therefore an important research priority. Such analysis could inform public policy aiming to meet the needs of this population with respect to supporting language development and intervention targeting (Bishop, Snowling, Thompson, & Greenhalgh, [<reflink idref="bib6" id="ref15">6</reflink>] ; Conti ‐ Ramsden, St. Clair, Pickles, & Durkin, [<reflink idref="bib9" id="ref16">9</reflink>] ).</p> <p>Previous studies have explored subgroups in language trajectories; however, methodological limitations exist with respect to both analytical approach and sampling. The most common approach to defining subgroups in language is a ‘categorical’ one; assigning children to either impaired or unimpaired groups at specific cut ‐ points in language scores at two or more time points (Beitchman et al., [<reflink idref="bib4" id="ref17">4</reflink>] ; Bishop & Edmundson, [<reflink idref="bib5" id="ref18">5</reflink>] ; Law, Rush, Anandan, Cox, & Wood, [<reflink idref="bib16" id="ref19">16</reflink>] ; Snowling, Duff, Nash, & Hulme, [<reflink idref="bib36" id="ref20">36</reflink>] ; Zambrana et al., [<reflink idref="bib42" id="ref21">42</reflink>] ; Zubrick et al., [<reflink idref="bib43" id="ref22">43</reflink>] ). While providing important insights, this approach has significant disadvantages. Measurement error inevitably leads to instability in group membership for children whose scores fall near a cut ‐ point. Regression to the mean can suggest changes in children's profile that are, in fact, artefacts of repeated measurement. Furthermore, the cut ‐ point at which ‘impairment’ is defined is necessarily arbitrary in such approaches and creates a bias a priori to finding a ‘disordered’ pathway.</p> <p>Advanced analytical approaches such as longitudinal trajectory latent class analyses and the multilevel model of change have been applied to understanding school ‐ age language trajectories. However, studies have either used clinical samples of children with Developmental Language Disorder (DLD) and considered trajectories within that group (Law, Tomblin, & Zhang, [<reflink idref="bib18" id="ref23">18</reflink>] ), or used matched cohorts of children with DLD and typically developing children (Beitchman et al., [<reflink idref="bib3" id="ref24">3</reflink>] ; Rice & Hoffman, [<reflink idref="bib30" id="ref25">30</reflink>] ), or have not identified subgroups (McKean et al., [<reflink idref="bib20" id="ref26">20</reflink>] ; Taylor, Christensen, Lawrence, Mitrou, & Zubrick, [<reflink idref="bib38" id="ref27">38</reflink>] ). Thus, estimates of prevalence of different subgroups have not been made nor are we certain that the full range of school ‐ age trajectories extant in a population has been uncovered. We address these challenges through the application of longitudinal trajectory latent class analysis, a method which minimises issues associated with measurement error and repeated measurement, to data from a specialist language longitudinal community cohort (the Early Language in Victoria Study – ELVS) considering children aged 4–11 years.</p> <hd id="AN0125199443-2">Predictors of prognosis</hd> <p>Recently, Zambrana et al. ([<reflink idref="bib42" id="ref28">42</reflink>] ) and Snowling et al. ([<reflink idref="bib36" id="ref29">36</reflink>] ) examined ‘trajectories’ of DLD in early to middle childhood (3–5 and 3–8 years respectively). They suggest a ‘late emerging’ trajectory may be most influenced by genetic mechanisms, as indicated by family history of language or literacy difficulties, and a ‘persisting’ trajectory may reflect multiple accumulative risks (Zambrana et al., [<reflink idref="bib42" id="ref30">42</reflink>] ) including social disadvantage (Snowling et al., [<reflink idref="bib36" id="ref31">36</reflink>] ). In addition to characterising the nature of subgroups in language trajectories, this paper also aims to build on these previous studies to identify the specificity with which predictors are associated with the language trajectories that children will follow. In this special edition, Bishop and colleagues advocate the term Language Disorder be used for children who are likely to have language problems ‘enduring into middle childhood and beyond’ (p. 1070). Bishop acknowledges a major challenge in operationalising this for practice is the ‘relatively limited evidence regarding prognostic indicators’ (p. 1076) making identification of children likely to have ‘enduring’ difficulties challenging. Indeed, as children transition into formal schooling, even those who may go on to receive diagnoses of co ‐ occurring conditions such as autistic spectrum disorder (ASD) or attention deficit hyperactivity disorder (ADHD) often have not been identified. Clinicians and educators may remain unsure as to which of the children they support are most at risk. We therefore examine whether clinically applicable predictors can be identified which indicate whether children's difficulties are likely to persist or indeed worsen over time to support implementation of Bishop et al.'s recommendations to practice.</p> <hd id="AN0125199443-3">Empirical analyses using ELVS</hd> <p>Given the limited knowledge of the subgroups of trajectories in school ‐ age language development across the range of language ability, we adopted an exploratory approach. In a community sample beginning as children transition into formal schooling and ending at the threshold of high school and adolescence, we asked:</p> <p>What are the subgroups in trajectories of language development from 4 to 11 years that may be identified within a community sample using longitudinal latent class analysis?</p> <p>What are the predictors of the trajectories children follow that hence can be used as indicators of prognosis?</p> <hd id="AN0125199443-4">Methods</hd> <hd id="AN0125199443-5">Participants and procedures</hd> <p>Participants were from the ELVS cohort, a specialist language longitudinal community cohort which is largely representative of children in Victoria. Detailed recruitment, sampling procedures and exclusion criteria are provided elsewhere (Reilly et al., [<reflink idref="bib28" id="ref32">28</reflink>] ). At baseline, 1,910 children aged 7.5–10 months were recruited (see Appendix S1 for participant flow chart and demographic data). Parents completed questionnaires at baseline, annually from 1 to 7 years, and then at 9 and 11 years. Direct child assessments were carried out at 4, 5, 7 and 11 years of age. Ethics approval was provided by the Human Research Ethics Committees at the Royal Children's Hospital, Melbourne and La Trobe University.</p> <hd id="AN0125199443-6">Measures</hd> <hd id="AN0125199443-7">Language</hd> <p>The Clinical Evaluation of Language Fundamentals (CELF) Australian Standardised Edition was administered at 4 (CELF P2) (Wiig, Secord, & Semel, [<reflink idref="bib39" id="ref33">39</reflink>] ), 5, 7 and 11 years of age (CELF 4) (Semel, Wiig, & Secord, [<reflink idref="bib33" id="ref34">33</reflink>] ). For the statistical analyses, the CELF raw score was standardised to a z ‐ score with respect to the sample at each wave, (Mean (M) = 0; Standard Deviation (SD) = 1) to ensure consistency and ease of interpretability across waves.</p> <hd id="AN0125199443-8">Predictors</hd> <hd id="AN0125199443-9">Child factors</hd> <p>At baseline, parents reported gender, low birth weight (<2,500 g) and birth position, and from 6 years indicated whether their child had ever been diagnosed with ADHD, a learning disability or ASD. All ASD diagnoses were later validated through telephone interview with a qualified clinician experienced in ASD. At 4 years, nonverbal cognition was assessed [Kaufman Brief Intelligence Test (K ‐ BIT) (Kaufman & Kaufman, [<reflink idref="bib15" id="ref35">15</reflink>] )], screening for speech disorder was undertaken [≤10th centile (Goldman & Fristoe, [<reflink idref="bib12" id="ref36">12</reflink>] )] and parents completed the Strengths and Difficulties Questionnaire (SDQ). Clinical cut ‐ points were used to determine the presence of socioemotional problems (Goodman, [<reflink idref="bib13" id="ref37">13</reflink>] ).</p> <hd id="AN0125199443-10">Maternal and family factors</hd> <p>A range of family and maternal factors were determined by parent report at baseline including: whether languages other than English were spoken in the home (non ‐ English ‐ speaking background ‐ NESB); family history of language and/or literacy difficulties (i.e. whether the mother, father or siblings had been late to talk, had ongoing problems with speech or language, stuttered or had problems learning to read); maternal age at birth (>24 years; ≤24 years); and maternal education (completed < year 12 – the last year of formal schooling in Australia; ≥ year 12). At 4 years, parents reported whether the main language spoken to the child was not English (non ‐ English ‐ speaking background ‐ NESB).</p> <p>Socioeconomic disadvantage was calculated using baseline postcodes and the census ‐ derived SEIFA Index of Relative Disadvantage (Australian Bureau of Statistics, [<reflink idref="bib2" id="ref38">2</reflink>] ) (M = 1,000, SD = 100: a lower score representing greater disadvantage). Family literacy was derived using a composite score calculated from mothers' and fathers' Mill Hill Vocabulary Scale at child age 2 years (Raven, Court, & Raven, [<reflink idref="bib27" id="ref39">27</reflink>] ) and the Wide Range Achievement Test Reading subtest at child age 4 years (WRAT ‐ 4) (Wilkinson & Robertson, [<reflink idref="bib40" id="ref40">40</reflink>] ). Measures were each scaled to a z ‐ score, then summed and a further z ‐ score calculated from the sum.</p> <p>Home learning environment factors included the number of books in the home (at 2 years: <10; 10–20; 21–30; >30 books); the frequency the child was read to [measured at all waves from 8 months to 4 years using the Brigance Infant and Toddler Screen (BITS; Glascoe & Brigance, [<reflink idref="bib11" id="ref41">11</reflink>] ) item ‘I look at or read children's books to my child’ (not very often; sometimes; often); and average child television exposure (hours per week) measured at 4 years. To aid data analysis, quintiles were derived from a composite score of the BITS item across data waves and from the average exposure to television each week.</p> <hd id="AN0125199443-11">Support/intervention factors</hd> <p>At each wave, parents reported on any additional help sought relating to the child's speech and language in the last 12 months.</p> <hd id="AN0125199443-12">Statistical analysis</hd> <hd id="AN0125199443-13">Latent trajectories</hd> <p>For the latent class trajectory analysis, modelling was conducted on the subset of children completing at least two language assessments at 4, 5, 7 and 11 years, consisting of 1,279 children (from the total of 1,910). Using the statistical software package ‘R’ (R Core Team, [<reflink idref="bib26" id="ref42">26</reflink>] ), the ‘hlme’ function of the ‘lcmm’ package (Proust ‐ Lima, Philipps, Diakite, & Liquet, [<reflink idref="bib25" id="ref43">25</reflink>] ) was used to model the language scores across time, identifying groups of children with similar patterns or trajectories. Parameter estimates were derived using a full information maximum likelihood (FIML) estimator which is a commonly accepted way to handle missing data (Schafer & Graham, [<reflink idref="bib31" id="ref44">31</reflink>] ).</p> <p>Latent class growth modelling was completed using standardised CELF Core z ‐ score (M = 0, SD = 1) as the outcome. Preliminary analysis of the distribution of scores at each time point supported assuming the groups were normally distributed. A quadratic trend over time allowed for curvilinear trajectories. Models were run with 1, 2, 3 and 4 groups with each group allowed to have different parameters (e.g. different intercept, linear, quadratic trend and variance). We additionally examined a number of alternative modelling approaches, including random effect models (allowing individual trajectories to be more variable than the group mean) and those allowing for autocorrelation (due to the repeated nature of the measurements) (Ohlssen, Sharples, & Spiegelhalter, [<reflink idref="bib24" id="ref45">24</reflink>] ; Wraith & Wolfe, [<reflink idref="bib41" id="ref46">41</reflink>] ). Model fit statistics for the alternative modelling approaches are presented in Appendix S2.</p> <p>For further analysis, we selected the best model using both statistical goodness of fit criteria and interpretability, the latter taking into account the size of the groups, the complexity of the model and the size of the difference between the groups. To assess the statistical goodness of fit, we used estimates of the log ‐ likelihood (LR), the Akaike information criteria (AIC) and the Bayesian information criteria (BIC) (Akaike, [<reflink idref="bib1" id="ref47">1</reflink>] ; Lo, Mendell, & Rubin, [<reflink idref="bib19" id="ref48">19</reflink>] ; Schwartz, [<reflink idref="bib32" id="ref49">32</reflink>] ). Lower estimates of all these measures indicate better fitting and in the case of AIC and BIC, more parsimonious models. Following these criteria, the fixed effects model allowing for autocorrelation and including three trajectory groups was chosen. This model was then used to calculate for each participant the posterior probability of following each language trajectory and identify the most likely trajectory.</p> <hd id="AN0125199443-14">Bivariable models</hd> <p>To identify predictor variables associated with group membership, a series of bivariable multinomial logistic regressions were completed in Stata (StataCorp., [<reflink idref="bib37" id="ref50">37</reflink>] ). The results are presented as relative risks (RR) in our analyses which may be similarly interpreted to standard odds ratios in logistic regression. To account for uncertainty in group memberships, we used weighting with the posterior probabilities of group membership representing the weights (Wraith & Wolfe, [<reflink idref="bib41" id="ref51">41</reflink>] ).</p> <hd id="AN0125199443-15">Multivariable model</hd> <p>To examine the unique impact of individual risks and possible effects of accumulative risk exposures, a multivariable multinomial logistic regression analysis was conducted. Variables significant at ≤.05 level in the bivariable multinomial logistic regression analyses were included. To account for collinearity, minimise the effect of missing predictors and not overfit the model given the small sample sizes in some of the groups, a highly conservative approach was taken. First variables most likely to account for differences in trajectories were included (NESB, and the neurodevelopmental disorders: learning disability, ADHD and ASD). We then identified the minimum number of predictors which represented factors from the Child, Family, Maternal and Support/intervention categories while also considering collinearity and missing data. At each stage, the multivariable model was assessed using model fit criteria, LR tests and pseudo ‐ R<sups>2</sups> values.</p> <hd id="AN0125199443-16">Results</hd> <hd id="AN0125199443-17">Participants</hd> <p>Compared to the entire ELVS cohort (N = 1,910), participant families in the latent class analysis (N = 1,279) were more likely to have higher SEIFA, and be more highly educated and older mothers Appendix (S1).</p> <hd id="AN0125199443-18">Language trajectories</hd> <p>Figure [NaN] illustrates the individual growth trajectories and Table [NaN] presents the numbers of children within the ELVS sample classified according to their most likely trajectory group.</p> <p>Number of children by group and threshold (with percentages) and change in scores from 4 to 11 years by group</p> <p> <ephtml> <table><tr><th align="left" /><th align="char">n (%)</th><th align="char">Change in scores from 4 to 11 years M (SD)</th></tr><tr><th align="char">All waves</th><th align="char">4 years</th><th align="char">5 years</th><th align="char">7 years</th><th align="char">11 years</th></tr><tr><td align="left">Overall (total sample)</td><td align="char">1,279</td><td align="char">1,239</td><td align="char">978</td><td align="char">1,188</td><td align="char">820</td><td align="char" /></tr><tr><td align="left">Above mean</td><td align="char" /><td align="char">717 (57.9)</td><td align="char">496 (50.7)</td><td align="char">627 (52.8)</td><td align="char">413 (50.4)</td><td align="char" /></tr><tr><td align="left">Below mean</td><td align="char" /><td align="char">522 (42.1)</td><td align="char">482 (49.3)</td><td align="char">561 (47.2)</td><td align="char">407 (49.6)</td><td align="char" /></tr><tr><td align="left">1.25 SD below mean</td><td align="char" /><td align="char">112 (9.0)</td><td align="char">116 (11.9)</td><td align="char">123 (10.4)</td><td align="char">63 (7.7)</td><td align="char" /></tr><tr><td align="left">Low ‐ decreasing group</td><td align="char">50</td><td align="char">49</td><td align="char">32</td><td align="char">45</td><td align="char">31</td><td align="char">−1.51 (.76)</td></tr><tr><td align="left">Above mean</td><td align="char" /><td align="char">6 (12.2)</td><td align="char">1 (3.1)</td><td align="char">0 (0.0)</td><td align="char">0 (0.0)</td><td align="char" /></tr><tr><td align="left">Below mean</td><td align="char" /><td align="char">43 (87.8)</td><td align="char">31 (96.9)</td><td align="char">45 (100.0)</td><td align="char">31 (100.0)</td><td align="char" /></tr><tr><td align="left">1.25 SD below mean</td><td align="char" /><td align="char">25 (51.0)</td><td align="char">26 (81.3)</td><td align="char">41 (91.1)</td><td align="char">31 (100.0)</td><td align="char" /></tr><tr><td align="left">Stable for consistency with figure and text group</td><td align="char">1,199</td><td align="char">1,161</td><td align="char">928</td><td align="char">1,115</td><td align="char">773</td><td align="char">−.20 (.74)</td></tr><tr><td align="left">Above mean</td><td align="char" /><td align="char">711 (61.2)</td><td align="char">494 (53.2)</td><td align="char">618 (55.4)</td><td align="char">405 (52.4)</td><td align="char" /></tr><tr><td align="left">Below mean</td><td align="char" /><td align="char">450 (38.8)</td><td align="char">434 (46.8)</td><td align="char">497 (44.6)</td><td align="char">368 (47.6)</td><td align="char" /></tr><tr><td align="left">1.25 SD below mean</td><td align="char" /><td align="char">61 (5.3)</td><td align="char">80 (8.6)</td><td align="char">74 (6.6)</td><td align="char">32 (4.1)</td><td align="char" /></tr><tr><td align="left">Low ‐ increasing group</td><td align="char">30</td><td align="char">29</td><td align="char">18</td><td align="char">28</td><td align="char">16</td><td align="char">1.96 (.75)</td></tr><tr><td align="left">Above mean</td><td align="char" /><td align="char">0 (0.0)</td><td align="char">1 (5.6)</td><td align="char">9 (32.1)</td><td align="char">8 (50.0)</td><td align="char" /></tr><tr><td align="left">Below mean</td><td align="char" /><td align="char">29 (100.0)</td><td align="char">17 (94.4)</td><td align="char">19 (67.9)</td><td align="char">8 (50.0)</td><td align="char" /></tr><tr><td align="left">1.25 SD below mean</td><td align="char" /><td align="char">26 (89.7)</td><td align="char">10 (55.6)</td><td align="char">8 (28.6)</td><td align="char">0 (0.0)</td><td align="char" /></tr></table> </ephtml> </p> <p>Between 4 and 11 years, ~4% of children were classified as having a low ‐ decreasing trajectory and ~ 2% having a low ‐ improving trajectory. The remaining children had a stable trajectory (~94%) with language scores ranging from 2 SD below or above the mean.</p> <p>The majority of the low ‐ decreasing group and all the low ‐ increasing group had language scores below the mean at 4 years. Approximately, 50% of children in the low ‐ decreasing class had either a learning disability, ASD or ADHD diagnosis, and 50% of the low ‐ improving class were from a NESB.</p> <hd id="AN0125199443-19">Bivariable analyses</hd> <p>Tables [NaN] and [NaN] present the findings of bivariable analyses testing the association between child, family and support/intervention factors and group membership using the stable group as the reference and including those factors which reach or approach significance at the p < .05 level.</p> <p>Results of bivariable analysis: child factors</p> <p> <ephtml> <table><tr><th align="left">Variables</th><th align="char">Groups</th><th align="char">Low ‐ decreasing compared to low ‐ increasing group</th></tr><tr><th align="char">Stable group (reference)</th><th align="char">Low ‐ decreasing group</th><th align="char">Low ‐ increasing group</th></tr><tr><th align="char">n (%) or M (SD)</th><th align="char">n (%) or M (SD)</th><th align="char">RR (95% CI), p ‐ value</th><th align="char">n (%) or M (SD)</th><th align="char">RR (95% CI), p ‐ value</th><th align="char">RR (95% CI), p ‐ value</th></tr><tr><td align="left">Gender</td></tr><tr><td align="left">Female</td><td align="char">615 (51.3)</td><td align="char">20 (40.0)</td><td align="char">(base)</td><td align="char">12 (40.0)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Male</td><td align="char">584 (48.7)</td><td align="char">30 (60.0)</td><td align="char">1.39 (0.96, 2.03), p = .08</td><td align="char">18 (60.0)</td><td align="char">1.54 (0.97, 2.45), p = .07</td><td align="char">0.91 (0.51, 1.61) p = .74</td></tr><tr><td align="left">Low birth weight</td></tr><tr><td align="left">No</td><td align="char">1,140 (96.5)</td><td align="char">39 (81.3)</td><td align="char">(base)</td><td align="char">28 (96.5)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">41 (3.5)</td><td align="char">9 (18.7)</td><td align="char">4.05 (1.97, 8.34), p < .001</td><td align="char">1 (3.5)</td><td align="char">0.78 (0.19, 3.24), p = .74</td><td align="char">5.18 (1.12, 24.02), p = .04</td></tr><tr><td align="left">Nonverbal cognition</td><td align="char">0.12 (0.91)</td><td align="char">−0.92 (1.32)</td><td align="char">0.48 (0.39, 0.58), p < .001</td><td align="char">−0.81 (1.27)</td><td align="char">0.53 (0.43, 0.66), p < .001</td><td align="char">0.90 (0.70, 1.17), p = .44</td></tr><tr><td align="left">Speech disorder</td></tr><tr><td align="left">>11th centile</td><td align="char">1,099 (94.9)</td><td align="char">37 (80.4)</td><td align="char">(base)</td><td align="char">25 (89.3)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">10th centile or less</td><td align="char">59 (5.1)</td><td align="char">9 (19.6)</td><td align="char">3.22 (1.73, 6.00), p < .001</td><td align="char">3 (10.7)</td><td align="char">2.35 (1.13, 4.86), p = .02</td><td align="char">1.37 (0.57, 3.33), p = .49</td></tr><tr><td align="left">Socioemotional problems</td></tr><tr><td align="left">Peer problems</td></tr><tr><td align="left">No</td><td align="char">1,039 (91.1)</td><td align="char">34 (73.9)</td><td align="char">(base)</td><td align="char">20 (74.1)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">101 (8.9)</td><td align="char">12 (26.1)</td><td align="char">2.90 (1.70, 4.97), p < .001</td><td align="char">7 (25.9)</td><td align="char">3.35 (1.81, 6.22), p < .001</td><td align="char">0.87 (0.41, 1.87), p = 0.72</td></tr><tr><td align="left">Emotional problems</td></tr><tr><td align="left">No</td><td align="char">1,077 (94.5)</td><td align="char">39 (84.8)</td><td align="char">(base)</td><td align="char">26 (96.3)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">63 (5.5)</td><td align="char">7 (15.2)</td><td align="char">2.20 (1.13, 4.29), p = .02</td><td align="char">1 (3.7)</td><td align="char">0.65 (0.23, 1.89), p = .43</td><td align="char">3.36 (1.01, 11.20), p = .05</td></tr><tr><td align="left">Conduct problems</td></tr><tr><td align="left">No</td><td align="char">1,022 (89.7)</td><td align="char">35 (76.1)</td><td align="char">(base)</td><td align="char">25 (92.6)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">118 (10.4)</td><td align="char">11 (23.9)</td><td align="char">2.14 (1.28, 3.57), p = .004</td><td align="char">2 (7.4)</td><td align="char">1.12 (0.53, 2.35), p = .77</td><td align="char">1.91 (0.81, 4.54), p = .14</td></tr><tr><td align="left">Inattention/hyperactivity problems</td></tr><tr><td align="left">No</td><td align="char">1,043 (91.5)</td><td align="char">32 (69.6)</td><td align="char">(base)</td><td align="char">27 (100.0)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">97 (8.5)</td><td align="char">14 (30.4)</td><td align="char">3.45 (2.04, 5.82), p < .001</td><td align="char">0 (0.0)</td><td align="char">0.73 (0.36, 1.47), p = .38</td><td align="char">4.71 (2.07, 10.70), p < .001</td></tr><tr><td align="left">Neurodevelopmental diagnoses</td></tr><tr><td align="left">ADHD</td></tr><tr><td align="left">No</td><td align="char">1,137 (98.8)</td><td align="char">38 (84.4)</td><td align="char">(base)</td><td align="char">27 (100.0)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">14 (1.2)</td><td align="char">7 (15.6)</td><td align="char">8.78 (3.66, 21.06), p < .001</td><td align="char">0 (0.0)</td><td align="char">1.21 (0.23, 6.37), p = .83</td><td align="char">7.29 (1.27, 41.79), p = .03</td></tr><tr><td align="left">Learning disability diagnosis</td></tr><tr><td align="left">No</td><td align="char">1,083 (95.0)</td><td align="char">21 (47.7)</td><td align="char">(base)</td><td align="char">23 (88.5)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">57 (5.0)</td><td align="char">23 (52.3)</td><td align="char">12.96 (7.99, 21.03), p < .001</td><td align="char">3 (11.5)</td><td align="char">2.66 (1.26, 5.63), p = .01</td><td align="char">4.87 (2.16, 10.98), p < .001</td></tr><tr><td align="left">Autism</td></tr><tr><td align="left">No</td><td align="char">1,175 (98.0)</td><td align="char">42 (84.0)</td><td align="char">(base)</td><td align="char">29 (96.7)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">24 (2.0)</td><td align="char">8 (16.0)</td><td align="char">9.33 (3.96, 21.98), p < .001</td><td align="char">1 (3.3)</td><td align="char">1.69 (0.22, 12.91), p = .61</td><td align="char">2.58 (0.80, 8.34), p = .11</td></tr></table> </ephtml> </p> <p>1 Stable group was the reference group against which the low ‐ decreasing and low ‐ increasing groups were compared in the weighted (multinomial) logistic regression expressed as relative risks (RR).</p> <p>2 Standardised (M = 0, SD = 1).</p> <p>Results of bivariable analysis: family and support/intervention factors</p> <p> <ephtml> <table><tr><th align="left">Variable</th><th align="char">Classes</th><th align="char">Low ‐ decreasing compared to low ‐ increasing group</th></tr><tr><th align="char">Stable group (ref)</th><th align="char">Low ‐ decreasing group</th><th align="char">Low ‐ increasing group</th></tr><tr><th align="char">n (%) or M (SD)</th><th align="char">n (%) or M (SD)</th><th align="char">RR (95% CI), p ‐ value</th><th align="char">n (%) or M (SD)</th><th align="char">RR (95% CI), p ‐ value</th><th align="char">RR (95% CI), p ‐ value</th></tr><tr><td align="left">NESB</td></tr><tr><td align="left">No</td><td align="char">1,121 (98.7)</td><td align="char">44 (95.7)</td><td align="char">(base)</td><td align="char">13 (48.2)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">15 (1.3)</td><td align="char">2 (4.3)</td><td align="char">3.14 (0.96, 10.23), p = .06</td><td align="char">14 (51.9)</td><td align="char">41.25 (20.10, 84.77), p < .001</td><td align="char">0.08 (0.02, 0.26), p < .001</td></tr><tr><td align="left">Social disadvantage</td><td align="char">1,043.24 (53.64)</td><td align="char">1,010.49 (71.43)</td><td align="char">0.99 (0.98, 0.99), p < .001</td><td align="char">995.23 (75.91)</td><td align="char">0.99 (0.98, 0.99), p < .001</td><td align="char">1.00 (1.00, 1.01), p = .33</td></tr><tr><td align="left">Family history</td></tr><tr><td align="left">No</td><td align="char">906 (75.6)</td><td align="char">35 (70.0)</td><td align="char">(base)</td><td align="char">18 (60.0)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">293 (24.4)</td><td align="char">15 (30.0)</td><td align="char">1.50 (1.00, 2.25), p = .05</td><td align="char">12 (40.0)</td><td align="char">1.75 (1.07, 2.85), p = .03</td><td align="char">0.86 (0.47, 1.58), p = .62</td></tr><tr><td align="left">Family literacy</td><td align="char">0.19 (0.87)</td><td align="char">−0.36 (0.90)</td><td align="char">0.57 (0.47, 0.70), p < .001</td><td align="char">−0.83 (1.39)</td><td align="char">0.43 (0.32, 0.58), p < .001</td><td align="char">1.33 (0.95, 1.87), p = .10</td></tr><tr><td align="left">Home learning environment</td></tr><tr><td align="left">Books in the home</td></tr><tr><td align="left">More than 30 books</td><td align="char">824 (71.3)</td><td align="char">20 (43.5)</td><td align="char">(base)</td><td align="char">10 (35.7)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">21–30</td><td align="char">180 (15.6)</td><td align="char">11 (23.9)</td><td align="char">1.95 (1.17, 3.24), p = .01</td><td align="char">4 (14.3)</td><td align="char">2.01 (1.09, 3.70), p = .03</td><td align="char">0.97 (0.45, 2.08), p = .94</td></tr><tr><td align="left">10–20</td><td align="char">135 (11.7)</td><td align="char">13 (28.3)</td><td align="char">2.72 (1.64, 4.53), p < .001</td><td align="char">8 (28.6)</td><td align="char">3.29 (1.74, 6.23), p < .001</td><td align="char">0.83 (0.38, 1.80), p = .63</td></tr><tr><td align="left">Less than 10</td><td align="char">16 (1.4)</td><td align="char">2 (4.4)</td><td align="char">2.62 (0.69, 9.94), p = .16</td><td align="char">6 (21.4)</td><td align="char">15.44 (6.30, 37.85), p < .001</td><td align="char">0.17 (0.04, 0.74), p = .02</td></tr><tr><td align="left">Frequency child read to</td></tr><tr><td align="left">2</td><td align="char">282 (26.7)</td><td align="char">7 (18.4)</td><td align="char">(base)</td><td align="char">1 (4.4)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">3</td><td align="char">211 (20.2)</td><td align="char">4 (10.5)</td><td align="char">0.89 (0.44, 1.78), p = .74</td><td align="char">2 (8.7)</td><td align="char">1.43 (0.55, 3.70), p = .46</td><td align="char">0.62 (0.20, 1.96), p = .42</td></tr><tr><td align="left">4</td><td align="char">341 (32.6)</td><td align="char">11 (29.0)</td><td align="char">1.22 (0.68, 2.21), p = .51</td><td align="char">8 (34.8)</td><td align="char">2.93 (1.37, 6.29), p = .006</td><td align="char">0.42 (0.16, 1.06), p = .07</td></tr><tr><td align="left">Low</td><td align="char">211 (20.2)</td><td align="char">16 (42.1)</td><td align="char">2.60 (1.41, 4.78), p = .002</td><td align="char">12 (52.2)</td><td align="char">6.23 (2.90, 13.37), p < .001</td><td align="char">0.42 (0.16, 1.07), p = .07</td></tr><tr><td align="left">TV watching/week</td></tr><tr><td align="left">High</td><td align="char">178 (15.9)</td><td align="char">13 (29.6)</td><td align="char">(base)</td><td align="char">5 (20.0)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">2</td><td align="char">94 (8.4)</td><td align="char">5 (9.1)</td><td align="char">0.65 (0.30, 1.41), p = .27</td><td align="char">0</td><td align="char">0.37 (0.15, 0.90), p = .03</td><td align="char">1.77 (0.57, 5.49), p = .32</td></tr><tr><td align="left">3</td><td align="char">176 (15.7)</td><td align="char">4 (36.4)</td><td align="char">0.45 (0.24, 0.84), p = .01</td><td align="char">5 (20.0)</td><td align="char">1.11 (0.51, 2.40), p = .79</td><td align="char">0.41 (0.16, 1.05), p = .06</td></tr><tr><td align="left">4</td><td align="char">425 (38.0)</td><td align="char">16 (36.4)</td><td align="char">0.52 (0.31, 0.87), p = .01</td><td align="char">12 (48.0)</td><td align="char">0.81 (0.40, 1.62), p = .55</td><td align="char">0.64 (0.28, 1.47), p = .30</td></tr><tr><td align="left">Low</td><td align="char">246 (22.0)</td><td align="char">6 (13.6)</td><td align="char">0.32 (0.17, 0.63), p = .001</td><td align="char">3 (12.0)</td><td align="char">0.44 (0.20, 0.97), p = .04</td><td align="char">0.73 (0.27, 1.99), p = .54</td></tr><tr><td align="left">Maternal factors</td></tr><tr><td align="left">Maternal education</td></tr><tr><td align="left">>year 12</td><td align="char">957 (80.0)</td><td align="char">35 (71.4)</td><td align="char">(base)</td><td align="char">23 (76.7)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">≤ year 12</td><td align="char">240 (20.1)</td><td align="char">14 (28.6)</td><td align="char">1.62 (1.06, 2.48), p = .03</td><td align="char">7 (23.3)</td><td align="char">1.14 (0.68, 1.92), p = .62</td><td align="char">1.42 (0.75, 2.70), p = .29</td></tr><tr><td align="left">Young Mum</td></tr><tr><td align="left">Age >24 years</td><td align="char">1,153 (96.4)</td><td align="char">45 (90.0)</td><td align="char">(base)</td><td align="char">27 (90.0)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Age ≤24 years</td><td align="char">43 (3.60)</td><td align="char">5 (10.0)</td><td align="char">2.11 (0.88, 5.04), p = .10</td><td align="char">3 (10.0)</td><td align="char">2.80 (1.23, 6.38), p = .01</td><td align="char">0.75 (0.24, 2.31), p = .62</td></tr><tr><td align="left">Sought help last 12 months</td></tr><tr><td align="left">4 years</td></tr><tr><td align="left">No</td><td align="char">952 (84.0)</td><td align="char">34 (73.9)</td><td align="char">(base)</td><td align="char">23 (85.2)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">181 (15.9)</td><td align="char">12 (26.1)</td><td align="char">2.03 (1.29, 3.18), p = .002</td><td align="char">4 (14.8)</td><td align="char">1.24 (0.71, 2.17), p = .45</td><td align="char">1.64 (0.83, 3.24), p = .16</td></tr><tr><td align="left">6 years</td></tr><tr><td align="left">No</td><td align="char">815 (88.4)</td><td align="c har">13 (35.1)</td><td align="char">(base)</td><td align="char">14 (73.7)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">107 (11.6)</td><td align="char">24 (64.9)</td><td align="char">8.54 (5.38, 13.56), p < .001</td><td align="char">5 (26.3)</td><td align="char">2.17 (1.04, 4.54), p = .04</td><td align="char">3.93 (1.72, 8.94), p = .001</td></tr><tr><td align="left">9 years</td></tr><tr><td align="left">No</td><td align="char">918 (92.0)</td><td align="char">22 (57.9)</td><td align="char">(base)</td><td align="char">22 (95.7)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">80 (8.0)</td><td align="char">16 (42.1)</td><td align="char">7.16 (4.37, 11.76), p < .001</td><td align="char">1 (4.3)</td><td align="char">1.11 (0.48, 2.59), p = .80</td><td align="char">6.44 (2.55, 16.24), p < .001</td></tr><tr><td align="left">11 years</td></tr><tr><td align="left">No</td><td align="char">740 (95.2)</td><td align="char">15 (48.4)</td><td align="char">(base)</td><td align="char">14 (87.5)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">Yes</td><td align="char">37 (4.8)</td><td align="char">16 (51.6)</td><td align="char">18.51 (10.09, 33.96), p < .001</td><td align="char">2 (12.5)</td><td align="char">2.72 (.94, 7.88), p = .07</td><td align="char">6.80 (2.18, 21.16), p = .001</td></tr></table> </ephtml> </p> <ulist> <item>3 NESB, non ‐ English ‐ speaking background.</item> <item>4 Stable group was the reference group against which the low ‐ decreasing and low ‐ increasing groups were compared in the weighted (multinomial) logistic regression expressed as relative risks (RR).</item> <item>5 Measured using SEIFA = socioeconomic index for areas (M = 1,000, SD = 100).</item> <item>6 Standardised M = 0; SD  1.</item> <item>7 Quintiles.</item> </ulist> <p>A large number of factors placed children at increased risk of being in either the low ‐ decreasing or low ‐ increasing group rather than the stable group. These included the child factors – speech disorder, peer problems, learning disability diagnosis and lower nonverbal cognitive score; the family factors – family history of language difficulties, lower family literacy and SEIFA, 10–30 books in the home (relative to having >30); higher average hours of TV viewing per week; and seeking additional support at 6 years. For these factors, RR were usually similar across the two groups except in the case of learning disability diagnosis (low ‐ decreasing RR = 12.96, p < .001; low ‐ increasing RR = 2.66, p = .01) and help ‐ seeking (low ‐ decreasing RR = 8.54, p < .001; low ‐ increasing RR = 2.17, p = .04). Factors associated only with an increased risk of low ‐ decreasing group membership were low birth weight, emotional problems, conduct problems, inattention and hyperactivity, ADHD or ASD diagnosis, and seeking help for the child's difficulties at ages 4, 9 and 11 years. Factors associated only with an increased risk of low ‐ increasing group were NESB (RR = 41.25, p < .001), a younger mother and <10 children's books in the home (RR = 15.44, p < .001).</p> <hd id="AN0125199443-20">Multivariable analysis</hd> <p>A multivariable analysis examined the unique impact and possible effects of cumulative exposures of individual factors on group membership. Analyses should be interpreted with caution given small group sizes (Table [NaN] ). Children were at increased risk of being members of the low ‐ decreasing group if they had lower nonverbal cognition, low birth weight, a learning disability diagnosis, lower family literacy, 10–20 children's books in the home and if parents had sought additional support (age 6 years). Children were at increased risk of being in the low ‐ increasing group if they were NESB, had lower nonverbal cognition and SEIFA scores, <10 children's books in the home and were not of low birth weight.</p> <p>Results of multivariable model</p> <p> <ephtml> <table><tr><th align="left">Variables</th><th align="char">Groups</th><th align="char">Low ‐ decreasing compared to low ‐ increasing group</th></tr><tr><th align="char">Low ‐ decreasing group</th><th align="char">Low ‐ increasing group</th></tr><tr><th align="char">RR (95% CI), p ‐ value</th><th align="char">RR (95% CI), p ‐ value</th><th align="char">RR (95% CI), p ‐ value</th></tr><tr><td align="left">Child factors</td></tr><tr><td align="left">Nonverbal cognition</td><td align="char">0.60 (0.46, 0.77), p < .001</td><td align="char">0.61 (0.45, 0.82), p = .001</td><td align="char">0.98 (0.71, 1.37), p = .91</td></tr><tr><td align="left">Low birth weight</td><td align="char">2.98 (1.23, 7.22), p = .02</td><td align="char">0.09 (0.03, 0.26), p < .001</td><td align="char">34.67 (9.21, 130.49), p < .001</td></tr><tr><td align="left">Neurodevelopmental diagnosis</td></tr><tr><td align="left">ADHD</td><td align="char">2.07 (0.75, 5.71), p = .16</td><td align="char">0.74 (0.09, 6.29), p = .78</td><td align="char">2.82 (0.36, 21.73), p = .32</td></tr><tr><td align="left">Autism</td><td align="char">1.00 (0.31, 3.24), p = .99</td><td align="char">1.15 (0.19, 7.08), p = .88</td><td align="char">0.87 (0.12, 6.30), p = .89</td></tr><tr><td align="left">Learning disability diagnosis</td><td align="char">2.83 (1.35, 5.92), p = .006</td><td align="char">2.46 (0.52, 11.61), p = .26</td><td align="char">1.15 (0.23, 5.69), p = .86</td></tr><tr><td align="left">Family factors</td></tr><tr><td align="left">NESB</td><td align="char">0.36 (0.06, 1.96), p = .24</td><td align="char">43.42 (14.68, 128.45), p < .001</td><td align="char">0.01 (0.00, 0.05), p < .001</td></tr><tr><td align="left">Social disadvantage</td><td align="char">1.00 (0.99, 1.00), p = .19</td><td align="char">0.99 (0.99, 1.00), p = .02</td><td align="char">1.00 (1.00, 1.01), p = .27</td></tr><tr><td align="left">Family literacy</td><td align="char">0.73 (0.54, 0.98), p = .03</td><td align="char">1.01 (0.69, 1.48), p = .95</td><td align="char">0.72 (0.45, 1.14), p = .16</td></tr><tr><td align="left">Home learning environment</td></tr><tr><td align="left">Books in the home</td></tr><tr><td align="left">More than 30 books</td><td align="char">(base)</td><td align="char">(base)</td><td align="char">(base)</td></tr><tr><td align="left">21–30</td><td align="char">1.30 (0.68, 2.49), p = .44</td><td align="char">2.13 (0.96, 4.70), p = .06</td><td align="char">0.61 (0.23, 1.63), p = .32</td></tr><tr><td align="left">10–20</td><td align="char">2.37 (1.19, 4.71), p = .01</td><td align="char">3.15 (1.28, 7.77), p = .01</td><td align="char">0.75 (0.26, 2.21), p = .61</td></tr><tr><td align="left">Less than 10</td><td align="char">0.62 (0.18, 2.20), p = .46</td><td align="char">5.75 (1.55, 21.37), p = .009</td><td align="char">0.11 (0.02, 0.54), p = .007</td></tr><tr><td align="left">Support/intervention factors</td></tr><tr><td align="left">Seeking help/extra support in last 12 months (6 years)</td><td align="char">2.99 (1.59,5.62), p = .001</td><td align="char">1.32 (0.42, 4.15), p = .64</td><td align="char">2.27 (0.65, 7.89), p = .20</td></tr></table> </ephtml> </p> <ulist> <item>8 Stable group was the reference group against which the low ‐ decreasing and low ‐ increasing groups were compared in the weighted (multinomial) logistic regression expressed as relative risks (RR).</item> <item>9 Standardised M = 0; SD = 1.</item> <item>10 Measured using SEIFA = socioeconomic index for areas (M = 1,000, SD = 100).</item> </ulist> <hd id="AN0125199443-21">Discussion</hd> <p>This study applied longitudinal trajectory latent class modelling in a community sample with repeated direct testing of children's language to identify subgroups in trajectories of language development from 4 to 11 years across the full range of ability. Three groups were identified: a large ‘stable’ group with wide ranging but relatively stable language ability which included the majority of children (94%), 5% of whom had language abilities falling >1.25 SD below the mean; a ‘low ‐ decreasing’ group; and a ‘low ‐ increasing’ group. Of significant concern was the small group of children following a low ‐ decreasing trajectory, starting with below average language abilities at 4 years and falling substantially over time so that all children in the group experienced severe language difficulties by age 11. Over the course of the study, approximately half of this group received a diagnosis of learning disability, ASD or ADHD. In contrast by 11 years, all of the children in the low ‐ increasing group (the smallest group) had language scores within the typical range, and by 7 years were indistinguishable from the stable group (McKean et al., [<reflink idref="bib20" id="ref52">20</reflink>] ). Around half were from a NESB providing further support for the argument that children from NESB require prolonged exposure to the language of instruction in preschool and school to consolidate skills in both languages.</p> <p>In terms of identifying which children are likely to have enduring language problems, these data suggest that the relative position in language ability of most children is established by 4 years of age: those with low language at 4 years are likely to stay low to 11 years. The clear exception was children from a NESB who were likely to catch up with their peers by 7 years. Although the mean trajectory in the ‘stable’ trajectory group was flat, a small degree of variability in rate of progress was present such that children would continue to move above and below any given cut ‐ point over time (McKean et al., [<reflink idref="bib21" id="ref53">21</reflink>] ). Indeed ~ 22% of the ‘stable’ group had a difference in score from 4 to 11 years of >.75 SD; a meaningful change in relative ability.</p> <p>Early identification of the vulnerable children in the low ‐ decreasing group would be beneficial for children and families, enabling access to earlier intervention and educational support. Children following this low ‐ decreasing trajectory were more likely to have socioemotional and behavioural problems, lower family literacy and be of low birth weight. These may be important ‘signals of risk’ for children presenting with mild ‐ moderate language difficulties at 4 years, indicating the need for monitoring, preventative interventions and multidisciplinary assessment. Especially given that only half this group received a neurodevelopmental diagnosis over the course of the study, many not doing so until 7 years or older. Targeting interventions should be guided by cumulative risk models based on child and family factors identified as important to prognosis. These factors are considered when children present to specialist services; however, many children with low language do not (Morgan et al., [<reflink idref="bib22" id="ref54">22</reflink>] ; Skeat et al., [<reflink idref="bib34" id="ref55">34</reflink>] ). The application of cumulative risk models to targeting in communities ‘at risk’ of both language difficulties and limited access to services should be considered.</p> <p>Despite the relatively large study sample, two subgroups contained small numbers. Combined with missing data for some predictors, this limited our ability to build a comprehensive multivariable risk model and so the findings regarding predictors of group membership should be interpreted with caution. Taking the bivariable and multivariable analyses together, there is tentative evidence to suggest that the low ‐ decreasing group was associated with biological risks (i.e. low birth weight; lower family literacy; and neurodevelopmental diagnoses) and the low ‐ increasing group with environmental factors (i.e. NESB; young mother; few children's books in the home; lower SEIFA scores) (Snowling et al., [<reflink idref="bib36" id="ref56">36</reflink>] ; Zambrana et al., [<reflink idref="bib42" id="ref57">42</reflink>] ). Larger samples and/or meta ‐ analyses are likely to be required to yield sufficient power to test these findings and those of previous studies.</p> <p>As no previous studies have attempted to define subgroups in longitudinal trajectory across a community sample of school ‐ age children, the approach taken to the identification of subgroups was exploratory. Replication in other samples is required to determine whether similar trajectory groups exist in different populations.</p> <hd id="AN0125199443-22">Conclusion</hd> <p>For most children, individual differences in relative language ability are established before 4 years. Those factors which drive individual differences would appear to exert their influences early or continue to act across development, maintaining children's relative position. By 4 years, services can be confident that children with low language will remain low over the primary years. However, using rigid cut ‐ points in language ability to determine eligibility to access support is not recommended due to continued individual variability. Our findings suggest that service delivery models should incorporate monitoring over time, targeting according to both language abilities and associated risks and delivery of a continuum of interventions across a continuum of need.</p> <hd id="AN0125199443-23">Acknowledgements</hd> <p>The authors sincerely thank the participating parents and children and acknowledge the contribution of the Victorian Maternal and Child Health nurses who supported recruitment. This study was supported by the Australian National Health and Medical Research Council (NHMRC grants: 237106, 9436958, 1041947 and 1023493). NHMRC fellowships supported C.McK. (Centre of Research Excellence in Child Language 1023493), F.M. (1037449, 1111160), D.W. (1035261) and S.R. (491210). Research at the Murdoch Childrens Research Institute is supported by the Victorian Government's Operational Infrastructure Support Program. The authors have declared that they have no competing or potential conflicts of interest.</p> <p>Key points</p> <p>There is considerable instability in child language profiles over development. This hinders the strategic targeting and design of interventions. There is emerging evidence to suggest differing child, family and societal factors may be associated with differing language trajectories.</p> <p>Three language trajectory groups were identified: a ‘stable’ trajectory (94% of participants), a low ‐ decreasing trajectory (4%) and a low ‐ improving trajectory (2%).</p> <p>A very vulnerable low ‐ declining group was associated with low birth weight, socioemotional and behavioural problems, and lower family literacy.</p> <p>By 4 years of age, services can be confident that most children with low language abilities will remain low over the primary years. However, using rigid cut ‐ points in language ability to target interventions is not recommended due to continued individual variability in rates of language development.</p> <p>Service delivery models should incorporate monitoring over time, targeting according to both language abilities and associated risks and delivery of a continuum of interventions across a continuum of need.</p> <ref id="AN0125199443-24"> <title>Footnotes</title> <blist> <bibl id="bib1" idref="ref47" type="bt">1</bibl> <bibtext>Conflict of interest statement: No conflicts declared. </bibtext> </blist> </ref> <ref id="AN0125199443-25"> <title>References</title> <blist> <bibtext>Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19, 716 – 723. </bibtext> </blist> <blist> <bibl id="bib2" idref="ref38" type="bt">2</bibl> <bibtext>Australian Bureau of Statistics (2001). Socio ‐ economic indexes for areas. Canberra : Author. </bibtext> </blist> <blist> <bibl id="bib3" idref="ref24" type="bt">3</bibl> <bibtext>Beitchman, J.H., Jiang, H., Koyama, E., Johnson, C.J., Escobar, M., Atkinson, L., … & Vida, R. (2008). Models and determinants of vocabulary growth from kindergarten to adulthood. Journal of Child Psychology and Psychiatry, 49, 626 – 634. </bibtext> </blist> <blist> <bibl id="bib4" idref="ref17" type="bt">4</bibl> <bibtext>Beitchman, J.H., Wilson, B., Brownlie, E.B., Walters, H., Inglis, A., & Lancee, W. (1996). Long ‐ term consistency in speech/language profiles: II. Behavioral, emotional, and social outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 815 – 825. </bibtext> </blist> <blist> <bibl id="bib5" idref="ref18" type="bt">5</bibl> <bibtext>Bishop, D.V.M., & Edmundson, A. (1987). Language ‐ impaired 4 ‐ year ‐ olds: Distinguishing transient from persistent impairment. Journal of Speech and Hearing Disorders, 52, 156 – 173. </bibtext> </blist> <blist> <bibl id="bib6" idref="ref15" type="bt">6</bibl> <bibtext>Bishop, D.V.M., Snowling, M.J., Thompson, P.A., Greenhalgh, T., & the CATALISE ‐ 2 consortium. (2017). Phase 2 of CATALISE: a multinational and multidisciplinary Delphi consensus study of problems with language development: Terminology. Journal of Child Psychology and Psychiatry, 58, 1068 – 1080. </bibtext> </blist> <blist> <bibl id="bib7" idref="ref4" type="bt">7</bibl> <bibtext>Bornstein, M.H., Hahn, C.S., & Putnick, D.L. (2016). Long ‐ term stability of core language skill in children with contrasting language skills. Developmental Psychology, 52, 704 – 716. </bibtext> </blist> <blist> <bibl id="bib8" idref="ref1" type="bt">8</bibl> <bibtext>Conti ‐ Ramsden, G., & Botting, N. (2008). Emotional health in adolescents with and without a history of specific language impairment (SLI). Journal Of Child Psychology And Psychiatry and Allied Disciplines, 49, 516 – 525. </bibtext> </blist> <blist> <bibl id="bib9" idref="ref16" type="bt">9</bibl> <bibtext>Conti ‐ Ramsden, G., St. Clair, M.C., Pickles, A., & Durkin, K. (2012). Developmental trajectories of verbal and nonverbal skills in individuals with a history of specific language impairment: From childhood to adolescence. Journal of Speech Language and Hearing Research, 55, 1716 – 1735. </bibtext> </blist> <blist> <bibl id="bib10" idref="ref5" type="bt">10</bibl> <bibtext>Ghassabian, A., Rescorla, L., Henrichs, J., Jaddoe, V.W., Verhulst, F.C., & Tiemeier, H. (2014). Early lexical development and risk of verbal and nonverbal cognitive delay at school ‐ age. Acta Paediatrica, 103, 70 – 80. </bibtext> </blist> <blist> <bibl id="bib11" idref="ref41" type="bt">11</bibl> <bibtext>Glascoe, F.P., & Brigance, A.H. (2002). Brigance infant and toddler screen: Parent ‐ child interactions form. North Billerica, MA : Curriculum. </bibtext> </blist> <blist> <bibl id="bib12" idref="ref36" type="bt">12</bibl> <bibtext>Goldman, R., & Fristoe, M. (2000). Goldman Fristoe 2: Test of Articulation (2nd edn). Circle Pines, MN : American Guidance Service. </bibtext> </blist> <blist> <bibl id="bib13" idref="ref37" type="bt">13</bibl> <bibtext>Goodman, R. (1997). The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581 – 586. </bibtext> </blist> <blist> <bibl id="bib14" idref="ref6" type="bt">14</bibl> <bibtext>Henrichs, J., Rescorla, L., Schenk, J.J., Schmidt, H.G., Jaddoe, V.W.V., Hofman, A., … & Tiemeier, H. (2011). Examining continuity of early expressive vocabulary development: The generation R study. Journal of Speech Language and Hearing Research, 54, 854 – 869. </bibtext> </blist> <blist> <bibl id="bib15" idref="ref35" type="bt">15</bibl> <bibtext>Kaufman, A.S., & Kaufman, N.L. (2004). Kaufman brief intelligence test, second edition (KBIT ‐ 2). Bloomington, MN : Pearson. </bibtext> </blist> <blist> <bibl id="bib16" idref="ref19" type="bt">16</bibl> <bibtext>Law, J., Rush, R., Anandan, C., Cox, M., & Wood, R. (2012). Predicting language change between 3 and 5 Years and its implications for early identification. Pediatrics, 130, 132 – 137. </bibtext> </blist> <blist> <bibl id="bib17" idref="ref3" type="bt">17</bibl> <bibtext>Law, J., Rush, R., Schoon, I., & Parsons, S. (2009). Modeling developmental language difficulties from school entry into adulthood: Literacy, mental health, and employment outcomes. Journal of Speech Language and Hearing Research, 52, 1401 – 1416. </bibtext> </blist> <blist> <bibl id="bib18" idref="ref23" type="bt">18</bibl> <bibtext>Law, J., Tomblin, J.B., & Zhang, X. (2008). Characterizing the growth trajectories of language ‐ impaired children between 7 and 11 years of age. Journal of Speech Language and Hearing Research, 51, 739 – 749. </bibtext> </blist> <blist> <bibl id="bib19" idref="ref48" type="bt">19</bibl> <bibtext>Lo, Y., Mendell, N.R., & Rubin, D.B. (2001). Testing the number of components in a normal mixture. Biometrika, 88, 767 – 778. </bibtext> </blist> <blist> <bibl id="bib20" idref="ref26" type="bt">20</bibl> <bibtext>McKean, C., Mensah, F.K., Eadie, P., Bavin, E.L., Bretherton, L., Cini, E., & Reilly, S. (2015). Levers for language growth: Characteristics and predictors of language trajectories between 4 and 7 years. PLoS ONE, 10, e0134251. </bibtext> </blist> <blist> <bibl id="bib21" idref="ref13" type="bt">21</bibl> <bibtext>McKean, C., Reilly, S., Bavin, E.L., Bretherton, L., Cini, E., Conway, L., … & Mensah, F. (2017). Language outcomes at 7 years: Early predictors and co ‐ occurring difficulties. Pediatrics, 139, e20161684. </bibtext> </blist> <blist> <bibl id="bib22" idref="ref54" type="bt">22</bibl> <bibtext>Morgan, P.L., Hammer, C.S., Farkas, G., Hillemeier, M.M., Maczuga, S., Cook, M., & Morano, S. (2016). Who receives speech/language pathology services by 5 years of age in the United States? American Journal of Speech ‐ Language Pathology, 25, 183 – 199. </bibtext> </blist> <blist> <bibl id="bib23" idref="ref11" type="bt">23</bibl> <bibtext>Norbury, C.F., Gooch, G., Wray, C., Baird, G., Charman, T., Simonoff, E., … & Pickles, A. (2016). The impact of nonverbal ability on prevalence and clinical presentation of language disorder: Evidence from a population study. Journal of Child Psychology and Psychiatry, 57, 1247 – 1257. </bibtext> </blist> <blist> <bibl id="bib24" idref="ref45" type="bt">24</bibl> <bibtext>Ohlssen, D.I., Sharples, L.D., & Spiegelhalter, D.J. (2007). Flexible random ‐ effects models using Bayesian semi ‐ parametric models: Applications to institutional comparisons. Statistics in Medicine, 26, 2088 – 2112. </bibtext> </blist> <blist> <bibl id="bib25" idref="ref43" type="bt">25</bibl> <bibtext>Proust ‐ Lima, C., Philipps, V., Diakite, A., & Liquet, B. (2015). lcmm: Extended Mixed Models Using Latent Classes and Latent Processes. R package version 1.7.2. Available from: <ulink href="http://CRAN.R-project.org/package=lcmm">http://CRAN.R-project.org/package=lcmm</ulink> [last accessed 12 February 2016]. </bibtext> </blist> <blist> <bibl id="bib26" idref="ref42" type="bt">26</bibl> <bibtext>R Core Team. (2014). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Available from: <ulink href="http://www.R-project.org/">http://www.R-project.org/</ulink> [last accessed 23 January 2016]. </bibtext> </blist> <blist> <bibl id="bib27" idref="ref39" type="bt">27</bibl> <bibtext>Raven, J.C., Court, J.H., & Raven, J. (1998). Mill Hill Vocabulary Scale. Oxford : Oxford Pyschology Press. </bibtext> </blist> <blist> <bibl id="bib28" idref="ref32" type="bt">28</bibl> <bibtext>Reilly, S., Eadie, P., Bavin, E.L., Wake, M., Prior, M., Williams, J., … & Ukoumunne, O.C. (2006). Growth of infant communication between 8 and 12 months: A population study. Journal of Paediatrics and Child Health, 42, 764 – 770. </bibtext> </blist> <blist> <bibl id="bib29" idref="ref7" type="bt">29</bibl> <bibtext>Reilly, S., Wake, M., Ukoumunne, O.C., Bavin, E., Prior, M., Cini, E., … & Bretherton, L. (2010). Predicting language outcomes at 4 years of age: Findings from the Early Language in Victoria Study. Pediatrics, 126, e1530 – e1537. </bibtext> </blist> <blist> <bibl id="bib30" idref="ref25" type="bt">30</bibl> <bibtext>Rice, M.L., & Hoffman, L. (2015). Predicting vocabulary growth in children with and without specific language impairment: A longitudinal study from 2;6 to 21 years of age. Journal of Speech Language and Hearing Research, 58, 345 – 359. </bibtext> </blist> <blist> <bibl id="bib31" idref="ref44" type="bt">31</bibl> <bibtext>Schafer, J.L., & Graham, J.W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7, 147 – 177. </bibtext> </blist> <blist> <bibl id="bib32" idref="ref49" type="bt">32</bibl> <bibtext>Schwartz, G.E. (1978). Estimating the dimension of a model. Annals of Statistics, 6, 461 – 464. </bibtext> </blist> <blist> <bibl id="bib33" idref="ref34" type="bt">33</bibl> <bibtext>Semel, E., Wiig, E.H., & Secord, W.A. (2006). Clinical evaluation of language fundamentals ‐ fourth edition, Australian standardised edition (4th edn). Sydney, NSW : Psychological Corporation. </bibtext> </blist> <blist> <bibl id="bib34" idref="ref55" type="bt">34</bibl> <bibtext>Skeat, J., Wake, M., Ukoumunne, O.C., Eadie, P., Bretherton, L., & Reilly, S. (2014). Who gets help for pre ‐ school communication problems? Data from a prospective community study. Child: Care Health and Development, 40, 215 – 222. </bibtext> </blist> <blist> <bibl id="bib35" idref="ref2" type="bt">35</bibl> <bibtext>Snowling, M.J., Adams, J.W., Bishop, D.V.M., & Stothard, S.E. (2001). Educational attainments of school leavers with a preschool history of speech ‐ language impairments. International Journal of Language and Communication Disorders, 36, 173 – 183. </bibtext> </blist> <blist> <bibl id="bib36" idref="ref20" type="bt">36</bibl> <bibtext>Snowling, M.J., Duff, F.J., Nash, H.M., & Hulme, C. (2016). Language profiles and literacy outcomes of children with resolving, emerging, or persisting language impairments. Journal of Child Psychology and Psychiatry, 57, 1360 – 1369. </bibtext> </blist> <blist> <bibl id="bib37" idref="ref50" type="bt">37</bibl> <bibtext>StataCorp. (2013). Stata Statistical Software: Release 13. College Station, TX : StataCorp LP. </bibtext> </blist> <blist> <bibl id="bib38" idref="ref27" type="bt">38</bibl> <bibtext>Taylor, C.L., Christensen, D., Lawrence, D., Mitrou, F., & Zubrick, S.R. (2013). Risk factors for children's receptive vocabulary development from four to eight years in the longitudinal study of Australian children. PLoS ONE, 8, e73046. https://doi.org/10.1371/journal.pone.0073046. </bibtext> </blist> <blist> <bibl id="bib39" idref="ref33" type="bt">39</bibl> <bibtext>Wiig, E.H., Secord, W.A., & Semel, E. (2006). Clinical evaluation of language fundamentals ‐ preschool: Australian standardised edition (2nd edn). Sydney, Australia : Harcourt Assessment. </bibtext> </blist> <blist> <bibl id="bib40" idref="ref40" type="bt">40</bibl> <bibtext>Wilkinson, G.S., & Robertson, G.J. (2006). The wide ranging acheivement test (4th edn). Lutz, FL : Psychological Assessment Resources. </bibtext> </blist> <blist> <bibl id="bib41" idref="ref46" type="bt">41</bibl> <bibtext>Wraith, D., & Wolfe, R. (2014). Classifying patients by their characteristics and clinical presentations; the use of latent class analysis. Respirology, 19, 1138 – 1148. </bibtext> </blist> <blist> <bibl id="bib42" idref="ref8" type="bt">42</bibl> <bibtext>Zambrana, I.M., Pons, F., Eadie, P., & Ystrom, E. (2014). Trajectories of language delay from age 3 to 5: Persistence, recovery and late onset. International Journal of Language and Communication Disorders, 49, 304 – 316. </bibtext> </blist> <blist> <bibl id="bib43" idref="ref14" type="bt">43</bibl> <bibtext>Zubrick, S.R., Taylor, C.L., & Christensen, D. (2015). Patterns and predictors of language and literacy abilities 4 ‐ 10 years in the longitudinal study of Australian children. PLoS ONE, 10, e0135612. </bibtext> </blist> </ref> <p>Graph: Plots of individual growth trajectories by class for Model M1 [on the x ‐ axis is age (years) and on the y ‐ axis is the CELF score, Group 1 = Low ‐ decreasing (n  = 50), Group 2 = Stable (n  = 1,199), Group 3 = Low ‐ increasing (n  = 30)]</p> <p>Graph: Appendix S1. Participant flow chart from 4 months to 11 years (denominator is number participating at baseline – 1,910). Appendix S2. Model fit statistics for the three modelling approaches applied.</p> <aug> <p>By Cristina McKean; Darren Wraith; Patricia Eadie; Fallon Cook; Fiona Mensah and Sheena Reilly</p> </aug>
Header DbId: eric
DbLabel: ERIC
An: EJ1154693
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Subgroups in Language Trajectories from 4 to 11 Years: The Nature and Predictors of Stable, Improving and Decreasing Language Trajectory Groups
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22McKean%2C+Cristina%22">McKean, Cristina</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0001-9058-9813">0000-0001-9058-9813</externalLink>)<br /><searchLink fieldCode="AR" term="%22Wraith%2C+Darren%22">Wraith, Darren</searchLink><br /><searchLink fieldCode="AR" term="%22Eadie%2C+Patricia%22">Eadie, Patricia</searchLink><br /><searchLink fieldCode="AR" term="%22Cook%2C+Fallon%22">Cook, Fallon</searchLink><br /><searchLink fieldCode="AR" term="%22Mensah%2C+Fiona%22">Mensah, Fiona</searchLink><br /><searchLink fieldCode="AR" term="%22Reilly%2C+Sheena%22">Reilly, Sheena</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Child+Psychology+and+Psychiatry%22"><i>Journal of Child Psychology and Psychiatry</i></searchLink>. Oct 2017 58(10):1081-1091.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: cs-journals@wiley.com; Web site: http://www.wiley.com/WileyCDA
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 11
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2017
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Language+Acquisition%22">Language Acquisition</searchLink><br /><searchLink fieldCode="DE" term="%22Child+Development%22">Child Development</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Aptitude%22">Language Aptitude</searchLink><br /><searchLink fieldCode="DE" term="%22Longitudinal+Studies%22">Longitudinal Studies</searchLink><br /><searchLink fieldCode="DE" term="%22Cohort+Analysis%22">Cohort Analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Regression+%28Statistics%29%22">Regression (Statistics)</searchLink><br /><searchLink fieldCode="DE" term="%22Predictor+Variables%22">Predictor Variables</searchLink><br /><searchLink fieldCode="DE" term="%22Environmental+Influences%22">Environmental Influences</searchLink><br /><searchLink fieldCode="DE" term="%22Biological+Influences%22">Biological Influences</searchLink><br /><searchLink fieldCode="DE" term="%22Non+English+Speaking%22">Non English Speaking</searchLink><br /><searchLink fieldCode="DE" term="%22Disadvantaged%22">Disadvantaged</searchLink><br /><searchLink fieldCode="DE" term="%22Family+Environment%22">Family Environment</searchLink><br /><searchLink fieldCode="DE" term="%22Childrens+Literature%22">Childrens Literature</searchLink><br /><searchLink fieldCode="DE" term="%22Emotional+Problems%22">Emotional Problems</searchLink><br /><searchLink fieldCode="DE" term="%22Family+Literacy%22">Family Literacy</searchLink><br /><searchLink fieldCode="DE" term="%22Learning+Disabilities%22">Learning Disabilities</searchLink><br /><searchLink fieldCode="DE" term="%22Intervention%22">Intervention</searchLink><br /><searchLink fieldCode="DE" term="%22Progress+Monitoring%22">Progress Monitoring</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Tests%22">Language Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Intelligence+Tests%22">Intelligence Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Behavior+Problems%22">Behavior Problems</searchLink><br /><searchLink fieldCode="DE" term="%22Child+Behavior%22">Child Behavior</searchLink><br /><searchLink fieldCode="DE" term="%22Screening+Tests%22">Screening Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Questionnaires%22">Questionnaires</searchLink><br /><searchLink fieldCode="DE" term="%22Achievement+Tests%22">Achievement Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Statistical+Analysis%22">Statistical Analysis</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Australia%22">Australia</searchLink>
– Name: SubjectThesaurus
  Label: Assessment and Survey Identifiers
  Group: Su
  Data: <searchLink fieldCode="SU" term="%22Clinical+Evaluation+of+Language+Fundamentals%22">Clinical Evaluation of Language Fundamentals</searchLink><br /><searchLink fieldCode="SU" term="%22Kaufman+Brief+Intelligence+Test%22">Kaufman Brief Intelligence Test</searchLink><br /><searchLink fieldCode="SU" term="%22Strengths+and+Difficulties+Questionnaire%22">Strengths and Difficulties Questionnaire</searchLink><br /><searchLink fieldCode="SU" term="%22Wide+Range+Achievement+Test%22">Wide Range Achievement Test</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1111/jcpp.12790
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0021-9630
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Background: Little is known about the nature, range and prevalence of different subgroups in language trajectories extant in a population from 4 to 11 years. This hinders strategic targeting and design of interventions, particularly targeting those whose difficulties will likely persist. Methods: Children's language abilities from 4 to 11 years were investigated in a specialist language longitudinal community cohort (N = 1,910). Longitudinal trajectory latent class modelling was used to characterise trajectories and identify subgroups. Multinomial logistic regression was used to identify predictors associated with the language trajectories children followed. Results: Three language trajectory groups were identified: "stable" (94% of participants), "low-decreasing" (4%) and "low-improving" (2%). A range of child and family factors were identified that were associated with following either the low-improving or low-increasing language trajectory; many of them shared. The low-improving group was associated with mostly environmental risks: non-English-speaking background, social disadvantage and few children's books in the home. The low-decreasing group was associated with mainly biological risks: low birth weight, socioemotional problems, lower family literacy and learning disability. Conclusions: By 4 years, services can be confident that most children with low language will remain low to 11 years. Using rigid cut-points in language ability to target interventions is not recommended due to continued individual variability in language development. Service delivery models should incorporate monitoring over time, targeting according to language abilities and associated risks and delivery of a continuum of interventions across the continuum of need.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Ref
  Label: Number of References
  Group: RefInfo
  Data: 43
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2017
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1154693
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1154693
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1111/jcpp.12790
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 11
        StartPage: 1081
    Subjects:
      – SubjectFull: Language Acquisition
        Type: general
      – SubjectFull: Child Development
        Type: general
      – SubjectFull: Language Aptitude
        Type: general
      – SubjectFull: Longitudinal Studies
        Type: general
      – SubjectFull: Cohort Analysis
        Type: general
      – SubjectFull: Regression (Statistics)
        Type: general
      – SubjectFull: Predictor Variables
        Type: general
      – SubjectFull: Environmental Influences
        Type: general
      – SubjectFull: Biological Influences
        Type: general
      – SubjectFull: Non English Speaking
        Type: general
      – SubjectFull: Disadvantaged
        Type: general
      – SubjectFull: Family Environment
        Type: general
      – SubjectFull: Childrens Literature
        Type: general
      – SubjectFull: Emotional Problems
        Type: general
      – SubjectFull: Family Literacy
        Type: general
      – SubjectFull: Learning Disabilities
        Type: general
      – SubjectFull: Intervention
        Type: general
      – SubjectFull: Progress Monitoring
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Language Tests
        Type: general
      – SubjectFull: Intelligence Tests
        Type: general
      – SubjectFull: Behavior Problems
        Type: general
      – SubjectFull: Child Behavior
        Type: general
      – SubjectFull: Screening Tests
        Type: general
      – SubjectFull: Questionnaires
        Type: general
      – SubjectFull: Achievement Tests
        Type: general
      – SubjectFull: Statistical Analysis
        Type: general
      – SubjectFull: Australia
        Type: general
      – SubjectFull: Clinical Evaluation of Language Fundamentals
        Type: general
      – SubjectFull: Kaufman Brief Intelligence Test
        Type: general
      – SubjectFull: Strengths and Difficulties Questionnaire
        Type: general
      – SubjectFull: Wide Range Achievement Test
        Type: general
    Titles:
      – TitleFull: Subgroups in Language Trajectories from 4 to 11 Years: The Nature and Predictors of Stable, Improving and Decreasing Language Trajectory Groups
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: McKean, Cristina
      – PersonEntity:
          Name:
            NameFull: Wraith, Darren
      – PersonEntity:
          Name:
            NameFull: Eadie, Patricia
      – PersonEntity:
          Name:
            NameFull: Cook, Fallon
      – PersonEntity:
          Name:
            NameFull: Mensah, Fiona
      – PersonEntity:
          Name:
            NameFull: Reilly, Sheena
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 10
              Type: published
              Y: 2017
          Identifiers:
            – Type: issn-print
              Value: 0021-9630
          Numbering:
            – Type: volume
              Value: 58
            – Type: issue
              Value: 10
          Titles:
            – TitleFull: Journal of Child Psychology and Psychiatry
              Type: main
ResultId 1