The Language Profile of School-Aged Children with FASD

The Language Profile of School-Aged Children

with Fetal Alcohol Spectrum Disorder (FASD)

Le profil linguistique d’enfants d’âge scolaire

KEY WORDS ayant un trouble du spectre de l’alcoolisation
FETAL ALCOHOL SPECTRUM
DISORDER (FASD) fœtale (SAF)
PARTIAL FETAL ALCOHOL
SYNDROME (PFAS) Shelley Proven
ALCOHOL RELATED
NEURODEVELOPMENTAL Carla Ens
DISORDER (ARND) Paul G. Beaudin
PRENATAL ALCOHOL
EXPOSURE
LANGUAGE DEVELOPMENT
LANGUAGE DISORDERS
Abstract
CLINICAL EVALUATION OF A population-based study of school age children diagnosed with FASD was conducted to evaluate
LANGUAGE FUNDAMENTALS-
the language abilities of these children and describe their language strengths and weaknesses.
4TH EDITION (CELF-4)
A retrospective chart review methodology was applied to examine language abilities of children
diagnosed with FASD. Secondary data from 124 children aged 5 to 18 years, who were diagnosed
with FASD between January 2005 and October 2010, were included in the study. Results from
the CELF-4 language assessment tool were analyzed to compare the language abilities of
these children. This study revealed globally poor performance across expressive and receptive
language abilities, suggesting that language development is significantly affected by prenatal
alcohol exposure. The Core Language Index Scores (total test scores) showed almost 70% of
the participants received a language rating of “severe” (indicating significant communication
impairments). About 20% had a rating of either ”moderate” or “mild”, and fewer than 15% had a
rating of “average”. Approximately 85% of the sample experienced mild to severe language delays
in the index categories. The 5 to 8 year old age group had the highest average scores in all index
Shelley Proven, M.Sc., categories, whereas the 9 year-olds consistently had the lowest average scores. The changing
S-LP(C), CCC-SLP profile by age group is significant with important ramifications on longitudinal language testing
Clinical Resource Speech and programming. A better understanding of language abilities in children with prenatal alcohol
Language Pathologist; exposure may lead to improved planning for language interventions.
Manitoba FASD Centre,
633 Wellington Crescent,
Winnipeg, MB
Canada
Abrégé
Une étude d’une population d’enfants d’âge scolaire ayant un diagnostic de SAF a été faite afin
Carla Ens, Ph.D. d’évaluer les habiletés langagières de ces enfants et de décrire leurs forces et leurs faiblesses
Epidemiologist, Manitoba au plan du langage. Une méthodologie d’examen rétrospectif des dossiers fut appliquée. Des
Health; Assistant Professor, données secondaires de 124 enfants âgés de 5 à 18 ans ayant le diagnostic de SAF, recueillies
Department of Community entre janvier 2005 et octobre 2010, furent inclues dans l’étude. Les résultats de l’outil d’évaluation
Health Sciences, Faculty langagière CELF-4 ont été analysés pour comparer les habiletés linguistiques de ces enfants.
of Medicine, University of Cette étude a révélé une performance généralement pauvre pour les habiletés langagières
Manitoba expressives et réceptives, ce qui suggère que le développement langagier est affecté de façon
Winnipeg, MB significative par l’exposition prénatale à l’alcool. Les scores de base des indices langagiers (CLIS)
Canada (résultat total des tests) démontrent que presque 70 % des participants ont reçu une cote
« sévère » (indiquant des troubles importants de communication). Environ 20 % ont reçu une
Paul G. Beaudin, Ph.D., cote de degré « modéré » ou « léger » et moins de 15 % ont reçu une cote de degré « moyen ».
MSc., S-LP(C) Environ 85 % des participants accusaient un retard de langage de léger à sévère dans les
Research Associate, catégories d’indices. Le groupe des cinq à huit ans avait les taux moyens les plus élevés dans
Research and Evaluation toutes les catégories d’indices, alors que les enfants de neuf ans avaient constamment les taux
Unit, Winnipeg Regional moyens les plus bas. Le changement de profil selon le groupe d’âge est significatif, ce qui a des
Health Authority (WRHA) ramifications importantes sur l’évaluation et la programmation longitudinale du langage. Une
Winnipeg, MB meilleure compréhension des habiletés langagières des enfants exposés à l’alcool en période
Canada prénatale pourra contribuer à l’amélioration des interventions dans le domaine du langage.

268 Canadian Journal of Speech-Language Pathology and Audiology | Vol. 37, N0. 4, Winter 2014
 The Language Profile of School-Aged Children with FASD

Introduction deficiency; the facial phenotype; central nervous system

damage or dysfunction; and gestational exposure to alcohol.
Fetal Alcohol Spectrum Disorder (FASD) is an umbrella
term that encompasses three specific medical diagnoses The following nine domains of the central nervous system
resulting from prenatal exposure to alcohol: Fetal receive neurobehavioural assessments: 1) communication;
Alcohol Syndrome (FAS), Partial Fetal Alcohol Syndrome 2) hard and soft neurological signs including sensory
(pFAS) and Alcohol Related Neurodevelopmental motor deficits; 3) brain structure; 4) cognition; 5) academic
Disorder (ARND). FAS describes those individuals with a achievement; 6) memory; 7) executive functioning; 8)
characteristic pattern of physical and neurological birth attention; and 9) adaptive behaviour. A central nervous
defects, including facial dysmorphology, growth deficiency,
system domain, commonly referred to as a “brain domain”
and neurobehavioural abnormalities; pFAS refers to
in clinical practice, is considered to be impaired when
those with facial dysmorphology and neurobehavioural
standardized scores are either two or more standard
abnormalities but no evidence of growth deficiency; ARND
deviations below the mean or where there is a discrepancy
pertains to those individuals who have characteristic
of at least one standard deviation between subtests within
neurodevelopmental abnormalities but no dysmorphology
domains. The Canadian Guidelines require impairment
or growth retardation. The spectrum of brain differences
in three different domains before a diagnosis can be
with FASD varies by individual and may cause different
considered. Over the years, there has been an improved
learning, behavioural and daily living challenges for each
understanding of each of the brain domains through
(Chudley et al., 2005). An estimated 9 in 1000 babies born
research. Without exception, research has improved
in Canada are affected by FASD (Public Health Agency
our understanding and assessment of the relationships
of Canada, 2005). Although there has been a substantial
between various language components.
body of literature examining behavioural, psychosocial
and cognitive impairments of FASD, there is a scarcity Assessing the Communication Brain Domain. Since
of research on prenatal alcohol exposure (PAE) and its FASD was first documented, a connection between
effect on language development. Large-scale language prenatal alcohol exposure (PAE) and impaired language
and communication deficiencies have been described in development has been considered. Early on, researchers
individuals with FASD yet no consistent or conclusive theorized a connection between PAE and language,
pattern of deficits has been identified. Therefore, the suggesting the need for special education services such as
objectives of this study were to: 1) examine the language speech and language interventions (Streissguth, Herman,
abilities of school age children who have a formal & Smith, 1978). Shortly thereafter, Sparks (1984) was among
diagnosis of either pFAS or ARND; and 2) describe the the first S-LPs to query if a specific relationship between
language strengths and weaknesses in this population. PAE and language difficulties existed. Her formative
The present study will contribute relevant and unique work identified a stronger link between FASD and speech
information to the growing body of research about and language problems in children than had previously
the language abilities of children with FASD. A better been reported. In the late 90s, further support came
understanding of the language profiles of school aged from Church & Kaltenbach (1997) who posited FAS may
children with FASD can assist the FASD diagnostic teams be the leading cause of hearing, speech, and language
in assessing communication abilities and improve the difficulties in children. Since then, studies examining
services offered by speech-language pathologists (S-LPs). deficiencies in children with FASD have described a
number of communication deficiencies encompassing
Background language areas such as: naming (Mattson, Riley, Gramling,
Making an FASD Diagnosis in Canada. The procedures used Delis, & Jones, 1998); verbal fluency (Mattson & Riley,
to diagnose a disorder due to prenatal alcohol exposure 1999; Schoenfeld, Mattson, Lang, Delis & Riley, 2001);
and determine an alcohol related diagnosis have changed grammar comprehension (Kodituwakku, 2009); central
considerably since FAS was first described by Jones and processing (Church & Kaltenbach, 1997); narrative
Smith (1973). The first Canadian Guidelines for the Diagnosis discourse (Thorne, Coggins, Carmichael Olson, & Astley,
of FASD and its related disabilities were published in 2005 2007); and inappropriate use of social language (Coggins,
and were established to assist Canadian diagnostic teams Timler, & Olswang, 2007). The results of these studies,
in providing a consistent and objective diagnosis (Chudley however, have not been consistent. Several studies have
et al., 2005). The Canadian Guidelines were harmonized shown a significant correlation between global language
with other international protocols resulting in a four digit development and PAE (Becker, Warr-Leeper & Leeper, 1990;
diagnostic code now widely used in Canada (Chudley et Carney & Chermak, 1991) while others have not (Greene,
al., 2005). This code addresses the severity of the four key Earnhart, Martier, Sokol, & Ager, 1990; Fried, O’Connell &
diagnostic features of FASD in the following areas: growth Watkinson, 1992). Limitations such as small sample sizes,

Revue canadienne d’orthophonie et d’audiologie | Vol. 37, N0. 4, hiver 2014 269
The Language Profile of School-Aged Children with FASD

low doses of prenatal alcohol consumption, and large age Group, 2009). As a collective, the Manitoba FASD Centre1
ranges, may have contributed to the inconclusive findings. selected the CELF-4 as one of the assessment tools to
evaluate language and communication.
The literature examining language development in
children with FASD has not used consistent methodologies Current Study
or instruments, nor have the existing studies applied
The specific goals of this research study were to: 1)
consistent methods of diagnosing FASD (there is variation
determine if there are significant differences across age
between and sometimes within countries). As the field can
groups for the different measures derived from the CELF-4;
still be considered to be in its infancy, inconsistencies should
2) examine if there are significant difference in diagnosis
be expected. Even so, language deficits in children with
for the different measures derived from the CELF-4; and 3)
FASD are regularly reported. Several different language
to analyze if there are significant differences in sex for the
tests (e.g. Test of Language Development-Primary-4th
different measures derived from the CELF-4.
edition, (Newcomer & Hammill, 2008) and Comprehensive
Receptive Expressive Vocabulary Test-2nd edition (Wallace
Methods
& Hammill, 2002)) have been utilized in studies and it is
unclear how comparable those results are to the CELF-4 Procedures. A retrospective chart review was conducted
(Adnams et al., 2007; Aragon, et al., 2008; Carney & Chermak, of 124 children between the ages of 5 and 18 years that had
1991; Coggins et al., 2007; Kodituwakku et al., 2006; Wyper been assessed over a five and a half year period (January
& Rasmussen, 2011). Assuming a broad level of similarity, it 2005 to October 2010) at the Manitoba FASD Centre. The
is not surprising that the literature indicates a widespread Manitoba FASD Centre assessments adhere to the Canadian
range of language and communication deficiencies in guidelines for FASD diagnosis (Chudley et. al., 2005) and
children with FASD. Knowing that language will likely be all participants were administered the CELF-4 as part
an issue for a child with FASD means that the role of S-LP of the communication domain assessment. A diagnosis
in assessing and developing individual programming is was only made collectively following the completion of
critical. Improved identification of language strengths and assessments from a multidisciplinary team including an
limitations through yearly assessments using a standard S-LP, occupational therapist, psychologist, developmental
protocol will allow for a strengths-based approach for pediatrician, and a geneticist. This is not a “blind” approach
a child with FASD. To encourage greater consistency in as clinicians are aware that they are assessing a child
evaluation of language and communication abilities, Pan who may have FASD. However, having multiple members
contribute to the final decision increases objectivity. The
Canadian consultations were held to identify standardized
subjects of this chart review were all assessed by the same
tools and diagnostic categories.
S-LP. Ethics approval was obtained for this study by the
In 2007, Pan Canadian consultations resulted in a list of Health Research Ethics Board, University of Manitoba.
standard tools appropriate for the assessment of children
Participants. A database was created by the S-LP team
between 4 to 18 years of age when an FASD diagnosis is
at the Manitoba FASD Centre in 2005, consisting of
being considered. The four tools identified for evaluating
demographic information and scores of various language
language and communication for children aged 6 to 11
assessment instruments. Individuals assessed between
years included: 1) the Clinical Evaluation of Language
January 2005 and October 2010 were extracted for the study
Fundamentals -4th edition (CELF-4) (Semel, Wiig, & Secord,
by the principal investigator. To be included in the sample,
2003); 2) the Test of Narrative Language (Gillam & Pearson,
participants needed to: a) be aged 5 to 18 years, b) have
2004); 3) the Test of Problem Solving 3 Elementary -3rd
the language portion of CELF-4 completed, c) be English
edition (Bowers, Huisingh, & LoGiudice, 2005); and 4) the
speaking, d) have been assessed by the same S-LP, and e)
Pragmatics Profile subtest from the CELF-4. For children
have received an FASD diagnosis based on the Canadian
ages 12 to 18 years, the three standard set of core language
Guidelines. If data were missing from any component of the
tests included: 1) the CELF-4; 2) Test of Problem Solving
CELF-4 language assessment, the individual was excluded
2 Adolescents (Bowers, Huisingh, & LoGiudice, 2007); and
from the study. Of the 1078 children evaluated by the S-LP,
3) the CELF-4 Pragmatics Profile (Canada NorthWest
124 (11.83%), met the inclusion criteria.
FASD Research Group, 2009). Although the CELF-4 does
have certain limitations in that it does not assess social Of the 124 children that met the criteria for the chart review,
language, problem solving or written language abilities, 23 (18.5%) had a diagnosis of pFAS and 101 (81.5%) had a
the Pan Canadian consultations with other S-LPs working diagnosis of ARND. No participants had a diagnosis of FAS.
in FASD diagnostic clinics determined that the CELF-4 Of the total group, 78 (62.9%) of the subjects were male and
was acceptable and appropriate to effectively test the 46 (37.1%) were female. Two-thirds of the sample (n=83) was
communication domain (Canada NorthWest FASD Research based in an urban setting and one-third (n=41) had some

270 Canadian Journal of Speech-Language Pathology and Audiology | Vol. 37, N0. 4, Winter 2014
 The Language Profile of School-Aged Children with FASD

form of intervention (e.g., consultation or direct therapy) Analysis. The CELF-4 data were analyzed using SPSS
with S-LP services prior to the assessment. With regards to (v.11.0). Frequency distributions were examined to identify
alcohol use, 47 (39.7%) of the biological mothers used alcohol and address potential outliers. In order to address the
only, 15 (12.1%) used a combination of alcohol and tobacco, first objective of this study, the evaluation of language
and 62 (50.0%) reported using alcohol and other substances abilities, cross-tabulations were completed to provide
such as cocaine or marijuana. Family structure varied; half the results by age group, including the core language
of the participants lived in foster care (n=63; 50.8%) with the score (total test score), and indices measuring receptive
remainder of the sample living with: 1) biological parent(s) language, expressive language, language content, language
(n=25; 20.2%); 2) an extended family member (n=27; 21.7%); structure, and language memory. Parametric tests (t-tests
or 3) an adoptive family (n=9; 7.3%). All of individuals in the and ANOVAs) were conducted to assess the significance of
sample spoke English as their first language, with a small between group differences. For example, it was important
proportion also speaking a second language (7.3%) (see Table to determine if scores differed significantly by sex of
1). The demographics of the sample were consistent with the participant as well as by age of participant. The second
population seen at the Manitoba FASD Centre. objective of the study was to describe strengths and
limitations based on the CELF-4 total and index scores.
Measure. The CELF-4 is the third revision of The Clinical
To address this objective, age group results were plotted
Evaluation of Language Fundamentals that was originally
to determine performance differences among age groups
published in 1980 (Semel, Wiig, & Secord, 2003). It is
and to compare each age group in the diagnosed sample to
an individually administered clinical tool used in the
results for the general population. For all statistical tests, a
identification, diagnosis and follow-up evaluation of
level of significance of p<0.05 was chosen.
language and communication disorders in individuals 5 to
21 years old. Two separate test forms are used for individuals
Results
aged 5 to 8 and 9 to 21 years. These test forms are then
differentiated by a variety of sub-tests which lead to further The CELF-4’s Core Language Index Score provides a broad
divisions in the age groupings: 1) 5 to 8 years; 2) 9 years; 3) assessment of an individual’s overall language performance.
10 to 12 years; and 4) 13 to 21 years (See Table 2). Following No significant differences were found based on age of the
the test administration, there are two important score participants (see Table 3). In this sample, two-thirds of the
categories. The first is the ‘Core Language Score’, which is participants (n= 84; 67.7%) received a severity rating of severe.
a measure of the general language ability that quantifies a Approximately 10% of the sample (n=12; 9.7%) received a
child’s overall language performance. The second important severity rating of moderate, 11 (8.9%) received a severity
category includes the ‘language indices’. There are five rating of mild, and 17 (13.7%) received a severity rating of
language indices that provide additional details on language average. The “communication brain domain” would be
and communication and are calculated using 15 sub-test considered impaired if children receive a rating of severe,
scaled scores (see Table 3). in that it is two standard deviations below the mean or
there is a discrepancy of at least one standard deviation
The CELF-4 classifies language delay using the following between subtests with domains. Although there was slight
severity rating scale: severe (standard scores < 70); moderate variation by indices, the trend was that the majority of the
(standard scores between 71- 77); mild (standard scores sample had some level of language delay with the greatest
between 78-85); average (standard scores between 86-114); proportion being severely delayed.
and above average (standard scores >114) (Semel, Wiig, &
Secord, 2003). As stated earlier, the Canadian Guidelines When examining chi-square results by age groups and index
require impairment in three different domains before a categories, the Receptive Language Index Scores varied
diagnosis can be considered. The “communication brain significantly by age group (p<0.01) as did the Language
domain” would be considered impaired if children receive a Structure Index Scores (p=0.02). The 5 to 8-year-olds had
rating of severe, in that it is two standard deviations below the highest average scores (higher scores indicates lower
the mean. For example, when using the CELF-4, a severity severity) in all three index categories yet the scores still
rating below a standard score of 70 indicates a severe indicated moderate to severe language difficulties. The
impairment, and thus the communication domain would average scores across all age groups and indices ranged from
be identified as a significant deficit when considering for 60.2 to 76.6, again indicating severe language issues. There
an FASD diagnosis. While internal consistency reliability appears to be a complex pattern emerging. The highest
estimates for the CELF-4 vary depending on the age group scores, indicating better performance, were seen in the
and subtest (α ranges from 0.77 to 0.92) they have shown youngest group. In the 9 year-olds, there is a dip in scores
adequate stability across time for all age bands (Semel, Wiig, which is followed by somewhat higher measures for the
& Secord, 2003). older group (see Table 4).

Revue canadienne d’orthophonie et d’audiologie | Vol. 37, N0. 4, hiver 2014 271
The Language Profile of School-Aged Children with FASD

As the sample contained participants who could be communication impairment as the focus is on the child’s
classified into dichotomous groups, t-tests were used to other areas of difficulties (ie. poor attention span, sleep
determine if there were differences in scores by diagnosis. difficulties, behavioural issues, etc.). The results from this
Although the scores for the pFAS sample (n=23) were study clearly suggest that language services need to be
consistently lower in all indices than those of the ARND viewed as a necessity given the needs of the population.
population (n=101), these differences were not statistically For example, a child who has access to language services
significant (Table 5). prior to the assessment may not receive a different
diagnosis but could have improvements in his or her
Finally, the test scores were also not statistically significant
language capabilities overall.
for differences by sex for all sub-groups. It also means that
language scores did not vary a great deal when comparing Second, a factor linked to lower scores on the CELF-4 was
children with different diagnoses. For example, a female age. Our sample indicated that younger children scored
with pFAS was not likely to have a significantly higher or better than older children. The literature, however, is not
lower score that a male with ARND. clear on whether language performance changes over
time, the direction of the change, or if language remains
Discussion stable. Davies et al. (2011) did note a consistent performance
This study examined the use of a standardized decrease in language testing over time. Wyper & Rasmussen
comprehensive language assessment tool, the CELF-4, for (2011) cites literature differentiating language deficits in
evaluation of the language abilities in children diagnosed both older and younger children; older children with FAS
with FASD between the years 2005 and 2010. It is the first showed deficits most specifically in syntax whereas younger
study to comprehensively examine language development children have more global language deficits. The concern
in a large cohort of school age children formally diagnosed is that language testing does not happen frequently
with FASD and using the Canadian guidelines. As such, it enough to ascertain changes in language development. In
represents an important contribution to what is known addition, it is unclear what factors, if any, may be impacting
about FASD, communication, and language measurement. language development. If there is a common trajectory,
Although it has been the perception that individuals with is it a product of environment, the nature of FASD, or a
FASD have higher expressive language skills relative to their unique combination as Coggins and his colleagues suggest
receptive language skills, this study did not support this (2007). For the S-LP, these factors should be considered,
perception. One key finding was that the majority of the giving way to annual testing and documentation
clinical sample had language deficiencies: over 65% of the on social environmental factors. Future research,
sample showed severe impairment and an additional 16% including longitudinal studies, could further impact our
of the sample demonstrated a core language deficit that understanding if comparisons would be made between the
ranged from mild to moderate. Another key finding was that developmental trajectories of typically developing children
age seemed to be an important factor in some indices of to those with FASD in similar social environments.
the CELF-4: the Receptive and Expressive Language Index
Third, a small proportion (13.7%) of the clinical sample was
Scores showed significant difference by age as did Language
considered “typical” based on this language assessment,
Structure Index Scores. The youngest group consistently
a finding that warrants further examination. Other
had the strongest results in all three subtests. A third key
researchers have noted similar findings and have reported
finding was that there were no significant differences in
no significant differences in controls and FASD cases
language scores when comparing children diagnosed with
(Kodituwakku et al., 2006). One study reported one-third
pFAS and ARND. This is a critical finding that warrants
of their clinical sample achieved language scores within
more investigation, particularly with a sample that includes
the expected range of language performance (Coggins et
children with FAS. In addition to profiling the language
al., 2007). However, it should be noted that the limitations
scores of FASD children, this study identified several
of the CELF-4 may not have captured weaknesses in
interesting findings that warrant further discussion.
written language or social language use. Future research
First, a particularly troubling finding was that should include an examination into the proportion of
approximately 85% of the sample had some level of FASD children who have typical language development. It
language impairment with almost 70% presenting with may be that certain protective factors are common in that
severe language deficits. However, we found that only particular population. Streissguth (2003) has posited that
33% of the entire sample had previously received some at least six protective factors are associated with lower
form of speech and language intervention. This lack of rates of secondary disabilities in the FASD population:
service provision may be the result of individuals involved living in a stable nurturing home of good quality, not
with the child not recognizing the significance of their having frequent changes of household, not being a victim

272 Canadian Journal of Speech-Language Pathology and Audiology | Vol. 37, N0. 4, Winter 2014
 The Language Profile of School-Aged Children with FASD

of violence, having received disability services, and having Guidelines—only about 10% of any clinical sample will
been diagnosed before the age of 6 years. Measuring those be diagnosed with FAS (Coggins et al., 2007) —having
factors could implicate practice. Knowing what contributes the diagnosed population limited to children with either
to typical development would both assist in determining ARND or pFAS was limiting. It may be that due to the
deficiencies in social environments and improve models of more prominent dysmorphia present in children with FAS,
assessment and diagnosis. It also highlights the importance diagnoses occur before five years of age. Expanding the
of large scale data collection for each clinical assessment, inclusion criteria to include pre-school children may have
to enable research to conduct multi-factorial studies to captured FAS children differently. Additional limitations
better understand both strengths and challenges as well as were that the study was unable to determine impact of
promote evidence informed best practice for S-LPs. co-morbid diagnoses (such as ADHD), social-economic
backgrounds, or home/family environment (e.g., adopted,
Since none of the children in the present sample had foster homes, etc). It also would have been useful to
FAS, we became differently focused on the ARND and complete a multi-factorial analysis that incorporates the
pFAS scores in order to better understand if differences impact of socioeconomic status. The information on family
by diagnosis were apparent. We found that there were type is useful in that regard but not a direct proxy. At the
few differences between the scores of participants with time of the study, however, the Manitoba FASD Centre did
either ARND or pFAS; language and communication not collect data on socioeconomic status. Finally, the study
seemed similar regardless of diagnosis. To the best of our did not have a typical control group of children from similar
understanding, prior research has not explored language by socio-economic backgrounds.
specific diagnosis which makes this result novel and worthy
of further exploration. It also has pragmatic implications Directions for Future Research. Future research on the
for assessment and therapy. Dysmorphology in FASD is language abilities of children with FASD would require
a contributor to the diagnosis of FAS and pFAS. Having more in-depth data pertaining to the language services
the physical traits of FAS and pFAS can provide outward children received. This data would allow for greater clarity
evidence of the brain disorder; children with FAS and on the role of S-LP in providing intervention. In addition,
pFAS may be differently supported as their disability is future studies should target children with FAS. Future
more recognizable. Because children with ARND appear research has many potential avenues. A longitudinal
methodology would be useful to understand changes
“typical”, there may be an assumption that language and
over time, the possible role or impact of environment, as
communication also follow along a “typical” trajectory.
well as the implications of socioeconomic status within
If children with ARND are not being recognized by
a changing demographic. Exploring language abilities as
parents and teachers as having language difficulties, they
opposed to deficiencies based on results of the subtest
are at greater risk for being misunderstood in both the
scores on the CELF-4 would also be an interest direction
classroom and home environment. Given the severity
with clinical application. Employing a qualitative analysis
of the language impairment in the non-dysmorphic
of children with typical communication behaviours would
diagnostic categories, there is no question that the non-
be an excellent way to explore less tangible variables
dysmorphic population needs the same level of support
that may impact language in a child with FASD. In depth
as those who are dysmorphic.
diagnostic assessment of multiple language areas is central
Limitations. This study contributes to the examination of to understanding the behavioural phenotype of school aged
language abilities and general language profile of school age children with FASD. Given that prenatal alcohol ingestion is
children with an FASD. The sample size of 124, collected over often paired with other teratogens, future research should
a five and a half year period, was appropriate to measure explore if language and communication outcomes vary by
significant differences. The CELF-4 is a validated tool and types of prenatal exposures.
useful in measuring communication capability. For the
scope of this study, it was a valuable tool in that it provided Conclusion
detailed data by age group pertaining to language strengths This study contributes to the emerging literature pertaining
and limitations in those children being assessed for FASD. to language and communication abilities in children with
One limitation, however, was that very few research studies FASD. This study revealed that children with FASD had
have utilized the CELF-4 instrument in analyses, making globally poor performances across expressive and receptive
comparisons limited in their specificity. This limitation language abilities, suggesting that language development is
impacts the study’s generalizability and applicability to significantly affected by prenatal alcohol exposure. The Core
populations outside of Canada. Additionally, none of the Language Index Scores of the CELF-4 showed almost 70%
children in the present sample had a diagnosis of FAS. of the participants received a language rating of “severe”
While FAS is not a common diagnosis with the Canadian (indicating significant communication impairments), about

Revue canadienne d’orthophonie et d’audiologie | Vol. 37, N0. 4, hiver 2014 273
The Language Profile of School-Aged Children with FASD

20% were rated as “mild”, and fewer than 15% had a rating children with fetal alcohol spectrum disorder. Language Speech and
Hearing Services in the Schools, 38, 117-138.
of “average”. In addition, we found that language abilities
changed with age: The five to eight year old age group had Davies, L., Dunn, M., Chersich, M., Urban, M., Chetty, C., Olivier, L., & Viljoen,
D. (2011). Developmental delay of infants and young children with and
the highest average scores in all index categories, whereas
without fetal alcohol spectrum disorder in the Northern Cape Province,
the nine year olds consistently had the lowest average scores. South Africa. African Journal of Psychiatry, 14(4), 298-305.
If we understand this finding to describe a change over time,
Fried, P. A., O’Connell, C. M., & Watkinson, B. (1992). 60-72 month follow-up
the clinical ramifications are the need for frequent language of children prenatally exposed to marihuana, cigarettes, and alcohol.
testing over time coupled with programming consistent Pediatrics, 13, 383-391.

with a child’s changing needs. The small proportion of Gillam, R. B., & Pearson, N. A. (2004). Test of narrative language, Austin TX:
our sample that exhibited average language development Pro-ed.
was a significant finding and one that requires further
Greene, T., Earnhart, C. B., Martier, S., Sokol, R., & Ager, J. (1990). Prenatal
investigation. It is imperative that we better understand alcohol exposure and language development. Alcoholism: Clinical
drivers of success, whether they relate to a child’s context, Experimental Research, 14, 937-945.

resilience, or to other factors. This research contributes Jones, K., & Smith, D. (1973). Recognition of fetal alcohol syndrome in early
to the development of evidence-informed practice for infancy. Lancet, 2, 999-1001.

providing specific S-LP services as well as comprehensive Kodituwakku, P. W., Coriale, G., Fiorentino, D., Aragon, A. S., Kalberg, W.
rehabilitation services. Strategies to support children with O., Buckley, D.,… May, P. A. (2006). Neurobehavioural characteristics
of children with fetal alcohol spectrum disorders in communities
FASD must target language and communication as part of a
from Italy: Preliminary results. Alcoholism: Clinical and Experimental
multidisciplinary collaborative community and educational Research, 30(9), 1551-1561.
supports for affected children and their families. With
Kodituwakku, P. W. (2009). Neurocognitive profile in children with fetal
the proper services in place, individuals with FASD can alcohol spectrum disorders. Developmental Disabilities Research
communicate to their fullest potential. Reviews, 15(3), 218-224.

Mattson, S. N., Riley, E. P., Gramling, L., Delis, D. C., & Jones, K. L. (1998). A
review of the neurobehavioral deficits in children with fetal alcohol
syndrome or prenatal exposure to alcohol. Alcoholism: Clinical and
References
Experimental Research, 22(2), 279-294.
Adnams, C. M., Sorour, P., Kalberg, W. O., Kodituwakku, P., Perold, M. D.,
Mattson, S. N., & Riley, E. P. (1999). Implicit and explicit memory functioning
Kotze, A.,… May, P. A. (2007). Language and literacy outcomes from
a pilot intervention study for children with fetal alcohol spectrum in children with heavy prenatal alcohol exposure. Journal of
disorders in South Africa. Alcohol, 41(6), 403-414. International Neuropsychological Society, 5(5), 462-471.

Aragon, A., Coriale, G., Fiorentino, D., Kalberg, W. O., Buckley, D., & Gossage, Newcomer, P. L., & Hammill, D. (2008). Test of Language Development
J. P. (2008). Neuropsychological characteristics of Italian children with Primary-4th ed., Torrance, CA: Western Psychological Services.
fetal alcohol spectrum disorders. Alcoholism: Clinical and Experimental
Research, 32(11), 1909-1919. Public Health Agency of Canada (2005). FASD: A Framework for Action.
Retrieved from http://www.phac-aspc.gc.ca/publicat/fasd-fw-etcaf-
Becker, M., Warr-Leeper, G. A., & Leeper H. A., Jr. (1990). Fetal alcohol ca/framework-eng.php
syndrome: A description of oral motor, articulatory, short term memory,
grammatical, and semantic abilities. Journal of Communication Schoenfeld, A. M., Mattson, S. N., Lang, A. R., Delis, D. C., & Riley, E. P.(2001).
Disorders, 23, 97-124. Verbal and nonverbal fluency in children with heavy prenatal alcohol
exposure. Journal of Studies on Alcohol, 62, 239-246.
Bowers, L., Huisingh, R., & LoGuidice, C. (2007). Test of Problem Solving
Adolescent-2nd ed., East Moline, IL: LinguiSystems Inc. Semel, E., Wiig, E. H., & Secord, W. A. (2003). Clinical Evaluation of Language
Fundamentals-4th ed., San Antonio, TX: The Psychological Corporation.
Bowers, L., Huisingh, R., & LoGuidice, C. (2005). Test of Problem Solving
Elementary-3rd ed., East Moline, IL: LinguiSystems Inc. Streissguth, A. (2003). Fetal alcohol syndrome: A guide for families and
communities. Baltimore, Paul H. Brooks Publishing Company.
Canada Northwest FASD Research Group. (2009). Canadian Symposium:
Refining the evaluation and coidentification of the brain in persons Streissguth, A. P., Herman, C. S., & Smith, D. W. (1978). Intelligence, behavior
exposed to alcohol in gestation. Retrieved from http://www.canfasd. and dysmorphogenesis in the Fetal Alcohol Syndrome: A report on 20
ca/files/PDF/BrainSymposium_e-version_12-09.pdf patients. Journal of Pediatrics, 92(3), 363-367.

Carney, L. J., & Chermak, G. D. (1991). Performance of American Indian Sparks, S. N. (1984). Speech and language in fetal alcohol syndrome.
children with fetal alcohol syndrome on the Test of Language American Speech and Hearing Association, 26(2), 27-31.
Development. Journal of Communication Disorders, 24, 123-134.
Thorne, J., Coggins, T. E., Carmichael Olson, H., & Astley, S. J., (2007).
Chudley, A. E., Conry, J., Cook, J. L., Loock, C., Rosales, T., & LeBlanc, N. Exploring the utility of narrative analysis in diagnostic decision making:
(2005). Fetal alcohol spectrum disorder: Canadian guidelines for Picture-bound reference, elaboration, and fetal alcohol spectrum
diagnosis. Canadian Medical Association Journal, 172, S1-S21. disorders. Journal of Speech, Language, and Hearing Research, 50,
459-474.
Church, M. W., & Kaltenbach, J. A. (1997). Hearing, speech, language, and
vestibular disorders in fetal alcohol syndrome: A literature review. Wallace, G., & Hammill, D. (2002). Comprehensive Receptive and Expressive
Clinical Experimental Research, 21, 495-512. Vocabulary Test-2nd ed., Torrance, CA: Western Psychological Services.

Coggins, T. E., Timler, G. R., & Olswang, L. B. (2007). A state of double Wyper, K. R., & Rasmussen, C. (2011). Language impairments in children with
jeopardy: Impact of prenatal alcohol exposure and adverse fetal alcohol spectrum disorder. Journal of Population therapeutics and
environments on the social communicative abilities of school-age Clinical Pharmacology, 18(2), 364-376.

274 Canadian Journal of Speech-Language Pathology and Audiology | Vol. 37, N0. 4, Winter 2014
 The Language Profile of School-Aged Children with FASD

Table 1: Demographic Overview by Age Group

5-8 Years 9 Years 10-12 Years 13-18 Years TOTAL

n=54 % n=19 % n=28 % n=23 % N=124 %

Diagnosis

pFAS 15 27.8 3 15.8 3 10.7 2 8.7 23 18.5

ARND 39 72.2 16 84.2 25 89.3 21 91.3 101 81.5

Sex

Male 38 70.4 11 57.9 18 64.3 11 47.8 78 62.9

Female 16 29.6 7 36.8 10 35.7 12 52.2 45 36.3

Location

Urban 36 66.7 12 63.2 18 64.3 15 65.2 81 65.3

Rural 18 33.3 6 31.6 9 32.1 8 34.8 41 33.1

Previous Involvement with S-LP

Yes 15 27.8 9 47.4 13 46.4 5 21.7 42 33.9

No 39 72.2 9 47.4 14 50.0 18 78.3 80 64.5

Alcohol Use

Alcohol only 21 38.9 5 26.3 11 39.3 9 39.1 46 37.1

Alcohol and tobacco 28 51.9 3 15.8 6 21.4 1 4.3 38 30.6

Alcohol and
4 7.4 8 42.1 10 35.7 13 56.5 35 28.2
other substances

Family Type

Biological Parent 15 27.8 2 10.5 5 17.9 2 8.7 24 19.4

Extended Family 15 27.8 4 21.1 5 17.9 3 13.0 27 21.8

Adopted 2 3.7 3 15.8 2 7.1 1 4.3 8 6.5

Foster Care 22 40.7 8 42.1 14 50.0 17 73.9 61 49.2

Language

English Only 49 90.7 16 84.2 26 92.9 22 95.7 113 91.1

Preferred Language
English (other 5 9.3 2 10.5 1 3.6 1 4.3 9 7.3
languages spoken)

Revue canadienne d’orthophonie et d’audiologie | Vol. 37, N0. 4, hiver 2014 275
The Language Profile of School-Aged Children with FASD

Table 2: CELF-4 Breakdown of Subtests for Each Age Group

5-8 Years 9 Years 10-12 Years 13-18 Years

Core Language Scores

Concepts and
X X X
Following Directions

Word Structure X

Recalling Sentences X X X X

Formulated Sentences X X X X

Word Classes- Total X X X

Word Definitions X

Receptive Language Index

Concepts and
X X X
Following Directions

Word Classes- Receptive X X X X

Sentence Structure X

Semantic Relationships X

Understanding
X
Spoken Paragraphs

Expressive Language Index

Word Structure X

Recalling Sentences X X X X

Formulated Sentences X X X X

Word Classes- Expressive X X X

Language Content Index

Concepts and
X
Following Directions

Word Classes-Total X X X

Expressive Vocabulary X X

Word Definitions X X

Sentence Assembly X

Understanding
X X X
spoken paragraphs

276 Canadian Journal of Speech-Language Pathology and Audiology | Vol. 37, N0. 4, Winter 2014
 The Language Profile of School-Aged Children with FASD

Language Structure Index

Word Structure X

Recalling Sentences X

Formulated Sentences X

Sentence Structure X

Language Content Index

Recalling Sentences X X X

Concepts and
X X
Following Directions

Formulated Sentences X X X

Semantic Relationships X

Table 3: CELF-4 Clinical Tool Overview (Semel, Wiig & Secord, 2003)

Test Form A Test Form B

5-8 Years 9 Years 10-12 Years 13-18 Years

Core Language Index measures general language ability and quantifies a student’s overall language performance. Each
(total test score) Composite Index Score consists of a different compilation of subtests to yield the standardized scores

Receptive Language Index measures overall ability to listen to and comprehend information

Expressive Language Index measures overall production of language and the ability to express thoughts, ideas and feelings

measures various aspects of semantic development, including vocabulary, concept and category
development, comprehension of associations and relationships among words, interpretation of factual
Language Content Index
and inferential information orally presented and the ability to create meaningful semantically and
syntactically correct sentences

LS: measures the receptive and expressive LM: measures the ability to recall spoken
components of interpreting and producing directions, formulate sentences with given
Language Structure (LS) &
sentence structures words, and identify semantic relationships. It also
Language Memory (LM)
provides a measure of the ability to apply working
memory to linguistic content and structure

Revue canadienne d’orthophonie et d’audiologie | Vol. 37, N0. 4, hiver 2014 277
The Language Profile of School-Aged Children with FASD

Table 4: Mean Standard Scores by Index Category and Age Group

Mean
Age Standard Confidence
Index Category Standard p-value
Group Deviation Interval
Score

Core Language Index Score 5-8 69.4 14.1 65.5 to 73.3 0.19

9 60.3 11.4 54.8 to 65.8

10-12 64.3 17.4 57.6 to 71.1

13-18 65.8 20.9 56.8 to 74.8

total 66.0 16.1 63.2 to 68.9

Receptive Index Score 5-8 76.6 12.5 73.2 to 80.0 <0.01

9 65.0 10.6 59.8 to 70.0

10-12 67.8 13.0 62.7 to 72.8

13-18 65.5 18.4 57.5 to 73.4

total 70.5 14.5 67.9 to 73.1

Expressive Index Score 5-8 71.6 14.2 67.7 to 75.5 0.15

9 62.4 12.4 56.5 to 68.4

10-12 67.3 17.1 60.7 to 73.9

13-18 65.1 18.4 57.2 to 73.1

total 67.9 15.7 65.1 to 70.7

Language Structure Index Score 5-8 74.0 14.1 70.1 to 77.9 0.02

9 60.2 10.2 55.3 to 65.2

10-12 65.5 16.9 59.0 to 72.1

13-18 67.6 24.4 57.0 to 78.2

total 68.7 17.2 65.6 to 71.2

Language Content Index Score 5-8 74.1 11.6 71.0 to 77.3 0.20

9 68.0 12.4 62.0 to 74.0

10-12 71.1 13.6 65.9 to 76.4

13-18 69.0 21.9 59.5 to 78.5

total 71.3 14.7 68.7 to 74.0

278 Canadian Journal of Speech-Language Pathology and Audiology | Vol. 37, N0. 4, Winter 2014
 The Language Profile of School-Aged Children with FASD

Table 5: T-test scores by Diagnosis and Index Category

Mean Standard Deviation Confidence Interval p-value (2-tailed)

Index Category Diagnosis
Standard Score

pFAS 65.2 13.1 59.6 to 70.8 0.80

Core Language Index Score
ARND 66.2 16.9 62.9 to 69.5

pFAS 67.9 11.2 63.0 to 72.7 0.33

Receptive Index Score
ARND 71.1 15.2 68.1 to 74.1

pFAS 67.1 13.7 61.2 to 73.0 0.79

Expressive Index Score
ARND 68.1 16.2 64.9 to 71.2

pFAS 69.7 12.6 64.3 to 75.2 0.56

Language Content
Index Score
ARND 71.7 15.2 68.7 to 74.7

pFAS 68.7 11.8 63.6 to 73.8 0.99

Language Structure
Index Score
ARND 68.6 18.3 65.0 to 72.2

End Notes Authors’ Note

1
The Manitoba FASD Centre (formerly the Clinic for Alcohol Correspondence concerning this article should be addressed
and Drug Exposed Children-CADEC) in Winnipeg Manitoba to Shelley Proven, M.Sc., S-LP(C), CCC-SLP, Clinical Resource
was founded in 1999 and has been assessing and diagnosing Speech Language Pathologist; Manitoba FASD Centre,
children with FASD for over 12 years. To date, the clinic has 633 Wellington Crescent, Winnipeg, Manitoba, R3M OA8,
assessed approximately 2300 children and diagnosed over Canada. Email: sproven@rccinc.ca
1200 individuals with FASD. Although the Centre has many
functions, its primary purpose is to provide comprehensive
multidisciplinary assessments, diagnosis and follow-up Received date: November 22, 2012
services to individuals who have had prenatal alcohol
exposure (PAE). Accepted date: August 9, 2013

Acknowledgements

First we would like to acknowledge and thank the families

who support and care for children with FASD. We would
also like to acknowledge the following organizations for
their support: 1) Manitoba FASD Centre; 2) Children’s
Hospital Foundation of Manitoba; and 3) Audiology and
Speech-Language Pathology Services, Health Sciences
Centre. We are incredibly grateful for the time that Brenda
Fjeldsted, Leslie Sarchuk, and Bonnie Gairns invested in
reviewing the work.

Revue canadienne d’orthophonie et d’audiologie | Vol. 37, N0. 4, hiver 2014 279