Articulo 3

JSLHR
Review Article
Effectiveness of Rehabilitation Approaches

Proposed to Children With Severe-to-
Profound Prelinguistic Deafness on the
Development of Auditory, Speech, and
Language Skills: A Systematic Review
Dominique Demersa,b and François Bergerona,b
Purpose: The purpose of this systematic review is to identify Evidence. Every step of the selection and analysis was
and evaluate the available scientific evidence on the made by 2 independent judges.
effectiveness of rehabilitation approaches proposed to Results: Of 1,739 articles listed in different databases,
children with severe-to-profound prelinguistic deafness 38 met the inclusion criteria and were selected for analysis.
on the hearing, speech, and language skills development. The majority of included articles present a relatively low
Method: Databases (PubMed, CINHAL, PsycInfo, Cochrane, level of evidence. Rehabilitation approaches that do not
ERIC, and EMBASE) were searched with relevant key words include signs appear more frequently associated with a
(children, deafness, rehabilitation approach, auditory, speech, better auditory, speech, and language development, except
and language). Studies published between 2000 and 2017 for receptive language, than approaches that included any
were included. The methodological quality of the studies was form of signs.
evaluated with the Quality Assessment Tool for Quantitative Conclusion: More robust studies are needed to decide
Studies, and the level of evidence was evaluated with the on the approach to prioritize with severe-to-profound deaf
Oxford Centre for Evidence-Based Medicine Levels of children.
H
earing loss affects two to six children out of Yoshinaga-Itano, Sedey, & Carey, 1998; Monteiro, Cordeiro,
1,000 children (Fortnum, Summerfield, Marshall, Silva, & Queiroga, 2016; Sarant, Holt, Dowell, Rickards, &
Davis, & Bamford, 2001; Hyde, 2005; Ministry Blamey, 2009). Furthermore, the World Health Organization
of Health and Social Services, 2012), making it one of the (2017) reports a more systemic impact, including functional,
most common deficiencies at birth (Vohr, 2003). Of these emotional, social, and economic spheres.
children, one is diagnosed with a severe-to-profound deaf- For more than 30 years, the development of hearing-
ness (≥ 70 dB HL; Ministry of Health and Social Services, related technologies such as digital hearing aids and cochlear
2012). Many studies have demonstrated the consequences of implants have reduced the impact of deafness by giving
this level of hearing loss on communication, language, and children with hearing loss access, at least partially, to the
education (Mayne, Yoshinaga-Itano, & Sedey, 1998; Mayne, sounds of speech. Furthermore, the introduction of systematic
newborn screening in many countries support early detection
of deafness and, thus, earlier intervention, a key factor in
a the prognostic for babies with hearing loss. While access to
Rehabilitation Department, Faculty of Medicine, Université Laval,
Québec, Canada
new technologies and to newborn screening has the potential
b
Centre Interdisciplinaire de Recherche en Réadaptation et to reduce the consequences of deafness, it does not conse-
Intégration Sociale, Québec, Canada quently imply a typical development (Hyde, 2005). Indeed,
Correspondence to Dominique Demers: the outcome of pediatric cochlear implantation is charac-
[email protected] terized by a large variability (O’Donoghue, Nikolopoulos,
Editor-in-Chief: Frederick (Erick) Gallun & Archbold, 2000). Thus, many children with hearing loss
Editor: Lori J. Leibold maintain an important gap in terms of auditory, speech,
Received April 11, 2018
Revision received October 15, 2018
Accepted July 15, 2019 Disclosure: The authors have declared that no competing interests existed at the time
https://doi.org/10.1044/2019_JSLHR-H-18-0137 of publication.
4196 Journal of Speech, Language, and Hearing Research • Vol. 62 • 4196–4230 • November 2019 • Copyright © 2019 American Speech-Language-Hearing Association
Downloaded from: https://pubs.asha.org Iberoamericana- Instituto Universitaria on 04/05/2024, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions
and language development in comparison with normal hearing However, these reviews did not consider a lot of
children (Kirk, Miyamoto, Ying, Perdew, & Zuganelis, 2000). existing literature on the subject. In fact, they included a
Some studies have explored the different factors that could limited range of methodological designs. Considering that
predict the development of these children: Age at implantation a lot of research made in the field of rehabilitation with
(Boons et al., 2012; Niparko et al., 2010) and residual pre- deaf children used cohort, cross-sectional, or case study
operative hearing (Chiossi & Hyppolito, 2017) have been designs, it is relevant to make a review with a larger scope
shown to influence the outcome of cochlear implants. in order to look at everything that has been made on the
Rehabilitation also appears as an important factor (Joint subject. Also, previous reviews provided few solutions on
Committee on Infant Hearing, 2007; Monteiro et al., 2016). how to develop more robust evidence on the subject. A
Thus, rehabilitation programs are frequently offered to review addressing limits and providing solutions in order
optimize the efficiency of hearing technologies and support to help and guide future researchers is missing.
the development of auditory, speech, and language skills. The primary purpose of this systematic review is to
Five main rehabilitation approaches are generally proposed explore the effectiveness of these five main approaches on
by clinicians on a continuum of spoken/visual language auditory, speech, and language development of severe-to-
(Gravel & O’Gara, 2003): profound deaf children. More specifically, it aims to identify
1. Auditory–verbal therapy promoting an exclusive the available scientific data with a wider scope than the
access to the sounds of speech, focusing on the auditory previous review, to make a critical analysis of the level of
signal, and developing auditory skills to help children the evidence of these data, and to support clinicians and
develop their abilities (Brennan-Jones, White, Rush, parents in their decision-making process.
& Law, 2014). Second, another purpose of this review article is to
identify and explore the limitations of existing literature
2. Auditory–oral approach where sounds of speech, and to guide future researchers in order to develop higher
lipreading, facial expressions, and natural gestures level findings. Indeed, the designs used, the lack of control
support communication. No form of signs is used. over many variables, and the small sample are often listed
Since most studies do not make distinctions between as important factors lowering the level of evidence in the
auditory–verbal therapy and auditory–oral approaches, those field of rehabilitation with deaf children.
were often grouped under the global term oral communication
(OC). The OC group was defined as oral-only approaches,
excluding approaches that used any kind of signs. Method
3. Total communication (TC), an eclectic intervention The systematic review was carried out according to
integrating sounds of speech, lipreading, facial expres- the systematic review guidelines proposed by Higgins and
sions, natural gestures, and some form of sign support. Green (2011) and Glasziou, Irwig, Bain, and Colditz (2004).
Because there were different uses of the term total These guidelines describe the process to realize a health
communication (TC), it was defined as any approach care systematic review. Both guidelines recommend formu-
that had a focus on spoken language, supported by lating a PICO (population, intervention, comparison,
any form of visual cues, such as sign-supported speech, outcome) question defining the relevant variables. The con-
cued speech, and manually coded English. tribution of two independent judges is emphasized to ensure
4. Bilingual–bicultural, a philosophy promoting a bilin- systematicity.
gual development in a bicultural environment. Sign Research question: As proposed in both guidelines,
language is learned as a first language, and oral lan- the following PICO question was formulated to determine
guage is taught as a second language to help reading/ the focus of the research: What is the efficiency of reha-
writing learning. This approach considers that the bilitation approaches proposed to severe-to-profound pre-
child lives in a deaf environment that is part of a lingual deaf children on auditory, speech, and language
larger community that mostly uses spoken language. development?
Eligibility criteria: The following criteria were used
5. Sign language, exclusively based on visual language, to select studies:
such as American Sign Language. Language: Articles published in French and English.
In the clinical and research contexts, there is no con- Year: Articles published after 1999 (in order to include
sensus on which approach to prioritize for individual children recent technological advances in hearing devices and the
with severe-to-profound deafness (Gravel & O’Gara, 2003; impact of newborn screening).
Hyde, 2005). Recent systematic reviews abound in the same Study design: Any scientific designs, going from case
direction. Brennan-Jones et al. (2014), Fitzpatrick et al. study to randomized controlled trial (RCT; previous system-
(2016), and Kaipa and Danser (2016) concluded that more atic reviews have concluded that there were not enough data
robust data are necessary to validate the efficiency of the in scientific literature to validate the effectiveness of some
auditory–verbal therapy approach. Likewise, Fitzpatrick approaches; these reviews only considered specific research
concluded that there is not enough high-quality data to sup- designs—in this review, all scientific studies are considered
port the choice between an oral or an oral and sign approach. in order to broaden the perspective).
Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4197

Population: Articles that included children aged 0– form included studies’ general information, design and ob-
12 years old with a severe-to-profound bilateral deafness jectives, participants, characteristics, intervention characteris-
(≥ 70 dB HL) that appeared before 3 years of age were tics, outcomes, conclusions, and limits.
included. Children should not have presented any other Quality assessment: Methodological quality was assessed
disability. with the Quality Assessment Tool for Quantitative Studies
Interventions: Any of the five rehabilitation approaches (Effective Public Health Practice Project, 2009). This tool
mentioned previously (auditory–verbal therapy, auditory– focuses on eight indicators (selection bias, study design, co-
oral, TC, bilingual–bicultural, sign language). The terms founders, blinding, data collection methods, withdrawals
oral communication or oral approaches were also included and dropouts, intervention integrity, analyses); a quality
since most of the studies grouped auditory–verbal therapy level (strong, moderate, weak) is assigned to each indicator,
and auditory–oral together. and a general quality level is given to the complete study.
Outcomes: As for study selection, the quality assessment was made
by the two authors on a 10% sample until a 90% interjudge
1. Auditory skills: Skills related to auditory information, agreement was reached.
except for detection. Discrimination, closed-set identi- Level of evidence: The level of evidence is used to
fication or open-set recognition for sounds or speech provide a strength level to the results of articles addressing
and auditory comprehension were included. a specific question in order to assist clinicians in their decision-
2. Speech skills: Skills related to the motor aspect of making process (Burns, Rohrich, & Chung, 2011). It is
speech production, including sound production determined using a ranking system based on the probability
(phonetic inventory and accuracy) and intelligibility. of bias. In the present review, studies were analyzed according
3. Language skills (oral and sign language): Skills including to the Oxford Centre for Evidence-Based Medicine Levels
phonology, vocabulary, morphosyntax, narrative, of Evidence (Phillips et al., 2009). This tool sets the strength
and receptive skills. Reading, writing, and metacogni- of evidence level on a scale from 1 to 5 according to the
tive skills were excluded. research design and the type of study (therapy, prevention,
prognostic, diagnostic, epidemiology, and economy). Level 1
Search strategy: Pubmed, PsycInfo, CINAHL, is given to a systematic review of RCT, Level 2 to RCT,
EMBASE, ERIC, and Cochrane Library were searched Level 3 to cohort studies, Level 4 to cross-sectional and case
between February 13 and February 15, 2017, and then on studies, and Level 5 to mechanism-based reasoning.
September 20, 2017. SpeechBITE, a database specialized in
speech and language pathology, was searched in second
place. Language (English or French) and year (2000 and Results
up) filters were used. Variations of the concepts “children,” Search results: A total of 1,739 articles were identified
“cochlear implant,” “hearing aid,” “deafness,” “language/ in the databases on the first search, excluding duplicates (see
auditory rehabilitation,” and “language/auditory skills” were Figure 1). Following titles and abstracts review, 73 articles
used as key words to find pertinent articles. Appendix A were selected for a complete analysis. Thirty-seven did not
shows key words and synonyms used for the search strategy. meet the including criteria, mainly because they were not
Study selection: Titles, then abstracts, were screened, related to the targeted population and variables of interest.
and every study that did not meet selection criteria was On the second search, a total of 69 new articles were identi-
excluded. Then, full texts of relevant studies were evaluated. fied, and two were included in the review. Thus, a total of
At all stages, the two authors independently reviewed a 10% 38 articles were included in the final review. No additional
portion of the sampling, and a 90% interjudge agreement study was found in the reference section of these articles.
was sought. When it was reached, the first author continued When studies explored more variables than those targeted
the selection alone. When it was not the case, another 10% for the review, only results and conclusions related to the
sample was reviewed until a 90% interjudge agreement was variables of interest were extracted. Table 1 presents the
met. Any disagreement was resolved by discussion and data collected from each selected study (see note at the
consensus. References of all the included articles were also end of Table 1 for abbreviations).
consulted. Study characteristics: Selected studies were published
Data extraction: A data collection form (see Appen- between 2000 and 2017; the majority of the studies were
dix B) was developed for this systematic review. The docu- published between 2000 and 2005 (see Table 2). Most studies
ment was developed based on Higgins and Green’s (2011) were based on a cohort or cross-sectional design; a few
recommendations and on the Data Collection Form for were case studies and one was an RCT. This impacted the
Intervention Reviews: RCTs and non-RCTs proposed by global level of evidence, which was low; on the scale going
(Cochrane, 2014). The form was adapted to the project by from 1 (strong) to 5 (weak), 26 were at Level 4, 11 were at
adding information on hearing loss, the age at onset of deaf- Level 3, and one study was at Level 2. The methodological
ness, and the technological devices used by the children in quality level of most studies was weak (n = 22) or moderate
the participants section and by adding information on reha- (n = 15). One study presented a strong level.
bilitation approaches and modalities of rehabilitation (e.g., As shown in Table 3, studies included between one
duration, providers) in the intervention section. The final and 288 participants, most of them including fewer than
4198 Journal of Speech, Language, and Hearing Research • Vol. 62 • 4196–4230 • November 2019
Figure 1. Flow chart.
50 subjects (n = 22). The majority included cochlear-implanted 2002; Geers, Brenner, & Davidson, 2003; Geers, Brenner,
children only (n = 36); three also included children with et al., 2002; Geers et al., 2017; Kirk et al., 2002, 2000;
hearing aids. Regarding intervention, rehabilitation approaches O’Donoghue et al., 2000; Osberger & Fisher, 2000; Percy-
were one of the main variables of many studies (n = 26), Smith, Cayé-Thomasen, Breinegaard, & Jensen, 2010;
with 12 considering these factors as predictive variables. Phan, Houston, Ruffin, Ting, & Holt, 2016; Tobey et al.,
Thirty-one articles studied OC approaches, 23 studied TC 2000). Geers, Brenner, et al. (2002) also observed a negative
approach, and seven studied the bilingual–bicultural approach. correlation between the number of signs used and the de-
A large range of outcomes were explored: 23 studies addressed velopment of auditory skills. As shown in Tables 4 and 5,
auditory outcomes, 21 addressed speech outcomes, and this observation is constant for all designs and for every pe-
25 addressed language outcomes. Outcomes were mostly riod. Young, Grohne, Carrasco, and Brown (2000) observed
assessed with formal and informal tests and collected by that OC tends to have a faster auditory development,
professionals. In few studies, observations in natural environ- whereas TC catches up after 12 months. The RCT study
ment and surveys for professionals or parents were used in compared OC approaches; results indicated that both groups
complement to the tests. had an equivalent development (Oryadi Zanjani et al., 2013).
Only a few studies reported absolute scores, which is
the score compared to the normative mean. From these
Auditory Skills results, it seems that OC reached the normal range more
The level of evidence, going from 1 to 5, of the studies often than TC (Archbold et al., 2000; Kirk et al., 2002,
focusing on these skills was mostly poor to moderate: 2000). However, Yanbay, Hickson, Scarinci, Constantinescu,
19 studies were at Level 4, eight were at Level 3, and one and Dettman (2014) observed that both OC and TC catch
was at Level 2. Auditory skills were mostly assessed with up 12 months after cochlear implantation. Studies that
repetition of words or sentences presented in quiet. looked at the development of children followed in an ap-
Almost all studies considering OC and TC/bilingual– proach that focused on auditory training (auditory–verbal
bicultural observed that OC was associated with better skill therapy) observed an improvement in auditory skills
development than TC/bilingual–bicultural (Archbold et al., (Diller, Graser, & Schmalbrock, 2001; Fairgray, Purdy, &
2000; Dettman, Wall, Constantinescu, & Dowell, 2013; Smart, 2010; Lertsukprasert, Kasemkosin, Cheewareungroj,
Dowell, Dettman, Blamey, Barker, & Clark, 2002; Geers, & Kasemsuwan, 2010). Diller et al. (2001) reported that

4200
Table 1. Included studies.

Journal of Speech, Language, and Hearing Research • Vol. 62 • 4196–4230 • November 2019
Evidence
Authors, year, Dependent Participants Intervention Conclusions level
country, title Design variables characteristics approach/CM Measures Results of the authors Metho qlty
Archbold Cohort, Speech n = 46, OC/ CM used at -Iowa Closed- Speech perception OC users 3
et al., 2000, prospective perception signs: T1 home and Set Sentence and intelligibility: performed moderate
United Kingdom, (pre-CI, 3–4 Intelligibility n = 22/24, school. Groups Test OC better than better than
Approach to –5 y post-CI) T2 n = classified at -Connected signs at each signs users
Communication, 16/10, T3 every interval. Discourse measure and on speech
Speech n = 13/7, OC: spoken lng. Tracking each interval perception and
Perception deaf: 7 mo, Signs: signs for -CAP ( p < .05) intelligibility.
and Intelligibility CI: 53 mo, all/part of the -Speech Compared OC Authors
After Paediatric CI HL: > day. Spoken Intelligibility group and concluded
60 dB HL English + sign Rating children who it was hard
support (TC or changed to say if OC
simultaneous from signs resulted in
communication). to OC, no better
significant performance
difference or if children
( p > .05) used OC
because they
performed
better.
Connor et al., 2000, Cohort, Speech n = 147, OC/ CM used at school. -Arizona Speech: TC Children 3
United States, prospective (consonant TC: n = 81/ OC: spoken lng Articulation contributed benefited moderate
Speech, (pre-CI, 6 mo production) 66, deaf: alone, with or Proficiency negatively to from using
Vocabulary, and post-CI, 1– Vocabulary 0.33/0.26 y, without the use Scale Revised score and CI in terms
the Education 10 y post-CI) (expressive CI: 5.2/ of lipreading -GFTA rate of growth of consonant
of Children and receptive) 5.69 yo, HL: and visual -Picture ( p < .001). accuracy,
Using CI: Oral severe-to- supports. Some Vocabulary TC had positive receptive and
or Total profound, programs used subtest of the contribution expressive voc.,
Communication? CI use auditory training. Woodcock on acceleration regardless
length: TC: use form of Johnson ( p < .05). of the CM
6 mo to 10 y, sign lng + Tests of The difference used by their
male: n = 71 spoken lng, Cognitive between school.
majority had Ability OC/TC increased
auditory training -PPVT-R/III over time.
and speech-lng Receptive voc.: No
teaching. difference in
scores and
rate of growth
( p > .05).
Expressive voc.:
TC had better
score than OC
( p < .001). No
difference in rate
of growth ( p > .05).
(table continues)
Table 1. (Continued).
Evidence
Cullington Cross-sectional Morphosyntax n = 24, OC/TC: CM used at school. -TACL-R Expressive voc.: This study 4
et al., 2000, (expressive n = 12/12, OC: oral program -Grammatical OC had less suggested weak
United States, and receptive) prelingually or AVT therapy. Analysis of delay than TC that OC had
Comparison of Vocabulary deaf, CI: 73/ TC: use of Elicited Lng (p = .032). better lng
Lng Ability in (expressive 81 mo, CI combination of -PPVT-ΙΠ Receptive voc.: skills than
Children With and receptive) use length: speech, signed -Expressive OC better TC programs,
CI Placed in 38/54 mo, English and Vocabulary than TC, but no even if TC
Oral and Total age at test: finger-spelling. Test significant had longer
Communication 111/136 mo, difference CI use.
Educational male: n = 4/7 (p = .817).
Settings Morphosyntax:
OC better
than TC, but no
significant
difference on
both test
(p = .175/.146/
.110).
Demers & Bergeron: Effectiveness of Rehabilitation Approaches
Kirk et al., 2000, Cohort, Speech n = 103, OC CM -GAEL-P Speech Performance 3

United States, prospective recognition (CI < 2/2–4, OC: No use of -Mr. Potato recognition: was better weak
CI in Young (pre-CI, every Vocabulary > 5 yo)/TC sign lng. Head Task No difference for OC than
Children: Effects 6 mo for at (receptive) (CI < 2/204, TC: combined -PPVT for rate of TC in speech
of Age at least 3 y post- Lng (expressive > 5 yo: n = use of Signed -Reynell Dev growth. OC recognition,
Implantation CI, groups and receptive) 8/37/11/6/ Exact English Lng Scales better than but not for
and CM divided by 33/11, deaf: and spoken TC in scores the lng
CM and age prelingually, English. (GAEL-P: measures.
at CI) CI: 1.61/ p < .02, May be
2.98/6.23/ Potato: explained by
1.72/3.19/ p < .01). the fact that
5.78 yo, HL: Receptive voc.: TC children
> 100, length No difference rely on visual
CI use: 2.5/ in rate of cues that
2.96/1.75/ growth. No were available
2.22/1.94/ effect of CM. for lng
1.72 y Receptive lng: measures,
No effect of but not for
CM. recognition.
Expressive lng:
No effect of
CM.
(table continues)
4201
4202
Evidence
O’Donoghue et al., Cohort, Speech n = 40, OC/TC: CM -Connecting Speech perception: Authors 3
2000, United prospective perception n = 14/26, TC: spoken lng + discourse Score at 5 y concluded moderate
Kingdom, (pre-CI, 3–4– deaf: 11,8 mo, signed input to tracking post-CI correlated that age at
Determinants 5 y post-CI) CI: 52 mo, whatever degree with CM (p = CI and OC
of Speech HL: > 100 dB, OC .00005). Significant are the most
Perception in male: n = 24 Children received at 3/4 y post-CI. important
Children After rehabilitation in CM significant factors to
CI local districts. between-individuals influence later
factor (p = .04). outcome.
Age at CI and CM Study showed
explained 43% of that decisions
the variance. about CI
should be
influenced by
these factors,
but others
yet to be
identified
might also
influence the
outcomes.
Osberger & Fisher, Cohort, Speech n = 57, deaf: CM -Glendonald Speech perception: OC had higher 4
2000, United retrospective perception 0.9 y, CI: Oral-cued speech Auditory CM predicted performance moderate
States, (pre-CI, 3/6/ 5.4 y, HL: TC Screening 12 mo score than TC.
Preoperative 12 mo post-CI) 110 dB HL. Procedure (p < .0001). Difference
Predictors of Divided into Oral children with between OC
Postoperative 4 groups and without pre- and TC less
Implant depending operative speech clear over
Performance on CM and perception had time.
in Children pre-operative smaller 12 mo Few children
score: oral/> post-CI difference with pre-
1%= 15, (77% vs. 48%) operative
oral/< 1% = than the TC groups speech
25, TC/> (61% vs. 14%). perception
1% = 3, TC/< skills used TC.
1% = 14
(table continues)
Evidence
Tobey et al., 2000, Cross-sectional Intelligibility n = 46, TC/AO: CM used in -Imitation of 36 Speech perception: Intelligibility 4
United States, Speech n = 23/23, educational sentences AO better than seemed better weak
Factors perception CI use length: program 3 y (3, 5, or 7 TC (p < .05). for AO than
Associated 4–6 y, age post-CI. syllables Intelligibility: AO TC, but the
With Speech at testing: AO: received 91/ long) better than TC first group
Intelligibility in 8–9 yo 96/88 hr of -LNT (p < .01). More had higher
Children With therapy in the variability across amount of
CI first 3 y post-CI. subjects in TC. therapy, which
TC: received 36/ Moderately strong suggested
49/56 hr of positive relationship that it might be
therapy. was evident an important
between the 2 factor to
variables. consider.
Young et al., 2000, Cohort, Speech n = 43, OC/TC: CM -Deaf Early Speech perception: Study suggested 3
United States, prospective perception n = 23/30 OC: AO, AV, cued Speech 6 mo post-CI, that OC moderate
Speech (pre-CI + 6– Speech congenitally speech during Perception significant acquired
Perception of 12 mo post- identification deaf, CI < the first year Battery difference between auditory skills
Young Children CI) 5 yo, absence after CI. groups (p < .05). faster than
Using Nucleus of medical TC 12 mo post-CI, TC, but TC
22-Channel or conditions no difference appeared to
CLARION (R) CI likely to (p > .05). catch up OC
impede Identification: after 12 mo.
progress 6 mo post-CI,
significant
difference between
groups for spondee
and monosyllable
(p < .000/.001).
12 mo post-CI,
no difference for
monosyllable
(p > .05) and
significant
difference for
spondee (p < .05).
(table continues)
4203
4204
Evidence
Diller et al., 2001, Cohort, Lng skills n = T1: 103, Early natural AV Survey for Hearing: At T2, The children’s 4
Germany, Early prospective Speech T2: 96, T3: program offered professionals, 38.5% reached dev is moderate
Natural AV (1–2–3 y) Hearing skills 61, prelingually in education including normal hearing influenced by
Education of deaf, CI/ institution for items from: lng dev levels. an educational
Children With hearing aids: children with -Denver Dev At T3, 50.5% program that
Profound < 24 mo, HL: hearing Scales reached normal includes
Hearing > 90 dB, CI/ impairment. -Sensomotoric hearing lng dev limited focus
Impairments in hearing aid AV: no sign lng, Dev Scale levels. on natural
the Federal use length: communication -Vademecum Lng/Speech: At T2, auditory–
Republic of T1: 6.2, T2: concentrated on -Munich 37.5% reached verbal. When
Germany: 28.3, T3: hearing, facial Functional normal dev levels. the best
Results of a 41.6 mo, age expression, Dev Test At T3, 38.5% conditions
4-Year Study at survey: natural gesture reached normal are present,
T1: 15, 2T2: and pointing to dev levels. children can
37.4, T3: a normal extend, achieve
50.6 mo, not normal articulation normal
multihandicapped and prosody, hearing/lng
multiple directions. skills with
the same
natural path
with their
own native
lng.
Lachs et al., 2001, Cross-sectional Speech n = 27, OC/TC: Education settings. -Common Intelligibility: OC Early education 4
United States, perception n = 12/15, OC: spoken Phrases Test higher is related to weak
Use of Intelligibility deaf: 0.51 yo, lng including (Auditory scores than TC the intelligibility
Audiovisual CI: 4.52 yo, lipreading. Alone, Visual (p = .04). Was of speech
Information HL: 112 dB HL TC: combination Alone, correlated with and the ability
in Speech of spoken and Audiovisual) audiovisual gain to combine
Perception by manually coded -LNT/MLNT for OC (p = .08), auditory and
Prelingually English. -PBK but not for TC. visual sources
Deaf Children -Beginner’s of information.
With CI: A First Intelligibility
Report Test
(table continues)
Evidence
Dowell et al., 2002, Cohort, Speech n = 102, OA/manual CM used at home. -Closed-set Speech perception: Post-CI CM 3
Australia, prospective perception supplement: n = OA: used an vowel ➔PBK phoneme: CM had a high moderate
Speech (every 6 mo 61/41, deaf: exclusively imitation significantly significant
Perception in post-CI) 1.4 yo, CI: oral CM in the -Closed-set predicted score association with
Children Using 5.9 yo, HL: first 2 y post-CI. monosyllable (p < .001), performance.
CI: Prediction profound, CI Manual supplement: recognition accounted for 18% OA performed
of Long-Term use length: 4 y, used sign lng in -PBK of the variance. better than
Outcomes age: > 4 yo addition to OA -NU-CHIPS OA better scores manual
post-CI. Closed-Set than manual supplement
Consonant supplement. at open-set
Discrimination ➔PBK words: CM speech
Test significantly perception.
-BKB predicted score. The cause and
➔BKB: CM effect
significantly relationships
predicted remained
score. open to
interpretation.
Geers, 2002, Cross-sectional Speech n = 136, OC/TC: CM used in -ESP Speech perception: CM = most 4
United States perception n = 67/69, classrooms. -WIPI CM significantly important weak
and Canada, Speech deaf: 0.4 yo, Rating from 1 -LNT and MLNT predicted score educational
Factors Spoken lng CI: 3.6 yo, age (mostly signs) to -BKB (p < .001). factor for
Affecting the Total lng at test: 9.0 yo, 6 (AV). 1.5 hr of -Video Game Speech: CM auditory/
Dev of Speech, (signs + CI use length: therapy/week. Test of significantly spoken lng
Lng, and spoken) 5.6 y TC: 1–3. Speech predicted score dev. OC
Literacy in Reading OC: 4–6. Pattern (p < .001). associated
Children With Contrast Spoken lng: CM with better
Early CI Perception significantly outcomes.
Geers, Brenner, -McGarr predicted score Use of TC
et al., 2002, Sentences (p < .001). did not
United States -IPSyn Total lng: CM did promote
and Canada, -Narrative not predicted auditory and
Rehabilitation Ability Score score (p > .05). speech dev
Factors -TACL-R Rehabilitation and did not
Contributing factors accounted result in an
to Implant for 12% of the advantage.
Benefit in variance. OC = important
Children choice for CI
children.
(table continues)
4205
4206
Evidence
Geers et al., 2002, Cross-sectional Intelligibility n = 27, deaf: CM used in -WIPI SP, intelligibility, TC did not 4
United States Receptive lng 0.81 yo, CI: classroom. -High-context receptive lng and ensure/prevent weak
and Canada, Expressive lng 3.68 yo, CI TC: sign lng sentences expressive lng: spoken lng
Use of Speech (morphosyntax, length use: along with devised by Significantly dev. Could
by Children lexical diversity) 5.47 y, age at speech for McGarr or and positively not conclude
From TC test: 9.14 yo, ≥ 3 y post-CI. sentences from associated with whether speech
Programs Who normal nonverbal Most children the Beginner’s the % of words or sign was
Wear CI IQ, monolingual had full-time Intelligibility Test produced in more important
English home special -TACL-R speech (p < .01), to initial lng dev.
environment, no education -IPSyn but negatively The influence
additional until 2 y correlated with % direction
educationally post-CI. of words produced between
significant in sign (p < .01). speech use
disabilities Successful CI and the
users preferred outcomes is
speech. unclear. It
There was a significant would be
negative correlation prudent to
(p < .001) between encourage
a child’s use of speech use
speech and use to optimize
of sign. auditory, lng,
and speech
dev.
Kirk et al., 2002, Cohort, Speech n = 73, OC (CI < 3/> CM -Mr. Potato Speech recognition: OC demonstrated 3
United States, prospective recognition 3)/TC (CI < 3/> OC: No use of sign Head More gradual rate more rapid weak
Effects of Age Lng (expressive 3): n = 25/17/16/ lng. -Reynell Dev of growth for TC gains in
at Implantation (pre-CI, every and receptive) 16, deaf: 0.18/ TC: combined use Lng Scales than OC. communication
in Young 6 mo for at 0.29/0.18/0.08 yo, of Signed Exact Receptive lng: No abilities than
Children least 3 y post- CI: 2.13/3.89/ English and difference in growth TC in speech
CI, groups 2.24/3.90 yo, spoken English. of rate. recognition
divided by CM HL: ≥ 108, no Expressive lng: Dev and expressive
and age at CI) additional was delayed for lng, but not in
handicapping TC in comparison receptive lng.
condition with OC.
(table continues)
Evidence
Cassandro et al., Case study Speech n = 1, deafness CM used at home, -First perception Speech perception: Study supported 4
2003, Italy, (pre-CI, perception diagnostic: in school and categories Pre-CI, had no that sign lng weak
Dev of 6–12 mo Phonetic/ 13 mo, CI: 7 yo, in therapy. -Infant Word perception of does not affect
Communication post-CI) phonology HL: severe OC: with father, at Identification speech structure. negatively dev
and Speech Vocabulary school, in therapy. Test At 6/12 mo post-CI, of speech if
Skills After CI (expressive Signs: Italian Sign -Phonologic could identify words. rehabilitation
in a Sign Lng and receptive) Lng with mother Evaluation Went from < 25% program is
Child Morphosyntax and deaf of Infant Lng of word adapted to
(receptive) members of the Tests identification to > the needs of
Nonverbal family. -PPVT 80% 12 mo post-CI. he child.
communication -MacArthur Phonetic: Pre-CI, Signs used by
questionnaire had 4 present the child
-Grammar phonemes and 4 seemed to be
Comprehension occasionally progressively
Test for present. At 12 mo replaced by
Children post-CI, had 17 OC
-Analysis of present and 3 The two CM
spontaneous occasionally appeared to

production present phonemes. have enriched
Vocabulary: Pre-CI, each other.
had 210 expressive
lexical items. At
12 mo post-CI,
had 590 terms.
Receptive lexical
age of 3.10 yo at
12 mo post-CI.
Morphosyntax: Pre-CI,
performance
corresponding
at the lower limit
of a child of 3.4 yo.
At 12 mo post-CI,
was corresponding
to a child of 5 yo.
➔expressive: MLU
went from 2.7
to 7.6 at 12 mo
post-CI.
(table continues)
4207
4208
Evidence
Chin, 2003, Cross-sectional + Phonology n = 12, OC/TC: n = Consecutive use -Picture Naming Phonology: TC were The inventory of 4
United States, descriptive 6/6, deaf: 0.48/ of the same CM. missing more OC contained weak
Children’s 0.53 yo, CI: 3.53/ OC: exclusively English segments more English
Consonant 3.73 yo, HL: spoken lng. than OC (p < .01). sounds than TC.
Inventories 110.32/114.5 dB TC: spoken + sign No significant TC tended to
After Extended HL, age at test: lng, any degree difference for non- have more
CI Use 8.95/10.77 yo, of formal sign English segments. non-English
CI use length: lng. OC had more sounds.
5.42/7.03 y, male: complex inventory
n=6 than TC.
Geers et al., Cross-sectional (A) Speech n = 181, OC/TC: CM used in classroom. (16) (16) Speech perception: (16) Education 4
2003, United perception n = 92/89, deaf: OC: from cued speech -Video Game Classroom CM focusing on weak
States and (B) Receptive lng 0.4 yo, CI: 3.5 yo, to AV. Test of was the only OC was the
Canada, Factors Expressive lng CI length use: TC: from sign Speech educational variable most important
Associated (spoken+ signs, 5.6 y, age at emphasis to speech Pattern independently rehabilitation
With Dev of morphosyntax, first hearing aid: emphasis supported Contrast contributing to the factor
Speech lexical diversity, 1.3 y, age at test: by signs. Perception score (p < .0001). associated
Perception narrative skills) 8.11 yo, male: -ESP Relation with CM with good
Skills in Children Verbal reasoning n = 90, IQ: 102, -WIPI even after all other dev.
Implanted by no additional -LNT factors have been Better use of the
Age Five educationally -BKB removed. information
Geers et al., 2003, significant -Children’s (17) Receptive lng: provided by
United States disabilities Audio-Visual No significant the CI to
and Canada, Enhancement differences understand
Lng Skills of Test between OC/TC in speech
Children With (17) lng comprehension. associated
Early CI -TACL-R Expressive lng: OC with OC
-IPSyn better than TC on program.
-WISC-III narrative ability (17) The study
-Narrative scores (p < .0009), showed that
Ability Score utterance length better linguistic
(p < .0001), lexical skills were
diversity (p < .0001), associated
bound morpheme with CM that
(p < .0001), emphasized
syntax complexity the dev of
(p < .0001). speech and
Educational variables auditory skills.
accounted for 12%
of the variance.
(table continues)
Evidence
Tobey et al., Cross-sectional Intelligibility n = 131, TC/AO: CM used in -Imitation of Intelligibility: CM Higher 4
2004, United pre-CI: n = 60/71, educational 36 sentences at the time of the intelligibility weak
States and 1 y post-CI: n = program used (3 or 5 study: AO better associated
Canada, 55/76, 2 y post-CI: pre-CI, 3 y syllables long) than TC (p < .001). with AO. Early
Mode of n = 61/70, 3 y post-CI and at CM pre-CI: AO better emphasis on
Communication post-CI: n = 59/72, the time of the than TC (p < .001). OC and
and Classroom current time: n = study. 1 y/2 y/3 y post-CI: auditory skill
Placement 57/74, CI: 3.63 yo, AO: cued speech, AO better than TC may have a
Impact on CI use length: AO and AV (p < .001). later impact
Speech 5.33 y, age at programs. No on the ability
Intelligibility testing: 8–9, male: signs. to use the
n = 66, normal IQ TC: any amount of auditory
signs. information to
produce more
intelligible
speech.
Dornan et al., Cohort, General lng n = 58 AV program: centers -PLS-4 OR General lng, The study 3
2007, Australia, prospective Vocabulary Matching groups: followed the CELF-3. receptive voc. supported moderate
Outcomes of (pre-CI, 9 mo (receptive) Deaf: n = 29, principles of the -PPVT-3 and speech: that children
an AV Program post-CI) Speech prelingually deaf, AV approach. -GFTA-2 Significant in an AV
for Children HL: 76.17 dB, 20 mo, weekly -Parent progress between program
with Hearing age: 3.79 yo, attendance to survey T0 and T1 in both made
Loss: A male: n = 21, AV therapy for developed groups for all significant
Comparative English-only lng min. of 6 mo. by the tests (p < .001) progress in
Study with a used by parents. Early intervention, authors and no significant lng dev and
Matched Group Hearing: n = 29, age: frequent audiology interactions were speech
of Children with 2.97 yo, male: n = follow-up. found between production
Normal Hearing 21, no phonetic groups for scores over a 9-mo
delay, no cognitive and rate of growth period.
or physical (p < .05). It supported
disabilities, that the
English-only lng progress is
used by parents the same for
normal
hearing/
deaf children.
Tobey et al., Cross-sectional Speech n = 173, TC/AO: n = CM used in -Imitation of Speech: AO Results continued 4
2007, United 88/85, deaf: < educational 36 sentences better than to confirm a weak
States and 3 yo. CI: 3.5 yo, settings. (3, 5, or 7 TC (p < .0001) better phoneme
Canada, CI use length: AO: cued speech, syllables long) in overall accuracy in AO
Phoneme 5.7 y, first hearing AO and AV phoneme than TC.
Accuracy as a aid: 1.3 yo, age programs. accuracy. More opportunities
Function of at testing: 8–9 yo, TC: any amount of Difference more to practice
Mode of normal nonverbal signs. reflected in auditory/speech
Communication intelligence consonant skills is an
in Pediatric CI (p > .01) than advantage for
vowels (ns). phoneme
accuracy.
4209
(table continues)
4210
Evidence
Jimenez et al., Cross-sectional Intelligibility n = 18, deaf: 10 mo, Educational -Induced Intelligibility: Children who 4
2009, Spain, A Vocabulary CI: 3.2 yo, length environment Phonological Monolingual better used spoken weak
Comparative (receptive) of CI use: 3,1 y, before and after Register than bilingual + sign lng
Study of Psycholinguistic age: 6.25 yo CI. -PPVT (p < .05). achieved
Speech Dev skills Monolingual: -Illinois Test of Vocabulary: No better results
Between Deaf Adaptive behaviors spoken lng. Psycholinguistic significant in aspects
Children With Bilingual: spoken + Abilities difference like manual
CI Who Have sign lng. between the groups expression,
Been Educated for average delay comprehension
With Spoken and performance of visual
or Spoken + (p > .05). symbols and
Sign Lng Psycholinguistic: verbal fluency.
Bilingual better Children who
than monolingual used only
in manual expression spoken lng
and verbal fluency achieved
(p < .05). Monolingual better
better in auditory pronunciation,
reception and oral
association, and comprehension
grammatical rules and use of
(p < .05). No grammatical
significant difference rules.
in visual closure, Signs can be
auditory and visual an important
sequential memories, tool to facilitate
and visual association communication
(p > .05). when technical
problems with
CI.
(table continues)
Evidence
Fairgray et al., Cohort Lng (receptive n = 7, deaf: 0:11 to Therapy approach. -CELF-4 Lng: Improvement Therapy with a 4
2010, New (pretherapy + and expressive) 3:2 yo, CI: 1:0 to AV: 1 hr/week. -Hodson in Understanding listening and weak
Zealand, posttherapy) Phonology 5:1 yo, HL: Practice goals Assessment Spoken spoken lng focus
Effects of AV Speech moderate-to- of the week. of Phonological Paragraphs seemed to be a
Therapy for Speech perception profound, age: Lipreading Patterns (p < .05), but successful form
School-Aged Reading 5–17 yo, male: minimized. Edition not in Concepts of intervention for
Children With n = 2, IQ in the Practice at home. -New Zealand and Following children with
Hearing Loss: average Articulation Directions and hearing loss.
An Exploratory Test expressive lng. Study highlighted
Study -LNT Phonology: Error the need for these

scores improved children to have
significantly rehabilitation, not
(p = .028). just assistance
Speech: Error in the first years.
score improved
significantly
(p = .018).
Speech
perception:
Significant
improvement
for easy
(p = .046) and
hard words
(p = .018).
(table continues)
4211
4212
Evidence
Lertsukprasert Cohort, Listening skills n = 27, HL: 104 dB Aural rehabilitation Not mentioned Listening skills: Following aural 4
et al., 2010, prospective, Vocabulary HL, male: n = 12, program. Age at After 6 mo, 20 rehabilitation, weak
Thailand, descriptive Morphosyntax hearing aid length: enrollment: could detect there was
Listening and (every 6 mo Speech 3 mo, age at 2.10 yo, length: sounds. At 18 mo, evidence of
Speaking Ability for 18 mo) testing: < 6 yo, no 1.5 y. 14 could identified dev of listening
of Thai Deaf other handicaps Groups of 4–6 speech sounds and and speaking
Children in paired by 6 were at the level ability. It was
Preschool Aural performance of comprehension. found to be
Rehabilitation level, once a Speech: After 6 mo, useful for deaf
Program week, 3 hr/ 21 could imitate children.
session. speech sounds. At Auditory training
Session: 45 min of 18 mo, 16 produced is important for
auditory training, spontaneous speech, improvement
50 min of group 11 produced of listening
conversation, prompted speech, and lng dev.
15 min of Morphosyntax:
individual training, After 6 mo, 18
10 min parent’s could produce
counseling. single words. At
18 mo, 11 could
produce phrases
and 12 could
produce sentences.
Vocabulary: After
6 mo, average of
≈ 20 words. At
18 mo, more than
120 words.
(table continues)
Evidence
Percy-Smith Cross-sectional Auditory awareness n = 155, spoken/ CM at home. -Tait video CM was The study 4
et al., 2010, Receptive lng spoken + sign Spoken only analysis significantly documented a weak
Denmark, Phonology support/spoken + Spoken + sign -Reynell test predicting all high influential
Parental Vocabulary sign lng: n = support -Phonological test scores. effect of the CM
Mode of Auditory skills 40/85/30, CI: 3 yo, Spoken + sign lng ‘Sproglydstesten’ Auditory used at home. It
Communication Intelligibility male: n = 70 (Speech Sound awareness: suggests a clear
Is Essential for Test) p < .0001. advantage of
Speech and Lng -Viborg materialet Odds of high spoken lngCM vs.
Outcomes in -CAP scoring: sign support/lng.
Cochlear- -Speech spoken > sign
Implanted intelligibility lng = 28 times.
Children rating Receptive lng:
p < .0001.
Odds of high
scoring:
spoken > sign
support/lng =
61.82 / >
100 times.
Phonology:
p = .0148.
Odds of high
scoring: spoken
> sign lng =
41,25 times.
Vocabulary:
p = .0025.
Odds of high
scoring:
spoken/sign
support >
sign lng = >
100/21.20
times.
Auditory skills:
p = .0010.
Odds of high
scoring:
spoken > sign
support/lng =
> 100 times.
Intelligibility:
p = .0017.
Odds of high
scoring:
spoken >
sign lng =
10,66 times.
4213
(table continues)
4214
Evidence
Boons et al., Cohort, Receptive lng n = 288, OC/TC/Bi: CM used at home -Reynell Dev Lng Receptive lng: CM CM predicted 4
2012, Belgium retrospective Expressive lng n = 102/166/20 OC: no signs. Scales significant word and moderate
and the (1–2–3 y (vocabulary, prelingually deaf, TC: signs to -Schlichting predictor 3 y comprehension
Netherlands, post-CI) morphosyntax) CI: < 5 yo, age support spoken Expressive post-CI (p = .006). dev 3 y
Predictors of at test: 2–8 yo, lng. Lng Test OC better than post-CI, but
Spoken Lng male: n = 152, Bi: spoken Dutch + TC/Bi (p < .001). not sentence
Dev Following normal intellectual Dutch sign lng. Vocabulary: CM dev.
Pediatric CI abilities significant Cannot state if
predictor 3 y OC caused
post-CI (p = .040). good lng or
OC better than that good
TC/Bi (p < .001/ lng leaded
p = .001). to OC.
Morphosyntax:
CM not significant
predictor (p > .05).
Kanda et al., Cross-sectional Interpersonal n = 33, higher/below: Type of education. -Test for question The group that had Results showed 4
2012, Japan, (2 groups; higher communication n = 14/19, CI: Higher vs. below and answer scores above the that 4 factors weak
What Factors vs. below skills 1.6–6.3 yo, CI average: interaction dev national average were different
Are Associated national average Vocabulary (fluency, length use: 76/ AV/oral education: -Word fluency had longer period between the
With Good for children receptive) 57 mo, age at first Average of 8 y vs. test of visits at the 2 groups.
Performance with hearing Syntax (receptive hearing aid: 4 mo 6.7 y in AV/oral -PPVT-R hearing center Higher group
in Children With difficulty) and expressive) to 5.4 yo, HL of education. -Standardized (p = .049), longer (a) had longer
CI? From the Reading CI side: 115/ Sign or cued speech comprehension wearing period visit periods
Outcome of Writing 113 dB HL, HL test of abstract for CI (p = .02), at the hearing
Various Lng Dev of nonoperation words educational center; (b) had
Tests, Research side: 102.1/97.1 -Syntactic methods their CI earlier;
on Sensory and dB HL, age: processing concentrated on (c) were
Communicative 48–155 mo old test of aphasia AV/oral education exposed to
Disorders -Criterion-referenced (p = .003), and more AV/OC;
Project in Japan: testing for attended regular (d) had been
Nagasaki Japanese lng kindergarten/ integrated in
Experience and math. nursery school a regular
before elementary kindergarten
school (p = .01). before
There was no elementary
significant school.
difference in the Authors
HL (p > .05). suggested
that taking
these
factors into
consideration
will have an
effect on
lng dev.
(table continues)
Evidence
Dettman et al., Cohort, Vocabulary n = 39, AV/AO/ Early intervention -PPVT Vocabulary: This study 4
2013, Australia, retrospective (receptive) bi-bi: n = 8/23/8, approach. -CNC word lists Significant relation supported moderate
Communication (post-CI, all Speech congenital AV: speaking lng -BKB between CM and focus on
Outcomes for collection perception deafness, CI: only, no visual age equivalent oral/aural
Groups of points 2/1.6/1.8 yo, cues. ( p =.001)/delay input for
Children Using available) HL: 98.5/104.8/ AO: spoken lng + ( p =.022). optimal
CI Enrolled in 107.6 dB HL, CI lipreading. AV better than AO/ spoken
AV, AO, use length: 4.3/3.4/ Bi-bi: natural sign bi-bi for score communication
and Bilingual- 3.2 y, age at test: lng as first lng, ( p < .05). AV better outcomes.
Bicultural Early 6.2/4.9/4.9 yo, then spoken lng. than bi-bi for delay It found that
Intervention male: n = 39, ( p < .05), but not CM with more
Programs no additional than OA ( p > .05). emphasis
disabilities, normal All equivalent in on oral lng
IQ, monolingual rate of growth performed

English speakers, ( p > .05). better on
same approach at Speech perception: speech
all collection significant relation perception
points between CM and and lng
AV had significantly phoneme/word measures.
more device scores ( p = .001). When subjects
experience and AV/OA better than were individually
were older in age bi-bi for phonemes matched on
than AO/bi-bi. ( p < .05), AV better demographic
than OA/bi-bi for variables,
word score (p < the age at
.05). AV/OA/bi-bi implantation
equivalent for became the
sentence ( p > .05). most significant
CM not significantly factor for lng
predicting outcomes performance.
when subjects
individually matched
( p > .05).
(table continues)
4215
4216
Evidence
Mouvet et al., Case study Pragmatic n = 1, deaf: prelingually, CM of the mother -Child–mother Expressive lng: Study showed 4
2013, Belgium, (when the Expressive lng CI: 10 mo HL: > and nursery. interactions Communicative that the child weak
The Lng Dev child was 90 dB Bi-bi: sign + spoken -Pragmatics outputs decreased seemed to
of a Deaf Child 6–9–12–18– lng. Profile of Everyday in the first year, be profiting
with a CI 24 mo) Communication then triples at 18 from a bi-bi
-Profile of Actual (29–95), but approach up
MacArthur–Bates dropped at 24 mo to 9 mo of
Communicative (65). age. The
Development In the first year, more child made
Inventory functions in visual– important
gestural. In the progress in
second year, both spoken
expressed more Dutch and
functions in oral. Flemish Sign
Total sentence Lng, with a
production went possible
from 6 at 9 mo to onset of
20 at 9 mo. functional
Comprehension code switch.
went from 122 at
18 mo to 304 at
24 mo.
Pragmatic:
Communicative
functions
Oryadi Zanjani RCT (pre-/ Lng (receptive n = 22, auditory Educational program -Newsha Lng: No difference Study 2
et al., 2013, postintervention) and expressive) only/auditory– at school and at Development in the intervention recommended strong
Iran, Comparing Morphosyntax visual: n = 11/11, home. Scale efficiency index not removing
the Effect of Speech deaf: congenital, Auditory only: no -Picture Description ( p = .0598). lipreading
Auditory-Only Auditory skills CI: 42.1/37.0 mo, lipreading. Conversation Auditory skills: No cues in
and Auditory– (detection, age: 45.45/ Auditory–visual: Methods difference between rehabilitation
Visual Modes discrimination, 42.81 mo, CI use access to -Tavana Test the groups ( p = programs.
in Two Groups identification, length: 2.36/ lipreading. -Speech Intelligibility .376). Both groups
of Persian comprehension) 4.64 mo, male: Same educational Measurement Morphosyntax: No acquired
Children Intelligibility n = 6/7, normal program: auditory Test difference between auditory, lng,
Using CI: A IQ, no other training, lng -Mother’s report the groups ( p = and speech
Randomized disabilities stimulation, speech .732). skills at the
Clinical Trial correction. Intelligibility: No same rate.
3½ hr/day, 13 mo. difference between Lipreading does
the groups ( p = not affect the
.237). outcomes of
Both groups intervention,
developed their so it is not
skills significantly necessary
( p < .001). to limit its
access.
(table continues)
Evidence
Park et al., 2013, Cohort, Speech n = 28, with deaf/ AVT: twice a week -CAP Speech: No Children with 4
Korea, Auditory retrospective Speech normal hearing for 2 y. But most -Ling’s stage difference between deaf parents moderate
and Speech (pre-CI, 3, 6, perception parents: n = of the children the groups at every and with
Performance in 12 and 24 mo 14/14, deaf: with deaf parents time ( p < .05). normal
Deaf Children post-CI) prelingual/ also used sign lng. Speech perception: hearing
With Deaf postlingual: n = No difference parents
Parents After CI 22/6 CI: 4.0/ between the develop
3.9 yo, male: groups at every similar
n = 6/6 time ( p < .05). speech
perception
and speech,
when they
have an
additional
normal
hearing
caregiver to
support oral
development.
Sign lng is
not harmful.
Davidson et al., Cross-sectional General lng n = 25, deaf/hearing: CM used at home. -PLS General lng: Age Early knowledge 4
2014, United Vocabulary n = 5/20, congenital Bi-bi: natural -Expressive equivalent was of a sign lng weak
States, Spoken (expressive) deafness, CI < sign lng (American Vocabulary Test predicted by age did not
English Lng Dev Speech 3 yo, age at test: sign lng) + spoken -GFTA-2 ( p < .001), but not prevent
Among Native Literacy skills 4–6.4 yo/6 yo, CI English. -Dynamic by hearing status subsequent
Signing Children Morphosyntax use length: 1.1– Indicators of ( p = .55). spoken lng
With CI 4.8 y, at least one Basic Early Vocabulary: Age dev using a
deaf parent. Literacy Skills equivalent was CI.
-IPSyn predicted by age Sign lng might
( p < .001), but not well lead to
predicted by greater
hearing status success
( p = .997). with such
Speech: Every dev.
subject performed The study
in the normal range. suggested
Morphosyntax: CI that bi-bi
children had a should be
score considered considered
successful for CI as a serious
users of the same option.
age (> 75).
(table continues)
4217
4218
Evidence
Rinaldi & Caselli, Case study Vocabulary n = 1, deaf: prelingually, Rehabilitation -Il Primo Vocabulary: Growth Study 4
2014, Italy, Lng (from 2.6 to (receptive CI: 2.5 yo, first approach. Vocabolario del rate in spoken suggested weak
Dev in a Bi-Bi 5.1 yo, every and expressive) hearing aid: 1.8 yo, Bi-bi: simultaneous Bambino Sordo lng equivalent to that the use
Child With CI: 5–8 mo) HL: severe, normal use of spoken (adaptation of normal of signs in
A Longitudinal IQ lng and lexical MacArthur– ➔Primo vocabolario: early stages
Study signs from sign Bates) At 2.6 yo, produced can help the
lng and use of -Picture Naming 17 labels (spoken construction
grammatical Game + signs). At 3.5 yo, of conceptual
structure of the -PPVT-R produced 100 representations
spoken lng. -Boston Naming labels. Spoken and support
Exposed to bi-bi Test labels went from the dev of
since 1 yo. 2 to 67. Shift of spoken lng.
Therapy started at CM from 2.6 to Signs gave the
1.8 yo. 3.5 yo. opportunity
➔Picture naming: to understand
At 2.6 yo, 90% of what is
correct answers happening
expressed by in the
signs. At 3 yo, environment,
46% expressed ask questions
in bimodal. Gradual and understand
shift in CM. answers.
➔PPVT: At 3.11 yo,
understood 29
spoken words. At
5.1 yo, understood
60 spoken words.
➔Boston naming: At
3.11 yo, produced
23 spoken words.
At 5.1 yo, produced
60 words.
(table continues)
Evidence
Yanbay et al., Cohort, Vocabulary n = 42, AO/Sign + Rehabilitation -PPVT Vocabulary: No Children with CI 4
2014, Australia, retrospective (receptive) spoken/AV: n = program. -PLS-4 significant can dev age- moderate
Lng Outcomes Auditory 14/10/18, deaf: AO: Natural OC. difference between appropriate
for Children comprehension 0.42/0.68/0.34 yo, Lipreading, groups ( p > .05). lng regardless
With CI Lng (expressive) CI: 1.21/1.53/ natural gesture, Auditory comp.: No of the approach
Enrolled in 1.04 yo, HL: > facial expressions. significant received.
Different 95 dB HL, CI use Began program at difference between
Communication length: 3.55/2.72/ 0.88 yo. In the groups ( p > .05).
Programs 2.59 yo, age at first program for 3.85 y. Expressive lng: No
hearing aid: 0.54/ Sign + spoken: significant
0.8/0.43 yo, age at Began program difference between
testing: 4.6/4.31/ at 1.33 yo. In the groups ( p > .05).
3.64 yo, male: n = program for 2.9 y.
7/4/8, no other Any use of sign +
handicaps that spoken lng.
could affect AV: Communication
communication dev via hearing alone.
Began program at
0.66 yo. In the
program for 2.92 y.
Klein & Wie, 2015, Cohort, Narrative skills n = 42, deaf/hearing: CM used by parents -Translation of Information: Parent Higher AV use (vs 3
Norway, Long- prospective (information, n = 18/18, deaf: and educators. Renfrew Bus and educator CM sign support or moderate
Term Predictors (3, 6, 9, 12, 24, MLU, number prelingually, CI: Scale from 1 to Story Test (only not correlated lipreading) soon
of Narrative 36, 48, 72 mo of subordinate 11.5 mo, first 5 Sign lng: 1. at 72 mo post- ( p > .05). after CI was
Skill in Children post-CI) clauses) hearing aid: TC: 2/3. CI) (Narrative MLU: Educator CM associated to
With Early 5.4 mo, age at AO: 4, lipreading/ Scoring Scheme) correlated at 12 increased lng
Bilateral CI last assessment: gestures +spoken -LittlEARS parent to 48 mo ( p < complexity after
6–7 yo, male: n = lng. questionnaire .01/.05/.05/.01). 72 mo post-CI.
8/8, no additional AV: 5, no visual support. -Phonetically Parent CM The use of
disabilities affecting Children all used balanced single- correlated at 48 complex AV
lng dev AO/AV as their syllable word mo ( p < .05). lng at school
main CM. recognition test Subordinate clauses: and home may
Parent CM help facilitate
correlated at children with
3/6 mo ( p = .010/ CI in their lng
.006), 12/24 mo dev. Spoken
( p < .05) and lng + visual
48 mo ( p < .01). support may
Educator CM communicate
correlated at 12/ discourse
24 mo ( p < .001/ content as
.05) and 48 mo well as AV, but
( p < .01). lng complexity
Overall narrative: was not as high.
not correlated to
parent or educator
CM ( p > .05).
4219
(table continues)
4220
Evidence
Phan et al., Cohort, Speech n = 21, early/later CM Synthetically Speech discrimination: CM might have a 3
2016, United prospective discrimination interval: n = 17/10, OC/cued produced token CM not predicting greater influence moderate
States, Factors (2-4 weeks OC: n = 13/8, TC: Sign/TC for the vowels significantly on speech
Affecting Speech post-CI, n = 4/2, deaf: /i/–/u/. performance at discrimination
Discrimination 6-9 mo post-CI) prelingually, CI: Habituation phase early interval ( p > skills, OC
in Children 15.6/15.8 mo, HL: (repetition of .05), but was the performing
With CI: 108.7/106.9 dB HL, one vowel until only significant better than TC.
Evidence From age at test: 16.4/ a max. of 15 trials) predictor of the
Early-Implanted 23.2 mo, male: + test phase later interval
Infants n = 8/3, (10 old trials + performance
4 novel trials). ( p = .031).
Mean looking OC better than TC.
times at monitor
during novel vs.
old trials.
(table continues)
Evidence
Geers et al., 2017, Cohort, Speech recognition n = 97, no sign/ Sign use by the -Imitation of 36 Speech recognition: No advantage of 3
United States, prospective Intelligibility short term/long parents/intervention sentences (3, 5, No significative using sign lng moderate
Early Sign Lng (pre-CI, Expressive lng term: n = 35/26/36, program. or 7 syllables difference pre-CI pre/post-CI.
Exposure and post-CI during (semantic, deaf: 0.3/1.2/ No sign: OA, AV long) and 12/24 mo Children with
CI Benefits elementary morphosyntax, 1.3 mo, CI: 19.3/ or cued speech. -Comprehensive post-CI between sign exposure
school) comprehension, 22.1/22.8 mo, No sign use from Assessment of the groups ( p > are more likely
nonliteral age at first hearing baseline to 3 y Spoken Lng .05). Significative to experience
lng, pragmatic) aid: 9.4/10.8/ post-CI. -Speech Recognition difference at delayed lng
Reading 11.5 mo, age at Short-term sign: TC, Index in Quiet 36 mo ( p = .03), elementary
testing: speech baby sign, signing no sign group grades.
recognition: 3 y exact English, better than long No sign exposure
post-CI, intelligibility: signed English, term ( p = .004). developed
6–8.9 yo, spoken sign lng, sign Intelligibility: better intelligibility,
lng: 5–7.9 yo and support, Pidgin Significative suggesting they
9–11.9 yo, male: sign. Sign use at difference between can use speech
n = 18/17/15, IQ: baseline and/or the groups to communicate
94.5/97.4/98.5 12 mo post-CI. ( p < .001). effectively.
Long-term sign: Long term less Focus on early
Same than short- intelligible than spoken input
term, but use at no sign ( p < .001) increases the
baseline and/or and short term probability of

12 mo, and 24/ ( p = .01). achieving spoken
36 mo. Expressive lng: communication.
Significative
difference between
the groups at the
two times ( p =
.04/p = .002).
Difference
increased with
time. No sign
better than long
term at both times
( p = .01/< .001)
and better than
short term at the
second time
( p = .13/.04).
Note. CM = communication mode; Metho qlty = methodological quality; CI = cochlear implant; y = years; OC = oral communication; TC = total communication; mo = months; lng =
language; CAP = capacity of auditory performance; yo = years old; GFTA = Goldman-Fristoe Test of Articulation; PPVT = Peabody Picture Vocabulary Test; voc. = vocabulary; TACL-R =
Test for Auditory Comprehension of Language–Revised; GAEL-P = Grammatical Analysis of Elicited Language Pre-Sentence Level; Reynell Dev. Lng Scales = Reynell Developmental
Language Scales; AO = auditory–oral; LNT = Lexical Neighborhood Test; AV = auditory–verbal; NU-CHIPS = Northwestern University Children's Perception of Speech; IPSyn = Index of
Productive Syntax; bi-bi = bimodal–bilingual; dev = development; MLNT = Multisyllabic Lexical Neighborhood Test; PBK = Phonetically Balanced Kindergarten; BKB = Bench–Kowal–
Bamford; ESP = CID Early Speech Perception Test; WIPI = Word Intelligibility by Picture Identification Test; PLS = Preschool Language Scale; CELF = Clinical Evaluation of Language
Fundamentals; ns = nonsignificant; MLU = mean length of utterance; CNC = consonant–vowel nucleus–consonant.
4221
Table 2. Characteristics of the studies.
studies, and one was a Level 2 study. Studies that looked
at OC and TC/bilingual–bicultural reported a similar pattern,
Design which is that OC approaches were associated with a better
Randomized controlled trial 1 speech development than TC/bilingual–bicultural (Archbold
Cohort, prospective 14
Cohort, retrospective 5 et al., 2000; Chin, 2003; Connor, Hieber, Arts, & Zwolan,
Cross-sectional 15 2000; Geers, 2002; Geers, Brenner, et al., 2002; Geers et al.,
Case study 3 2017; Jimenez, Pino, & Herruzo, 2009; Lachs, Pisoni, &
Year of publication Kirk, 2001; Percy-Smith et al., 2010; Tobey, Rekart, Buckley,
[2000–2005] 20
[2005–2010] 3
& Geers, 2004; Tobey et al., 2007, 2000). As shown in
[2010–2017] 15 Tables 4 and 5, this observation is constant for all designs
Language and periods. Another study that looked at TC observed
English 38 a negative correlation between speech development
French 0
Country
and the use of signs (Geers, Spehar, & Sedey, 2002).
United States 19 RTC compared OC approaches and concluded that they
Australia 4 rendered a similar development (Oryadi Zanjani et al.,
United Kingdom 2 2013).
Other 13
Published
Every study that looked at absolute scores observed
Published 38 that OC was related to more intelligibility, to a higher ex-
Unpublished 0 pected speech (e.g., sound production), or to more chances
of reaching the normal range than TC (Archbold et al.,
2000; Chin, 2003; Connor et al., 2000; Tobey et al., 2004,
50% of the children reached the normal range after 3 years, 2007). Children followed in program focusing on auditory
and Park et al. (2013) observed no difference between the training (auditory–verbal therapy), and using auditory
normal and the deaf groups (with cochlear implant) with technology would also get closer to the normal range or
and without deaf parents. develop typical speech (Diller et al., 2001; Fairgray et al.,
In summary, OC was mostly associated with better 2010; Lertsukprasert et al., 2010; Park et al., 2013). Arti-
auditory development than TC/bilingual–bicultural. Few cles that studies bilingual–bicultural, in which children
studies looked at the absolute level, but it seems that auditory– develop both sign and oral language, reported a posi-
verbal therapy, which focuses on auditory training, could tive impact of the approach used, and some of them
lead to a typical development. observed a typical speech development (Cassandro et al.,
2003; Davidson, Lillo-Martin, & Chen Pichler, 2014; Mouvet,
Matthijs, Loots, Taverniers, & Van Herreweghe, 2013).
Speech Skills (Phonetic Inventory
In short, OC seems to be associated with a better
and Accuracy, Intelligibility) development of speech skills (phonetic inventory and accu-
Most studies focusing on speech skills had a low level racy, intelligibility) than TC. Studies looking at absolute
of evidence: 16 were Level 4 studies, four were Level 3 levels observed that OC had more chances to develop a
Table 3. Sample size distribution.
Table 4. Result distribution per design.
Auditory skills Speech skills Expressive lng skills Receptive lng skills
Design OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC
Cohort, 7 0 2 3 0 1 2 1 3 0 0 5
prospective (77.77%) (0%) (22.22%) (75%) (0%) (25%) (33.33%) (16.66%) (50%) (0%) (0%) (100%)
Cohort, 3 0 1 — — — 1 0 2 2 0 0
retrospective (75%) (0%) (25%) (33.33%) (0%) (66.66%) (100%) (0%) (0%)
Cross-section 6 0 0 7 0 0 7 0 1 2 0 1
(100%) (0%) (0%) (100%) (0%) (0%) (87.5%) (0%) (12.5%) (33.33%) (0%) (33.33%)
Note. Em dashes indicate that no retrospective cohort articles analyzed speech skills. lng = language; OC = oral communication; TC = total
comunication.
typical speech. According to other studies, OC and bilingual– OC approaches and they concluded that there was no dif-
bicultural could lead to typical speech development. ference between them.
When looking at absolute levels, the results are also
mixed. Connor et al. (2000) observed that TC was associ-
Language Skills ated to a better development than OC, but both groups
Expressive language: Twenty-six studies were interested showed some delay. Two studies mentioned that OC and
in expressive language development. They mostly were low- TC showed a similar development with some delay in both
level studies: 19 were at Level 4, six were at Level 3, and only groups (Kirk et al., 2000; Yanbay et al., 2014). Kirk et al.
one was at Level 2. Language skills were mostly evaluated (2002) observed that OC had less delay than TC, and Geers
with spoken language, and signs were sometimes permitted et al. (2017) mentioned that the use of signs was associated
for children with TC/bilingual–bicultural approaches. with a more delayed expressive language. Other studies that
Studies that looked at OC and TC/bilingual–bicultural were interested in bilingual–bicultural, in which both sign
obtained mixed results; nine mentioned that OC was related and oral language are developed, or OC individually
to a better development than TC/bilingual–bicultural (Boons all mentioned that it helped the children’s development
et al., 2012; Cullington, Hodges, Butts, Dolan-Ash, & (Cassandro et al., 2003; Davidson et al., 2014; Diller et al.,
Balkany, 2000; Geers, 2002; Geers, Brenner, et al., 2002; 2001; Dornan, Hickson, Murdoch, & Houston, 2007;
Geers et al., 2017; Geers, Nicholas, & Sedey, 2003; Kanda Fairgray et al., 2010; Lertsukprasert et al., 2010; Mouvet
et al., 2012; Kirk et al., 2002; Percy-Smith et al., 2010), et al., 2013; Rinaldi & Caselli, 2014).
four observed a similar development (Jimenez et al., 2009; Receptive language: Thirteen studies looked at receptive
Kirk et al., 2000; Klein & Wie, 2015; Yanbay et al., 2014), development; most showed a low level of evidence (nine at
and only one study mentioned that TC was associated Level 4 and four at Level 3).
with a better development than OC (Connor et al., 2000). Most studies that were interested in OC and TC
Likely, Geers, Spehar, et al. (2002) looked at TC and observed that neither was associated with a better develop-
concluded that signs were negatively related to expressive ment (Connor et al., 2000; Cullington et al., 2000; Geers,
development. Table 4 shows that higher level studies had Nicholas, et al., 2003; Percy-Smith et al., 2010; Yanbay
mixed results, while cross-sectional studies mostly found et al., 2014). However, four studies observed that OC was
that OC had an advantage on TC. As shown in Table 5, related to a better development than TC/bilingual–bicultural
results were mixed in earlier and more recent studies. Oryadi and one observed a negative correlation between receptive
Zanjani et al. (2013) made an RCT in which they compared language development and the use of signs (Boons et al.,
Table 5. Result distribution per year.
Auditory skills Speech skills Expressive lng skills Receptive lng skills
Year of
publication OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC
[2000–2005] 10 (83.33%) 0 2 (16.66%) 7 0 0 6 1 2 1 0 5

(0%) (100%) (0%) (0%) (66.66%) (11.11%) (22.22%) (16.66%) (0%) (83.33%)
[2006–2010] — — — 1 0 1 0 0 1 0 0 1
(50%) (0%) (50%) (0%) (0%) (100%) (0%) (0%) (100%)
[2011–2017] 4 0 1 2 0 0 4 0 4 3 0 0
(80%) (0%) (20%) (100%) (0%) (0%) (50%) (0%) (50%) (100%) (0%) (0%)
Note. Em dashes indicate that no articles published between 2006 and 2010 analyzed auditory skills. lng = language; OC = oral communication;
TC = total communication.

2012; Geers, Spehar, et al., 2002; Kanda et al., 2012; Kirk would be the evolution of the technology that now gives
et al., 2002, 2000). Table 4 shows that studies with higher some better access to the sound of speech, increasing the
design level all observed an equivalent development for proportion of children in OC groups. Those using TC might
OC and TC, while other designs had more mixed results. be children with more difficulties, introducing a bias in the
As shown in Table 5, OC and TC seems to be equal in selection of the participants. Again, the causality cannot be
earlier studies, but OC appears to be associated with a better established.
development than TC in more recent studies. This review also shows that more research interested
Some studies mentioned absolute scores; most of in absolute level is needed to determine if an approach
them observed that OC and TC had a similar development would lead to skills in the normal range or a smaller delay.
and were equally lower than the normal range (Connor Most of them observed that both OC and TC have delayed
et al., 2000; Kirk et al., 2002, 2000; Yanbay et al., 2014). auditory, speech, and language development, but that
However, three studies observed that TC was related to a OC would catch up faster than TC for auditory and speech
bigger delay (Cullington et al., 2000; Dettman et al., 2013; skills.
Jimenez et al., 2009). All studies on the auditory–verbal The question of children who were in a sign language
therapy approach, in which auditory training takes main environment is still not answered. Many studies reported
places, mentioned that it had been effective (Dornan et al., that signs did not interfere with the children’s development
2007; Fairgray et al., 2010). Another study that looked at (Cassandro et al., 2003; Davidson et al., 2014; Jimenez
a child using bilingual–bicultural approach concluded that et al., 2009; Mouvet et al., 2013; Park et al., 2013; Rinaldi
signs supported oral development (Rinaldi & Caselli, 2014). & Caselli, 2014). In an environment where parents master
To summarize, OC appears to be related to a better sign language, signs would help the child to build a com-
expressive language development than TC/bilingual– munication and language before he accesses oral language.
bicultural, but results are not homogenous. For receptive However, most studies of this review included children
language, most studies found that development was similar with hearing parents and were thus probably exposed to a
for OC and TC. Auditory–verbal therapy and bilingual– limited degree of signs, but the degree of parent’s expertise
bicultural approach could both lead to typical development. in sign language was rarely mentioned. This might have
underscored the performance of children in sign approaches.
Furthermore, sample size of studies including children
Discussion evolving in a sign language environment was usually small,
The objectives of this systematic review were to explore limiting the weight of their conclusions. More research is
the effectiveness of the main rehabilitation approaches on still needed to explore the potential difference in the impact
auditory, speech, and language development of severe-to- of sign approaches between children with deaf parents and
profound deaf children in order to provide evidence to hearing parents.
support clinicians in their decision-making process. The sci- It is important to mention that the overall level of
entific literature shows a general trend, which is that OC evidence of included studies is low. Only one study had a
seems to be linked to a better auditory, speech, and expres- good level of evidence. Therefore, despite the distinct trend
sive language potential than TC/bilingual–bicultural. As in favor of oral approaches, available level of evidence
mentioned previously, more studies showed that OC was does not seem to be sufficient to support the existence of
linked to better skills than TC/bilingual–bicultural. However, differences in auditory, speech, and language development
because of the designs used by studies, it is not possible to from the rehabilitation approaches. This observation is
establish the causality link between these variables; do OC consistent with conclusions of recent systematic reviews,
help children develop better skills, or do children with which all concluded that more data were needed to be able
better skills more likely use OC? Some authors suggest that to guide clinicians and parents in their decision-making
OC leads to a better development of auditory, speech, and process (Brennan-Jones et al., 2014; Fitzpatrick et al., 2016;
expressive language development because children are Kaipa & Danser, 2016).
immersed in an oral–auditory environment (Jimenez et al., Furthermore, the trends noted in Fitzpatrick et al.’s
2009). Furthermore, other authors mention that TC would (2016) review were similar to the present observations for
limit linguistic environment (Kirk et al., 2000). receptive language. On the other hand, trends identified
One exception to this trend is observed for receptive by both reviews were different for expressive language.
language; indeed, TC and OC were more likely to be similar In fact, in the present review, OC seemed to be related to
on this skill. The reasons remain unclear, but it could be a better performance than TC/bilingual–bicultural, but
explained by the fact that receptive tasks were often presented Fitzpatrick et al. observed no difference between the inter-
in the preferred modality, whereas expressive tasks were ventions. This difference could be explained by the fact
presented only in oral language (Kirk et al., 2002). Other the present review included wider range of experimental
authors proposed that TC gives children early access to designs and time lapse, resulting in a larger number of in-
meanings via visual access to language, helping developing cluded studies (n = 38 vs. 11). Still, more data are needed
receptive language and coping hearing loss (Connor et al., to explore these trends.
2000; Rinaldi & Caselli, 2014). This trend appears more con- Studies included in the present review present some
sistent in recent studies (2011–2017). A potential explanation methodological limitations that had an impact on the level
of evidence. Because of the methodological design used in Conclusion
most studies, it is usually difficult to isolate the rehabilitation
factor and certify that the approach led to the difference The present review emphasizes the lack of robust
observed between the groups. A good example of this situa- scientific literature in the field of rehabilitation intervention
tion is emphasized by Dettman et al. (2013), who compared with severe-to-profound deaf children. In fact, even if an
the outcomes of three groups (auditory–verbal therapy, overall trend seems to advantage oral approaches, there is
auditory–oral, bilingual–bicultural). When the performances no strong evidence available to establish which approach is
of the three groups were initially compared, an auditory– to prioritize with these children. However, the results of
verbal therapy advantage was observed. However, when this review revealed a trend in the included articles: OC
subjects were matched on key factors (age at implantation, seems to be related to better auditory, speech, and expressive
age at first hearing aid, age at testing, duration of hearing aid language development than TC approach. The approaches
use), no significant difference between groups was found. appeared to be similar for receptive language.
Another limit lies in the large range of tests used to Further research is thus needed to address the issue
quantify auditory, speech, and language outcomes; it is thus of rehabilitation approaches for deaf children. As discussed,
difficult to compare the results directly from the included the present review showed that studies controlling the effect
studies. Finally, most studies did not clearly define the ap- of potential covariables did not observe the same trend.
proach used, neither the intervention settings (frequency, Studies involving large samples, matched groups, and well-
length, provider) that children were involved in. Since inter- controlled interventions are essential to isolate the interven-
vention is rarely clearly portrayed, it is difficult to come to tion factor and be able to generalize findings. Meanwhile,
a robust conclusion and guide clinicians on their practice. since scientific literature does not support a specific reha-
Future research should try to avoid these limitations bilitation approach with severe-to-profound deaf children,
to bring more robust data to the field and new perspective the choice of the approach should take into consideration
on it. Since many studies were based on retrospective data the particularities and needs of the child, his family, and the
or cross-sectional designs, an important element that should context they are evolving in (Joint Committee on Infant
be considered by future researchers is using a study design Hearing, 2007).
that has fewer methodological limits. Focusing on prospec-
tive and longitudinal data would allow a better control of
the studied variables (e.g., age at onset of deafness, age at Acknowledgments
testing, age at cochlear implantation), enhance the level of The authors would like to acknowledge Marie Denis Lavoie,
evidence, and allow researchers to match the groups on who helped with the search in the databases.
key variables to reduce their influence on the results.
Another element that should be addressed is the
sample size. Since small samples make it more difficult to
References
ensure the representativeness of the population and to mini-
mize sampling bias, having bigger samples would help Archbold, S. M., Nikolopoulos, T. P., Tait, M., O’Donoghue, G. M.,
Lutman, M. E., & Gregory, S. (2000). Approach to communica-
avoid these limits and reduce effects of other variables. The
tion, speech perception and intelligibility after paediatric cochlear
recent systematic review also pointed out the challenge of implantation. British Journal of Audiology, 34(4), 257–264.
rehabilitation approach definitions. Most studies classified Boons, T., Brokx, J. P. L., Dhooge, I., Frijns, J. H. M., Peeraer, L.,
them with the question “did the child use any form of Vermeulen, A., . . . van Wieringen, A. (2012). Predictors of spoken
signs/gestural language,” separating them into OC and TC. language development following pediatric cochlear implantation.
However, rehabilitation approaches are defined by more Ear and Hearing, 33(5), 627–639.
than the use of signs and have very different intervention Brennan-Jones, C. G., White, J., Rush, R. W., & Law, J. (2014).
philosophy. Future studies should define them more clearly Auditory–verbal therapy for promoting spoken language develop-
so that readers know exactly what intervention was in place ment in children with permanent hearing impairments. Cochrane
Database of Systematic Reviews, 3, CD010100.
and compare the different rehabilitation approaches (auditory–
Burns, P. B., Rohrich, R. J., & Chung, K. C. (2011). The levels of
verbal therapy, auditory–oral, TC, bilingual–bicultural). evidence and their role in evidence-based medicine. Plastic
Another element that remains unclear concerns the Recontruction Surgery, 128(1), 305–310. https://doi.org/10.1097/
causes and impacts of changing from an approach to another. PRS.0b013e318219c171
Many studies included children who changed their commu- Cassandro, E., Nicastri, M., Chiarella, G., Genovese, E., Gallo, L. V.,
nication mode but did not include this variable in their & Catalano, M. (2003). Development of communication and
analyses. Archbold et al. (2000) included that variable in speech skills after cochlear implant in a sign language child.
their analyses and observed that children who go from TC Acta Otorhinolaryngologica Italica, 23(2), 88–93.
to OC have an equivalent auditory and speech develop- Chin, S. B. (2003). Children’s consonant inventories after extended
cochlear implant use. Journal of Speech, Language, and Hearing
ment to those who have always been in OC. Other studies Research, 46(4), 849–862.
excluded them, limiting the external validity. It would be Chiossi, J. S. C., & Hyppolito, M. A. (2017). Effects of residual
important to explore this issue to get a broader view of the hearing on cochlear implant outcomes in children: A systematic
difference between the approaches since switching approach review. International Journal of Pediatric Otorhinolaryngology,
frequently happens in the rehabilitation process. 100, 119–127.

Cochrane. (2014). Data collection form for intervention reviews: Geers, A. E., Nicholas, J. G., & Sedey, A. L. (2003). Language
RCTs and non-RCTs. Retrieved from http://training.cochrane. skills of children with early cochlear implantation. Ear and
org/resource/data-collection-forms-intervention-reviews Hearing, 24(1), 46S–58S.
Connor, C. M., Hieber, S., Arts, H. A., & Zwolan, T. A. (2000). Geers, A. E., Spehar, B., & Sedey, A. (2002). Use of speech by
Speech, vocabulary, and the education of children using cochlear children from total communication programs who wear cochlear
implants: Oral or total communication? Journal of Speech, implants. American Journal of Speech-Language Pathology, 11,
Language, and Hearing Research, 43(5), 1185–1204. 50–58.
Cullington, H., Hodges, A. V., Butts, S. L., Dolan-Ash, S., & Glasziou, P., Irwig, I., Bain, C., & Colditz, G. (2004). Systematic
Balkany, T. J. (2000). Comparison of language ability in chil- reviews in health care: A practical guide. Cambridge, United
dren with cochlear implants placed in oral and total communi- Kingdom: Cambridge University Press.
cation educational settings. Annals of Otology, Rhinology & Gravel, J. S., & O’Gara, J. (2003). Communication options for
Laryngology. Supplement, 185, 121–123. Retrieved from http:// children with hearing loss. Mental Retardation and Developmental
www.ncbi.nlm.nih.gov/pubmed/11140984 Disabilities Research Reviews, 9(4), 243–251.
Davidson, K., Lillo-Martin, D., & Chen Pichler, D. (2014). Spoken Higgins, J. P., & Green, S. (Eds.). (2011). Cochrane Handbook
English language development among native signing children for Systematic Reviews of Interventions (Version 5.1.0). London,
with cochlear implants. Journal of Deaf Studies and Deaf United Kingdom: The Cochrane Collaboration. Retrieved from
Education, 19(2), 238–250. http://www.handbook.cochrane.org
Dettman, S., Wall, E., Constantinescu, G., & Dowell, R. (2013). Hyde, M. L. (2005). Newborn hearing screening programs: Over-
Communication outcomes for groups of children using cochlear view. Journal of Otolaryngology, 34(2), S70–S78.
implants enrolled in auditory–verbal, aural–oral, and bilingual– Jimenez, M. S., Pino, M. J., & Herruzo, J. (2009). A comparative
bicultural early intervention programs. Otology & Neurotology, study of speech development between deaf children with cochlear
34(3), 451–459. Implants who have been educated with spoken or spoken + sign
Diller, G., Graser, P., & Schmalbrock, C. (2001). Early natural language. International Journal of Pediatric Otorhinolaryngology,
auditory–verbal education of children with profound hearing 73(1), 109–114.
impairments in the Federal Republic of Germany: Results of a Joint Committee on Infant Hearing. (2007). Year 2007 position
4-year study. International Journal of Pediatric Otorhinolaryn- statement: Principles and guidelines for early hearing detection
gology, 60(3), 219–226. and intervention programs. Pediatrics, 120(4), 898–921.
Dornan, D., Hickson, L., Murdoch, B., & Houston, T. (2007). Kaipa, R., & Danser, M. L. (2016). Efficacy of auditory–verbal
Outcomes of an auditory–verbal program for children with hearing therapy in children with hearing impairment: A systematic
loss: A comparative study with a matched group of children with review from 1993 to 2015. International Journal of Pediatric
normal hearing. The Volta Review, 107(1), 37–54. Otorhinolaryngology, 86, 124–134.
Dowell, R. C., Dettman, S. J., Blamey, P. J., Barker, E. J., & Kanda, Y., Kumagami, H., Hara, M., Sainoo, Y., Sato, C.,
Clark, G. M. (2002). Speech perception in children using cochlear Yamamoto-Fukuda, T., . . . Takahashi, H. (2012). What factors
implants: Prediction of long-term outcomes. Cochlear Implants are associated with good performance in children with cochlear
International, 3(1), 1–18. implants? From the outcome of various language development
Effective Public Health Practice Project. (2009). Quality assessment tests, research on sensory and communicative disorders project
tool for quantitative studies. Retrieved from https://www.ephpp.ca/ in Japan: Nagasaki experience. Clinical and Experimental
quality-assessment-tool-for-quantitative-studies/ Otorhinolaryngology, 5(Suppl. 1),, S59–S64.
Fairgray, E., Purdy, S. C., & Smart, J. L. (2010). Effects of auditory–
Kirk, K. I., Miyamoto, R. T., Lento, C. L., Ying, E., O’Neill, T.,
verbal therapy for school-aged children with hearing loss: An
& Fears, B. (2002). Effects of age at implantation in young
exploratory study. The Volta Review, 110(3), 407–433.
children. Annals of Otology, Rhinology & Laryngology. Supplement,
Fitzpatrick, E. M., Hamel, C., Stevens, A., Pratt, M., Moher, D.,
189, 69–73.
Doucet, S. P., . . . Na, E. (2016). Sign language and spoken
language for children with hearing loss: A systematic review. Kirk, K. I., Miyamoto, R. T., Ying, E. A., Perdew, A. E., &
Pediatrics, 137(1), 1–17. Zuganelis, H. (2000). Cochlear implantation in young children:
Effects of age at implantation and communication mode. The
Fortnum, H. M., Summerfield, A. Q., Marshall, D. H., Davis, A. C.,
Volta Review, 102(4), 127–144.
& Bamford, J. M. (2001). Prevalence of permanent childhood
hearing impairment in the United Kingdom and implications Klein, K. E., & Wie, O. B. (2015). Long-term predictors of narra-
for universal neonatal hearing screening: Questionnaire based tive skill in children with early bilateral cochlear implants.
ascertainment study. BMJ, 323(7312), 536–540. European Journal of Special Needs Education, 30(2), 202–219.
Geers, A. E. (2002). Factors affecting the development of speech, Lachs, L., Pisoni, D. B., & Kirk, K. I. (2001). Use of audiovisual
language, and literacy in children with early cochlear implan- information in speech perception by prelingually deaf children
tation. Language, Speech, and Hearing Services in Schools, with cochlear implants: A first report. Ear and Hearing, 22(3),
33(3), 172–183. 236–251.
Geers, A. E., Brenner, C., & Davidson, L. (2003). Factors associated Lertsukprasert, K., Kasemkosin, N., Cheewareungroj, W., &
with development of speech perception skills in children implanted Kasemsuwan, L. (2010). Listening and speaking ability of Thai
by age five. Ear and Hearing, 24(Suppl. 1),, 24S–35S. deaf children in preschool aural rehabilitation program. Jour-
Geers, A. E., Brenner, C., Nicholas, J., Uchanski, R., Tye-Murray, N., nal of the Medical Association of Thailand, 93(4), 474–480.
& Tobey, E. (2002). Rehabilitation factors contributing to Mayne, A., Yoshinaga-Itano, C., & Sedey, A. (1998). Receptive
implant benefit in children. Annals of Otology, Rinology & vocabulary development of infants and toddlers who are deaf
Laryngology. Supplement, 189, 127–130. or hard of hearing. The Volta Review, 100(5), 29–52.
Geers, A. E., Mitchell, C. M., Warner-Czyz, A., Wang, N.-Y., Mayne, A., Yoshinaga-Itano, C., Sedey, A., & Carey, A.
Eisenberg, L. S., & CDaCI Investigative Team. (2017). Early (1998). Expressive vocabulary development of infants and
sign language exposure and cochlear implantation benefits. toddlers who are deaf or hard of hearing. The Volta Review,
Pediatrics, 140(1), 1–9. 100(5), 1–28.
Monteiro, C. G., Cordeiro, A. A., Silva, H. J., & Queiroga, B. A. cochlear implants: Evidence from early-implanted infants. Jour-
(2016). Children’s language development after cochlear im- nal of the American Academy of Audiology, 27(6), 480–488.
plantation: A literature review. CoDAS, 28(3), 319–325. Phillips, B., Ball, C., Sackett, D., Badenoch, D., Straus, S., Haynes,
Mouvet, K., Matthijs, L., Loots, G., Taverniers, M., & Van Herreweghe, B., & Dawes, M. (2009). Oxford Centre for Evidence-Based
M. (2013). The language development of a deaf child with a Medicine—Levels of evidence. Retrieved from http://www.
cochlear implant. Language Sciences, 35, 59–79. cebm.net/index.aspx?o=1025
Ministry of Health and Social Services. (2012). Programme québé- Rinaldi, P., & Caselli, M. C. (2014). Language development in a
cois de dépistage de la surdité chez les nouveau-nés. La Direction bimodal bilingual child with cochlear implant: A longitudinal
des communications du ministère de la Santé et des Services study. Bilingualism: Language and Cognition, 17(4), 798–809.
sociaux [Quebec’s newborn deafness screening program. Ministry Sarant, J. Z., Holt, C. M., Dowell, R. C., Rickards, F. W., &
of Health and Social Services’ communications direction]. Blamey, P. J. (2009). Spoken language development in oral
Retrieved from http://publications.msss.gouv.qc.ca/msss/document- preschool children with permanent childhood deafness. Journal
000483/ of Deaf Studies and Deaf Education, 14(2), 205–217.
Niparko, J. K., Tobey, E. A., Thal, D. J., Eisenberg, L. S., Wang, Tobey, E. A., Geers, A. E., Douek, B. M., Perrin, J., Skellet, R.,
N.-Y., Quittner, A. L., & Fink, N. E. (2010). Spoken language Brenner, C., & Toretta, G. (2000). Factors associated with
development in children following cochlear implantation. JAMA, speech intelligibility in children with cochlear implants. Annals
303(15), 1498–1506. of Otology, Rinology & Laryngology. Supplement, 185, 28–30.
O’Donoghue, G. M., Nikolopoulos, T. P., & Archbold, S. M. (2000). Tobey, E. A., Rekart, D., Buckley, K., & Geers, A. E. (2004).
Determinants of speech perception in children after cochlear Mode of communication and classroom placement impact on
implantation. The Lancet, 356(9228), 466–468. speech intelligibility. Archives of Otolaryngology—Head &
Oryadi Zanjani, M. M., Hasanzadeh, S., Rahgozar, M., Shemshadi, H., Neck Surgery, 130(5), 639–643.
Purdy, S. C., Mahmudi Bakhtiari, B., & Vahab, M. (2013). Tobey, E. A., Wiessner, N., Lane, J., Sundarrajan, M., Buckley,
Comparing the effect of auditory-only and auditory–visual modes K. A., & Sullivan, J. (2007). Phoneme accuracy as a function
in two groups of persian children using cochlear implants: of mode of communication in pediatric cochlear implantation.
A randomized clinical trial. International Journal of Pediatric Audiological Medicine, 5(4), 283–292.
Otorhinolaryngology, 77(9), 1545–1550. Vohr, B. (2003). Overview: Infants and children with hearing loss?
Osberger, M. J., & Fisher, L. (2000). Preoperative predictors of Part I. Mental Retardation and Developmental Disabilities
postoperative implant performance in children. Annals of Otology, Research Reviews, 9(2), 62–64.
Rinology & Laryngology. Supplement, 185, 44–46. World Health Organization. (2017). Deafness and hearing loss.
Park, G. Y., Moon, I. J., Kim, E. Y., Chung, E.-W., Cho, Y.-S., Retrieved from http://www.who.int/news-room/fact-sheets/detail/
Chung, W.-H., & Hong, S. H. (2013). Auditory and speech deafness-and-hearing-loss
performance in deaf children with deaf parents after cochlear Yanbay, E., Hickson, L., Scarinci, N., Constantinescu, G., &
implant. Otology & Neurotology, 34(2), 233–238. https://doi. Dettman, S. J. (2014). Language outcomes for children with
org/10.1097/MAO.0b013e31827b4d26 cochlear implants enrolled in different communication pro-
Percy-Smith, L., Cayé-Thomasen, P., Breinegaard, N., & Jensen, J. H. grams. Cochlear Implants International: An Interdisciplinary
(2010). Parental mode of communication is essential for speech Journal, 15(3), 121–135.
and language outcomes in cochlear implanted children. Acta Young, N. M., Grohne, K. M., Carrasco, V. N., & Brown, C. (2000).
Oto-Laryngologica, 130(6), 708–715. Speech perception of young children using Nucleus 22-Channel
Phan, J., Houston, D. M., Ruffin, C., Ting, J., & Holt, R. F. or Clarion cochlear implants. Annals of Otology, Rhinology, and
(2016). Factors affecting speech discrimination in children with Laryngology, 185, 77–79.

Appendix A
Search Strategies
Main terms used for database research and synonyms used.
Children
Child* Preschool child* Kid*

Pediatric*/paediatric* School child* Infant*/infancy
Toddler* Baby/babies
Hearing aid*/cochlear implant*

Auditory prosthesis Cochlear prosthesis
Deafness
Hearing impair* Deaf* Bilateral deaf*
Hearing loss Sensorineural hearing loss Bilateral hearing loss
Hearing disorder* Sensorineural deaf*
Intervention approach
Speech therapy Communication mode Speechreading
Language therapy Communication approach* Cued speech
Rehabilitation of hearing impaired Mode of communication Bicultural/bi-cultural
Auditory rehabilitation Communication method* Bilingual/bi-lingual
Early intervention Auditory–verbal/Auditory verbal Bilingual–bicultural
Hearing therapy* AVT Bi bi/bi-bi
Listening train* Auditory only Sign language
Auditory train* Oral approach* Manual communication
Rehabilitation Aural approach* ASL
Language rehabilitation Aural-oral American Sign Language
Speech rehabilitation Aural rehabilitation LSQ
Child* intervention Auditory–oral Langue des signesquébécoise
Early child* intervention Total communication Visual language
Simultaneous communication
Lipreading/Lip Reading
Auditory skill*
Listening ability* Listening skill* Auditory localization
Speech perception Auditory perception Sound localization
Speech discrimination Auditory discrimination Comprehension
Identification Recognition
Speech
Intelligibility* Phonetic* Articulation
Speech development*
Language
Oral language Language acquisition Syntax
Spoken language Verbal ability* Morphology*
Expressive language Communication Phonology*
Receptive language Vocabulary Fluency*
Language development* Semantic Pragmatic*
Appendix B ( p. 1 of 2)
Data Collection Form
General data:
- Authors:
- Title:
- Published: ▢ Yes ▢ No
Journal:
- Year:
- Country:
- Language: ▢ French ▢ English
- Contact:
Design:
- Study design:
Commentary:
Objective(s):
Participants:
- n total:
- n/group:
Equivalent?
Controlled variables:
- Recruitment method:
- Allocation method:
- Inclusion/exclusion criteria:
For each group:
- Characteristics:
Age:
Sex:
Race:
Socio-economical status:
Mother’s education:
Parent implication:
Deafness level:
Age at onset:
Technology device: ▢ Hearing aid ▢ Cochlear implant
Age at CI:
CI use length (months):
- Representative of population: ▢ Yes ▢ No
Explanation:
Commentary:
Intervention:
For each group:
- Approach/intervention:
Details (e.g., frequency, intensity, provider, group vs. one to one):
- Duration of intervention:
Commentary:
Measures:
- Variables:
- Assessor:
For each variable:
- ▢ pre-
Assessment tool(s):
Validation: ▢ Yes ▢ No
Reliability: ▢ Yes ▢ No
Relevant?
Commentary:

Appendix B ( p. 2 of 2)
Data Collection Form
Results:
- Abandon (n/%):
Reason:
- Statistical analysis:
Relevant?
p clearly mentioned?
- Main results:
Commentary:
Conclusions:
- Main conclusions:
Linked to the results?
Commentary:
Limits:
- Reported by the authors: ▢ Yes ▢ No
- Main limits:
Commentary:
Methodological quality (Effective Public Health Practice Project, 2009):

- Selection bias: Strong Moderate Weak
- Study design: Strong Moderate Weak
- Cofounders: Strong Moderate Weak
- Blinding: Strong Moderate Weak
- Data collection method: Strong Moderate Weak
- Withdrawals and dropouts: Strong Moderate Weak
Level of evidence (Phillips et al., 2009):

1 2 3 4 5
General commentaries:

Articulo 3

Uploaded by

Copyright:

Available Formats

Articulo 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Articulo 3

Uploaded by

Copyright:

Available Formats

JSLHR

Effectiveness of Rehabilitation Approaches

Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4197

Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4199

Table 1. Included studies.

Kirk et al., 2000, Cohort, Speech n = 103, OC CM -GAEL-P Speech Performance 3

spontaneous occasionally appeared to

Study -LNT Phonology: Error the need for these

IQ, monolingual rate of growth performed

baseline and/or and short term probability of

Table 3. Sample size distribution.

Design OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC OC > TC TC > OC TC = OC

Table 5. Result distribution per year.

[2000–2005] 10 (83.33%) 0 2 (16.66%) 7 0 0 6 1 2 1 0 5

Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4223

Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4225

Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4227

Main terms used for database research and synonyms used.

Child* Preschool child* Kid*

Hearing aid*/cochlear implant*

Demers & Bergeron: Effectiveness of Rehabilitation Approaches 4229

Methodological quality (Effective Public Health Practice Project, 2009):

Level of evidence (Phillips et al., 2009):

You might also like

Hearing aid/cochlear implant