On the Benefits of Speech-Language Therapy for Individuals Born With Cleft Palate: A Systematic Review and Meta-Analysis of Individual Participant Data
Publication: Journal of Speech, Language, and Hearing Research
Volume 65, Number 2
Pages 555-573
Abstract
Purpose:
Cleft lip and/or palate (CLP) is a common birth defect, and after reconstructive surgery, about 50% of children at 5 years of age have speech deviations and are referred to speech-language therapy (SLT). The peer-reviewed evidence for the benefit of SLT has been uncertain. Our objective was to systematically review and meta-analytically summarize the benefit of SLT for individuals born with CLP.
Method:
A systematic search was conducted (last search on February 19, 2021) on studies evaluating SLT with pre and post measures on speech production, language ability, intelligibility, and/or patient-reported outcomes. We sought individual participant data (IPD) and evaluated on an individual level if the outcome measure had improved to a clinically relevant degree during SLT and if the outcome measure was on a level with peers or not after SLT. Meta-analyses and meta-regressions were applied to synthesize IPD across studies.
Results:
Thirty-four eligible studies were found. Nineteen studies provided IPD (n = 343) for the main analysis on speech production. The synthesized information suggests that, during SLT, speech production improved to a clinically relevant degree for many individuals (95% CI [61%, 87%]) and that speech production was on a level with peers for some individuals after SLT (95% CI [10%, 34%]).
Conclusions:
The main strength of this meta-analysis is that we evaluated on an individual level pre- and post-intervention data based on considerations of clinical relevance. This approach allowed us to conclude that many individuals benefit from SLT and that further work on evaluating SLT in this patient group is meaningful.
Supplemental Material
Cleft lip and/or palate (CLP) has an incidence rate of approximately one to two per 1,000 births (Mossey et al., 2009; Mossey & Modell, 2012). It can affect a person's facial appearance, speech development, hearing, and well-being. Children born with CLP are at an increased risk of demonstrating speech and language problems, including phonological/articulation delay or disorder, resonance disorders, and expressive language delay or disorder (Kuehn & Moller, 2000). After reconstructive surgery, a large number of children have no such problems and develop in line with peers. However, approximately 50% of children with repaired CLP have persisting speech deviations at 5 years of age (Britton et al., 2014; Lohmander, Persson, et al., 2017), perhaps influenced by age at surgery, surgical technique, or learned speech behaviors, and are often referred to speech-language therapy (SLT; Hardin-Jones et al., 2009). Thus, the incidence of speech deviations is high among children born with CLP, in comparison to peers of the same age. Typically developing children show a gradual progress in speech sound production with age, and after 4 years of age, most speech sounds has been acquired correctly (Stoel-Gammon & Sosa, 2007). However, sibilant fricatives and the /r/ sound are typically established during a wide time period, and distortions may occur until around 6 years of age. This general description has been verified in a study of 684 English-speaking typical children aged 3;0–6;11 (years;months; Dodd et al., 2003) and in 473 Swedish typical children aged 3–6 years (Blumenthal & Lundeborg Hammarström, 2014).
In the group of children with repaired CLP, some persisting speech problems require secondary surgery, such as problems related to velopharyngeal incompetence, where the normal balance between the oral and nasal cavity is compromised, resulting in passive speech errors such as hypernasal resonance, audible nasal emission, weak pressure on consonants. These are the unavoidable consequence of an unwanted oral–nasal coupling on speech when no effort is made by the speaker to change it. Active speech errors, “compensatory errors,” occur when consonants are realized via a change in place and/or manner of articulation in order to compensate for the lack of adequate intraoral air pressure required for production of stop and fricative consonants (Harding & Grunwell, 1998). Active speech errors are one of the contributors to reduced intelligibility and should be treated with SLT. Problems with expressive language may further compromise intelligibility, and children might develop less positive attitudes to speech and communication compared to those without a cleft (Havstam & Lohmander, 2011; Whitehill et al., 2011).
Two major SLT approaches have been adopted to treat active speech errors: a motor-phonetic–based approach grounded in the theory that a motor skill is learned through repeated actions of that specific skill using a hierarchical structured therapy (Albery & Enderby, 1984; Ruscello & Vallino, 2014) and a language (linguistic)–phonological approach that focuses on the organization and representation of the sound system of a particular language (Chapman, 1993). In 1965, Prins and Bloomer observed that the majority of individuals with CLP who receive SLT do not achieve a speech production at the level of their peers. They stated, “The burden of proof to demonstrate the effectiveness of therapy for oral cleft patients is increasing, and a prerequisite for this is the development of meaningful procedures for studying changes in speech behavior” (Prins & Bloomer, 1965, p. 357). Unfortunately, the peer-reviewed evidence for the benefit of SLT for individuals born with CLP is still uncertain (Vallino-Napoli, 2011). In 2013, Bessell et al. conducted a systematic review of the published intervention studies, and despite evaluating 17 studies, they found little evidence to support the effectiveness of SLT in children with CLP. Bessell et al. (2013) diagnosed this uncertainty as stemming from very heterogeneous measuring and analytical procedures and argued that more rigorous methodology is needed. We certainly agree with calls for more rigorous and standardized methods for studying changes in speech behavior (Allori et al., 2017; Sell, 2005). However, we also believe that the unclear literature stems from how interventions have been evaluated on a fundamental level and that intervention outcomes mostly have been evaluated based on statistical significance (e.g., Alighieri et al., 2020; Hardin-Jones & Chapman, 2008; Pamplona et al., 2005, 2014), rather than clinical relevance.
Clinical Versus Statistical Evaluation
In many intervention studies on SLT, null hypothesis significance tests were applied to group aggregated data (e.g., t tests, analyses of variance, and Wilcoxon signed-ranks tests), and a statistically significant pre–post difference in the outcome measure was interpreted as supporting that the intervention was beneficial for the individuals. Although this analytical approach has been widespread, a statistically significant pre–post difference is actually theoretically and practically unrelated to the research interests many clinical researchers have (Angst et al., 2017; Wasserstein & Lazar, 2016). There are two research issues. The first issue is about which, if any, of the individuals enrolled in the study benefited from the intervention to a clinically relevant degree. The second issue is about what proportion of future individuals, based on the study at hand, are estimated to benefit from similar interventions.
The first issue is not a statistical query. It requires us to define what pre–post changes in a particular outcome measure should be considered clinically relevant. This is a research aspect that has been somewhat overshadowed in both SLT research and clinical research at large (Ioannidis, 2016; McShane et al., 2019; Revicki et al., 2008). It is a complex issue because what improvement constitutes a clinically relevant benefit must be sensitive to, for example, the patient group in question, the particular outcome of interest, and the dosage of the intervention. In the specific context of individuals born with CLP, it is also important to consider, for example, the type of cleft, the surgical technique, and the age at intervention. Interpreting clinical benefit based only on a statistical analysis does not take these considerations into account (Wasserstein et al., 2019). Furthermore, the focus of statistical analysis on aggregated data often obfuscates the fact of large interindividual variability in treatment outcomes (see Box 1). Thus, to evaluate SLT, we cannot only statistically evaluate treatment results but also need to evaluate clinical benefit on an individual level. Finally, in SLT contexts, we are interested not only in whether a pre–post change is clinically relevant or not but also in whether the outcome measure (e.g., speech production) is on a level with peers or not after SLT. This is again a complex issue that requires us to consider prior knowledge of typical speech development and the individual in question, not only statistical concerns.
The second issue is a statistical query, but one quite easily answered once the more complex first issue has been dealt with: Simply count the proportion of individuals that benefitted from the intervention. To get a statistic that summarizes what should be generalized from the study, calculate a confidence interval (CI) around this proportion. The CI should be interpreted as a “compatibility interval” illustrating what parameter values are most compatible with the data (Amrhein et al., 2019; Greenland, 2019). In this context, the CI can be interpreted as summarizing what possible proportions of future individuals might benefit from similar interventions.
Box 1. The difference between statistically significant and clinically relevant.
Participant PCC before SLT PCC after SLT Difference A 20% 26% 6%-points B 25% 29% 4%-points C 30% 35% 5%-points D 35% 41% 6%-points E 40% 45% 5%-points M (SD) 30% (8%-points) 35% (8%-points) 5%-points (0.8%-points) In the table above, we have illustrated five individuals' pre- and post-intervention speech production measured as percentage of consonants correct [PCC] in a picture-naming task. The pre–post difference is consistent between individuals and, thus, statistically significant (t 4 = 13.9, p < .001). However, none of the individuals improved to a clinically relevant degree.In the table below, we have illustrated a similar set of data. Here, the pre–post difference is inconsistent between individuals and, thus, not statistically significant (t 4 = 1.6, p = .18). Some individuals have, however, improved to a clinically relevant degree; an increase from 25% or 35% to 95% should be important in everyday settings.
Participant PCC before SLT PCC after SLT Difference A 20% 20% 0%-points B 25% 95% 70%-points C 30% 30% 0%-points D 35% 95% 60%-points E 40% 40% 0%-points M (SD) 30% (8%-points) 56% (36%-points) 26%-points (36%-points)
Purpose
Our primary goal in this systematic review was to obtain individual participant data (IPD) to evaluate the treatment effectiveness of SLT in individuals born with CLP. Our first aim was to meta-analytically estimate what proportion of individuals receiving SLT improved to a clinically relevant degree during treatment. Our second aim was to meta-analytically estimate what proportion of individuals had outcome measures on a level with peers after SLT. We also asked if the dispersion in treatment effects could be related to individual factors or intervention details. Using clinical considerations to evaluate IPD is novel in this area because previous studies and systematic reviews have focused either on inferential group-level statistics or on abstract effect size measures. Our secondary goal of this review was to narratively summarize the intervention studies IPD could not be obtained from. Our outcomes of interest were speech production, language ability, intelligibility, and patient-reported outcomes (PROs).
Method
A protocol of our systematic review was registered in 2019 at PROSPERO (https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=165754). See Supplemental Material S1 for the PRISMA IPD checklist.
Inclusion Criteria
The inclusion criteria for this systematic review was that a study had to (a) evaluate a conventional SLT intervention; (b) measure an outcome that related to speech production, language ability, intelligibility, or PROs; and (c) be readable in English.
As conventional SLT, we considered therapy aiming at the correct production of deviant consonant sounds using either a motor-phonetic approach (Ruscello & Vallino, 2014) or a language (linguistic)–phonological approach (Chapman, 1993). We did not include studies evaluating the use of speech bulbs, continuous airway pressure, and aerodynamic treatment, for example.
Regarding eligible outcomes, as measures of speech production, we included measures of articulation and consonant proficiency, consonant errors, and compensatory articulation. We did not include acoustical measures of speech production or physiological measures of tongue position, lateral pharyngeal wall movements, or measures of oral pressure. As measures of ability in different aspects of language, we included all expressive aspects, such as consonant inventory and other measures of phonology, lexicon or vocabulary measures, sentence complexity, and measures of grammar. As measures of intelligibility, we included caregiver or clinician ratings or other measures of intelligibility or understandability. As measures of PROs, we included self- or caregiver-reported indications of communicative participation or quality of life.
Study design (e.g., randomized controlled trial [RCT]) was not an inclusion criterion, instead we included evidence from all types of studies from RCT to case reports. This was the case because our IPD-focused analysis often implied a comparison other than the original study comparison anyway.
On a participant level, the inclusion criteria for the IPD-focused analysis were that participants had to (a) have been born with a CLP, (b) have outcome measures below the level of their peers before intervention, and (c) not have additional malformations or syndromes. We included individuals born with unilateral or bilateral cleft lip and palate and cleft palate but excluded individuals with noncleft velopharyngeal dysfunction, for example. We excluded individuals with syndromes, for example, 22Q11.2 deletion or CHARGE. After obtaining IPD, we concluded that most included participants were young (below 20 years old) and therefore excluded one participant because of his age (39 years old).
Search Strategy
To perform our electronic search, we engaged the Karolinska Institute University Library search consultation group. Our search targeted Medline, Embase, Cochrane, Web of Science, PsycInfo, and CINAHL. The specific search strategies for each database can be found in Supplemental Material S2, but they generally consisted of two “blocks”: one describing the population group (e.g., “cleft palate,” “cleft lip,” and “velopharyngeal insufficiency”) and the other one describing the intervention/outcomes of interest (e.g., “speech therap*” and “language training”). The MeSH terms identified for searching Medline (OVID) were adapted in accordance to corresponding vocabularies in Embase, PsychInfo, and CINAHL. Each search concept was also complemented with relevant free-text terms, and these were, if appropriate, truncated and/or combined with proximity operators. We had no restriction regarding year of publishing, language, or type of article (dissertations, theses, etc.) in our search strategy as including studies reported in all types of publication will reduce potential bias (Boutron et al., 2021). Our last search was on February 19, 2021. This search yielded 4,728 results.
Our electronic search was supplemented by reviewing the reference lists from included articles and published reviews and by an expert's insights. To gain access to potentially unpublished data (Boutron et al., 2021), we also contacted several researchers who had described their studies as “preliminary” or “pilot studies.” We did not, however, through these means identify any further studies.
Study Selection
We conducted two rounds of screening using Rayyan (Ouzzani et al., 2016). In the first round, we excluded, based on title and abstract information, articles that did not meet our inclusion criteria. One author (A.S.) screened all articles, and another author (E.H.) independently screened a randomly selected subset of 433 articles (10% of the initial search). The two authors strongly agreed in this first round of screening; they agreed on 431 out of 433 articles (Cohen's kappa = .94).
In the second round, we excluded, based on full-text information, articles that did not meet our inclusion criteria. Two authors (A.S. and E.H.) independently reviewed these articles. When there was disagreement, the third author (A.L.) helped settle this. See Figure 1 for a flowchart of the reviewed studies. For more information on why we excluded some studies, including some studies included in the review by Bessell et al. (2013), please see Supplemental Material S3.
Data Collection
We found 34 eligible studies that are listed in Supplemental Material S4. Twenty-one of these studies contributed with IPD. From 19 studies, IPD could be tabulated from the article and from two studies; IPD was received upon request from the author. These 19 studies contributed with IPD on one or more of the outcomes of interest (see Figure 1 and Supplemental Material S4). Thirteen studies did not contribute with IPD: In seven cases, we contacted the authors but were not able to obtain IPD, and in six cases, where the studies were published between 13 and 50 years ago, we did not request IPD because we judged it unlikely that we would obtain it.
On a study level, we tabulated type of therapy (motor-phonetic vs. phonological/language), the delivery of therapy (directly by a speech-language pathologist or by a speech assistant or caregiver, and if the therapy was delivered in a clinic or some other setting), and the dosage of the intervention. Because therapy details were heterogeneous between studies, we were not able to systematically tabulate more detailed information on dosage (Allen, 2013) than the total intervention duration.
On the IPD level, we tabulated demographic data regarding age, type of cleft, gender, syndrome, and the outcome measures outlined above pre- and post-intervention. When there were uncertainties in the reporting, we contacted authors to verify that we correctly interpreted the information.
The main aspect of the IPD-focused analysis was that we evaluated on an individual level separately (a) if the individual improved to a clinically relevant degree during treatment and (b) if the individual's outcome measure was on a level with peers or not after SLT. These evaluations were based on the second and third authors' (E.H. and A.L.) clinical expertise, knowledge of the studied therapies and measures, and knowledge on typical speech and language development, complemented with reference data for specific measures where available. In evaluating (a), the IPD information we considered was the age of the individual, the individual's pre–post change, and the dosage of the intervention (in relation to the prior information described below). In evaluating (b), the IPD information we considered was the age of the individual and the outcome measure after SLT (in relation to the prior information described below), disregarding the dosage of the intervention and any idiosyncratic speech deviations of the patient before intervention. We were strict in evaluating (b), requiring the outcome to be completely on the same level as peers without a history of clefts or other speech deviations. However, one speech production error was accepted as being on peer level at all ages as one error could be an error of measurement.
The prior information we used to evaluate the IPD was, in short, that after 4 years of age the phonemic inventory is completed in most children, who thereafter gradually reach adultlike precision of consonant sounds (Stoel-Gammon & Sosa, 2007). Although many consonants are similar across languages, there are differences regarding voicing, aspiration, and subtle differences in the place and manner of articulation (Hutters & Henningsson, 2004). Such differences could not be fully considered in all included studies. Reference values on PCC and consonant errors among typically developed children between 3 and 19 years of age (Dodd et al., 2003; Lohmander, Lundeborg, & Persson, 2017) were used to interpret the included IPD and to grossly evaluate relevant improvement during treatment and if speech production was on a level with peers or not after SLT. Typically, a strong progress in consonant production between 3 and 5 years of age is seen and again from 7 years of age where the median number of consonant errors in this group is zero.
As mentioned, speech production was measured in different ways in the included studies. Below, we exemplify our evaluation for three types of measures and studies. Prathanee et al. (2014) reported the number of consonant errors for their patients. For example, they reported on a 6-year-old child who had 13 consonant errors before and one error after SLT. We evaluated this child's pre–post change as being a clinically relevant improvement and the child's speech production as being on a level with peers after SLT. In comparison, for a 7-year-old child who had 15 errors before and three errors after SLT, we evaluated this child's pre–post change as being a clinically relevant improvement, but the child's speech production as not being on a level with peers after SLT. No consideration was taken to type of consonant error.
Van Demark and Hardin (1986) reported PCC. For example, they reported on an 8-year-old child who had a PCC of 77% before and 96% after SLT. We evaluated this child's improvement as being clinically relevant and her speech production as being on a level with peers after SLT. In comparison, for an almost 9-year-old child who had 71% before and 87% after SLT, we evaluated the improvement as being clinically relevant, but the child's speech production as not being on a level with peers after SLT.
Pamplona et al. (2014) reported a rating of the quality of consonant articulation in their studies. We evaluated a change from constant compensatory articulation/a single phoneme correct (Scale Steps 0 and 1) before SLT to correct consonant production in syllables, words, or short phrases in a closed context (Scale Steps 2 or 2–4 in different scales used) after SLT, as an improvement at all ages. For peer-level evaluation, however, this level was only accepted in 3-year-olds, whereas correct production in an open context (Scale Steps 3–5 in the different scales) was required at 4 years of age, and appropriate production or with only a single/inconsistent error (Scale Steps 5–7) was accepted from 5 years of age as being on a level with peers.
In summary, we could establish more formal procedures for our clinical evaluations in some studies, but for many studies, the evaluation was made in an informal manner. In a few studies in which the primary researchers made, formal or informational, evaluations of clinical improvement, we based our evaluation on their evaluation. We also tried to contact the primary researchers to understand how they evaluated their data from a clinical perspective. IPD tabulated from the included studies, and all our clinical evaluations can be found in Supplemental Material S5.
Risk of Bias Assessment
We were inspired by the ROBINS-I tool (Sterne et al., 2016) in assessing the risk of bias in the reviewed studies. ROBINS-I mainly targets studies in which two groups of patients are compared, often one intervention group versus a no-intervention group. The comparison that we tabulated from the included studies, however, was uncontrolled before–after comparisons; in other words, it only contained an intervention group. Thus, some domains of the ROBINS-I tool were not relevant in this review. Instead, we focused on two domains of the ROBINS-I tool: bias due to confounding factors (Question 1.1 in the ROBINS-I tool) and bias in measurement of outcomes. We used the categories for risk of bias judgments in line with the ROBINS-I tool (Sterne et al., 2016). We did not include the risk of bias judgments in the quantitative analysis by, for example, weighing down studies with greater risk of bias.
We identified two probable confounding factors in the included studies. (a) We assessed whether maturation (i.e., that the cleft-related speech behaviors or language delay resolved spontaneously) could explain any improvement occurring during treatment. We judged this to be more likely if the patients were very young and/or if the total intervention duration was long. (b) We assessed whether familiarity with the testing procedure or material could explain any improvement occurring during treatment. We judged this to be more likely if the patient was repeatedly tested with short intervals between the tests. We also assessed whether the outcome measure could have been influenced by the assessors not being blinded toward the patient's treatment or time point of measurement (Questions 6.1–6.2 in the ROBINS-I tool).
Analytical Procedure
For each study that we were able to obtain IPD (see Supplemental Material S4), we calculated the proportion of individuals who improved to a clinically relevant degree during treatment and the proportion of individuals whose outcome measures were on a level with peers after SLT. For the outcomes other than speech production, we found too few studies to carry out any formal quantitative synthesizing and instead only present the result of our evaluations in the Results section. For the studies on speech production for which we were not able to obtain IPD, we only narratively summarized the aggregated results presented in the articles.
We performed a proper meta-analysis on the IPD for the outcome speech production. All statistical analyses were conducted in R (R Core Team, 2013) and are available as Supplemental Material S6. We carried out two parallel analytical procedures on (a) what proportion of individuals improved to a clinically relevant degree during treatment and (b) what proportion of individuals whose speech production was on a level with peers after SLT. In both analyses, we first calculated a meta-analytical average across studies and the dispersion (prediction intervals) around these estimates (Borenstein, 2019). Briefly, the meta-analytical average proportion was estimated by weighting each individual study by its inverse variance (for more information on meta-analyses, please see Borenstein, 2019; Viechtbauer, 2010; Wang, 2018). Because the studied interventions differed in type, delivery, and dosage of intervention and the studied patient groups were heterogeneous, we assumed there to be large heterogeneity in intervention effects and used a random-effects meta-analytical model. Because sample sizes were often small and extreme proportions were included (e.g., 0.0 and 1.0), we applied a double arcsine transformation on the proportions (Viechtbauer, 2010) but present the back-transformed results. We fitted the models using restricted maximum-likelihood estimation (Laplace approximation). The meta-analyses were conducted using the “metafor” package in R (Viechtbauer, 2010; Wang, 2018).
In both analyses, our second step was to investigate to what extent individual factors and intervention details could explain the dispersion in the intervention effect. For individual-level factors (e.g., age at intervention), we applied mixed-effects logistic regression models that allow studies to have random intercepts rather than meta-regression, because of the large within-study heterogeneity of these factors. For study-level factors, we applied a meta-regression model. The mixed-effects logistic regressions were conducted using the “lme4” package and the meta-regression using the “metafor” package in R (Bates et al., 2015; Viechtbauer, 2010).
Results
Overview of the Included Studies
Nineteen studies contributed with IPD on speech production. As a rough overview of these studies, the therapy was implemented in the form of intensive summer camps in three studies (Pamplona et al., 2014, 2017; Van Demark & Hardin, 1986), the SLT was implemented in the form of conventional clinical care in four studies (Derakhshandeh et al., 2016; Pamplona & Ysunza, 2018; Roxburgh et al., 2016; Sweeney et al., 2020), the therapy was implemented through an initial short workshop led by SLTs and then carried out by parents or speech assistants over a long duration in nine studies (Dobbelsteyn et al., 2014; Hanchanlert et al., 2015; Makarabhirom et al., 2015; Prathanee, 2011; Prathanee et al., 2014, 2020; Pumnum et al., 2015; Scherer et al., 2008; Sritacha et al., 2016), and the therapy was very brief in three studies (Alighieri et al., 2019; Lindeborg et al., 2020; Luyten et al., 2016). See Table 1 for a general overview of these studies, and see Table 2 for risk of bias assessment of these studies. In all studies, possible biases favored treatment.
| Study | Therapy type | Delivery | Intervention duration | N a | Age range (months) | Speech measure | Intervention description |
|---|---|---|---|---|---|---|---|
| Alighieri et al. (2019) | Motor-phonetic | SLP (clinic) | 3 days | 2 | 137–140 | PCC based on 15 sentencesb in the SNAP test | Six 1-hr sessions SLT in patients' secondary language |
| Derakhshandeh et al. (2016) | Motor-phonetic | SLP (clinic) | 10 weeks | 5 | 54–108 | PCC in 14 oral and two nasal sentencesb (by picture naming) | Four sessions of 45-min per week for 10 weeks |
| Dobbelsteyn et al. (2014) | Motor-phonetic | Parent (home) | 4 months | 7 | 51–192 | PCC in SAF containing 11 oral phrasesb | Parents completed a 4-hr workshop and then gave daily 10-min sessions for 4 months. |
| Hanchanlert et al. (2015) | Motor-phonetic | Speech assistant and parent (home) | 9 months | 6 | 47–124 | Number of articulation errors on sentencesb on the TUPS | One- to 3-day speech camps with three to four 45-min sessions daily with SLP Speech assistants and caregivers gave weekly or biweekly 30-min sessions for 9 months. |
| Lindeborg et al. (2020) | Motor-phonetic | SLP and speech assistant (camp) | 1 week | 38 | 36–216 | Number of articulation errors on repeated single words | One week speech camp with daily sessions of 30–60 min administered by speech assistants assisted by SLPs |
| Luyten et al. (2016) | Motor-phonetic | SLP (clinic) | 3 days | 5 | 88–235 | PCC based on 15 sentencesb in the SNAP test | Six 1-hr sessions over 3–4 days SLT in patients' secondary language |
| Makarabhirom et al. (2015) | Motor-phonetic | Speech assistant and parent (home) | 9 months | 16 | 40–147 | Number of articulation errors on sentencesb on the TUPS | Three-day speech camps with three to four 45-min sessions daily with SLP Speech assistants and caregivers then gave weekly 30- to 45-min sessions for 9 months. |
| Pamplona & Ysunza (2018) | Language/phonological | SLP (clinic) | No information | 32 | 43–81 | 20 min of spontaneous speech rated according to severity of compensatory articulation | Fifteen sessions of 45 min (unclear interval) |
| Pamplona et al. (2014) | Language/phonological | SLP (camp) | 4 weeks | 90 | 36–80 | 20 min of spontaneous speech rated according to severity of compensatory articulation | Summer camp with 4-hr therapy per day, 5 days a week for 4 weeks |
| Pamplona et al. (2017) | Language/phonological | SLP (camp) | 3 weeks | 41 | 36–77 | 20 min of spontaneous speech rated according to severity of compensatory articulation | Summer camp with 4-hr therapy per day, 5 days a week for 3 weeks |
| Prathanee et al. (2014) | Motor-phonetic | Speech assistant (home) | 9 months | 16 | 49–96 | Number of articulation errors on sentencesb on the TUPS | Three-day speech camps with SLP Speech assistants then gave six 1-day sessions at home over 9 months. |
| Prathanee et al. (2020) | Motor-phonetic | Speech assistant and parent (home) | 1 year | 12 | 43–192 | Number of articulation errors on words on the TUPS | Three-day speech camps with six sessions of 45 min per day with SLP Speech assistants then gave six 1-day sessions at home over 9 months. Three 1-day follow-up speech camps with SLP |
| Prathanee (2011) | Motor-phonetic | Speech assistant and parent (home) | 6 months | 9 | 54–150 | Number of articulation errors in an articulation test (by picture naming) | Four-day speech camps with 18 hr of treatment with SLP Caregivers and health providers were responsible for children's speech program for 6 months. |
| Pumnum et al. (2015) | Motor-phonetic | Speech assistant and parent (home) | 9 months | 4 | 68–104 | Number of articulation errors on sentencesb on the TUPS | One- to 3-day speech camps with three to four 45-min sessions daily with SLP Speech assistants and caregivers then gave weekly or biweekly 30-min sessions for 9 months. |
| Roxburgh et al. (2016) | Motor-phonetic | SLP (clinic) | 2 weeks | 2 | 74–110 | PCC on 36 untrained words selected specifically for the patient's articulation errors | Eight sessions of 60 min over 2 weeks |
| Scherer et al. (2008) | Motor-phonetic | Parent (home) | 3 months | 10 | 18–35 | PCC-R measured from 30 min of recorded language sample from parent–child interaction | Parents completed two to four training sessions and implemented this for 3 months. |
| Sritacha et al. (2016) | Motor-phonetic | Speech assistant and parent (home) | 9 months | 6 | 49–72 | Number of articulation errors on sentencesb on the TUPS | One- to 3-day speech camps with three to four 45-min sessions daily with SLP Speech assistants and caregivers then gave weekly or biweekly 30-min sessions for 9 months. |
| Sweeney et al. (2020) | Motor-phonetic | SLP (clinic) vs. parent (home) | 3 months | 29 | 35–85 | PCC calculated for sentencesb in CAPS-A | One group (n = 12a) received six therapy sessions over 12 weeks. In the other group (n = 17a), parents attended 2 days of training and then carried out the intervention for 12 weeks. |
| Van Demark & Hardin (1986) | Motor-phonetic | SLP (camp) | 6 weeks | 13 | 80–144 | PCC articulation on ICPAT sentences. | Summer camp with 4-hr therapy per day for 6 weeks |
Note. SLP = speech-language pathologist; PCC = percentage of consonants correct; PCC-R = percentage of consonants correct–revised; SNAP = simplified nasometric assessment procedures; SLT = speech-language therapy; SAF = speech assessment form; TUPS = Thai universal parameters of speech outcomes for people with cleft palate; CAPS-A = Cleft Audit Protocol for Speech–Augmented; ICPAT = Iowa Cleft Palate Articulation Test.
a
Number of individuals included in our synthesis; we excluded individuals because of syndromes, because pretherapy speech production was already on a level with peers (according to the presented information), because of missing values, or because of patient overlap between studies. See Supplemental Material S5 for more information.
b
When both were available, we selected speech production on sentence rather than word level, because we believe this to have more ecological validity.
| Study | Bias due to confounding | Bias in measurement of outcomes | Judgment across domains | Note | |||
|---|---|---|---|---|---|---|---|
| Maturation | Familiarity | Judgment | Blind | Judgment | |||
| Alighieri et al. (2019) | N | Y | Serious | Y | Low | Serious | Possible familiarity from using same speech material at pre and post. The short revision before assessment might enhance this. (Contacted authors; no reply.) |
| Derakhshandeh et al. (2016) | N | N | Low | N | Serious | Serious | |
| Dobbelsteyn et al. (2014) | Y | N | Moderate | N | Serious | Serious | Possible maturation over the 4 months of intervention, especially for the younger children. |
| Hanchanlert et al. (2015) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 9 months of intervention, especially for the younger children. |
| Lindeborg et al. (2020) | N | N | Low | Y | Low | Low | |
| Luyten et al. (2016) | N | Y | Serious | N | Serious | Serious | Possible familiarity from using the same speech material at pre and post. One of three assessors was blinded. |
| Makarabhirom et al. (2015) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 9 months of intervention, especially for the younger children. |
| Pamplona & Ysunza (2018) | No information | N | No information | N | Serious | Serious | Unclear interval of treatment, so we cannot judge risk of maturation effects. |
| Unclear if and how assessors were blinded. | |||||||
| Pamplona et al. (2014) | N | N | Low | N | Serious | Serious | Unclear if and how assessors were blinded. |
| Pamplona et al. (2017) | N | N | Low | N | Serious | Serious | Unclear if and how assessors were blinded. |
| Prathanee et al. (2014) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 6 months of intervention, especially for the younger children. |
| Prathanee et al. (2020) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 9 months of intervention, especially for the younger children. |
| Prathanee (2011) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 1 year of intervention, especially for the younger children. |
| Pumnum et al. (2015) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 9 months of intervention, especially for the younger children. |
| Roxburgh et al. (2016) | N | N | Low | N | Moderate | Moderate | Two of three assessors were blinded. |
| Scherer et al. (2008) | Y | N | Serious | Y | Low | Serious | Possible maturation over the 3 months of intervention, especially for the younger children. |
| Sritacha et al. (2016) | Y | N | Serious | N | Serious | Serious | Possible maturation over the 9 months of intervention, especially for the younger children. |
| Sweeney et al. (2020) | N | N | Low | Y | Low | Low | |
| Van Demark & Hardin (1986) | N | N | Low | N | Serious | Serious | |
Note. Y = yes; N = no.
Five, eight, and one studies contributed with IPD for the outcomes language ability, intelligibility, and PROs, respectively. We elected to not perform a quantitative synthesis of these studies and instead only present the study results in Table 3.
| Outcome measure | Study | Total N a | Improved during therapy | At peer level after SLT | ||
|---|---|---|---|---|---|---|
| n | Proportion (95% CIb) | n | Proportion (95% CIb) | |||
| Language aspects | Hanchanlert et al. (2015) | 1 | 0 | 0.00 [0.00, 0.98] | 0 | 0.00 [0.00, 0.98] |
| Pamplona & Ysunza (2000) | 41 | 28 | 0.68 [0.52, 0.82] | 7c | 0.17 [0.07, 0.32] | |
| Prathanee et al. (2020) | 7 | 3 | 0.43 [0.10, 0.82] | 3 | 0.43 [0.10, 0.82] | |
| Pumnum et al. (2015) | 1 | 1 | 1.00 [0.03, 1.00] | 1 | 1.00 [0.03, 1.00] | |
| Sritacha et al. (2016) | 3 | 2 | 0.67 [0.09, 0.99] | 2 | 0.67 [0.09, 0.99] | |
| Intelligibility | Alighieri et al. (2019) | 2 | 0 | 0.00 [0.00, 0.84] | 0 | 0.00 [0.00, 0.84] |
| Hanchanlert et al. (2015) | 0 | — | — | |||
| Luyten et al. (2016) | 5 | 2 | 0.40 [0.05, 0.85] | 2 | 0.40 [0.05, 0.85] | |
| Prathanee et al. (2020) | 8 | 3 | 0.38 [0.09, 0.76] | 3 | 0.38 [0.09, 0.76] | |
| Prins & Bloomer (1965) | 10 | 5 | 0.50 [0.19, 0.81] | 0 | 0.00 [0.00, 0.31] | |
| Pumnum et al. (2015) | 2 | 0 | 0.00 [0.00, 0.84] | 0 | 0.00 [0.00, 0.84] | |
| Sritacha et al. (2016) | 1 | 1 | 1.00 [0.03, 1.00] | 1 | 1.00 [0.03, 1.00] | |
| Sweeney et al. (2020) | 24 | 10 | 0.42 [0.22, 0.63] | 8 | 0.33 [0.16, 0.55] | |
| PROs | Sweeney et al. (2020) | 15 | 10 | 0.67 [0.38, 0.88] | 8 | 0.53 [0.27, 0.79] |
Note. Em dashes signify that no participant was included from Hanchanlert et al. (2015), as all participants either were included in other studies or had outcome measures before SLT already on reference level.
a
We excluded participants whose outcome measures before SLT already were on reference level. See Supplemental Material S5 for more information.
b
Clopper–Pearson “exact intervals.”
c
A conservative estimate because of lack of information on the individual children's age.
Twelve studies measured speech production but did not contribute with IPD. We elected to not quantitatively synthesize the aggregated results of these studies and instead narratively describe them in Supplemental Material S7. One study measured language ability, but we were not able to obtain IPD (Pamplona et al., 2015).
Aim 1: To Estimate the Proportion of Individuals Who Improved to a Clinically Relevant Degree During Intervention
Figure 2 summarizes the meta-analysis on the proportion of individuals who improved to a clinically relevant degree during intervention. On average, across intervention details and individual demographics, 75% of individuals improved during treatment (95% CI [61%, 87%]). However, the dispersion around this meta-analytical average was large; the prediction interval indicates that, for some interventions and demographic groups studied, the proportion could be as low as 20%, and for other cases, it could be as high as 100%. Our second step was to address factors potentially able to explain the variation in intervention effect. No individual study was deemed an outlier in the sense that it influenced the meta-analytical estimate to a disproportionate degree.
Figure 3 summarizes the mixed-effects logistic regression we used to assess how age at intervention was related to the probability to improve during intervention. In summary, age at intervention was strongly related to the probability that an individual would improve during intervention: For children 6 years of age or younger, the model estimated on average roughly 80% probability that they would improve during SLT. For individuals older than 14 years of age, the predicted probability was only about 40%. This model was based on 343 individuals from 19 studies. The intercept in this model (in log odds) was 1.99 (95% CI [1.02, 3.07]), and the regression coefficient (in log odds) was −0.011 (95% CI [−0.021, −0.002]) per month of age. As illustrated in Figure 3, there was considerable between-studies variation; the standard deviation between the random intercepts was (in log odds) 1.12.
Table 4 summarizes the mixed-effects logistic regression we used to assess how cleft type was related to the probability of improving during intervention. In this analysis, we dummy-coded types of cleft into unilateral CLP, bilateral CLP, and cleft palate, and we used unilateral CLP as reference and included studies as random intercepts. In summary, the model suggests that individuals with a bilateral CLP were slightly less likely to improve during therapy; however, note that there is considerable uncertainty in how strong this effect is; the 95% CI around the coefficient of having a bilateral CLP, compared to unilateral CLP, was [−1.67, 0.44] in log odds. This model was based on 303 individuals from 18 studies; Lindeborg et al. (2020) did not include information on type of cleft and could thus not be included in this analysis. The intercept in this model (in log odds) was 1.35 (95% CI [0.63, 2.16]), and the regression coefficients (in log odds) were −0.61 (95% CI [−1.67, 0.44]) and −0.19 (95% CI [−1.19, 0.86]) for bilateral CLP or cleft palate, as compared to unilateral CLP. The standard deviation between the random intercepts was (in log odds) 1.06.
| Type of cleft | Total N a | Improved | Peer level | ||
|---|---|---|---|---|---|
| n (%) | Predictedb | n (%) | Predictedb | ||
| Unilateral CLP | 240 | 193 (80) | 79% | 52 (22) | 17% |
| Bilateral CLP | 28 | 18 (64) | 68% | 8 (29) | 16% |
| Cleft palate | 35 | 25 (71) | 76% | 14 (40) | 36% |
Note. CLP = cleft lip and/or palate.
a
One individual with submucous cleft palate and one individual with soft palate cleft were coded as having a cleft palate; one individual with a cleft lip was excluded.
b
The probabilities reported under “Predicted” are the model predictions (back-transformed to the probability scale) that take between-studies variability into account (as random intercepts).
Aim 2: To Estimate the Proportion of Individuals Whose Speech Production Was on a Level With Peers
Figure 4 summarizes the meta-analysis on the proportion of individuals whose speech production was on a level with peers after SLT. On average, across intervention details and individual demographics, 21% of individuals had speech production on a level with peers after SLT (95% CI [10%, 34%]). However, the dispersion around this meta-analytical average was large; the prediction interval was between 0% and 71%.
Figure 5 summarizes the mixed-effects logistic regression we used to assess how age at intervention was related to the probability to have speech production on a level with peers after SLT. In summary, age at intervention was weakly related to this probability. For example, a 6-year-old child was not that much more probable to achieve this than a 14-year-old teen (note that they are compared to peers of their respective age). Figure 5 illustrates that there is more variation in the studies intercepts than there is along the age range, indicating that most of the between-studies heterogeneity is unaccounted for. This model was based on 343 individuals from 19 studies. The intercept in this model (in log odds) was −0.77 (95% CI [−1.94, 0.34]), and the regression coefficient (in log odds) was −0.01 (95% CI [−0.020, 0.003]) per month of age. As mentioned, there was considerable between-studies variation; the standard deviation between the random intercepts was (in log odds) 1.24.
Table 4 summarizes the mixed-effects logistic regression we used to assess how cleft type was related to the probability to have a speech production on a level with peers after SLT. In summary, individuals with cleft palate only seem to have a greater chance of achieving this. This model was based on 303 individuals from 18 studies; Lindeborg et al. (2020) did not include information on type of cleft and could thus not be included in this analysis. The intercept in this model (in log odds) was −1.59 (95% CI [−2.65, −0.78]), and the regression coefficients (in log odds) were −0.06 (95% CI [−1.18, 0.99]) and 1.01 (95% CI [0.03, 2.03]) for having a bilateral CLP or cleft palate, as compared to unilateral CLP. The standard deviation between the random intercepts was (in log odds) 1.33.
Figure 6 summarizes the meta-regression we used to assess how intervention duration was related to the probability to have a speech production on a level with peers after SLT. In summary, longer intervention durations were related to a larger proportion of patients achieving this. This model was based on 18 studies; Pamplona and Ysunza (2018) did not include information on intervention duration and could thus not be included in this analysis. The intercept in this model (in percentage points) was 0.33 (95% CI [0.19, 0.47]), and the regression coefficient was 0.01 (95% CI [0.006, 0.018]) per week of intervention. In conclusion, intervention duration was able to explain parts of the heterogeneity in the intervention effect (Q 1 = 14.75, p = .0001).
Discussion
Although our original intent was to treat the four outcomes (speech production, language ability, intelligibility, and PROs) in a similar manner, we found that we could only meta-analytically analyze data on speech production. Thus, large parts of the Results section and our inferences concern only this outcome.
Speech Production
We found 34 eligible studies on SLT for individuals born with CLP. The studies were heterogeneous with regard to intervention details, demographics of the studied individuals, specific outcome measures, and statistical reporting. We decided that transformations of the various reported effect sizes to a standardized mean difference scale would be impossible to interpret in relation to clinical relevance. Instead, we evaluated IPD from 19 studies regarding improvement during treatment and speech production after SLT based on clinical considerations and meta-analytically summarized these results. We obtained only aggregated data in 12 studies (see Supplemental Material S7), and although the authors interpreted their results as being supportive of SLT benefiting patients to some degree, we were not able to infer how many (if any) individuals benefited to a clinically relevant degree in those studies.
Our main conclusion from the quantitative synthesis of the IPD is that SLT does benefit the speech production of many individuals born with CLP. We do, however, rate the overall quality of evidence for this conclusion to be low due to (a) serious risks of bias in the studies, (b) imprecision in many of the studies, and (c) serious inconsistency in the implementation of the therapies between the included studies.
Clinically Relevant Improvement in Speech Production
Because the between-studies heterogeneity was substantial, we cannot deliver a definitive estimate of the benefit of SLT, but we have provided some reasonable limits in the intervention effect. Based on our meta-analysis, even in cases in which SLT is estimated to be the least helpful, about 20% (lower limit of the prediction interval) of individuals are predicted to improve to a clinically relevant degree. On average, across intervention details and study designs, about 61%–87% (95% CI) of individuals are predicted to improve during SLT. Thus, although we cannot specify the parameters of SLT that provide the best results for the patients, SLT does seem to have positive effects for many or most individuals who receive it.
The main factor explaining the dispersion in the estimated proportion of individuals who improved from intervention was the individual's age at intervention: Young children (e.g., younger than 6 years old) had a greater probability to improve than teens receiving intervention (e.g., older than 14 years old). Note, however, that the relationship between age and beneficial outcome could be biased by (a) the fact that our evaluation of clinical improvement considered age at intervention, (b) a greater risk of bias in studying younger children who receive long intervention durations (see the Limitations section below), and (c) fewer studies included teens, and these studies differed from the other studies in other important details. For example, two studies with the oldest individuals (Alighieri et al., 2019; Luyten et al., 2016) used a relatively low intervention dosage (six 1-hr sessions over 3 days), and the intervention was conducted in the patients' secondary language. In other studies with older individuals (Dobbelsteyn et al., 2014; Makarabhirom et al., 2015; Prathanee, 2011), speech assistants or parents were the persons who delivered most of the SLT. In comparison, many of the studies with younger individuals (Pamplona & Ysunza, 2018; Pamplona et al., 2014, 2017) used intensive and relatively high-dosage therapies delivered by speech-language pathologists. Still, it seems quite reasonable that younger individuals are more likely to benefit from SLT (Peterson-Falzone et al., 2010).
Speech Production After SLT
SLT should not only improve the speech of patients but also bring their speech production up to the level of their peers. Less positive, then, is our estimate that, on average, around 10%–34% (95% CI) of individuals are predicted to have speech production on a level with peers after SLT. To understand this somewhat negative result, we first want to note that none of the studied interventions were designed to produce satisfactory speech production but were, rather, planned to end in a specified time frame and with a predetermined intervention dosage. Thus, it is not unreasonable that the reviewed studies underestimate the proportion of individuals who would reach peer-level speech after SLT if the dosage of therapy were to be increased. However, this is far from demonstrated: In the Scandcleft Project, many of the children who had the largest amount of speech therapy visits were those with persistent speech problems (Persson et al., 2020; Willadsen et al., 2017).
The prediction interval for the proportion of individuals who have speech production on a level with peers was very large (0%–71%), indicating that the intervention effect was very inconsistent between studies. On a study level, intervention duration was able to explain some of the dispersion in speech production following SLT (see Figure 6). This is reasonable both because (a) longer intervention duration should be able to produce better outcomes (but see the Discussion section above) and (b) longer intervention durations were related to a greater risk of bias from maturation (see the Limitations section). On an individual level, type of cleft was somewhat related to speech production after SLT: Individuals being born with cleft palate only, as compared to unilateral or bilateral CLP, perhaps have a slightly better prognosis (see Table 4). We must point out, however, that this result is based on very few individuals with cleft palate (n = 35) and therefore is far from certain.
We were thus not able to fully explain the inconsistency in intervention outcomes between studies, but this should not be too surprising. The between-studies heterogeneity regarding context, intervention details, and patient groups was very large. For example, in some studies, individuals were specifically included because of their severe articulatory deviations (Pamplona & Ysunza, 2018; Pamplona et al., 2014, 2017) and would accordingly a priori be less likely to reach peer-level speech production than patients in other studies. Such study idiosyncrasies were too particular for us to be able to adequately model. Thus, because of the large heterogeneity in the reviewed studies, we do not believe that we have produced a stable estimate of the probability to reach peer-level speech production after SLT but have at least summarized the information in the literature.
Other Outcomes
We found very little information in the published literature on the effects of SLT on language ability, intelligibility, and PROs. The take-home message from Table 3, we believe, is that the published literature, so far, has very low evidential value regarding these outcomes.
This is a problematic situation. First, the end goal of SLT is to increase the patient's ability to communicate and participate positively in real-world settings (Havstam & Lohmander, 2011). Ecologically valid measures of intelligibility, attitudes toward communication, and communicative participation are necessary to evaluate whether SLT leads to these effects. Second, from a methodological perspective, ecologically valid measures of intelligibility in everyday settings are necessary to validate the different measures of speech production that are used in clinical research today (Neumann & Romonath, 2012).
We do not, however, want to devalue the great impact that many of the interventions in the reviewed studies have had on the quality of life of many participants, especially in areas where SLT or cleft-related interventions are otherwise rare (e.g., Lindeborg et al., 2020; Luyten et al., 2016; Prathanee, 2011). We only would prefer a scenario where we have more systematized data on these effects.
Limitations
The main aspect of this project was our evaluations of improvement during treatment and outcome measures immediately after SLT on IPD, and the certainty of our results is totally founded on these evaluations. We based these evaluations on the reported data in the included studies. This is necessarily a flawed approach. As reviewers of published studies, we have no contact with the individuals enrolled in the study, little insight into the particulars of the intervention, and imperfect information on the actual outcome instruments. Furthermore, we could not operationalize any formal criteria for “improvement to a clinically relevant degree” or “outcome measures at peer level” across studies, or indeed within some studies. Thus, we believe our IPD evaluations should be seen as a “proof of concept,” rather than as a “gold standard,” and as highlighting the need for standardized ways to evaluate improvement on an individual level. We encourage all second opinions on the set of IPD we analyzed here (which can be found as Supplemental Material S5) and invite any discussion on this important topic. A stronger methodology in the future, clearly, would be if primary researchers evaluated which, if any, of their participants improved to a clinically relevant degree and could do so based on standardized criteria (see the Recommendations for Future Research section). Nevertheless, in lieu of this, our evaluations, we believe, have allowed us to distill that the published literature demonstrates that SLT and further work on evaluating SLT in this patient group are meaningful endeavors.
We must acknowledge that the results in many of the primary studies included in the quantitative synthesis were at a serious risk of bias (see Table 2). In our individual evaluations, we could not, nor did we try to, establish whether any clinically relevant improvement during therapy or peer-level outcome measures following SLT were caused by the SLT. For example, patients may have matured spontaneously over long intervention durations (Lohmander et al., 2006). We have only evaluated the magnitude of improvement that occurred during treatment and the outcome measures at a particular time point after SLT. Thus, we must acknowledge that the benefits of SLT, especially for younger age groups, may be overestimated in this synthesis. However, it is not clear exactly what study design could be implemented in this research context to control for this confound as the intervention duration can be much longer than what is reasonable for a waitlist control group to wait for treatment, for example, and possibly biased estimates may be the only estimates we can hope for (Sterne et al., 2016).
In most of the primary studies, we also judged the risk of bias to be serious because assessors were not blinded regarding the patients' treatment status. This is something that can and needs to be addressed in future studies.
Recommendations for Future Research
Our primary recommendation for future intervention studies is that primary researchers use their expert and clinical insight to evaluate which, if any, of their patients benefitted to a clinically relevant degree from the studied intervention. We cannot rely solely on “context blind” statistical approaches in evaluating intervention outcomes (Angst et al., 2017; McShane et al., 2019; Wasserstein et al., 2019).
As we noted above, what pre–post changes should constitute clinically relevant benefit is a complex issue—that should take individual factors, intervention details, and PROs into account—but it is an issue that clinical research on SLT must confront if we are to adequately evaluate SLT. Several approaches to delineate clinically relevant change in other medical fields have been proposed and discussed and could form baselines for discussions in SLT research (Angst et al., 2017; Crosby et al., 2003; Guyatt et al., 2002; Jayadevappa et al., 2017; Revicki et al., 2008). Although the reliability of the outcome measures must be considered, we hope that discussions on this important topic will not overly rely on statistical definitions of benefit, for example, minimal differences based on measurement error or statistical change compared to reference data (Jacobson & Truax, 1991; Weir, 2005). Instead, our preliminary recommendation would be that clinically relevant benefit should be context sensitive to a particular patient group and established in relation to patient reports, where possible, supplemented by caregiver or clinician insight, where reasonable (Angst et al., 2017; Guyatt et al., 2002). We also believe that absolute outcome levels after SLT should also be reported and related to reference standards to note which, if any, patients reached peer levels.
Second, we should strive to make intervention studies on SLT less heterogeneous and to evaluate the global benefits of SLT, not only related to speech production but also regarding intelligibility and communicative participation in everyday settings. A standard outcome set to measure speech production, intelligibility, and PROs and standardized criteria for evaluating these outcomes are necessary for future systematic reviews and meta-analyses of SLT. The ability to stringently compare and systematize intervention outcomes across studies outweighs the drawback that no standard set may perfectly fit any particular study setting. One suggested standard outcome set is the ICHOM Standard Set for Cleft Lip & Palate, which contains measures of speech production, intelligibility, and PRO (Allori et al., 2017; https://www.ichom.org/portfolio/cleft-lip-palate/). In this set, articulation is measured via the modified PCC based on single words (Klintö et al., 2011; Shriberg et al., 1997), velopharyngeal competence is measured with the Velopharyngeal Competence Scale (e.g., Lohmander, Hagberg, et al., 2017; Lohmander et al., 2009), intelligibility is measured via the Intelligibility in Context Scale (McLeod et al., 2012), and communicative attitude is measured via the Cleft Q Speech Distress and Speech Function Scales (Klassen et al., 2018; Stiernman et al., 2021).
Conclusions
Based on our clinical evaluations on individual intervention outcomes, the peer-reviewed studies included in this meta-analysis indicate that SLT benefits the speech production of many individuals born with CLP. Especially young children (e.g., below 6 years of age) are likely to benefit from SLT to a clinically relevant degree. It is less certain, from the published literature, what proportion of individuals can reach speech production on a level with peers after SLT. The imprecision in the estimates and overall quality of evidence in the literature should encourage more research on this important topic.
Supplemental Material on Figshare
Acknowledgments
We thank the Karolinska Institute University Library search consultation group and Sabina Gillsund, specifically, for formulating and conducting our literature search. We also want to thank all primary researchers who were helpful in our correspondence.
References
Albery, L., & Enderby, P. (1984). Intensive speech therapy for cleft palate children. British Journal of Disorders of Communication, 19(2), 115–124.
Alighieri, C., Bettens, K., Bruneel, L., Sseremba, D., Musasizi, D., Ojok, I., & Van Lierde, K. (2020). Comparison of motor-phonetic versus phonetic-phonological speech therapy approaches in patients with a cleft (lip and) palate: A study in Uganda. International Journal of Pediatric Otorhinolaryngology, 131, 109849.
Alighieri, C., Bettens, K., Bruneel, L., Vandormael, C., Musasizi, D., Ojok, I., D'haeseleer, E., & Van Lierde, K. (2019). Intensive speech therapy in Ugandan patients with cleft (lip and) palate: A pilot-study assessing long-term effectiveness. International Journal of Pediatric Otorhinolaryngology, 123, 156–167.
Allen, M. M. (2013). Intervention efficacy and intensity for children with speech sound disorder. Journal of Speech, Language, and Hearing Research, 56(3), 865–877.
Allori, A. C., Kelley, T., Meara, J. G., Albert, A., Bonanthaya, K., Chapman, K., Cunningham, M., Daskalogiannakis, J., De Gier, H., Heggie, A. A., Hernandez, C., Jackson, O., Jones, Y., Kangesu, L., Koudstaal, M. J., Kuchhal, R., Lohmander, A., Long, R. E., Magee, L., … Wong, K. W. (2017). A standard set of outcome measures for the comprehensive appraisal of cleft care. The Cleft Palate–Craniofacial Journal, 54(5), 540–554.
Amrhein, V., Trafimow, D., & Greenland, S. (2019). Inferential statistics as descriptive statistics: There is no replication crisis if we don't expect replication. The American Statistician, 73(Suppl. 1), 262–270.
Angst, F., Aeschlimann, A., & Angst, J. (2017). The minimal clinically important difference raised the significance of outcome effects above the statistical level, with methodological implications for future studies. Journal of Clinical Epidemiology, 82, 128–136.
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1), 1–48.
Bessell, A., Sell, D., Whiting, P., Roulstone, S., Albery, L., Persson, M., Verhoeven, A., Burke, M., & Ness, A. R. (2013). Speech and language therapy interventions for children with cleft palate: A systematic review. The Cleft Palate–Craniofacial Journal, 50(1), 1–17.
Blumenthal, C., & Lundeborg Hammarström, I. (2014). LINUS LINköpingsUnderSökningen Ett fonologiskt testmaterial för barn från 3 år [LINUS LINköpingsUnderSökningen A phonological test material for children from 3 years of age.] [Master thesis in Swedish, Linköping University Electronic Press ]. http://www.diva-portal.org/smash/get/diva2:737467/FULLTEXT01.pdf
Borenstein, M. (2019). Common mistakes in meta-analysis and how to avoid them. Biostat.
Boutron, I., Page, M., Higgins, J., Altman, D., Lind, A., & Hróbjartsson, A. (2021). Chapter 7: Considering bias and conflicts of interest among the included studies. In Cochrane handbook for systematic reviews of interventions Version 6.2 (updated February 2021). Cochrane. http://www.training.cochrane.org/handbook
Britton, L., Albery, L., Bowden, M., Harding-Bell, A., Phippen, G., & Sell, D. (2014). A cross-sectional cohort study of speech in five-year-olds with cleft palate ± lip to support development of national audit standards: Benchmarking speech standards in the United Kingdom. The Cleft Palate–Craniofacial Journal, 51(4), 431–451.
Chapman, K. L. (1993). Phonologic processes in children with cleft palate. The Cleft Palate–Craniofacial Journal, 30(1), 64–72.
Crosby, R. D., Kolotkin, R. L., & Williams, G. R. (2003). Defining clinically meaningful change in health-related quality of life. Journal of Clinical Epidemiology, 56(5), 395–407.
Derakhshandeh, F., Nikmaram, M., Hosseinabad, H. H., Memarzadeh, M., Taheri, M., Omrani, M., Jalaie, S., Bijankhan, M., & Sell, D. (2016). Speech characteristics after articulation therapy in children with cleft palate and velopharyngeal dysfunction—A single case experimental design. International Journal of Pediatric Otorhinolaryngology, 86, 104–113.
Dobbelsteyn, C., Bird, E. K.-R., Parker, J., Griffiths, C., Budden, A., Flood, K., & Stilson, A. (2014). Effectiveness of the corrective babbling speech treatment program for children with a history of cleft palate or velopharyngeal dysfunction. The Cleft Palate–Craniofacial Journal, 51(2), 129–144.
Dodd, B., Holm, A., Hua, Z., & Crosbie, S. (2003). Phonological development: A normative study of British English-speaking children. Clinical Linguistics & Phonetics, 17(8), 617–643.
Greenland, S. (2019). Valid P-values behave exactly as they should: Some misleading criticisms of P-values and their resolution with S-values. The American Statistician, 73(Suppl. 1), 106–114.
Guyatt, G. H., Osoba, D., Wu, A. W., Wyrwich, K. W., Norman, G. R., & Clinical Significance Consensus Meeting Group. (2002). Methods to explain the clinical significance of health status measures. Mayo Clinic Proceedings, 77(4), 371–383.
Hanchanlert, Y., Pramakhatay, W., Pradubwong, S., & Prathanee, B. (2015). Speech correction for children with cleft lip and palate by networking of community-based care. Journal of the Medical Association of Thailand, 98, S132–S139.
Harding, A., & Grunwell, P. (1998). Active versus passive cleft-type speech characteristics. International Journal of Language & Communication Disorders, 33(3), 329–352.
Hardin-Jones, M. A., & Chapman, K. L. (2008). The impact of early intervention on speech and lexical development for toddlers with cleft palate: A retrospective look at outcome. Language, Speech, and Hearing Services in Schools, 39(1), 89–96.
Hardin-Jones, M. A., Karnell, M. P., & Peterson-Falzone, S. J. (2009). Cleft palate speech (4th ed.). Mosby.
Havstam, C., & Lohmander, A. (2011). Communicative participation. In Cleft palate speech: Assessment and intervention (pp. 305–315). Wiley.
Hutters, B., & Henningsson, G. (2004). Speech outcome following treatment in cross-linguistic cleft palate studies: Methodological implications. The Cleft Palate–Craniofacial Journal, 41(5), 544–549.
Ioannidis, J. P. A. (2016). Why most clinical research is not useful. PLOS Medicine, 13(6), e1002049.
Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59(1), 12–19.
Jayadevappa, R., Cook, R., & Chhatre, S. (2017). Minimal important difference to infer changes in health-related quality of life—A systematic review. Journal of Clinical Epidemiology, 89, 188–198.
Klassen, A. F., Riff, K. W. Y. W., Longmire, N. M., Albert, A., Allen, G. C., Aydin, M. A., Baker, S. B., Cano, S. J., Chan, A. J., Courtemanche, D. J., Dreise, M. M., Goldstein, J. A., Goodacre, T. E. E., Harman, K. E., Munill, M., Mahony, A. O., Aguilera, M. P., Peterson, P., Pusic, A. L., … Forrest, C. R. (2018). Psychometric findings and normative values for the CLEFT-Q based on 2434 children and young adult patients with cleft lip and/or palate from 12 countries. Canadian Medical Association Journal, 190(15), E455–E462.
Klintö, K., Salameh, E.-K., Svensson, H., & Lohmander, A. (2011). The impact of speech material on speech judgement in children with and without cleft palate. International Journal of Language & Communication Disorders, 46(3), 348–360.
Kuehn, D. P., & Moller, K. T. (2000). Speech and language issues in the cleft palate population: The state of the art. The Cleft Palate–Craniofacial Journal, 37(4), 1–35.
Lindeborg, M. M., Shakya, P., Pradhan, B., Rai, S. K., Gurung, K. B., Niroula, S., Rayamajhi, B., Chaudhary, H., Gaire, B., Mahato, N., Rana, L., Rokaya, P., Shrestha, N., Shrestha, R., Tamang, J., Joshi, H. D., Gaha, P., Khorja, D. K., Nakarmi, K. K., … Shaye, D. A. (2020). A task-shifted speech therapy program for cleft palate patients in rural Nepal: Evaluating impact and associated healthcare barriers. International Journal of Pediatric Otorhinolaryngology, 134, 110026.
Lohmander, A., Friede, H., Elander, A., Persson, C., & Lilja, J. (2006). Speech development in patients with unilateral cleft lip and palate treated with different delays in closure of the hard palate after early velar repair: A longitudinal perspective. Scandinavian Journal of Plastic and Reconstructive Surgery and Hand Surgery, 40(5), 267–274.
Lohmander, A., Hagberg, E., Persson, C., Willadsen, E., Lundeborg, I., Davies, J., Havstam, C., Boers, M., Kisling-Møller, M., Alaluusua, S., Aukner, R., Pedersen, N. H., Turunen, L., & Nyberg, J. (2017). Validity of auditory perceptual assessment of velopharyngeal function and dysfunction—The VPC-Sum and the VPC-Rate. Clinical Linguistics & Phonetics, 31(7–9), 589–597.
Lohmander, A., Lundeborg, I., & Persson, C. (2017). SVANTE—The Swedish Articulation and Nasality Test—Normative data and a minimum standard set for cross-linguistic comparison. Clinical Linguistics & Phonetics, 31(2), 137–154.
Lohmander, A., Persson, C., Willadsen, E., Lundeborg, I., Alaluusua, S., Aukner, R., Bau, A., Boers, M., Bowden, M., Davies, J., Emborg, B., Havstam, C., Hayden, C., Henningsson, G., Holmefjord, A., Hölttä, E., Kisling-Møller, M., Kjøll, L., Lundberg, M., … Semb, G. (2017). Scandcleft randomised trials of primary surgery for unilateral cleft lip and palate: 4. Speech outcomes in 5-year-olds—Velopharyngeal competency and hypernasality. Journal of Plastic Surgery and Hand Surgery, 51(1), 27–37.
Lohmander, A., Willadsen, E., Persson, C., Henningsson, G., Bowden, M., & Hutters, B. (2009). Methodology for speech assessment in the scandcleft project—An international randomized clinical trial on palatal surgery: Experiences from a pilot study. The Cleft Palate–Craniofacial Journal, 46(4), 347–362.
Luyten, A., Bettens, K., D'Haeseleer, E., Hodges, A., Galiwango, G., Vermeersch, H., & Van Lierde, K. (2016). Short-term effect of short, intensive speech therapy on articulation and resonance in Ugandan patients with cleft (lip and) palate. Journal of Communication Disorders, 61, 71–82.
Makarabhirom, K., Prathanee, B., Suphawatjariyakul, R., & Yoodee, P. (2015). Speech therapy for children with cleft lip and palate using a community-based speech therapy model with speech assistants. Journal of the Medical Association of Thailand, 98, S140–S150.
McLeod, S., Harrison, L. J., & McCormack, J. (2012). The intelligibility in context scale: Validity and reliability of a subjective rating measure. Journal of Speech, Language, and Hearing Research, 55(2), 648–656.
McShane, B. B., Gal, D., Gelman, A., Robert, C., & Tackett, J. L. (2019). Abandon statistical significance. The American Statistician, 73(Supp_1), 235–245.
Mossey, P. A., Little, J., Munger, R. G., Dixon, M. J., & Shaw, W. C. (2009). Cleft lip and palate. The Lancet, 374, 1773–1785.
Mossey, P. A., & Modell, B. (2012). Epidemiology of oral clefts 2012: An international perspective. Frontiers of Oral Biology, 16, 1–18.
Neumann, S., & Romonath, R. (2012). Application of the International Classification of Functioning, Disability and Health–Children and Youth Version (ICF-CY) to cleft lip and palate. The Cleft Palate–Craniofacial Journal, 49(3), 325–346.
Ouzzani, M., Hammady, H., Fedorowicz, Z., & Elmagarmid, A. (2016). Rayyan—A web and mobile app for systematic reviews. Systematic Reviews, 5(1), Article 210.
Pamplona, C., Silis, S. C., Ysunza, P. A., & Morales, S. (2015). Metacognitive strategies for enhancing language development in children with cleft palate. European Journal of Plastic Surgery, 38(5), 377–384.
Pamplona, C., & Ysunza, A. (2000). Active participation of mothers during speech therapy improved language development of children with cleft palate. Scandinavian Journal of Plastic and Reconstructive Surgery and Hand Surgery, 34(3), 231–236.
Pamplona, C., & Ysunza, A. (2018). Deliberate practice: Preliminary results of a useful strategy for correcting articulation in children with cleft palate. Journal of Craniofacial Surgery, 29(6), 1490–1494.
Pamplona, C., Ysunza, A., & Morales, S. (2014). Strategies for treating compensatory articulation in patients with cleft palate. International Journal of Biomedical Sciences, 10(1), 43–51.
Pamplona, C., Ysunza, A., & Morales, S. (2017). Audiovisual materials are effective for enhancing the correction of articulation disorders in children with cleft palate. International Journal of Pediatric Otorhinolaryngology, 93, 17–23.
Pamplona, C., Ysunza, A., Patiño, C., Ramírez, E., Drucker, M., & Mazón, J. J. (2005). Speech summer camp for treating articulation disorders in cleft palate patients. International Journal of Pediatric Otorhinolaryngology, 69(3), 351–359. 2
Persson, C., Pedersen, N.-H., Hayden, C., Bowden, M., Aukner, R., Vindenes, H. A., Abyholm, F., Withby, D., Willadsen, E., & Lohmander, A. (2020). Scandcleft Project Trial 3: Comparison of speech outcomes in relation to sequence in 2-stage palatal repair procedures in 5-year-olds with unilateral cleft lip and palate. The Cleft Palate–Craniofacial Journal, 57(3), 352–363.
Peterson-Falzone, S. J., Hardin-Jones, M. A., & Karnell, M. P. (2010). Cleft palate speech (4th ed.). Mosby/Elsevier.
Prathanee, B. (2011). Cost effectiveness of speech camps for children with cleft palate in Thailand. Journal of the Medical Association of Thailand, 94, S33–S39.
Prathanee, B., Makarabhirom, K., Pummnum, T., Seepuaham, C., Jaiyong, P., & Pradubwong, S. (2014). Khon Kaen: A community-based speech therapy model for an area lacking in speech services for clefts. Southeast Asian Journal of Tropical Medicine and Public Health, 45(5), 1182–1195.
Prathanee, B., Pumnum, T., Yoodee, P., & Makarabhirom, K. (2020). Speech therapy model for patients with cleft palate in Lao People's Democratic Republic: Lack of speech services. International Journal of Pediatric Otorhinolaryngology, 138, 110366.
Prins, D., & Bloomer, H. H. (1965). A word intelligibility approach to the study of speech change in oral cleft patients. The Cleft Palate Journal, 2, 357–368.
Pumnum, T., Kum-ud, W., & Prathanee, B. (2015). A networking of community-based speech therapy: Borabue District, Maha Sarakham. Journal of the Medical Association of Thailand, 98, S120–S127.
R Core Team. (2013). R: A language and environment for statistical computing. R Foundation for Statistical Computing. http://www.R-project.org/
Revicki, D., Hays, R. D., Cella, D., & Sloan, J. (2008). Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. Journal of Clinical Epidemiology, 61(2), 102–109.
Roxburgh, Z., Cleland, J., & Scobbie, J. M. (2016). Multiple phonetically trained-listener comparisons of speech before and after articulatory intervention in two children with repaired submucous cleft palate. Clinical Linguistics & Phonetics, 30(3), 398–415.
Ruscello, D., & Vallino, L. (2014). The application of motor learning concepts to the treatment of children with compensatory speech sound errors. SIG 5 Perspectives on Speech Science and Orofacial Disorders, 24(2), 39–47.
Scherer, N. J., D'Antonio, L. L., & McGahey, H. (2008). Early intervention for speech impairment in children with cleft palate. The Cleft Palate–Craniofacial Journal, 45(1), 18–31.
Sell, D. (2005). Issues in perceptual speech analysis in cleft palate and related disorders: A review. International Journal of Language & Communication Disorders, 40(2), 103–121.
Shriberg, L. D., Austin, D., Lewis, B. A., McSweeny, J. L., & Wilson, D. L. (1997). The percentage of consonants correct (PCC) metric: Extensions and reliability data. Journal of Speech, Language, and Hearing Research, 40(4), 708–722.
Sritacha, P., Pumnum, T., & Prathanee, B. (2016). Speech correction for children with cleft lip and palate in community: Kantharawichai networking. Journal of the Medical Association of Thailand, 99, S1–S8.
Sterne, J. A. C., Hernán, M. A., Reeves, B. C., Savović, J., Berkman, N. D., Viswanathan, M., Henry, D., Altman, D. G., Ansari, M. T., Boutron, I., Carpenter, J. R., Chan, A.-W., Churchill, R., Deeks, J. J., Hróbjartsson, A., Kirkham, J., Jüni, P., Loke, Y. K., Pigott, T. D., … Higgins, J. P. T. (2016). ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. BMJ, 355.
Stiernman, M., Klintö, K., Persson, M., & Becker, M. (2021). Comparison of corresponding scores from the Cleft Hearing Appearance and Speech Questionnaire (CHASQ) and CLEFT-Q in Swedish patients with cleft lip and/or palate. The Cleft Palate–Craniofacial Journal, 58(6), 736–745.
Stoel-Gammon, C., & Sosa, A. V. (2007). Phonological development. In E. Hoff & M. Shatz (Eds.), Blackwell handbook of language development (pp. 238–256). Blackwell.
Sweeney, T., Hegarty, F., Powell, K., Deasy, L., Regan, M. O., & Sell, D. (2020). Randomized controlled trial comparing Parent Led Therapist Supervised Articulation Therapy (PLAT) with routine intervention for children with speech disorders associated with cleft palate. International Journal of Language & Communication Disorders, 55(5), 639–660.
Vallino-Napoli, L. D. (2011). Evaluation and evidence-based practice. In A. Lohmander & S. Howard (Eds.), Cleft palate speech—Assessment and intervention (pp. 317–358). Wiley-Blackwell.
Van Demark, D. R., & Hardin, M. A. (1986). Effectiveness of intensive articulation therapy for children with cleft palate. The Cleft Palate Journal, 23(3), 215–224.
Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1–48.
Wang, N. (2018). How to conduct a meta-analysis of proportions in R: A comprehensive tutorial. John Jay College of Criminal Justice.
Wasserstein, R. L., & Lazar, N. A. (2016). The ASA statement on p-values: Context, process, and purpose. The American Statistician, 70(2), 129–133.
Wasserstein, R. L., Schirm, A. L., & Lazar, N. A. (2019). Moving to a world beyond “p < 0.05.” The American Statistician, 73(Suppl. 1), 1–19.
Weir, J. P. (2005). Quantifying test–retest reliability using the intraclass correlation coefficient and the SEM. Journal of Strength and Conditioning Research, 19(1), 231–240.
Whitehill, T. L., Gotzke, C. L., & Hodge, M. (2011). Speech intelligibility. In Cleft palate speech: Assessment and intervention (pp. 293–304). Wiley.
Willadsen, E., Lohmander, A., Persson, C., Lundeborg, I., Alaluusua, S., Aukner, R., Bau, A., Boers, M., Bowden, M., Davies, J., Emborg, B., Havstam, C., Hayden, C., Henningsson, G., Holmefjord, A., Hölttä, E., Kisling-Møller, M., Kjøll, L., Lundberg, M., … Semb, G. (2017). Scandcleft randomised trials of primary surgery for unilateral cleft lip and palate: 5. Speech outcomes in 5-year-olds—Consonant proficiency and errors. Journal of Plastic Surgery and Hand Surgery, 51(1), 38–51.
Information & Authors
Information
Published In
Journal of Speech, Language, and Hearing Research
Volume 65 • Number 2 • February 2022
Pages: 555-573
PubMed: 34990556
Copyright
Copyright © 2022 The Authors. 
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License.History
- Received: Jul 1, 2021
- Revised: Sep 2, 2021
- Accepted: Sep 23, 2021
- Published online: Jan 6, 2022
- Published in issue: Feb 9, 2022
Authors
Metrics & Citations
Metrics
Article Metrics
View all metricsCitations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
Citing Literature
- Cassandra Alighieri, Febe Vandewiele, Valerie Pereira, What works for whom? A systematic review on personalized speech intervention in children and adolescents with a cleft palate, International Journal of Pediatric Otorhinolaryngology, 10.1016/j.ijporl.2025.112401, (112401), (2025).
- Griffin Lowry, Erika Rezende Silva, Jeannene Matthews, Ayumi Shoji, Timothy Turvey, George Blakey, David Zajac, Jeff Mielke, Laura Anne Jacox, A Lingual Ultrasound Study of Speech in Patients With Cleft Lip and Palate Following Orthognathic Surgery, Orthodontics & Craniofacial Research, 10.1111/ocr.12926, (2025).
- Ye-Sol Song, Jae-Woo Ryu, Yoon Ju Cho, Proposal of a Non-Face-To-Face Language Rehabilitation Service for the Recovery of Language Skills in Stroke Patients, Journal of Digital Contents Society, 10.9728/dcs.2025.26.3.717, 26, 3, (717-729), (2025).
- Cassandra Alighieri, Andrew Hodges, Jolien Verbeke, Katrien Kestens, Rica Albite, Raphaelle May Tan, Kim Bettens, Kristiane Van Lierde, Immediate individual effects of intensive group speech intervention on speech and health-related quality of life in adolescents with cleft palate: a descriptive study in the Philippines, Logopedics Phoniatrics Vocology, 10.1080/14015439.2025.2453134, (1-13), (2025).
- Rajani Singh, Cleft Lip and Cleft Palate: Incidence, Etiology and Development, Recent Advances in the Treatment of Orofacial Clefts, 10.5772/intechopen.114339, (2024).
- Chelsea L. Sommer, Natalie R. Wombacher, Prenatal to Adulthood: The Responsibility of the Speech-Language Pathologist on the Comprehensive Cleft Palate and Craniofacial Team, American Journal of Speech-Language Pathology, 10.1044/2024_AJSLP-24-00230, 34, 1, (12-31), (2024).
- Kathy L. Chapman, Thomas Sitzman, Adriane Baylis, Mary Hardin-Jones, Richard Kirschner, M'hamed (Hamy) Temkit, A Comparative Effectiveness Study of Speech and Surgical Outcomes: Study Overview, The Cleft Palate Craniofacial Journal, 10.1177/10556656241274242, (2024).
- Anusha Sundarrajan, Liz Nevolo, Gender-affirming voice training for trans women: effectiveness of training on patient-reported outcomes and listener perceptions of voice, Evidence-Based Communication Assessment and Intervention, 10.1080/17489539.2024.2403366, 18, 1, (23-32), (2024).
- Claire Cordella, Manuel J. Marte, Hantian Liu, Swathi Kiran, An Introduction to Machine Learning for Speech-Language Pathologists: Concepts, Terminology, and Emerging Applications, Perspectives of the ASHA Special Interest Groups, 10.1044/2024_PERSP-24-00037, 10, 2, (432-450), (2024).
- Louise Lendt, Magnus Becker, Marie Eriksson, Kristina Klintö, Foreign‐born 5‐year‐old children with cleft palate had poorer speech outcomes than their native‐born peers, Acta Paediatrica, 10.1111/apa.17385, 113, 12, (2628-2636), (2024).
- See more
View Options
View options
Sign In Options
ASHA member? If so, log in with your ASHA website credentials for full access.
Member Login
