Cleft palates with or without cleft lips (CP±L) result from structural impairments of the palate and mouth during the first trimester of pregnancy. An impaired fusion between maxillary and nasal elevations, due to insufficient neural crest cells, leads to cleft lip formation. Meanwhile, defective formation of lateral palatine processes causes cleft palate. After surgical repair, many children with CP±L continue to experience structural and functional abnormalities in the soft palate and pharyngeal wall, leading to speech and language deficits. These deficits typically include compensatory articulation disorders (CAD), abnormal resonance, delayed language development, voice disorders, and hearing impairments. Among these, CAD is the most prevalent speech disorder in children with CP±L [1]. Speech therapy should ideally commence early to improve speech and language skills before children reach school age. However, the limited number of speech and language pathologists (SLPs) in Thailand negatively impacts early intervention, affecting both duration and frequency of therapy sessions. Typically, children with CP±L receive only one speech therapy session every 1–3 months, delaying speech correction and necessitating extended treatment durations [2,3]. Previous strategies have attempted to address these limitations through training, community development, infrastructure improvements, and managing human resource shortages using diverse training models. Examples include speech summer camps [4], telepractice [5], community-based speech therapy models, and networking [2].

However, during pandemics or epidemics such as coronavirus disease 2019 (COVID-19), which necessitate social distancing, access to face-to-face speech therapy becomes significantly limited and challenging. Many countries have responded by implementing telemedicine services, utilizing technology and video conferencing to enable healthcare professionals to deliver diagnostic, treatment, and preventive healthcare remotely, overcoming temporal and spatial constraints. Telemedicine, or telepractice, is thus a viable strategy to facilitate speech therapy, as well as long-distance medical services in many fields [6]. Telepractice reduces the risk of communicable diseases and the expenses of having face-to-face speech therapy were reduced [7]. Speech therapy applications that increase children’s engagement and yield positive responses present a practical alternative during periods of social distancing. Furthermore, parents can assist their children’s speech practice at home using these applications [8]. Therefore, telepractice using articulation therapy applications should be considered for children with CP±L [5].

The objective of this study was to compare the percentage of correct consonants (PCC) before and after telepractice-based speech therapy using the Application for Articulation Therapy–Thai (AAT-T) [9] in children with CP±L.

This research study was approved by the Ethics Committee on Human Research of Khon Kaen University (project number HE654002, approved on April 22, 2022). All participating children and their parents received clear information about their involvement in the study. Children under 13 years old or those unable to read independently received detailed explanations from the research assistant.

This research employed a pre-experimental and post-experimental design with a single group. The study population consisted of children with CP±L.

Children were eligible for inclusion if they were between 5 and 13 years of age, had undergone palatoplasty at least 1 year prior, and demonstrated articulation errors. Exclusion criteria included: (1) conditions that could affect articulation, such as bilateral hearing loss greater than 40 dB, global developmental delay, or attention deficit hyperactivity disorder; (2) fewer than two articulation errors (excluding the /r/ sound, which is commonly and inconsistently acquired in Thai); or (3) symptoms of a cold, runny nose, or fever at the time of assessment.

Twenty-one children with CP±L were initially enrolled in the study. Two children were excluded based on the exclusion criteria, and one child withdrew due to caregiver inability to participate in speech therapy activities. A total of 18 children completed the study.

The sample size was calculated based on the estimation of the primary outcome, focusing on participation in speech therapy activities. Comparisons were conducted between pre- and post-intervention outcomes. Parameters included an alpha error of 0.01, statistical power of 95%, an effect size of 1.524, and a two-tailed analysis. These values were based on parameters from a previous study [10]. Accordingly, the required sample size was determined to be 18 participants.

Eighteen children with CP±L (6 females, 12 males) participated in this study, with a mean age of 7.69 years (standard deviation [SD], 2.13). The cleft palate types among participants included isolated cleft palate (three individuals), unilateral cleft lip and palate (nine individuals), and bilateral cleft lip and palate (six individuals).

Language and speech development at pre-assessment and speech and language skills results are presented in Table 1.

The articulation patterns identified in participants are summarized in Table 2. Descriptive statistics, including minimum, maximum, means, and SDs for pre- and post-intervention outcomes, are also detailed. Comparisons showed statistically significant improvements in PCC at both word and sentence levels, with mean differences (MDs) of 17.20 (SD, 17.37; 95% confidence interval [CI], 8.56–25.83) and 20.90 (SD, 16.51; 95% CI, 12.69–29.11), respectively (Table 3). Fig. 1 provides box plots illustrating PCC scores before and after the intervention. An analysis of improvement levels with respect to age, sex, and cleft type revealed no statistically significant differences related to sex or cleft type. However, significant findings included: (1) older children (aged 9–13 years) demonstrated better outcomes than younger children (aged 5–8 years) at the word level (p<0.05); and (2) children with normal development exhibited greater progression in PCC at the sentence level compared to children with delayed speech and language development.

Children with CP±L exhibit structural defects in the facial region, potentially affecting speech and language development. In this study, five out of 18 children (27.78%) showed delayed language development, consistent with previous research indicating that children with CP±L are at a higher risk for delayed speech and language development compared to typically developing peers [1,13]. Oral examinations revealed a high prevalence of soft palate abnormalities, both structural (83%) and functional (50%). These abnormalities likely contributed to the high incidence of hypernasality (77.78%) observed, aligning with findings from Rezaei et al. in 2022 [14] and surpassing previously reported hypernasality rates (36.11%–48%) [1,13]. This suggests that residual structural defects and functional limitations of the soft palate directly affect hypernasality [15,16]. The rate of voice abnormalities was 22.22%, slightly lower than in previous research (30.56%) [17]. Such abnormalities are often caused by soft palate dysfunction and resonance issues, potentially leading children to speak louder or shout in an effort to compensate for nasal acoustic energy or velopharyngeal insufficiency (VPI) [16]. Facial grimaces were notably high at 77.78%, compared to the 35.70% found in an earlier study [18]. This increase may reflect children’s attempts to control facial movements to compensate for speech deficiencies, resulting in an elevated prevalence of facial grimace [19]. Rates of understandability (normal: 55.5%; abnormal: 44.44%) were similar to those previously reported [20,21]. Acceptability deviations were found in 83.33% of cases, consistent with findings by Ooppanasak et al. in 2019 [22]. The higher prevalence of speech and language abnormalities observed in this study compared to earlier studies could be attributed to the specific inclusion of children with CP±L who exhibited articulation errors directly impacting these skills.

Interestingly, hypernasality, facial grimace, understandability, and acceptability improved following the intervention, with voice disorders being the exception (Table 1). These results indicate that specifically targeting articulation errors through placement and manner correction in speech therapy for CAD can improve overall speech and language abilities [16,23]. Voice disorders did not significantly benefit from the intervention, possibly due to four children experiencing temporary abnormal voice quality from colds during the assessment or previous vocal misuse at school.

The most common articulation error patterns identified were trilling errors and functional articulation disorders, confirming findings from previous studies [24]. Approximately 80% of 11-year-old children exhibit errors with the /r/ sound, potentially because it requires precise tongue tip manipulation and develops more slowly even in typically developing children [24]. Functional articulation disorders persisted in children with CP±L due to residual oral physiological defects causing deviations in manner and placement, consistent with earlier research [25,26]. The most common CAD pattern identified was velar substitution (66.67%–83.33%), consistent with Klinto et al. in 2016 [27]. Unlike previous studies, where glottal stops were more commonly reported [14,15], this study highlights that most CAD patterns in children with CP±L involve posterior oral placements, likely due to velopharyngeal dysfunction. This mechanism compensates for air leakage into the nasal cavity and hypernasality [27]. After speech therapy, a reduction in CAD patterns was observed (Table 2), suggesting improved pronunciation accuracy, phonological awareness, and effective use of specific therapeutic strategies [14]. Post-training analysis revealed that articulation errors decreased in children with mild and moderate VPI, specifically for nasalized voiced pressure consonants. Most patients with moderate hypernasality improved by at least one level following therapy, with one achieving normal resonance. However, one child with severe VPI (A21C) exhibited no improvement in PCC at the sentence level and negative results at the word level. Contributing factors to these negative outcomes included poor cooperation from the child and caregiver in home practice and severe VPI coupled with delayed speech and language development. This suggests that speech therapy effectively reduces articulation errors and might reduce the necessity for surgical VPI repair in patients with moderate hypernasality. However, speech therapy alone may be insufficient for severe hypernasality, which involves obligatory articulation errors requiring surgical reconstruction.

The main finding of this study indicates that all participating children demonstrated a reduction in articulation errors, reflected by increased PCC scores (Fig. 1). Statistically significant improvements were observed at both word and sentence levels (word level: MD, 17.20; SD, 17.37; 95% CI, 8.56–25.83; sentence level: MD, 20.90; SD, 16.51; 95% CI, 12.69–29.11). These findings align with previous research [3,5,28], confirming that telepractice using AAT-T effectively corrects articulation errors and significantly reduces the number of errors post-intervention. An important aspect of this approach is the integration of telepractice with face-to-face speech therapy sessions. This combination is particularly beneficial for adjusting sound patterns, such as weak oral high pressure or phoneme-specific nasal air emissions, which typically require visual feedback. Solely relying on auditory cues through an online platform might not suffice. Face-to-face sessions enable SLPs to correct specific sounds directly, provide tailored training recommendations, and guide parents more effectively in adapting training techniques for home practice. This study highlighted that using technological tools (such as the AAT-T) for home speech practice can effectively motivate and engage both children and caregivers. Using the application at home helps maintain pronunciation consistency, enhances training efficiency, and supports ongoing therapy between appointments. Future studies should explore the long-term efficacy of the intervention, potentially extending study durations to 1 year to achieve more pronounced outcomes [2,3].

Regarding satisfaction, caregivers and participants rated highly the duration of training sessions, suitability of the AAT-T application, ease of using the Zoom application for telepractice, convenience of communication via Zoom or Line for inquiries, and reduced costs and travel requirements for speech therapy services. This training model is particularly applicable in regions with similar contexts requiring social distancing and in developing countries experiencing limited access to cleft palate speech therapy services. Furthermore, adapting the AAT-T or similar articulation therapy applications to other languages could enhance accessibility and effectiveness in various linguistic contexts.

Occasional instability in internet connectivity, especially at participants’ homes, may have caused disruptions and negatively affected outcomes. Additionally, this study lacked a face-to-face-only control group for comparison due to social distancing measures necessitated by the COVID-19 pandemic. Future research incorporating a control group is recommended to validate these findings further.

In conclusion, telepractice using the AAT-T application significantly improved PCC scores among children with CP±L.