Multidomain Developmental Indicators in 4- to 9-Year-Old Children with Oral Motor and Speech Sound Disorders

Abstract

Introduction

Persistent speech sound disorders (SSDs) are common in childhood and affect communication, literacy, and social development. Identifying risk indicators (RIs) and predictors for atypical oral motor and speech development is crucial for early intervention. This study examined medical, developmental, oral-behavioural, and hereditary RIs in children with SSD and compared diagnostic subgroups to typically developing peers.

Methods

A clinical cohort of 198 children (ages 4–9) referred for specialist assessment was compared to 77 age-matched controls. Children were classified into four subgroups: motor speech disorder, with/without oral motor developmental delay and/or language-oriented disorder (MSD+), language-oriented disorder with oral motor developmental delay, no MSD (LD + ODD), ODD-only, and LD-only. Thirty-one RIs were analyzed using the chi-square or Fisher’s exact tests with Bonferroni correction. Multivariate binary backwards logistic regression identified predictors of group membership (clinical vs. control).

Results

The clinical group showed 11 of 31 RIs significantly more prevalent than controls (p < 0.0016). Children with multiple diagnoses (MSD+ and LD + ODD) had the highest RI counts (12 and 11, respectively), while single-diagnosis groups showed fewer RIs (LD-only: 2; ODD-only: 6). Significant RIs included medical (adenoid surgery, pulmonary disease), oral behaviours (mouth stimuli, selective eating), and developmental indicators (abnormal or absent crawling, delayed bladder control, fine/gross motor delay, non-canonical babble, poor attention during story listening). Family history of speech/language delay and literacy difficulties was also significant. The regression model demonstrated an excellent fit (Nagelkerke R² = 0.69; classification accuracy = 86.8%). Strong predictors for having an oral motor and/or a speech disorder included adenoid surgery (OR = 63.49), ear tube surgery (OR = 60.53), mouth stimuli behaviours (OR = 9.77), non-canonical babble (OR = 10.45), abnormal crawling (OR = 12.20), and family history of speech/language delay (OR = 9.89).

Conclusion

Children with SSD often present multiple RIs across medical, developmental, oral-behavioural, and hereditary domains, especially those with combined diagnoses. Findings highlight the need for early, multidomain assessment and suggest that RIs such as oral motor behaviours and early developmental delays may inform screening and intervention strategies, supporting clinicians in identifying children who could benefit from tailored early intervention.

Plain Language Summary

Some children have ongoing difficulties saying speech sounds clearly. These problems can affect how they communicate, learn to read, and interact with others. This study explored what might increase the chances of these speech difficulties happening. Researchers looked at almost 200 children aged 4–9 who were referred for specialist help and compared them with 77 children who developed speech normally. They examined many possible risk indicators (RIs), such as medical history, early development, mouth habits, and family background. Children with speech difficulties often had more RIs than other children. Common ones included having surgery for adenoids or ear tubes, unusual mouth habits like putting objects in the mouth, delays in crawling or bladder control, and early speech signs such as missing typical babbling patterns. Family history of speech or language delay was also important. Children with several diagnoses had the highest number of RIs, while those with only one diagnosis had fewer. The study shows that speech difficulties often occur together with other developmental challenges. These findings suggest that certain early experiences and behaviours may be linked to later speech problems. Understanding these patterns can help researchers and clinicians learn more about how speech difficulties develop and why some children are more affected than others. This knowledge may also guide future studies on how different areas of development such as movement, attention, and oral behaviours interact with speech and language growth over time.

Introduction

Atypical speech sound development in children includes a range of conditions that affect the planning, coordination, and execution of speech movements. Central among these are speech sound disorders (SSDs), which encompass both phonological and motor-based impairments [1], including motor speech disorders (MSDs), such as childhood apraxia of speech (CAS), dysarthria, oral dyspraxia affecting speech gestures, and speech motor delay [2, 3]. These conditions are increasingly recognized as neurodevelopmental in origin, arising from disruptions in the complex interplay between genetic, neurological, and environmental factors that support speech and language acquisition [2, 4, 5]. Children with SSD frequently show delays in language development [6, 7], as well as in fine and gross motor milestones [8, 9], and atypical sensory processing has also been observed in this population [10, 11]. These delays in language, oral and general motor milestones, and difficulties in sensorimotor integration may reflect underlying neurobiological vulnerabilities that constitute potential risk indicators (RIs) [9, 10, 12–15]. Such indicators often co-occur with other neurodevelopmental disorders, compounding challenges in speech, language, and socio-emotional functioning [16, 17]. Importantly, speech disorders have also been linked to increased psychiatric vulnerability during adolescence [18].

Identifying RIs and predictors for persistent SSD in combination with thorough differential diagnosing assessments [19–21] is essential for guiding clinical decisions, particularly when determining the timing and necessity of intervention. For clinicians, the challenge lies in accurately assessing prognosis and deciding whether to intervene early or adopt a wait-and-see approach [22, 23], especially given that many children eventually catch up with their typically developing peers by around age six [24–27]. Which RIs to take into consideration has somewhat been researched [5, 28–30]. Several studies have explored demographic, familial, developmental, medical, and functional factors that may predict persistent SSDs, though findings have not always been consistent [29]. In their study, Wren and colleagues presented a thorough background that synthesized existing literature on RIs in children with SSD. Strong indicators included abnormal early vocalization, delayed language development, literacy difficulties, family history of literacy delay, and general motor development. Less consistent associations were found for socioeconomic status, gender (e.g., being male), ear, nose, throat (ENT) problems and feeding difficulties. Persistent hearing problems have emerged as a particularly significant medical indicator, logically linked to increased risk for both expressive and receptive language difficulties [31, 32]. Asthma has also been reported as a common condition among children with SSD [33]. Genetic contributions have received increasing attention, particularly in more severe conditions such as CAS [5, 12]. Although CAS is considered one of the less common diagnoses within the SSD group [2], it has received comparatively more research attention in recent decades. This focus has contributed to a deeper understanding of MSDs per se but has also meant that other MSD conditions remain less explored in the literature [2, 3, 21]. Although no single gene has been definitively linked to CAS, genetic studies suggest a heterogeneous phenotype with overlap across neurodevelopmental disorders, such as autism spectrum disorder and developmental coordination disorder and neurological, psychiatric and metabolic conditions [4, 5]. These findings support the use of family history and neurodevelopmental profiling in early identification and risk assessment. Understanding medical conditions, developmental patterns and family history is crucial for improving early identification and tailoring intervention strategies. Although oral motor development is increasingly recognized as important, its role as a RI for early identification of SSDs remains underexplored. Evidence suggests that delays in oral motor milestones, such as reduced chewing efficiency, weak sucking, and general oral motor immaturity may signal risk for persistent speech disorders [13, 15, 34]. Motor development supports language acquisition through activities like hand-to-mouth and oral exploration, which are linked to play, cognition, and speech [35–39]. Limited oral exploration has been noted in preterm children, while prolonged oral stimulation at later ages may relate to severe neurodevelopmental conditions [40, 41]. Other RIs include pacifier use, which can affect oral motor skills, though findings on its link to phonological impairment are mixed [42, 43]. There is a pressing need to better understand how sensory, motor, and language domains converge in children with speech and language difficulties. This study addresses that gap by investigating oral motor and speech sound development in a clinical cohort, with the aim of identifying RIs that can inform prognosis and guide timely intervention. By focusing on exploring the RIs, the study contributes to a more nuanced understanding of which children may benefit from early support, helping clinicians make more informed decisions in the critical early years of development. Given that much of the existing SSD literature is based on English-speaking populations, this study also helps address the need for large-scale investigations in Swedish.

Research Questions

(1) Which medical, oral-behavioural, developmental, and hereditary indicators distinguish children with oral motor and SSDs from typically developing peers?

(2) Are there differences in these indicators across diagnostic subgroups?

Method and Procedures

Recruitment of Participants

The present study was part of a larger research project conducted at the Centre for eating, speech and oral motor function (OMC), within the Division of Speech and Language Pathology, Department of Neurology, Danderyd Hospital, Stockholm, Sweden. Participants in the study group (SG) were drawn from a consecutive sample of children referred for routine differential diagnostic assessment of oral and speech motor disorders. All children between 4 and 9 years of age, with at least one caregiver who could speak and read Swedish, were invited to participate between January 2022 and October 2023. Five speech language pathologists (SLPs), working in their regular clinical roles at OMC, supported the data collection as part of their standard diagnostic procedures. Informed consent was obtained from the child’s caregiver when visiting the clinic. A total of 211 patients agreed to participate. Children that did not receive an SLP-diagnosis or had received a diagnosis for structural upper airway and oral conditions (ankyloglossia, hypertrophy of adenoid or tonsils) (n = 13) were not included in the study, leaving 198 patients (139 boys and 59 girls) meeting the inclusion criteria. Data from an age-matched control group (CG) of 77 typically developing children (37 boys and 40 girls), also being part of the larger research project, were collected between September 2022 and March 2024 in a consecutive manner during their regular dental examinations at the Division of Paediatric Dentistry at Karolinska Institutet, at a private dental clinic, and from colleagues’ children at the Division of Speech and Language Pathology. Prior to the data collection, the responsible clinicians were calibrated in the procedure to ensure consistency. To be included in the CG, one parent had to be able to speak and read Swedish, the child should not have any known delay of speech and language development nor have received any intervention from an SLP, and they should not have any known general developmental delay or condition. The final sample included 275 children (176 boys and 99 girls) with a mean age of 6:2 years.

Assessment Procedure

All participants of the SG were assessed according to the specialist clinic’s standard intake protocols. Three different regular standardized assessment batteries are used at the OMC dependent on the referral question. These batteries aid differential diagnosis of motor speech disorders, phonological disorders, oral motor developmental delay (ODD), and structural oral dysfunctions. Assessments lasted 30 min to 2 h, depending on referral question and assessment battery that was used. The timelier extensive battery always included breaks. All assessments were stopped if the child refused to cooperate. Results were documented in medical records, and after analysis, a diagnosis and detailed feedback were provided to caregivers and the referring clinician (see Björelius et al. [11] for details). For the clinical group, the standardized anamnesis questionnaire was sent by post to the caregivers prior to the assessment, in accordance with routine procedures at the clinic. Caregivers completed the form at home, and the responses were reviewed during an interview with the responsible SLP, usually conducted by phone before the visit, but occasionally in person at the clinic. The outcome of the anamnesis provides clinicians with contextual information that may be relevant, such as additional diagnoses, medical history, early development, oral behaviours, feeding issues, family situation, and heredity. Importantly, these caregiver responses are not used for diagnostic decisions; they serve only as background information to support understanding of the child and their family. For the CG group, the same anamnesis questionnaire was used. Caregivers completed the questionnaire and were interviewed by the responsible clinician during the child’s scheduled appointment. Responses were manually transferred to Excel for data processing and statistical analysis.

Diagnostics and Terminology

SLPs in Sweden use criteria and diagnostic terms from the international classification system ICD-10 (WHO, 2004). A thorough overview of the Swedish diagnoses translated to English is presented in Figure 1. Note that these do not directly correspond to the diagnostic terms used in the English SLP literature. Some children had been diagnosed with a specific language impairment prior to the OMC assessment, and these were grouped under the umbrella term language-oriented disorders (LDs) (Fig. 1). In Sweden, the criteria for diagnosing phonological language disorder include the child having dysfunctions in the auditory discrimination and perception of speech sounds affecting the production of specific speech sounds. Details of the OMC guidelines and the symptom criteria applied to determine the SLP diagnoses, including the clinical differentiation between MSDs and ODD, are presented in online supplementary material 1 (for all online suppl. material, see https://doi.org/10.1159/000551061).

Fig. 1.

The Swedish diagnostic system. International classification ICD-10 (WHO). The diagnoses used by Swedish SLPs in clinical settings and the translation to English terms.

Statistical Methods

The children in the entire sample of oral and SSDs were divided into four diagnostic subgroups. MSD+ are children with a MSD diagnosis (childhood apraxia of speech; dysarthria; speech motor delay and oral dyspraxia) (Fig. 1) that can also have an ODD and/or a language-oriented diagnosis (LD) [phonological language disorder; expressive language disorder and developmental language disorder] (Fig. 1). Children with LD + ODD do not have an MSD diagnosis and ODD-only have solely an ODD diagnosis. LD-only are children with no MSD nor ODD diagnoses. The distribution and overlap of diagnoses are illustrated in Figure 2 [44].

Fig. 2.

Distribution of diagnostic groups and overlaps in clinical sample. Area-proportional Euler diagram [43] visualizing the distribution of the clinical sample across main diagnostic groups: motor speech disorders (MSD), oral motor developmental delay (ODD), and language-oriented diagnoses (LDs). Overlapping cases between these groups are also shown. MSD subgroups in the figure include: MSD (only); MSD + ODD; MSD + LD; and MSD + ODD + LD. Additional subgroups are LD + ODD, LD (only), and ODD (only). Age range: 4–9 years.

All statistical analyses were performed using IBM SPSS Statistics, version 29.0.2.0. Descriptive statistics were used to summarize demographic characteristics for the entire clinical group as well as for each diagnostic subgroup separately, including age and gender. A CG of typically developing children was included for comparison. All children were between 4 and 9 years of age. Mean age is presented in Table 1.

Table 1.

The table presents the number and percentage of children with each circumstance or RI across 5 groups: control group (CG); motor speech disorder-only or with concomitant language-oriented disorder and oral motor developmental delay (MSD+); language-oriented disorder with oral motor developmental delay (LD+ODD); language-oriented disorder only (LD-only), and oral motor developmental delay only (ODD-only) and the entire sample with oral motor and SSDs

Circumstance/risk indicator	CG, N = 77	MSD+, N = 110	LD + ODD, N = 34	LD-only, N = 21	ODD-only, N = 33	Total SG, N = 198
	mean age (6:5)	mean age (6:1)	mean age (6:3)	mean age (5:9)	mean age (5:9)	mean age (6:0)
	(n/%)	(n/%) vs. CG p value	(n/%) vs. CG p value	(n/%) vs. CG p value	(n/%) vs. CG p value	(n/%) vs. CG p value
Medical history/illness
Several otis media	6 (7.9)	25 (23.1) 0.006**	2 (5.9)	6 (28.6) 0.020*	7 (21.2)	40 (20.4) 0.013*
Op. ear tubes	1 (1.3)	6 (5.5)	1 (2.9)	4 (19.0) 0.007**	4 (12.1) 0.029*	15 (7.6) 0.047*
Op. adenoid	1 (1.3)	15 (13.8) 0.003**	7 (20.6) <0.001***	4 (19.0) 0.007**	6 (18.2) 0.003**	32 (16.2) <0.001***
Op. tonsils	1 (1.3)	11 (10.0) 0.029*	3 (8.8)	1 (4.8)	4 (12.1) 0.029*	19 (9.6) 0.018*
Op. tongue tie	2 (2.6)	11 (10.1)	3 (8.8)	3 (14.3)	3 (9.4)	20 (10.2) 0.040*
Colic	5 (6.6)	10 (9.3)	0	0	4 (12.1)	14 (7.2)
Obstipation	6 (8.0)	17 (15.6)	8 (24.2)	2 (9.5)	5 (15.2)	32 (16.3)
Pulmonary (asthma/croup)	4 (5.3)	34 (31.2) <0.001***	13 (39.4) <0.001***	6 (28.6) 0.006**	9 (27.3) 0.002**	62 (31.6) <0.001***
RS virus	0	7 (6.4) 0.043*	2 (6.1)	2 (9.5)	3 (9.1) 0.026*	14 (7.1) 0.013*
Allergy	5 (6.6)	14 (13.1)	4 (12.1)	1 (4.8)	1 (3.0)	20 (10.3)
Sum significant	​	5/10	2/10	4/10	5/10	7/10
After Bonferroni correction	​	1/10	2/10	0/10	0/10	2/10
Oral behaviour/feeding
No breastfeeding	7 (9.2)	25 (22.9) 0.015*	12 (36.4).<0.001***	3 (14.3)	7 (21.2)	47 (24.0) 0.006**
Pacifier first year/s in life; high frequence day and night	13 (17.6)	32 (30.2)	12 (37.5) 0.026*	3 (14.3)	7 (24.1)	54 (28.7)
Selective eater	1 (1.3)	27 (25.5) <0.001***	9 (26.5) <0.001***	5 (23.8) 0.002**	6 (18.2) 0.003**	47 (24.2) <0.001***
Mouth stimuli: sucking, chewing, and/or biting on objects or fingers	7 (9.3)	46 (41.8) <0.001***	14 (41.2) <0.001***	6 (26.1)	15 (45.5) <0.001***	81 (40.9) <0.001***
Sum significant	​	3/4	4/4	1/4	2/4	3/4
After Bonferroni correction	​	2/4	3/4	0/4	1/4	2/4
Developmental domains
Hearing dysfunction	1 (1.4)	2 (1.9)	1 (3.0)	1 (4.8)	2 (6.3)	6 (3.1)
Vision dysfunction	5 (7.1)	15 (14.2)	5 (14.7)	2 (9.5)	5 (16.1)	27 (14.1)
Delayed bladder control development	0	16 (14.8) <0.001***	5 (15.2) 0.002**	2 (9.5) 0.045*	4 (12.1) 0.007**	27 (13.8) <0.001***
Gross motor development	0	33 (30.0) <0.001***	10 (29.4) <0.001***	3 (14.3) 0.009**	10 (30.3) <0.001***	56 (28.3) <0.001***
Fine motor development	0	38 (34.9) <0.001***	9 (26.5) <0.001***	3 (14.3) 0.009**	9 (27.3) <0.001***	59 (29.9) <0.001***
Crawling: abnormal/non	7 (9.5)	35 (33.3) <0.001***	9 (36.0) 0.004**	3 (15.8)	9 (31.0) 0.013*	56 (31.5) <0.001***
Walked early (<10 months)	9 (12.2)	13 (12.4)	4 (12.9)	2 (10.0)	3 (9.1)	22 (11.6)
Walked late (>18 months)	6 (8.1)	23 (21.9) 0.014*	5 (16.1)	1 (5.0)	9 (27.3) 0.014*	38 (20.1) 0.019*
No canonical babble or quiet	3 (4.4)	36 (35.0) <0.001***	10 (34.5) <0.001***	10 (52.6) <0.001***	6 (22.2) 0.014*	62 (34.8) <0.001***
Late >18 months first words	8 (10.5)	18 (18.0)	8 (26.7)	2 (9.5)	6 (20.7)	34 (18.9)
Concentration difficulties when listening to story/book	1 (1.3)	30 (27.3) <0.001***	10 (29.4) <0.001***	4 (19.0) 0.007**	8 (24.2) <0.001***	52 (26.3) <0.001***
Do not play with friends	8 (11.0)	27 (25.5) 0.016*	5 (15.6)	5 (26.3)	10 (32.2) 0.009*	47 (25.0) 0.013*
Sum significant	​	8/12	6/12	5/12	8/12	8/12
After Bonferroni correction	​	6/12	4/12	1/12	3/12	6/12
Family history
Speech and language delay	11 (14.5)	65 (59.6) <0.001***	19 (67.6) <0.001***	14 (66.7) <0.001***	18 (56.3) <0.001***	116 (59.5) <0.001***
Reading and writing delay/dyslexia	12 (15.8)	44 (40.4) <0.001***	16 (58.5) <0.001***	7 (33.3)	9 (28.1)	76 (39.0) <0.001***
Stuttering	7 (9.2)	14 (12.8)	3 (9.1)	1 (4.8)	9 (28.1) 0.017*	27 (13.8)
ADHD/autism	19 (25.0)	44 (40.4) 0.030*	12 (36.4)	7 (33.3)	14 (43.8)	77 (39.5) 0.025*
Sum significant	​	3/4	2/4	1/4	2/4	3/4
After Bonferroni correction	​	2/4	2/4	1/4	1/4	2/4
Concomitant NDD
NDD diagnoses	0	21(19.1) <0.001***	2 (5.9)	1 (4.8)	7 (21.2) <0.001***	31 (15.7) <0.001***
Sum significant	​	1/1	0/1	0/1	1/1	1/1
After Bonferroni correction	​	1/1	0/1	0/1	1/1	1/1
Total sum significant	​	20/31	14/31	11/31	18/31	22/31
After Bonferroni correction	​	12/31	11/31	2/31	6/31	11/31

p values represent comparisons between each clinical group and the CG using chi-square tests. Bonferroni correction was applied across variables and applied to adjust for multiple testing. N = group sizes. Age 4–9 years.

p values Bonferroni adjusted across 31 variables.

p = <0.05*, <0.01**, <0.0016***.

NDD diagnoses by group: (i) MSD+ (n = 21): 5 autism, 1 ADHD, 2 ADHD + autism, 1 ADHD + DCD, 1 ADHD + ID, 1 autism + ID, 1 autism + DCD + genetic, 1 ADHD + autism + ID + CP, 1 ID + CP + genetic, 2 CP, 2 DCD, 1 ID, 2 genetic. (ii) LD + ODD (n = 2): 2 DCD. (iii) LD-only (n = 1): 1 DCD. (iv) ODD-only (n = 7): 1 ADHD, 1 ADHD + autism, 1 autism + CP, 1 autism + ID, 1 DCD + genetic, 1 DCD + ID + genetic, 1 genetic.

Potential differences in the prevalence of RIs were examined using Chi-square or Fisher’s exact tests. In total, 31 variables were compared across the entire clinical sample, the diagnostic subgroups, and the CG. Additional analyses were also conducted for the MSD+ subgroups found in Figure 2 (MSD-only; MSD + ODD; MSD + LD and MSD + ODD + LD). See online supplementary material 2 for details. A Bonferroni correction was applied to control for multiple testing, setting the adjusted significance threshold at p < 0.0016. All p values are reported; variables meeting this stricter criterion are flagged as Bonferroni-significant (Table 1).

Multivariate binary logistic regression with backward likelihood ratio stepwise selection analyses were conducted for the total SG to identify which RIs significantly predicted group membership (clinical vs. control) while controlling for multiple variables simultaneously. Predictors with very low counts (e.g., one or a few cases) were retained but interpreted cautiously due to wide confidence intervals (CIs) associated with sparse data. Model performance was assessed using Nagelkerke’s pseudo-R² as an indicator of explanatory power, the Hosmer-Lemeshow test to evaluate calibration, and the classification table to estimate predictive accuracy. Both sensitivity (proportion of correctly identified diagnostic cases) and specificity (proportion of correctly identified controls) were reported to provide a balanced evaluation of model performance. Odds ratios (ORs), p values, and 95% CIs were reported for each retained predictor (Fig. 3). The significance thresholds for the regression analyses were set at α < 0.05.

Fig. 3.

Multimodal RIs predicting oral motor and SSDs. Forest plot showing RIs associated with the clinical group oral motor and SSDs (N = 198), based on multivariate binary logistic regression with backward likelihood ratio (LR) stepwise selection. Twenty-six indicators are included (variables with zero counts were removed); see Table 1. p values, odds ratios (ORs), and 95% confidence intervals (CIs) are presented for each variable. Age range: 4–9 years. An arrow indicates when the upper CI extends beyond the plotted range.

Results

The chi-square analyses with Bonferroni correction revealed that the SG group had 11 of 31 RIs significantly more prevalent compared to the CG group (Table 1). Within diagnostic subgroups, children with multiple SLP diagnoses showed the highest number of RIs: MSD+ had 12 RIs and LD + ODD had 11 RIs. Children with a single diagnosis had fewer RIs: LD-only had 2 RIs and ODD-only had 6 RIs (all p’s < 0.001). The significant RIs identified for the entire SG were in the Medical Domain, operation of the adenoid and history of pulmonary disease. In the Oral Behaviour/Feeding Domain, selective eater and using mouth stimuli (chewing, sucking, licking on fingers, and/or objects) were found to be significant. In the developmental domain several RIs were identified as significant, delayed bladder control development, fine and gross motor development, abnormal or no crawling, no canonical babble and difficulties with concentration when listening to story/book. Furthermore, significant RIs in the family history domain were heritage of delayed speech and language development as well as delay in reading and writing or a diagnosis of (all p’s < 0.001). The presence of one or multiple concomitant NDD diagnoses also showed a significant association. Table 1 presents details on the identified RIs for the entire SG group and each diagnostic subgroup. Additional chi-square analyses were performed for the subgroups within MSD+ (e.g., MSD-only, MSD + ODD, MSD + LD, MSD + ODD + LD). These detailed results are presented in online supplementary material 2. A multivariate binary logistic regression was conducted to identify which RIs were associated as predictors for children with atypical oral motor and speech sound development. The logistic regression model showed an excellent fit to the data for the entire study sample (n = 198), Hosmer-Lemeshow χ²(8) = 3.70, p = 0.88, with a high explanatory value (Nagelkerke R² = 0.69). The model correctly classified 86.8% of cases overall, with a sensitivity of 93.8% for the entire SG and a specificity of 72.1% for the CG. Several indicators were significantly associated with atypical oral motor and speech sound development. Operations involving ear tubes (OR = 60.53, p = 0.006) and adenoids (OR = 63.49, p = 0.004) showed particularly strong associations, although the wide CIs suggest limited precision so these results should be interpreted with caution. Mouth stimuli behaviours such as chewing, sucking and licking on objects or fingers (OR = 9.77, p < 0.001), non-canonical babbling (OR = 10.45, p = 0.002), and abnormal or absent crawling (OR = 12.20, p = 0.002) were also significantly associated. A family history of speech and language delay (OR = 9.89, p < 0.001) and issues with concentration (OR = 15.17, p = 0.026) further supported the relevance of these developmental RIs. Detailed statistical results on all predictors are presented in Figure 3.

Discussion

This study explores a broad set of RIs associated with atypical oral motor and speech sound development in children. The findings contribute to the growing understanding of how medical, developmental, oral behaviours/feeding, and hereditary factors may interact in shaping speech and language outcomes [5, 29]. While some of the identified RIs have been previously described [31], the current results offer a more detailed picture of how these indicators may cluster, particularly in children with complex diagnostic profiles. The identification of 11 significant RIs in the entire SG, and even more in subgroups with multiple diagnoses, highlights the multifactorial nature of these developmental difficulties. The results also resonate with developmental theories suggesting that early motor and sensory processes are closely linked to the emergence of speech and language [13, 35, 38, 39]. The strong associations found in the logistic regression model further support the clinical relevance of these RIs, particularly in guiding early intervention strategies.

Co-Occurring Diagnoses and Neurodevelopmental Profiles

The high prevalence of co-occurring diagnoses in the clinical sample, where 67% of children received two or more diagnoses, reflects the complexity of speech and language disorders in real-world clinical settings [6]. The overlap between MSDs, language-oriented disorders (LDs), and ODD suggests that these conditions rarely occur in isolation [6, 34]. Notably, ODD was present in 69% of the children, yet this diagnosis and its potential role as part of a broader developmental delay is seldom highlighted in previous research. The clinical relevance of ODD may be underestimated, despite its frequent co-occurrence with other speech and language difficulties. The high incidence of ODD may reflect missed oral-motor and oral-sensory milestones that are suggested to influence speech development in R. Kent’s model “Developmental Functional Modules” [13]. These findings point to a dimensional understanding of speech and language disorders, where symptoms often overlap across motor, linguistic, and sensory domains rather than fitting neatly into separate diagnostic categories [11, 45]. That 16% of the children had one or more co-occurring neurodevelopmental diagnoses further underscores the importance of including a broad neurodevelopmental perspective when assessing children with oral motor and speech sound difficulties.

Risk and Predictive Indicators

The study identified several RIs that were significantly more prevalent in the clinical group compared to controls. These included medical (e.g., adenoid and tongue-tie operations, pulmonary disease), oral behaviours (e.g., mouth stimuli), developmental delays (e.g., abnormal crawling, non-canonical babble), and family history of speech and language difficulties. Some of these RIs have been described in earlier studies [12, 29], and the current results offer further support for their relevance in children with speech and language difficulties.

The multivariate logistic regression model demonstrated excellent fit and predictive power, correctly classifying nearly 87% of cases. Strong associations were found for procedures such as adenoidectomy, tongue-tie release, and ear tube insertion indicating possible vulnerabilities that may influence auditory input and, consequently, speech development [31]. These findings are consistent with evidence that sleep-disordered breathing caused by adenoid hypertrophy can contribute to delays in overall development [46]. Together, these patterns suggest that early ENT conditions could be one of several indicators with potential relevance for developmental trajectories [46], but their specific contribution is not yet well understood. Furthermore, several RIs tended to co-occur in children with more complex diagnostic profiles, especially those with multiple SLP diagnoses. This clustering is consistent with broader developmental involvement in such cases and underscores the value of considering combinations of indicators rather than single features. Among these, prolonged oral behaviours such as mouth stimuli (i.e., chewing, sucking, and licking on objects) and selective eating may reflect underlying oral-motor developmental vulnerabilities relevant to speech development and warrant clinical attention. These oral and eating behaviours may also represent features of the condition itself. Similarly, abnormal crawling stands out as a relevant predictor supporting the view that early sensorimotor patterns may be closely linked to speech outcomes, as also reported by other studies [13, 36, 38]. The connection between oral motor behaviours and broader developmental processes may hold important implications for early identification [2]. Furthermore, a family history of speech and language delay emerged as a significant predictor, reinforcing previous findings and underscoring the role of genetic factors in shaping developmental trajectories [5, 16].

Clinical Implications

These findings have direct implications for clinical practice. First, they underscore the importance of multidimensional assessment, particularly for children presenting with multiple symptoms across motor, language, and sensory domains. Second, the identification of early and prolonged signs of delayed development such as abnormal crawling, non-canonical babble, and mouth stimuli behaviours may inform screening and help clinicians identify children at risk before speech difficulties become entrenched. Third, the presence of multiple RIs in children with MSD+ and LD + ODD profiles suggests that these children may benefit from earlier and more intensive intervention strategies. Finally, a multidisciplinary approach is needed to understand the true nature of these conditions and plan intervention accordingly.

Strengths and Limitations

A major strength of this study is the large clinical sample and the systematic mapping of RIs across multiple domains. The use of both descriptive and inferential statistics, including multivariate logistic regression, provides robust evidence for the associations identified. However, some limitations should be noted. The reliance on the caregiver’s memory of the retrospective data collected through the questionnaire and the interview for certain RIs, such as family history and early developmental milestones, may introduce recall bias. It is also possible that parents of children with clinical difficulties recall early challenges differently than parents of typically developing children, which may have influenced the reports. As early speech and motor difficulties are part of the clinical presentation of children with MSDs, using such variables to predict clinical vs. control status may introduce some conceptual overlap. This may contribute to higher predictive values and should be considered when interpreting the findings. Another limitation is that detailed early language measures (e.g., CDI vocabulary scores) were not collected, and these can be sensitive early indicators of later language and speech outcomes. Such data were not available for this cohort. Finally, the wide CIs for some odds ratios suggest limited precision, and future studies should aim to replicate these findings in larger and more diverse samples.

Future Directions

Further research is needed to explore the long-term impact of identified RIs on speech and language outcomes. Longitudinal studies could clarify which indicators are most predictive of persistent difficulties, and which may resolve with development. There is also a need to examine how oral motor development interacts with other neurodevelopmental domains such as attention, sensory processing, and executive function. Future studies may benefit from integrating structured RI screening into assessment protocols to improve early identification and support more individualized intervention planning.

Conclusion

This study highlights the complex interplay of medical, developmental, oral behaviour, and hereditary RIs in children with atypical oral motor and speech sound development. These indicators tended to co-occur more often in children with multiple diagnoses, which aligns with the broader developmental profiles typically observed in this subgroup. One important observation is the presence of persistent mouth-related behaviours, which may reflect delayed oral motor development and atypical sensorimotor integration. These findings may support clinicians in identifying children who could benefit from early and tailored intervention.

Acknowledgments

We are grateful to the children and caregivers from the SG approving to take part in this study. We want to thank the SLPs at the division of speech language pathology who helped collect the data. A special and warm thank you to Isabell Hammarlund.

Statement of Ethics

The Swedish Ethical Review Authority reviewed and approved the study protocol (Dnr: 2020-03398; 2022-00861; 2023-029-02). Written informed consent was obtained from the caregivers to participate with their child in this study.

Conflict of Interest Statement

The authors report no competing interests to declare. Hayo Terband was a member of the journal’s Editorial Board at the time of submission.

Funding Sources

This study was supported by grants resaved from Samariten Foundation, Promobilia Foundation, Aina Börjeson Foundation for Speech Language pathology, The Sven Jerring Foundation and from the group of development and research, department of Neurology, Danderyd Hospital. The funders had no role in the design, data collection, data analysis, and reporting of this study.

Author Contributions

H.B. had primary responsibility for assessments of the patients, outcome assessment, data analysis and writing the manuscript. H.T. supervised and contributed to the writing of the manuscript. F.J. contributed to supervision regarding statistics and to the data analyses. G.T. supervised and contributed to the writing of the manuscript. S.T. supervised and contributed to the writing of the manuscript.

Funding Statement

This study was supported by grants resaved from Samariten Foundation, Promobilia Foundation, Aina Börjeson Foundation for Speech Language pathology, The Sven Jerring Foundation and from the group of development and research, department of Neurology, Danderyd Hospital. The funders had no role in the design, data collection, data analysis, and reporting of this study.

Data Availability Statement

The participants of this study did not provide written consent for their data to be shared publicly; therefore, so due to the sensitive nature of the research supporting data is not available. Further enquiries can be directed to the corresponding author.