Examining factors affecting parental satisfaction with speech therapy in children with Phelan-McDermid Syndrome

Abstract

Objectives: This paper explores the relationship between speech therapy intensity and parent satisfaction with speech therapy (ST) in children with Phelan-McDermid Syndrome (P-MS), a rare genetic disorder.

Methods: ST intensity (ST Dose [minutes per session]) × (ST Dose Frequency) × (ST Length [years]) and parent satisfaction (modified PSQ-18) with ST were measured by online questionnaire. Non-parametric correlation, partial correlation, and linear regression calculations were performed.

Results: Significant correlations between ST Dose and parent satisfaction were observed in the subscales of Time Spent with ST (r = .36, p < .05) and Accessibility and Convenience (r = .40, p < .05) in children with P-MS controlling for child age. ST Dose was also a significant independent predictor of parent satisfaction with ST in specific subscales.

Conclusion: Significant positive correlation and linear regression results indicate increases in ST Dose (minutes per session) represent a mechanism for increasing parent satisfaction with ST in children with P-MS.

Phelan-McDermid Syndrome (P-MS), also called 22Q13 deletion syndrome, is a rare genetic disorder with many phenotypic traits resulting in severe learning disabilities (Kolevzon et al. 2014). Due to a terminal genetic deletion on the 22nd chromosome, P-MS results in decreased expression of the SHANK-3 gene that codes for scaffolding proteins in the post-synaptic density in the brain (Phelan and McDermid 2011). The reduced neurological function causes delayed nerve conduction, resulting in impaired memory and learning disabilities (Soorya et al. 2013). Children with P-MS typically have severe intellectual disability (ID), autism spectrum disorder (ASD), and profound learning and speech impairment (Kolevzon et al. 2014). The profound limitation of expressive speech is one of the central components to P-MS that affects intellectual growth and social development (Phelan and McDermid 2011; Soorya et al. 2013). Due in part to advances in clinical genetic diagnostic testing, including the availability of chromosomal microarray and fluorescence in situ hybridization (FISH) (Kolevzon et al. 2014), the number of P-MS diagnoses is rising (Shaw et al. 2011). Estimates of P-MS prevalence in the general population are inexact and underrepresented; however, at least 1,800 diagnosed cases of P-MS have been recorded worldwide (Phelan-McDermid Syndrome Foundation 2018). Though investigative therapies for the treatment of the underlying neurodevelopmental cause of P-MS are ongoing, there is no pharmacological treatment specific to P-MS (Rossi 2018; Schmidt et al. 2008). As a result, therapeutic treatments including speech, occupational, physical, and other therapies are recommended for adjunctive and supportive management of P-MS (Phelan and McDermid 2011).

Speech therapists have a unique challenge assisting children with intellectual disabilities learn expressive language. Research performed on individuals with intellectual disabilities supports that speech therapy (ST) is more effective when targeted ST interventions coincide with individuals’ unique speech patterns and intellectual needs (Coppens-Hofman et al. 2013). However, the varied presentation of ID can result in clinical speech deficits unique to each individual, making targeting of therapy difficult. As a result, even experienced speech therapists can disagree on the most effective communication techniques and ST interventions for individual children within this population (Goldbart et al. 2014; Koski and Launonen 2012). In addition, learning disabilities resulting from global developmental delays can make the treatment of expressive speech delays in P-MS challenging.

Speech development in children with P-MS often follows a pattern of early development of babbling and limited vocabulary until age 3, after which speech regression and loss of expressive speech often occur (Phelan and McDermid 2011). Expressive speech impairment can be troubling for children and parents/caretakers, because the inability to express wants and desires can lead to child frustration and social isolation. However, standard therapeutic approaches used for expressive speech delay in the general population may be ineffective or inappropriate in children with P-MS, presenting unique barriers to skill acquisition and retention in individuals with P-MS. Speech interventions to treat speech delays in children with ID fall into several major categories. These include intensive interaction, objects of reference, music therapy, micro-switch technology, augmentative communication therapies, and symbolic approaches (Goldbart et al. 2014). Comorbid conditions common to P-MS including fine motor, gross motor, and intellectual disability (Phelan and McDermid 2011) may yield typical approaches to ST ineffective in this population compared to children with conditions associated with singular deficits in functioning. For example, individuals with solitary intellectual disability often use manual sign systems to foster improved acquisition and communication of language (Vandereet et al. 2011). However, hypotonia and other gross/fine motor deficits are common in P-MS (Kolevzon et al. 2014) and may cause difficulties with the administration of manual ST techniques such as sign language. Despite these challenges, Phelan and McDermid (2011) suggest that assistive screen-based speech technologies, such as AAC devices, may be useful in assisting children with P-MS to better communicate. However, empiric research investigating the use of assistive ST interventions in children with P-MS that are commonly used in populations with ID is lacking. This remains an area in need of further investigation in children with P-MS. Taking into account the concomitance of the above speech challenges, learning and attention deficits, and the severity of disability common to P-MS, current recommendations suggest that therapy sessions be intensive, short duration, and frequent, in order to manage P-MS-specific learning impairments (Kolevzon et al. 2014). However, these recommendations are based on observational studies and are not the result of empiric comparative research. Further research is needed to delineate efficacy and specify the preferred intensity, duration, and frequency of therapy regimens for children with P-MS.

There are multiple components to the quantification of ST services, including speech therapist motivation, speech therapist experience, intervention techniques, attitude of the client, and ST intensity (To et al. 2012). Of these, the intensity of ST is a corollary indicator of therapy quality. Research on ST intensity has suggested that increased intensity of ST improves outcomes in childhood apraxia (Namasivayam et al. 2015) and childhood speech sound disorder (Allen 2013). Intense ST is recommended in the management of P-MS learning disabilities (Kolevzon et al. 2014), though no known studies have directly examined the outcomes of increased intensity. ST intensity is a function of ST frequency and duration of ST session (To et al. 2012).

Parent perceptions of ST services in children with intellectual disability may influence the effectiveness of the ST delivered, although direct relationships between these variables are lacking (Carroll 2010). As a result of the inherent challenges children with ID experience with expressive speech, parents of children with ID can serve as proxies to indicate child satisfaction with therapy services. In this manner, parent satisfaction represents an important measure of quality of care for children with ID (Bairati et al. 2011). Parent satisfaction measures the degree to which parents view their expectations of children’s therapy needs are being met and can therefore provide a measure of quality care that can be useful to help tailor therapy services independent of therapist feedback (Bairati et al. 2011). As a result, parent satisfaction with therapy services has been linked with both therapy adherence and overall health outcomes (Bairati et al. 2011; Marshall and Hays 1994).

Research by Mazer et al. (2017) suggests that higher rates of parent satisfaction with rehabilitation services in children with primary language impairment are linked with parental involvement and collaboration with therapy. This relationship indicates that parent satisfaction may augment the benefits of rehabilitation services, including speech language pathology (Mazer et al. 2017). Furthermore, parent satisfaction is a critical indicator of quality of care related to pediatric therapy services by acting as a measure of the structure, processes, and outcomes of care (Schreiber et al. 2011). Identifying the salient components of parent satisfaction with therapy services can provide useful guidance to create the most supportive therapy environment for children with disabilities (Schreiber et al. 2011). Research in children with Fragile X syndrome, another genetic developmental disorder with similarities to P-MS, indicates decreased parent satisfaction with therapy in the areas of speech, communication, and language, with corresponding parental preference to increase the amount of speech-language therapy for their children (Faundes et al. 2018). This highlights the effect of ST intensity as a potential major modifying factor in parent satisfaction with ST services in children with P-MS.

A more thorough examination of the relationship between ST regimens and parent satisfaction may establish the correlational and predictive relationships between components of ST intensity that affect therapy adherence and overall speech outcomes. Identifying relationships between ST and higher rates of parent satisfaction may identify factors affecting parent satisfaction regarding the provision of ST services to children with P-MS. This research explores the relationship between ST intensity and parent satisfaction with ST in 40 children with P-MS.

Methods

Research design

A quantitative, cross-sectional survey methodology was utilized for this study. As a correlational study, the aim was to describe the relationship between the independent variable of the ST intensity and the dependent variable of parent satisfaction with ST services in children with P-MS. Due to logistical challenges with the long-term direct assessment of speech skills and therapy intensity in children with P-MS in this study, parental report of speech skills and therapy intensity was collected using online questionnaire as the preferred method of data collection.

Study participants

A total of 46 parents/caregivers with children age from birth to 21 years diagnosed with P-MS returned online questionnaires on behalf of their children, with 40 responses suitable for data analysis. Questionnaires containing unusable or out-of-range responses were removed from data analysis.

Inclusion criteria

Pediatric guidelines define the age limit for pediatric care in individuals with disabilities from birth to 21 years old (Reider-Demer et al. 2008), so individuals with P-MS up to age 21 were included. Inclusion criteria were verified based on parent/caregiver responses to initial demographic questions on a questionnaire assessing the type and intensity ST services delivered. All study data were collected from the parents/caregivers of children with P-MS, on behalf of their children, who were aged from birth to 21 years old, had received a previous diagnosis of P-MS by some form of genetic testing, and who participated in ST as a method of managing expressive speech delay.

Exclusion criteria

Exclusion criteria were verified based on parent/caregiver responses to initial demographic questions on the questionnaire assessing the type and intensity ST services delivered. Parents/caregivers were excluded from study participation if their children had expressive speech delays without coexisting P-MS, were older than 21-years-old, or had never received ST services.

Sampling methodology

This study was a nonrandomized, purposive sampling of parents/caregivers of children with P-MS. Participants were recruited for the study using online social media support groups and through snowball sampling.

Institutional review board

This study was approved by the Institutional Review Board (IRB) at A.T. Still University.

Data collection

Online questionnaires were distributed to parents of children with P-MS using SurveyMonkey. The data were compiled and downloaded into Microsoft Excel. After the data were cleaned, coded, and checked for accuracy, they were imported into IBM SPSS Statistics Version 24.0. Descriptive statistics were used to analyze nominal and ordinal variables for frequencies, averages, standard deviation, and percentages.

Scores for measures of parent satisfaction with ST services were measured using the Patient Satisfaction Questionnaire Short Form (PSQ-18), which is an independently validated tool used for the measurement of patient satisfaction with medical care (Marshall and Hays 1994). The PSQ-18 is based on the long-form 50 item PSQ-III that utilizes questions with five-point Likert-style responses to measure patient satisfaction with medical care (Marshall and Hays 1994). Parent satisfaction was measured using 17 of 18 items from the PSQ-18, an externally validated, internally reliable short-form questionnaire. One question on the PSQ-18 was removed due to awkward rephrasing in adaptation, redundancy in measuring the subscale of general satisfaction, and to facilitate shortening in questionnaire length (See Table 1). The PSQ-18 was adapted for this study to rate total parent satisfaction with ST services (Table 1). Using scores from coded values ranging from 0 (very dissatisfied) to 5 (very satisfied), PSQ-18 scores were calculated by adding the coded sums for responses to each item. Items on the PSQ-18 are worded with alternating agreement with satisfaction (Marshall and Hays 1994). This tool was adapted for rating parent satisfaction with ST services. The standard phrasing in the PSQ-18 was replaced, with all references to ‘medical care’ in the PSQ-18 removed and replaced with ‘ST services,’ and the term ‘physician’ with ‘speech therapist,’ in order to more directly assess satisfaction with ST services. While this is a minor adaptation of the PSQ-18 for use specific to ST services, these novel and uninvestigated changes limit the applicability of the original psychometric properties of this tool.

Table 1.

Statistics for PSQ-18 questionnaire adapted to assess parent satisfaction with speech therapy services

Subscale and Item	Frequency Likert Response					Median Response
	Strongly Agree	Agree	Uncertain	Disagree	Strongly Disagree
General Satisfaction
3. The speech therapy care my child has been receiving is just about perfect.	7.5% (3/40)	37.5% (15/40)	25% (10/40)	25% (10/40)	5% (2/40)	2.0
Technical Quality
2. I think my child’s speech therapist has everything needed to provide complete speech therapy services.	12.5% (5/40)	22.5% (9/40)	30% (12/40)	27.5% (11/40)	7.5% (3/40)	3.0
4. Sometimes speech therapists make me wonder if their speech therapy services are correct.	7.5% (3/40)	50% (20/40)	30% (12/40)	12.5% (5/40)	0% (0/40)	2.0
6. When my child goes for speech therapy services, they are careful to check everything when providing speech therapy.	2.6% (1/39)	22.5% (9/40)	50% (20/40)	22.5% (9/40)	0% (0/40)	3.0
14. I have some doubts about the ability of the speech therapists who treat my child.	5% (2/40)	27.5% (11/40)	20% (8/40)	15% (15/40)	10% (4/40)	3.0
Interpersonal Manner
10. Speech therapists act too businesslike and impersonal towards my child.	0% (0/40)	7.5% (3/40)	10.0% (4/40)	50% (20/40)	32.5% (13/40)	4.0
11. My child’s speech therapist’s treat my child in a very friendly and courteous manner.	0% (0/40)	5% (2/40)	5% (2/40)	45% (18/40)	45% (18/40)	4.0
Communication
2. Speech therapists are good about explaining the reason for specific types speech therapy.	2.5% (1/40)	27.5% (11/40)	17.5% (7/40)	47.5% (19/40)	5% (2/40)	3.0
13. Speech therapists sometimes ignore what I tell them.	0% (0/40)	27.5% (11/40)	22.5% (9/40)	40% (16/40)	10% (4/40)	3.0
Financial Aspects
5. I feel confident that my child can get the speech therapy services that they need without being set back financially.	30% (12/40)	27.5% (11/40)	15% (6/40)	20% (8/40)	7.5% (3/40)	2.0
7. I have to pay for more of my child’s speech therapy than I can afford.	12.8% (5/39)	33.3% (13/39)	5.1% (2/39)	30.8% (12/39)	17.9% (7/39)	2.0
Time Spent with Speech Therapist
12. Those who provide my child’s speech therapy services sometimes hurry too much when they provide speech therapy to my child.	2.5% (1/40)	20% (8/40)	27.5% (11/40)	40% (16/40)	10% (4/40)	3.0
15. Speech therapists usually spend plenty of time with my child.	5.1% (2/39)	33.3% (13/39)	17.9% (7/39)	25.6% (10/39)	17.9% (7/39)	3.0
Accessibility and Convenience
8. I have easy access to the speech therapists my child needs.	25% (10/40)	30% (12/40)	5% (2/40)	32.5% (13/40)	7.5% (3/40)	2.0
9. Where my child gets speech therapy, people have to wait to long for speech therapy services.	15.3% (6/39)	25.6% (10/39)	20.5% (8/39)	33.3% (13/39)	5.12% (2/39)	2.5
16. I find it hard to get an appointment for the speech therapy services for my child right away.	12.8% (5/39)	30.8% (12/39)	12.8% (5/39)	28.2% (11/39)	15.4% (6/39)	2.0
17. My child is able to get speech therapy services whenever they need it.	17.5% (7/40)	35% (14/40)	17.5% (7/40)	22.5% (9/40)	7.5% (3/40)	2.0

Note. An additional question that appears in the original PSQ-18 was removed from the PSQ-18 for this study. This question typically appears as Question #17 and reads, ‘I am dissatisfied with some things about the medical care that I receive.’

Appropriate items on the questionnaire were reverse coded for parity, so higher responses reflected higher satisfaction with ST services for all items. The PSQ-18 was used to assess seven subscales of satisfaction with ST services that included general satisfaction (Item 3), technical quality (Items 2, 4, 6, and 14), interpersonal manner (Items 10 and 11), communication (Items 1 and 13), financial aspects (Items 5 and 7), time spent with speech therapist (Items 12 and 15), and accessibility and convenience (Items 8, 9, 16, and 18).

Data analysis

Summary scores for measures of total parent satisfaction and subscales of parent satisfaction with ST services were determined by calculating sums and average scores for these categories. Summative values for cumulative intervention intensity were also calculated. Total ST intervention duration, originally measured in ranges (<12 months, 1–2 years, 3–4 years, 5–6 years, 7–8 years, 9–10 years, >10 years), were converted into interval estimates of total intervention duration by multiplying ordinal categories by a relative numeric multiplier (0.5, 1.5, 3.5, 5.5, 7.5, 9.5, 10.5). For example, 0.5 = 12 months, 1.5 = 1–2 years, 3.5 = 3–4 years. Total ST intervention duration, referred to in this study as ST length in years, was measured in years. The purpose of creating an interval estimate of total ST intervention duration was to enable the calculation of ST cumulative intervention intensity, which is typically calculated as the product of dose × dose frequency × total intervention duration (Warren et al. 2007). Performing this calculation using a converted interval estimate of intervention duration, parents were able to provide an estimated range of the length of total therapy duration that was still usable as an exact calculation of total ST length and ST cumulative intensity. In addition, due to the use of an online questionnaire, minor changes were required to calculate ST cumulative intervention intensity. Specifically, ST cumulative intervention intensity, referred to as ST cumulative intensity, was calculated using parent estimates of therapy instead of direct values gathered through retrospective chart review or direct ST observation. Therefore, in this study ST cumulative intensity is calculated as the parent estimated sum of (ST Dose [minutes per session]) x (ST Dose Frequency) x (ST Length [years]).

Inferential statistics were used to analyze all interval and ratio variables. Shapiro–Wilk tests indicated that not all continuous variables were normally distributed (p>.05), and participants were enrolled through nonrandom, purposive sampling. Therefore, more conservative nonparametric statistics were used for data analysis. Following trends in statistical analysis of Likert scale responses, non-normally distributed data were described using frequency distributions (Sullivan and Artino 2013). Summary scores and/or averages were calculated for appropriate variables including participant age, total parent satisfaction, subsets of ST satisfaction, ST dose frequency, ST dose time per session, ST total intervention duration, and ST cumulative intensity. Two-tailed Spearman’s rank correlation coefficients were calculated between each continuous variable. Correlations with value of p<.05 using the Spearman’s rank correlation coefficient were considered statistically significant. Partial correlation testing was then calculated for all variables that showed significant correlation coefficients to control for the effect of age. Correlation coefficients were also calculated for subgroups by age, to further investigate and describe age-based relationships concerning parent satisfaction with ST services. Partial correlations with a value of p<.05 were considered statistically significant. Eta and Eta² values were calculated between nominal variables of parent-rated objective/subjective speech skills with parent satisfaction to identify trends concerning parent satisfaction with ST services and speech skills. Finally, bivariate linear regression was used to analyze whether ST cumulative intensity significantly predicted total parent satisfaction. Multiple linear regression analysis was used to analyze whether ST intensity subsets were significant independent predictors of total parent satisfaction, and stepwise forward regression testing was conducted to evaluate whether variables displaying statistically significant prediction of parent satisfaction were collective or independent predictors of parent satisfaction. Partial correlations with a value of p<.05 were considered statistically significant. Regression was considered statistically significant with a value of p<.05.

Results

Participant characteristics

A total of 46 parents of children with P-MS responded to the online questionnaire for this study. All children (n= 46, 100%) in the study had received a formal diagnosis of P-MS, by either genetic microarray or FISH (fluorescence in situ hybridization) testing. Questionnaire responses described children from heterogeneous backgrounds, possessing variable levels of baseline speech, who participated in differing types, intensities, and settings of ST services (See Table 1). Educational experiences and setting for ST also varied, with children receiving ST through a blend of home, school, and other clinical settings (Table 2). A total of four subjects were lost due to attrition due to partial completion of the questionnaire and were excluded from analysis of this study. Two additional subjects were excluded due to child age exceeding the inclusion criteria of younger than age 21. In total, 40 completed questionnaires were used for data analysis. Some respondents also intermittently left random responses blank; therefore when accounting for incomplete answers to individual questions, most items received between 36 and 40 responses.

Table 2.

Characteristics of participants

	Children with P-MS (n= 39)
Female	19/39 (48.7)
Mean Age [Minimum-Maximum] (SD)	6.92 [1–17] (3.90)
Race/Ethnicity
Caucasian	33/39 (84.6)
African American	1/39 (2.6)
Asian/Pacific Islander	1/39 (2.6)
Multiple Ethnicities	4/39 (10.3)
Years of Participation
<12 months	5/39 (12.8)
1–2 years	5/39 (12.8)
3–4 years	14/39 (35.9)
5–6 years	6/39 (15.4)
7–8 years	4/39 (10.3)
9–10 years	0/39 (0)
>10 years	5/39 (12.8)
Subjective Speech
Very Poor	19/39 (48.7)
Poor	12/39 (30.8)
Fair	6/39 (15.4)
Good	1/39 (2.6)
Very Good	1/39 (2.6)
Objective Speech
Nonverbal, noninteractive with environment	3/39 (7.7)
Nonverbal, interactive, no ACD	12/39 (30.8)
Nonverbal, interactive, ACD	16/39 (41.0)
Verbal, blended speech/ACD	5/39 (12.8)
Verbal, no ACD	3/39 (7.7)

Note. All values are reported as n (%) unless otherwise noted.

Children of survey respondents were 51.3% (n= 20) male and 48.7% (n= 19) female. Average child age of survey respondents was 7 years (X= 6.92, SD= 3.90), with age ranging from 1 to 17 years. The majority of children of survey respondents were Caucasian (84.6%), although other race/ethnicities were represented (10.3% multiple ethnicities, 2.6% Asian/Pacific Islander, and 2.6% Black/African American).

The baseline level of speech varied across children of survey respondents. For this study, subjective speech refers to parents’ subjective assessment of children’s speech abilities from very poor to very good. Most respondents rated children’s subjective speech level as very poor (48.7%), poor (30.8%), or fair (15.4%; see Figure 1, Table 1). Only two parents rated subjective speech above fair, with separate parents rating speech as good (2.6%, n= 1) or very good (2.6%, n= 1). Objective speech in this study referred to parents’ assessment of children’s speech abilities. Specifically parents were asked to identify whether their children are verbal, nonverbal, or use an augmentative communicative device (ACD) as the preferred method of communication (See Figure 2), and parents’ assessment of children’s objective speech levels shadowed these findings, supporting known trends of increased incidence of significant speech impairments in children with P-MS (Soorya et al. 2013). In total, 79.5% of children of survey respondents were nonverbal, with 41.0% of nonverbal children using an ACD. Of all children of survey respondents, 17.4% of children were verbal, with 10.9% using a blend of nonverbal and verbal speech. A small subset comprised of 6.5% of children of survey respondents (3/39) was verbal without the assistance of an ACD (Table 2). These findings correspond to the existing summaries of speech abilities in children with P-MS. For example, in the research sample described by Soorya et al. (2013), a minority of children with P-MS achieved spontaneous word usage and none developed verbal speech skills at the level of spontaneous phrases. Due to the high prevalence of severe speech impairment common to P-MS, no data were collected pertaining to the verbal speech abilities of children in this study. Therefore, detailed sub-analyses of verbal speech abilities of the children in this study were not performed.

Figure 1.

P-MS child subjective speech level as rated by parents

Figure 2.

P-MS child objective speech level as rated by parents.

Types of speech therapy services

Utilization of ST services also varied. Survey respondents indicated that ST services were provided to children in a wide variety of settings, with school (21.1%), home and school (21.1%), and hospital/clinic and school (18.4%) as the most common locations of ST therapy delivery. Individuals who were the primary providers, or those who spent the most time providing ST to children relative to other ST providers, were most commonly a combination of speech therapists/teachers/parents (28.2%), speech therapists without teachers/parents (25.6%), and speech therapists and parents (23.1%). Different ST techniques were used for children of survey respondents, with the picture exchange communication system (PECS; 30.8%), objects of reference (17.9%), and ACDs (10.3%) the most common. Although sign language is a common method of promoting nonverbal communication in children with intellectual disability (Vandereet et al. 2011), sign language was used as the primary method of ST communication by only one child (2.6%, n= 1; see Figure 3). The most common length of ST participation in children of survey respondents was 3–4 years (35.9%), with an almost equal distribution in the length of ST participation in the ranges outside this time period. Average ST intensity, which comprised of ST dose frequency, ST dose (minutes per session), and ST length (years) of ST, was represented with a mean of 2.32 sessions of ST per week (SD = 2.91), 46.11 minutes per session (SD = 35.64), and 4.53 years (SD = 3.16).

Figure 3.

Type of speech therapy technique used by P-MS children.

Satisfaction with speech therapy services

The highest possible PSQ-18 score is a mean score of 5/5 on all 18 items, in all 7 subscales. PSQ-18 data are designed for analyses by individual subscale (Marshall and Hays 1994). Values relating to total or aggregate PSQ-18 score, referred to as cumulative parent satisfaction, are also listed to provide a complete review of the data obtained in this study, but are not intended to represent general parent satisfaction which has a separate PSQ-18 subscale. Total PSQ-18 score was represented with a mean satisfaction score of 3.01 (maximum score of 5), with variable satisfaction with ST represented in scores ranging from 1.88 to 4.65 (SD = 0.68) in total satisfaction with ST services (Table 4). Subscales of satisfaction with ST services also showed variation between respondents. The subscales with the highest average levels of parent satisfaction were ST’s interpersonal manner (4.18/5), communication (3.29/5), and time spent with ST (3.26/5). The subscales with the lowest average level of parent satisfaction were the financial aspects of ST (2.74/5), general satisfaction (2.79/5), and accessibility and convenience (2.81/5; See Table 4).

Table 4.

Satisfaction with speech therapy services

	Range (Min–Max)	Mean (0–5)	Standard Deviation
PSQ-18 Score	1.88–4.65	3.09	1.20
General Satisfaction	1.0–5.0	2.79	1.04
Technical Quality	1.25–4.5	2.87	0.74
Interpersonal Manner	2.0–5.0	4.18	0.74
Communication	2.0–5.0	3.29	0.86
Financial Aspects	1.0–5.0	2.74	1.20
Time Spent With ST	1.0–5.0	3.26	0.99
Accessibility and Convenience	1.25–5.0	2.81	1.03

Speech therapy services and parent satisfaction – correlation

ST dose (time per session) (r=.42, p<.05), ST cumulative intensity (r=.43, p<.01), and ST dose frequency (r= .34, p<.05) and general parent satisfaction were significantly correlated (r=.43, p<.01) (See Table 5). In addition, ST dose (time per session) showed significant correlations with parent satisfaction subscales. ST dose (time per session) was significantly correlated with the parent satisfaction subscales of technical quality (r=.38, p<.05), interpersonal manner (r=.39, p<.05), communication (r=.35, p<.05), time spent with speech therapist (r=.52, p<.01), and accessibility and convenience (r=.40, p<.05). The satisfaction subscale of accessibility and convenience was also significantly correlated with both ST cumulative intensity (r=.44, p<.01), ST dose frequency (r=.41, p<.05), and ST dose (minutes per session) (r=.40, p<.05; see Table 5).

Table 5.

Correlations between PSQ-18 total parent satisfaction, parent satisfaction subsets, speech therapy intensity, and P-MS child age

Variables	General Parent Satisfaction	Technical Quality	Interpersonal Manner	Communication	Financial Aspects	Time Spent with ST	Accessibility & Convenience	P-MS Child Age
ST Cumulative Intensity	.43** (.20)	.16	.07	–.04	.17	.11	.44** (.32)	.51**
ST Dose Frequency	.34* (.27)	.11	.10	.12	.14	.14	.41* (.29)	–.18
ST Dose (minutes per session)	.42* (.32)	.38* (.32)	.39* (.16)	.35* (.31)	–.01	.51** (.36*)	.40* (.40*)	–.08
ST Total Intervention Duration (Years)	.08	.02	–.05	–.19	.20	–.13	.17	.78**

Note. p<.05. * p<.01. **. Partial correlations performed on all statistically significant correlations, displayed as (r) to correct for the effect of child age.

Total ST intervention duration, originally measured in ranges (<12 months, 1–2 years, 3–4 years, 5–6 years, 7–8 years, 9–10 years, >10 years) was converted into interval estimates of total intervention duration by multiplying ordinal categories by a relative numeric multiplier (0.5, 1.5, 3.5, 5.5, 7.5, 9.5, 10.5). After this conversion, ST cumulative intervention intensity, referred to as ST Cumulative Intensity, was calculated as the sum of (ST Dose [minutes per session])×(ST Dose Frequency)×(ST Length [years]).

ST child age was the only other independent variable identified that showed significant correlations with ST cumulative intensity (r=.51, p<.001) and ST total intervention duration (r=.78, p<.001). However, child age was nonsignificantly negatively correlated with total parent satisfaction, ST dose frequency, and ST dose (time per session). These age-specific results indicated the need for further statistical analysis to identify the effect of P-MS child age on significant correlations. As a result, partial correlation calculations were performed. Partial correlation showed a consistent negative effect of P-MS child age on all significant correlation coefficients, also decreasing the statistical significance of all correlational relationships (Table 5). When controlling for the effect of P-MS child age on parent satisfaction with ST services and ST dose (time per session), the only independent variables that retained statistically significant correlation coefficients were cumulative parent satisfaction (r=.42, p<.05), time with ST (r=.36, p<.05), and access and convenience (r=.41, p<.05; see Table 5).

Age range appears to contribute with additional significant subscales of parent satisfaction with ST services. In regard to parent satisfaction with ST services, these effects in the P-MS population appear to be inconsistent, with absent categorical carryover of significant correlations from younger to older age group populations. For example, parent satisfaction with ST in the subscale of communication was significantly negatively correlated (r=–.68, p=.031) with ST total intervention duration in P-MS children aged 1–4. These findings may be interpreted as a reflection of difficulties in ST communication with parents of children in this younger age cohort. However, this significant correlation was absent in older age ranges and in overall grouped data. In a similar manner, parent satisfaction with ST services in parents with children aged 5–8 showed significant positive correlations between the satisfaction domains of general parent satisfaction (r=.64, p=.004) and interpersonal manner (r=.58, p=.014) with ST total intervention duration. This indicates increased parent satisfaction with ST services when total intervention duration (years of therapy) alone was increased. However, this effect was absent in parents of children younger than 5 years and older than 8 years of age.

Parents of children aged 9–12 were more highly influenced by ST cumulative intensity, with positive significant correlations in the domains of general parent satisfaction (r=.93, p=.006), technical quality (r=.82, p=.046), interpersonal manner (r=.94, p=.006), and time spent with ST (r=.92, p=.010). The strength of these correlations shows the relative importance of ST cumulative intensity to parents of children aged 9–12, which is absent in parents of children younger than 9 and older than 12 years of age.

Finally, parent satisfaction of parents with children aged 13–17 were influenced by ST dose frequency in the domain of accessibility and convenience (r=.88, p=.047), showing that increased frequency of ST favorably influences parent perception of accessibility and convenience with ST services in this older study subpopulation. In addition, ST total intervention duration was highly positively correlated with parent satisfaction with interpersonal manner (r = 1.00, p<.001), indicating that parents of children receiving the longest total ST intervention duration were highly satisfied with the interpersonal manner of STs. Again, these findings were novel to parents of children within this older age cohort.

Subjective and objective speech levels were strongly positively correlated (r=.61, p= <.001), indicating that parents consistently rate the level of speech in children with P-MS based on both subjective views of speech abilities and objective level of speech skills. Parent determined objective child speech skills showed no significant positive correlations with parent satisfaction subscales. Significant negative correlations were present between objective speech and parent satisfaction with STs interpersonal manner (r=–.41, p=.1) and time spent with ST (r=–.34, p=.04)(See Table 6 ). These findings indicate that as objective speech skills advance, parent satisfaction decreases related to STs interpersonal manner and time spent with STs.

Table 6.

Correlation values between PSQ-18 Parent Satisfaction Subsets and Subjective and Objective Speech Level in children of all ages

Variables	General Parent Satisfaction	Technical Quality	Interpersonal Manner	Communication	Financial Aspects	Time Spent with ST	Accessibility & Convenience
Objective Speech Level	.06	.03	–.41* (.010)	–.28	.14	–.34* (.04)	.21
Subjective Speech Level	.25	.18	–.12	.04	.33* (.045)	–.08	.37* (.02)

Note. p<.05. * p<.01. **. All statistically significant correlations, displayed as (r)

Total ST intervention duration, originally measured in ranges (<12 months, 1–2 years, 3–4 years, 5–6 years, 7–8 years, 9–10 years, >10 years), was converted into interval estimates of total intervention duration by multiplying ordinal categories by a relative numeric multiplier (0.5, 1.5, 3.5, 5.5, 7.5, 9.5, 10.5). After this conversion, ST cumulative intervention intensity, referred to as ST Cumulative Intensity, was calculated as the sum of (ST Dose [minutes per session])×(ST Dose Frequency)×(ST Length [years]).

By contrast, parent determined subjective child speech skills were positively correlated with parent satisfaction subscales in the domains of financial aspects (r=.33, p=.045) and accessibility and convenience (r=.37, p=.02)(See Table 6). There were no significant negative correlations between subjective speech and parent satisfaction. These findings suggest increased parent satisfaction in the domains of financial aspects of ST and accessibility and convenience when parents subjectively felt their children’s speech skills were stronger.

Setting, provider, and technique of speech therapy services and parent satisfaction – eta/eta²

In regard to parent satisfaction with ST services and ST setting, ST provider, and ST technique, eta² values were calculated to determine the variance accounted for by these nominal categorical independent variables on the dependent variable of parent satisfaction with ST services (See Figure 4). These analyses display relative trends between ST categories, showing small differences in the effects of ST setting, ST provider, and ST technique on subscales of parent satisfaction with ST services across all ages. Of these analyses, small trends can be identified, including a larger relative amount of variance in general parent satisfaction accounted for by ST provider (38%, eta²=.38) than by ST setting (26%, eta²=.26) and ST technique (18%, eta²=.18), with similar relative trends displaying a smaller magnitude of effect of ST provider on parent satisfaction in the subscales of accessibility and convenience (See Figure 4). ST setting accounted for the largest relative amount of satisfaction with STs interpersonal manner (30%, eta²=.30), compared to ST provider (14%, eta²=.14) and ST technique (24%, eta²=.24) (See Figure 4). In addition, ST technique accounted for the largest relative amount of variance in the parent satisfaction subscale of technical quality (23%, eta²=23%) (See Figure 4). However, it should be noted that the generalizability of these findings to the larger P-MS population is difficult to interpret, given the small magnitude of difference between eta/eta² values among these independent variables and the inability to perform further testing to assess statistical significance of these findings.

Figure 4.

Variance (eta²) of parent satisfaction accounted for by ST setting, ST provider, and ST technique.

Speech therapy services and parent satisfaction – regression

Bivariate linear regression was used to analyze if ST cumulative intensity significantly predicted total parent satisfaction. Results of the regression indicated that ST cumulative intensity was a nonsignificant independent predictor of parent satisfaction, explaining 26% of the variance of parent satisfaction (R² = .26, F[1,35] = 2.39, p >.05). Multiple linear regression analysis was used to analyze if ST intensity subsets were significant independent predictors of total parent satisfaction. The results of the multiple regression indicated that ST dose (time per session), ST dose frequency, and ST total intervention duration explained 18% of the variance (R² = .18, F[3,35] = 2.36, p =.09) and were together nonsignificant predictors of total parent satisfaction. As independent predictors of parent satisfaction, ST dose (time per session) significantly predicted total parent satisfaction (β = .01, Beta = .54, p < .05), while ST dose frequency (β = -.04, Beta = -.17, p = .80) and ST total intervention duration (β = -.03, Beta = -.70, p = .49) did not independently predict total parent satisfaction. Stepwise forward regression testing was conducted to evaluate whether ST dose frequency, ST dose (time per session), and ST total intervention duration were all necessary to predict total parent satisfaction. At Step 1 of the analysis, ST dose (time per session) entered into the equation and was a significant independent predictor of parent satisfaction (R² = .16, F[1,35] = 6.55, p <.05), with a multiple correlation coefficient of .40, accounting for 16.2% of variance of total parent satisfaction (See Table 7). Stepwise regression further excluded total intervention duration (t = -.46, p >.05) and ST frequency (t = -.53, p >.05) from Step 2 of the analysis as predictors of variance.

Table 7.

Multiple regression between ST Dose, ST Intensity, and ST Duration and Parent Satisfaction controlling for the effect of P-MS child age

Dependent Variable	Independent Variables	R	R2	p	B	β	t
Cumulative Parent Satisfaction		.40	.16*
	ST Dose			.01*	.01	.40	2.59
	P-MS Child Age	Excluded from Step 2 of regression analysis
Technical Quality		.35	.12
	ST Dose			.05*	.01	.32	2.00
	P-MS Child Age	Excluded from Step 2 of regression analysis
Interpersonal Manner		.24	.06
	ST Dose			.37	.00	.15	.91
	P-MS Child Age	Excluded from Step 2 of regression analysis
Communication		.38	.14*
	ST Dose	Excluded from Step 2 of regression analysis
	P-MS Child Age			.02*	–.08	–.38	–2.43
Time Spent with ST		.34	.11*
	ST Dose			.04*	.01	.36	2.28
	P-MS Child Age	Excluded from Step 2 of regression analysis
Accessibility and Convenience		.40	.16*
	ST Dose			.01*	.01	.40	2.58
	P-MS Child Age	Excluded from Step 2 of regression analysis
Accessibility and Convenience		.32	.11
	ST Cumulative Intensity	Excluded from Step 1 of regression analysis
	P-MS Child Age	Excluded from Step 2 of regression analysis
Accessibility and Convenience		.29	.09
	ST Dose Frequency	Excluded from Step 1 of regression analysis
	P-MS Child Age	Excluded from Step 2 of regression analysis

Note. p<.05. * p<.01. *

Additional stepwise regression was performed on all variables displaying significant correlations, to control for the effect of P-MS child age on ST dose (time per session) when predicting cumulative parent satisfaction. These tests were consistent with the partial correlation results, displaying that for significantly correlated variables, P-MS child age had a limited effect on independent covariables and lacked criteria for inclusion in the stepwise regression. The only exception was in the satisfaction subset of communication, where P-MS child age was a significant independent predictor of parent satisfaction (R² = .14, F[1,35] = 5.91, p <.05) and ST dose (time per session; t = 1.93, p >.05) was excluded in Step 2 of the analysis as a predictor of variance (See Table 7). P-MS child age was found to be a significant independent predictor of parent satisfaction with ST services for the domain of communication. In total, when using stepwise regression to control for the effect of P-MS child age, ST dose (time per session) was a significant independent predictor of parent satisfaction with ST services for the domains of cumulative parent satisfaction, technical quality, time spent with ST, and accessibility and convenience.

Discussion

The purpose of this study was to investigate the correlational relationship between measures of ST intensity and parent satisfaction with ST services in children with P-MS. Children in this study possess significantly disordered speech (See Figure 1, Table 2, Figure 3), consistent with prior research involving children with P-MS displaying significant speech and learning disabilities (Kolevzon et al. 2014; Phelan & McDermid, 2011). ST for these children involves differing types of therapy providers, settings, and therapy intensities, without identifiable trends displaying obvious preferred therapy regimens (See Tables 2 and 3). Given these divergent ST regimens, parent satisfaction with ST services also varies widely (See Table 1). Despite the significant variation in ST treatment plans, this research was able to identify significant positive correlations linking ST intensity and parent satisfaction with ST services in children with P-MS (Table 5).

Correlations were identified between several submeasures of ST intensity and subscales of parent satisfaction. However, child age was found to interact significantly as a covariable in many of these correlations (See Table 5). ST intensity was calculated in this study as ST dose (minutes per session)×ST frequency × ST length (years). ST length, measured in years, yields increases in ST intensity as P-MS child age increases. For this reason, partial correlation calculations were used to control for P-MS child age as a contributing covariable, displaying a clearer picture of the correlational relationship between these variables. Specifically, significant correlations exist between ST dose and parent satisfaction in the domains of time spent with ST and accessibility and convenience, independent of P-MS child age (See Table 5).

Parent satisfaction with ST services in the subscale of time spent with ST is a similar construct to ST dose (minutes per session). Therefore, significant positive correlational and predictive relationships between parent satisfaction with time spent with ST and ST dose (minutes per session) were predicted. However, positive significant correlations and linear regression calculations between ST dose (minutes per session) and parent satisfaction in the subscales of accessibility and convenience are the novel findings. These results support recommendations for increased intensity of therapy in the treatment of learning disabilities in children with P-MS (Kolevzon et al. 2014), and prior research performed on children with developmental disabilities, which highlight the benefits of increases in ST intensity in ST regimens (Allen 2013; Namasivayam et al. 2015).

The positive directionality of the correlation between parent satisfaction with ST services, accessibility and convenience, and ST dose suggests that a higher dose of therapy, delivered in a convenient manner, is preferred by parents. Increases in accessibility and convenience may be directly positively linked to increased doses of ST, as barriers in accessibility/convenience would lead to decreased ST dose. This aligns with previous research showing parental preference to increase the amount of speech language therapy for children with fragile X Syndrome due to disproportionate challenges in this population related to baseline poor speech, language, and communication in this population (Faundes et al. 2018). Given similar challenges in speech, language, and communication inherent to P-MS, the positive correlation between parent satisfaction and ST dose/accessibility/convenience would indicate a similar desire among parents of children with P-MS for increased amount of ST.

There were a number of significant correlations identified specific to age grouping of children with P-MS, with little carryover between age groups. One explanation of these varied age group-specific findings may be the variability in speech, gross motor, and fine motor abilities among P-MS children from the same age cohorts. Due to wide variability of abilities and heterogeneity of phenotypes inherent to P-MS (Kolevzon et al. 2014), children of a similar age may be better suited to different preferred types and settings of ST. This is supported with the demographic data in this study (See Table 3 and Figure 3), showing a wide range of settings and techniques within this study population. Similar trends are also present in the provision of ST in children with other developmental disabilities, such as cerebral palsy, where different subtypes of CP are associated more highly with a preference for the ST technique of AAC compared to conventional approaches such as speech articulation (Cockerill et al. 2014).

Table 3.

Types of speech therapy

	Children with P-MS (n= 39)
Speech therapy setting
Hospital/Clinic	0/38 (0)
Home	5/38 (13.2)
School	8/38 (21.1)
Home and School	8/38 (21.1)
Hospital/Clinic and Home	6/38 (15.8)
Hospital/Clinic and School	7/38 (18.4)
Hospital/Clinic and Home and School	3/38 (7.9)
Other	1/38 (2.6)
Speech Therapy Professional
Speech Therapist	10/39 (25.6)
Teacher	1/39 (2.6)
Parent/Caregiver	2/39 (5.1)
Speech Therapist + Teacher	3/39 (7.7)
Speech Therapist + Parent	9/39 (23.1)
Teacher + Parent	3/39 (7.7)
Speech Therapist + Teacher + Parent	11/39 (28.2)
Speech Therapy Technique
Sign Language	1/39 (2.6)
ABA + ST	3/39 (7.7)
Objects of Reference	7/39 (17.9)
PECS	12/39 (30.8)
Micro-switch	0/39 (0)
ACD	4/39 (10.3)
Other Symbolic	2/39 (5.1)
Music Therapy	0/39 (0)
Other	6/39 (15.4)
Verbal Speech Therapy	4/39 (10.3)

Note. All values are reported as n (%) unless otherwise noted.

This study has limitations that restrict the application of results to guide treatment recommendations for speech disorders in children with P-MS. No direct measurement of speech abilities was collected, limiting the ability to draw conclusions related to the effect of increased ST intensity on ST outcomes. Measurements of parent satisfaction utilized in this study are the only indirect measures of ST treatment effectiveness and do not necessarily result in improved ST treatment outcomes. However, research has shown that parent satisfaction is associated with external factors that may contribute to parental acceptance of ST (Bairati et al. 2011). Due to awkward rephrasing in adaptation, redundancy in measuring the subscale general satisfaction, and facilitate shortening in questionnaire length, only 17 out of 18 items on the PSQ-18 were used to assess parent satisfaction (See Table 1), yielding an incomplete measurement of general parent satisfaction.

Age-based findings may indicate general parent satisfaction trends specific to P-MS children within a similar aged cohort; however, it should be noted that the generalizability of these findings to the larger P-MS population is difficult to predict given the small sample sizes of children included in these research subpopulations (ages 1–4 n = 10, ages 5–8 n = 16–18, ages 9–12 n = 6, ages 13–17 n = 6). Additional study limitations may also relate to the lack of diversity of the sample. The sample population in this study was from a predominantly Caucasian background. With the exception of the subscale of parent satisfaction with ST related to financial aspects of ST, data pertaining to socioeconomic status were not collected. Therefore, the relationship between the effects of socioeconomic background and race/ethnicity regarding satisfaction with ST in children with P-MS was largely unaddressed in this research and may contribute to satisfaction with ST services. Further research incorporating socioeconomic variables may help to address this relationship.

As described in the methods section, ST length (years) was converted from approximate ranges to interval estimates of total ST length (years). While this enabled the calculation of ST cumulative intensity, the conversion may have resulted in inexact estimations of ST length, resulting in increased variation in the calculation of ST cumulative intensity, potentially affecting the reliability of this measurement. The use of interval estimates to facilitate the use of online questionnaires affects the precision of the calculation of ST cumulative intensity, leading to a less exact calculation of overall ST cumulative intensity compared to data utilizing specific measurements of ST intervention duration.

Another study limitation is related to the measurement of ST dose in this study, which was measured in minutes per session, instead of a more definite measure such as trials per session. Previous related research has also utilized minutes per session as a corollary measure of ST dose (Namasivayam et al. 2015), as trials per session generally increase proportionally with session length. Nevertheless, measuring ST dose in minutes per session is an incomplete measure of ST dose compared to direct measures of the number of trials per session. Despite the clear advantages of using a more direct measure of ST dose, measuring ST dose using trials per session would necessitate observation of multiple ST sessions by a health care professional, which is outside the scope of research utilizing a parent completed Internet-based questionnaire.

Conclusion

Numerous significant correlations between ST intensity, parent satisfaction, and P-MS child age indicate that P-MS child age interacts as a significant covariable in the relationship between parent satisfaction with ST services and ST intensity. These age-specific findings may be due to the increase in ST intensity that results from increases in P-MS child age, as it acts a covariable with ST dose to exert significant positive correlational and predictive effects on some subscales of parent satisfaction. In spite of these qualifications, parents of P-MS children may be reassured that this research identifies positive correlations in subscales of satisfaction with ST services as both ST intensity and age increase.

Significant partial correlations were discovered between ST dose (minutes per session) and parent satisfaction in the subscales of time spent with ST (r=.36, p<.05) and accessibility and convenience (r=.40, p<.05) in children with P-MS controlling for P-MS child age. Analysis of parent satisfaction with ST services related to linear regression analysis showed parity, with ST dose a statistically significant predictor of cumulative parent satisfaction, technical quality, time spent with ST, and accessibility and convenience, largely independent of child age. Objective and subjective speech is also significantly correlated with parental satisfaction with ST services, speech therapists and other health care professionals implementing ST regimens with children with P-MS, or other similar developmental disabilities, may find these research findings useful when devising and implementing treatment plans to address barriers to parent satisfaction. ST frequency and ST length (years) show no significant correlation with increased parent satisfaction when controlling for P-MS child age (Table 5), indicating that increases in ST dose (minutes per session) alone represent a possible mechanism for increasing parent satisfaction with ST services in children with P-MS. Parent satisfaction with ST services was also found to be dependent on a number of factors that are age specific, language skill specific, and dependent on ST setting, and ST technique. These findings highlight the individual preferences of P-MS parents are dependent on factors related to age and language abilities. However, across all ages and settings of ST, only therapeutic increases in ST dose (minutes per session) were found to significantly influence parent satisfaction with ST services across a number of ST satisfaction subsets. These findings indicate that ST dose (minutes per session) may represent a mechanism for increasing parent satisfaction with ST services in children with P-MS

Parent perceptions of ST in children with intellectual disability augment the efficacy of ST delivery (Carroll 2009); however, further research is needed to determine whether increases in parent satisfaction with ST services in the P-MS population yield improved objective outcomes in ST adherence, expressive speech, or receptive speech. Future research utilizing direct measures of ST trials per session, exact interval measures of ST length in years, and objective measures of speech abilities may further elucidate the relationship between ST intensity and speech outcomes in children with P-MS. Nevertheless, this study represents a first step examining the effects of increased ST intensity, specifically increased ST dose, as a potential future therapeutic modality for improving speech outcomes in children with P-MS.