Early Interventionists’ Knowledge of Evidence-Based Practices for Autism

Abstract

This online survey study examined early intervention providers’ knowledge and training needs surrounding evidence-based practices for autism spectrum disorder (ASD). The researchers analyzed data from 87 early intervention providers (speech-language pathologists and developmental specialists) in a rural Southwestern U.S. state. The survey included both quantitative (i.e., rating and multiple-choice questions) and qualitative (open-ended) components. Survey sections included demographics, self-ratings of knowledge of evidence-based practices/self-reported training needs and barriers to working with children with ASD, and directly assessed knowledge of evidence-based practices. Findings support specific knowledge gaps and training needs (e.g., practices for addressing challenging behaviors, strategies for working with children with limited communication skills) and suggest developmental specialists who provided special instruction services may have greater training needs than speech-language pathologists. Findings have important implications for professional development and training programs for early intervention providers.

Keywords

autism spectrum disorder early intervention survey evidence-based practices

Best practice for autism spectrum disorder (ASD) involves intervening early using evidence-based strategies (Bradshaw etal., 2015; Landa, 2018). In the United States, Part C of the Individuals with Disabilities Education Improvement Act (IDEA, 2004) ensures that children ages 0 to 3 with developmental delays have access to early intervention services such as occupational therapy, speech-language pathology, and special instruction. Although IDEA does not require services to be evidence-based, it does require that, to the best extent possible, interventions are based in peer-reviewed research. In a review of ASD intervention research, Wong et al. (2015) identified 27 evidence-based practices for children with ASD. Practices were also disaggregated based upon their evidence-base for different age groups. Although the review did not identify practices with specific support for children ages 0 to 3, it did determine practices for children 0 to 5 years old. Although other reviews such as the one conducted by the National Autism Center (2015) classify evidence-based practices in slightly different terms, there is considerable overlap in the practices identified by reviews. A recent updated review added some new practices and collapsed others, but still supports the strategies from the 2015 review (Steinbrenner etal., 2020).

Although it is critical that early intervention providers understand and apply evidence-based strategies, there is often a programmatic research-to-practice gap in ASD (Luskin-Saxby & Paynter, 2018; Stahmer etal., 2005). Barriers that contribute to this gap may include limited knowledge, lack of resources, and limited specialized training (Luskin-Saxby & Paynter, 2018; Paynter etal., 2018). Service providers may rely on professional development sources that vary in quality and content (Paynter etal., 2017, 2018). In addition, there is variability in pre-service training and licensure or certification requirements for providers. Although some early intervention providers like speech-language pathologists (SLPs) have specific certifications, qualifications for individuals providing special instruction (which focuses on working with families and children to promote growth across developmental domains) are not specified by IDEA (2004) and vary by state. For instance, in New York, special instruction is delivered by certified special educators (New York Department of Health, 2021). Other states, such as Texas and New Mexico, do not require that early intervention providers delivering special instruction have specific certifications or licensures (New Mexico Department of Health, 2013; Texas Health and Human Services, 2021). Educational degree requirements for providers of special instruction also differ by state. For instance, early intervention specialists, who deliver special instruction in Texas must have a bachelor’s degree and at least 18 credits related to early childhood (Texas Health and Human Services, 2021). In contrast, in New Mexico, entry-level developmental specialists who deliver special instruction need only a high school degree (New Mexico Department of Health, 2013). Developmental specialists in New Mexico with an associate’s, bachelor’s or graduate degree can have backgrounds in education or a variety of related fields such as social work, counseling, or music therapy (New Mexico Department of Health, 2013).

Understanding the knowledge gaps and training needs of different early intervention providers in states that have minimal qualifications for some providers may lead to changes in preservice training requirements. Such knowledge might also aid in the creation of differentiated professional development programs that can attract and retain qualified personnel. Although a variety of providers (e.g., occupational and physical therapists) work with children with ASD, some evidence-based practices may be outside the scope of practice for these therapies. However, as SLPs address communication skills that affect multiple domains (e.g., social, cognitive, and adaptive) a larger array of evidence-based practices might be applicable. Similarly, as providers delivering special instruction (e.g., developmental specialists) intervene across domains, knowledge of a variety of practices is critical.

A limited number of studies have examined SLPs’ self-reported measures of competency, confidence, or training in ASD (cf. Plumb & Plexico, 2013; Schwartz & Drager, 2008). While revealing potential training needs, these studies did not explicitly assess knowledge of intervention strategies. Ray (2010) directly tested SLPs’ knowledge of ASD characteristics, intervention, and principles related to applied behavior analysis (ABA). Although most SLPs correctly answered questions about characteristics, they did not perform as well on intervention and ABA questions. For example, a majority of participants incorrectly answered questions relating to reinforcement and extinction. A majority of SLPs also indicated they had not received training in a variety of evidence-based practices (e.g., picture exchange communication system, functional communication training). Paynter et al. (2018) asked allied health professionals (e.g., SLPs, occupational therapists, behavior analysts) working in various settings to classify ASD practices as supported, emerging, or unsupported. Although there were high accuracy rates, behavior analysts were most accurate and most likely to report using evidence-based practices. Although there are no studies solely examining the knowledge of providers delivering special instruction, some studies have assessed knowledge across a range of early intervention providers (e.g., therapists and special educators). In one study, providers in California often reported being unaware of evidence-based practices for ASD (Stahmer etal., 2005). In a study by Paynter et al. (2017), Australian providers reported greater knowledge/use of evidence-based practices than emerging or unsupported practices but rated themselves as having limited knowledge of some evidence-based practices (e.g., pivotal response training, technology-aided instruction).

Based upon gaps in prior research, the primary aim of the current study was to determine early intervention providers’ knowledge and training needs in evidence-based practices for ASD. The current study focused specifically on assessing the knowledge of SLPs and developmental specialists (i.e., providers of special instruction) working in a Part C early intervention program in a rural Southwestern state. Comparisons across these groups were considered critical because while SLPs have extensive pre-service educational and clinical practice requirements, developmental specialists with varying educational backgrounds and experiences are often the primary providers serving families in the state where the study was conducted. Evidence indicating differences in ASD knowledge across the two provider groups, might, therefore suggest the need for differential changes to pre- and post-service training. Primary research questions included the following:

Research Question 1 (RQ1): What are early intervention providers’ self-reported and directly assessed knowledge levels of evidence-based practices for ASD?

Research Question 2 (RQ2): Do self-reported and directly assessed knowledge levels differ among SLPs and developmental specialists?

Research Question 3 (RQ3): What training interests and barriers to working with children with ASD do early intervention providers report?

Post hoc analyses examining the effects of educational level and years of experience were conducted. Qualitative analysis of open-ended questions was used to enhance the social validity of quantitative findings.

Method

This study consisted of an online survey distributed via QualtricsXM (2018) software to early intervention providers (SLPs and development specialists) in a rural Southwestern state. Due to minimal risk, this study was determined to be exempt from additional review by an Institutional Review Board. An informed consent statement was provided at the start of the survey, and participants had to indicate that they understood that continuing to complete the survey constituted consent.

Recruitment and Survey Distribution

Initially, we contacted all 36 early intervention program directors in the state where the study was conducted. We provided directors with a study description and asked them to forward the survey to developmental specialists and SLPs employed or contracted to work in early intervention. Twenty-one directors (58%) distributed the email across agencies representing all state counties. Based on directors’ report, they sent 324 total emails. The total number of unique providers who received the link could not be determined as many providers contract with multiple agencies. Emails sent to providers included a description of the study and incentives (a US$20 Amazon gift card) along with the survey link. We asked directors to send one reminder a week later. If participants began but did not complete the survey, we sent one reminder.

Participants

A total of 111 participants (75 developmental specialists, 36 SLPs) across the state completed the demographics section of the survey (e.g., questions related to discipline, caseload, education, and experience). Attrition occurred, with 68 developmental specialists and 34 SLPs completing a self-report section of the survey, and 55 developmental specialists and 32 SLPs completing the entire survey (87 total). We excluded data from participants who did complete the entire survey. Exact response rates could not be determined as we do not know the number of unique providers who received the link. A conservative response rate (assuming no one received the link more than once) would be 27% for the full survey.

Caseloads

Average total early intervention caseloads were 19 children (developmental specialists) and 23 children (SLPs). Across both participant groups, providers reported that more children on their caseloads displayed characteristics of ASD than had formal diagnoses. For example, although only 44% of developmental specialists reported that they were currently working with children with an ASD diagnosis, 87% noted having one or more children on their caseload with ASD symptoms. In comparison, 66% of SLPS reported having at least one child with an ASD diagnosis and 81% reported having one or more children with ASD symptoms on their caseload.

Experience and education

For developmental specialists, the years of experience in early intervention included 1 to 5 years (55%), 6 to 10 years (27%), 11 to 15 years (5%), and 16+ years (13%). For SLPs, years of experience included 1 to 5 years (31%), 6 to 10 years (38%), 11 to 15 years (9%), and 16+ years (22%).

Across the 55 developmental specialists, a bachelor’s degree was most common (55%), followed by master’s (38%) and associate’s (4%) degrees. One participant reported graduate certificates in early childhood education and psychology but did not note full degrees. Data are missing for one participant. For bachelor’s degrees (n = 30), the most common fields were elementary education (n = 7), psychology (n = 7), and human development and family studies (n = 5). The most common fields for master’s degrees (n = 21) were special education (n = 6), early childhood education (n = 5), and social work (n = 4). All SLPs had a master’s degree in speech and hearing sciences or communication disorders, and one also had a doctorate in special education.

Sixty-four percent of developmental specialists and 53% of SLPs reported that they had not taken any college-level courses fully devoted to ASD. Only 23% of developmental specialists and 25% of SLPs reported having taken more than one course fully devoted to ASD. Participants were more likely to report taking college-level courses that included ASD topics (i.e., ASD was discussed in class, but not the main topic of the course). For instance, 66% of developmental specialists and 75% of SLPs reported having taken more than one college course that included ASD topics. However, 16% of developmental specialists and 12% of SLPs reported having taken no such courses. With regard to continuing education, 51% of developmental specialists and 94% of SLPs reported having completed professional development hours related to ASD.

Survey and Development

The researchers built the survey on the Qualtrics^XM online survey platform. A copy of the full survey is available upon request. To develop the demographics section, we examined information commonly reported in prior survey studies (e.g., Schwartz & Drager, 2008). To develop the self-report section, the first author first examined the review by Wong et al. (2015) to determine evidence-based practices for young children (0–5) with ASD related to social behaviors, communicative behaviors, and challenging behaviors. This included 24 practices (for definitions see Wong etal., 2015). Although these practices are distinct from one another, more broad practices (e.g., naturalistic intervention) involve the understanding of other practices (e.g., prompting, modeling). In addition, while one of these practices is described as technology-aided interventions, for this study we focused on speech-generating devices (SGDs) as these are most applicable in early intervention. The first author and three additional experts independently sorted the identified practices into larger broad categories based upon the primary purposes of the practice. All experts were doctoral-level behavior analysts with 5 to 10 years of experience. Three had worked in Part C early intervention programs across the United States, and the fourth worked with young children with ASD for private agencies in Texas. Behavior analysts were recruited because many practices are derived from ABA. Using feedback from the experts, the 24 practices were grouped into four broad categories: general instructional formats (ways to deliver treatment such as discrete trial or naturalistic intervention), general instructional strategies (practices such as modeling or prompting used to address a variety of skills), challenging behavior practices (practices like functional communication training or extinction used primarily to decrease challenging behaviors), and social communication practices (practices such as SGDs and social narratives used primarily to increase social or communication skills). After grouping practices, we developed 24 Likert-type scale questions asking respondents to rate their knowledge of each practice using a scale from 1 to 5 (1 = not at all knowledgeable to 5 = extremely knowledgeable). Participants also ranked the four broad categories from most to least desired area for training. In addition, participants were provided with a list of practices under each broad category, and asked to select one practice within each category in which they most desired further training. Finally, the section included two open-ended questions: “What topics would you like to see included in an ASD intervention training program?” and “What are the biggest challenges in working with children with ASD?”

To develop the knowledge assessment, the first author and the three additional experts described above wrote questions for evidence-based practices within their area of expertise. Each expert reviewed definitions of their assigned practices (derived from Wong etal., 2015) and created a pool of three multiple choice questions applicable to early intervention contexts for each practice. Multiple choice questions were selected as they are less time-intensive and easier to objectively assess than other question types. The authors compiled the list of 72 potential questions and sent them to an additional set of experts for validation. The second set of experts included a dually certified behavior analyst/SLP who specializes in ASD, and two university clinic SLPs who specialize in ASD (including one with prior experience as a developmental specialist). These experts were asked to note any questions that might be confusing, were less applicable to early intervention, or had answer choices with which they disagreed. Based upon feedback, the first author narrowed the list down to 48 questions. (i.e., two per practice), which were then sent back to the ABA experts. Experts selected the question they felt was (a) most representative of the practice and (b) most applicable to early intervention. Using feedback, the first author narrowed the pool to 24 questions (i.e., one per practice). We included one question per strategy to reduce attrition due to survey length. Samples of four knowledge assessment questions (one from each broad category) are provided in supplementary materials.

Final Survey Expert Review and Pilot Study

The authors entered all survey sections into Qualtrics^XM. Three university-based experts completed the survey online and answered questions regarding the survey validity. These experts included a research coordinator in a speech and hearing sciences department who had worked as a developmental specialist and an SLP, a university clinical SLP who specializes in ASD, and a professor of educational psychology who specializes in ASD and has conducted prior survey research. After each question, experts were asked to provide feedback. Experts also noted agreement with the strategy classification into the four broad categories. Based upon feedback, we made minor edits (e.g., adding a question regarding total caseloads). Following the final expert review, we conducted a small pilot study with an early intervention agency in another Southwestern U.S. state. The agency director distributed the survey link to 10 SLPs and early intervention specialists (i.e., providers of special instruction). Pilot participants received a US$10 Amazon gift code upon survey completion. Two early intervention specialists and four SLPs completed the survey. Pilot participants were asked follow-up questions regarding any suggested changes. We made minor changes to study questions following the pilot study (e.g., rewording one question).

Mixed Methods Approach

This study was primarily “quantitatively driven,” but an embedded mixed methods design was used wherein qualitative data were collected concurrently to enhance quantitative findings (Schoonenboom & Johnson, 2017). Qualitative data served to provide support that the evidence-based practices assessed in closed-ended questions were socially valid (e.g., related to real-world training needs and challenges) and to determine additional provider needs.

Quantitative Analysis

For the self-report section, within and across participant groups, we computed: (a) mean Likert-type scale scores for self-reported knowledge ratings, (b) the percentage of participants who ranked each broad category as their first or second training interest, and (c) the percentage of individuals who ranked a specific practice as their highest training interest within a broad category. For the knowledge assessment, within and across participant groups, we computed the mean percentage correct across all questions and for each of the broad categories. We also determined the percentage of participants, within and across groups, who answered specific questions correctly. We computed Pearson’s coefficient to assess the relationship between self-report ratings and percentage correct on the knowledge assessment. We used two-tailed independent t tests (unequal variance) to determine differences between SLPs’ and developmental specialists’ mean self-ratings and knowledge scores for the full assessment and for each broad category. To explore the effects of educational level, we also conducted two-tailed independent t tests (unequal variance) to determine differences between mean self-ratings and mean knowledge scores of SLPs and developmental specialists with a master’s degree. To determine the effects of years of experience (categorized into groups) a Kruskal–Wallis test was conducted. Dunn’s test with Bonferroni corrections was used for pairwise comparisons.

Qualitative Analysis

Responses to open-ended questions were uploaded to QSR International’s NVivo 12 software (2018). Following methods described by Miles et al. (2014), descriptive codes detailing the topic/focus of distinct phrases within a participant response were applied. Descriptive coding can involve both a priori codes, as well as codes that emerge, are refined, and collapsed during an iterative and cyclical process (Miles etal., 2014). Primary and subcodes are also developed and applied (Miles etal., 2014). For the question “What topics would you like to see included in an ASD intervention training program?” a priori codes derived from the broad categories developed for the close-ended questions (e.g., challenging behavior practices) were initially applied. The second author read all participant responses and labeled distinct phrases/concepts within responses with a priori codes and created additional broad codes (e.g., parent support) when phrases did not fit an a priori code. After broad codes were applied, the second author sorted phrases into subcodes. A priori subcodes using the 24 evidence-based practices were applied first and additional subcodes were created as needed. To reduce the total number of subcodes, related topics with fewer responses were collapsed into larger subcodes (e.g., Picture Exchange Communication System [PECS] and SGDs collapsed to augmentative and alternative communication). For the question “What are the biggest challenges in working with children with ASD?” a priori codes were not used because responses to this question were not a list of practices. The second author first read responses and used short labels (e.g., engagement issues) to describe the primary focus of distinct phrases within a response. These labels were then collapsed to create a smaller list of subcodes (e.g., engagement/motivation) that were then applied to each distinct phrase. The coder then grouped related subcodes together to create larger broad codes such as limited foundational skills. For both questions, both authors collaboratively reviewed all codes and any disagreements were discussed and modified until 100% agreement was reached.

Results

Quantitative Findings

Mean self-rating across all evidence-based practices was 2.85 out of 5 (SD = .84) across all participants, 2.68 (SD = .91) for all developmental specialists, 2.76 (SD = 1.14) for masters level developmental specialists, and 3.15 (SD = .61) for SLPs. SLPs had significantly higher ratings than all developmental specialists, t(83.44) = 2.87, p = .005, but not masters’ level developmental specialists, t(27.48) = −1.412 p = .17. Mean ratings across participants, were moderately positively correlated with knowledge assessment scores; r (85) = .373, p < .0001.

The mean percentage correct on the knowledge assessment was 64% (SD = 14%) for all participants, 59% (SD = 13%) for developmental specialists, 60% (SD = 11%) for developmental specialists with a master’s degree, and 73% (SD = 10.58%) for SLPs. SLPs had significantly higher scores than all developmental specialists, t(74.44) = 5.32, p < .0001, and developmental specialists with a master’s degree, t(41.82) = −4.16 p < .0001.

The Kruskal–Wallis test did not provide evidence of a significant difference between groups with varying years of experience on self-ratings, χ²(3) = 3.87, p = .237, but did indicate that at least two groups differed significantly on knowledge scores, χ²(3) = 8.897, p = .031. Dunn’s test indicated that the only significant differences (at p < .05) on knowledge scores were between individuals with 1 to 5 years of experience and those with 16+ years (p = .019). Providers with 16+ years (n = 14) had a mean score of 72% (SD = 15%) and those with 1 to 5 years (n = 40) had a mean score of 61% (SD = 12%).

Broad category findings

Table 1 provides a summary of self-assessment ratings and knowledge scores across the four broad categories for developmental specialists and SLPs. Across groups, participants rated themselves as most knowledgeable about general instructional strategies (M = 3.40, SD = .86) and least knowledgeable about general instructional formats (M = 2.49, SD = .83) and challenging behavior practices (M = 2.49, SD = .98). The mean knowledge ratings for social communication practices was 2.97 (SD = 1.05). SLPs were significantly more likely to rate themselves as more knowledgeable than developmental specialists across general instructional formats, t(84.56) = 2.57, p = .012, general instructional strategies, t(79.007) = 3.56, p = .001, and social communication practices, t(78.82) = 3.51, p = .001. There were no significant differences for challenging behavior practices, t(76.98) = .64, p = .53.

Table 1.

Mean Self-Assessment Ratings and Mean Percentage Correct on Knowledge Assessment for Broad Categories and All Practices.

Category	Mean self-reported knowledge rating for practices in category^a			Mean percentage correct on knowledge questions in category
Category	Developmental specialists	SLPs	All	Developmental specialists	SLPs	All
General instructional formats	2.36 (SD = .93)	2.78* (SD = .57)	2.51 (SD = .84)	56.8% (SD = 19.80%)	67.60%* (SD = 18.60%)	60.60% (SD = 20.00%)
General instructional strategies	3.19 (SD = .89)	3.80* (SD = .67)	3.41 (SD = .87)	70.71% (SD = 21.28%)	84.86%* (SD = 15.14%)	75.86% (SD = 20.42%)
Challenging behavior practices	2.39 (SD = 1.07)	2.53 (SD = .85)	2.44 (SD = 1)	54.86% (SD = 20.14%)	67.00%* (SD = 17.71%)	59.28% (SD = 20.14%)
Social communication practices	2.74 (SD = 1.10)	3.48* (SD = .83)	3.01 (SD = 1.06)	50.20% (SD = 18.60%)	68.20%* (SD = 19.20%)	56.80% (SD = 20.80%)
All practices	2.68 (SD = .91)	3.15* (SD = .61)	2.85 (SD = .84)	58.86% (SD = 12.7%)	72.52%* (SD = 10.58%)	63.88% (SD = 13.67%)

Note. SLP = speech-language pathologist.

Scale = 1 to 5.

SLP mean ratings or scores significantly higher than developmental specialist at p <.05.

On the knowledge assessment, general instructional strategies had the highest mean percentage correct (M = 76%, SD = 20%). Means in other areas were 61% (SD = 20%) for general instructional formats, 59% (SD = 20%) for challenging behavior practices, and 57% (SD = 21%) for social communication practices. Across all categories, SLPs had higher percentages correct. Differences were significant for general instructional formats, t(67.93) = 2.52, p = .014, general instructional strategies, t(81.36) = 3.56, p = .001, challenging behavior practices, t(71.71) = 2.9, p = .005, and social communication practices, t(62.81) = 4.18, p < .0001.

Specific strategy findings

Table 2 provides mean ratings and the percentage of participants who answered each knowledge question correctly, across and within groups. Across groups, for general instructional formats participants rated themselves as most knowledgeable about parent-implemented intervention (M = 3.17, SD = 1.1) and least knowledgeable about pivotal response training (M = 2.06, SD = 1.08). Fewer than 60% of the total participants correctly answered questions on discrete trial teaching, naturalistic intervention, and pivotal response training. Developmental specialists also had high rates of incorrect responding to the peer-mediated intervention question. For general instructional strategies, participants rated themselves as most knowledgeable about modeling (M = 3.83, SD = .93) and least knowledgeable about video modeling (M = 2.56, SD = 1.17). There were no questions that fewer than 60% of the total participants answered correctly, however, developmental specialists had high rates of incorrect responding on video modeling and prompting. For challenging behavior practices, participants rated themselves as most knowledgeable about functional behavior assessment (M = 2.68, SD = 1.44) and response interruption/redirection (M = 2.68, SD = 1.38), and least knowledgeable about extinction (M = 2.35, SD = 1.14) and differential reinforcement (M = 2.35, SD = 1.14). Fewer than 60% of the total participants correctly answered questions on differential reinforcement, extinction, and functional behavior assessment and developmental specialists also had high rates of incorrect responding on functional communication training. For social communication practices, participants rated themselves as most knowledgeable about the Picture Exchange Communication System; PECS (M = 3.25, SD = 1.16.) and least knowledgeable about SGDs (M = 2.66, SD = 1.15). Fewer than 60% correctly answered questions on PECS and SGDs.

Table 2.

Mean Self-Assessment Ratings and Percentage of Participants Answering Knowledge Question Correctly for Individual Practices.

Practice	Mean self-rating^a			Percentage answering question correctly
Practice	Developmental specialists	SLPs	All	Developmental specialists	SLPs	All
General instructional formats
Discrete trial instruction	2.06 (SD = 1.19)	2.63 (SD = 0.86)	2.27 (SD = 1.11)	58.18%	46.88%	54.02%
Naturalistic intervention	2.49 (SD = 1.17)	3.25 (SD = 0.94)	2.77 (SD = 1.15)	27.27%	56.25%	37.93%
Parent-implemented intervention^b,c	3.00 (SD = 1.18)	3.44 (SD = 0.97)	3.16 (SD = 1.12)	94.55%	90.32%	93.02%
Peer-mediated intervention	2.31 (SD = 1.11)	2.38 (SD = 0.89)	2.33 (SD =1.04)	50.91%	81.25%	62.07%
Pivotal Response Training	2.00 (SD =1.12)	2.19 (SD =1.04)	2.07 (SD =1.10)	54.72%	65.63%	58.82%
General instructional strategies
Modeling	3.56 (SD = 0.93)	4.34 (SD = 0.64)	3.85 (SD = .92)	80.00%	96.88%	86.21%
Prompting	3.45 (SD = 0.97)	4.31 (SD = 0.68)	3.77 (SD = .97)	59.26%	75.00%	65.12%
Reinforcement	3.51 (SD = 1.02)	4.03 (SD = 0.77)	3.7 (SD = .97)	69.09%	87.50%	75.86%
Task analysis	2.65 (SD = 1.19)	3.09 (SD = 1.07)	2.82 (SD = 1.17)	73.36%	68.75%	73.56%
Time delay	3.31 (SD = 1.06)	4.09 (SD = 0.80)	3.60 (SD = 1.04)	75.93%	93.75%	82.56%
Video modeling^c	2.42 (SD = 1.20)	2.84 (SD = 1.12)	2.57 (SD = 1.19)	56.36%	75.00%	63.22%
Visual supports^b	3.42 (SD = 0.99)	3.88 (SD = 0.99)	3.59 (SD = 1.01)	80.00%	96.88%	86.21%
Challenging behavior practices
Antecedent based intervention	2.37 (SD = 1.16)	2.59 (SD = 0.93)	2.45 (SD = 1.09)	78.18%	87.50%	81.61%
Differential reinforcement	2.31 (SD = 1.26)	2.22 (SD = 0.93)	2.28 (SD = 1.15)	41.82%	31.25%	37.93%
Exercise	2.52 (SD = 1.15)	2.25 (SD = 0.94)	2.42 (SD = 1.08)	74.55%	81.25%	77.0%
Extinction	2.20 (SD = 1.18)	2.47 (SD = 1.15)	2.30 (SD = 1.17)	42.59%	62.50%	50%
Functional behavior assessment^b	2.48 (SD = 1.24)	2.88 (SD = 1.14)	2.63 (SD = 1.11)	12.73%	31.25%	19.54%
Functional communication training^c	2.24 (SD = 1.25)	2.63 (SD = 1.14)	2.38 (SD = 1.22)	54.55%	75.00%	62.07%
Response interruption/redirection	2.61 (SD = 1.27)	2.66 (SD = 1.05)	2.63 (SD = 1.19)	80.00%	100.00%	87.36%
Social communication practices
PECS^c	3.05 (SD = 1.30)	3.78 (SD = 0.74)	3.32 (SD = 1.18)	16.36%	56.25%	31.03%
Scripting	2.55 (SD = 1.27)	3.16 (SD = 1.09)	2.79 (SD = 1.24)	61.82%	75.00%	66.67%
Social narratives	3.00 (SD = 1.29)	3.63 (SD = 0.89)	3.23 (SD = 1.20)	72.73%	81.25%	75.86%
Social skills training^b	2.79 (SD = 1.26)	3.47 (SD = 1.09)	3.02 (SD = 1.25)	83.33%	87.50%	84.88%
SGDs	2.36 (SD 1 = .20)	3.38 (SD = 0.82)	2.74 (SD = 1.18)	18.87%	40.63%	27.06%

Note. SLP = speech-language pathologist; DS = developmental specialist; PECS = Picture Exchange Communication System; SGDs = Speech-generating devices.

Scale = 1 to 5. ^bHighest interest DSs. ^cHighest interest SLPs.

Training interests

Social communication practices (64% of participants) and challenging behavior practices (63% of participants) were most often ranked as first or second training interests. For each broad category, the specific strategies participants most often selected as the top training interest were parent-implemented intervention for general instructional formats (48%), visual supports for general instructional strategies (23%), functional communication training for challenging behavior practices (26%), and social skills training for social communication (42%).

Qualitative Findings

Open-ended responses regarding training interests were most commonly coded as broadly relating to challenging behavior practices (n = 51). An example quote related to this topic included: “how to deal with specific behaviors—for example, screaming, head-banging.” The most common subtopics for this broad code were general behavioral modification strategies (applied to quotes like that above that did not mention a specific strategy), response interruption/redirection, functional communication training, and functional behavior assessment. The second most commonly coded broad topic was social communication practices (n = 36), with common subtopics such as social skills training, augmentative and alternative communication, and language/communication. For example, “No agency I have ever worked for has provided things like PECS” described an interest in augmentative and alternative communication. The broad code general instructional formats (n = 24) most commonly included quotes related to parent implemented intervention and naturalistic intervention, and general instructional strategies (n = 14) most commonly included phrases related to modeling. Other broad topics of interest included any evidence-based intervention (n = 11), parental support (n = 9), ABA practices (n = 7), multidisciplinary collaboration (n = 5), and diagnosis(n = 4).

The most commonly reported challenges to working with children with ASD centered on issues related to working with parents and families (n = 34). These involved challenges related to parental understanding and acceptance of ASD, parent involvement and follow-through, and helping parents meet child and family needs. An example quote coded as parental understanding and acceptance of ASD included “They [parents] really just want their baby fixed. They [parents] appear stuck in a certain stage of the grieving process.” The next most common barriers surrounded challenging behaviors (n = 20), limited foundational skills (n = 19), and limited social communication skills (n = 19). Responses coded as relating to challenging behaviors typically included general responses such as “behaviors” or “aggression,” while limited foundational skills most commonly focused on issues related to engagement/motivation, attention/joint attention, and task completion. For instance, one participant wrote “Finding what is motivating to them [the children] and using it in your [provider’s] favor. I feel like what is motivating will change.” The code limited social communication skills applied to quotes such as “buy-in for SGDs,” relating to issues working with minimally verbal children, or quotes like “how to support them [children] socially,” relating to the social deficits of ASD. Additional broad challenges included limited training or resources (n = 15), diagnostic issues (n = 6), and working in teams (n = 5).

Discussion

Findings from this study support prior research suggesting that early intervention providers may have gaps in their knowledge of evidence-based practice for ASD (Luskin-Saxby & Paynter, 2018 ; Stahmer etal., 2005) The study highlights specific knowledge gaps and training needs of developmental specialists and SLPs in a rural Southwestern U.S. state. The study extends prior peer-reviewed research (e.g., Paynter etal., 2017; Plumb & Plexico, 2013; Schwartz & Drager, 2008) by directly assessing provider knowledge rather than relying on self-report alone. Although self-rating scores in this study varied by topic and were higher for SLPs, SLP mean ratings across practices were in line with findings from Schwartz and Drager (2008). Ratings were moderately correlated with knowledge scores, but in some instances, higher mean ratings (i.e., above 3 out of 5) did not align with knowledge scores for topics such as PECS. Qualitative findings also enhanced the social validity of the study as frequently mentioned training interests (e.g., challenging behavior and social communication practices) aligned with knowledge gaps, and suggested barriers (e.g., working with parents, addressing challenging behaviors, and teaching foundational skills) can be addressed by evidence-based strategies.

Additional findings indicate differences across provider groups. For instance, SLPs performed significantly better than developmental specialists on both total knowledge scores and scores for each of the broad categories. SLP mean self-ratings across all categories except challenging behavior practices were also significantly higher than developmental specialists. Although a smaller number of SLPs participated, developmental specialists are more commonly employed in the state’s early intervention program. By nature of the role, all SLPs had a master’s degree, compared with only 38% of developmental specialists. Despite this fact, having a graduate degree alone did not appear to influence knowledge scores, as developmental specialists with master’s degrees still scored significantly lower than SLPs. Given the diversity of developmental specialists’ graduate programs of study, and the fact that SLPs have required coursework and field experiences focused on intervention, this may not be surprising. SLPs were also about twice as likely to have professional development related to ASD. In addition, a higher percentage of SLPs also reported more years of experience, more children diagnosed with ASD on their caseload, and more full courses in ASD. Similar to Ray (2010), years of experience did appear to have an effect on knowledge, but the only significant differences were between providers with 1 to 5 years and those with 16+ years (more experienced providers had higher mean scores). The categorical measurement used, may, however, have limited detection of additional effects.

When examining training needs across groups, a consistent finding was that training in challenging behavior practices is highly desired and warranted. Participants reported this as a top training need in closed and open-ended questions, had low mean self-ratings (2.49 out of 5), and low mean percentage correct on the knowledge assessment (59%). Challenging behaviors was also one of the most commonly reported barriers to working with children with ASD. Specific knowledge gaps across groups included functional behavioral assessment, differential reinforcement, and extinction. Developmental specialists also tended to lack knowledge in functional communication training. Although expertise in some of these strategies (e.g., extinction) may not be expected nor desired, participants reported training interests in functional behavior assessment and functional communication training. Overall, these findings align with those of Ray (2010) which indicated SLPs often lacked training in several of these topics.

Training in social communication practices was also frequently nominated as an area of interest on closed and open-ended questions, and limited social communication skills were frequently indicated as a barrier. A low mean percentage correct on the knowledge assessment (57%) primarily related to difficulty with questions related to two augmentative and alternative communication (PECS and SGDs). A need for further training in augmentative and alternative communication was also reflected in open-ended questions. Although Ray (2010) reported that a majority of SLPs had training in augmentative and alternative communication, most did not have prior training in PECS. Early intervention providers in the Paynter et al. (2017) study also reported having less knowledge of technology-aided instruction (which includes SGDs).

In terms of general instructional formats, there is evidence for knowledge gaps in discrete trial teaching, naturalistic intervention, and pivotal response training. The fact that only 38% answered the question on naturalistic intervention correctly may be particularly concerning given that this is critical in early intervention. Incorrect answers indicated that providers may not fully understand that naturalistic intervention can be systematic and address specific targets and goals. Findings from Ray (2010) indicate that a majority of SLPs reported that they had not received training in this topic. Another interesting finding was that although the mean self-rating for parent-implemented intervention was 3.16, and 93% answered this knowledge question correctly, this topic was the most common general instructional format nominated for further training. Working with parents and families was also the most commonly reported barrier.

In comparison to other topics, there appeared to be less interest and need for training in general instructional strategies (e.g., modeling, reinforcement). This may not be surprising given that many of these strategies are broadly applicable and used across children with a variety of developmental delays. Providers may still benefit from additional training in some topics such as prompting or video modeling (relative areas of weakness for developmental specialists).

Finally, additional qualitative information on training needs suggests that some participants are interested in any evidence-based strategies, and some want to learn more about ABA and collaboration. Additional barriers suggest that providers may benefit from training in topics such as parent counseling, and how to apply evidence-based strategies to improve limited foundational skills such as joint attention and motivation. However, programmatic issues such as diagnostic challenges and limited resources may also affect services in rural states.

Implications for Research and Practice

Findings support a need for developing ASD-focused professional development and training opportunities for early intervention providers in a rural Southwestern U.S. state. Although results may be more likely to generalize to other rural states or those with similar qualification requirements for providers delivering special instruction (e.g., developmental specialists), findings are in line with other studies supporting research-to-practice gaps in early intervention (e.g., Luskin-Saxby & Paynter, 2018; Stahmer etal., 2005). Options for differentiating community-based training to service providers with various educational backgrounds, certifications, and experiences should be considered, and additional supports (e.g., supervision and coaching) should be offered for individuals who lack training and experience. Replications of the current study could examine knowledge gaps in states with different pre- and post-service training requirements. For instance, as special instruction providers in New York must be certified educators, it is possible that they might perform more similarly to SLPs. Although such evidence might suggest that some states should consider revising qualifications, policy changes and funding to support providers in obtaining degrees in specific fields are needed. Future studies should consider a community-participation approach in which researchers and community agencies collaboratively assess gaps and create individualized training programs (Vivanti etal., 2018). Although providers from agencies across a rural state participated in this study, due to wide variability in agency size and response rates, and the fact that some providers worked for more multiple agencies, we could not assess knowledge differences across agencies. As most agencies employ both SLPs and developmental specialists, agency alone likely did not affect the differences between the two groups. However, future research could examine differences in the quality and content of training across agencies.

Training programs must also consider parental involvement (Vivanti etal., 2018). Although there are many studies in which researchers train parents to implement comprehensive or targeted early intervention models, there is limited research that focuses on teaching community providers to train parents (Bradshaw etal., 2015; Vivanti etal., 2018). Findings from this study suggest that (a) collaboration with parents is often seen as a barrier to working with children with ASD and (b) providers desire more training in methods for supporting parents. In addition, research should also focus on training in targeted evidence-based practices such as challenging behavior practices, augmentative and alternative communication, and naturalistic interventions. In rural states, the efficacy of online or short-term training programs should be explored. Although Hamad et al. (2010) demonstrated that online modules in general behavioral intervention strategies (e.g., reinforcement, prompting) improved early intervention providers’ knowledge of practices, these topics may not align with high priority needs. A recent study suggests that free online training modules on practices identified by Wong et al. (2015) may improve knowledge of evidence-based practices (Sam etal., 2020). Replications across specific provider groups that assess the effects of these trainings on applied outcomes is needed.

Given overlaps in scopes of practice, and the educational and experience differences among providers, the efficacy of collaborative service models should also be explored (Cox, 2012; Gerenser & Koenig, 2019). For instance, although SLPs play an important role in the implementation of behavior plans that focus on methods such as functional communication training (within their scope of practice), consultation with behavior analysts may be helpful (Gerenser & Koenig, 2019). In rural areas, behavior analysts might provide short-term consultative training or telehealth coaching to SLPs or developmental specialists. SLPs with ASD expertise could also provide coaching to developmental specialists. As services for children with ASD become more collaborative, shared beliefs, knowledge and frameworks are needed (Cox, 2012; Gerenser & Koenig, 2019). Unfortunately, misunderstandings of evidence-based approaches may lead to resistance to collaboration (Gerenser & Koenig, 2019). To promote collaboration, universities should consider interdisciplinary training in ASD.

Limitations

There are several limitations of this study. First, we relied upon one specific review for determining evidence-based practices (Wong etal., 2015). This review was selected because it was comprehensive and systematic and included detailed definitions of practices that have support for young children (ages birth to 5 years). However, the fact that the review did not disaggregate findings for birth to 3 years of age might suggest that some practices assessed in this study do not have as strong of an evidence base for infants and toddlers. In addition, while there is a crossover in practices identified by Wong et al. (2015) with other reviews, researchers have identified additional promising practices, and have noted that factors such as social validity should also be considered when selecting evidence-based practices (Callahan etal., 2017). Furthermore, it could be argued that some practices (e.g., extinction) would not be appropriate for SLPs or developmental specialists to consider. In addition, while we had several levels of validation of our broad categories, some practices might have secondary purposes. For instance, while the primary reason to introduce functional communication training may be to decrease challenging behavior, this practice also teaches social communication. In addition, we included only one knowledge question per practice to prevent attrition. There were several rounds of expert review to ensure content validity, but it is possible that some questions were more difficult than others. Although we may be able to make less conclusive statements about knowledge of specific practices, aggregating findings across broad topic areas showed consistencies in terms of areas of need (e.g., challenging behavior practices) and strengths (general instructional strategies). Questions in this study also did not focus on ASD characteristics. Research efforts should consider the benefits of training programs that embed these topics. In addition, although we could not determine an exact response rate due to duplication across agencies, a conservative response rate was 27%. A meta-analysis of survey research with health care professionals reported a mean 38% response rate for online surveys, but indicated rates are on a downward trend and are often lower in the United States (Cho etal., 2013). Although our rate may be slightly below average, prior related studies did not report rates or reported lower rates (cf. Paynter etal., 2018; Ray, 2010; Schwartz & Drager, 2008). Given employment replication, it is probable that our actual response rate was higher, but attrition was still a concern particularly for developmental specialists. Greater monetary incentives may have increased rates and decreased attrition (Cho etal., 2013). Finally, in addition to geographical limitations, it is unclear whether findings would generalize to other service providers. Replication across states and different providers is necessary.

Conclusion

Findings support a research to practice gap in ASD intervention among early intervention providers in a rural, Southwestern state. Additional pre-professional and professional supports such as training and coaching programs are needed to ensure that early intervention providers can support young children with ASD and their families. Collaborative service models as well as interdisciplinary preprofessional training programs (e.g., between SLP and special education departments) should also be further explored. Training for early intervention providers should (a) address high priority areas such as practices for reducing challenging behaviors and increasing social communication, (b) be differentiated to providers, and (c) incorporate parents. In addition, while some states may want to consider revising the educational, training, or certification requirements of early intervention providers who deliver special instruction to children with ASD, funding and resources to train, attract, and retain qualified providers are necessary.

Supplemental Material

sj-docx-1-foa-10.1177_10883576221099895 – Supplemental material for Early Interventionists’ Knowledge of Evidence-Based Practices for Autism

Supplemental material, sj-docx-1-foa-10.1177_10883576221099895 for Early Interventionists’ Knowledge of Evidence-Based Practices for Autism by Cindy Gevarter, Mariah G. Siciliano and Erin Stone in Focus on Autism and Other Developmental Disabilities

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The first author is a board certified behavior analyst as well as a former early childhood special teacher and early interventionist. The second two authors are speech-language pathologists. The authors’ training in these specific disciplines may have impacted the study’s design and interpretation.

Funding

This study was supported by a research allocation grant from the University of New Mexico. The second two authors were students at the University of New Mexico when data collection occurred.

ORCID iD

Cindy Gevarter

Supplemental Material

Supplemental material is available on the Focus on Autism and Other Developmental Disabilities webpage with the online version of the article.

References

Bradshaw

Steiner

A. M.

Gengoux

Koegel

L. K.

(2015). Feasibility and effectiveness of very early intervention for infants at-risk for autism spectrum disorder: A systematic review. Journal of Autism and Developmental Disorders, 45(3), 778–794. https://doi.org/10.1007/s10803-014-2235-2

Callahan

Hughes

H. L.

Mehta

Toussaint

K. A.

Nichols

S. M.

P. S.

Wang

H.-T.

(2017). Social validity of evidence-based practices and emerging interventions in autism. Focus on Autism and Other Developmental Disabilities, 32(3), 188–197. https://doi.org/10.1177/1088357616632446

Cho

Y. I.

Johnson

T. P.

VanGeest

J. B.

(2013). Enhancing surveys of health care professionals: A meta-analysis of techniques to improve response. Evaluation & the Health Professions, 36(3), 382–407. https://doi.org/10.1177/0163278713496425

Cox

D. J.

(2012). From interdisciplinary to integrated care of the child with autism: The essential role for a code of ethics. Journal of Autism and Developmental Disorders, 42(12), 2729–2738. https://doi.org/10.1007/s10803-012-1530-z

Gerenser

J. E.

Koenig

M. A.

(2019). ABA for SLPs: Interprofessional collaboration for autism support teams (1st ed.). Paul H. Brookes.

Hamad

C. D.

Serna

R. W.

Morrison

Fleming

(2010). Extending the reach of early intervention training for practitioners: A preliminary investigation of an online curriculum for teaching behavioral intervention knowledge in autism to families and service providers. Infants and Young Children, 23(3), 195–208. https://doi.org/10.1097/iyc.0b013e3181e32d5e

Individuals with Disabilities Education Improvement Act, 20 U.S.C. § 1400 et seq. (2004)

Landa

R. J.

(2018). Efficacy of early interventions for infants and young children with, and at risk for, autism spectrum disorders. International Review of Psychiatry, 30(1), 25–39. https://doi.org/10.1080/09540261.2018.1432574

Luskin-Saxby

Paynter

(2018). An introduction to the knowledge-to-action strategic framework for disseminating evidence-based practice in autism spectrum disorder early intervention centers and its relevance to the National disability insurance scheme. Evidence-based Communication Assessment and Intervention, 12(4), 152–166. https://doi.org/10.1080/17489539.2018.1558953

10.

Miles

M. B.

Huberman

A. M.

Saldana

(2014). Qualitative data analysis: A methods sourcebook. SAGE.

11.

National Autism Center. (2015). Findings and conclusions: National standards project, Phase 2. National Autism Center.

12.

New Mexico Department of Health. (2013). Developmental specialist certification manual. https://www.nmhealth.org/publication/view/guide/3718/

13.

New York Department of Health. (2021). Service providers frequently asked questions. https://www.health.ny.gov/community/infants_children/early_intervention/service_providers/frequently_asked_questions.htm

14.

Paynter

J. M.

Ferguson

Fordyce

Joosten

Paku

Stephens

. . Keen

(2017). Utilization of evidence-based practices by ASD early intervention service providers. Autism, 21(2), 167–180. https://doi.org/10.1177/1362361316633032

15.

Paynter

J. M.

Sulek

Luskin-Saxby

Trembath

Keen

(2018). Allied health professionals’ knowledge and use of ASD intervention practices. Journal of Autism and Developmental Disorders, 48(2), 2335–2349. https://doi.org/10.1007/s10803-018-3505-1

16.

Plumb

A. M.

Plexico

L. W.

(2013). Autism spectrum disorders: Experience, training, and confidence levels of school-based speech-language pathologists. Language, Speech, and Hearing Services in Schools, 44(1), 89–104. https://doi.org/10.1044/0161-1461(2012/11-0105)

17.

QSR International. (2018). NVivo 12 [Computer software]. www.qsrinternational.com.

18.

QualtricsXM. (2018). Survey software. [Computer software]. https://www.qualtrics.com

19.

Ray

J. M.

(2010). Knowledge and confidence of speech-language Pathologists Regarding Autism 2015. (Doctoral Dissertation, University of North Texas). Available from Proquest Dissertations and Theses Global.

20.

Sam

A. M.

Cox

A. W.

Savage

M. N.

Waters

Odom

S. L.

(2020). Disseminating information on evidence-based practices for children and youth with autism spectrum disorder: AFIRM. Journal of Autism and Developmental Disorders, 50, 61931–61940. https://doi.org/10.1007/s10803-019-03945-x

21.

Schoonenboom

Johnson

R. B.

(2017). How to construct a mixed methods research design. KZfSS Kölner Zeitschrift für Soziologie und Sozialpsychologie/kzfss Cologne Journal for Sociology and Social Psychology], 69(Suppl. 2), 107–131. https://doi.org/10.1007/s11577-017-0454-1

22.

Schwartz

Drager

K. D.

(2008). Training and knowledge in autism among speech-language pathologists: A survey. Language, Speech, and Hearing Services in Schools, 39(1), 66–77. https://doi.org/10.1044/0161-1461(2008/007)

23.

Stahmer

A. C.

Collings

N. M.

Palinkas

L. A.

(2005). Early intervention practices for children with autism: Descriptions from community providers. Focus on Autism and Other Developmental Disabilities, 20(2), 66–79. https://doi.org/10.1177/10883576050200020301

24.

Steinbrenner

Hume

Odom

Morin

Nowell

Savage

M. N.

(2020). Evidence-based practices for children, youth, and young adults with autism. National Clearinghouse on Autism Evidence & Practice. https://ncaep.fpg.unc.edu/research-resources

25.

Texas Health and Human Services. (2021). Early intervention specialist IPDP credentialing. https://www.hhs.texas.gov/doing-business-hhs/provider-portals/assistive-services-providers/early-childhood-intervention-programs/training-technical-assistance/early-intervention-specialist-ipdp-credentialing

26.

Vivanti

Kasari

Green

Mandell

Maye

Hudry

(2018). Implementing and evaluating early intervention for children with autism: Where are the gaps and what should we do? Autism Research, 11(1), 16–23. https://doi.org/10.1002/aur.1900

27.

Wong

Odom

S. L.

Hume

K. A.

Cox

A. W.

Fettig

Kucharczyk

Schultz

T. R.

(2015). Evidence-based practices for children, youth, and young adults with autism spectrum disorder: A comprehensive review. Journal of Autism and Developmental Disorders, 45(7), 1951–1966.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB