Abstract
This article examines the recommended eligibility determination for learning disabilities (LD) in both the Individuals with Disabilities Education Improvement Act and the proposed changes in diagnostic criteria in the DSM-5. The focus is on the inclusion of the criterion of responsiveness to intervention (RTI) and the implications for practice of school psychologists and general and special education teachers. The research base on RTI for diagnostic purposes is examined, and considerations of changing roles for clinicians and school-based practitioners are discussed.
The proposed changes to the DSM-5 criteria for the identification of students with learning disabilities (LD) reflect an alignment with options for LD identification outlined in the 2004 reauthorization of the Individuals with Disabilities Education Improvement Act (IDEA) in the United States. Specifically, IDEA (2004) recommended that state and local agencies adopt an alternative method to the IQ–achievement discrepancy formula, such as response to intervention (RTI). Section 1414(b)(6)(B) of IDEA specifically stated, “In determining whether a child has a specific learning disability, a local educational agency may use a process that determines if the child responds to scientific, research-based intervention a part of the evaluation process.”
The changes to IDEA followed reports from the President’s Commission on Excellence in Special Education (2002) and the Donovan and Cross (2002) report that both emphasized the importance of a contextualized assessment process that considered a child’s opportunity to learn. The option of using RTI as a means to identify LD represented a shift in institutional priorities and was, as suggested by Johnston (2011), a response to four factors: lack of utility of the IQ–achievement discrepancy assessment for instructional planning, the growth in numbers of youth identified as LD, the overrepresentation of minority youth in special education, and a growing research base documenting early intervention effectiveness.
The inclusion of the option to measure a student’s responsiveness to intervention as a method for identifying LD outlined in IDEA was, in part, a response to criticism of the IQ–achievement discrepancy model for identification of LD (Fletcher, Lyon, Fuchs, & Barnes, 2007; Francis, Fletcher, & Steubing, 2005). Related to this, the recommendation for the proposed change for diagnostic criterion for LD in the DSM-5 that allows for the consideration of school-based reports of responsiveness to intervention was also based on critiques in the research literature regarding the utility of the IQ–achievement discrepancy (see Tannock, 2013) Thus, the proposed changes of diagnostic criteria for LD in the DSM-5 reflect congruence with legislative changes and greater alignment with options for LD identification as provided in IDEA (2004). The American Psychiatric Association (APA) specifically noted in the rationale for the proposed changes to diagnostic criteria for LD that the proposed
diagnostic criteria do not depend upon comparisons with overall IQ and are consistent in the USA’s reauthorized IDEA regulations (2004) which state that: “the criteria adopted by the State must not require the use of a severe discrepancy between intellectual disability and achievement for determining whether a child has a specific learning disability, as defined in 34 CFR 300.8 (c)(10).” (APA, 2012)
Proposed Changes to DSM-5 for LD Identification
As noted by Tannock (2013), the recommendation for the continued inclusion of the low achievement criterion for determining LD eligibility required the consideration of additional diagnostic criteria given the unsuitability of low achievement as sole criterion. Thus, the key issues considered by the LD work group in an examination of the validity of alternative diagnostic criteria for LD were RTI, psychological processing deficits, and a stringent standard (cutoff) for the identification of low achievement.
Specifically, the proposed recommendations for the DSM-5 diagnosis of LD note that diagnosis should be made by clinical synthesis of developmental, medical, family, and educational reports, which may include school-based reports of test scores and observations, including RTI. The proposed diagnostic criteria include the following:
History or current presentation of persistent difficulties in the acquisition of reading, writing, arithmetic, or mathematical reasoning skills during the formal years of schooling (i.e., during the developmental period). The individual must have at least one of the following:
Inaccurate or slow and effortful word reading Difficulty understanding the meaning of what is read (e.g., may read text accurately but not understand the sequence, relationships, inferences, or deeper meanings of what is read) Poor spelling (e.g., may add, omit, or substitute vowels or consonants) Poor written expression (e.g., makes multiple grammatical or punctuation errors within sentences, written expression of ideas lack clarity, poor paragraph organization, or excessively poor handwriting) Difficulties remembering number facts Inaccurate or slow arithmetic calculation Ineffective or inaccurate mathematical reasoning Avoidance of activities requiring reading, spelling, writing, or arithmetic
Current skills in one or more of these academic skills are well below (the current recommendation is 1.5 standard deviations) the average range for the individual’s age or intelligence, cultural group or language group, gender, or level of education, as indicated by scores on individually administered, standardized, culturally and linguistically appropriate tests of academic achievement in reading, writing, or mathematics.
The learning difficulties are not better explained by Intellectual Developmental Disorder, Global Developmental Delay, neurological, sensory (vision, hearing), or motor disorders.
Learning difficulties identified in Criterion A (in the absence of the tools, supports, or services that have been provided to enable the individual compensate for these difficulties) significantly interfere with academic achievement, occupational performance, or activities of daily living that require these academic skills, alone or in any combination (APA, 2012).
The inclusion of the proposed criterion D that states that learning difficulties (as identified in criterion A) must be present in the “absence of tools, supports, or services” suggest that documentation of intervention or support as well as some measure of student responsiveness should also be considered in the synthesis of developmental, medical, family, and educational reports as these reports may “include school-based reports of test scores and observations, including responsiveness to intervention.” Therefore, together, Criteria A and D suggest that school-based RTI data may be an integral part of the proposed changes to criteria as the documentation of school-based interventions and measures of student school-based progress could be necessary data to be considered in the “clinical synthesis” of reports for diagnosis. These legislative changes and proposed changes to DSM-5 of diagnostic criteria for LD identification have implications not just for clinicians but also school-based practice of school psychologists and both general and special education teachers implementing RTI.
Response to Intervention Framework
RTI has been defined as a systematic general education model characterized by research-based tiered interventions tailored to meet the needs of individual children (Mellard, 2004). One assumption behind RTI was that the instructional accountability found in empirically validated curriculum and interventions would address student learning needs prior to academic failure and help to ensure that poor academic achievement that resulted in referrals to special education was not a result of poor instruction. The National Association of State Directors of Special Education articulated that RTI implementation should be based on the primary assumptions that (a) all students can learn in the general education classroom, (b) early intervention is critical to preventing problems from getting out of control, (c) the implementation of a multitiered service delivery model is necessary, (d) a problem-solving model should be used to make decisions between tiers, (e) research-based interventions should be implemented to the extent possible, (f) progress monitoring must be implemented to inform instruction, and (g) data should drive decision making.
The U.S. National Research Center on Learning Disabilities (NRCLD) recommended core features of RTI. These features were universal screening, high-quality classroom instruction that is research based, continuous progress monitoring of students’ classroom performance, research-based interventions if progress is not being made, and the use of fidelity measures for instruction and interventions (Johnson, Mellard, Fuchs, & McKnight, 2006). Proponents of RTI have recommended the use of a dynamic model built around the documentation of students’ response to research-based instructional interventions in general education, primarily in reading (Fuchs & Fuchs, 2007). The tiered model of RTI is designed to provide intensified evidence-based interventions and support based on students’ responsiveness to instruction as identified through systematic progress monitoring.
All states responded to the IDEA (2004) policy change allowing RTI models to be used as an alternative or supplement to discrepancy models for the identification of LD. By 2008, all states had reported a timeline and state plan for RTI implementation (National Center on Response to Intervention, 2008). The U.S. Department of Education reported in July 2011 that 37 states reported using the IQ–achievement discrepancy model and RTI for LD identification, six states reported using the discrepancy model but only with the inclusion of RTI data prior to comprehensive assessment that includes IQ testing, and seven states had mandated RTI and disallowed the use of the discrepancy model for LD determination (National Center for Education Evaluation and Regional Assistance, 2011). Although features of RTI have been recommended, there is no federal guidance or consensus related to how to best implement RTI as a means to identify youth with LD. This variability in RTI model features (especially related to assessment procedures across the tiers) has raised some concern among researchers. Hughes (2008) noted a key issue in particular was that the degree of evaluation at each tier varies across models.
RTI as a Method of Identification: Emerging Issues
As noted by Tannock (2013), researchers have not reached consensus regarding the utility of RTI as an alternative model of LD identification (Johnson, Mellard, & Byrd, 2005; Mastropieri & Scruggs, 2005; McKenzie, 2009; Swanson, 2008). Although there has been much support for the universal screening and instructional approach implemented within RTI models as an early intervention prior to any referral for special education assessment (e.g., Fletcher, Denton, & Francis, 2005; Fuchs, 2003; Yssledyke, Burns, Scholin, & Parker, 2010), there has been much discussion centered around issues related to the primary or exclusive use of an RTI approach in LD identification. The issues include concerns related to measurement error and threats to validity, a need for research-based interventions across multiple academic domains, measurement of treatment fidelity, a lack of standardized criteria for determining nonresponsiveness, and methods for culturally responsive instruction for linguistically and culturally diverse students within standardized RTI protocols (Burns, Jacob, & Wagner, 2008; Klingner & Edwards, 2006; McKenzie, 2009; Reynolds, 2008).
RTI Empirically Validated for LD Identification?
Researchers in both psychology and education are consistent in the support of RTI as an effective early intervention model but are also in agreement on issues and concerns related to the use of RTI models as a means of identification of LD (Brieger & Majovski, 2008; Burns et al., 2008; Fletcher, Denton, et al., 2005; Reynolds & Shaywitz, 2009; Swanson, 2008). Brieger and Majovski (2008) noted, “RTI as an approach to providing intervention . . . can be highly valuable and can advance the improvement of instructional practice [but] as a method of diagnosis of an individual child, it is flawed” (p. 150). Burns et al. (2008) critiqued the use of RTI data for formal eligibility decisions as this practice represented a use of data that is different from what they had previously been validated for. Concerns related to the variability of assessment measures utilized for universal screening (e.g., Walker-Dalhouse et al., 2009) and progress monitoring include concerns related to consistency of predictive validity of assessments across diverse subgroups (e.g., Hosp, Hosp, & Dole, 2011) and the lack of an empirical base of curriculum-based measures to determine RTI (Ball & Christ, 2012; Reynolds & Shaywitz, 2009) or their reliability and validity for initial diagnosis (e.g., Kemp & Korkman, 2008).
A concern raised related to the inclusion of RTI as criteria for LD identification in the proposed changes to the DSM-5 is that this inclusion of the RTI option does not address or solve the issues around LD identification noted above. However, the diagnostic criteria proposed in the DSM-5, while including RTI methods, do also include a recommendation that academic skills be substantially and quantifiably below those expected as based on standardized measures, thus potentially ruling out RTI as a sole method for identification (if the RTI model does not include standardized measures of performance), and maintain the DSM-IV rule-out criterion of intellectual disability. However, many experts in the field also suggest that the appropriate operationalization of a LD should include a presentation of cognitive processing deficit (Johnson, Humphrey, Mellard, Woods, & Swanson, 2010; Reynolds, 2008), and the proposed changes to the DSM-5 diagnostic criteria do not specify assessment for cognitive processing. Tannock (2013) discusses the mixed support for the inclusion of cognitive processing deficits in the diagnostic criteria for LD found in the research literature (e.g., Fuchs et al., 2012; Johnson et al., 2010) that informed the exclusion of the diagnostic criterion of cognitive processing deficit. Rather, the proposed criteria remove the IQ–achievement discrepancy requirement (in alignment with the literature) and add nonresponsiveness to RTI and a measure of academic skills that are “well below” (the current recommendation is 1.5 SD) the average range for the individual’s age as indicated by scores on individually administered standardized tests of academic achievement in reading, writing, or mathematics. Therefore, although the proposed criteria do not rule out the use of cognitive processing assessment, they provide no guidelines for inclusion or interpretation as suggested by Johnson et al. (2010).
In sum, the inclusion of the RTI model in the proposed DSM-5 criteria is aligned with legislative changes in the U.S. (IDEIA); however, the questions (e.g., predictive validity of progress-monitoring measures, criteria for measuring responsiveness, and methods for measuring fidelity of intervention delivery) related to the validity of RTI for diagnostic purposes remain. If RTI is to be used as even one component of the eligibility process, these issues must be addressed. In the meantime, the large-scale implementation of RTI has already begun in the United States; therefore, effective implementation of RTI models for early intervention or any stage of pre-referral assessment will require appropriate training for implementation and clearly defined roles for school psychologists, interventionists, and teachers that emphasize effective methods of collaboration.
School-Based Practice: Changing Roles
The proposed change to DSM-5 criteria for LD identification that include the consideration of RTI has implications for practice of both clinicians and school-based practitioners who must navigate a collaborative model for interdisciplinary decision making for the “clinical synthesis of developmental, medical, family, and educational reports” (APA, 2012) to be used in the eligibility determination that may occur in either schools or clinical settings. Hosp et al. (2011) emphasized the importance of consistency in decision making for both service delivery and eligibility determinations within RTI models. Yet, as noted, one of the criticisms of RTI as a method for LD identification is the variability of assessment measures and criteria for decision making related to responsiveness to instruction and intervention. In addition, changing roles of school psychologists and teachers in RTI systems must be clarified to support effective decision making (Fletcher, Francis, Morris, & Lyon, 2005). Mastropieri and Scruggs (2005) highlighted challenges RTI presents for changing roles of teachers, diagnosticians, and school psychologists and noted the lack of clarity in most RTI models related to outlined responsibilities for instructional delivery, intervention selection and implementation, progress monitoring of RTIs, protocols for movement of students among tiers, and diagnosis. Yet these challenges remain, and guidance for implementation of an RTI framework that requires emphasis on selection of appropriate screening and progress monitoring assessments for diverse students, instructional delivery, appropriate selection of “evidence-based” intervention(s), and the ability to effectively access personnel and school resources and collaborate systematically with other school professionals remains variable across schools and districts implementing RTI (e.g., Cavendish, Harry, Menda, Espinosa, & Mahotiere, 2012; McKenzie, 2009). The need for increased collaboration in eligibility determination is aligned with the guiding principles of the National Joint Committee on Learning Disabilities (NJCLD) for comprehensive assessment and evaluation that include the stated principle that
Professionals with expertise in learning disabilities are necessary to conduct a comprehensive assessment and evaluation system for students suspected of having learning disabilities. These professionals from various disciplines must make up a multidisciplinary team along with the family and the student (as appropriate). To make identification and eligibility decisions, the team must possess the range of competencies necessary for evaluation and identification. (NJCLD, 2010)
Unfortunately, most state RTI plans and manuals for implementation do not provide guidance related to specific roles of clinicians or practitioners (e.g., Berkeley, Bender, Peaster, & Saunders, 2009; Cavendish et al., 2012; Jones, Yssel, & Grant, 2012). Therefore, districts and schools implementing RTI must explicitly identify the roles and responsibilities of general and special educators, school psychologists, reading specialists, and interventionists. Defined roles and responsibilities in an RTI system will likely emphasize an increased need for training in collaboration across the silos of general education, special education, and school psychology.
These changing roles and the need for greater collaboration also suggest a likelihood of increased training and professional development needs for school and clinical practice. Effective collaboration in educational decision making should consist of interaction between professionals with different areas of expertise yet who share goals and responsibilities (Murawski & Hughes, 2009), and parity is critical (Friend & Cook, 2007). As the legislative change to IDEA (2004) has already led to the inclusion of RTI models in some aspect of the identification process in most districts and schools, the inclusion of RTI among the diagnostic criteria in the proposed changes to the DSM-5 is likely to permeate clinical practice in much the same way.
Therefore, teacher preparation and school psychology programs must provide not just instruction related to collaborative models but also opportunities for practice in field-based experiences in schools and as part of the school psychology internship experience (Hawkins, Kroeger, Musti-Rao, Barnett, & Ward, 2008). Hawkins et al. (2008) developed and evaluated a preservice interdisciplinary RTI-focused program that included contextualized (in general education) RTI field experiences for school psychology and special education students. The program was developed to integrate previous and concurrent course content related to “(a) collaboration and consultation; (b) elements of RTI: evidence-based prevention, instruction, and intervention; and (c) data-based decision making (including technical adequacy)” (p. 747). Hawkins et al.’s findings related to this integrated program support interdisciplinary training using a systematic approach, the need for both general and special education preservice teachers to be trained in field-based and data-based team decision making, and preservice programs’ inclusion of explicit instruction within assessment-related coursework that includes methods for determining intervention strength. Thus, the need for collaboration in data-based decision making and “teaming” within RTI should inform both school psychology and preservice teacher training and professional development for general and special education teachers (Nellis, 2012).
Conclusions
The in-depth review of the research literature and consideration of professional organizations and stakeholder input as part of the process for developing the proposed changes to the DSM-5 criteria for the identification of LD are laudable. In addition to the considerations of antecedent, concurrent, and predictive validators, this review has allowed for the consideration of a greater research base on RTI model development and implementation since the authorization of IDEIA in 2004 given the marked increase in RTI model development and research on RTI processes since that time. Thus, the time frame for the scholarly discourse around the proposed changes is advantageous for the consideration of research on the utility of RTI as part of the evaluation process. Although RTI has not been shown to be a valid diagnostic paradigm, the consideration of the research base on effective early intervention contributed to the recommendation that RTI be included as one criterion for LD identification (Tannock, 2013). The proposed changes to the DSM-5 do not limit LD identification to RTI methods but provide for the inclusion of both standardized measures and RTI in the proposed diagnostic criteria for LD that do not limit assessment or evaluation options. Although the inclusion of RTI does not provide clarity or a decisive framework for clinicians and practitioners in schools to navigate the operational definition of LD and the selection of the related appropriate assessments within RTI frameworks, the retention of the DSM-IV low achievement criterion with the addition of “standardized measurement” and the quantification of low achievement as 1.5 standard deviations below the population mean for age do provide a constructive framework on which practitioners and clinicians can base their collaborative decision making for eligibility. The proposed changes to the DSM-5 criteria would not be expected to provide resolution to the long-standing debate in the field related to the operational definition of LD, but rather they provide for the inclusion of multiple diagnostic criteria with demonstrated support (if not consensus) in the field.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
