Specific learning disabilities in children: Evaluation

INTRODUCTION — Learning disabilities (LD) are a heterogeneous group of disorders characterized by the unexpected failure of an individual to acquire, retrieve, and use information competently. They are the most severe, pervasive, and chronic form of learning difficulty in children with average or above-average intellectual abilities [1,2].

LD have a multifactorial etiology [3]. They typically manifest as a failure to acquire reading, writing, or math skills at grade- and age-expected levels. Learning problems that are outside of these traditional core domains, such as memory problems, attention problems, and difficulty managing social interactions, are not typically considered to be LDs. However, they may affect reading, writing, and math and may also require intervention.

The evaluation of children with LD will be presented here. The clinical features, management, and prognosis of LD and the role of the primary care provider are discussed separately. (See "Specific learning disabilities in children: Clinical features" and "Specific learning disabilities in children: Educational management" and "Specific learning disabilities in children: Role of the primary care provider".)

Educational definitions for LD and a review of special education law, which describes how students can access special education services in school settings, also are provided separately. (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States".)

TERMINOLOGY — In this topic review, the term "learning disability" (or "specific learning disability") refers to a heterogeneous group of disorders characterized by the unexpected failure of an individual to acquire, retrieve, and use information competently.

However, in some countries, the term "learning disability" is used to refer to intellectual disability (formerly "mental retardation"). Intellectual disability is discussed separately. (See "Intellectual disability (ID) in children: Clinical features, evaluation, and diagnosis" and "Intellectual disability in children: Evaluation for a cause" and "Intellectual disability (ID) in children: Management, outcomes, and prevention".)

DEFINITION — The core feature of a specific learning disability (LD) is an intrinsic cognitive difficulty that results in academic achievement in reading, writing, or math at a level less than expected for the individual's intellectual potential [4,5]. Several definitions of LD agree with this core concept (table 1) [6-8]. Despite conceptual agreement, the current definitions of LD are not specific enough for consistent case definition. (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Educational definitions'.)

OVERVIEW — Early identification of learning disabilities (LD) is crucial to providing interventions to optimize learning and prevent secondary emotional problems [4,9]. Because early recognition may affect ultimate outcome, pediatric health care providers should have a low threshold for considering LD in children who are at risk and those who have problems at school (whether the problems are in academic achievement, behavior, attention, or social interaction). (See "Specific learning disabilities in children: Clinical features", section on 'Risk factors'.)

LD usually are identified when the child fails to achieve academic milestones alongside his or her peers. Any child who is failing at school who has had adequate educational exposure should be presumed to have an LD until proven otherwise. Qualitative observations and/or the child's report card often are sufficient to identify learning failure, but psychometric testing is used to make a formal diagnosis. (See 'Comprehensive evaluation' below.)

The formal definitions of LD (table 1) do not specify which psychometric tests to use to identify an LD; which of the specific skills within reading, writing, or math need to be impaired to qualify for LD; what cutoff scores to use on the psychometric tests (ie, how many standard deviations below the mean should be used to indicate LD); or how qualitative observations of the child's performance should be used to supplement the standard measure findings. The identification of LD is thus left to the individual tester, who needs to synthesize information from more than one source to draw a firm conclusion. As a result, there is significant variability in the diagnosis of LD in research studies, school districts, and states of the United States. This variability affects which students are included in research studies. It also affects which students are identified by schools as being in need of services [10-13]. (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Terminology' and "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Educational definitions'.)

School districts are legally required to use information collected from the student's educational history (eg, report card), from classroom observations, and from standardized measures. However, other factors can influence decisions about eligibility for services. These factors further add to the variability in identification of LD, which has not changed substantially over time [14-16]. (See 'Comprehensive evaluation' below.)

INITIAL PRESENTATION

Parent concerns — Children with learning disabilities (LDs) frequently present to their health care provider when the parent expresses concerns about the child's learning failure. Such concerns should always be taken seriously. Parental concerns are a sensitive indicator of developmental and learning problems [17-20]. The pediatric health care provider can help the family by informing them of their right to evaluation/testing through the child's school district and by providing access to resources that help families of children with disabilities (table 2) [21-23]. In so doing, the pediatric clinician empowers the family to find ways to identify the LD and to access services on their own. (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Laws affecting the education of students with disabilities' and 'Resources' below.)

Teacher concerns — Learning problems usually are first identified by teachers and/or become evident at the time that the student's report card is issued. The teacher may then initiate referral for a special education evaluation. The teacher's decision to make a referral depends upon his or her impression of the student's needs (eg, the amount of explicit instruction and type of classroom accommodations that are necessary for the student to succeed) [24].

Teachers typically are required to use the "prereferral" system, or the Student Support Team (the name of the team varies by state), before making a referral for a formal special education evaluation. The prereferral team may make suggestions to the teacher or may decide to provide remedial services (eg, using a "responsiveness to intervention" model) to help improve the student's performance before a decision is made to conduct a multidisciplinary evaluation. (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Every Student Succeeds Act'.)

Teachers are highly accurate in identifying children with LDs and in differentiating students with learning problems from students with behavior disorders or low overall cognitive ability [25,26]. However, general education teachers may not always request a prereferral team meeting or an evaluation when it is needed. Teacher nonreferral may occur for a number of reasons. Teachers may not refer a student if the prereferral team is not perceived to be helpful, if the referral results in an increased work load for colleagues who already have too many students to serve, or when the teacher has been discouraged from making referrals because of a lack of resources for special education in that district [27]. (See 'Determination of service eligibility' below.)

Clinician concerns — Pediatric clinicians are not likely to identify an LD unless the child is already known by others to have learning problems and/or the child's parents express concerns about the child's failure to learn. However, because early recognition may affect ultimate outcome, pediatric health care providers should be quick to consider LD in children who are at risk and children who have problems at school (whether the problems are in academic achievement, behavior, attention, or social interaction). (See "Specific learning disabilities in children: Clinical features", section on 'Risk factors' and "Specific learning disabilities in children: Role of the primary care provider", section on 'Early identification'.)

Specific historical information and physical examination features can help confirm the need for formal evaluation. (See "Specific learning disabilities in children: Role of the primary care provider", section on 'Medical evaluation'.)

When formal evaluation is needed, there should be a low threshold for requesting a "prereferral" or "Student Support Team" evaluation (the name of the team varies by state). (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Every Student Succeeds Act'.)

In schools that do not have an effective prereferral system, it may be necessary to make a request for a formal evaluation by the special education team. Parental requests for such an evaluation have as much legal weight as a request from the primary care provider. However, the primary care provider's input is valuable, particularly if the provider includes historical information or pertinent risk factors that indicate to the school team why the request for an evaluation is justified [28].

ROLE OF THE PRIMARY CARE PROVIDER — The role of the primary care provider in the evaluation and management of the child with learning disabilities is discussed separately. (See "Specific learning disabilities in children: Role of the primary care provider".)

EVALUATION AND IDENTIFICATION OF LD IN SCHOOL SETTINGS — Special education law requires that school teams conduct a "comprehensive" evaluation of "all areas of suspected disability" in a student with learning difficulty [29], including:

●Intellectual ability

●Academic achievement; this can include an evaluation of the student's vocational potential

●Speech/language skills

●Occupational and/or physical therapy assessment to evaluate gross and fine motor skills

●Psychologic and/or emotional factors that may interfere with learning

●Medical or neurologic conditions that may require accommodations and/or treatment

The areas of suspected disability are identified by conducting a review of the student's developmental skills and weaknesses. This can be done by reviewing the student's report card, by reviewing responses to teacher rating scales, by conducting classroom observations, and/or by discussing the student's performance with the classroom teacher. Based on the information gathered, a decision is made about which areas require a formal evaluation. The school team should consider all of the possible causes for learning failure (including associated behavioral and mental health conditions). The team may need to conduct formal evaluation in several areas to determine the student's needs.

In schools, the "comprehensive evaluation" can be conducted by any or all of the following team members [30]:

●Educator (general and special education)

●Psychologist

●Social worker

●School nurse

●Speech pathologist

●Occupational or physical therapist

●Guidance

●School administrator

In some circumstances, schools use the services of a consulting primary care provider, child psychiatrist, or developmental-behavioral pediatrician to conduct part of the school-based evaluation and to identify specific medical or mental health conditions that may be a focus of treatment and/or affect learning. More commonly, the input from clinicians outside the school system is requested by the student's parents and is based upon an office-based evaluation, with varying amounts of information obtained from the school. The clinic-based clinician can contribute to the evaluation by signaling to the family and school team any mental health, neurologic, genetic, or other medical condition that may be related to the learning disability or affect the student's overall functioning.

COMPREHENSIVE EVALUATION

Objective — The purpose of the comprehensive evaluation is to assess the student's areas of need and determine eligibility for services under education and/or disability rights laws. The assessment should provide a comprehensive review of the student's learning weaknesses and strengths. It should include a review of the basic skills of reading, writing, and math (each of which consists of several subcomponents or subskills). It should also review the student's speech and language skills, attention span (ie, working memory, impulse control, etc), capacity to integrate information from multiple sources, and, where applicable, motor skills. Emotional factors and capacity for social interactions also must be considered.

Diagnostic process — An accurate diagnosis of learning disability (LD) depends upon a synthesis of three types of information: history of the learning problem (educational history), classroom observations, and performance on a standardized or psychometric measure.

The diagnosis LD is made primarily by history. Classroom observations and psychometric measures help to confirm the presence of LD and identify targets for intervention. The criteria for identification of an LD are not explicitly defined by special education law or LD researchers, although the literature provides some opinions [31,32]. In practice, the choice of psychometric measures, the types of subskills assessed by the measure, the cutoff scores used in psychometric measures, and the type of qualitative information used to make the diagnosis are all determined by the individual researcher, state, or school district and can be quite variable [33].

A psychometric measure, on its own, does not always identify LD successfully. Special education law (ie, the Individuals with Disabilities Education Act [IDEA]) encourages the use of clinical or educational judgment in addition to using test scores in the identification of children with LD [31,32]. Although qualitative information may or may not be used to identify children with LD for research purposes, both qualitative and quantitative measures should be used to identify students in need of services in the school setting [31]. Qualitative measures, such as the student's performance in the classroom setting (classroom observations) and a review of the student's educational history, help to put test performance (psychometric measures) into context.

The validity of the identification and service needs of children with LD is increased when information from each of the following sources is included:

●Standardized psychometric measures (quantitative information).

●Review of the student's educational history (eg, report cards/grades over time, grade retention); important aspects of the educational history include the child's access to education and the quality of education provided. Absenteeism related to illness, family stress, anxiety, or other problems reduces exposure to instruction and can result in poor academic achievement. Poor quality instruction may result from being in a disruptive class, staffing changes, large class size, and factors related to the teacher. (See "Specific learning disabilities in children: Clinical features", section on 'Differential diagnosis'.)

●Description of classroom observations (eg, participation behaviors, success in the completion of classroom and home assignments, etc).

Ideally, the identification also includes an assessment of student performance during responsiveness to intervention (RTI) services. RTI and standardized measures are not mutually exclusive; they should be used together. By including regularly collected student data during delivery of RTI, the influence of measurement error at a single point in time is reduced. The method for identifying LD using data collected through RTI is uncertain and is highly variable across the United States [34].

Psychometric tests — Psychometric tests provide standardized procedures for measuring knowledge and abilities in various areas (eg, reading, writing, math, executive functions, language skills, etc). Psychometric measures are used to estimate how the student will perform in real-time situations. On their own, psychometric measures are not "proof" of the presence or absence of a skill or set of skills and therefore also are not proof of the presence or absence of LD. Psychometric measures can sometimes uncover areas of disability that were not suspected from the qualitative review of the student's performance; however, psychometric measures should always be compared with the student's performance in the classroom setting and the student's performance over time [13]. This is the only way to verify that the student's performance during testing corresponds to the student's performance in classroom settings. Some commonly used tests are listed in the table (table 3).

Psychometric measures are used in conjunction with other information to define LD in the school setting. The manner in which psychometric tests are used may vary from school district to school district.

Formulas used to identify LD — A number of models may be used to identify students with LD in the school setting with psychometric measures [31]. These models include discrepancy formula, low achievement, intraindividual differences, and responsiveness to intervention [31]. The psychometric measure does not, on its own, identify LD.

●Discrepancy formula model – When using the discrepancy formula model, the evaluator determines whether there is a discrepancy between the student's overall intelligence and his or her academic achievement. The discrepancy formula relies on the use of a measure of intelligence (eg, an intelligence quotient [IQ] test) to measure the student's "ability," which is then compared with the student's "achievement" in a particular academic area. For example, a student may have an IQ score of 100 but have a score of 80 in a specific academic domain. This (statistically significant) discrepancy indicates the presence of an LD in that academic domain, because the student's IQ (ability) would predict an achievement score of 100.

When a discrepancy is present, a discrepancy formula can facilitate identification of LD. However, the absence of a discrepancy between intelligence and achievement is not proof of the absence of an LD.

The discrepancy model is flawed because the presence of the LD can lower the student's performance on the IQ measure. This reduces the validity of the comparison between IQ and achievement [35]. In addition, the standard error of measurement increases when two tests are compared, particularly when the two tests have not been standardized on the same population [36]. Thus, when a test of intelligence and a test of educational achievement are compared with one another, the validity of the comparison is reduced if the two tests have not been standardized on the same population. Intelligence and achievement tests that are standardized on the same population include the Wechsler Intelligence Scale for Children (WISC) and the Wechsler Individual Achievement Test (WIAT); and the Woodcock Johnson Tests of Cognitive Abilities and the Woodcock Johnson Tests of Achievement. Using these paired test instruments reduces measurement error when a discrepancy formula is used to define LD [36].

Partly because of the limitations discussed above, the 2004 revisions to the IDEA stipulate that a discrepancy formula cannot be used as the sole means of identifying a learning disability [29]. (See "Definitions of specific learning disability and laws pertaining to learning disabilities in the United States", section on 'Individuals with Disabilities Education Act'.)

●Low achievement model – In low achievement models, the student's overall educational performance on standardized tests is compared with national or local norms. When the student's performance is "significantly" lower than the performance expected by national or local norms, she or he may be identified as having an LD. What constitutes a "significant" difference is determined locally. This "difference" can vary between less than one standard deviation (SD) below the mean (eg, standard score of 90, or 25^th percentile) to two SDs below the mean (eg, standard score of 70, or 3^rd percentile).

●Intraindividual differences model – When using intraindividual differences models, the evaluator determines whether the student's learning profile (on whatever cognitive or academic measure was used) demonstrates varying strengths and weaknesses (also known as "subtest scatter"). Intraindividual differences are not a sensitive indicator of the presence or absence of an LD [31]. However, if confirmed by qualitative observations, intraindividual differences provide important information about the student's profile of learning strengths and weaknesses.

●Responsiveness to intervention (RTI) model – When using an RTI model, the evaluator determines the student's progress when she or he is provided with specific educational interventions, such as more intense instruction in reading decoding, more small group instruction in general, etc. However, the optimal method of monitoring the student's progress with RTI is not clear [31].

Limitations of psychometric measures — Psychometric measures may provide useful information about the student's achievement and abilities. However, they have a number of limitations and should be used in conjunction with clinical and educational judgment when identifying students with LD who are in need of services. The limitations of psychometric measures are described below.

●Identification of LD in young students – Psychometric tests are less reliable for identifying LD in students in first or second grade (or younger) than in older students. Students in the early grades have not participated in the formal education process long enough to demonstrate learning delays on standardized measures as clearly as older students. This is especially true when LD is defined on the basis of a discrepancy formula [37]. (See 'Formulas used to identify LD' above.)

●National versus local norms – The population with which a student's performance on a standardized test is compared may affect whether or not the child is identified as having an LD [38]. Some school teams use standardized tests with national norms, and others use tests with local norms or may weigh their decisions more heavily on one set of norms than the other. In other words, students who are not learning successfully in a school district with high educational standards are more likely to be identified as having LD. Students who are not learning successfully in a school district with low educational standards are less likely to be identified as having LD.

●Performance during testing versus performance in the classroom – The student's performance on a psychometric measure can vary from one evaluator to the next and may not reflect the student's actual performance in the classroom. The context of the testing situation differs from that in the classroom, with the testing situation typically being more supportive.

For example, in a test situation, an evaluator is likely to speak more slowly, has more time to assure the student's comprehension before asking the student to complete a task, and clearly sets up the testing environment to assure correct execution of tasks. In so doing, the evaluator can circumvent language weaknesses or executive dysfunctions that would normally interfere with the student's performance in a classroom setting.

A thoughtful evaluator makes comments on any improvements noted in the student's performance in the testing situation and describes the strategies used during the testing situation that may have helped to enhance the student's performance. Although it can be difficult to quantify what strategies were the most helpful, recommendations such as this should be made to the classroom teacher to assure optimal performance in the classroom setting and/or to advocate for eligibility for services.

●Arbitrary cutoff scores – Standardized measures use test scores to determine the presence or absence of disability. Scores are reported in different ways. The two most commonly reported scores are the standard score (average performance score = 100) or percentiles (average performance score = 50^th percentile). The cutoff scores used to define an LD on psychometric tests are chosen somewhat arbitrarily. In any case, given the standard measurement error inherent in any test and performance variation over time, scores close to the cutoff are suspect [13]. The learning performance and treatment needs of students just above or just below cutoff scores are not different [39].

Evaluators may use a self-selected cutoff score on standardized test instruments [12] or the cutoff criteria determined by school district or state. The selected cutoff scores vary. A very strict definition of impairment may be defined as performance that is two (SDs below the mean (ie, performance below a standard score of 70, or below the 3^rd percentile), whereas a more generous definition of impairment may be defined as a standard score below 90, or below the 25^th percentile, which is between the mean and one SD below the mean. In observational studies, scores below the 25^th percentile correlate with teacher and parent perceptions of learning problems at school [35]. However, a cutoff of one SD below the mean (ie, standard score of below 85, or below the 16^th percentile) is commonly used to define LD in school settings. The use of different cutoff scores may account for different conclusions between evaluators.

●LD may affect overall performance – A student's overall performance on a measure of cognitive skills (eg, IQ score) is affected by the student's neurodevelopmental weaknesses in one area or another. Thus, the fact that a student has an LD, or has executive dysfunctions, can affect his or her global cognitive score. The global cognitive performance of the student cannot be measured separately from any LD that is affecting the student's performance. As such, the discrepancy model under-identifies students with LD and, on its own, cannot be relied upon to identify students with LD. For this reason, the IDEA no longer allows the discrepancy formula as a sole criterion for the identification of LD. (See 'Formulas used to identify LD' above.)

●Lack of diagnostic stability – Diagnostic stability is one measure of the validity of a psychometric measure: If the LD persists over time, the psychometric measure may be considered more valid. However, LD is quite unstable over time and varies by LD type and by study. Estimates of diagnostic stability are too wide to be clinically meaningful, ranging from 30 to 70 percent [40,41].

The wide variation is related to the age at diagnosis, different measures used to define LD, intervention/treatment effects, and changes in LD expression over time [42-44]. As an example, reading LD usually first manifests as difficulty with decoding in grades 1 and 2. As the student ages into higher grades (eg, grades 3 and 4), reading LD may manifest as difficulty with reading fluency or reading comprehension. Thus, the degree of impairment in reading decoding, reading fluency, and reading comprehension varies by age, but also varies by treatment effects. A student with reading LD can improve his or her performance in all three areas – reading decoding, reading fluency, and reading comprehension – if remediation is offered for each of these areas. Given these changes in the expression of LD, psychometric measures at a single point in time are relatively unreliable when they are used as the sole criterion for the identification of LD.

Determination of service eligibility — In order for a student with LD to be eligible for special education services under the IDEA, the student must be diagnosed with LD, the disability must "adversely affect the child's education," and the child must "require specialized instruction and related services." School districts vary in how they define the "adverse impact" of the student's disability on the student's education. The school district may emphasize educational history and classroom observations as much as the psychometric measure in their eligibility determination or may emphasize the psychometric measure alone. School districts also use varying psychometric measure cutoff scores to define the degree of adverse impact. These differences help to explain variable identification patterns of LD and variable service delivery patterns across school districts and states.

In addition to the many caveats listed above related to the identification of LD (see 'Limitations of psychometric measures' above), children with LDs may be inappropriately excluded from services (ie, not found eligible for services). Factors that affect eligibility determination include the following:

●Availability of services – Availability or lack of availability of services can both influence the identification of students in need of services. When services are available, students may be over-identified, whereas when services are not available, students may be under-identified [38,45].

●Team dynamics – The identification of LD in schools is a team decision. As is true for any team, the opinions of some members may be given greater weight than the opinions of others (eg, a staff member who also serves a supervisory role) and may have extra influence on the decision-making process. Also, a staff member who provides remedial services may be more likely or less likely to identify a student as eligible, depending in part upon how the decision will affect her or his overall caseload.

●The degree of involvement of the student's parents in the identification process and whether or not family members have prior relationships with the team members.

●The degree to which the teachers see themselves as being personally responsible for ensuring the success of the student.

Factors such as those listed here can affect the consistency with which an LD is identified and the consistency with which the student is identified as being eligible for services, both within and between school teams [46,47].

RESOURCES — Resources for health care providers, patients, and families of children with learning disabilities are provided in the table (table 2).

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient education" and the keyword[s] of interest.)

●Basics topic (see "Patient education: Learning disabilities (The Basics)")

SUMMARY

●The core feature of a specific learning disability (LD) is an intrinsic cognitive difficulty that results in academic achievement at a level less than expected for the individual's intellectual potential. (See 'Definition' above.)

●Early identification of LD is crucial to providing interventions to optimize learning and prevent secondary emotional problems. Any child who is failing at school should be presumed to have an LD until proven otherwise. (See 'Overview' above.)

●The primary care provider should consider the possibility of LD if there is parental concern about any school-related problem (eg, behavioral regulation, poor peer interactions, learning problems); there is a family history of learning problems; or the child has a behavioral/mental health condition, developmental delay, or a neurologic or genetic condition that is associated with LD. (See 'Initial presentation' above and "Specific learning disabilities in children: Role of the primary care provider", section on 'Early identification'.)

●The diagnosis of LD is made primarily by history. Psychometric measures help to confirm the presence of LD and identify targets for intervention. An appropriate assessment for LD includes information from standardized psychometric measures, a review of the student's educational history, and a description of classroom observations. These three components are required for accurate identification and description of the LD and are also required by the Individuals with Disabilities Education Act (IDEA). In addition, schools can use data collected from responsiveness to intervention services or by following the student's progress over time. (See 'Diagnostic process' above.)

●In order for a student with LD to be eligible for special education services under the IDEA, the student must be diagnosed with LD, the disability must "adversely affect the child's education," and the child must "require specialized instruction and related services." School districts vary in how they define the "adverse impact" of the student's disability on the student's education. This helps to account for variable identification patterns of LD, and variable service delivery patterns, across school districts and states. (See 'Determination of service eligibility' above.)

●Resources for health care providers, patients, and families of children with LDs are provided in the table (table 2). (See 'Resources' above.)