This article describes a systematic approach to subgroup classification based on a classification framework and sequential steps involved in the subgrouping process. The sequential steps are stating the purpose of the classification, identifying the classification elements, using relevant information, and using clearly stated and purposeful subgroup classification terms. This systematic approach reflects current changes in the field of intellectual disability (ID), the modern and social understanding of ID, and the multiple purposes for classification.
Introduction and Overview
The field of intellectual disability (ID) is changing in regard to terminology, the model used to explain human functioning, and the approach taken to classifying the severity of the condition. In reference to terminology, the term “intellectual disability” is currently used by the American Association on Intellectual and Developmental Disabilities (AAIDD; Schalock et al., 2010) and the Diagnostic and Statistical Manual–Fifth Edition (DSM-5; American Psychiatric Association, 2013). In the International Classification of Diseases–11th Edition's (ICD-11) proposed system, the term “disorders of intellectual development” is used (World Health Organization, 2015a). Diagnostically, three criteria are used by all three organizations to define ID operationally as characterized by significant limitations in intellectual functioning, significant limitations in adaptive behavior, and early age of onset (prior to 18 according to AAIDD; during the developmental period according to the American Psychiatric Association [APA] and World Health Organization [WHO]).
Models of human functioning are developed to clarify thinking. Beginning with the ninth edition of the AAIDD's (formerly the American Association on Mental Retardation) manual, Mental Retardation: Definition, Classification, and Systems of Supports (Luckasson et al., 1992), and continuing in their 2002 (Luckasson et al., 2002) and 2010 (Schalock et al., 2010) editions, a social-ecological, multidimensional model is used to explain human functioning and ID. A multidimensional approach broadens the conceptualization of ID to position it within human functioning and allows for a more comprehensive and purposeful approach to classification. As currently envisioned, this multidimensional model includes five dimensions: intellectual abilities, adaptive behavior, health, participation, and context (Buntinx, 2006; Luckasson & Schalock, 2013a; Schalock et al., 2010).
During the same period the 1992 AAIDD manual was developed, the World Health Organization was developing a comparable multidimensional approach to disability that was eventually published in 2001 and 2007. The purpose of the International Classification of Functioning, Disability, and Health (ICF) was to complement the International Classification of Diseases–10th Edition (ICD-10), which represented a classification of health conditions, diseases, and disorders. The ICF, however, extended beyond the medical perspective and included societal and environmental perspectives encompassing activities, participation, environmental factors, and personal factors.
A multidimensional approach to understanding and explaining ID can be used not only to conceptualize ID but also to classify the severity of ID. As the field of ID has shifted to a social-ecological approach to disability, the supports paradigm, and a supports-based service delivery model, the historical reliance on IQ scores as the sole basis for a classification scheme has diminished. For example, the DSM-5 specifies the severity level of autism spectrum disorder according to the following three levels of needed supports: supports needs, substantial supports needs, and very substantial supports needs (American Psychiatric Association, 2013, p. 52). These three levels are reflective of the four levels of needed supports initially proposed in the 1992 AAIDD manual (Luckasson et al., 1992): intermittent, limited, extensive, and pervasive. These levels did not appear in subsequent editions of the AAIDD manuals due to the then lack of standardized data on which to establish psychometrically sound classification bands. As discussed in a later section of this article, the current availability of standardized supports needs intensity scores permits a data-based approach to classifying support intensity levels.
Another important paradigm shift in the DSM-5 relates to abandoning the use of IQ scores to determine the severity level of ID. In place of the IQ score, the individual's level of adaptive behavior is proposed as the guide for determining severity level. Using a four-level system of mild, moderate, severe, and profound, DSM-5 asserts that adaptive behavior is a better classification element to use. This four-level system is also proposed in the ICD-11, but with the severity level determined by considering both the individual's level of intellectual ability and level of adaptive behavior (World Health Organization, 2015a, p. 4).
How the clinician classifies an individual with ID involves high-stakes decisions about these persons and their families. What's at stake for these individuals is an improved understanding of the person, rationally linking subgroup characteristics to important actions (e.g., planning supports, research, outcomes evaluation, and communication), fairness, and the equitable distribution of resources (Luckasson & Schalock, 2013b; Schalock & Luckassson, 2014). A classification system also significantly impacts the approach the field takes to answering the basic question in classification: “How is the total group that was defined as ‘in' the category (i.e., diagnosed as an individual with ID) now to be subdivided or categorized into smaller groups on the basis of criteria that are relevant to a specified purpose?” Without a systematic approach to classification, the field is hampered in the use of best practices related to answering this basic question and addressing the high stakes involved in the classification process.
Reflective of the changes that are occurring in the field of ID, the purpose of this article is to describe a systematic approach to subgroup classification. The importance of such a systematic approach is that it ensures that classification best practice standards are met. These standards are: a classification system addresses the question at hand, serves an important purpose, uses relevant information, and helps the clinician better understand the person (Schalock & Luckasson, 2014). Throughout the article, classification is defined as using a systematic framework and process to subdivide the group of individuals with ID into smaller subgroups based on classification elements and information that are relevant to and aligned with a specified purpose.
Components of a Classification System
We suggest that there are two components to a classification system. The first component is that the system incorporates five human functioning dimensions (i.e., intellectual abilities, adaptive behavior, health, participation, and context) and supports needs. The second is a classification framework that aligns each of these components with classification elements, relevant information, and common classification subgrouping terms.
Human Functioning Dimensions and Supports Needs
A multidimensional model of human functioning that can be used for subgroup classification was first proposed by AAIDD in the 1992 manual (Luckasson et al., 1992) and further refined in the 2002 and 2010 editions (Luckasson et al., 2002; Schalock et al., 2010). The five dimensions of the AAIDD model, which are consistent with the International Classification of Diseases (ICD) model (Buntinx, 2006; World Health Organization, 2001, 2007), are:
Intelligence—a general mental ability that includes reasoning, planning, solving problems, thinking abstractly, comprehending complex ideas, learning quickly, and learning from experience
Adaptive behavior—the collection of conceptual, social, and practical skills that have been learned and are performed by people in their everyday lives
Health—a state of complete physical, mental, and social well-being
Participation—the performance of people in activities in social life domains
Context—the interrelated personal and environmental conditions within which people live their everyday lives
We have added supports needs as a framework component. This addition is necessitated due to the significant impact of the supports paradigm on current service delivery policies and practices (Arnold, Riches, & Stancliffe, 2014; Arnold, Vivenne, & Stancliffe, 2011; Brown, Ouellete-Kuntz, Bielska, & Elliot, 2009; Thompson, Schalock, Agosta, Teninty, & Fortune, 2014). Supports needs are defined as the pattern and intensity of supports needed to be more successful in major life activities, to maintain or improve exceptional medical conditions, and/or to prevent or stabilize complex behavioral needs. Supports are defined as resources and strategies that aim to promote the development, education, interests, and personal well-being of a person and to enhance individual functioning. Services are one type of support provided by professionals and agencies.
The proposed classification framework is based on the five previously described human functioning dimensions plus supports needs. As shown in Table 1, each of these dimensions is aligned with the three processes involved in subgroup classification: classification element, relevant information, and classification subgrouping terms. As described in the following section, these processes are used as an organizing framework for the four sequential steps involved in implementing a systematic approach to subgroup classification.
Sequential Steps Involved in Subgroup Classification
We have suggested elsewhere (Schalock & Luckasson, 2014) that a classification system should be aligned with a clearly stated purpose, result from a logical and sequential series of steps, allow for multiple classification subgroupings, and be useful to the individual. The four sequential steps described next meet these criteria and also represent best practices related to determining (a) the specific purpose of the classification, (b) what is to be classified (i.e., classification element), (c) what information is required for classification, and (d) the specific terms used to categorize subgrouping.
The four primary purposes for subgroup classification are describing functional levels, operationalizing the level of supports needs, defining health status, and determining legal status. Additional reasons include understanding outcomes based on one or more human functioning dimensions, understanding related health status, establishing eligibility for supports, or conducting multivariate research studies.
Subgrouping is related to purpose. If, for example, the purpose is to determine individual budget allocations and matching needs with resources, classification into subgroups should be made on the basis of assessed supports needs. If, however, the purpose is to conduct research on health problems associated with intellectual functioning or adaptive behavior, purposeful classification into subgroups should be made on the basis of levels of intellectual functioning or adaptive behavior. If the purpose is to determine whether or not the person is competent to parent, subgroup classification should be based on the legal elements for competence to parent and the intensity of supports needs. As a final example, if one desires a holistic view of the person, classification would be based on a subgrouping profile across the six framework components.
As the field of ID continues to move to an ecological focus, the supports paradigm, and a functionality approach to ID that emphasizes human functioning dimensions and personal outcomes, there is a corresponding need to incorporate into best practices a broader approach to classification. For example, in reference to funding, individuals today are more likely to be grouped for funding on the basis of a weighted combination of factors that include either assessed supports needs or adaptive behavior, health status, and context (e.g., residential platform or geographical location). In reference to research, the identification of the significant predictors of human functioning and personal and family outcomes is increasingly based on grouping predictor variables into one or more of the dimensions of human functioning or levels of needed supports. In regard to planning supports, individualized support strategies and support objectives are increasingly based on the pattern and intensity of assessed supports needs across human functioning dimensions, life activity areas, and/or the domains of personal well-being (Schalock & Luckasson, 2014; Schalock & Verdugo, 2012).
Once the purpose of the classification is stated clearly, the next sequential step involves determining what is to be classified. Is it the individual's level of intellectual functioning, adaptive behavior, health status, participation, contextual facilitators or inhibitors, or supports needs? As summarized in Table 1 (column 2), this is a straightforward step.
For intellectual abilities, the classification element is IQ range.
For adaptive behavior, the element is adaptive behavior levels.
For health, the element is health status.
For participation, the element is level of participation.
For context, the element is personal or environmental facilitators or inhibitors.
For supports needs, the element is the intensity of supports needs.
Relevant information is essential to subgroup classification. To that end, we have included in Table 1 (column 3) relevant information that is aligned with the respective framework component, classification element, and subgrouping. As noted in column 3, information sources can be based on (a) a thorough history of the individual and include information obtained from previous evaluations, examinations, reports, receipt of services or supports, and current living/work environments; and (b) information obtained from broad-based assessment measures.
It is beyond the scope or purview of this perspectives article to list or summarize the plethora of broad-based assessment instruments. However, it is within the article's purview to discuss recent work in two human functioning dimensions in which, at this point, the field lacks standardized instruments: participation and context.
Participation refers to the performance of people in social life domain activities. Participation involves roles and interactions in the areas of home living, work, education, leisure, spiritual, and cultural activities. Currently, AAIDD has a task force addressing the conceptualization and measurement of participation, as well as the parameters of instruments to assess (a) participation levels in activities, events, and organizations; (b) interactions with friends, family, peers, and neighbors; and (c) social roles in regard to home, school, community, work, leisure, and recreation (M. Wehmeyer & K. Shogren, personal correspondence, December 18, 2014). Such participation-referenced information when obtained from standardized instruments could be used meaningfully to establish subgroups.
In reference to context, significant work is being done regarding how one can view context as an independent variable, an intervening variable, or an integrative approach to unifying systems of supports (Shogren, 2013; Shogren, Luckasson, & Schalock, 2014). As an independent variable, context includes personal and environmental characteristics that are not usually manipulated, such as age, language, culture and ethnicity, gender, and family. As an intervening variable, context includes (a) personal strengths and assets that can be maximized through personal growth and development opportunities; and (b) policies and practices at the community, organization, and societal level that can be manipulated to enhance personal outcomes. As an integrative approach, context can be used to align disability policy goals, personal outcome domains, personal outcome indicators, and influencing factors.
Based on this multifaceted approach to context, we have begun to identify contextual factors influencing personal outcome domains (Shogren, Luckasson, & Schalock, in press). This expanded understanding of context provides an additional way to think about context and classification subgroupings. For example, subgroupings could involve (a) levels of personal outcomes in the outcome domains of self-determination, full citizenship, education/life-long learning, productivity, well-being, inclusion in society and community life, and human relationships; and/or (b) personal and environmental factors influencing each of these outcome domains.
Classification Subgrouping Terms
This fourth sequential step is still emerging and is in a state of flux regarding not just the categorical terms used, but also the criteria used to operationalize the terms. As reflected in Table 1 (column 4), some of the subgrouping terms listed are those used historically (e.g., mild, moderate, severe, and profound), some are becoming more widely used (e.g., supports needs, substantial supports needs, very substantial supports needs), and some are still emerging (e.g., degrees of community integration, levels of social integration, levels of personal outcomes, levels of personal or environmental facilitators or inhibitors).
The reader is familiar with the IQ bands that have historically been used as subgroup classification terms. Analogous bands are available regarding adaptive behavior levels (Tassé et al., 2012) and support intensity levels (Arnold et al., 2014; Thompson et al., 2015; Thompson et al., in press). In terms of health, the ICD-10 (World Health Organization, 1999) and the DSM-5 (American Psychiatric Association, 2013) provide classification subgroupings for a multitude of disorders, diseases, or injuries. The ICD-11 has yet to be published (World Health Organization, 2015b), but, as mentioned earlier, the level of intellectual ability and the level of adaptive are currently being proposed to establish subgroupings (World Health Organization, 2015a). DSM-5 uses a descriptive approach to establish severity levels for autism spectrum disorders. For example, in reference to Level 3 (requiring very substantial supports) and in reference to social communication, Level 3 is defined as “deficits in verbal and nonverbal social communication skills, cause severe impairments in functioning, very limited initiation of social interactions, and minimal response to social overtures from others” (p. 52).
The authors anticipate that further significant work will occur in efforts to subcategorize in order to more accurately capture the complexities of human functioning and the multiple purposes of classification. These efforts will include conceptualizing sound approaches to subgrouping, developing standardized data-based approaches to establish subgroups, establishing the criteria used to operationalize the subgrouping(s) conceptualized, and reaching consensus terminology (Arnold et al., 2014; Schalock & Luckasson, 2014; Tassé, Luckasson, & Nygren, 2013). As work in these areas merge with the use of Item Response Theory as an alternative to Classical Test Theory, the use of statistical procedures involving cluster analysis and structural equation modeling, and the movement toward data-based decision making, the authors also anticipate that the field will consider that subgrouping classification (a) does not have to be based on a single variable cut-off score, (b) is not an absolute phenomenon based on criterion bands, (c) can be based on metrics other than standard deviations or percentiles, and (d) requires a new or significantly modified terminology.
Advantages of the Proposed Approach
Classification should reflect the modern scientific and social understanding of human disability that includes recognizing the complexity of human functioning and the need to address multiple classification purposes. The field of ID is increasingly embracing an ecological focus that encompasses the multidimensionality of human functioning, a supports paradigm, evidence-based practices, and outcomes evaluation. Knowledge and application in both service delivery and research endeavors are increasingly reflecting these changes and require us to answer three questions related to classification in new ways (Simmeonsson et al., 2006):
Does the classification contribute to a systematic, hierarchical way of organizing knowledge?
Is the classification based on a coherent theoretical framework (such as the multidimensionality of human functioning and the supports paradigm)?
Does the classification system reflect the efficient organization of knowledge and a relevant purpose for the classification?
We conclude that the components of a classification system and the four sequential steps involved in subgroup classification discussed in this article address these questions and present a heuristic model that will enhance the precision, quality, validity, and relevance of the respective classification scheme used.
In addition, the suggested approach engages the power of alignment and the critical thinking skill of synthesis. As discussed more fully in Schalock and Luckasson (2014) and Schalock and Verdugo (2012), alignment involves placing or bringing clinical processes into a logical sequence and aligning them to the questions asked. Synthesis involves integrating information from multiple sources to create a more complete, accurate, and coherent form or pattern.
In reference to the power of alignment, the processes described in reference to Table 1 articulate the operative relationships among the essential elements of a classification system, identify the information and thinking skills involved in that system, and emphasize core processes involved in subgroup classification. These processes allow for transparency, a clearer understanding of classification among stakeholders, and a systematic way to address multiple classification purposes. The ability to demonstrate this flexibility will not only allow one to be more specific in regard to answering classification questions, but will also strengthen the clinician's professionalism since the classification system used reflects best practices and is relevant to the individual.
Synthesis plays a central role in classification wherein information is integrated from multiple sources in order to create a more complete, accurate, and relevant classification subgrouping. Synthesis involves integrating (a) classification purpose with information needed to formulate a relevant subgrouping; (b) components of a classification framework with the criteria used to define operationally the categories used in the subgrouping; (c) current best practice mental models that emphasize the social-ecological nature of disability, the supports paradigm, and the values of community inclusion, equity, and self-determination; and (d) the available information in light of the individual's beliefs, judgment, personal goals, and circumstances. Thus, synthesis involves not just using information to divide individuals with ID into subgroups, but also to give meaning to both the information and the subgrouping.
Because of the potential uses (and misuses) of subgroup classifications, it is important that best practices guide their development and use. To that end, the four sequential steps outlined above involving classification purpose, classification elements, relevant information, and aligned subgrouping terms meet the standards of best practices. These steps also address the question at hand, serve an important purpose, and help the clinician better understand the person.
Individuals working today in the field of ID are faced with answering classification-related questions about an individual's functional level, supports needs, health care, funding level, outcome predictors, and competency (Arnold et al., 2014; Ellis, 2013; Schalock & Luckasson, 2013). Answering these questions requires attention to clinical judgment standards, extensive knowledge and interactions with the person, a familiarity with best practices in the field regarding the expanded uses of classification, and the use of critical thinking skills related to alignment and synthesis (Luckasson & Schalock, in press).
We have suggested in this article that classification needs to be purposeful and relevant to the individual. Fulfilling these two criteria requires that classification must be (a) considered within the context of a valued community life and valued personal outcomes; (b) based on valid assessment that considers cultural and linguistic factors; (c) aligned with a purpose that is relevant to the individual and the individual's well-being; and (d) systematic, logical, and transparent. We have also suggested that a thorough and more systematic discussion of classification will result in moving the field toward a much needed consensus on the definition of classification and its purpose, classification elements, relevant information to inform decision making, subgroup classification terminology, and clearly articulated and operationalized subgrouping criteria.
Professionals within the field of ID are challenged today more than ever before to deeply analyze what subgroup classification really means, keeping in mind that classification is the net result whenever one takes an entire diagnostic group and categorizes them into subgroups. Thus, because classification begins with where one starts, one must determine clearly what important purpose the subgrouping serves; what information is most relevant to form the subgroupings; and, ultimately, whether the classification process allows one to better understand the person, enhances the individual's well-being, and facilitates responsive system change.
Robert L. Schalock, Hasting College; Ruth Luckasson, University of New Mexico.