Classification and Assessment
PRINCIPLES OF CLASSIFICATION
Classification helps us to be better observers and to formulate hypotheses and principles. Shared approaches to classification help us communicate more effectively and develop better theories. In mental health disciplines, and indeed in medicine in general, the process of giving a label may be associated with some sense of relief on the part of the patient or the patient’s parents when the patient is a child. Assigning a label, however, does not imply having an explanation. Like all human theories, classification approaches have their limitations or can be misused. There is no single “right” way to classify disorders in childhood. Systems vary, depending on the purpose of classification and what is being classified. Official diagnostic systems, such as the World Health Organization’s International Classification of Diseases (ICD-11) (World Health Organization, 2019-2020) and the American Psychiatric Association’s Diagnostic and Statistical Manual (DSM-5) (American Psychiatric Association, 2013) are generally categorically oriented. The dimensional approaches that are based on considerations of how behaviors or emotional traits are distributed in the population are quite useful for understanding psychopathology. To be useful, classification schemes must provide adequate descriptions of disorders (so they can be reliably differentiated from each other) and must be useful across the range of age and severity. Deviant behavior itself does not necessarily constitute a disorder unless it is a manifestation of dysfunction within the individual person that significantly impaired this person’s life.
MODELS OF CLASSIFICATION
Various approaches to classification can be used (Table 4.1). These are not necessarily incompatible with each other. For example, a continuous variable such as IQ or blood pressure can be used to define levels of severity (e.g., of intellectual disability) or disorder (hypertension).
The issue of which model works best depends on the specific situation. For example, structured rating scales and diagnostic interviews have been developed for many disorders. In addition, a series of well-designed, psychometrically sound structured interviews “keyed” to DSM-5 diagnostic concepts have been developed; these are particularly useful and often required in research studies (Angold et al., 2018). Another approach has focused on assessment
of psychopathology with derivation of more basic “factors” such as internalizing and externalizing disorders on the Child Behavior Checklist (Achenbach, 1966). Such instruments have both research and clinical utility and may have importance for screening but do not usually translate straightforwardly into DSM-type diagnoses, which require integration of clinically relevant information by an expert (or at least experienced clinician). As we shall discuss later, different informants—parents, teachers, and the children themselves—can, and often do, provide different information that can inform assessment and diagnoses in important and complementary ways.
of psychopathology with derivation of more basic “factors” such as internalizing and externalizing disorders on the Child Behavior Checklist (Achenbach, 1966). Such instruments have both research and clinical utility and may have importance for screening but do not usually translate straightforwardly into DSM-type diagnoses, which require integration of clinically relevant information by an expert (or at least experienced clinician). As we shall discuss later, different informants—parents, teachers, and the children themselves—can, and often do, provide different information that can inform assessment and diagnoses in important and complementary ways.
TABLE 4.1 Approaches to Classification | ||||||
|---|---|---|---|---|---|---|
|
Developmental issues are also of considerable importance in classifying disorders in children and adolescents (and occasionally adults). Disorders such as autism and attention-deficit hyperactivity disorder (ADHD) originate during specific times of development, and other disorders, such as Tourette’s disorder, and age of onset and developmental trajectory can provide important diagnostic context. In other cases, preexisting disorders may complicate the diagnosis of other conditions (e.g., a child with oppositional defiant disorder who goes on to develop depression in adolescence, which might be overlooked owing to a history of irritability). For some forms of developmental psychopathology, diagnostic guidelines have a strong developmental orientation, but for others, the nature of the symptoms predominates.
Theoretically based classification systems tended to be more common in the past. For example, Anna Freud had a model of classification based on her psychoanalytically informed understanding of child development (Freud, 1965). More phenomenologically based classification systems can be traced to Kraeplin’s delineation of schizophrenia and bipolar disorder (Hoff, 2015). Theoretically based classification systems tend to be most useful for clinicians working with that specific theoretical framework, but they may be less useful for clinicians who do not share the same orientation. For several decades, the official classification systems, such as the one reflected in the Diagnostic and Statistical Manual of the American Psychiatric Association, have focused less on theories and more on research-based approaches to classification.
It is generally recognized that there is no single “right” way to classify disorders in childhood (Rutter, 1965). Classification systems vary depending on the purpose of classification and what is being classified. “Official” diagnostic systems have tended to adopt, on the whole, a categorical approach, but a dimensional approach would be equally applicable, though perhaps less useful for clinical purposes.
It is often assumed that classification systems of psychopathology are developed to approximate some ideal diagnostic system in which the cause could be directly related to clinical condition. This is not, in fact, the case, because the cause need not be included in classification systems. Different etiologic factors may result in rather similar conditions, and the same etiologic factor may be associated with a range of clinical conditions. With a few exceptions (e.g., posttraumatic stress disorders), etiologic factors are not generally included in official diagnostic systems.
Environmental or contextual factors are particularly important in understanding childhood psychopathology. Thus, variables such as family, school, or cultural setting can serve as major modifiers of clinical presentation. For example, a child who had attentional difficulties because
of an inappropriate school placement should not have a diagnosis of attention-deficit disorder. Contextual variables are particularly problematic in disorders of infancy and early childhood in which child and parent variables often interact with each other. Cultural differences may also be important and may interact in complex ways with child vulnerability and family variables.
of an inappropriate school placement should not have a diagnosis of attention-deficit disorder. Contextual variables are particularly problematic in disorders of infancy and early childhood in which child and parent variables often interact with each other. Cultural differences may also be important and may interact in complex ways with child vulnerability and family variables.
Finally, it should be emphasized that disorders, not children, are classified. This may seem a subtle point, but it is not. There are potential negative effects (and positive ones) related to labeling. Clearly, it is children, and not labels, who need help, and it is not appropriate to equate people with their problems. Labels can have some social stigma or other untoward effects or may be associated with more realistic expectations on the part of parents and teachers and provision of potentially more appropriate services.
IDEOGRAPHIC APPROACHES
Ideographic approaches to diagnosis are common in clinical practice. In the broader sense of clinical diagnosis, most clinicians target certain problems or issues for intervention that relate only in part to categorical or even dimensional diagnosis (Bostic et al., 2018). In some ways, such approaches are more practical in clinical setting or in conjunction with certain forms of psychotherapy such as family counseling. Ideographic approaches that require careful documentation of signs and history of clinical issues as well as detailed description of contexts that may affect or be affected by symptoms such as family, school, and neighborhood are also an important part of clinical case formulation (Henderson & Martin, 2018). Beyond the individual cases, the utility of ideographic approaches is limited to, for example, research purposes, where information needs to be communicated concisely and systematically.
RESEARCH ISSUES
As official classification systems have become more complex and sophisticated, issues of reliability and validity have assumed increasing importance. For example, both the DSM-5 and the ICD-10 use results of large national or international field trials in providing definitions of disorders. Categorical and dimensional approaches to classification share certain statistical concerns, including validity and reliability. Validity refers to the extent to which the diagnostic category captures the phenomena it purports to (e.g., does the diagnostic category have some meaning relative to course and treatment or family history or associated conditions?) Reliability refers to the ability of different individuals to use the diagnostic approach in the same way. This is also referred to as inter-rater reliability, a statistical index that addresses an important question about a diagnosis such as whether two doctors who examine the child within a relatively short interval (e.g., in the same week) would arrive at the same diagnosis. As with validity, various kinds of reliability are identified, and these are impacted by various factors, including who is conducting the assessment as well as variability in symptoms and time course of the disorder. Test-retest reliability refers to a question of whether or not diagnosis remains the same or changes if assessments are conducted on two occasions separated by a longer period. This issue of testretest reliability is linked to the issues of stability and chronicity of mental health disorders. For example, conditions such as ADHD and obsessive-compulsive disorder tend to remain relatively stable over extended periods if untreated. And some disorders such as Tourette’s syndrome or major depression can go through periods of remission and exacerbation. The waxing and waning course may be an important classification feature of the disorder that needs to be reflected in diagnostic assessment.
Various statistical techniques have been applied to data derived from assessment methods (Nunnally & Bernstein, 1994). These techniques are theoretically of considerable interest in that they can provide more rational and empirical approaches to the derivation of diagnostic schemes. The fundamental assumption of such techniques is that the variables of interest lie
along some dimension of function and dysfunction that all persons exhibit to some degree. For many types of problems, this assumption is probably justified, such as relative to anxiety or depression. However, the usefulness of such techniques is limited in important ways. In the first place, these methods are highly dependent on both the sample and the type of data entered in the analysis. For example, factor analysis of an instrument designed to detect conduct problems would not likely produce a factor related to anxiety and vice versa (Fabrigar & Wegener, 2012). Similarly, cluster analysis of even a very large normative sample would not likely produce a cluster that corresponded to autism, given the low base rate of this disorder in the population.
along some dimension of function and dysfunction that all persons exhibit to some degree. For many types of problems, this assumption is probably justified, such as relative to anxiety or depression. However, the usefulness of such techniques is limited in important ways. In the first place, these methods are highly dependent on both the sample and the type of data entered in the analysis. For example, factor analysis of an instrument designed to detect conduct problems would not likely produce a factor related to anxiety and vice versa (Fabrigar & Wegener, 2012). Similarly, cluster analysis of even a very large normative sample would not likely produce a cluster that corresponded to autism, given the low base rate of this disorder in the population.
In summary, classification in child and adolescent psychiatry has multiple meanings and functions. Complications in the classification of child and adolescent disorders are myriad: The child is often not the person complaining, different kinds of data may be used in making a diagnosis, developmental factors may have a major impact on the expression of disorders, and certain features (e.g., beliefs in fantasy figures) are normative at certain ages but not at others. Additional complications are posed by the unintended, but no less real, uses to which diagnostic concepts are put, such as their inclusion in legislation and their use as mandates for services in educational programs or for purposes of insurance reimbursement for services. Different kinds and levels of classification are needed for different purposes.
RESEARCH DOMAIN CRITERIA (RDOC) FRAMEWORK
The Research Domain Criteria (RDoC) approach is a relatively new initiative that has been launched by the National Institute of Mental Health (NIMH) in order to gather empirical data to ultimately develop a new classification of psychopathology based on the dimensions of neurobiology and behavior that cut across traditional categories of mental disorders (Insel et al., 2010; Sanislow, 2020). In contrast to the DSM and ICD classification systems, which are intended for clinical use, the RDoC approach is intended for researchers to provide a working model to stimulate research on the core dimensions (or constructs) of psychopathology. First, in 2009, an NIMH work group devised a proposal for a new system and outlined five major domains of functioning such as cognition and emotions and units of analysis ranging from genes to neural circuits to behavior. Then, five groups of experts in basic and behavioral sciences representing each of the domains were convened for a series of workshops and asked to determine and define specific dimensions to be included in each domain. In order to be included, the domains (e.g., “fear”) had to reflect validated behavioral functions and show evidence for a neural circuit or system responsible for implementing this function.
The ensuing RDoC constructs are organized in a two-dimensional matrix of five domains that include constituent constructs in the rows and seven units of analysis as the columns (Cuthbert, 2015). The units of analysis include genes, molecules, cells, circuits, physiology, behavior, and self-report, reflecting a range of methodologic approaches from genetics to psychology. The first three levels of analysis pertain to the neurobiologic mechanism of circuitry essential to each construct. In turn, neural circuit is a central unit that can be studied with neuroimaging methods such as functional magnetic resonance imaging or indices of circuit activity such as fear-potentiated startle. Physiology refers to variables such as event-related potentials or heart rate variability that can be validated as indirect measures of neural circuits. Measures of observable behavior and performance on laboratory task are grouped under the unit of analysis termed “behavior.” Lastly, self-report category is reserved for interviews, rating scales, and other psychometric instruments of various aspects of the construct of interest that represent signs and symptoms of psychopathology. This emphasis on the multifaceted characterization of constructs is part of the RDoC’s aim to incorporate methods of genetics, neuroscience, and cognitive science into the future classification scheme of mental disorders.
The RDoC constructs are grouped into domains of functioning that reflect key aspects of emotion, motivation, cognition, social behavior, and regulatory systems (National Institute of Mental Health, 2020). The first domain, negative valence systems, includes constructs defined by responses to acute threat (“fear”), potential threat (“anxiety”), sustained threat, loss, and frustration. The second domain, positive valence system, contains constructs defined
by reward learning and habit formation. For example, approach motivation construct encompasses goal pursuit behaviors subserved by the mesolimbic dopamine system. A dysfunction of this system can lead to abnormally low (e.g., avolition) or high (e.g., addition) goal-directed behaviors. Another construct within this domain, called “habit,” is viewed as a manifestation of reinforcement learning where repetitive motor or cognitive behaviors occur and manifest without serving an adaptive goal. The cognitive systems domain encompasses six broad constructs: attention, perception, declarative memory, language, cognitive control, and working memory. The social processes domain includes constructs that bridge social behavior with neural circuits of attachment, social dominance and perception, and understanding of self and others. For example, facial communication may include receptive aspects of facial affect recognition and productive aspects of eye contact and gaze following. The neural underpinnings of these behaviors have been studied with an array of neuroimaging and electrophysiologic methods in order to establish biomarkers of psychiatric disorders such as ASD and schizophrenia. The fifth domain includes arousal and regulatory systems that subserve many of the other domains and are central in sleep and wakefulness.
by reward learning and habit formation. For example, approach motivation construct encompasses goal pursuit behaviors subserved by the mesolimbic dopamine system. A dysfunction of this system can lead to abnormally low (e.g., avolition) or high (e.g., addition) goal-directed behaviors. Another construct within this domain, called “habit,” is viewed as a manifestation of reinforcement learning where repetitive motor or cognitive behaviors occur and manifest without serving an adaptive goal. The cognitive systems domain encompasses six broad constructs: attention, perception, declarative memory, language, cognitive control, and working memory. The social processes domain includes constructs that bridge social behavior with neural circuits of attachment, social dominance and perception, and understanding of self and others. For example, facial communication may include receptive aspects of facial affect recognition and productive aspects of eye contact and gaze following. The neural underpinnings of these behaviors have been studied with an array of neuroimaging and electrophysiologic methods in order to establish biomarkers of psychiatric disorders such as ASD and schizophrenia. The fifth domain includes arousal and regulatory systems that subserve many of the other domains and are central in sleep and wakefulness.
The RDoC framework is based on three core assumptions: (1) mental illnesses are presumed to be disorders of brain circuits, (2) neuroscience tools can identify the pathophysiology, and (3) the discovery of biosignatures will supplement diagnoses based on clinical signs and symptoms and direct assessment and treatment of mental illness. Since its inception, the RDoC approach has stimulated an increasing body of research on developing and validating the proposed constructs along the lines that are outlined in the RDoC matrix. In addition, RDoC offers a new perspective to study co-occurring disorders. For example, autism commonly co-occurs with anxiety, but when using symptom-based criteria, it is hard to know whether anxiety is a true comorbidity or a feature of autism. However, measures that have been validated to test relevant functions within anxiety/fear circuitry and social brain networks can provide an alternative way to characterize anxiety in autism (Lau et al., 2020).
DIAGNOSTIC ASSESSMENT
Clinical and diagnostic evaluation of children and adolescents should carefully consider the context of functioning within the family, school, peer, cultural, and community settings with a goal of identifying specific forms of psychopathology and developing an appropriate treatment plan if one is needed. Depending on the reason for referral and presenting concerns, the examiner may need to prioritize areas for assessment and intervention (e.g., the presence of suicidal thoughts or high-risk behaviors). Some assessments are conducted for a very specific purpose (e.g., custody assessments or evaluation of suicidal risk of a child in the emergency department) and require very specialized expertise. More typically, the diagnostic assessment process requires the examiner to take a broad view, taking into account presenting complaints (of the child, parents, teachers, or others), the child’s history and level of development, and family and cultural factors.
The assessment of a child differs from the psychiatric assessment of an adult in several important ways. Typically, parents or sometimes schools have initiated a referral, and the child may or may not be as troubled by the problem. The assessment also depends on the child’s chronologic age and developmental level so that the approach to a preschool child will often involve play or games; a school-age child may prefer some combination of discussion and activities; and an interview with an adolescent may be more like that of the adult. It is important that the child understand, at whatever level they can, the purpose of the assessment and that, as appropriate, the clinician conduct interviews in a way designed to facilitate discussion. Unlike adults, children can act very differently depending on the setting. A child who is having real trouble sitting still in school may be well behaved and popular on the playground. As a result, it is important to collect information from different sources, including the child, parent(s), and school. Consequently, a major task for the clinician becomes the reconciliation of views when they diverge. Also, as a result of this process, the clinician needs to form a working relationship with multiple parties while maintaining, as appropriate, the
child and family’s confidentiality. In contrast to interviews with adults, developmental issues can loom large either as presenting complaints (e.g., continued temper tantrums or delayed speech) or as important considerations in the assessment itself (e.g., a child with autism who is minimally verbal). For younger children without developmental delays, it is important that the clinician have an awareness of normative cognitive processes and common childhood fears, beliefs, and fantasies. The assessment should, of course, be tailored to the circumstances of the individual case, but several key components should be considered (Table 4.2) with the aim of identifying the variables relevant to the child’s presentation.
child and family’s confidentiality. In contrast to interviews with adults, developmental issues can loom large either as presenting complaints (e.g., continued temper tantrums or delayed speech) or as important considerations in the assessment itself (e.g., a child with autism who is minimally verbal). For younger children without developmental delays, it is important that the clinician have an awareness of normative cognitive processes and common childhood fears, beliefs, and fantasies. The assessment should, of course, be tailored to the circumstances of the individual case, but several key components should be considered (Table 4.2) with the aim of identifying the variables relevant to the child’s presentation.
TABLE 4.2 Content Components of The Psychiatric Assessment of Children and Adolescents | |||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||||||||||||||||||||||||||||||||||
Typically, the assessment begins with a review of the reasons for referral. This helps clarify the nature of the presenting problem(s) and expectations for what the assessment will provide. The history can be obtained from relevant persons and perspectives (e.g., child, parents, other family members, school personnel). The examiner should be alert to the context or circumstance in which problem behavior emerges. In some fundamental way, the examiner tries to assemble, and constantly revise, a narrative with attention first paid to the “facts” as they present themselves (the who, what, where, and when of the narrative) with an eventual formulation (the why). The clinician should be alert to important clues about what sets off or maintains problem behaviors. A history of previous treatment should be included if relevant. At some time, the examiner will also wish to obtain a developmental history to help clarify any potential developmental difficulties contributing to current problems and any long-standing issues that may shed light on current problems.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
