Issues in the diagnosis and assessment of anxiety disorders in children and adolescents

doi:10.1016/S0272-7358(99)00037-9

Clinical Psychology Review

Volume 20, Issue 4, June 2000, Pages 453-478

https://doi.org/10.1016/S0272-7358(99)00037-9 Get rights and content

Abstract

Accurate diagnosis and assessment of anxiety disorders in children and adolescents is important for both treatment and research. In this paper, we review research related to the discriminant validity of the childhood anxiety disorders, and outline methods of assessment for children, including diagnostic interviews, self-report instruments, and behavioural, cognitive, and psychophysiological measures. Particular attention is given to psychometric and developmental issues, and their influence on assessment. The evidence provides support for the validity of the anxiety disorders as a whole, but only partial support for differentiation between specific anxiety disorders in children. Similarly, assessment methods used with anxious children provide reasonably accurate information on anxious symptoms, however are limited by poor discriminant validity and lack of sensitivity to developmental levels.

Section snippets

Diagnosis

In the Diagnostic and Statistical Manual of Mental Disorders, 4th ed.(DSM-IV; American Psychiatric Association, 1994), the specific category that was allocated to the childhood anxiety disorders in both editions of DSM-III, (American Psychiatric Association, 1980), was removed. The current diagnostic system views childhood anxiety disorders as essentially downward extensions of adult anxiety disorders. Separation anxiety disorder is the only diagnosis that remains a disorder classified under

Description of disorders and prevalence

Separation anxiety disorder (SAD) is characterised by excessive worry about separation from major caretakers. Children with SAD fear the loss of, or harm to, parents or other significant caretakers. The separation-anxious child may become tearful, throw tantrums, or complain of stomachaches or headaches in anticipation of or in the event of separation from major attachment figures. School refusal due to fear of separation is common. Children with SAD may also experience nightmares involving the

Discriminant validity

Discriminant validity refers to the degree to which the childhood anxiety disorders can be differentiated from each other, and from other disorders of childhood. If the childhood anxiety diagnoses do not demonstrate discriminant validity, there is little point in classifying them as distinct disorders. Evidence related to discriminant validity can come from many sources, including demographic data, psychological data, and comorbidity rates. Research in these areas has addressed the extent to

Stability over time

Another avenue for examining the validity of the childhood anxiety diagnoses, is to determine how stable the diagnoses are over time. That is, do children with anxiety disorder ‘grow out of it’? There are limited studies that examine the stability of anxiety diagnoses over time, however, some information is available. Caspi, Henry, McGee, Moffit, and Silva (1995) examined children at 5 years of age and then followed them through childhood and adolescence. Shy 5-year-olds were found to be more

Assessment measures and critical issues in assessment

Adequate assessment of anxiety in children and adolescents is critical for clinical diagnosis, treatment plan formulation, treatment evaluation, and research. Clearly, there is a need for assessment measures of child anxiety which are both reliable and valid. Current thinking on child assessment emphasises the importance of obtaining information from children and adolescents, in addition to information from other significant individuals in the child's life (La Greca, 1990). Children and

Structured diagnostic interviews for children

Structured or semistructured diagnostic interviews have several advantages over unstructured interviews. The reliability of unstructured interviews has generally been reported to be poor (Achenbach, 1980). Even though relatively unstructured interviews allow for greater flexibility in examining the problem, their lack of objectivity is clearly a major limitation. Structured interviews provide data which is easily quantifiable, and which assists in making a formal diagnosis based on current

Self-report measures for children

There are many self-report measures designed specifically to assess anxiety in children and adolescents, as shown in Table 1. These measures are quick and easy to administer, provide useful normative data, and are helpful in assessing treatment outcome. Nevertheless, most of these measures have limitations, primarily related to their ability to discriminate between diagnostic groups, and to their sensitivity to developmental issues. As outlined in Table 2, test-retest reliability for each of

Behavioural measures

As an alternative to structured interviews and self-report measures, there are several forms of behavioural assessment of anxiety in children. The behavioural assessment of anxiety primarily falls into two categories: direct observation and behavioural avoidance tests (BATs). Direct observation involves observation of the behavioural aspects of anxiety, such as facial expression, posture, and active avoidance. Children are usually observed in natural settings, however, specific settings can

Cognitive measures

The role of cognitive factors in the development and maintenance of anxiety disorders in children and adolescents has become the focus of investigation in recent research, however much is still unknown in this area. One line of work has centered on the development of self-report measures designed to assess the automatic thoughts, or self-statements, of anxious children. While several self-report measures of anxious cognitions have been developed for adults, there are few parallel measures for

Psychophysiological measures

The physiological measures of assessment most commonly used with anxious children, are those that assess cardiovascular and electrodermal responding (King, 1994). The physiological responses by normal children to threatening situation include increased heart rate (Beidel, 1988), changes in perspiration (Tal & Miklich, 1976), and increases in palmar sweating (Lore, 1966). In addition, children with anxiety disorders show increased adrenergic activity (Rogeness, Cepeda, Macedo, Fischer, & Harris,

Critical developemental issues in the assessment of child anxiety

The researched reviewed highlights several points of concern in the assessment of anxiety in children and adolescents. These include the lack of adequate reliability in children's self-reports across some measures, the low level of agreement between parent and child reports, and the lack of discriminant validity across diagnostic groups. The evidence points to the conclusion that children, particularly under the age of 11 years, have difficulty answering complex questions, suggesting that items

Cognitive and language development

Research into cognitive development has shown that distinct changes occur in an individual's ability to construct and manipulate representations of the environment, from birth through to adulthood (McShane, 1991). A child's ability to reason, to form concepts, and to communicate and comprehend, develop at a rapid rate (Bee & Mitchell, 1980). The quality and quantity of information obtained from a child during an assessment depends on the sensitivity of the measure to these cognitive

Concept of self, self-awareness, and emotion

An assumption underlying the majority of assessment methods with anxious children is that children have developed a concept of the self, have an understanding of emotion, and have some insight into processes within themselves. Children are also assumed to be able to communicate these concepts to others. These assumptions need to be examined in light of developmental research in these areas. Empirical data show distinct changes in children's self-understanding from infancy to adolescence (Damon

Cultural and gender issues

Work with children throughout the world shows that self-report measures of child anxiety are sensitive to cultural factors (Fonseca, Yule, & Erol, 1994). To illustrate, Dong, Yang and Ollendick (1994) found that Chinese children reported higher levels of social evaluative fears, compared to their Western counterparts, particularly for ages 11–23 years. Dong et al. (1994) argued that this may have been due to Chinese cultural and educational practices, which place great emphasis on performance

Social desirability

Assessment methods with children and adolescents aim to reflect the individual's unique perspective and subjective experiences. However, it has been increasingly recognised that children may respond in socially desirable ways, in an effort to please parents or adult examiners. This is a very important issue with respect to child anxiety, where children can be particularly concerned about their performance and negative evaluation. Children with social phobia will be especially likely to respond

Developmental changes in children's fears

A final issue in the assessment of anxiety in children, relates to normal fears during development, and to changes in the experience of anxiety from infancy through to adolescence. Numerous investigations have looked at the content, prevalence, and development of normal fear in children (e.g., Gullone & King, 1993). Clear changes have been found in content, from fears of loud noises, strangers and separation during infancy, to fears of evaluative and social situations during the school years,

Summary and conclusions

Studies into the validity and reliability of the current diagnostic system, provide partial support for the classification of anxiety disorders in children and adolescents. There is support for the validity of the anxiety disorders as a whole, but only partial support for differentiation between specific anxiety disorders in children. Therefore, more research is needed to justify the current system for the anxiety disorders. Most forms of assessment provide reasonably accurate information on