Axis I

Axis I contains what the DSM-IV-TR labels “Clinical Disorders”: all psychiatric disorders included in the DSM except for the personality disorders and mental retardation, which are recorded on Axis II (Box 17-2). In addition, Axis I is the location to record “Other Conditions That May Be a Focus of Clinical Attention,” which includes medication side effects and a variety of social and relational problems (Box 17-3). All Axis I diagnoses that are relevant to care should be listed in order of clinical concern, with the most important diagnosis listed first.

Various algorithms are available to help maneuver through the DSM, including those in Appendix A of the DSM-IV-TR, “Decision Trees for Differential Diagnosis.” An example for the anxiety disorders is shown in Figure 17-1; others are delineated elsewhere in this textbook. When making the determination of which DSM-IV-TR Axis I diagnoses best fit a particular patient, it is important to attend to which diagnoses are hierarchical and mutually exclusive and which may be co-morbid/co-existing in an individual patient at the same time. For example, a patient may have a long-standing specific phobia of heights, develop obsessive-compulsive disorder (OCD) in adolescence, and then develop post-traumatic stress disorder (PTSD) following an injury in adulthood, resulting in three co-morbid anxiety disorders at the time of presentation at age 45. On the other hand, the criteria for adjustment disorder clearly state that before assigning this diagnosis, the clinician must determine that the symptoms do not fulfill criteria for another Axis I disorder and are not related to an exacerbation of another Axis I or II disorder. Likewise, a nonpsychiatric medical or substance-induced etiology must typically be ruled out to the clinician’s clinical satisfaction before any other Axis I disorder can be diagnosed.

Figure 17-1 Anxiety disorders algorithm. This algorithm demonstrates that some diagnoses are hierarchical and mutually exclusive, whereas others may be co-morbid in an individual patient.

Most sections of the DSM also include “Not Otherwise Specified (NOS)” diagnoses, such as alcohol-related disorder NOS, dissociative disorder NOS, or eating disorder NOS. When used alone or in combination, these may be used to describe disorders that, although clinically significant, do not fit neatly into the major diagnostic categories, or where more data are required before a more precise diagnosis can be assigned. When applicable, diagnosis deferred, no diagnosis, or unspecified mental disorder (nonpsychotic) could be recorded on Axis I. Another option to signal diagnostic uncertainty is to mark a diagnosis as provisional, meaning that the selected diagnosis is expected to emerge over time or with new information. An example of how Axis I diagnoses may evolve in the setting of clinical uncertainty and change is shown in Box 17-4.

Specifiers may also be added to Axis I diagnoses to provide additional information about the clinical course, features, and severity. Current severity can be indicated with specifiers (i.e., mild, moderate, or severe). When symptoms have substantially improved, the following specifiers may be used: in partial remission, in full remission, or prior history. While guidance is included in the DSM-IV-TR about the definition of these specifiers for some conditions (e.g., conduct disorder, substance dependence, and manic episode), the psychiatrist may apply them to other disorders using his or her clinical judgment. Some disorders have their own specifiers listed in the DSM-IV-TR, such as “with delayed onset” for PTSD. Others have various mutually exclusive subtypes, such as blood-injection-injury type or natural environment type for specific phobia (see Figure 17-1). In some cases, as with mood disorders, the DSM-IV-TR provides instructions for representing specifiers or subtypes via the fifth digit of the ICD-9-CM numerical diagnostic code.

Axis II

Axis II contains personality disorders (Table 17-2) and mental retardation (Table 17-3). Borderline intellectual functioning, although not considered a mental disorder, is also coded on Axis II. As with Axis I, multiple diagnoses should be listed on Axis II if present. If an Axis II diagnosis, rather than one or more co-morbid Axis I disorders, is the primary clinical concern, this may be noted by qualifying it in parentheses as principal diagnosis or reason for visit. Given that additional evaluation time or clinical information may be needed to diagnose Axis II disorders, it may be appropriate to specify no diagnosis or diagnosis deferred. In addition, personality traits that do not meet full criteria for a personality disorder, but are nonetheless maladaptive, may be listed on Axis II without the use of a diagnostic code, as may defensive patterns.⁴ The DSM-IV-TR provides examples of specific defensive patterns in Appendix B, in the “Glossary of Specific Defense Mechanisms and Coping Styles.” The Defensive Functioning Scale (pp. 807-810)⁵ is included as a “Proposed Axis for Further Study,” with the suggestion that this hierarchical ranking of defensive styles be placed below Axis V. In practice, inclusion of specific defensive patterns or a defensive level might be more easily incorporated into Axis II.⁴ Box 17-5 is an example of using Axis II in a way that may enhance clinical communication within a mental health care system.

Table 17-2 Axis II: Personality Disorders

Table 17-3 Axis II: Mental Retardation

Axis III

Axis III lists all relevant “General Medical Conditions,” meaning nonpsychiatric illnesses; Table 17-4 lists some of the disorders commonly recorded on this axis. ICD-9-CM codes for frequently used medical diagnoses can be found in Appendix G of the DSM-IV-TR.⁵ If the evaluator considers the medical problem and psychiatric problem to be closely interrelated, listings on both Axis I and Axis III may be appropriate (Box 17-6). This may occur when the medical problem is thought to cause the psychiatric problem, resulting in a diagnosis of “Psychiatric Disorder due to a General Medical Condition” on Axis I. It may also occur when an Axis I or II psychiatric disorder, subthreshold symptoms (e.g., anxiety or overeating), personality traits or defenses, or physiological stress responses are judged to have a negative impact on the course of a general medical condition; this scenario results in a diagnosis of “Psychological Factor Affecting Medical Condition” on Axis I. Particularly salient psychosocial factors that affect coping with medical illness may also be appropriate to list on Axis IV.

Table 17-4 Axis III: Selected General Medical Conditions

Axis IV

Psychosocial and environmental problems are listed on Axis IV (Table 17-5).¹⁸ As many stressors as are relevant to diagnosis and clinical management may be included. In some treatment settings, a checklist that contains common types of problems that has space for adding specifics, such as the one contained in the DSM-IV-TR “Multiaxial Evaluation Report Form” (p. 36),⁵ can be useful in facilitating clinical communication. Much of the time, Axis IV will include only current problems, but in some cases, past stressors are particularly relevant to diagnosis and clinical management and should be listed as well. For example, though it occurred several years previously, a contentious parental divorce that has exacerbated a teenager’s anxiety disorder and is critical to understanding the current family environment should be included on Axis IV, with the anxiety disorder listed as usual on Axis I. Sometimes the psychosocial or environmental problem is important as a main target of treatment, rather than primarily via its effects on a separate psychiatric disorder; in this case, the problem may also be listed on Axis I, using a code from the category “Other Conditions That May Be the Focus of Clinical Attention” (Box 17-7).

Table 17-5 Axis IV: Psychosocial and Environmental Problems

Axis V

The Global Assessment of Functioning (GAF) is recorded on Axis V (Table 17-6).¹⁹ The GAF is a number between 0 and 100 that summarizes the clinician’s view of the patient’s current degree of impairment in terms of psychosocial and occupational or educational function. Generally, normal function is coded in the 70-to-100 range, mild psychiatric symptoms fall in the 70-to-80 range, and moderate symptoms are assigned a number between 60 and 70. Severe symptoms are coded as 50 and below; higher levels of psychiatric support (such as intensive community-based treatment, residential settings, or inpatient hospitalization) are often required as function drops further down the scale. Although it may be difficult to determine in complicated cases, the DSM-IV-TR specifies that functional difficulty that results from “physical (or environmental) limitations”⁵ should not be considered in assigning a GAF score; the intent is to focus on the effects of mental illness. So, a patient who is no longer able to work or to live independently because of paraplegia may nonetheless receive a GAF score well above 70 if he has a wide social network, remains active in local politics, feels satisfied with life, and experiences frustration but takes effective action in the face of setbacks related to his physical problems.

Table 17-6 Axis V: Global Assessment of Functioning (GAF) Scale

The GAF score is not a static number; it changes for each patient over time. It can be used to track and succinctly communicate the course of psychiatric disorders over time and to monitor response to treatment. An increase in the GAF score from 35 to 55 during the course of an inpatient hospitalization, for example, would suggest that the treatment plan has been at least somewhat effective, whereas a decline from 35 to 20 might suggest that it has not. Reporting the highest and lowest GAF score in the past year along with the current GAF score can also help paint a concise picture of overall function. However, it is important to consider the limitations of the GAF scale in clinical decision-making (Box 17-8). Neither the GAF score alone nor, for that matter, the entire multiaxial assessment can be a substitute for psychiatric clinical skill and judgment.

Psychodynamic Approaches

One criticism of psychiatric evaluation geared too rigidly to making DSM-IV–based multiaxial diagnosis is that it may neglect psychodynamic formulation, resulting, for example, in the possibility of psychiatrists’ “feeling more confident that they know what the patient ‘has’ than they are of who the patient is” (Gordon and Riess³⁷); in this scenario, there may be difficulty in determining which combinations of biological, spiritual, social, and/or psychological approaches to treatment might be indicated.³⁷ Some have suggested that psychodynamic principles be returned to the DSM. The Defensive Functioning Scale appears in Appendix B as an axis to be studied for potential incorporation into future versions of the DSM, and alternative multiaxial schemas designed for psychotherapy planning have been proposed as well.³⁸ Meanwhile, a coalition of psychoanalytically oriented organizations has developed its own diagnostic manual separate from the DSM and ICD.³⁹

Behavioral Approaches

Behaviorally oriented clinicians argue that describing a behavior (such as crying) divorced from its context and function is meaningless, and that behavioral functional analysis should be incorporated into diagnostic classification.^33,40 In this view, the kinds of syndromic disorders created in DSM-IV-TR by grouping together of lists of symptoms is minimally useful. For example, it may be more helpful in behavioral treatment planning to create categories that account for what is reinforcing rule-breaking behavior in a child, rather than listing the types or numbers of rules broken (fire setting vs. lying), to create a category such as “Conduct Disorder.”⁴¹

Family/Systems Theory

Despite caveats within the text of the DSM as to the difficulty of satisfactorily defining the boundaries of the term mental disorder, it is clear that this diagnostic system identifies mental disorders as occurring solely “in the individual” (pp. xxx–xxxi).⁵ Critics operating from a family therapy or systems orientation have questioned this fundamental assumption. The DSM-IV-TR currently allows problematic relationship patterns, while not recognized as mental disorders, to be listed in the multiaxial system. This can be done using items from “Other Conditions That May Be a Focus of Clinical Attention” on Axis I to specify that problems exist in a particular relationship (e.g., “Parent-Child Relational Problem,” or “Partner Relational Problem”). Relational difficulties could also be recognized in a more quantitative manner by using the proposed Global Assessment of Relational Functioning (GARF) Scale from Appendix B as a supplement to Axis V. However, use of these strategies in the absence of an individual disorder listed on Axis I or II may result in the practical problem of difficulty receiving third-party reimbursement for treatment, no matter how impairing the relational difficulties may be for the individuals involved.⁴ While some have argued that this problem invites increased attention to the limitations of the DSM system or to creating alternative systems of assessment,^34,42 others have proposed that disordered family systems could be categorized and operationalized with reliable diagnostic criteria based on research, perhaps becoming included on Axis II or elsewhere within a revised DSM.^43,44

Evolutionary/Developmental Approaches

The issue of environmental influence on the individual raises the possibility of reconsidering psychiatric diagnosis from evolutionary, adaptational, and neurodevelopmental perspectives.^24,25,45 This issue may be particularly vexing in the area of child psychiatry; while children’s distress may often arise from impaired adaptation, trauma, or a lack of fit between developmental needs and environmental resources, these early experiences may also go on to affect neurobiological development.²⁵ In turn, temperament and biological substrate help shape future experience and the emergence of symptomatology. Knapp and Jensen⁴⁶ argue that a useful diagnostic scheme for children and adolescents must capture not only symptoms, but also developmental stage and the child’s adaptation to his or her sociocultural context. They suggest adding a Relationship Axis or descriptive material within the DSM text that builds on the Diagnostic Classification of Mental Health and Developmental Disorders of Infancy and Early Childhood.⁴⁷

Dimensional Models

The introduction to DSM-IV-TR briefly discusses the limitations of the categorical model used to organize the DSM and the possibility of incorporating dimensional approaches in future editions. Despite the clinical and administrative complications that such a major change would likely entail,⁴⁸ some authors have argued strongly that dimensional models using standardized instruments would aid the field in differentiating pathological from normal psychological function and creating more useful boundaries between conditions. Rather than particular diagnoses, Widiger and Clark³ imagine a revised DSM that would “consist of an ordered matrix of symptom-cluster dimensions.” There has been particular work along these lines in the area of personality disorders research, with various proposed systems of dimensional classification that may rate the degree of traits (such as extroversion, novelty seeking, persistence, mistrust, aggression, and social avoidance).^21,44,49

Cultural Models

As the DSM becomes more widely used internationally and there is increasing attention to cultural diversity in the United States, researchers have argued the need to better integrate cultural matters into DSM-V.⁵⁰ For example, Lewis-Fernández and colleagues⁵¹ delineate how the syndrome of ataque de nervios found in many Spanish-speaking cultures both overlaps with and differs from panic disorder, psychiatric disorders that feature dissociative symptoms, and normal reactions to stressful situations. Recognition of this syndrome, and of its inability to be simply “translated” into one of the DSM diagnoses, may have important implications for culturally appropriate treatment. By contrast, depressed Chinese-Americans often have somatic symptoms for a variety of reasons. When asked directly, however, many readily endorse the psychological symptoms of major depressive disorder as defined in the DSM, creating the possibility of negotiating acceptable and effective treatments⁵²; in this scenario, it may be more important to change the delivery of services than to change the diagnostic criteria themselves.