Key to the development of any effective plan of treatment for mental health and substance abuse patients is the ascertainment of an accurate diagnosis of the problem(s) for which the patient is seeking intervention. As in the past, assisting in the differential diagnosis of psychiatric disorders continues to be one of the major functions of psychological assessment (Meyer et al., 1998). In fact, managed behavioral health care organizations (MBHOs) are more likely to authorize reimbursement of testing for this purpose than for most other reasons (Maruish, 2002). Assessment with well-validated, reliable psychological test instruments can provide information that might otherwise be difficult (if not impossible) to obtain through psychiatric or collateral interviews, medical record reviews, or other clinical means. This is generally made possible through the inclusion of (a) test items representing diagnostic criteria from an accepted diagnostic classification system, such as the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American PsychiatricAssociation, 1994)or(b) scales that either alone or in combination with other scales have been empirically tied (directly or indirectly) to specific diagnoses or diagnostic groups.
In most respects, considerations related to the use of psychological testing for diagnostic purposes are the same as those related to their use for screening. In fact, information obtained from screening can be used to help determine the correct diagnosis for a given patient. As well, information from either source should be used only in conjunction with other clinical information to arrive at a diagnosis. The major differentiation between the two functions is that screening generally involves the use of a relatively brief instrument for the identification of patients with a specific diagnosis, a problem that falls within a specific diagnostic grouping (e.g., affective disorders), or a level of impairment that falls within a problematic range. Moreover, it represents the first step in a process designed to separate those who do not exhibit indications of the problem being screened for from those with a higher probability of experiencing the target problem and thus warrant further evaluation for its presence. Diagnostic instruments such as those just mentioned generally tend to be lengthier, differentiate among multiple disorders or broad diagnostic groups (e.g., anxiety disorders vs. affective disorders), or are administered further along in the evaluation process than is the case with screeners. In many cases, these instruments also allow for a formulation of description of personality functioning.
There are many instruments available that have been specifically designed to help identify individuals with disorders that meet a diagnostic classification system's criteria for the disorders). In the vastmajority of the cases, these types of tests will be designed to detect individuals meeting the diagnostic criteria of DSM-IV or the 10th edition of the International Classification of Diseases (ICD-10; World Health Organization, 1992). Excellent examples of such instruments include the Millon Clinical Multiaxial Inventory-III (MCMI-III; Millon, 1994), the Primary Care Evaluation of Mental Disorders (PRIME-MD; Spitzer et al., 1994), the Patient Health Questionnaire (PHQ, the self-report version of the PRIME-MD; Spitzer, Kroenke, Williams, & Patient Health Questionnaire Primary Care Study Group, 1999); the Mini-International Neuropsychiatry Interview (MINI; Sheehan et al., 1998).
Like many of the instruments developed for screening purposes, most diagnostic instruments are accompanied by research-based diagnostic efficiency statistics—sensitivity, specificity, PPP, NPP, and overall classification rates—that provide the user with estimates of the probability of accurate classification of those having or not having one or more specific disorders. One typically finds classification rates of the various disorders assessed by any of these types of instrument to vary considerably. For example, the PPPs for those disorders assessed by the PRIME-MD (Spitzer et al., 1999) range from 19% for minor depressive disorder to 80% for major depressive disorder. For the self-report version of the MINI (Sheehan et al., 1998), the PPPs ranged from 11% for dys-thymia to 75% for major depressive disorder. Generally, NPPs and overall classification rates are found to be relatively high and show a lot less variability across diagnostic groups. For the PRIME-MD, overall accuracy rates ranged from 84% for anxiety not otherwise specified to 96% for panic disorder, whereas MINI NPPs ranged from 81% for major depressive disorder to 99% for anorexia. Thus, it would appear that one can feel more confident in the results from these instruments when they indicate that the patient does not have a particular disorder. This, of course, is going to vary from instrument to instrument and disorder to disorder. For diagnostic instruments such as these, it is therefore important for the user to be aware of what the research has demonstrated as far the instrument's classification accuracy for each individual disorder, since this may vary within and between measures.
There are a number of instruments that, although not specifically designed to arrive at a diagnosis, can provide information that is suggestive of a diagnosis or diagnostic group (e.g., affective disorders) or can assist in the differential diagnosis of complicated cases. These include multiscale instruments that list symptoms and other aspects of psychiatric disorders and ask respondents to indicate if or how much they are bothered by each of these, or whether certain statements are true or false as they apply to them. Generally, research on these instruments has found elevated scores on individual scales, or patterns or profiles of multiple elevated scores, to be associated with specific disorders or diagnostic groups. Thus, when present, these score profiles are suggestive of the presence of the associated type of pathology and bear further investigation. This information can be used either as a starting place in the diagnostic process or as additional information to support an already suspected problem.
Probably the best known of this type of instrument is the Minnesota Multiphasic Personality Inventory-2 (MMPI-2; Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989). It has a substantial body of research indicating that certain elevated scale and subscale profiles or code types are strongly associated with specific diagnoses or groups of diagnoses (see Graham, 2000, and Greene, 2000). For example, an 8-9/9-8 highpoint code type (Sc and Ma scales being the highest among the significantly elevated scales) is associated with schizophrenia, whereas the 4-9/9-4 code type is commonly associated with a diagnosis of antisocial personality disorder. Similarly, research on the Personality Assessment Inventory (PAI; Morey, 1991, 1999) has demonstrated typical patterns of PAI individual and multiple-scale configurations that also are diag-nostically related. For one PAI profile cluster—prominent elevations on the DEP and SUI scales with additional elevations on the SCZ, STR, NON, BOR, SOM, ANX, andARD scales— the most frequently associated diagnoses were major depression (20%), dysthymia (23%), and anxiety disorder (23%). Sixty-two percent of those with a profile cluster consisting of prominent elevations on ALC and SOM with additional elevations on DEP, STR, and ANX were diagnosed with alcohol abuse or dependence.
In addition, there are other well-validated, single- or multi-scale symptom checklists that can also be useful for diagnostic purposes. They provide means of identifying symptom domains (e.g., anxiety, depression, somatization) that are problematic for the patient, thus providing diagnostic clues and guidance for further exploration to the assessing psychologist. The BDI-II and STAI are good examples of well validated, single-scale symptom measures. Multiscale instruments include measures such as the Symptom Checklist-90-Revised (SCL-90-R; Derogatis, 1983) and the SymptomAssessment-45 Questionnaire (SA-45; Strategic Advantage, Inc., 1996).
Regardless of the psychometric property of any given instrument for any disorder or symptom domain evaluated by that instrument, or whether it was developed for diagnostic purposes or not, one should never rely on test findings alone when assigning a diagnosis. As with any other psychological test instruments, diagnosis should be based on findings from the test and from other sources, including findings from other instruments, patient and collateral interviews, reviews of psychiatric and medical records (when available), and other pertinent documents.
Was this article helpful?
Suffering from Anxiety or Panic Attacks? Discover The Secrets to Stop Attacks in Their Tracks! Your heart is racing so fast and you don’t know why, at least not at first. Then your chest tightens and you feel like you are having a heart attack. All of a sudden, you start sweating and getting jittery.