Moreover, clinicians are greatly dependant upon parents’ perceptions of the nature of the presenting problems. Parents may differ, for example, in the extent to which they perceive very active behavior as problematic versus being “all boy,” or view a quiet and introspective child as being “shy” versus having a “social disorder.” Subjectivity in diagnosis also is manifest in the variable diagnoses received by white patients and individuals who are members of ethnic minorities. African American patients with manic-depressive illness, for example, have been found to be at higher risk for being misdiagnosed as having schizophrenia than are whites (Bell and Mehta, 1980, 1981; Mukherjee et al., 1983). Such racial differences have tended to disappear when structured interviews rather than clinical diagnoses are used (Adebimpe, 1994; Simon and Fleiss, 1973), suggesting the existence of differences in clinician assessment by patient ethnicity.
A number of factors account for variations in diagnosis of M/SU illnesses. Foremost, in contrast with general health conditions, relatively few laboratory, imaging, or other physical measures can detect the presence of a mental illness or substance dependence.10 Accurate diagnosis relies instead upon descriptive methods whereby patients or their caregivers inform clinicians about symptoms, and clinicians apply their expert judgment to determine whether diagnostic criteria for a condition are met. Moreover, individual clinicians vary in the breadth, depth, and theoretical basis of their training (see Chapter 7). Because diagnosis requires a subjective interpretation of reported symptoms, these variations result in inconsistency and unreliability in how individuals are diagnosed. Administrative rules and financial incentives can also influence diagnostic practices.
Criteria for diagnosing M/SU problems and illnesses reliably are found in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM), which has been a highly significant milestone in the diagnosis and treatment of mental and substance-use problems and illnesses and is now in its revised fourth edition (DSM-IV-R). However, adherence to these guidelines is not uniform. Fully 56 percent of primary care physicians in Michigan surveyed in 2002 reported that they did not use DSM criteria to diagnose ADHD (Rushton et al., 2004). This may be because DSM-IV is not easy to use in primary care settings, in part because of its focus on specialty care, its length, and its complexity (Pincus, 2003).
Several different approaches have been undertaken to improve the accuracy of diagnosis of M/SU illnesses. System-level interventions, such as routine screening, have been shown to help (Gilbody et al., 2001; Rollman et al., 2001). Structured diagnostic interview instruments have also been developed to reduce variability in information gathering and biases that can inadvertently influence individual clinicians’ decision making. While these