Clinical diagnosis: A marker for disease?

In the science of psychiatry, a careful logic has linked items of phenomenology to diagnoses and hypothesized disease constructs. However, the resultant improved reliability of diagnostic assessment does not directly ensure the specificity and validity of diagnosis, and diagnoses are often thought of as identical to diseases. Although specificity is enhanced by the use of patterns of items of phenomenology operationalized into diagnoses to classify psychiatric disease, evidence suggests that diagnostic heterogeneity remains significant, limiting specificity. Furthermore, diagnostic concepts may bias assessments of phenomenology. Theoretically, psychiatric diseases are pathological states of neurobiological, social, and psychological processes and interactions. Although advances in social and psychological science continue, neurobiological science is advancing rapidly and is expected to contribute substantially to validity. But the relative psychometric distance and sources of measurement error between these underlying aspects of diseases and manifest phenomenology suggest that the use of clinical diagnoses, derived from phenomenology, may not be optimal as a standard from which to test neurobiological hypotheses, limiting validity. The authors submit that an approach utilizing measurement theory to combine standardized assessment of symptom items with neurobiological observations for classification will assist in efforts to understand the complexities of psychiatric disorders. Incorporation of neurobiological (and other) measures into the diagnostic logic has potential to extend and refine definitions of psychiatric diseases. A subtle shift of paradigm is suggested, in which clinical diagnosis is thought of not as the definition of disease, but as a marker for disease.