The article by Regier et al1 on the limitations of diagnostic criteria and assessment instruments for mental disorders raises important and critical questions. In particular, the use of diagnostic criteria that were developed primarily for clinical decision-making, in epidemiology and other research projects, predictably raises questions about the appropriate choice of thresholds and the resulting rates of false positives and false negatives.2 The article was nicely balanced by commentaries by Frances3 and Spitzer.4 In particular, Frances' discussion of the DSM-IV "clinical significance criteria" was largely undiscussed in the Regier article. It would be useful to understand more explicitly how the DSM-IV clinical significance criteria could have been operationalized in the Epidemiologic Catchment Area study and Diagnostic Interview Schedule, and how they might have affected the data they presented.
Perhaps more importantly, we agree with Spitzer that the authors mistakenly conflate "need for treatment" with "presence of a disorder." Although there is considerable overlap between these 2 concepts, they are certainly separable and have different public health and clinical implications. Whereas the presence of a disorder depends on the establishment of a dysfunction in the individual that results in distress or disability, determining the need for treatment necessitates consideration of additional nondiagnostic factors, such as the presence of social supports and the availability of effective therapeutic options. This is true throughout all of medicine.
Where psychiatry differs from the rest of medicine is in the paucity of objective measures that might serve to establish the presence or absence of a disorder. This is further complicated by the fact that the majority of psychiatric syndromes are made up of symptoms which, at the milder end of the severity spectrum, are part and parcel of normal daily functioning. Thus, the DSM-IV clinical significance criterion serves to remind the clinician of the need to differentiate "normal" psychological, emotional, and behavioral symptoms (for example, moderate anticipatory anxiety associated with air travel) from more pathological states (intense airplane phobias that interfere with job requirements).
Recently, one of us (H.A.P.) received a letter from Robert Spitzer asking whether we were satisfied with having added the clinical significance criteria to many of the DSM-IV criteria sets. In responding to him, the answer was both no and yes. We are not satisfied because, in fact, the criterion is imprecise and not easily measured in a standardized way. The development of newer approaches to measurement—in both clinical and research situations—poses a challenge for the future (and is, in part, being examined in the American Psychiatric Association development of the Handbook of Psychiatric Measures and Outcomes). Nonetheless, overall, we are pleased with the decision because it places the decision-making about "caseness" in the hands of the individual clinician in evaluating specific patients.
The introduction to DSM-IV (which too often goes unread) takes great pains to disabuse people of the notion that diagnoses can be made with the mindless checking off of symptoms and to encourage a healthy skepticism toward the rigid application of DSM-IV. As one of the participants of the DSM-IV process suggested, we should subliminally implant the word "think" on every page.
1.Regier
DAKaelber
CTRae
DSFarmer
MEKnauper
BKessler
RCNorquist
GS Limitations of diagnostic criteria and assessment instruments for mental disorders: implications for research and policy.
Arch Gen Psychiatry. 1998;55109- 115
Google Scholar 2.Zarin
DAEarls
F Diagnostic decision making in psychiatry.
Am J Psychiatry. 1993;150197- 206
Google Scholar 3.Frances
A Problems in defining clinical significance in epidemiological studies.
Arch Gen Psychiatry. 1998;55119
Google Scholar 4.Spitzer
RL Diagnosis and the need for treatment are not the same.
Arch Gen Psychiatry. 1998;55120
Google Scholar