Clinically significant distress or impairment: essential to a diagnosis or redundant?

by Joye C. Anestis

We've spent a good amount of time covering some potential diagnostic changes in the upcoming DSM-V.  The proposed changes discussed on PBB so far include altering the format of the mood and anxiety disorders, changing the PTSD diagnostic criteria, shifting to a dimensional model of personality disorders, removing some of the autism spectrum disorders, and creating a hierarchical model of eating disorders.  Besides diagnosis-specific changes, rumblings of broader changes are also underfoot. 

In an effort to reduce the false positive rate of diagnoses (i.e., to rule out individuals who met the diagnostic criteria for a disorder but whose symptoms are not harmful enough to be considered disordered), the DSM-IV added a clinical significance criterion to almost all symptom sets, requiring that the symptoms must cause clinically significant distress and impairment to warrant a diagnosis.  For example, criterion C of a major depressive episode states: "The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning" (APA, 2000, p. 356).  The authors of the DSM-IV were attempting to not over-pathologize milder forms of symptoms that may be present but aren't getting in the way of an individual's ability to function.  Despite the laudability of this intention, no field trials were conducted to test how effective this addition was in actually eliminating false positives.

One proposed change for the DSM-IV is the removal of this clinical significance criterion from the diagnostic criteria and making it a separate dimension.  There are two prominent arguments promoting such consideration.  First, some argue that the clinical significance criterion is redundant.  The wording of the symptoms themselves already ensure impairment and distress.  Second, significant distress or impairment can be part of a "normal" reaction to loss or stress - thus, the goal of eliminating false positives may not be being met.  Another reason the DSM-V Task Force is reportedly considering this change is to make the DSM-V  more compatible to the ICD (the International Statistical Classification of Diseases and Related Health Problems), the classification system used by the rest of the world, which avoids reference to role impairment in its criteria (it is also under revision, and the revision task force is reportedly considering completely eliminating any statements about functioning or disability).

A recent study in the American Journal of Psychiatry aimed to empirically assess the effectiveness of the clinical significance criterion to reduce community prevalence rates of major depression.  Specifically, authors Wakefield, Schmitz, and Baer wanted to test the redundancy hypothesis, proposing that neither distress or impariment would have a significant effect on the prevalence rate of major depression in the community.  A community sample was chosen for this study because they are the most likely to suffer from false positive rates.  It is less likely that a person presenting at a clinic would receive a false positive diagnosis of depression.

• Method: The authors obtained data from the National Comorbidity Survey Replication (NCS-R), a community-based epidemiological survey of the U.S.  For unexplained reasons, analyses were restricted to respondents aged 18-54 (n = 6,707).  The administration of the interviews, and subsequently the analyses, got a little complicated.  First, participants were given a screening questionnaire which asked questions about persistent sadness.  If a participant answered yes to any of the persistent sadness questions, then they were asked questions about clinical distress (e.g., How severe was your emotional distress?).  Those who reported distress were then evaluated for major depression.  A depression diagnosis required 2 weeks of 5 or more of the DSM-IV symptoms (DSM-IV criterion A).  The NCS-R did not utilize the DSM-IV's bereavement exclusion (an individuals cannot be diagnoses with a major depressive episode if they have lost a loved one within the previous 2 months), allowing for the possibility of more false positives than usual.  Questions regarding impairment were asked after the symptom interview.  A positive response to any of 10 distress and impairment questions satisfied the clinical significance criterion. 

• Results: Interestingly, even though an earlier DSM-III-R-based National Comorbidity Survey did not include a clinical significance criterion, the prevalence rate in the current sample (18.3%) was higher than that in the previous sample (15.2%).  The redundancy hypothesis for distress was strongly supported in this sample.  Of the respondents who reported at least 1 mood symptom for 2 weeks (n = 2,071), 97.2% (n = 2,016) satisfied the distress clinical significance criterion.  Of these 2,071 folks who were evaluated for distress, 60.5% (n = 1,254) met criteria for a diagnosis of major depression.  Of the 817 folks who did not meet diagnostic criteria for major depression, 93.5% satisfied the distress criterion.  The redundancy hypothesis was also strongly supported for the impairment clinical significance criterion.  Of the 1,542 participants who satisfied the Criterion A requirement, 96.2% (n = 1,487) satisfied the impairment criterion.

The findings reported by Wakefield and colleagues indicate that the clinical significance criterion does virtually nothing to distinguish normal sadness from major depression.  I am intrigued by this finding, as the requirement of clinically significant distress and impairment is often emphasized when clinicians are being trained to make diagnoses.  It's very rare to apply a diagnosis to an individual who is not distressed and/or imparied by the symptoms (for example, while a clinical significance criterion is included in the general diagnostic criteria for a personality disorder, some are not bothered by their symptoms - although others might be).  The authors are careful to point out that they do not support the idea that the distress/impairment criterion should be uniformly removed from all symptom sets in the DSM-V.  Instead they propose a disorder-by-disorder analysis of this issue (I find it hard to believe that this will actually happen...that's a lot of work!).  Based on this data (although replication is definitely needed), it does appear that the major depressive episode symptoms cover distress and impairment to the extent that a separate criterion isn't necessary (although I don't see how it would hurt anything to leave it in).  But I can definitely see the necessity of it for other diagnoses...some folks have specific phobias that don't interfere with their lives and some personality disorders could cause us to over-pathologize "different" behavior without such a criterion.  What do you think?  Are you aware of other data on this issue?  Can you think of other disorders that may really need a specific clinical significance criterion?

************************

If you would like to learn more about this and other topics discussed on PBB, Psychotherapy Brown Bag recommends that you consult their online store of scientifically-based psychological resources.

Joye Anestis is a doctoral candidate in clinical psychology at Florida State University.