The Significance of Clinical Significance Establishing the Boundary with Normal

Psychiatric symptoms are fairly ubiquitous in the general population-most normal people have at least one, many have a few. When present in isolation, a single symptom (or even a few) does not a psychiatric disorder make. Two additional conditions must also be met before a symptom can be considered to be part of a mental disorder.

First, there has to be a characteristic cluster of the symptoms--as laid out in the pertinent DSM criteria set. Isolated symptoms of depression, or anxiety, or insomnia, or memory difficulties, or attention problems, or whatever, are never by themselves sufficient to justify the diagnosis of a mental disorder.

Second-and our main subject here-the symptoms must cause clinically significant distress or clinically significant impairment in social or occupational functioning. This caveat is so important that it has been included as a separate item in the majority of the DSM criteria sets. It is not enough to have symptoms. . .  they also have to create serious problems in your life.

Why not define mental disorder just on the presence or absence of the characteristic cluster? Why was it felt to be necessary to also require distress or impairment?

Most of the DSM disorders present along a graduated spectrum of severity. At the severe end, the person’s suffering and the impairment occasioned by the symptoms are so obvious that there can be no possible doubt that the presentation qualifies as a mental disorder. But at the mild end of most disorders, there is no clear boundary distinguishing normality from mental disorder.

But how do we define what is clinically significant?

Unfortunately we don’t. This is a necessarily vague term with no precise markers. Deciding whether someone is experiencing enough distress or impairment to have a clinically significant mental disorder can be an inherently a tough and subjective judgment that has to be made without objective criteria.

Here are some hints that may help. First off, there are no obvious right answers. Accept the inevitability of at least some uncertainty in answering the question whether someone’s condition is severe enough to warrant diagnosis and possible treatment. This awareness leads to several important implications. Watchful waiting may be the best first step- -far. . .  better than jumping to a conclusion one way or the other. Tincture of time often provides a cheap, side-effect free cure, especially for milder conditions that have not lasted long and are not very impairing.

Next, this decision often requires other informants to supplement the patient’s subjective judgment. Some people are stoics and would literally have to be at death’s door before accepting that they are in trouble. At the other extreme are those who may seek a diagnosis and a treatment for what may be the expectable aches, pains, disappointments, and sufferings of everyday life.

In toss-up situations, it is useful to do a risk/benefit analysis of the pluses and minuses of making the diagnosis. The basic question boils down to “will making this diagnosis more likely help or hurt the patient?” All else being equal when the decision could go either way, it makes sense to provide a diagnosis when there is a recommended treatment proven to be safe and effective. And it make sense to avoid a questionable diagnosis if there is no proven treatment or if the available treatment has potentially dangerous side effects or complications.
 
It may be tempting to do a trial of treatment even if the presence of a diagnosis is not clear-cut. One possible rationale is that if the treatment makes the patient feel better, who cares whether the diagnostic criteria were fully met? Another rationale is that a positive response proves that the diagnosis was accurate and that the treatment was needed.

These are both inaccurate and misleading arguments. Patients with milder disorders have a very high placebo response rates-often about 50%, which is very close to the response rate achieved by medication. Immediately starting medicine for someone with a mild disorder prevents ever determining what caused the subsequent improvement (time and the placebo effect or the active ingredient in the medication?). Patients routinely misattribute improvement to medication effect and continue to take pills unnecessarily and for too long-- risking side effects. So for milder conditions, the best order of approach is time and watchful waiting first (an exercise regimen is always a good part of this package); psychotherapy next; and then medication as a third and last resort.
 
The bottom line: if the patient’s psychiatric status is betwixt and between-not all well, but not clearly disordered-don't jump to make a decision regarding diagnosis and treatment. Give it some time. Things usually sort out within a month or two of watchful waiting.
 
DSM-5 is threatening to remove the clinical significance requirement from the many criteria sets where it is currently included. This is a serious mistake that will likely exacerbate diagnostic inflation and the provision of unnecessary treatment. However imperfect and unreliable, clinical significance is a necessary gatekeeper that protects against excessive and unwarranted diagnosis.