Leave PMDD in DSM Appendix

This letter is in response to the article “PMDD May Leave DSM Appendix and Become New Diagnosis” in the May 18 issue.

I am not a researcher and have no career interest in the inclusion or exclusion of premenstrual symptoms in DSM. I have been involved with the issue since DSM-III-R was developed during my term as chair of the APA Committee on Women. There were very good reasons that premenstrual dysphoric disorder (PMDD) was put in the appendix of DSM at that time, and the same reasons explain why it should stay there.

The study of this issue entails an insuperable methodological problem. No matter what exclusion criteria you include and what instruments you use, you cannot blind women to their menstrual status nor can you obviate the culturewide belief in PMS. Americans strongly believe that the menstrual cycle causes unacceptable changes in women’s moods and behavior. Time and again, both in research and in clinical practice, more women present with symptoms they believe to be premenstrual than actually have symptoms limited to the premenstruum. When my 9-year-old (and distinctly not pubertal) granddaughter was in a bad mood at a restaurant recently, the waitress opined that she must be “PMS-ing.”

The failure of mood changes to track with the menstrual cycle is reflected both in the failure of attempts to validate the diagnosis by altering the cycle with exogenous hormones (which was attempted at NIMH) and by the proposal to loosen the criteria for the linkage between symptoms and menstrual-cycle status. The proposed criteria would allow for a woman to be symptomatic for the majority of days of each cycle and for symptoms neither to arise nor end at the times hormone levels change during the cycle. At no time is it required that symptoms be absent. At best, these criteria would define cyclical variations in an existing mood disorder.

Women have good reason to be irritable and unhappy. Women assume a disproportionate burden of household and dependent-management and nurturing responsibilities, receive lower pay for the same work, and are underrepresented in positions of authority. At the same time, women are expected to be pleasant, compliant, and helpful. Anger and irritability are less acceptable in women than in men. Women who are not able to fulfill these expectations can and do become irritable, anxious, and depressed. They are uncomfortable with those reactions and seek reasons for them. They greatly prefer “hormonal” reasons to “psychiatric” reasons. Therefore, Eli Lilly changed the name of fluoxetine from Prozac to Sarafem when marketing it for PMS.

Research into the relationships of hormones to deleterious changes in mood and behavior should include both sexes. The effects of male hormones can be readily discerned from the auto insurance rates for adolescent males, who cause a greatly disproportionate percentage of collisions and resulting injuries and deaths. If there are to be any specifically hormonally related disorders diagnostically singled out for women, they should certainly include postpartum depression and psychosis, which are scientifically well established but poorly understood by the courts, landing desperately ill women in prison rather than in treatment, after tragic sequelae.

There is very little to be gained by enshrining PMS in our diagnostic system. Symptomatic women are and will continue to be treated—with the same treatments that would be offered were they to be diagnosed with the disorders most of them actually have. However, there is a lot to be lost in terms of social attitudes toward, and the well-being of, all women—and girls like my granddaughter.

Nada Stotland, M.D., M.P.H.Chicago, Ill.

I appreciate the letter written by my friend and colleague, Dr. Stotland. I share her concerns about women’s place in society and the burdens women must shoulder. Even in 2012, we continue to fight for pay equality and to live in homes that are free of interpersonal violence. As chair of the subcommittee that evaluated whether the category premenstrual dysphoric disorder (PMDD) should be placed in DSM-5 or remain in the appendix, my charge was to work with experts and evaluate the scientific evidence for such a change. Theoretical, political, and sociological arguments aside, our evaluation found ample scientific evidence, grounded in biological, epidemiological, and treatment results, for the movement of the category to the manual. Much of this evidence was outlined in the May American Journal of Psychiatry (AJP) and written by our sub-work group.

As a brief rebuttal, I would like to outline several points. First, recall bias is a well-known phenomenon in medical research and is not limited to PMDD. Thus, it is not surprising that retrospective and prospective reports do not always agree. I daresay that there would be differences in the case rate for a number of diagnoses if prospective symptom reports rather than retrospective reports only were required.

Nevertheless, epidemiological studies that blind women to the menstrual cycle focus of the research find that a small subset of women (1 percent to 3 percent) have moderate to severe symptoms that begin during the premenstrual phase of their cycle and offset with menses or shortly thereafter. These studies included daily symptom charts from participants and prospectively documented symptom expression over the course of several months. As a corollary, this work repeatedly shows that the symptoms outlined in the category for PMDD are frequently endorsed as the most problematic symptoms.

Second, the criteria require that a woman have a postmenstrual week that is either completely symptom free or characterized by minimal symptoms. The reason for inclusion of “minimal symptoms” is that women are subjected to multiple stressors, medical conditions, and life events that can affect mood at any point in their menstrual cycle. A headache during the postmenstrual week should not disqualify a woman from the diagnosis of a repeated and impairing premenstrual condition characterized by a constellation of symptoms.

Third, Dr. Stotland points to an earlier study from NIMH that found women’s premenstrual symptoms were entrained to their usual premenstrual phase even if the underlying biology was changed. However, this same group has shown that the addition of either estrogen or progesterone, rather than placebo, can trigger symptoms in women with PMDD but not women free of PMDD, who took medication to stop ovulation and the subsequent hormonal changes that occur in the luteal phase of the cycle. Additionally, gonadotropin-releasing-hormone agonists that cease menstrual cyclicity are effective treatments for PMDD.

Fourth, despite the lack of relationship between hormone levels and mood that Dr. Stotland correctly notes, research shows that serotonin reuptake inhibitor treatment limited either to the luteal phase of the menstrual cycle, or even the last week of the menstrual cycle, is more effective than placebo for treatment of PMDD symptoms. Such “targeted” treatment is not typical of other mood or anxiety disorders and suggests a different psychopathology for PMDD.

In decades past, women with severe menstrual cramps and pain were told that it was “all in their head” or that “they are conflicted about being a woman.” We now know more about the physiology of the uterus and that contractions can lead to pain. Simple administration of a prostaglandin inhibitor provides profound relief for many women. Are we to go back to the paternalistic attitude of dismissing women’s complaints again by ignoring our patients and the accumulated science that now relates to PMDD? If a person calls another individual “psychotic” as an insult, are we to ignore psychotic illnesses? Instead, we should address stigma related to mental health issues, whether they relate to depression, psychosis, or other arenas and follow where the science leads us. There is much to be gained by recognizing PMDD, and that is the relief I see when a patient feels understood and experiences successful treatment.

“Premenstrual Dysphoric Disorder: Evidence for a New Category for DSM-5” is posted at http://ajp.psychiatryonline.org/article.aspx?articleid=1065356 .