Aligning Bodies With Minds:
APA Presentation May 2009


Rebecca Allison, M.D., FACC, FACP

Chair, American Medical Association Advisory Committee
On Gay, Lesbian, Bisexual, and Transgender Issues
President-Elect, Gay and Lesbian Medical Association

As a physician who has successfully completed the process of transition from male to female, I find it strange and inappropriate that I may still be considered “mentally ill” by those who would take literally the diagnosis of “gender identity disorder” in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). I believe such a diagnosis is incorrect, and submit this review to support my belief.

I’d like to begin by offering an example from my medical practice in support of the idea that “outcomes do matter” – that a favorable clinical outcome, reproducible from one patient to the next, validates the medical and/or surgical treatment prescribed for the diagnosis.

I intend to show that a cardiologist may initiate a course of treatment based on a patient’s reported symptoms, despite a lack of objective findings to support the working diagnosis; and that the accuracy of such a diagnosis is confirmed by the favorable response to what some might consider “empiric” treatment.

So much of my practice centers on a relatively small number of diseases of the cardiovascular system. One of the most common conditions I see is called angina pectoris.

Angina Pectoris is simply Latin for “Chest Pain.” But it’s a specific type of discomfort: mid-chest; radiates to the neck, jaw, or left arm; worse with physical or emotional stress. It’s relieved by rest or by the medication nitroglycerin.

When I see a patient with such typical symptoms, I know that it’s very likely he or she has interference with blood flow to the heart through clogged coronary arteries. I begin a series of tests including electrocardiograms, stress testing, and cardiac catheterization. Usually – over 95 percent of the time – the blockages are identified and managed appropriately with medications or surgery.

However, a small number of persons will have completely normal testing – no blockages at all. Even the electrocardiograms and stress testing do not indicate a physical cause for their symptoms. My next step for these persons is to look for other causes for their pain, with evaluation of the esophagus, GI tract, or musculoskeletal system. Even after this evaluation is complete, there will remain a group of people who have no discoverable cause for their pain, yet it is very real, severe, and sometimes disabling. What to do then?

I can tell you that what a cardiologist does then is treat the patient. We treat with medications, just as we would use for a patient with documented coronary artery disease. We give long acting nitroglycerin, calcium channel blockers, and aspirin. And our treatment works! The patients report prompt improvement. They are able to go about their normal lives without the anxiety and fear of the disabling pain. Outcomes do matter!

Numerous review articles in the cardiology literature confirm the effectiveness of medical therapy for the “syndrome of chest pain with normal coronary arteries.”

It has been suggested that this syndrome may be due to coronary vascular spasm, particularly in the small vessels of microscopic size. This is not a condition which can be objectively documented with any ease or safety, and so we do not perform provocative tests to induce spasm and “prove” our diagnosis, since such induced spasm could be harmful to the patient. We accept the diagnosis based on the response to our treatment. We name the condition “Cardiac Syndrome X,” which simply shows cardiologists aren’t very imaginative when it comes to original names. If we are more creative, we call it “microvascular angina,” and either way we give it an ICD Code of 413.9.

Notice that we do not call microvascular angina a mental disorder, although it has been reviewed frequently in the psychiatric literature, due to the high prevalence of serious anxiety in persons suffering from this condition. (I dare say, if I had frequent, severe, chronic chest pain, I’d be anxious too.) IF we use anti-anxiety medications, they are secondary to the primary medical treatment, and we would not prescribe them in lieu of nitroglycerin.

So, let’s just suppose someone comes to see me in my office, and he tells me, “Dr. Allison, I keep having these chest pains.” Suppose, after going through the complete evaluation, I tell him, “There’s nothing physically wrong. Perhaps we just need to help you adapt to your current circumstances. Let me give you a benzodiazepine [a mild tranquilizer].” What will he say? He most likely would tell me, “Listen here, either you give me some nitroglycerin, or I am going to find someone who will.” And he would be entirely correct. Because he knows that nitroglycerin will give him a good outcome, and he knows that outcomes matter.

Like microvascular angina, Gender Identity Disorder (or Gender Variance) is a diagnosis which is not based on objective findings. There’s no blood test for Gender Variance. There’s no chromosome analysis, no radiographic diagnosis, no nuclear scan. Pathologic changes in the hypothalamus, such as the “bed nucleus of the stria terminalis” or BSTc, are of academic interest but cannot be applied to living patients.

Some have suggested that a magnetic resonance image of the brain may show an appearance of the corpus callosum in male-to-female persons which is similar to genetic females. Even if a statistically significant correlation were to be proven, however, it would not have one hundred per cent sensitivity, meaning that some persons with true Gender Variance would have negative findings on MRI examination.

The best diagnosis remains subjective, the history as reported by the patient: a constant awareness of a sense of self which is not congruent with one’s physical body and the expected social role associated with that body.

Or, if you prefer, “a strong and persistent cross-gender identification;” and “persistent discomfort about one’s assigned sex or a sense of inappropriateness in the gender role of that sex.” You may recognize these as Criteria A and B of Gender Identity Disorder in the DSM-IV.

What about Criterion D, you ask? I maintain that “Clinically significant distress or impairment in social, occupational, or other important areas of functioning” should not be a necessary criterion for a diagnosis of Gender Variance. Many young persons are blessed with enlightened parents who facilitate their transition and help them avoid all the distress and impairment which was unavoidable for those of us who transitioned years ago. That is not to say that persons with Gender Variance no longer experience distress. We still do, but it is not a primary component of Gender Variance. It is secondary to the rejection we experience from family, friends, employers, and religious organizations.


What should be the appropriate treatment for persons who experience Gender Variance? Just as with microvascular angina, outcomes matter!

What constitutes a good outcome? Surely it is a well adjusted person, able to function socially and professionally, with no incongruity between sense of self and physical body.

What treatment results in a good outcome?

Does behavior modification result in a good outcome? “Reparative Therapy”? Aversion therapy? Electroconvulsive therapy? Antipsychotic drugs? What data suggest that persons undergoing such treatments and remaining in their birth sex role experience a good outcome? I would suggest that no such data exist.

What about psychotherapy? Is it helpful as an isolated treatment modality, in persons who are not going through transition?

On a more positive note, does psychotherapy, or any long term counseling relationship, help produce a good outcome in persons who are going through transition? Even so, do some persons who complete transition without the need of therapy experience equally good outcomes?

Numerous published studies document the outcomes of transition, and these studies are consistent in their findings.

Cohen-Kettenis and colleagues (Journal of the American Academy of Child and Adolescent Psychiatry, 1997) interviewed 22 consecutive adolescent patients of their Netherlands gender clinic who underwent sex reassignment surgery. Postoperatively they reported no gender dysphoric symptoms and were socially functioning well. None of the patients expressed any feelings of regret over transition.

From the same clinic, Smith and colleagues (Psychological Medicine, 2005) followed 162 adults (146 male to female, 76 female to male) who completed medical and surgical reassignment in the course of transition. They found that body image scores and psychological functioning were significantly improved. Only 1.6% of patients expressed any regrets.

Krege and colleagues at the University of Essen (BJU International, 2001) found no regrets among 66 male to female persons followed between 1995 and 2000.

Lawrence (Archives of Sexual Behavior, 2003) evaluated 232 male to female patients who had sex reassignment surgery between 1994 and 2000. No patient reported outright regret. Interestingly, compliance with the requirements for sex reassignment surgery as outlined in the WPATH Standards of Care was not associated with more favorable subjective outcomes.

At its 2008 Annual Meeting, the American Medical Association adopted Resolution 122 regarding treatment of Gender Identity Disorder. The AMA noted, among other concerns, that “An established body of medical research demonstrates the effectiveness and medical necessity of mental health care, hormone therapy and sex reassignment surgery as forms of therapeutic treatment for many people diagnosed with GID,” and “Health experts in GID, including WPATH, have rejected the myth that such treatments are ‘cosmetic’ or ‘experimental’ and have recognized that these treatments can provide safe and effective treatment for a serious health condition.” The document notes “Delaying treatment for GID can cause and/or aggravate additional serious and expensive health problems, such as stress-related physical illnesses, depression, and substance abuse problems, which further endanger patients’ health.”

The AMA concluded with the Resolution “That our American Medical Association support public and private health insurance coverage for treatment of gender identity disorder as recommended by a physician.”

As I review these documents, I see evidence of the effectiveness of the transition process. Persons who complete transition are more likely to be well-adjusted, successful in work and with social relationships, and without regrets. Our own life experiences, as well as those of our friends and colleagues, confirm these successful outcomes. Transition works! Outcomes matter!

The effectiveness of medical and surgical treatment for Gender Variance should mandate a medical diagnosis and inclusion in the ICD, rather than the DSM. Again, this does not imply that persons with Gender Variance do not benefit from behavioral health care. Many such persons experience depression, anxiety, or adjustment reactions due to the stress of rejection. These conditions are not, however, central to the diagnosis of Gender Variance. They are identical to the same conditions seen in persons who do not have Gender Variance. They will likely respond to appropriate counseling and/or medical management, but such treatment is adjunctive to the primary concern of working to achieve successful transition.

If Gender Variance were not a psychiatric diagnosis, then what other roles can the psychiatrist or psychologist play in the care of persons with Gender Variance?

We feel that one important role a therapist can perform is differential diagnosis. While the large majority of persons begin the transition process appropriately, there are a few who may have other psychological conditions which they confuse with Gender Variance. If these persons can be identified and directed to more appropriate courses of action before they take irreversible steps, we will not read about them as examples of regret for transition.

As we know, the WPATH Standards of Care require letters of approval from a behavioral health practitioner before a person may have sex reassignment surgery. While all North American surgeons, and most surgeons in Europe, require such letters, there are many doctors in other parts of the world who do not have such a requirement. The numbers of persons with Gender Variance who travel to these surgeons without going through a therapy relationship are increasing. In the future we may expect to see outcome studies relative to satisfaction or regret for transition without therapy.

IN CONCLUSION, I have used a common example from my practice of cardiology to illustrate that a diagnosis can be correctly made, and treatment can be successfully initiated, based on subjective symptoms as reported by the patient, without confirmation by specific diagnostic testing. Using the example of microvascular angina as my model, I contend that the same principles may be applied to the condition I call Gender Variance (rather than Gender Identity Disorder). It is not necessary to achieve certain measurements on a diagnostic image of the brain to know that a physical condition exists and is treatable by physical (medical and surgical) measures. The results of treatment are obvious and measurable. Hormone therapy works. Surgery to modify primary and secondary sex characteristics works. Transition works. Transition produces good outcomes, and outcomes matter.

In a perfect world, a diagnosis of Gender Variance, which does not carry the stigma of “disorder,” might exist as a medical condition in the ICD. Medical doctors could treat such persons with appropriate hormone management; surgeons could perform the operations essential for a normal life in the appropriate gender; and all such treatments would be covered by health insurance.

In a perfect world, psychiatrists and psychologists would appropriately manage the anxiety, depression, or other emotional conditions which occur in persons who have Gender Variance. These conditions would have appropriate codes in the DSM, but Gender Variance would not.

In a perfect world, psychiatrists might still treat persons WITH Gender Variance; but they would not treat persons FOR Gender Variance.

©2009 Rebecca Anne Allison

blog index