Box Turtle Bulletin

Gays Excluded From Some Clinical Trials

It’s a common practice to limit participants in studies according to specific parameters set for by the researchers conducting the studies. Typically, these parameters may include such factors as age, gender, health history, ethnicity and other factors which may be specifically relevant to the study. A study of diabetes among Native Americans, for example, would be restricted to Native American populations. But three researchers with the Fox Chase Cancer Center in Philadelphia have learned that many studies restrict LGBT people from participating, and it appears that in many cases these exclusions may be unjustified.

According to a research letter published in the March 18 edition of the New England Journal of Medicine, Drs. Brian L. Engelston, Roland L. Dunbrack, Jr., and Michael J. Hall recently encountered some study proposals which explicitly excluded people in same-sex relationships. They decided to try to understand how common that exclusion is and what effects it may have in the body of research. They began by performing exploratory searches of the ClinicalTrials.gov database (which contains information on more than 80,000 studies) to identify the types of studies gay men and women were more likely to be excluded. This is what they found:

We found that when we performed searches using the terms “couples,” “erectile dysfunction,” and “hypoactive” (related to hypoactive sexual disorder), we identified 243 studies, of which 37 (15%) had explicit exclusionary language (Table 1). In these 243 studies, the results of Fisher’s exact tests indicated that industry-sponsored trials, multiregion trials (according to census definitions), and phase 3 trials were the most likely to exclude lesbians and gay men (P<0.05). Other variables, such as the year the study opened, whether the study is open to those over 65 years of age, and whether the study accepts healthy volunteers, were not associated with exclusionary criteria.

The researchers found that such exclusions weren’t uniform across the board. For example, when they used the search term “asthma” to examine 1019 studies, they found that none of them excluded lesbians or gay men.

It’s unclear what effect the exclusion of gays and lesbians from these three specific study areas would have. Studies identified using the search term “Couples”, for example, bring up 142 results covering a wide range of topics, including breast cancer and post-traumatic stress disorder. The search term “Erectile Dysfunction” brought up studies in prostate cancer and arterial disease. If a particular study intended to observe the effects that specific relationship dynamics might might have on, for example, breast cancer survivorship, it’s concievable that the researchers may wish to try to isolate that study’s results from possible differences such dynamics which may exist between heterosexual and same-sex couples. But on the other hand, it would be useful to know whether relationship characteristics of lesbian couples have the same effect on breast cancer survivorship as relationship characteristics of heterosexual couples.

Unfortunately, this brief research letter does not explore the impact these exclusions may have in our medical knowledge. It only identifies a potential problem, and not its scope or impact. The researchers limited their search to studies from which lesbians and gay men were excluded, and those from which they were not excluded. We don’t know if there were any studies in the database which focused exclusively on same-sex couples by excluding heterosexual couples. Nor do they quantify the number of studies which used relationship status or characteristics as an independent variable. These would be useful data points, but that information could only come from a larger, more extensive (and expensive) study.

But for the majority of medical studies, it is unlikely that these exclusions have much of a rational basis to begin with. One clinical trial of a drug to treat attention deficit–hyperactivity disorder required that participants be “in a reciprocal relationship with a person of the opposite sex,” the team found. It’s hard to imagine what kind of a justification would support excluding gays and lesbians from such a trial.

Egleston pointed out to the Associated Press that when researchers apply for federal grants for their work, they must explain why their study excludes participants based on gender, race or ethnicity. But no explanation for exclusion based on sexual orientation is needed. Furthermore:

Exclusion can become self-perpetuating: Researchers designing a study often “cut and paste” participation criteria from earlier trials on a similar subject. “It becomes the way it’s done,” and any bias gets repeated, Egleston said.

Our results indicate that exclusion of lesbians and gay men from clinical trials in the United States is not uncommon, particularly in studies with sexual function as an end point. It is likely that most gay and lesbian patients are unaware that their sexual orientation is being used as a screening factor for participation in clinical trials. Researchers should be held to careful scientific reasoning when they develop exclusion criteria that are based on sexual orientation.

Letters to the editor are not peer-reviewed.