May 012013
 

One way to reduce stress and cortisol - CC BY 2.0 - flickr user eamoncurry123Summary: Research now indicates that cross-sex hormone therapy is associated with a lower cortisol awakening response in trans people, regardless of attachment style. Many confounding variables, however, were present in this study.

Transgender people have long asserted that gender dysphoria can be extremely distressing and that transition, including hormone therapy, helps relieve that dysphoria. Hormone therapy is known to improve self-reported quality of life, as measured by questionnaire. To my knowledge no other study has looked at stress-related biological factors in trans people. Biological factors are important because self-report is notorious for validity problems. This study looked at one such biological factor, called the cortisol awakening response.

What is the cortisol awakening response? Readers of the blog may remember the last time I spoke about cortisol (paragraph #2). For those who don’t remember…. cortisol is a “stress hormone.” When we’re stressed, whether by speaking in public or running from a lion, cortisol is released. It helps our body be ready for immediate survival by increasing blood sugar and helping with metabolism. High cortisol levels over a long period of time can have many negative effects on health, including weakening the immune system. The cortisol awakening response is part of the daily cycle, when blood levels spike about 20-30 minutes after waking in the morning. The cortisol awakening response is larger in stressed people than in non-stressed people and can be affected by many things, including burn out, fatigue, aspirin, and sleep schedule. Awakening response is thought to be a good indicator of general stress levels and as a good indicator for stress-related disease risks.

Participants in this study were 70 trans people seen at the Gender Identity Unit of the University of Bari Psychiatric Department, roughly 64% trans women. All the participants had the same hormonal treatment; transdermal estradiol gel and cyproterone acetate (an anti-androgen) for trans women, intramuscular testosterone esters for trans men. They were assessed before hormone therapy and 12 months after starting hormone therapy. There was no significant difference in age, education, or occupation between the two groups.

The researchers measured perceived stress (a self-report of how stressed a person feels) in addition to the cortisol awakening response. The cortisol awakening response was measured by a blood test at 8:00am on three consecutive days, 1 hour after waking.

The results were striking. Before treatment, both perceived stress and cortisol levels were above the  “normal” range. After twelve months of hormone therapy, both were much lower and back within normal ranges. There were no statistically significant differences between trans men and trans women.

However there are a number of confounds for this study. Cortisol levels vary with sex hormones. For example, the cortisol levels of menstrual women will vary depending on which part of the menstrual cycle they’re in. Could cross-sex hormone therapy have caused this change in cortisol levels? Maybe, but then I’d expect there to be a difference between the trans men and trans women in this study and there weren’t.

The researchers also did not appear to attempt to control for other factors which could have impacted the cortisol awakening response. Changes in sleep patterns (e.g., naps) or sleep quality (e.g., a noisy environment) have effects on the cortisol awakening response. As far as I can tell the researchers did not screen for these changes.

Cortisol and stress were not the only things measured in this study. The researchers also looked at attachment styles. Attachment styles are a psychological concept. The idea is that when we are children our interactions with parents, and how they respond to our needs, affects the type of “attachment” we have. Attachment styles are secure or insecure. A secure attachment often results in happy adult relationships. Insecure attachments include avoidant, anxious, and unresolved/disorganized styles. Attachment styles may influence how we respond to stress, so they could have been a confound in this study if not examined.

The researchers determined the attachment style of the participants with a structured interview. They found that trans people are more likely to have an insecure attachment (70%) than the general population with no psychiatric diagnoses (44%). Attachment style did not, however, appear to be correlated with cortisol awakening response or perceived stress.

In other words, the relationship trans people have with their parents did not appear to affect the stress-reducing effects of hormone therapy.

I do not really understand why these researchers chose to examine attachment style in this study. I think that knowing attachment styles may be useful for therapy or for the development of effective variations on therapies for trans people. But I don’t feel that the inclusion of attachment style was sufficiently justified in this study. Why look at attachment and not, for example, socioeconomic status or social support? I would think either of those would be more likely to have an impact on stress levels than attachment.

On the whole: I think that the cortisol results of this study are decent validation of the anecdotal evidence from trans people themselves, but that the exploration of attachment style in this context is a red herring.

The abstract is publicly available.

Feb 102013
 

CC BY-ND 2.0 - TjookAs a group non-heterosexual people have poorer mental health than heterosexuals do. LGB (lesbian, gay, bisexual) folk have higher rates of mood/anxiety disorders, suicidal ideation/attempts, and substance use. Why? The dominant theory is minority stress: simply being a minority is stressful, especially if one is a minority who faces discrimination. Higher levels of stress are associated with poorer mental health. For LGB folk, the fact that their minority status is invisible is an additional factor. Research is conflicted on whether “coming out” improves mental health or not. While coming out in a supportive environment may improve mental health, coming out in a discriminatory environment may do the opposite.

Before I jump into the actual study, a little background on stress. Stress that comes from a psychological or social source is called psychosocial stress. Like all stress, it isn’t just psychological. There’s a biological component too! In the laboratory, there are at least two different ways of measuring stress. The first, and easier, is through cortisol. Cortisol is a stress hormone made from cholesterol that is released by the adrenal glands. It’s a small, non-polar molecule, so it passes right cellular membranes into every cell in the body. Because of this, cortisol can be measured in saliva, making its collection easer, cheaper, and less riskier in research than other measures. Cortisol also has a cyclical pattern; it’s lowest in the morning but rises in concentration through the day. Cortisol measurement is not without its problems. Levels of cortisol, and the reactivity of cortisol concentration to stress, varies between men and women, and between women of varying menstrual cycles and oral contractive use. It’s also worth noting that cross-sex hormones may be a confounding variable for cortisol testing, which is why this study did not include transgender people.

Another way of measuring stress requires a blood sample and lots of blood tests. Stress affects many different body systems. Everything from sex hormones to triglycerides to insulin can be affected, so those levels can be used to help detect stress levels in participants. Non-blood tests such a blood pressure may also be used. These non-cortisol factors were referred to as “allostatic load” (AL) in this study. AL broadly refers to the cumulative biological effects of being ready for “fight or flight“, or in other words, stress.

So what about this study in particular? This study has two goals:

  1. Compare the stress levels of LGB people with heterosexual people
  2. Compare the stress levels of “closeted” LGB people with “out” LGB people.

Participants were 87 people, roughly evenly divided between lesbian/bisexual women, heterosexual women, gay/bisexual men, and heterosexual men. The researchers measured a variety of demographics including age, race, sex, occupational status, socioeconomic status, physical and mental health, substance use, religion, and family. They measured sexual orientation with the Klein scale, and asked about disclosure status (i.e., whether participants were “out” or not). Psychiatric variables included perceived chronic stress, anxiety symptoms, depression symptoms, burnout symptoms, and conscientiousness. Conscientiousness is a personality trait that has been found to be a confounding variable in these kinds of studies. Biological variables were salivary cortisol, measured three times a day to track cortisol’s daily cycle, and allostatic load, as I described above.

And what did they find? There were few statistically significant differences between the groups; only sexual orientation and oral contraceptive use were different. That’s important! Any differences between groups would be a confounding variable. They also verified some expected results. For example, that anxiety symptoms are associated with depression and burnout symptoms, and that elevated cortisol levels were correlated with burnout.

First the researchers reported their results for comparing LGB folk to heterosexual folk (goal #1). They found that gay/bisexual male participants had more depression symptoms than the heterosexual male participants. In contrast, their lesbian/bisexual women participants had fewer depression symptoms than the heterosexual female participants. They also found that allostatic load levels were lower in gay/bisexual men than in heterosexual men. They found no other differences between their LGB participants and their heterosexual participants.

Second the researchers reported their results for comparing out LGB folk to closeted LGB folk (goal #2). In this case, they did not separate by sex or orientation. Out LGB people had fewer anxiety symptoms, depression symptoms, burnout symptoms, and lower cortisol levels than closeted LGB folk did. No other differences were detected.

Every study has its limitations. This study was no different. Limitations and potential confounds included:

  • Combining homosexuals and bisexuals into one group for analysis. While some issues overlap, bisexuals can face different stressors than homosexuals do (e.g., bisexual folk report facing discrimination from both the gay and straight communities where gay folk don’t; ).
  • Relatively small sample size may have made accurately detecting statistical significance difficult.
  • Their sample was from the Montéal area, an area that has been called “one of the most gay-friendly places on Earth.” Results may have been different in a less tolerant area. This means that results from this study can’t necessarily be applied to people in other areas (e.g., Uganda, the American South).
  • Both age and conscientiousness were found to be confounding variables.
  • Variables like gender presentation (e.g., butch vs femme lesbians) were not considered. They could affect how much active discrimination an individual faces and thus might affect their stress load. Other variables, such a family acceptance, were also not considered.

All of this is interesting, but what does this mean? If we interpret these results as true, then there are some interesting dynamics at play. LGB people who are out of the closet have better psychiatric health than closeted LGB people. However, closeted LGB people don’t seem to be at a disadvantage when it comes to the physical effects of stress. As for comparing heterosexuals with non-heterosexuals, gay/bisexual men seem to have poorer mental health than heterosexuals, who have poorer mental health than lesbian/bisexual women. Coming from an American viewpoint, it seems to me that that might be explained by the cultural acceptance of lesbian/bisexual women and rejection of gay/bisexual men. I don’t know how true that is in Canada, though. Do the results support the minority stress hypothesis? Somewhat, but only for the out/closeted comparison. The heterosexual/LGB comparison results partially support minority stress and partially don’t.

I think these results should be interpreted with a large grain of salt. I don’t think it’s justifiable to make conclusions about all LGB people from this one study. These results are curious, certainly. There are factors at play which bear greater examination (e.g., why don’t closeted LGB people show higher cortisol and AL levels?). I’m curious to see what a study replication in a different area and more participants would show.

Abstract. Full text (PDF).