The inclusion, for the first time, of a question about
sexual orientation in a regularly-conducted national health survey in the US
has allowed researchers to establish that people who define themselves as
lesbian, gay or bisexual do have considerably higher rates of mental
ill-health than the general population.
The 2013 and 2014 National Health Interview Surveys (NHIS) found that gay men were nearly three
times as likely to have symptoms of ‘severe psychological distress’ as heterosexual men and were also more likely to smoke and to drink heavily.
Bisexual men were even more strongly affected.
In the case of women, those defining as lesbian were more
likely to be heavy smokers and drinkers. Bisexual women were more likely to have severe
psychological distress. Lesbian and bisexual women, unlike men, were also
somewhat more likely to suffer from poor physical health.
The NHIS is 60 years old next year, having been conducted
every year since 1957. Households throughout the US are randomly selected to
answer a core questionnaire on basic health and disability, and one member of
each household over 18 is then randomly selected to answer another, more detailed
survey that includes information on specific health conditions, health
behaviours, and access to healthcare. Completion rates for the general survey and more detailed survey are about 75% and 81% respectively.
The detailed survey was the one that included, for
the first time, a question on sexual orientation in 2013. Respondents were
asked whether they thought of themselves as heterosexual; lesbian or gay;
bisexual; something else; didn’t know; or refused to answer. Gender identity
was not asked about, so this is an LGB survey rather than an LGBT one.
This study compares answers from lesbian/gay and bisexual
respondents with heterosexual ones on a number of health indices: self-rated
heath, disability, the presence of one or more chronic conditions, psychological
distress, drinking and smoking.
The detailed Sample Adult surveys from 2013 and 2014 were amalgamated.
Altogether 68,814 adults answered the surveys. Of these, 2566 said they were
something other than heterosexual or refused to answer (3.7% or one person in
27); 1808 said they were non-heterosexual (i.e. lesbian/gay, bisexual or
‘something else’’ – 2.6% or one in 38). Of these, 1149 said they were lesbian/gay (1.7% or
one in 60) and 515 bisexual (0.7% or one in 133).
Lesbians, gays and bisexuals were more likely to be aged
under 26 than heterosexuals and less likely to be aged over 65. Although there
was no difference in ethnicity according to sexual orientation, LGB people were
less likely not to have been interviewed in English. Gay/bisexual men, though
not lesbians/bisexual women, were more likely to have a bachelor’s degree or
Gay/bisexual men were not more likely to be unemployed than
heterosexuals and did not differ in income; in women there was an interesting
difference between lesbians and bisexual women. While both, perhaps
unsurprisingly, were less likely to be unemployed than heterosexual women,
lesbians were more likely to be in full-time jobs and have high income while
bisexual women were more likely to have part-time jobs and low income. This may
be an age effect, as non-heterosexual women under 35 were considerably more
likely to describe themselves as bisexual rather than lesbian, whereas women
over this age were more likely to describe themselves as lesbian than bisexual.
Lesbian and bisexual women were more likely to have no health insurance and to
have been unable to afford medical care in the last year than other women;
there was no difference in men.
There were few differences in physical health between men of different sexual orientations. Lesbian and bisexual women were more likely to have multiple chronic conditions than heterosexual women, and lesbians were slightly more likely to complain of poor physical
This was very much not the case when they were asked about
psychological distress. In multivariate analysis, controlling for other factors like
age, gay men were 2.8 times more likely to say they had severe psychological
distress than heterosexual men, and bisexual men 4.7 times more likely. Bisexual
women, though not lesbians, were 3.7 times more likely.
Gay men and bisexual women were twice as likely to be heavy
drinkers, and lesbians and bisexual men three times as likely, as
And gay men and lesbians were twice as likely to smoke, and
bisexual men and lesbians twice as likely to be heavy smokers, than
Although a number of previous surveys of the health of LGBT
people have been done before in the US, they have either been self-referred, or
studies of small clinics or locality populations. So, while these surveys
generally reported higher rates of mental distress in LGBT people than the
national average, there was no denominator and no direct comparison. This is
the first national survey that establishes for sure that self-described LGB
people are more likely to have severe psychological distress.
It also confirms
another finding from previous surveys, namely that bisexual people – who may be
less certain of their sexuality and more socially isolated – are more severely
affected than lesbian or gay men.
It is worth defining 'severe psychological distress': it means that respondents said
there had been more than 13 days in the last month when they had felt “nervous,
hopeless or worthless; too sad to be cheered up; restless or fidgety; or that
everything was an effort.” In other words, we are talking about a significant
burden in terms of mood and subjective ability.
The researchers comment that “this study finds substantial health
disparities in LGB adults in the US, potentially due to the stressors that LGB
people experience as a result of interpersonal and structural discrimination.”
They draw particular attention to the fact that, as is not the case in other
minorities, there was no significant association between income, education or
ethnicity and psychological distress; it seems to hit people of all backgrounds
and classes and therefore may be caused by non-economic stressors.
“Health practitioners should pay particular attention to the
current and future health outcomes of this small, diverse and vulnerable
population,” they conclude.