New Data. New Opportunities.

So, I'm working on a new analysis, and my plan is to describe what I'm doing, why I'm doing it, and what I'm learning along the way, as it happens.

Well, I guess it's not entirely "new", depending on how you count it, I've been working on this for years already.

It's an extension of prior work I've talked about a lot on this blog, the extraordinary finding that gay men are more likely to be in "excellent" health than heterosexual men.

I've looked at this issue in many datasets, and published a paper on it in one particular dataset, the Behavioral Risk Factor Surveillance Survey, where I looked at the self-reported general health of men in same-sex and mixed-sex couples, based on their marital status and household composition.

In that study, I found that about half the general health advantage of men in same-sex couples was explained by the facts that men in same-sex couples were younger on average, better educated, and wealthier. When I looked at some of the specific health characteristics, men in same-sex couples were less likely to be overweight or obese, but more likely to smoke. Taking all the health factors I could into account didn't explain why man in same-sex couples were healthier than men in mixed-sex couples.

But what do I want to do now, in this new analysis? Well, it may yet shift and wiggle from this idea, but I'd like to do a couple of things. First, there's a lot more data to work with now. Instead of having to look at marital status and household composition, we can now look at sexual orientation directly, because a number of states have started to ask people not just who you live with and your marital status, but also whether you think of yourself as heterosexual, gay, lesbian, bisexual, or something else.

Second, because we now have both sexual orientation and marital status, and same-sex couples can get married in a lot more places than when I did the earlier analyses, I can look at single gay, lesbian and bisexual people, bisexual people in same-sex and mised-sex relationships, and a variety of household composition structures.

Third, I'd like to try to fold three different surveys into a single analysis, the BRFSS (which is the largest of the surveys, but the sexual orientation questions are asked in only select states), the National Health Interview Surveys (which are smaller, but comprehensively asked all across the country), and the National Health and Nutrition Examination Surveys (which are smaller still, but have a great deal of depth to them).

There are other ideas I'd like to look at too, like the degree to which heterosexuals would be misidentified as being sexual minorities because they live in same-sex households (like my heterosexual roommate). And what geographic and demographic factors predict that misclassification.

Also on the issue of misclassification, the BRFSS asks interviewers to guess the sex of the person they're on the phone with, and these data give us an estimate of how often they guess wrong.

Another set of issues I could look at with these data are things like basic demographics - Who do gay men and lesbians live with? How many are in relationships? Believe it or not, these basic issues have barely been touched in the academic literature, because data like this has not been available before.

Seeing Healthy Gay Men

I've been invited to take part in a panel discussion on Wednesday. Here's a preview of my talk. I'd love feedback on the flow, grammar, anything to make it roll off the tongue easier. I don't have a lot of time to deliver the talk, and I want to get some key ideas out there clearly.

SEEING HEALTHY GAY MEN
Introduce yourself to the person next to you, just your name. Hold their hand, and look right into their eyes, take a good deep look. Think about how healthy this person is. Through all of life's adversities, here is this healthy person in front of you.

What I've just asked you to do is very different from how I, as a public health researcher, look at gay men. To put it bluntly, in public health, we don't see healthy gay men.

First of all, we are not accustomed to look at health in anyone, queer or bland. We almost always look at sickness and death, and very rarely at health. Sickness is easier to measure than health. Sickness is seen as objective, while 'health' and 'well-being' are subjective states. Death is the most objective health outcome, and thus the most trustworthy object of study. So it's nothing special about gay men. We're not used to seeing anyone as healthy.

But there other reasons that public health, broadly written, is not used to seeing healthy gay men.

One of these is that a little over a hundred years ago, the fields of medicine (and later psychology) tried to exercise authority over homosexuality, to claim it as their turf, a claim widely supported by homosexuals at the time. It was a humane alternative to being regarded as the embodiment of sin, or being treated like a criminal.
Old habits die hard, and it has been hard to let go of these 'humane' impulses, despite the official depathologization of homosexuality in 1975.

Another reason that public health has a hard time seeing healthy gay men is that we actually have a lot invested in gay men being ill. That may sound strange at first, but think about it - citing health disparities is how we raise awareness about the health challenges facing us. Playing up health disparities is how we get money for programs. It's often how we make claims that legal protections are needed, how we describe ourselves in media stories to gather support.
So seeing healthy gay men risks upsetting the apple cart, derailing the gravy train, if you will.
And then there's the right wing. Not only are we a threat when we're sick, but if we're not sick, then we don't need "special" protections. We just can't win with those guys.

The last reason that public health has a hard time seeing healthy gay men that I want to raise is the perceived 'failure' of HIV prevention. While it would make sense to look at the relatively small number of new infections these days compared to the early 1980's and claim a great deal of success in HIV prevention, there is a tendency in public health circles to treat each and every new infection as a failure.
After all, we know a lot more about the biology and psychology of how HIV is transmitted. Hundreds of millions, probably billions of dollars have been spent on research and prevention efforts in the US, so even a single new infection represents some sort of failure in our prevention efforts.
Often implicit in our conversations about the 'failure' that each new infection implies is some blame directed at the man who got infected, and perhaps towards the man who infected him. Were these guys high? Careless? Selfish? Immature? Maybe they were just looking for love in a harsh world. Maybe they were possessed by the worst affliction a gay man can have in the eyes of public health: being a bug-chaser. All of these explanations imply that the infection was enabled by an individual weakness. So we are used to seeing gay men as sick, or at least as perpetually 'at risk'. And any attempt to break free from a life constantly 'at risk' must be a little crazy.

There may be other reasons that we public health types have trouble seeing gay men, but these are some of the big ones: 1) public health has a hard time seeing anyone as healthy, 2) the historical disciplinary claim of medicine and psychology implies that there is something inherently pathological about homosexuality, even though an explicit claim of pathology would be rejected in most circles. 3) seeing healthy gay men undercuts our ability to raise awareness within the community, and funds for services from public and private granting agencies, and 4) how else does one explain 'failures' in HIV prevention?

EVIDENCE OF HEALTHY GAY MEN
And so, a little over a year ago, when I came across evidence that gay men might well be healthier that straight men, I was dubious.
I was looking at data from the 2003/2005 California Health Interview Survey, the idea I had was to systematically go through the CHIS, documenting the disparity of each health outcome for gay and bisexual men relative to straight men, and for lesbians and bisexual women relative to straight women.
The first one on the list was pretty boring, at least from an epidemiologic standpoint: "in general, would you say your health is excellent, very good, good, fair, or poor?" Although this is by far the most common question on health surveys, analytically, we usually treat it like a throw-away question, basically a way to build rapport with the person on the other end of the phone before getting in to the more personal and revealing questions.
I expected to see what others before me had seen, that gay and bisexual men were more likely to report "poor" overall health. And they were in this dataset, too. But more striking was that gay men were also more likely to report being in excellent health. And that difference was larger.

But like I said, I was skeptical. Maybe it was a statistical fluke. I did what any sensible epidemiologist would do. I ignored it and moved on.

But, it stuck in the back of my mind.
When the 2007 data came on-line a short while later, I took another look, and the same pattern held. So now, I began to wonder if there might be something to it, these healthy gay men, not less healthy than straight men as everything had led me to expect, not equally healthy as straight men, but actually more likely to be healthy.
So then I began to look at other datasets. In most datasets, but not all, gay men were more likely to report excellent health.
I was still dubious.
Perhaps gay men were more likely to be in excellent health because they were younger, more educated, affluent. In most studies, men who say they are gay tend to be younger, more educated, affluent, and these things also tend to produce better health. So now I needed to find a dataset that was large enough to break people down by age and socioeconomic position.
I found two very large datasets, the Current Population Survey, and the Behavioral Risk Factor Surveillance System.
In both of these datasets, gay men were more likely to be in excellent health than straight men in all age groups, and within all levels of educational attainment. The BRFSS dataset was so large that I could stratify by both age and educational attainment simultaneously. Same result.
So now I knew it wasn't a fluke, it wasn't just because men who say they are gay to a stranger on the phone are just demographically healthier. Gay men were actually more likely to report being in excellent health.

Once I'd convinced myself that gay men are more likely to be in excellent health, that it was a 'real' finding, the next question is "Why?".
With all the HIV, STD's, alcohol abuse, drugs, depression and anxiety, even suicide documented in the literature, it still baffles the public health mind.
I'm going to leave the question "Why?" open, and move on instead to a related question: "So What?".
If gay men are more likely to be in excellent health, what are the implications of that?
One implication is that in the face of the long, oft-repeated litany of health-related faults, some gay men have found a way to thrive, arguably with no assistance from us in public health.
Perhaps we can learn from these healthy gay men. But that requires being able to see healthy gay men, and then to talk with them, and ask the right questions.
The salutogenic (health-creating) processes that these men have undertaken could be a blueprint for the rest of us, a novel approach to addressing the troublesome health disparities that gay men face.

Another implication is that gay men may well be doing some things 'right' that heterosexuals could do well to emulate. Perhaps the ways that we gay men have learned to structure our lives, with strong friendship networks, provides space for us to excel. Perhaps our more nuanced approach to monogamy and relationship structures gives us permission to be truer to ourselves and the ones we love. Pure speculation on my part, but if more of us are in excellent health despite all the social forces arrayed against us, we've got to be doing something right, and maybe if we knew what that something was, it would help society at large.

{I haven't quite figured out how to wrap it up yet}

Healthy Gay Men

What Explains the General Health Advantage of Gay Men in New York City?

In 2007, the New York City Community Health Survey asked men (and women) how they identified their sexual orientation, and also the standard question "Would you say that your health in general is: Excellent, Very Good, Good, Fair, or Poor?"
44 of 153 self-identified gay men (29%) said they were in excellent health, while 563 of 3,049 self-identified heterosexual men (18%) said they were in excellent health. So, gay men were almost twice as likely to report excellent health (odds ratio = 1.95, 95% confidence interval 1.36-2.79).

What gives? Well, aside from gay men being in better health, there might be a lot of other potential explanations. Like, what if men who identify as gay are just on average younger? Or less likely to live in poverty? Or less likely to be unemployed? All those things are strongly associated with one's subjective assessment of health.

So, I tried controlling for age, race/ethnicity, employment status, educational attainment, being foreign-born, and the ratio of the household's income to the poverty level set for a household of that composition (the poverty threshold for a household with two adults and three kids is a lot higher than the poverty threshold for someone living alone).

Of these, controlling for the ratio of income to poverty made the biggest difference. After controlling for the income:poverty ratio, self-identified gay men were 60% more likely to report excellent health (odds ratio 1.60, with a confidence range of 1.11-2.31).

Controlling for all of them simultaneously explained more of the gap, with gay men being almost half-again as likely as straight men to report excellent health (odds ratio 1.46, confidence range 0.99-2.16).

Still, all things equal, wouldn't you rather be gay in New York City?

Then, I decided to break it down by borough. Gay men in Manhattan were 67% more likely to report excellent health (odd ratio 1.67, confidence range 0.99-2.80), gay men in Brooklyn were 37% more likely to report excellent health (odds ratio 1.37, confidence range 0.58-3.23), and gay men in Queens were only 15% more likely to report excellent health (odds ratio 1.15, confidence range 0.40-3.28). There weren't enough men in this particular sample to say anything worthwhile about the Bronx or Staten Island.

Hmmm... looks like where you live as a gay man has a big influence on how you feel.

Excellent Health among Polish MSM

In the Polish General Social Survey, they asked about the gender of sexual partners in 1994.
Extreme caution should be used interpreting these numbers, because I'm sure there are cultural and linguistic differences in how the questions are interpreted and answered between the US and Poland.
That and the numbers are fairly small. In general I try to avoid combining gay and bi, or MSM with MSMW, but in this case, the numbers were small enough that I wanted to get more stable estimates.

Only 2 of 31 men who said they had sex with at least one man since age 18 reported excellent health (6% {0%-15%}), while 48 of 322 men who said they never had sex with a man since age 18 reported excellent health (15% {11%-19%}).

I'm nervous about saying much of anything about this data, but it seems like the proportion of men reporting excellent health is lower than the US data, but I would attribute that to language differences until proven otherwise. However, within that context, it does appear that men who have had sex with men in Poland are less likely to report excellent health, perhaps speaking to the general attitudes towards homosexuality in Poland - although not having been to Poland I'm not in a position to say. Anyone with more relevant experience want to comment?

Also interesting is that 31 of 322 men said they had had sex with a man, a much higher percentage than we usually see in US telephone surveys. Should I interpret that to mean that more men have sex with men in Poland, that more are willing to admit it during a telephone survey, or that the question wording is confusing enough that a relatively large proportion gave answers incongruent with their experiences?

I got this data from ICPSR.

Measuring Well-Being

While looking for more data on the well-being of gay men relative to str8 men, I came across this initiative in Europe to measure national (and individual) well-being. I don't ordinarily put much stock in these quiz yourself things, but I think the questions they are asking are useful, and it was kind of fun to compare myself to the various European nations that participate.
Although I have Polish, Irish, Scottish, English & Norwegian heritage (and probably more), my own profile more closely resembles the Hungarians - anhedonic but satisfied. Maybe I should visit there, when I can afford to travel again.
And I score really high on autonomy, which is probably a main (non-material) component of why wealthier people generally have much better health than poorer people.

Excellent health in the Current Population Survey (March Supplement)

Another data source that has data on general health is the "Current Population Survey", which is the dataset the US Gov't uses to estimate the unemployment rate, among other things. They interview about 47,000 to 57,000 households every month from all over the country.
Mostly they ask about who in the household is employed, unemployed, looking for work, etc. But every March, they ask a lot more questions, mostly about income & benefits, including health insurance.
Near the very end of the survey, they ask about the general health of everyone in the household, and that's what I'm interested in. The datasets are pretty huge relative to the capacity of my computer, so I've only been able to look at one year, 1998, so far.
In 1998, out of 64,659 households, I could identify 25,821 couple-headed households with non-imputed data on health. 23,124 of these were mixed-sex married, 1,366 were mixed-sex unmarried, and 24 were same-sex unmarried.
Admittedly, 24 couples out of 64,659 households is a pretty disappointing yield, but I'm told that as the years go on, more and more same-sex couples get accurately recorded, so I'm looking forward to more data from the other 10 years. But, on the plus side, 24 couples have 48 men in them, so that makes the sample size a bit more robust.
As you can see from the chart, there's really not much difference between the proportion of men in male couples who report excellent health (33%, 20%-46%), and the proportion of men in mixed-sex unmarried couples (33.6%, 31.1%-36.1%), or the proportion of men in mixed-sex marriages who do (31.0%, 30.4%-31.6%).
So, the fact that these results don't support the findings I'm seeing from California is mildly disappointing, but it's also clear that gay men (at least the ones in live-in relationships) aren't in dramatically worse health, which is what you'd expect to find given how we have been portrayed in the medical literature thus far.

I'm especially excited about the prospects for this dataset for a few reasons. First, it uses the same methodology (with minor twists) over an 11 year time span, so there is a lot of data to work with, even if the gay men in relationships are probably not representative of all gay men (but neither are married men representative of all straight men, for that matter), and since it spans an 11-year period, it might be possible to look at trends in 'excellent' health over time.
Second, they try to interview about half the same households the following year, so it would be possible to look at changes in health status over time (though this would only be true for one year's change for any individual), and also health in relationships that last vs. those that dissolve.
Finally, and best of all, is that the survey covers the whole geography of the United States, so I can look at health in relation to homophobia levels much more in-depth than is possible within the state of California. Perhaps the health of gay men is better in California and New York City than straight men, and worse than straight men in Wyoming and Nebraska, that might help explain why many of the national studies don't show much difference...
What do y'all think?

More Results on Gay Men in Excellent Health

So, I've looked at two more datasets in examining this phenomenon which started with noting that gay men were more likely to be in excellent health than straight men in the California Health Interview Study (CHIS).
Many thanks to Randy Sell at Drexel who has collated information on a large number of national and state-level studies that ask people about their sexual orientation and/or sexual partners at www.gaydata.org.

In the General Social Survey (GSS), a large telephone survey that has been on-going since 1972, they asked about the sex of respondent's sexual partners in the last year, the last 5 years, and since age 18, in various combinations since 1988. (I used this same dataset to look at the phenomenon of gay men giving blood, which can also be considered as an indicator of well-being).

In the GSS, 33% of men who had sex exclusively with other men over the prior five years (26%-40%) were in excellent health, virtually identical to the 33% (31%-34%) of men who had sex exclusively with women over the prior five years.
Looking at sexual partners just in the last year, 35% of men who had sex only with men in the previous year (29%-42%) were in excellent health, about the same as 34% (33%-35%) as men who had sex only with women in the previous year.

In the National Household Survey on Drug Abuse (NHSDU), on on-going telephone survey used to understand drug use in the US, they asked about sexual partners in one year, 1996. For reasons I don't understand, they never asked before or since. If they had, this survey could have been one of the best sources of information on LGB population health.

52% of the 58 men who reported sex only with males were in excellent health (39%-65%), which is much higher than the 36% of the 3,196 men who reported sex only with women (35%-38%).

The proportion of men who had sex with men who reported excellent health in this survey was a lot higher than in other surveys, too, so I'd be a little cautious in putting much faith in these particular results.

New York City MSM more likely to be in excellent health

More evidence that gay men are more likely to be in excellent health than straight men...
The New York City Department of Health and Mental Hygiene kindly makes data from its BRFSS available online.
Since 2002, they have been asking men their sexual orientation and also whether they have sex with men. Although the sexual orientation data is in there, it is not accessible on the website broken down by gender. But the sex with men variable is, so I used that as the closest approximation until I can get my hands dirty with the SAS programming.

The upshot is that in 5 out of 6 years, men who have sex with men were more likely to be in excellent health, after age-adjusting, than their (mostly) heterosexual counterparts.

When averaging the six years together, 27.5% (24.4%-30.6%) of New York City's MSM population were in excellent health, compared to 21.5% (20.7%-22.3%) of mostly heterosexual men in New York City.

That's very similar to the results from CHIS. After combining data from 2003,2005 & 2007, the proportion of California gay men in excellent health is 26.6% (23.1%-30.2%), compared to 22.1% (21.5%-22.6%) of straight men in California. These results are not age-adjusted like the NYC results, but I don't think that would make much difference.

Anyone want to speculate as to why gay men (at least in California and New York City, if not Canada) are more likely to be in excellent health than straight men? What lessons can straight men learn from us to be healthier?

Canadian Gay Men Only Slightly Healthier than Straight Men

Came across a neat study done by Statistics Canada. They interviewed about 260,000 Canadians for the study in 2003 & 2005 (Canadian Community Health Survey, CCHS). They found that gay men were slightly more likely to be in "very good or excellent" physical health (65.4%) than heterosexual men (63.9%, difference not statistically significant) but slightly less likely to be in "very good or excellent" mental health (73.8% vs. 75.4%, also not statistically significant).

I used the population from the California Health Interview Survey (CHIS) that most closely matches the Canadian study for comparison, adults aged 18-59 interviewed in 2003 & 2005, and lumped "excellent" and "very good" the same way they did. In California, 61.2% of gay-identified men reported "very good or excellent" general health, versus 54.3% of heterosexual men, a much larger difference.

The first thing that jumps out at me is that Canadian men seem to be a much healthier lot no matter what their sexual orientation is.

But it does leave me scratching my head - why are gay men more likely to be in excellent health than straight men in California - but not necessarily so in Canada, which has had a much friendlier political climate? Hmmmm...

Gay Men in Excellent Health - More from CHIS

So, I decided to look at this phenomenon of gay men being more likely to report being in excellent health than straight men from a slightly different angle - namely does it make a difference where one lives.
In particular, does the social environment of heteronormativity play a role in gay men being more likely to report being in excellent health?

The recent vote in California on whether to restrict the legal definition of marriage to "one man and one woman" (OMOW) seems like a pretty good way to define an area's heteronormativity, the more people who endorsed proposition 8, the higher the level of heteronormativity. Got this data from the California Secretary of State's Statement of the Vote.

So, using AskCHIS, I got estimates for the proportion of gay men and straight men reporting themselves to be in excellent health for 11 regions of the state: the counties of Los Angeles, San Francisco, and Riverside, and then 8 groups of counties (East Bay, South Bay, North Bay, Sacramento area, North & Sierras, Central Coast, Orange & San Bernardino, San Diego & Imperial, and the San Joaquin (Central) Valley). I had to use groups of counties in order to get enough sample size to estimate the proportion of men reporting excellent health within each region.

The proportion of gay men reporting excellent health was highest in San Francisco county, 34%, but this needs to be taken with a pretty big grain of salt, because the CHIS survey data would be compatible with a range from 25% to 42%. The lowest proportion of gay men in excellent health was in Riverside county, 13%, with a range compatible with the survey from 5% to 21%.
These two seem to line up with my hypothesis pretty well, in that only 25% of San Francisco voters endorsed a OMOW definition of marriage, while 65% of Riverside county voters did.
But, voters in the San Joaquin Valley (San Joaquin county down to Kern) endorsed OMOW by 70%, and yet had the 2nd highest proportion of gay men reporting excellent health: 32%, range 12% to 51%. So, it's not like there's a one-to-one correspondence.

Among straight men, the highest proportion reporting excellent health was also in San Francisco, 26% (23%-30%), and the lowest was in the San Joaquin Valley, 19% (17%-21%). In general, there was a much tighter correlation between the proporotion of str8 men reporting excellent health and how voters endorsed the OMOW restriction of Prop8. The tight correlation has mostly to do with the fact that there are more str8 men, and therefore better estimates of how healthy they are.

Of course there are many other potential explanations for why gay men (and str8 men) seem to be healthier in areas that are less homophobic (as measured by endorsing Prop8) in the CHIS data. For instance, it could be that older people are more likely to live in the more homophobic areas, and since old people are less likely to report being in excellent health, that's why you see fewer gay and str8 men reporting excellent health in the more homophobic areas.
Also, the mode of analysis I used is very preliminary - not only no control for potential confounders, but also the estimates of how likely men are to report being in excellent health is based on some pretty small numbers. But, it's enough to get the brain juices flowing, and perhaps to help me convince CHIS to give me access to individual-level data files...

more healthy gay men

I just looked at the data in the MIDUS-2 study, and the sample is just too small to say, but it looks like gay men and straight men are just about equally likely to be in 'excellent' health. As in CHIS, bisexual men may be somewhat less likely to report excellent health.
On to the next dataset...

gay men in excellent health

Just had an idea for a research project.

I was chatting with Rachel about why my research seems so alone in the world - nobody in the Public Health world seems to have picked up on the utility of looking at empirically measured changes in societal heteronormativity as having the potential to affect health impacts - and I don't know why. Gay & lesbian people 'get it' instantly. Public health types seem to get confused by not knowing precisely which biochemical pathway a change in law would need to activate.

Anyway, I was working last week on data from the CHIS, and noticed that gay men (in the aggregate) are more likely to report being in 'excellent' health than straight men. My public health training told me to ignore it - 'excellent' health is not a disease, after all. It's a squishy endpoint, who knows what it means?

But something sparked in me during the conversation with Rachel.
Why are gay men more likely to be healthy than str8 men? Maybe it's not a strange question. Maybe it's an endpoint that's worth a little more investigation.

I know that nobody in the epidemiology world will believe it if it's a single finding. As large as CHIS is, it is only one dataset. So now I need to go poking around in other datasets which ask similar questions - and fortunately there's a lot of them.

I'll keep you posted...