Health promotion often talks about 'men who have sex with men'. One reason might be to emphasise that we've known for over twenty years that a very large proportion of men who have sex with men do not have a sexual identity in the way that the vast majority of LGB activists do. If you say 'gay and bisexual' – or worse, just 'gay' – all over your material or just use the gay scene/media, a big chunk of your target audience are much, much less likely to see it, think of it as relevant to them if they do, and look at it.
Much more commonly, saying your work is with MSM or 'homosexually active men' is used to enable you to ignore bisexual men and just go after the gay ones while pretending that's not what you're doing.
For example, from 1996 millions of pounds were given by the Department of Health for the CHAPS project for "HIV health promotion for gay, bisexual and other homosexually active men". Almost all of it went on campaigns on the gay scene and in the gay media for gay men.
Indeed, CHAPS partners made a positive virtue of the fact that it was only interested in the sex with other men its target audience were having. It would monitor the difference between 'exclusively* homosexually active men' and 'behaviourally bisexual men', but it would not – ever – do anything about their sex with anyone other than men. Or accept that the differences it found and the recommendation that behaviourally bisexual men be a priority group be taken seriously, like the other handful of priority groups.
The result included material with the appalling definition of anal sex as "when you put a man's cock up your arse or put your cock up his". This is despite the fact that we know that bisexually behaving men are more likely to have anal sex with women than the general male population. (And most anal sex in the UK is between men and women.) When I asked the author of that one what they were saying women did not have – anuses or anal sex – I was told I was being rude.
When the funding for CHAPS was ended a few years ago, the HIV Prevention England project that replaced it combined CHAPS with the equivalent programme for African communities. HPE started off being a bit more honest: it had work "for gay men across the South and West of England, and across the North and Midlands" and campaigns "focused on encouraging more Africans and gay men to test for HIV" before a token mention of an "MSM community forum".
But while its new official site talks a lot about "gay and bisexual men", the biggest and most important survey it funds still has lines like "While the respondents' sex with women clearly has the potential to transmit HIV, [the survey] is concerned with mapping male homosexual behaviour and not heterosexual behaviour".
It's also either stopped asking about whether its respondents have had sex with both men and women – something it has done for all of its available past reports over nearly 20 years – or doesn't report that. Instead, they're just asked about how they identify. Obviously, this underestimates the bisexual behaving population.
This is what erasure looks like.
* 'Exclusively' meant 'in the past year'. CHAPS funded reports showed, for example, that more gay-identified men had sex with women in any particular year than did BDSM with anyone else, so it couldn't make the definition cover a much longer period of time, or there wouldn't be very many of them. Yes, CHAPS materials covered BDSM… with other men.