I suppose one would need to find HIV prevalence rates in homosexual men and compare them with prevalence rates in other groups who are excluded on the basis of things other than sexual orientation (i.e. drug users, hemophiliacs, etc). If this really is a question of homophobia then we would expect that the HIV prevalence in gay men is significantly lower than in the drug addicts, such that gay men must be being excluded from giving blood from reasons other than public health.
I'm not sure it tells us anything to point out groups who have a higher prevalence rate but are not excluded (i.e. black women as in your example). It is possible after all that blood donation is completely misadministered without it being homophobic. To show that this is a case of homophobia I think you would need to do something similar to what i described above.
Just on a side-note, I notice on the FDA website a claim:
quote:In a recent "HIV/AIDS Surveillance Report" CDC's states that men who have sex with men account for the largest proportion (38%) of new AIDS cases reported in the United States from 1996-1997
It would be interesting to know how this data ties in with the data you presented in pie charts. Using these data it should be easy to calculate, for example the "efficiency" of excluding gay men - by which I mean the number of HIV infected people removed from the total pool divided by the number of people in the excluded group, or something like that. We should expect the FDA to choose the most efficient groups to exclude, and if they aren't doing that then it would be evidence of homophobia.
One last thought - don't these guys test for HIV in the blood they collect? The test can reliably show HIV status after about two months since infection. So if it truly were necessary to exclude gay men, it would surely only be necessary to exclude men who have had sex in the last two months.
I can't seem to find any information on how long blood banks sit on their donations in order to screen them for HIV or other diseases, but it must be working pretty damn well whatever they do because there haven't been any recent cases (as far as we know) of HIV transmission through the donated blood supply.
According to the FDA, the risk of transfusion transmission of HIV is "one unit per 2 million donations" - that is, 0.0000005 per unit of blood used in a transfusion. In comparison, the probability of being killed by a car this year in the US is 0.00015.
You know, the easiest way of solving this question is to look at Switzerland and Spain, two countries which allow gay men to donate blood. Unfortunately I am unable to find any data on HIV transmission in transfusions in these countries. I'll wager that their HIV transmission is at similar levels to any other country but let's see if I can turn up any data.
It would be interesting to know how this data ties in with the data you presented in pie charts.
The data you presented is almost a decade old.
Sure, I just noticed that sexual orientation was missing from the pie charts, and surely that is central to the issue isn't it?
If we are going to exclude a group then by definition its going to be unfair on members of that group who feel they're all painted with one brush, and naturally desire an individualized treatment from the health authorities. For example drug addicts who always use clean needles might want to give blood and feel it unfair, silly and discriminatory that they're not allowed to do so.
But I can understand declaring a group persona non grata for efficiency reasons - say, the cost of individual testing is outweighed by the relatively high HIV prevalence in that group, so it's cheaper and more efficient to exclude the whole group without individual testing. That is precisely the decision that has been made for drug users.
If we're going to exclude groups at all, then the only justifiable reason for it is this kind of efficiency calculation. The efficiency calculation depends upon a) how frequently HIV is found in the group and b) what is the residual number of non-infected potential donors who will be unnecessarily excluded as part of the group.
You can only answer this question if you know a) how many gay men have HIV;and b) how many gay men who wish to give blood and are free of HIV are unable to do so because of the blanket ban. My personal feeling is that the blanket ban is silly, but on the other hand I don't think that excluding a group per se means discrimination against that group. Excluding drug addicts is (presumably) a public health choice based on these kinds of calculations. If you use needles there's a relatively high chance you have HIV, and there is a relatively small pool of heroin addicts who wish to donate. It would be "drugophobia" if the policy is enacted solely because the populace will be happier if they think that drug addicts can't donate because they're "sinful". But it's not "drugophobia" if it's done just to keep the costs of the service down.
In principle I see no reason whatsoever why these kinds of calculations should not be made for other groups defined by sexual orientation or whatever. Whether it amounts to discrimination depends on two things: first, is the exclusion is justified by the cost-benefit analysis? second, are the groups based on real, existing epidemiological clusters, or are they based on prejudice?
Finally, a note on the pie charts you provided. The area of the pie is proportional the percentage of new cases of HIV diagnosed from 2001-2004. These proportions do NOT correspond to the prevalence of HIV in those groups. 50% of new HIV diagnoses in 2004 were african american, while 1% of new HIV diagnoses were Asian/Pacific Islander. This does not mean that african americans are 50 times more likely to have HIV than asians, because the number of new diagnoses is not scaled by population size! These pie charts tell us nothing about what groups might be sensibly excluded from donating blood unless they are combined with demographic data.