HIV Home Test Kit: Debate Today

[Luke T.]

It seems there is a very simple home test kit for HIV which a company wants to make available over the counter. For some reason, the FDA is being reticent about making such a kit available. Why that is, I don't know.

OraSure Technologies Inc. wants to sell the first 20-minute, at-home test that screens for two HIV strains using a swab device that tests saliva.


Some AIDS groups have concerns about home testing, and a Food and Drug Administration advisory committee is set to discuss the idea this week, starting today.

http://www.freep.com/news/nw/hivtest3e_20051103.htm

The article doesn't say why the AIDS groups are concerned.

Oh. Found one that does. It talks about a similiar debate going on in Canada.

Most Canadian provinces have anonymous HIV test sites, which may reduce demand for home kits. Jane Greer of the Hasslefree Clinic in Toronto worries if people take an HIV test in isolation, they may not get counselling when faced with a positive result.

"They won't have any pretest counselling, what options they have for care and support if it's positive," said Greer. "Even for negative testers, the opportunity for prevention is lost if they haven't had a sit-down around what their risks are, how they can keep themselves safe in the future."

Dawn Archambault, a former prostitute in Toronto, sees pros and cons to home tests. She regularly went for HIV testing at a clinic. If a home test had been available, Archambault said she might have chosen it.

"You feel a little uncomfortable saying I need to have an HIV test and then there's the whole process of counselling that goes along with it," said Archambault, who now helps to keep prostitutes safe and healthy in Halifax.

http://www.cbc.ca/story/science/national/2005/11/02/HIV-tests051102.html

 

[Stitch]

I guess some of the concerns relate to "at-home test that screens for two HIV strains " This would suggest there are others?? So people may think they are OK, when they are not and go off and infect other people. The other issue of course is somebody gets a false positive, with no support and counselling and decides to take up base jumping but without the parachute.

 

[Luke T.]

I guess some of the concerns relate to "at-home test that screens for two HIV strains " This would suggest there are others?? So people may think they are OK, when they are not and go off and infect other people. The other issue of course is somebody gets a false positive, with no support and counselling and decides to take up base jumping but without the parachute.

The flip side is that there are people who suspect they may be infected, but are too embarassed to go to a doctor to find out, and therefore may go on to infect others.

I don't know if the AIDS groups mentioned in the article who oppose the home kit are the same AIDS groups which oppose testing everyone for AIDS who walks into a doctor's office for any reason (such as the flu) because it violates a person's privacy, but here is the perfect protection of privacy that should be made as widely available as possible, in my opinion.

There are a lot of people walking around who don't know they are infected. One way to get that under control would be to test everyone who comes to a doctor's office. Barring that, this is the second best solution.

 

[Luke T.]

If I took the home test and it read positive for HIV, you can bet I would take the test again.

There are home pregancy kits, so why not home HIV kits?

 

[Soapy Sam]

Because I need frequent visas for work and many countries demand an HIV test, I've been taking one every six months for about ten years. At first anyone asking for a test had to have two hours counselling. As I was never in any doubt about the result, I managed to dodge this, so don't know what it entailed. Now it's easier. As a regular blood donor, I just realised the NHS will mail me the result of routine screening on blood donated. So as of October 25th I can tell you I was clear of HIV I, II, HTLV I, Hep B and C.

All of which is dandy.

But what if I thought I actually might have HTLV? What if I take the test and get a false positive? Or a true positive? Am I now legally obligated to inform my insurers? My wife? My neighbour? His wife? (Hey she always stops to chat...)
What are my obligations?

I think the difference between a home pregnancy kit and a home AIDS kit, is that if you are pregnant and I sleep with you- (brief pause while Sam and Luke consider that bizarre concept in silence before moving on) -I can't catch your pregnancy.

 

[Rolfe]

Obvious question, what are the sensitivity and specificity of the test in question? (Sensitivity is the percentage of infected individuals who are correctly identified, specificity is the percentage of uninfected individuals who are correctly identified. I think.) If both of these numbers are 100%, then you can start having your discussion about the sociological and psychological implications of allowing the test to be marketed. However, if not, you have to realise that some people will be given a wrong result.

"Wrong" results are recognised in proper laboratory screening protocols, and samples will be taken through two or three stages of increasingly complex testing to get a definite result, where necessary. However, to do this, the sample has to be in a competent lab, and someone has to initiate these protocols. None of this is going to happen with a home-testing kit.

Pregnancy testing is a convenience. Some false results do occur, but not that many, and usually the consequences aren't actually all that vital. Someone suspecting they have a wrong result has plenty of options as regards checking it, and it usually isn't such a huge great deal.

HIV testing is a lot more critical than that. I'd be very wary of a test like that unless sensitivity and specificity were very very high indeed. Alternatively, if you have a test with virtually perfect sensitivity, but less good on the specificity front, AND YOU DON'T CALL A POSITIVE POSITIVE, YOU CALL IT SUSPECT, that might be useful in weeding out the negatives while just gently leading the possible positives in for further investigation.

But given the public's propensity to view a positive/negative test result in black and white terms, I'd be very unhappy indeed about allowing an HIV kit to be marketed unless that was essentially the case. (And in a field trial too, that is showing that operator error is virtually impossible.)

Rolfe.

 

[CBL4]

Every reason against having a home test is true but I think the fact that more people would find out that they have AIDS is a much greater benefit.

In any case, unless there is a definite and significant downside to society, then I think that people companies and people should have the to sell and buy what they want. The anti-home tester have not made their case convincly, so the test should be sold.

CBL

 

[Rolfe]

I think that people companies and people should have the to sell and buy what they want. The anti-home tester have not made their case convincly, so the test should be sold.

Irrespective of the sensitivity and specificity? Should I be able to market a test that was wrong more often than it was right? That was giving completely random results?

Where do you draw the line here?

Rolfe.

 

[CBL4]

Where do you draw the line here?

I would draw the line pretty low - perhaps 90 or 95%. However, this is much more accurate:

In the clinical studies by the manufacturer (OraSure Technologies, Inc.), the OraQuick test correctly identified 99.6% of people who were infected with HIV-1 (sensitivity) and 100% of people who were not infected with HIV-1 (specificity). The Food and Drug Administration expects clinical laboratories to obtain similar results.

http://www.cdc.gov/hiv/PUBS/faq/oraqckfaq.htm

I should note that this is not with home use.

CBL

 

[Rolfe]

I would draw the line pretty low - perhaps 90 or 95%. However, this is much more accurate:
http://www.cdc.gov/hiv/PUBS/faq/oraqckfaq.htm

I should note that this is not with home use.

CBL

I agree that that is very good. Though I have two reservations.

First is the obvious one, I would also like to see if these figures are maintained on a field trial. That is, is the test idiot-proof. Bitter experience tells me that many things work well in the lab, but when you let some idiot go away and play on his own, things can deteriorate pretty fast.

Second is that, good though the stats are, they are the wrong way round. It is the sensitivity that is not perfect. Therefore, although you seem to be able to rely on the test when it tells you that you are positive, there is room for doubt with the negatives.

This means that in fact you may have to advise all people testing negative that they should have the result checked by a reference method. This is a lot more problematical that having to advise that the positive results be checked - first because there will probably be a lot more negatives than positives, so you haven't cut down your workload much at all, and second because it is the positives you want to bring into the system for further care, not the negatives.

Even so, these stats are good enough that the argument does become more sociological and psychological than technical.

(BTW, I'd insist on an excellent sensitivity for a test like this, preferably better than 99% - but then we do in fact have that. I could live with a relatively poor specificity, maybe 95%, because I wouldn't tell the positives they were infected in that case, I'd just be calling them in for follow-up testing. Below 95% specificity I'd start to worry that the number of false positives would get too large to be acceptable, given that in a population where most people are uninfected, you can rack up a lot of false positives very quickly. 90% for either is much too low, 90% sensitivity in particular is hopeless, because if you can't believe the negatives then most of the usefulness of the test has disappeared.)

Rolfe.

 

[toddjh]

I think discussions of the sensitivity and specificity are a little bit hard to apply to home HIV tests, because of the relatively long incubation period. The last I heard, the virus is undetectable for the first two or three months.

Unless you find some way of testing for HIV besides looking for antibodies, any HIV test will give false negatives if taken too soon after infection -- whether in the lab or at home. Unless the person being tested has completely abstained from sex or IV drug use for a matter of months, 100% sensitivity is an unattainable goal in any environment. And, honestly, the people most at risk for contracting HIV are unlikely to have abstained for that period of time.

Jeremy

 

[toddjh]

I'd also like to add that I think a home test would be useful even if the number of false negatives is relatively high. I don't buy the argument that a false negative will drive an HIV-positive person to infect others: High-risk individuals tend to act as if they're clean until they find out they're not, so they'll already be infecting others. The damage is already being done, and a false negative won't change anything since it will simply appear to confirm what they already believed.

I suppose it's possible that a false negative might inspire some lower-risk people to engage in riskier behavior (for example, a negative test result might convince a monogamous couple it's safe to stop using condoms), but the lower-risk people aren't the ones that warrant the most attention.

Does anyone dispute that the availability of a home HIV test would save a lot of lives? Call me a pragmatic utilitarian with ruggedly handsome features if you want, but I think all other considerations should take a back seat to that.

Jeremy

 

[epepke]

(BTW, I'd insist on an excellent sensitivity for a test like this, preferably better than 99% - but then we do in fact have that. I could live with a relatively poor specificity, maybe 95%, because I wouldn't tell the positives they were infected in that case, I'd just be calling them in for follow-up testing. Below 95% specificity I'd start to worry that the number of false positives would get too large to be acceptable, given that in a population where most people are uninfected, you can rack up a lot of false positives very quickly. 90% for either is much too low, 90% sensitivity in particular is hopeless, because if you can't believe the negatives then most of the usefulness of the test has disappeared.)

That's an important point.

For a home test, it's not just the raw numbers that matter; you have to take into account Bayes' theorem. I'm not sure how many pysicians understand Bayes' theorem, but the results are kind of worked into standard practice anyway. Most people sure don't. John Allen Paulos wrote about this in his book Innumeracy. A false positive in a home test is a very big deal.

I just looked it up and, from coarse numbers, without regard to risk factors, it seems that the HIV infection rate in the US is about 0.4%. Now, if the test only gives 0.4% false positives, and a randomly selected person takes the test and gets a positive, there's about a 50% chance that it's a false positive.

This would not be a problem if people understood this and responded to risks rationally, but they don't. I'd be concerned about suicides of people who got false positives.

There's only one way I can see this would work, and that is for such a test never to tell a user that he and/or she has HIV with a greater probability than is justifiable from Bayes' theorem. Make it like the HbA1C home test with a digital readout.

 

[Roboramma]

I'd also like to add that I think a home test would be useful even if the number of false negatives is relatively high. I don't buy the argument that a false negative will drive an HIV-positive person to infect others: High-risk individuals tend to act as if they're clean until they find out they're not, so they'll already be infecting others. The damage is already being done, and a false negative won't change anything since it will simply appear to confirm what they already believed.

The problem I see is that people will now be more likely to take the home test than to go in and be tested professionally.

If the home test is more likely to give a false positive, those infected who take it are more likely to become confident that they are actually not infected and not get a professional test.

In balance will the fact that more people are likely to get the test more than make up for the fact that some will get false results? I don't know.

 

[toddjh]

In balance will the fact that more people are likely to get the test more than make up for the fact that some will get false results? I don't know.

I think it would. Getting an HIV test is a very embarrassing thing for some people. I think there are an awful lot of people who are too shy to go to a lab to get tested, but who would gladly order a test off the internet and do it themselves.

The main argument I've heard against home testing hasn't been reliability, but rather the need for health agencies to gather accurate data about HIV infection rates in different demographic groups. Home tests could make that information hard to come by.

Jeremy

 

[Art Vandelay]

I don't know if the AIDS groups mentioned in the article who oppose the home kit are the same AIDS groups which oppose testing everyone for AIDS who walks into a doctor's office for any reason (such as the flu) because it violates a person's privacy,

Do you seriously think that proposal even deserves consideration? Why test people who walk into a doctor's office? Won't that just make people less likely to see a doctor? Why not just force everyone to get a test? And then we can force everyone who tests positive to list every sexual partner (right before being shipped off to a quarantined island, that is).

As far as privacy is concerned, it strikes me that there is another aspect that no one has mentioned: if I have a home testing kit that I can use to test myself, how hard would it be to test someone else?

 

[Perpetual Notion]

I worked at an HIV testing site for a year and I think the home kits are a good idea. They actually already exist but you have to send the swabs into a lab and then call an 800 number for the results. They have counselors available when you call. I've never used the kits so I don't know how good the counseling is or if they would have resources specific to the areas where people live which I think would be important. You need to know where your local HIV/AIDS case management group is.

The drawback to having people come in for testing is that a lot of people no show. They make the appointment right after they realize they made a mistake or are having an attack of guilt, but then they have to wait for a week and they rationalize everything or go into denial and that's the end of it. If they do show up, they then have to come back sometimes up to two weeks later to get their results. Where I live the county can't afford the more expensive in office tests and you have to come back a week later. The interesting thing is that the vast majority of people who I had who tested postive didn't come back for their results. They had a very good idea what their status was and didn't want to get the official word.

A lot of people don't want to go to their doctor because they don't want it showing up on their insurance and they don't want to go to public clinics because of the stigma. The major clinic in my town is in a not so safe area and I have been there when gang members have been hanging out in the parking lot. If you have kids with you then I can understand not wanting to go there.

When you tell someone their status they pretty much stop listening. People who are negative, a lot of them, are like "alright, back to the party!" (we'd get a big rush of college students right before Spring Break) and people who are positive are in shock and can't absorb anything you say. You provide them with resources and information on not spreading the virus and make sure they have a safe way to get home. People will always see things in black and white and whether they're testing in an office or at home probably won't change that. As long as they know how to access resources in their area then let them test.

 

[Melendwyr]

If worse comes to worst and some fool kills himself over a false positive, what of it? That's just poor judgment on his part.

I realize it's somewhat callous to simply disregard people's lives, but when dealing with a sufficiently large population, deaths are the consequence of almost every public policy and action. It's not as if the test itself has a risk of personal injury. If people choose poorly, we're not obligated to take responsiblity for their decisions.

I also suspect doctors and therapists are vastly overestimating the value of their counseling, but that's merely a suspicion.

 

[CBL4]

Originally posted by Melendwyr
I realize it's somewhat callous to simply disregard people's lives, but when dealing with a sufficiently large population, deaths are the consequence of almost every public policy and action. It's not as if the test itself has a risk of personal injury. If people choose poorly, we're not obligated to take responsiblity for their decisions.

I love it when someone takes what is a muddled thought in my brain and expresses it succinctly.

CBL

 

[Perpetual Notion]

If there is no way to produce a home test which doesn't produce a false positive then we are left with testing people in clinics and doctors offices. Doctors and therapists tell people that they have terminal illnesses all the time. Are they not supposed to do that because the patient might go off and kill themselves? If we so overestimate the value of our counseling should we just pitch it in? There is no way to know how someone is going to react down the road to their diagnosis. If you know someone is going to harm themselves then you can take action but you can't prevent it if you don't know what's going on. You can do everything in your power, hospitalization, counseling, no-harm contracts and they can still walk out the door and kill themselves.

I do know that most people who are HIV+ and have it together enough to take care of themselves and others will do it. I don't want anyone to be told they're positive when they're not, but I want as many people to get tested as possible and the more safe options we have for this to happen, the better.