An Evaluation of Routine HIV Testing and Its Impact on HAART Initiation

A Discussion With Bonnie Goldman, Editorial Director, The Body PRO

May 14, 2009

Multimedia Toolbox

Listen to Audio (52 min.)

Please note: These files can be quite large. Allow some time for them to download.

Table of Contents

Note: This CME/CE activity expired on May 14, 2010. For a list of currently available activities, click here.

What follows is a discussion between clinician researcher David Alain Wohl, M.D., from the University of North Carolina at Chapel Hill, and Bonnie Goldman, editorial director of The Body PRO, on the topic of routine HIV testing and its impact on HAART initiation in the United States.


David Wohl, M.D.
David Wohl, M.D.
Today we will be discussing the revisions to the recommendations for routine HIV testing in health care settings in the United States, which were issued by the CDC (the U.S. Centers for Disease Control and Prevention) in 2006;1 their implications for the clinical management of HIV; and our ability to contain this epidemic.

We're going to break this up into a few different sections in order to try to:

  • Describe the landscape in which HIV testing is being conducted nowadays and explain the motivations behind the CDC recommendations issued almost three years ago, but only slowly emerging now into practice.
  • Explain any evidence that these changes are having their intended effect.
  • Explain the delay in the implementation of these recommendations and detail a few of the obstacles that are preventing the recommendations from being fully implemented.
  • Discuss the ways in which people are working around the implementation delays.

Motivation for CDC HIV Testing Recommendations

To start off, it's important to explain why we need to expand HIV testing in the U.S., though I think we're preaching to the choir, since the people who are most likely to be reading or listening to us right now are well aware that we're in the midst of a really horrific epidemic that we haven't been able to contain in the way that we would like to.

Even as HIV therapies have become more potent and more effective -- preventing millions of deaths worldwide -- we've learned recently that the CDC was off by about 40% in its annual estimates of how many people are getting infected with HIV every year in the U.S.: Rather than around 40,000 people per year getting newly infected, we discovered it may be around 55,000 - 56,000 -- maybe even as high as 60,000 -- persons getting infected with HIV per year.2 So we're feeding the pool and more and more people are living with HIV in our country.

As many of you are aware, there are some subgroups where we see more HIV infections, where the increase has been much more dramatic -- namely, in men who have sex with men. We're seeing subpopulations of men who have sex with men that are at increased risk. These are data that have come out over the last year or so, and have helped us target HIV testing.

So when we talk about HIV testing, we're not only talking about expanding testing and making it more routine, we're keeping in mind that there are certain groups that we have to pay more attention to, as well.

Prevalence of Undiagnosed HIV Infection Remains High

One thing that has been fairly consistent in most of the data that has been presented to date is that there is a sizable population of HIV-infected persons who are unaware that they're infected.3

Most of the data indicate that, of the one million or so people with HIV in the United States, about 21% to 33% don't know that they're infected. Some estimates and some modeling indicate that they account for more than half of the number of new HIV infections that occur every year. This is a big deal, and identifying those persons who are unaware that they are HIV infected has become paramount in our thinking about how to contain this epidemic.


Again, there are some epidemiological clues, of people who don't fall into any of the high-risk categories, who can be HIV infected and are being missed with our current HIV testing policies. Many women, for instance, do not have a concentration of high-risk behaviors, or risk factors, yet have become HIV infected because a partner has had some risk factors. But unless women are pregnant, they are not offered routine HIV testing. So we must think beyond the specific categories of high-risk groups.

Thus, there has been an effort to think about how to reach those people who don't know they are HIV infected, and who unwittingly may be transmitting their virus to others.

Recognizing the data and where they were trending, the CDC rightly convened a series of meetings to address whether the old paradigm for HIV testing was outmoded and needed to be updated and streamlined. The old testing paradigm had been designed to respond to the initial aspects of the epidemic. HIV testing is somewhat cumbersome when done in the traditional manner of extensive pretest counseling, followed by the actual test, a waiting period if a rapid test was not used, and then bringing people back and doing post-test counseling. Written informed consent has to be in there too, and that was proving to be rather burdensome.

There are clinicians in the community for whom those extra steps made HIV testing a challenge to order and follow up on -- and emergency rooms and urgent care centers were basically ruled out from performing HIV tests because of the need to track down people, bring them back in, etc.

Recognizing this, there's been a new set of recommendations, and that's what we're discussing today.1 I think the pressing need is that we're seeing this unabated incidence of HIV infection that has not changed at all in aggregate. We have seen plateauing of the epidemic in some subpopulations, but in others we're seeing an increase in numbers -- namely, in men who have sex with men.

Presenting to Care With Advanced HIV Disease

Even those who are getting tested for HIV seem to be getting tested pretty late, and that is causing a lot of difficulty, as well. In the U.S., the median CD4 cell count on presentation with HIV is estimated to be around 190.4 Thus, not only is HIV testing missing people who don't know that they are HIV infected, but people who are motivated to get tested are getting tested rather late in their disease.

Research has shown that these late testers are somewhat younger than people who test with less advanced disease. According to CDC data from 16 states, late testers are more likely to be heterosexual, less educated and African American or Hispanic.5

Were there regional differences? Were these late testers mostly in the South?

No. This was a diverse collection of different states. It wasn't concentrated in the South. Again, it fits with what we would think -- that heterosexuals may perceive themselves to be at less risk. Those who are less educated may be less informed, or have less access, because it may be a surrogate for poverty or living in places where there are not as many resources. Younger folks perceive themselves to be at less risk than older individuals. African Americans and Hispanics may have issues with access to care, perceptions of risk, and acceptance of wanting to get tested that can differ based upon a whole bunch of different socio-demographic factors.

It's also interesting that -- when you look at the reasons for testing, and compare late testers versus early testers (people who tested less than a year before an AIDS diagnosis versus people who tested greater than five years before an AIDS diagnosis) -- the late testers are more likely to be tested in the context of an illness. So they were sick, and they got tested. Many people still only get tested when they're sick with an AIDS-defining illness.

In comparison, people who were early testers, tested way before they got an AIDS diagnosis. They were more likely to come in for testing because their partner was perceived at risk; they perceived themselves to be at risk; they just wanted to know; or importantly, for the discussion we're having today, they were tested during a routine checkup. So the routine checkup, the routine medical encounter, was an opportunity for them to be tested, whether they brought that motivation to the encounter, or their clinician suggested it to them.

So there was really a difference between late and early testers. Late testers very rarely got tested during a routine checkup, or because they wanted to know, or because they thought they were at risk. They are much more motivated by symptoms, and so it's sort of imposed upon them.

I've seen some interesting studies about late testers. Many late testers seemed completely unaware that they were participating in high-risk activities.

This is the paradox, as I call it, of our current epidemic. There's a disconnect in HIV knowledge and understanding among the general population, and despite many more opportunities to know about HIV via the Internet, TV, movies and other media, there's a lack of knowledge in the subpopulations that are much more at risk for HIV than others. In comparison, 10 to 15 years ago -- when we didn't have as ready access to some of these technologies -- there seemed to be quite a bit of understanding about HIV in the communities affected.

So we are seeing this knowledge gap, though not everywhere, and not always. There are differences between rural settings and metropolitan settings, and different parts of the country, and different age groups.

This is a very prevalent illness, with one out of every 300 persons being HIV infected in the United States.2 How do we approach that? I think the CDC's very rational approach has been to make this a much more routine aspect of medical care. As we'll get into, there's some precedent for this. I think the CDC is also being driven by data that show that even those of us who are in the know, even when clinicians are confronted with patients who are basically screaming out to be HIV tested, we are just not doing as good a job as we should.

Failing to Test for HIV in Middle-Income People Diagnosed With AIDS-Defining Illnesses

There are data that are pretty striking, looking at a large group of individuals with private insurance who came into the hospital with conditions that are associated with AIDS.6 By that, I'm talking about Burkitt lymphoma, primary CNS [central nervous system] lymphoma, encephalopathy, wasting, histoplasmosis, Mycobacterium avium complex, all the way down to CMV [cytomegalovirus] retinitis and PCP [Pneumocystis pneumonia]. Remarkably, the HIV screening rate for these individuals was fairly low, under 25%. And usually fewer than 10% of individuals coming into the system were not tested for HIV, even though they had indications for HIV testing.

Again, as a health care system, we just haven't always risen to the occasion, and there has been unevenness in our ability to diagnose people with HIV. Varicella-zoster virus infection, tuberculosis [TB] or a positive PPD [purified protein derivative] skin test for TB, a sexually transmitted disease -- these are absolute indications for HIV testing, or at the very minimum, taking a very detailed sexual history and drug use history.

Rationale for Expanding HIV Testing Now

Early on in the HIV epidemic there was a sense by many patients that there would be little point to discovering whether or not they were HIV infected, since those were the days when we had only AZT [zidovudine, Retrovir]. In other words, there was a limit to what could be done to help these people.

We have come a long way, of course. Now there's additional motivation for individuals to get HIV tested, because we can:

  1. Improve the health of HIV-infected patients by getting their virus under control;
  2. Try to prevent them from getting opportunistic infections by using not only antiretrovirals, but prophylaxis when needed; and
  3. Reduce their infectiousness through treatment and risk modification interventions so that they don't infect someone else.

So there are a whole lot of good reasons and rationale for expanding HIV testing now that maybe weren't as operative way back when.

CDC Recommendations Address Missed Opportunities to Test Undiagnosed HIV-Infected People

Dr. Wohl, can we discuss the California study a little bit? I understand that 59% of the study sample was comprised of middle-aged, middle-income people in the U.S. South who were properly diagnosed with AIDS-defining illnesses, but never given an HIV test. If the same thing is going on among the poor and among the middle class, then it seems like we have a really big problem on our hands. How are we going to change that?

I think what happens is -- and it's hinted at in both the study that was reported from California and in other data -- and this is what we're hoping to neutralize with the CDC recommendations for HIV testing -- very, very young people oftentimes are not tested.

That may be because there is a perception that they are at less risk, or they don't want to get tested. This is also the case for older individuals. Oftentimes -- and this is pretty consistent with most of the studies I've seen -- when a patient is over 50 or 60, he or she is not as likely to be asked to get an HIV test -- because there's embarrassment, because we think that older people, for some funky reason, don't have sex.

Physicians just haven't done a really good job of approaching people in an evenhanded way. There are these biases that clinicians bring to this encounter when we talk about HIV testing. And we really have to get over that.

There are data that show that if we have an objective approach to HIV testing, we do much better, and we actually find more people who are (a) willing to get tested, and (b) at risk. And some of those people are infected.

I don't necessarily know that this is a regional issue. I think that this happens all over the U.S. There is a perception by health care providers when they look at a patient that: "You're not at risk, so I'm not going to offer you the test, and I'm not going to offend you by asking you about it." This is what we have to really get over.

That's a centerpiece of the new recommendations -- that HIV testing isn't targeted to specific groups based on risk, necessarily (and that could be a criticism of this), but that HIV testing, while still voluntary, is routine for all persons age 13 to 64 in health care settings.

People who are at risk for HIV should be screened much more often, even annually. However that, again, should be part of the conversation with an individual. To get around some of the biases that are brought up by that study, and others, we need to test everyone on a routine basis without that filtering of risk assessments, which clinicians oftentimes don't do very well.

What's ironic is that many people think that when they receive a full medical checkup, they're being tested for HIV. Most people are unaware that there are these protections in place. I've heard many anecdotes from those who are diagnosed with HIV years after they were infected. They are surprised and invariably say that they had had many "full" medical examinations along the way during which they thought they were being tested for "everything," including HIV.

Yes. Exactly right. In fact, there was a fascinating study from South Carolina a couple of years ago.7

Many people in South Carolina who are HIV infected depend upon Medicaid for their care. So the researchers were able to look back at the public health insurance records, and look at the time at which patients were diagnosed and then backward to how many encounters with the health care system these people had had. It was remarkable. Between visits to the ER [emergency room] and urgent care centers, as well as routine clinician appointments, the patients had had an average of four encounters with the health care system in the previous year or so prior to their HIV diagnosis.

The title of the presentation was "Missed Opportunities." And the missed opportunities were for HIV diagnoses. These were individuals who presented to care with a typical CD4 cell count of around 200. So they had been HIV infected for a while -- certainly, at least a year -- and had encountered clinicians, encountered medical facilities, but had never been tested for HIV. Those were missed opportunities to diagnose them at higher CD4 cell counts, and before they came in with, oftentimes, AIDS-defining illnesses.

This is a really, really important point that I think we have to stress. When the old system was in place, we missed a lot of people with HIV infections. A consequence of that is that a sizable proportion of the HIV-infected population doesn't know that they are infected. Another consequence is that people come in late.

We also see the same misperception that you're describing among people who are hospitalized or given a blood test -- "Well, they must have checked me for that."

We see the same among the incarcerated. People aren't always tested for HIV when they are incarcerated. States have different systems. Jails certainly don't do a whole lot of HIV testing. When you look at jails across the country, some do, some don't. There are these misconceptions about people getting tested; there is a lot less testing than, I think, people think there is.

The other part of the revised recommendations that I want to highlight goes beyond just who we should be testing -- i.e., that sexually active group, aged 13 to 64. That doesn't mean, in my mind, that we should stop at 64 and we shouldn't be screening 65-year-olds, or 68-year-olds. It depends upon the situation, and what's going on. I think that there's no upper limit, I think starting at 13 makes a lot of sense, for obvious, and disturbing, reasons.

CDC Recommendations Streamline Informed Consent Process: From "Opt-In" to "Opt-Out" HIV Testing

But the other part of it is that the voluntary nature is still operative, though the informed consent process has been streamlined, and made much more efficient.

This, I think, took a little bit of guts for the CDC committee to pursue. What we're talking about is switching from an opt-in scenario in which people are asked, "Do you want to get an HIV test?" and they have to say yes, to opt-out screening where you can order an HIV test without jumping through hoops, as long as the person has an opportunity to say no, which is much like we do with most of our testing.

For example, if I wanted to test someone for hepatitis B, hepatitis C or syphilis, I would tell that person that I'm going to do that. If he or she said, "I really would rather you not do it," then I wouldn't do it.

Opt-out HIV screening, with the opportunity to ask questions, and the option to decline, is another major aspect of this change that was recommended by the CDC a couple of years ago.

Dr. Wohl, could you describe what you mean by opt-out? Is an individual just tested, and then told about it? How does that work?

I think we have to stress this over and over again: Opt-out testing is not mandatory testing. When done properly, a clinician will tell the patient that he or she is going to draw an HIV test, unless the patient objects, unless the patient says no, which every patient has a right to do.

As we'll talk about, there are still some legal obstacles in some places. But that's changing over time, too. It's also a sea change in our thinking about HIV as a special entity. Certainly, there are things about HIV that are special. On the other hand, the more we treat HIV like other important diseases, the more we can perhaps move away from the discrimination and the stigma that attends this particular viral infection.

Importantly, as well, the guidelines indicate that written informed consent need not be obtained. In the case of syphilis, hepatitis C, prostate-specific antigen [PSA], or even some genetic testing -- sensitive tests -- we don't typically require informed consent to be written. It is a verbal agreement.

HIV, as well, could be handled under that sort of paradigm. I think that is, again, helpful, because it treats HIV like other infections and diseases, and also removes this hurdle that health care providers would have to jump over of getting the form, printing it out, having the person sign it and read it, of course. That just took time, and I think that really made it very difficult for people to do this as much as they should.

Particularly in emergency rooms, and places like that.

Yes. Also, the man-hours and the time involved, and the interruption of flow, made it very difficult.

It's that exceptionalism of HIV that's reinforced with that approach. What is key, I think, is that the recommendations are, for the moment, based largely in health care settings. We're talking about emergency rooms, urgent care clinics, inpatient facilities, certainly STD [sexually transmitted disease] clinics, TB clinics, public health clinics, substance abuse facilities, correctional facilities and primary care settings. Essentially, we're deputizing clinicians at all these different centers and saying to them, "This is your opportunity." This is where you have the chance to offer HIV testing to that 25% or so of the U.S. population who don't know that they're HIV infected.

There are some special considerations that I think we have to also have, beyond what we're talking about with respect to expanding HIV testing, making it more routine and testing individuals across a very, very broad age range.

HIV Testing in Pregnancy: CDC Recommends Opt-Out Testing

Pregnant women often get prenatal care, and that's another opportunity. We have done a really good job of offering HIV testing to pregnant women. That's had a profound effect, as we know, on perinatal transmission of HIV, because there are certain interventions we can do that are very effective at reducing the risk of HIV transmission from mother to child. I think we could take a page from that book. We often now provide pregnant women with a sort of opt-out scenario for HIV testing. We are certainly talking to them a lot more about it.

The revised CDC recommendations talk to this with opt-out rapid testing -- with an option to decline -- for all women presenting for prenatal care with undocumented HIV status. We know that these women are sexually active, because they are pregnant.

HIV testing should occur early on in women's pregnancy care, and again in the third trimester since transmission is much more likely during delivery.

Rapid testing of the newborn should also be considered if the mother's status is unknown at delivery, because a woman may present at delivery without prenatal care, or not have been offered HIV testing. Newborn testing with a rapid test can be done, and is fairly feasible.

In the state of New York, there is mandatory HIV testing of newborns.

Right. And there have been many discussions about this. Infants are tested mandatorily for a long list of things, some of which are a lot less common than HIV, especially in certain areas of our country. So it just makes sense that, if this is a bigger health risk than some of the things that we test for, we should test for it.

What's interesting about the testing of newborns, in states where it is mandatory, is that there's no opt-out. Most people aren't even aware their infant is being tested.

Yes. I think this is really interesting. It engenders a lot of different debating points that we could talk about with respect to autonomy. Pregnant women are being tested either themselves or via their newborns. Because if you test a baby for HIV antibodies, you're really testing the mother for HIV antibodies.

This has been an age-old debate in New York. Different states are approaching it differently. Even in places with opt-out testing, there are rigorous efforts to test women for HIV who are pregnant.

Certainly, there are data that indicate that women who are offered voluntary testing -- when counseled about the benefits of HIV testing and interventions to reduce transmission and improve their health -- have extremely high rates of testing acceptances. This is a group that, generally speaking, you do not mandatorily have to test. With the proper resources and the proper instruction, almost universally, women who are of unknown HIV status say yes. That's according to data that is fairly old by now, but that I don't think is out of date, at all.

Data on Opt-Out Testing for Pregnant Women

Moving on, I think that we have to look at what evidence there is that this new way of thinking about HIV testing actually works. There are a number of different experiences that have been reported where opt-out testing under this scenario has been studied. We know that there have been randomized control trials of opt-out screening for pregnant women.

In the trials that have been done, there is about a 35% acceptance of HIV testing using opt-in consent8 (the traditional way in which we have been doing HIV testing). With opt-in testing, women reported feeling that agreeing to an HIV test was an acknowledgement of high-risk behavior on their part.

Whereas, when women were tested for HIV under an opt-out scenario, the rate of acceptance of testing was about 88%, and there was significantly less anxiety about testing. HIV testing was perceived as being much more acceptable when it was part of routine prenatal care. Opt-out testing was not looked at by the women as something that was imposing.

Comparing Opt-Out and Opt-In HIV Testing

Opt-out and opt-in testing have been compared in a number of other settings, including STD clinics. This has been done in the United States and elsewhere. I think one of the nicest studies we've seen came from the Netherlands, where in an extremely busy STD center in Amsterdam, with 25,000 new visits annually, an opt-out testing program was instituted in 2007.9

The study looked at the pre-2007 HIV testing acceptance rate among the clinic's clients, which included men who have sex with men and heterosexual patients.

Remember, these are men and women getting STD testing at an STD center, so they all have a pretty high risk for having HIV. The study then looked at the HIV testing acceptance rate after January 2007, when the new opt-out testing policy was instituted. Remarkable changes in the rate of HIV testing were seen.

Among both men who have sex with men and heterosexuals, the researchers noted a dramatic uptake in HIV testing with the opt-out scenario. What's interesting is that the proportion of people with HIV did not change. By testing more people, even with the same prevalence, we identify more people who have HIV.

Thus in 2007, with the full opt-out system in place in this Amsterdam clinic, only 12% of men who have sex with men and 4% of heterosexuals opted out of HIV testing. Both figures are a fraction of what was seen beforehand, when 40% or more of men who have sex with men and over 25% of heterosexuals did not test for HIV in this high-risk setting.

A study from the United States, with data mostly from Texas, also looked at opt-out testing compared to opt-in testing in STD clinics.10 These researchers found dramatic increases in the number of people saying that they would get tested with the opt-out scenario.

Here in North Carolina, we have looked at this as well in our correctional system. The data are unpublished, but have been verified. The study looked at HIV testing under a voluntary opt-in situation before Nov. 1, 2008, and then subsequently after that date, when an opt-out testing policy came into place. HIV testing among inmates coming into the prison increased from around 58% to over 90%. Again, opt-out just seems to streamline things and gets people tested much more efficiently than the old paradigm of opt-in.

What some health care providers are doing -- and the CDC recommendations speak to this -- is, that when patients come in for general care, most, if not all, have to sign a consent to care -- a form that says, "You consent to have this, this and this done, and be treated by this clinician."

On that list could be, or should be, "... and we may do HIV testing." Now, you don't have to sign that form, or you can sign the form and indicate, "But not this test." Again, that doesn't negate that clinician's responsibility to later tell the patient before the actual HIV test is given.

Again, opt-out testing, incumbent upon that is the option to defer that testing. It's not mandatory testing. That's the way it should be practiced. If someone is drawing an HIV test and not telling the individual, that's mandatory testing, and that's not what we're talking about.

Can you explain how that would be mandatory testing? If a patient visits a clinic and there's a suspicion of something -- and it may not be HIV -- the clinician doesn't usually get detailed permission from the patient to do specific tests.

I think there are some things that are a little bit different. Again, we don't want to put HIV in a very, very special category, but we can't ignore that it does have some special aspects to it. I would argue that if I was going to test someone for hepatitis C, hepatitis B or syphilis, I would probably mention that to that individual.

We generally tell people when we're going to check PSA or cholesterol. Now, of course, we don't necessarily tell people that we're going to check their sodium or potassium levels. I think we have to explain things when they have important ramifications.

If I'm going to test you for cancer, that's a big deal. You'd want to know that I was testing you for that. I think it is the same with HIV testing. Genetic testing is another instance in which a clinician should talk to the patient before performing the tests.

I think HIV is not that different of a disease that it has to be in its own category. I think it's very similar to some of these other diseases in that the result that comes back -- whether positive or negative -- has important implications for that person's life, and you would want to prepare that person for it.

In this case, especially as we're switching from a system in which we had to go out of our way to get a patient's permission, to a new system in which we're making it so that the onus is on the patient to refuse the test, I think we should do a good job of advising that person. "This is why I want to do this test. It's routine. We do it for everyone coming into care." Or, "I'm a little worried about this." Whatever the reason is, the patient should be informed. Another nice thing about doing that is that patients then know that HIV testing is becoming routine.

If you do it blindly, not only do I think you've crossed an ethical line, but I also think you're doing a disservice to the whole rationale for this HIV testing system, which is to make it familiar, in that people get tested all the time, rather than something out of the ordinary.

Dr. Wohl, don't you think we need a sea change in society before HIV testing becomes the norm and is not stigmatized. At St. Vincent's Medical Center in New York, as they were attempting to integrate HIV testing for inpatients, they found that house staff were resistant to testing everyone. This took place not long after New York's governor at the time, Eliot Spitzer, was found to have been frequenting sex workers with possibly inconsistent condom use. Trainers used the governor's wife as an example of someone who, although perhaps at risk for HIV, would ordinarily never be tested for HIV. But even with this example, the clinicians still were hesitant about testing people they felt were not at risk.

There are two kinds of stigma. One is the traditional way we think about it, which is: I don't want to get tested, because if I find out that I'm positive, I won't be able to get a job, I won't be able to get housing, no one will sleep with me ... all that sort of stuff.

But the other sort of stigma, which you're talking about, is really a bias: People -- including health care providers -- think they know what HIV-infected people look like. You're right, in that this is a big problem.

What I love about these recommendations is, they completely address what you're talking about, and hit it head on. Which is saying: you should be testing everyone. It's just so sad that the recommendations were issued at the end of 2006, and here we are at the beginning of 2009 and this is only slowly being implemented.

Now, there are structural impediments for emergency rooms to do what we would like, what would be ideal. And those impediments are fairly obvious, and we could talk about those. But, in the case of Silda Spitzer, when she goes to her local primary care office, this is something that should be on a checklist with a mammogram and other types of screening. This is something that should be part and parcel of the maintenance of wellbeing that is in our medical care system.

The lead that I have for this -- I keep coming back to it -- is prenatal screening. Because we really have done a pretty darn good job of offering HIV testing to a sizable proportion of pregnant women in the U.S.

Now, someone may come in and say, "I'm not at risk for HIV, and I don't want you to test me for HIV, even though I'm pregnant." That's fine. But at least we should be offering it.

What we found is that by offering HIV testing as widely as we have and then doing something about it when people are discovered to be HIV positive, we have reduced new cases of pediatric HIV to almost an endangered species in most parts of the country.

I really have learned a lesson from what we've done with pediatric HIV. It's important to look at how we did that and what we can do to implement routine HIV testing for the rest of the population. This report is part of the effort. The idea is to do what we did to get obstetricians, gynecologists and primary care family practice docs to think about HIV as part of the prenatal screening. We have to do that for the care of individuals coming into primary care. So I think we have to target clinicians.

This is not something you put on the side of a bus: "Ask your doctor about HIV testing." No. This is something that should go into medical journals: "Ask your patient to get HIV tested." This is what we should be doing. I think this has to be clinician-led.

It probably should also be taught in medical school, because it seems that one of the problems is that health care providers feel awkward asking patients to get tested. It's just too embarrassing for some clinicians.

Three words: Get over it. We should feel comfortable talking about blow jobs. We should be comfortable talking about rimming. We should be talking about things that our patients sometimes want to talk to us about. We think they don't want to talk about these things, but often patients do want to talk to us about sex. They want to talk to us about death.

There are things we think are taboo. But you know what? When you start to ask people these questions, you get answers. When you ask about sexual orientation -- even if you may be afraid to do it, you find out that your patient is at risk. You start to get more and more information.

These blinders, these perceptions that we have, they keep us from helping people. If you can identify one person at risk for HIV, test them, find out they are HIV infected, get them on treatment, save that life, and prevent them from transmitting it to someone else, then it's worth the thousands of times you had to blush a little bit when you had to ask some older woman, "Are you at risk for HIV? Are you sexually active?"

There are instances where elderly women do turn out to be HIV infected. An example is the following response: "Well, now that you mention it, my husband died 10 years ago from a mysterious illness." This situation has happened and will continue to happen.

Or they were sexually assaulted, and they don't want to tell anybody. So I really think we should get over it. This is a routine part of medical care.

One out of every 300 persons in the United States is HIV infected.2 Not all these people have the same skin color. Neither do they all have the same sexual orientation or even track marks on their body. We don't have X-ray vision.

We have to make this much more routine. We may then find out it's a little bit more than one out of every 300 people; that's an estimate. But we won't know until we look at this. We have a big problem in this country. I think clinicians here can take a fantastic lead.

There's a wonderful study11 that was done in Boston, looking at exactly this issue. Providers were asked to screen people coming into emergency rooms for HIV. This was compared to having a dedicated HIV counselor who staffs the emergency room and tests most everyone who comes in.

As you would imagine, the providers didn't do a very good job. The researchers looked at over 8,000 eligible people who came in. Fifty-nine percent agreed to participate in this study, and were randomized to either having their provider offering them the HIV testing, or the dedicated counselor. Eighty percent of those who were randomized to the counselor were offered an HIV test, compared to a paltry 35% in the provider arm.

Acceptance of HIV testing was comparable in the provider arm (75%) and the counselor arm (71%). So again, what we're seeing is that acceptance of testing was pretty high among people who agreed to it. But so many more people in the counselor arm agreed to be tested.

I think that this is really important. It shows that if we have non-denominational, objective people offering HIV testing without those preconceived notions, acceptance rates are fairly high.

It's pretty ironic, because all clinicians use universal precautions. In effect, they practice medicine based on the assumption that anyone can have HIV.

But then there's this awkwardness around actually offering an HIV test.

I think some providers do a really good job, and some don't do a really good job. There are still a lot of patients who are offered HIV testing who refuse it. Data show that people need some education, too. The community needs education. People still are not getting HIV tested because they are afraid, or because they feel that they don't have much of a risk and are not worried about it. We have to get around those obstacles to HIV testing, too.

Even in the Dutch study,9 there was a small proportion of people who did not get tested for HIV. And these are men who have sex with men getting STD testing in an STD clinic in Amsterdam! Amsterdam must be one of the STD epicenters of the world -- yet they said, "I don't think I'm at risk." That's just madness.

So it's not all about the clinician. Clinicians can do better. Clinicians have done well before, and many clinicians are leading the way on this. We just have to make this easier and easier.

In emergency rooms, it is difficult. Part of the reason is that even rapid testing still takes somebody to do that test, or send it to a lab and then get it back. And then there's a time element. Different ERs have to work out different ways of doing this. There are data12,13 trickling in from different ERs in different hospitals showing that this is feasible, but that systems have to be introduced to do this, and to follow up on positive results in real time and, importantly, link people who test positive to care.

It's not good enough just to say, "Well, you're HIV positive." There has to be a route by which identified people can get into care in a really, really quick way. Because lots of people can get HIV tested, but that doesn't mean they are going to get into the care they need. It's not automatic.

I think there's a lot of work that's going to continue to be done, especially in urgent care centers and emergency rooms. But we shouldn't focus just on those centers, important as they are. We've got to be doing this at STD centers, which I think are really on board, and also primary care settings, because those are the places where a lot of people do receive medical care.

I think this is part of a bigger package, and we have to get over some of these obstacles. Where there's rapid testing with dedicated personnel, as in the Boston study,11 these things cost money. This is a question of resources. Especially in these economic times, it's very difficult. But I think those are the things we need to move forward with. Money doesn't solve everything, but money well spent could get you there a lot quicker.

The other thing is that there are states where HIV testing under an opt-out scenario -- without written informed consent -- is still not permitted. Again, this is changing.

There was a nice report by John G. Bartlett, from Johns Hopkins, in the August 2008 issue of JAMA that showed state by state where there are legislative restrictions to the CDC recommendations, where there is an evolving change in the legislation, and where the CDC recommendations have already been implemented.14

Here in North Carolina, we did have a change that allowed us last year to provide opt-out HIV testing in the state. It's still pretty spotty. Some hospitals do it, like mine at the University of North Carolina at Chapel Hill. Other hospitals haven't been able to adopt it yet.

I think we need to apply some pressure to get opt-out HIV testing to be more routine. I think this is important for our professional organizations. When we talk about clinicians, we're not only talking about individuals, but we're talking about the august bodies that represent them. This should be something of a priority for the AMA [American Medical Association] and for other groups that represent clinicians, whether they be family practice docs, internists, OB/GYNs, whatever.

New HIV Testing Options

Dr. Wohl, can you provide a quick review of some of the new HIV testing technologies?

Technology helps us in many ways. The rapid HIV tests are a wonderful example of that. We have seen an expansion of HIV testing since the introduction of rapid testing.

Try to start up a rapid HIV testing program, an outreach program, in your community, and you'll find someone else is already there doing it, whether it be a small operation offering testing at health fairs, or a much more comprehensive program like we have in many parts of rural North Carolina, where people are doing HIV testing in bars.

There are all sorts of initiatives going on, in correctional facilities, in emergency rooms, in the urgent care centers we talked about and in community health care centers. That was really helped a lot by the rapid tests.

Rapid tests, however, aren't perfect. One of the big problems of rapid tests, of course, is that they miss people with acute HIV infection. There aren't going to be a lot of people who are in the acute phase of HIV infection, but it can be sizable.

We have data that show that around 5% of all people getting voluntarily tested, who are HIV positive, are in the acute phase of the infection and they don't have antibodies that would lead them to be diagnosed.15

There are new technologies that are trying to combine assays or techniques for detecting evidence of the virus itself, either using p24 or other techniques, such as the pooling techniques, where blood from different people is pooled and tested for nucleic acids that are present in the virus. These can help detect acute infection in the absence of antibodies. I think this is really, really important. These combined tests that are looking for the p24 antigen in the virus, as well as HIV antibodies, hold great promise in that they can help us cut down on the number of people who are told, "You're HIV negative," but are in reality in the throes of acute HIV infection.

Could you define acute infection and talk about the tests? What is the usual window period?

Acute HIV infection, by that I am referring to people who recently became infected with HIV -- within the last days or weeks. We know that it could take three weeks, or a little bit longer, before antibodies to HIV are detectable by our standard EIA, or ELISA, test. So the definition of HIV positive relies upon these antibody tests.

But we know that there's this window period. It used to be as long as six months. The newer generation of assays in the last few years has closed that window down to about three to six weeks. By three to six weeks after an HIV risk, most people have started to evolve antibody responses to HIV that allow these HIV antibody tests to be positive -- not only by ELISA, but then confirmed by Western Blot.

It's this short period of time that we worry about. That's because it's soon after HIV infection, when there's a lot of virus -- in the millions sometimes -- and no antibodies because our standard HIV algorithms don't detect them.

In certain places -- including in North Carolina -- we have a pooling system. Blood is pooled in an aliquot that may have as many as a hundred people represented in a single tube. The blood is gathered from people who are HIV tested voluntarily at public venues. A viral load test can be done on that tube. If it's negative, no one who contributed to that pool, whose blood was put into that tube, has evidence of HIV nucleic acid. Thus, they are antibody negative and they are RNA negative. They don't have HIV.

But if it's positive, then you can go up generations of these aliquots because they are broken into aliquots of 10. Then, from the 10, some are put into a master tube. You can go back and find out who has nucleic acid sequences that indicate that they have HIV present in their blood, but no antibodies, and detect acute HIV.

So there are different ways of doing this. The idea is to not miss these folks, because there's pretty decent evidence that people who don't know that they are HIV infected unwittingly are responsible for maybe half of new infections.16

There's also data17,18 that show that people with acute HIV infection are also a leading edge of the epidemic, because they are so much more infectious during this acute period and likely to be engaging in the same activities that got them infected in the first place a week ago.

So here you are with 8 million HIV viruses in your blood, and if you're male, probably a similar amount of virus in your semen, and you're going out again on Saturday night, doing what you did last Saturday night when you got infected. Thus you may very likely infect someone else. You may not be sick. You may have some symptoms, but you may not have very many symptoms.

Technologies that can help us reduce that window and identify people with acute HIV infection have really very important public health significance.

This method was first employed by the state of North Carolina. Are there any other states that do this kind of testing?

There is no state that I know that does it throughout the state. There are cities that do pooling, and there are sometimes facilities that do pooling. But pooling has not been widespread. It's not necessarily a very expensive venture. It does add on a few dollars more to the cost of every HIV test, on the order of maybe three to four dollars. But data show that opt-out testing, pooling and things like that are cost effective.

In North Carolina, it's all done with robots, and the state runs the program. Many different states are interested in it. Certainly, some cities have taken it on. San Francisco is an example.

But I think that that's not the answer, necessarily, for everyone, although I think it's a really good answer.

The people who developed that deserve a lot of kudos. But I think that there are other techniques that people are trying to use so they can do point-of-care testing, where you can get an answer much more rapidly and you don't have to wait the days that it takes to do this. I think that's where the technology is going to drive us -- towards being able to do this in the field, where you can get an immediate answer. We're all about now. We don't want to lose the opportunity of that encounter -- to have someone leave us, and then have to try to find that person, because he or she has acute HIV, or chronic HIV.

What can a health care provider do if he or she suspects that a patient has acute infection?

I would draw a regular old HIV test. But at the same time, I would draw an HIV RNA level. I think that's the best thing to do. Nothing beats RNA right now. Even p24 has a little bit of a lag, compared to RNA. So I would get a viral load test at the same time. I think, again, you really want to be thinking about this in people who come in, especially to emergency rooms, with a viral-like illness and any of the sentinel signs of acute HIV.

Though it could also look like just a viral illness, but then, that's where history comes in. It's one of those things where we have to train health care providers -- and we're working on this, believe me -- on what to do when people come in with regular old viral syndromes.

If they have a patient who comes in with what looks like acute viral syndrome, they have to say, "I'm just curious, because there is some stuff that this could also be. Any new sexual partners within the last month or so? Have you ever injected drugs? If you have, have you shared needles within the last month or so?"

It doesn't take too long. Get over the embarrassment thing. Get over the perception thing. I don't care what they're wearing, or what they look like. Just ask it routinely, and surprisingly, you may find, once in a bit, an answer you didn't expect.

Would an RNA test in an HIV-uninfected person be covered by private insurance, or by public insurance?

It depends. This is an issue that comes up all the time -- whether or not any of these things would be covered. Even the opt-out HIV testing has insurance implications for people.

Generally, I think, what we have to do is drive that argument and not react to it as much. And sometimes, we have to tell people that this is going to be an added cost. We have to tell people about it. We have to say that this is indicated.

We're not ordering HIV RNAs on every single person who comes into an emergency room. What we're talking about is selective use in situations where we think there's acute HIV infection. You may have to tell someone that this may not be covered, because it's not an indicated use of this test.

I think you have to be honest with people about it, and just do it. That's something that I think is a very individual decision.

Dr. Wohl, I guess this wraps up our discussion. You've done a terrific job in summarizing the data and highlighting the important issues regarding making HIV testing routine.

Good, because I think this is a really important area. It's a moving area, and I do think people have to advocate. I really feel strongly about this. Clinicians have to advocate for these changes and be on the forefront of making this occur. It's going to make or break at the level of the clinicians.

This is not a community-directed initiative. This is trying to get clinicians to change what they do. I would like to see more and more evidence-based interventions directed towards clinicians, and helping them figure out how to remember to offer HIV testing. There have been some things that have been looked at that are useful for a while, and then over time, they extinguish.

There are a lot of things clinicians have to talk to their patients about -- whether it be to stop smoking, wear a seatbelt, wear a condom. I know there's a lot. This just has to be something on the list that we check off every once in a while.

Thank you so much.

This transcript has been lightly edited for clarity.


  1. Branson BM, Handsfield HH, Lampe MA, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep. September 22, 2006;55(RR14):1-17.
  2. Hall HI, Song R, Rhodes P, et al, for the HIV Incidence Surveillance Group. Estimation of HIV incidence in the United States. JAMA. August 6, 2008;300(5):520-529.
  3. Campsmith M, Rhodes P, Hall I. Estimated prevalence of undiagnosed HIV infection: US, end of 2006. In: Program and abstracts of the 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montréal, Canada. Abstract 1036.
    View poster: Download PDF
  4. Egger M. Outcomes of ART in resource-limited and industrialized countries. In: Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, Calif. Abstract 62.
  5. Nakashima AK, Campsmith ML, Wolfe MI, Nakamura G, Begley EB, Teshale EH. Late versus early testing of HIV -- 16 sites, United States, 2000-2003. MMWR Morb Mortal Wkly Rep. June 27, 2003;52(25):581-586.
  6. Chen J, Tian H, Dahlin-Lee E, Everhard F, Mayer K. HIV testing and monitoring in privately insured members recently diagnosed with suspected AIDS-defining events. In: Program and abstracts of the 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montréal, Canada. Abstract 1044.
    View poster: Download PDF
  7. Duffus W, Kettinger L, Stephens T, et al. Missed opportunities for earlier diagnosis of HIV infection -- South Carolina, 1997-2005. MMWR Morb Mortal Wkly Rep. December 1, 2006;55(47):1269-1272.
  8. Simpson WM, Johnstone FD, Goldberg DJ, Gormley SM, Hart GJ. Antenatal HIV testing: assessment of a routine voluntary approach. BMJ. June 19, 1999;318(7199):1660-1661.
  9. Heijman T, Stolte I, Thiesbrummel H, et al. Opting out increases HIV testing in a large STI outpatient clinic. Sex Transm Infect. December 22, 2008. [Epub ahead of print
  10. Branson BM. Revised recommendations for HIV testing in healthcare settings in the U.S. [Centers for Disease Control and Prevention Web site]. June 2006. Available at: Accessed May 2009.
  11. Walensky R, Arbelaez C, Reichmann W, et al. Counselor-versus provider-based HIV testing in the emergency department: results from the universal screening for HIV in the emergency room trial. In: Program and abstracts of the 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montréal, Canada. Abstract 1039.
  12. Heffelfinger J, Boyett B, Delaney K, Patel P, White D. Factors associated with acceptance of opt-out HIV screening and completion of testing in a large urban emergency department. In: Program and abstracts of the 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montréal, Canada. Abstract 1038.
    View poster: Download PDF
  13. Christopoulos K, Schackman B, Lee G, Green R, Morrison E. Results from a New York City emergency department rapid HIV testing program. In: Program and abstracts of the 16th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2009; Montréal, Canada. Abstract 1040.
    View poster: Download PDF
  14. Bartlett JG, Branson BM, Fenton K, Hauschild BC, Miller V, Mayer KH. Opt-out testing for human immunodeficiency virus in the United States: progress and challenges. JAMA. August 27, 2008;300(8):945-951.
  15. Pilcher CD, Fiscus SA, Nguyen TQ, et al. Detection of acute infections during HIV testing in North Carolina. N Engl J Med. May 5, 2005;352(18):1873-1883.
  16. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS. June 26, 2006;20(10):1447-1450.
  17. Pilcher CD, Joaki G, Hoffman IF, et al, for the UNC Project, Malawi. Amplified transmission of HIV-1: comparison of HIV-1 concentrations in semen and blood during acute and chronic infection. AIDS. August 20, 2007;21(13):1723-1730.
  18. Pilcher CD, Tien HC, Eron Jr JJ, et al, for the Quest Study and the Duke-UNC-Emory Acute HIV Consortium. Brief but efficient: acute HIV infection and the sexual transmission of HIV. J Infect Dis. May 15, 2004;189(10):1785-1792.

This article was provided by


The content on this page is free of advertiser influence and was produced by our editorial team. See our content and advertising policies.