HIV+ question

Factual Questions
21

[[The problem is that all projections of progression to HIV omit the vast untested population. While a deadly disease, and one that does some in quite quickly, the statistics for time to morbitity and mortality are always working from an unrepresented total population of infected persons.]]

I don’t understand this paragraph. Are you talking about the progression from HIV to AIDS? What do you mean by “time to morbidity and mortality”? The statistics of course mostly reflect those who have tested, but we also do have a pretty good idea - at least in certain populations - of how many infected persons there are, from seroprevalence studies.

I think the data we have about HIV and progression to AIDS are actually quite good. For example, as this poster alluded to, we have a good idea of when the virus showed up in various parts of the world by looking at blood samples collected over time for other reasons (the Hepatitis B studies of gay men in California, for example). All the evidence I’ve seen supports the idea that this virus showed up in the US no earlier than the 70s. Where do you get the idea that “party and drug behavior” are good predictors of progression?

22

I was unclear Jill, I meant progression to AIDS (once infected), not progression to HIV, as stated.

Time to morbitity and mortality refers in this case, to average time from HIV infection to AIDS (morbitity) and death from AIDS (mortality). This is common terminology used around all kinds of disease progression estimates. The problem even with seroprevalence studies, is that the total time a person has been infected can only be determined with accuracy, if they have been tested regularly from before infection, such as the MACS study referred to above, and a single cohort (group of people) is followed. These studies have been of gay men and IVDUs in urban centers. Any broad based sero-prevalence study of the general population must have the same characteristcs (following a group over time) to effectively estimate a true figure of seroprevalance in society. Simply looking at new HIV+ test result recipients does not accomplish this - we do not know how long a person has been infected by the time they take their first test. I know of no such study of a cohort that is assembled to represent the population in general, if you do please refer me to it, I am interested in following this up. I think it is an important thing to do, but it has not been done as far as I know.

The problem with estimating the potential deadliness of the virus over the total infected population comes in when people who are outside of these traditional long-term study cohorts (closeted or married gay men, people who do not identify their own risk factors, partners of infected individuals who are kept in the dark) become sick or die of the disease. Their mortality figures up the total death figure, yet they are often not counted in the studies that follow HIV positive individuals, thus skewing the numbers of progressors to total infected.

Drug use is correlated with sero-conversion. Not all drug users convert, that is not implied by the above statement. However, seroconversion is highest in populations that exhibit high drug use, especially hard drugs, from both needle sharing and unsafe sex. This is why many interventions are targeted at these high risk groups.

As far as the progression to AIDS among HIV pos. populations in the early stages of the epidemic, these were gay urban centers where drug use and promiscuity were a cultural norm. Yes this was the late 70’s, and when we begin to see HIV in blood samples. Many of these people (IMO) may have progressed quickly to AIDS by the early 80’s when it was still called GRID, for several reasons that are separate from the virulence of HIV (which may have been less so at that time).

Poor health habits related to drug use

Exposure to multiple STDs that weakened the immune system

Psychological despair from lack of a treatment, and social ostracism

23

[[Time to morbitity and mortality refers in this case, to average time from HIV infection to AIDS (morbitity) and death from AIDS (mortality). This is common terminology used around all kinds of disease progression estimates.]]

Then you would be referring to time from morbiDity to mortality? Yeah, I’ve heard that terminology.

24

[[Drug use is correlated with sero-conversion. Not all drug users convert, that is not implied by the above statement. However, seroconversion is highest in populations that exhibit high drug use, especially hard drugs, from both needle sharing and unsafe sex.]]

You seem to be pretty knowledgeable and to have thought a lot about this. But you are also misusing the term “seroconversion” here. Seroconversion refers to the point at which an infected person’s blood will show evidence of the infection - it usually does not refer to when the person becomes symptomatic.

One would assume that people who use hard drugs and party a lot would become symptomatic earlier than those who take care of themselves, but I haven’t seen any studies that demonstrate that this is so. Have you? And from what I’ve seen (admittedly a limited sample), it isn’t so.
Jill

25

[QUOTE]
Time to morbitity and mortality refers in this case, to average time from HIV infection to AIDS (morbitity) and death from AIDS (mortality). This is common terminology used around all kinds of disease progression estimates.{/QUOTE]

Okay, I think I see what you’re trying to say. You’re talking about time from infection to becoming symptomatic and, separately, time from becoming symptomatic to an AIDS diagnosis. I (and I think most who work in the HIV/AIDS field) consider “morbidity” to include all those who are HIV infected. The disease is a continuum. The AIDS definition is a somewhat arbitrary term, and is becoming less and less meaningful.

26

Jill, did you have any information on any study that follows a cohort meant to be representative of the general population (men women, gay bi straight, etc.) over time to determine with as much accuracy general infection rates - much like the smaller specific risk group studies?

I’d be curious - as I see the lack of any study like this (I may just be ignorant of it) as a stumbling block to a true global picture on progression rates.

Thanks.

27

[[Jill, did you have any information on any study that follows a cohort meant to be representative of the general population (men women, gay bi straight, etc.) over time to determine with as much accuracy general infection rates - much like the smaller specific risk group studies?]]

No, but I’m not sure exactly what the point would be. Seroprevalence studies repeated several times give us an idea of how many in certain populations are infected, and whether that number is growing. We’ve conducted such studies on injection drug users, patrons in gay bars, heterosexual clients in STD clinics, and youth coming through the juvenile detention center. Studies like this are conducted all over the US, along with many studies of those known to be HIV+. It would not make sense - and it would be very expensive with probable low yield - to conduct such studies on low risk populations… if there were infections in those groups, those people would be getting sick, and they’re not.

One of the largest HIV studies of any population - recently stopped for political reasons - was a blinded study testing blood collected from newborn babies (hospitals collect a drop of blood from babies when they’re born to test for metabolic disorders, and some of this blood was tested for HIV without identifying the child or mother, but only collecting data on the region they were from and some demographics). Babies of HIV infected mothers will carry maternal antibodies to the virus whether the babies are infected or not. So the purpose of the study was really to find out how many women of childbearing age were infected with HIV, not how many babies were. This was an extremely useful study, and it was conducted all over the US. In this way, we had an early warning system to discover how and where the virus was spreading among heterosexuals. This information could be used to beef up testing and counseling in those areas, and public education, particularly among pregnant women (who can prevent a large percentage of perinatal transmissions by going on therapy during pregnancy).

Some people don’t understand the purpose or idea of seroprevalence studies that are blinded, though, and threatened to unblind it. So the CDC stopped it before that could happen.

Anyway, it would be really pointless and unethical(in the same way the Tuskeegee syphilis experiment was) to study infected persons to see how long it takes them to progress to AIDS. We have medical treatments that are pretty effective at stopping (or slowing quite a bit) that progression now.

28

Here’s the article I was referring to in my first post about new recommendations for when to initiate therapy in HIV infected individuals. It came out in the Journal of the American Medical Association (JAMA) day before yesterday:
[urlhttp://jama.ama-assn.org/issues/v283n3/full/jst90023.html

Jill

29

Sheesh, I sure have problems with links on this thread.
http://jama.ama-assn.org/issues/v283n3/full/jst90023.html
Jill

30

RobRoy asks:

I can’t answer that question, but I can tell you where much of the early data on HIV seropositivity came from, as well as the then-heretical idea of heterosexual transmission.

The US Army (DoD, actually, but the key researchers were Army docs) initiated mass screening of all AD servicemen & women. They followed every soldier who was found to become HIV+ & developed the first “staging” system. (NB: regarding the ethics of simply observing the progression of HIV infections, that was all that was possible at the time - this was pre-AZT). Their initial findings that a high percentage of men were getting their infections from prostitutes was initially rejected by the scientific community, but later accepted. Female -> Male transmission was not thought to exist at the time. They screened family members & reported fairly high rates of infected spouses (mostly wives). They raised all kinds of red flags about HIV in women when they screened applicants & found equal rates of HIV infection in male & female 18-20 year olds from high-prevalence areas.

Sue from El Paso

Experience is what you get when you didn’t get what you wanted.

31

Thanks for the info MajorMD - I can take your info with more than a grain of slat judging by your handle.

With all due respect, I’d still be sceptical about the data. No doubt the women could be infected from husbands who were self identified as exclusively heterosexual, untested and indenial about their behavior. These men are unlikely to report same sex encounters - I mean loose your health care due to a dishonorable discharge - just when you find out you have a potentially terminal disease? The study to determine risk factors should have been handled by outside scientific consultants, with assurance of anonymity.

Female to male transmission does occur, but social stigmas against reporting homosexual acts (esp. in the military) skews a study. The almost equal female and male infection rates does raise a question. Were most (or nearly all of the females) partnered to male service men who were positive? Hmmm. To find a cohort with equal numbers of male and female infected is unusual. Tell me, where did the women report their infection came from? (men, women, or IVDU?).

This is interesting stuff.

I think the need for large double blind studies is still needed to really teaseout the ambiguities, as the military cannot expect to get accurate self reported information, especially from men.

As a gay man who sees the “other side”: married men, closeted military men and such, I realize that information reported outside the “gay ghetto” can be innaccurate.

32

Oh another note. The almost equal infection rates in Africa for men and women are thought not to be just from heterosexual sex (or unreported homosexual sex) but from tainted blood, and innoculations which are widespread in Africa. This mechanism does not exist to such a great degree in the US.

33

Could be, to some extent, but it is also theorized that some of the substrains of HIV prevalent in Africa (and rare in the US) spread more easily heterosexually than the substrains here do. Add that to high rates of untreated, sexually transmitted genital ulcerative diseases and it seems more plausible that HIV would hit females and males equally. BTW, in New Jersey, about 30% of the cases are female. They attribute it mostly to injection drug use, though.

The comprehensive double-blind tests you’re advocating, Rob, would be very expensive and there would be political opposition to them anyway. Why would they tell us more than we’re learning from the targeted studies being done now, RobRoy? There are some small communities/populations I know of who claim to have a much higher rate of seropositivity than I believe to be true. They don’t trust the epidemiological data that say it isn’t so, and are fighting for more services and prevention programs in their communities. I think that in some cases they are motivated by the dollars available for HIV/AIDS programs. I don’t doubt, however, that many of them believe it and are concerned about their people. If there were all these unknown cases in populations not generally accepted to be at high risk, wouldn’t we be seeing some illness in these groups? Wouldn’t they be showing up in hospitals with pneumocystis carinii pneumonia?
Jill

34

It’s not so much that there are vast numbers of undiagnosed HIV people out there - just that current studies are limited to pre-targeted groups and estimates of the average progression to AIDS in these groups may be an artifact of these groups. People with HIV AND NOT AIDS are undercounted compared to people with HIV AND AIDS, thus I think possibly skewing our general conception of total (or general) average time to progression.

I think the groups mentioned, in Africa, large numbers of people with untreated STD’s and poor health care, the NJ women who are IVDU’s, gay men who are out of the closet and part of a gay urban culture, all of these may be groups that not only contract HIV more, but also progress faster than those who do not enter the “epidemiological landscape” until they have a surprise diagnosis (through denial or honest ignorance) of AIDS.

This is speculation on my part I know, but an area that has not been investigated. Current studies have been made on the basis of some untested ideas. That “gay men” are the only ones engaging in homosexual sex. A recent Gay Men’s Health Crisis poster in NY omitted the word gay and was targeted at Latino and Black men. Result - phones off the hook. Good work, but it points out that studies that look at high risk groups tell us just about the disease in those groups, not the diseae in general or the disease itself.

As a personal observation, much of the gay subculture (I’m gunna get flamed for this, but I am gay, so it’s not homophobia) is centered around a bar culture and it there, in bath houses and gay magazines that cater to a bar going public that advertising and recruitement for studies cheifly occurs. I think this demographic represents a sub-culture that is multiply impacted, with HIV AND alcohol AND drug use. Same for the IVDU population.

Projections of progression to AIDS and death may be skewed in these populations. As opposed to those who - because they are not identified as gay, do not become part of the picture until they are sick with full blown AIDS, and are then ironically counted in a risk group.

It is a hallmark of alcoholism and addiction (sexual as well) that people under report the extent of their problem - that is generally true and has an influence on any estimate of the factors that are actually contributing to poor health in those studied. I’ve been amazed to see friends who were on meds and still drank alcohol, in spite of complications.

35

I know this is a complicated point so i will try to sum it up more clearly.

We need a comprehensive, general population study to determine the REAL time to progression of AIDS - why?

Populations that are routinely followed, IVDU’s, out gay men in urban centers, etc., may also be multiply impacted with other factors that negatively influence health and progression time to AIDS. (Result - overly pessimistic projections of progresion time)

AND

Populations that do not generally enroll in studies may only show up in statistics at the time of death from AIDS. This inflates the number of deaths from AIDS to the total pop. infected with HIV. (result - overly pessimistic projections of number of deaths per number of estimated HIV pos. individuals).

36

The reason to not attempt such a study, Rob, is pretty simple.

In order to start such a study now, you would need to enroll a large cohort of newly infected patients. You would need to have several options for management (observation, early one-drug suppression, early cocktail suppression) AND not deviate from those tracks while you followed the enrollees for 10-20 years or more.

This is NOT going to happen because:

The HIV+ community is extremely well-educated on treatment options, and very insistant upon deciding for themselves what they would like to do based on the best available data. They are very likely to use alternative treatments in addition to their standard treatments & may or may not share that info with their docs. Some get several different standard treatments from different docs without telling the docs involved. I do not see them committing themselves to limiting their options to those allowed within the study.

The second big problem is that the available treatments are changing so rapidly, that no one can study them long enough to know what the long-term effects are. Realize that 10 years ago, an AZT trial was stopped prematurely because researchers found that those receiving treatment did so much better than those receiving placebo that it was unethical to continue to observe the placebo group off treatment. That’s right. 10 years ago, we had NO anti-retroviral drugs. How things have changed. And remember, 10 years is the minimum length of time you could observe people now & draw meaningful conclusions. What new options will become available in the next 10 years? I don’t know, but if past history is any predictor, there will be an exponential growth in the treatments available (unless a single successful cure is discovered, of course).

Actually, since I was at Walter Reed Army Medical Center from 1996-91, I can attest to the fact that the service members being interviewed as to how they might have acquired their HIV were NOT facing discharge for homosexual activity. Soldiers were NOT discharged for being HIV+; they were kept on active duty, and when the disease progressed to the point where they couldn’t work, they were medically retired from active duty with full medical benefits.

I am not not expecting anyone to accept this solely on my say-so, but the Army felt gathering information on how soldiers (and all Americans) were spreading HIV was more important than keeping gays out of the ranks. The one soldier caught en flagrante in his hospital bed with another man was an exception to this open-mindedness, but that was just the last straw in a series of problems for this individual (beginning with a 45 day AWOL period).

The female=male rates of HIV+ were not for soldiers, but for applicants to active duty from high prevalence areas like NYC & NJ. On questioning, past IV drug abuse and a high number of sexual partners were found to both be present in a large number of these applicants.

If you do a MEDline search of HIV articles from the 80s & early 90s, you will see the name Robert Redfield as an author on many of these. He was the senior researcher of the Walter Reed HIV project, and an Active duty Army Colonel.

While the miltary is not a perfect cross-section of the American population, it was the first population with serial HIV testing done on a regular basis so that typically, the time an infection was acquired could be pinpointed pretty easily. A lot of the points you raise are valid concerns; however, the scientific community has recognized the value of this data, and the fact that they are unlikely to get any better data in the near future for the reasons listed above.

Sue from El Paso

Experience is what you get when you didn’t get what you wanted.

37

[[I know this is a complicated point so i will try to sum it up more clearly.]]

Rob, I am an HIV/AIDS epidemiologist for a living. It may be that I am misunderstanding you here, though. Let’s see if I’m getting this right.

I am assuming that you are not suggesting that we find people who are newly infected and follow them without treating them to see how long until they develop end-stage illness. Not only incredibly unethical, but pointless, as we have medications to stop that progression.

It sounds like - and I could be wrong - you are concerned that people are still thinking this is a “gay disease” and that there may be many other cases in other populations that we don’t know about. The only way to get at incidence (new cases) as opposed to prevalence (how many cases there are in the population all together) is to do seroprevalence studies, and repeat them over time. This IS being done. It’s also being done in emergency rooms in some places where there are a lot of cases, to see how many cases there may be out there that we don’t necessarily know about. Again, it is not cost effective to do a study on the entire population. We have extremely good - the best in the world - surveillance of HIV in the United States, and there is strong evidence that our data reflect what is really going on. There are different patterns to this epidemic on the east and west coasts, in the south, and where I am, in the Rocky Mountain west. Epidemiologists do smaller studies in different areas. If there is something new and shocking that comes out of those studies, they are replicated elsewhere.

[[We need a comprehensive, general population study to determine the REAL time to progression of AIDS]]

This is an outdated concept, I think, because again, we have good therapies now. From what I have seen - and this is anecdotal, not statistical - from having personally interviewed over 200 people with an in-depth survey that asks about behavior and health, there is no obvious relationship between drug use, partying and progression to ill-health. Some of the heroin users were in great health, and some of the wealthy, educated, health-conscious gay white men progressed quickly to death. There may be studies to support or counter this observation. Again, it’s just what I’ve seen. I would say, however, that nowadays with the importance of strictly following what may be a complicated regimen of drugs, those who have addiction or depression problems may not do this, and may indeed develop resistance to drugs and progress more quickly.

38

Annuity companies used to insure HIV people. people would actually invest their money in them. However with the new drugs, HIV people are living much longer & the investors are a bit miffed at not getting paid.

39

Yeah, they’re called “viatical settlements,” if that’s what you’re talking about. These companies buy life insurance policies (at less than thier value) from HIV+ people, and get the money when they pay off. I still see them advertised in magazines for HIV positive people, but it seems like a less profitable business nowadays. Fortunately.

40

[[In order to start such a study now, you would need to enroll a large
cohort of newly infected patients. You would need to have several
options for management (observation, early one-drug suppression, early
cocktail suppression) AND not deviate from those tracks while you
followed the enrollees for 10-20 years or more.

This is NOT going to happen because:…]]

I don’t think you stated clearly enough the main reason this is NOT going to happen, Sue. It’s because it would be horrifically unethical (and illegal, of course) to conduct such a study in which therapies that are known to be effective are withheld to “see what happens.” I think the U.S. Public Health Service finally kind of got that message after the Tuskegee Syphilis Experiment finally ended in 1972.
Jill