HIV+ question

Factual Questions
41

Actually, Jill, observation is NOT unethical as long as the observation arm included provisions for treatment once CD4 counts dropped below a certain level. More & more, docs & patients alike are questioning the wisdom of starting drugs as soon as an infection is documented.

From pre-AZT data, we know that the time from infection to opportunisitic infections averaged about 5-10 years. If patients were closely followed & drugs started as soon as CD4 counts dropped, fewer people would be on drugs, lowering costs & slowing the development of resistance. What is unknown is whether this would result in any overall decrease (or increase) in longevity of patients managed this way.

The major stumbling block is researchers inability to design a study today that is flexible enough to accomodate new developments over the next 5-10 years, and to attract HIV+ patients willing to accept randomization, blinding, and non-deviation from the assigned treatment arm.

Sue from El Paso

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

42

I’m talking more about a purely epidemiological study. The only ethical consideration is pre and post test counseling, disclosing a positive or negative result (if that’s what the person wants) and referral to medical programs.

If you wanted to use a general population cross section, you could also give people the option of not getting tested at all, yet following them over time. This would create a picture of the general population, including people who ask not to be tested.

43

To clarify, there would be no arm of the study that:
1)tested
2)got a positive result
3)and withheld that information involuntarily - or forced the patient to go with out treatment

It would be a merely observational study, and not tied to testing a drug. Those do require putting patients on treatments tht work better and discontinuing ones that don’t.

It may be another string, but studies sponsored by drug companies carry their own biases…

44

[[Actually, Jill, observation is NOT unethical as long as the observation arm included provisions for treatment once CD4 counts dropped below a certain level. More & more, docs & patients alike are questioning the wisdom of starting drugs as soon as an infection is documented]]

Offering mono-therapy instead of combined therapy is not the standard of care, and would be unethical in my opinion.

45

[[More & more, docs & patients alike are questioning the wisdom of starting drugs as soon as an infection is documented.]]

ps – Yes, I included a link to the latest article about this and discussed it in an earlier post.

46

Rob: [[I’m talking more about a purely epidemiological study. The only ethical consideration is pre and post test counseling, disclosing a positive or negative result (if that’s what the person wants) and referral to medical programs.]]

You are not describing a "purely epidemiological study. To get an unbiased “section of the population,” it would have to be a blinded study, without people having the ability to decline testing. As soon as we know persons are infected, it is unethical not to follow a protocol of therapy that is the latest standard. I still don’t understand, at all, what you’re trying to learn from this study. If it’s to learn how different groups vary in how they progress with the disease, you could find those already in therapy and simply compare their outcomes. My guess is that it would be more tied to socio-economics than anything, and I’m not sure what important information this would give you.

[[If you wanted to use a general population cross section, you could also give people the option of not getting tested at all, yet following them over time. This would create a picture of the general population, including people who ask not to be tested.]]

This doesn’t make any sense at all to me.

[[It would be a merely observational study, and not tied to testing a drug. Those do require putting patients on treatments tht work better and discontinuing ones that don’t.]]

What are you suggesting? I still can’t figure it out. I think you might want to go to www.cdc.gov and do a search. I believe you’ll find that there are a multitude of studies that have been are are being conducted all over the world to get at the questions you have. There is simply not a big mystery about who is becoming infected with HIV and how. If you are concerned about “biases,” then there are others I’d point the finger at way before I’d be concerned about epidemiologists.

47

Chill, Jill.

Studies that focus on people who are in treatment or based on a pre-defined risk group, may be biased. Not in a hatefully political way (no finger pointing here - jeesh) but biased in a scientific way.

There is no mystery how transmission occurs, that is not the point, never was, never will be in my mind. It is about the subtle statistical biases that can creep into any study if a rigorous double blind scenario is not followed. Using sick people ready to take therapy, or only pre-defined risk groups (urban out gay men and IVDU’s) is hardly an accurate picture. While making up most of the infected in the US, they are not all. Closeted gay men, non-IVDU, non-gay-identified men, etc. are not routinely followed in any way. Until they enter treatment, and are then counted as being in a risk group (or often innacurately outside of it). But in either case their lack of prophylactic treatment, early intervention etc. makes another HIV+ statistic appear on the “radar” not long before another death from HIV does. For example, I know of an African American closeted gay male (a minister), who did not get tested, finally got sick (years from HIV to AIDS totally UNKNOWN in this case - exactly my point), died 3 months later. His previous “demographic” (pos and healthy), never entered the national statistics until shortly before his death also did, same year.

To create an extreme continuum, with reality lying somewhere in between, if ALL HIV people were untracked until severe illness, the statistical figure for time from HIV+ infection to death by AIDS would be considered incredibly (innacurately) short. If on the other end, an entire cohort, representing a cross section of the population (still hard to do) was followed from ) rate of infection (let’s say we start with a cohort of 16 year olds, where the infection rate (while not zero is low for the general pop.) and follow them, tracking infection rates and progression to AIDS we would have a more accurate picture for a cross section of the population over race and gender for that age cohort.

In the MACS study on Hep B the study itself was not responsible for administering treatment. People were educated and referred to treatment. It was neither withheld, nor supplied.

Thanks for the reference - check out:
http://www.cdc.gov/hiv/pubs/hivser92.pdf

This highlights a national study - using various methods. (Unfortunately I cannot cut and paste from Adobe PDF documents - please do read for yourself, there are several interesting issues raised)

Sources:
Blood sampling from mothers who are giving birth, Federal job core applicants, runaway shelters, accident victims, etc.

All attempted to cancel out “skewing factors” - I will quote here - “Previous studies in many settings suggest that those who know or suspect they may be infected with HIV or at risk for infection may be less likely to participate in HIV studies, possibly causing the sero-prevalence to decrease”. The CDC sought to overcome that SCIENTIFIC BIAS in their studies. Apparently they agree with my anaylsis of the problem.

I found the study I was looking for - I’m glad they are doing it - it’s a good thing. Good but not perfect. Read on. (You don’t have to believe me, just check the reference)

Interestingly enough, the samples that were sent to the CDC were anonymous - to cancel the bias against people who are afraid to be tested. I’m not sure if they were asked for a general waver for “a national blood survey” etc.

People were not given treatment.

People were not told their status.

Testing was done on “discarded blood”.
Some demographic information was unavailable to the testing agency, although the general profile of the clinic from which it came was.

One arm of the study used emergency room and general hospital visit discarded blood for their sample - hmmm, that wouldn’t bias a higher seropositivity profile from IVDU’s and people with OIs related to AIDS would it?

In the case of some arms of the study, independent (“unlinked”) HIV testing and counseling was offered. Just like the MACS study - unlike double blind drug studies.

There is a ton of material in this study admitting (like any good one does) sources of potential bias. They unlike Jill do not call addessing a possibility of scientific bias “finger pointing”, I’ll repeat “jeeesh!”. Issues of reticense to test in certain populations normally in studies, untracked behavior, and how the relative age of a cohort, even when accrurately assayed, can confuse rates of infection within a given time period, and rates of total infection over time (“attrition”).

This study conscientiously address the scientific hurdles to an accurate study as outlined in my original posts - and then some. The number crunching required to randomize the data must be Herculean, but I commend them for their work. I do not confuse the results with a totally accurate picture of real HIV infection rates and time to progression to AIDS, for the general US population. In fact, more acurate than previous studies, they themselves admit possible biases.

In reply to another post -
The tuskegee experiment involuntarily withheld test results and treatment - highly unethical.

48

probably you meant the Tuskegee experiment voluntarily witheld test results and treatment.

In fact they deliberately misled the participants about the test results and treatment options, falsely claiming to be treating them when in fact all they were doing was cataloging their slow death.

A dark chapter in in american medicine.

49

Tuskegee - Involuntary to the participants.

Unlike the AIDS study cited above, at those locations where HIV tests were given (adolescent health centers, etc.) testing was offerred with conseling and referrals.

In the sites noted above which were at hospitals and the like the use of “discarded” blood was all involuntary, anonymous, and did not attempt to inform patients of their status, although I am sure that some were tested in the course of medical treatment and possibly informed under the rules governing that.

I wonder if some of the fine print on admission is a legal release of things like lost blood, removed organs etc.?

50

[[In the sites noted above which were at hospitals and the like the use of “discarded” blood was all involuntary, anonymous, and did not attempt to inform patients of their status, although I am sure that some were tested in the course of medical treatment and possibly informed under the rules governing that.
I wonder if some of the fine print on admission is a legal release of things like lost blood, removed organs etc.?]]

No. In blinded sero-prevalence studies (which - again - are being done in many populations, including some “general population” sites such as emergency rooms in areas of high HIV prevalence), persons are not identified, so their rights are not being violated. There is no reason to ask for permission, and the bias caused by voluntary testing is removed. If you only collect information on the results of voluntary testing programs, as Rob said, you only learn about HIV prevalence in certain populations who do not delay or avoid testing, and that doesn’t give us the whole picture.

51

I see I misunderstood Rob’s point. He was questioning whether hospitals ask patients to sign releases for discarded blood or organs. I don’t believe they do, but my last post wasn’t really addressing that. Sorry.

52

Rob, while all of the study groups you mention do avoid either positive selection bias(I’m worried so I’ll this study and maybe free drugs for while), or negative avoidance bias(I’m worried I won’t get a job if there’s a record anywhere of my being HIV+ so I won’t get tested), none of them provide cross-sections of American society.

Also, while they give us some information about prevalence, they give no hint towards answering the some of the questions you raised earlier about whether widespread drug therapy & resistance is impacting the length of infection before immunocompromise occurs, because the studies do not identify individuals, or track them longitudinally.

You are right in suggesting that current longitudinal studies may underestimate (or overestimate - bias can work both ways) the denominator for calculating complication & death rates. BUT the sero-prevalence studies you cite provide NO information about numerators.

And Jill, the various study arms I suggested were off the top of my head and meant to be examples; this was not a formal proposal for a study. However, I’m not sure that the long term effects of initial therapy with one drug which is not part of the standard cocktail are well enough established to make it unethical to formally study this & prove or disprove that concept.

Sue from El Paso

Does being married to another poster make me part of a clique?

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

53

Nother coupla points…

It’s pretty simple: People who test early for HIV, access treatment and take it properly don’t progress to AIDS, or do so very slowly. Those who are infected but delay or avoid testing and/or don’t get treatment (combination therapies) progress more quickly to AIDS. There is abundant evidence that this is so.

Blinded seroprevalence studies tell us what the prevalence of cases is in a population, but they don’t tell us when people got infected. You might be able to get at incidence (new infections) if you repeated studies over time in large groups, and some do that, but again, it’s very expensive.

Stepping away from sero-surveillance and just looking at surveillance (cases that are reported to the Health Department), in states where AIDS cases are counted but HIV is not yet a reportable condition, you can get a really slanted view of the epidemic. As an example, you might see a trend in which AIDS cases among white men who have sex with men are dropping sharply and cases among black females are rising quickly. This does NOT necessarily mean that white gay males are not still being infected or are even being infected at a lower rate than black females (it could be true, too, but these numbers alone don’t necessarily show that). It may simply mean that the white males are testing earlier and getting treatment and the black females aren’t.

Some people will look at such “trends over time” and misinterpret them to mean that prevention programs are no longer needed in the gay community. This is one of many reasons that more and more states are adopting HIV reporting instead of just counting AIDS cases, which are outdated and almost meaningless data.

A majority of “new AIDS cases” that are reported to my office are people who never tested or got treated for their infection until they came into a hospital with a possibly preventable acute illness, giving them an HIV and AIDS diagnosis at the same time.

54

[[However, I’m not sure that the long term effects of initial therapy with one drug which is not part of the standard cocktail are well enough established to make it unethical to formally study this & prove or
disprove that concept.]]

According to the HIV doctors I work with, mono-therapy (treating an HIV patient with only one drug, be it an antiretroviral, a protease inhibitor, or one of the others) is generally considered to be substandard care. The exceptions would be possibly pregnant women (though I believe some are now getting combination therapy) and those who have developed resistance or contra-indications with all but one drug (pretty rare). I would say this - no one would ever get a study approved in which one group received combination therapy and the other, mono-therapy.

Sue, if you disagree with me, give Abraham Verghese a call and ask him.

55

Jill, I absolutely agree that multi drug regimens are the standard of care right now. BUT, increasingly, whether it is optimal to start patients on these regimens as soon as HIV is diagnosed is coming into question because:

  1. These regimens are extremely expensive, to the point of being totally cost-prohibitive for most patients in subSaharan Africa, where the highest disease burden exists.
  2. Side effects are common, and can significantly impact quality of life.
  3. Widespread use increases the likelihood of newly infected persons harboring resistant strains so that even if taking a multi-drug regimen, they are getting de facto monotherapy.

While to the best of my knowledge, it is still in the “I wonder if” stage, people are considering deferred multi-drug therapy. The simplest possible study would compare outcomes between persons started immediately on standard cocktails vs persons started on standard cocktails only after developing persistantly low CD4 counts. But if it is considered ethical to defer cocktail therapy at all, I think it would also be ethical to include an arm with montherapy initially, followed by standards cocktails when/if CD4 counts fall IF the monotherapy agent is not part of the standard cocktail(s), or so closely related that resistance to it would confer resistance to cocktail agents.

This arm might provide vitally important results for patients in places where standard cocktail regimens from the time infection is diagnosed are completely out-of-reach.

Sue from El Paso

Does being married to another poster make me part of a clique?

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

56

Sue, you didn’t read my earlier post in which I also cited the recent JAMA article about changing recommendations for initiating therapy in newly diagnosed patients. I also included a link to this article.

Here’s another one, “Rationale for Drug Combinations in Anti-HIV Drug Therapy” http://hivinsite.ucsf.edu/medical/case_studies/2098.256e.html

This article describes how HIV becomes resistant to even very aggressive drugs if they are used alone, as monotherapy. If several drugs are used together, there is improved suppression of replication and also more mutations will have to accumulate in an HIV strain before it becomes resistant.

Experts now believe that there must be at least three drugs in a patient’s
anti-HIV treatment regimen, and some recommend many more (unless there are tolerance problems). A quote from this article, “Until 1995, a common recommendation for anti-HIV therapy was monotherapy (therapy with a single anti-HIV drug, e.g. with zidovudine or didanosine alone). However, because of the rate of replication of HIV and the observation that resistance emerges rapidly with currently available drugs used alone, monotherapy is no longer recommended.”

With recommendations like these, based on many, many studies, I say it would be unethical and pointless to put any group on monotherapy.

Tons of studies have been done about all of this.
Jill

57

I think MajorMD addressed and summed up my previous points well.

I did not think the study cited was ideal. It sems the ideal study may well be impossible for logistical and legal reasons. The study cited was better than those that use already infected people who are progressing and trying to access drugs etc…

The “ideal” study which would most accurately estimate the average time to progression accross all demographics, would be some sort of huge study that did not select people in any way EXCEPT to create a huge cross-section of America (or the planet - let’s think globally). Cohorts would have to be started when nearly all members were neg. 16 year olds for example. This does mean that, the generation of long term data would take a long time. Not logistically possible. And not very fruitful.

My argument, Jill is not for such a study. I meant to draw attention to the fact that anything less is more or less accurate, depending on the scientific bias. Depending on the method of data gathering. All studies should be viewed with an educated, critical eye. This in now way disparages the scientists and clinicians involved.

The first scientific projections of time from exposure to death by HIV were given in MONTHS, and this time frame was projected from the data available at the time, which as I pointed out above, was mainly from people who appeared as pos in the demographics, shortly before death. Or, were parts of the demographics who were urged to test - so called risk groups. Yet many who were infected then, are still alive today, and modeling the disease has become more complex in response.

Jill stated in more than a few posts that she did not understand what I was saying, yet spent much effort in replying, she also took issue and called it “finger pointing” at the epidemiologists on my part(?).

She then provided a source for a report that proved my point about scientific bias. (Thanks)

Now she has taken to arguing her point via personal email - respectfully and without personal attack I might add in all truth (I do appreciate that Jill).

I would ask that you put aside whatever professional buttons my sincere scientific inquiry posed, read the study cited, re-read my and MajorMd’s posts, and not address counter arguments to statements I have not made, i.e., the means of infection.

The point I am making is subtle, it is not conclusive. Like a hypothesis it could be used to model more accurate studies (as it has in the one cited).

58

Don’t know if anybody is actually reading this thread, but wanted to chime in here on an early comment.

egkelly:

Nobody dies of them because of vaccination programs. Remember all those shots you got as a kid?

Also, I’ve had a doctor comment to me she doesn’t think HIV can be transmitted through regular, healthy heterosexual contact. She claims the transmissions are due to undiagnosed STD’s. Of course her experience is limited to poor women in a free service health clinics in Mexico and Texas, so she’s got an observation bias working against her.

59

The main point I’m trying to make is that with the great therapies we have now, most people AREN’T progressing to AIDS. This is a good thing. And those “unknown” cases who aren’t being tested and aren’t being treated? They show up sick pretty quickly. So no “scientific bias” is going to miss them. It’s hard to miss pneumocystis pneumonia. We’re not seeing lots of surprises here… we aren’t seeing many cases outside the known transmission categories. When we do, we actively investigate them (such as occupational exposure or blood transfusion cases). This was the point I made to Rob in email - that I’d be happy to give him more details about how this is done. Also, the other point I’ve made repeatedly and I guess I should quit now, is that the studies Rob is proposing are already being done, in every single population imaginable, though maybe not in the numbers he’d like to see. I wish he’d peruse the literature and see what I mean.
Jill

60

There really isn’t a limit you can place on the amount of time that a person lives with HIV. HIV is a retrovirus and it’s common with many retroviruses to lie dormant in the person’s body until something causes them to become active. So you could be HIV+, have the virus in your system, and it could take you years or decades to die or you could never die from that because it never became active.

When are you going to realize being normal isn’t necessarily a good thing?