Thanks for your help - if you know a source of data that would be the "gold standard" for baseline acquisition risk under different scenarios, that would be really helpful . . . thanks again
Perhaps the "vaccine" trial they reference was a random population - i.e. not people aware of the status of their partner (a factor I have seen people say reduces transmission risk by 1/2) . . . that would make the numbers make more sense if it was the case, but that would seems like a very questionable starting point . . .
Is 1498 not a large enough population to draw conclusions about baseline acquisition rates? It is hard to imagine it would be so different . . . but perhaps some other study is the "gold standard" for establishing baseline acquisition risk?
Ok, that makes sense then as to why the risk reduction is the same . . . but the placebo / control group acquisition rate assumption / data is what makes the acquisition risk numbers look very different . . . not sure the best source for this number . . .
Both studies seem reliable to me - the risk reduction % is about the same. But the assumed/demonstrated base level / placebo acquisition rate is significantly different - what is the best source for this number?
It is just one number, but it is the number everything else is derived from and it actually makes a big difference . . .
I've seen this, and I'm fairly sure this is merely the write up and publication of the original study. Both this article and the GSK presentation to the FDA cite 1498 as the original number of enrollees in the study. I'd have to read it more carefully to find the discrepancies you note.
I think the 2004 study summary mentions a separate vaccine trial as a source of the acquisition rate that they mention, then they apply the "reduction in risk" that they establish for the drug in the study . . .
Even if the 2004 study is just to establish a "reduction of risk", wouldn't they want to show an actual acquisition rate also for different sub groups? Why reference a separate vaccine study to set the baseline acquisition rate?
The thing that confuses me is the 11/389 8-month acquisition rate in the placebo group (4.2% annual rate) in this study . . . the sex frequency is about the same int he two studies, so it isn't that . . . I'm not sure why this is so low, the 2004 study says it should be more like double that rate or a bit higher . . .
This is the NEJM page on the 2004 study:
http://www.nejm.org/doi/full/10.1056/NEJMoa035144#t=articleResults
No, the FDA did not fund this study. The results were presented to the FDA during the application for approval.
The GSK labs didn't administer the drug and do the swabs and blood tests. The research teams at the University of Washington, and probably at Westover Heights Clinic (Terri Warren's facility), did. Hard to get more trustworthy than that.
Also this is an FDA controlled study, right? The 2004 study was by the drug maker to prove efficacy and win extended approval for use for suppression therapy, correct?
Would a drug company ever structure a study to suggest a higher baseline acquisition risk than the FDA would? Perhaps if they did, it might help in marketing the drug?
It is easy to make numbers tell a story, perhaps the FDA study is more unbiased?
Shedding isn't consistent, so acquisition can't be consistent across different studies. We can be shedding all day, or an hour or two over a number of different days. There are about a dozen other factors to consider as well - how often the couples had sex, how often they forgot to take the pills, etc.
I suspect they arrived at the risk reduction figure using the acquisition figures in the same study each time. Do you have the link to the 2004 study offhand? I want to make sure it isn't actually a writeup of the same study with finalized figures.
I think the reduction in risk was particularly significant to report, to show the difference between using suppression and not using suppression. It was quite a breakthrough for the community when it was first officially reported.
Thank you, but isn't the acquisition rate the important one? The "reduction in risk" is only important if you know what the risk is (i.e. the acquisition rate) . . . it is the acquisition rate that seems to differ between the two studies - the "risk reduction" may be consistent, but from different baseline acquisition rates . . . but is that just due to the structure of the study?
Why would the rate of acquisition for the placebo groups be different?
I agree - the stats are helpful as information to give to a new partner at the time of disclosure.
Yes, I agree disclosure is required in all cases - but the statistics are very helpful to have on hand, provided they can be accurately understood as they relate to different subgroups . . .
It seems like a "small risk" to say something has a 2% chance over the course of a year . . . but 2% is significant, it is double 1%, quadruple the risk of 0.5% . . . If there was a 2% chance to win a $10 lottery worth $100MM, there would be 200,000 winners to split the pot . . . obviously, the odds in a typical lotto of that size are miniscule compared to 2% . . .
I'm not aware that they're inconsistent. Remember that what has been reported as a statistic is the *reduction in risk,* not the acquisition rate. Better marketing for the drug - but also a useful statistic to encourage antiviral use for prophylaxis. Also, as I mentioned, official numbers may have been adjusted to take actual demographics into consideration.
FYI, the official statistic on condom use is 30% protection (that's different than rate of acquisition of course). I don't know how that study dovetails with this one. I doubt antivirals were a factor in the condom study.
Thank you very much for posting the link to that study. Slide 27 is an interesting one.
The outcome in the study for Valacyclovir and "sometimes" or "nearly always" condom use was 0 transmissions out of 302 couples over 8 months (pretty good). The "never" condom use Valacyclovir group was 4 transmissions out of 401 (or about 1%, but more like 1.5% on an annual basis - also pretty good).
The placebo group with "never" condom use was 11 out of 389 (4.2% on annual basis), "sometimes" was 2/102 (2.9% annual rate), "nearly always" was 3/212 (2.1% annual rate) - this is supportive of condoms reducing risk by more than 50% if used "nearly always" . . . but the transmission numbers across the board are lower than what was shown in the 2004 study - is there a reason why?
This study has a little bit of a small sample size issue on a sub-population level, but 0/302 is interesting because it "should have been" a number like 2 to 3 or more out of 302 based on the findings of the 2004 study. And the 4/401 with Valacyclovir and no condoms used over 8 months is also surprisingly low.
Was there a difference in the way the studies were conducted? The overall risk of transmission in all groups of this study seems to be less than in the 2004 study . . . . the populations size overall is large enough, and the reduced rate similar across subgroups, so I am not sure why this would be the case . . .
Thank you again for posting that - the difference in implied transmission rates between this study and the 2004 study would appear to be significant . . . do you know of any analysis of the differences between the two studies that has been published?
Thanks again.
I should express the caveat that this raw data may not have been recalculated for accurate demographics. Also, regardless of statistics, it's still important to disclose one's status to a non-H partner before sexual contact. They have the right to choose to take the risk, however miniscule.
Again I ask, to what end? You can quote the ranges and so on to your partners but in the end, they either take the risk or they don't. Your actions will minimize the risk of transmission.
The reduction in the shedding rate is less important than what your shedding rate is in the first instance, which you will not know. You may be very high or low, even a non shedder. Your partners will also have different propensities to being infected.
Thank you - I am trying to figure out what applies to my situation. There is the average in the study, but then there are personal factors. The study average included people who forgot to take meds every day, didn't always use condoms, etc. (very few people in the study reported 100% compliance) . . . I understand that it was enough to prove efficacy of the drug and the benefit of condom use . . . I just want to know what impact compliance had on the numbers - I assume the least compliant subjects transferred the virus much more readily than the most compliant - so if you are 100% careful and compliant, the risk SHOULD be less than the study average . . . but how much less I can't estimate without the numbers . . .
I understand the data, and that the risk isn't ever going to zero - I just also understand the limitations of the average reported in the summary results of the 2004 study, and I would like to review additional data - or hear an experts take on what ranges of risk might represent a reasonable confidence interval assuming 100% compliance . . . not sure if this is the right place to ask this
Try not to believe every random website out there that claims to be an authority on herpes. The studies published in the peer-reviewed journals and archived at PubMed are the ones you want.
The official Valtrex study (which is similar to the studies on the other two antivirals) put BOTH the transmission risk reduction AND the shedding rate reduction at at least 50%. The 75% reduction is for acquisition with clinical (visible) symptoms. It's as low as 50% for subclinical infection.
Stats don't mean anything, however. At any moment a non-infected person could be a one of the 50% who does acquire the virus. Disclose, use barriers, abstain during symptoms, to help further the reduction of both transmission rate and shedding.
I have not seen better information than an average reduction of 50% from the study you refer to. Individuals though will vary, some will experience up to a 90% reduction.
To what purpose are you thinking of applying these percentages?
In real life the answer is either infected or not infected. All that can be done as an individual is discuss the issues and what minimizes transmission risk and see what your partner is comfortable with.
http://www.fda.gov/ohrms/dockets/ac/03/slides/3950S2_04_FDA-Valtrex.ppt