AIDS Education..Or Condom Promotion?(Viacom)

MRC ^ | February 14, 2003 | by L. Brent Bozell III

[RAT Patrol]

Before you go.....if you have a better understanding of person years will you post it before you leave? Thanks.

[eastsider]

Before you go.....if you have a better understanding of person years will you post it before you leave? Thanks.

The "person years" is obscure to me as well, but I'll look at it again over the weekend. I also find it as difficult to believe as you do that the incidence of HIV transmission without a condom is less than 7%.

Have a good one, RP.

[fight_truth_decay]

Because pyears may have several time variables, it is necessary that all of them be in the same units. For instance in the call py <- pyears(futime ~ rx + ratetable(age=age, sex=sex, year=entry.dt)) with a ratetable whose natural unit is days, it is important that futime, age and entry.dt all be in days. Given the wide range of possible inputs, it is difficult for the routine to do sanity checks of this aspect. A special function tcut is needed to specify time-dependent cutpoints. For instance, assume that age is in years, and that the desired final arrays have as one of their margins the age groups 0-2, 2-10, 10-25, and 25+. A subject who enters the study at age 4 and remains under observation for 10 years will contribute follow-up time to both the 2-10 and 10-25 subsets. If cut(age, c(0,2,10,25,100)) were used in the formula, the subject would be classifed according to his starting age only. The tcut function has the same arguments as cut, but produces a different output object which allows the pyears function to correctly track the subject. The results of pyears() are normally used as input to further calculations. The print routine, therefore, is designed to give only a summary of the table. (your link)

[goodnesswins]

You got that right (about condoms breakage, etc.) - I have a grandson to prove it....

[eastsider]

"Never" using condoms is ~93% effective, "always" using condoms is ~99% effective, therefore "always" is 85% MORE effective than "never."

The major and minor premises of the above-stated syllogism are both false.

Major Premise: Never using condoms is never effective, as can be demonstrated by a simple syllogism:

Every effect has a cause;
Inaction is not a cause;
Therefore, inaction has no effect.

To say that "never" using condoms is 93% effective is like saying that condoms are 93% effective when not used, which is absurd. As the poster in #6 above correctly pointed out, condoms are 0% effective when not used.

Minor Premise: The effectiveness of a prophylactic is calculated by (a) subtracting the total number of unfavorable events that occur in the group using the prophylactic from the total number of unfavorable events that occur in the control group (i.e., the group not using the prophylactic, and (b) dividing the difference by the total number of unfavorable events that occur in the control group. For example, if we wanted to know how effective seat belts were in preventing traffic fatalities and we knew that 7 of 100 motorists who never wore seatbelts were killed in traffic accidents but only 1 of 100 motorists who always wore seatbelts were killed, we would subtract 1 from 7 and divide the difference by 7, giving us an effective rate of .85, or 85%.

The data in our example do not support the onclusion that seatbelts are 99% effective. To determine the effectiveness, we have to compare the results of the test group to the results of the control group. Comparing the actual number of traffic deaths in the test group (1) with the total number of motorists in the test group (100) doesn't tell us how effective seatbelts are, only the chances of surviving a traffic accident while wearing a seatbelt -- that is, 99 out of 100 times, or 99%.

The data support Bozell's conclusion that condoms fail to prevent HIV transmission 15% of the time . Condoms are 85% effective against HIV transmission, not 99%.

[RAT Patrol]

Nice job, eastsider. I'm impressed!

[eastsider]

Hey, RAT Patrol! I got called in to work to cover during the blizzard we're having, and have some down time to FReep. Did you catch fight_truth_decay's explanation of person years in #45?

[RAT Patrol]

Yes. That was just a cut and paste of the link Madg provided. They need to translate it into english for me.

[RAT Patrol]

I take it to mean that, basically, one person in a hundred will get the disease in the course of a year if they always use condoms. I am unclear if every subject was always exposed to the virus or what the average exposure frequency would be. But, the longer the years of exposure, the higher the risk.

[eastsider]

What I find most stunning about the NIH study is that there were there were actually HIV-negative people who knowingingly had unprotected sex with HIV-positive partners. Can we say "death wish?"

[RAT Patrol]

I know.

Maybe these were couples who found themselves in the situation of one partner being infected and one not. Yet, that would mean either someone had a blood transfusion, is an IV drug user, or has not been monogamous. There would be study problems with some of those?

I haven't read the study, just its conclusions. I didn't want to download the whole thing. Maybe I will later if I have time.

[eastsider]

You're kinda missing the point.

Your point was that Bozell is distorting the test data, and he is not.

Less than one chance in a hundred over a hundred years. That is NOT "85%."

One chance in a hundred hundred over a hundred years v. 7 chances in a hundred over a hundred years is 7 minus 1, which is 6, divided by seven, which yields an 85% effectiveness against transmission.

[RAT Patrol]

less than one chance in a hundred over an extended period of time.

No, Madg, I think that's PER YEAR. It is not possible that they mean to stretch the risk ratio out over 100 years. Most people don't even live that long. It is just a way of calculating your risk over time. I think this study says you have approx. a one percent chance PER YEAR of getting the disease. If your years of risk increases, then your total risk increases.

I'll get back to you when I have time to find the facts on time versus risk factor.

[eastsider]

Whether "person hours" means a year, a century or a millennium, person hours will alter the effectiveness ratio only to the extent that it alters the number of HIV transmissions in either group. Presumably, the greater the number of person hours in the study, the more reliable the data.

[eastsider]

Sorry about that; "person hours" should read "person years."

[eastsider]

Correct and consistent condom use is a HECK of a lot better than 85%.

The NIH data don't support such a statement, so your beef is with the accuracy of the NIH data, not with Bozell's interpretation. Assuming the accuracy of the NIH data, Bozell correctly states that condoms failed to prevent HIV transmission 15% of the time. There's nothing even remotely deceptive about that.

A 1% transmission rate in the "always" group tells us that the risk of transmitting HIV in that group is 1%. A 7% transmission rate in the "never" group tells us that the risk of transmitting HIV in that group is 7%. Neither group by itself can tell us anything about the effectiveness of condoms. One can only assess the effectiveness of a condom by comparing the number of transmissions in the "always" group with the number of transmissions in the "never" group. Assuming the accuracy of the NIH data, condoms were only effective 85% of the time.