I need MATH help! (Probability and Statistics) for AIDS discussion

none | 10/15/03 | self

[r9etb]

Thanks, I figured that out about 50 posts back. :^)

Well, perhaps you can assuage my guilt by changing your name to "Mr. Dead Horse...." ;-)

[walden]

Thank you so much, everyone! These responses have been an enormous help.

[Maelstrom]

Condom failure is much higher than 3%.

The 3% figure only accounts for the probability of a condom to fail due to manufactured faults in condoms.

[Doctor Stochastic]

That's OK. It gave us the chance to expostulate on the difference between cumulative probability and single event probablity. (Maybe the lurkers also got something out of it.)

[NathanR]

Using my numbers, if you have random sex once a day for a year, the probablity that you would not get AIDS in that time period is .997**365 or about .334. (using the calculator in Windows it is .997^365)

In real life, the numbers would be totally differant, and other *unknown* factors would apply.

[general_re]

Always neglecting taxes and present value, of course.

Well, if we're going to talk about the odds of winning in imaginary lotteries, we might as well exempt ourselves from imaginary taxes and imaginary inflation and so forth too ;)

[general_re]

You can win a little money with games like this, and a lot of money if you can figure out how to apply the game outside of a drinking establishment.

As always, the only guaranteed way to win at the lottery is to be the one running the game. Unfortunately, my state, like all states, prefers to maintain its monopoly on lotteries, and so when I try organizing a lottery, they call it "running the numbers", and they come visit me with the subpoenas and the warrants and the handcuffs and so forth ;)

[r9etb]

Well, if you win a $2 Billion Lottery, I suppose that, before you actually collect, you could probably offer million-dollar contributions to a working majority of the US Congress to exempt yourself from both (contingent on their passing it, of course) and you'd still come out ahead....

[wideminded]

So when are you planning to receive an injection of pure HIV in order to prove your point?

By the way, how do you account for the many people who have contacted AIDS through tainted blood and blood-factor transfusions?

[Old Professer]

An effort made for the happiness of others lifts us above ourselves.

[Lonesome in Massachussets]

Jeeze, I hain't bought the paper today, you spoilt it fer me.

I find the Word Jumble embrassing, since my wife, who rarely does them (only when looking over my shoulder), comes up the answers before me, while I do it everyday.

[Lonesome in Massachussets]

Lydia M. Child

[NukeMan]

Sure do! Behold the power of Google...

Drs. Norman Hearst and Stephen Hulley, writing in the Journal of the American Medical Association (JAMA, April 22, 1988), have attempted to define the statisti- cal chances of becoming HIV positive after sex with various partners. Your risk of getting HIV, they say, depends primarily upon your partner’s probability of being infected—and also whether you use a condom. Their JAMA article confirms that your chance of contracting AIDS from a single act of intercourse can vary enormously. Their conclusion: Choosing a “low-risk partner” is the most important strategy. If your partner is HIV positive, they estimate your chance of picking up the AIDS virus from one-time intercourse at 1 in 500. Receptive anal intercourse and the presence of genital ulcers and venereal warts will increase your risk tenfold, or more. If your partner is known to be HIV negative, the risk of contracting AIDS from such an HIV-negative partner (using a condom) is estimated at 1 in 5 billion. There’s still a slight risk since people can be infectious before their test turns posi- tive—and condoms can fail.

[Stone Mountain]

Odds of condom failure 3%

Remember in your calculations, that this 3% failure rate does not mean that at any given time, a condom has a 3% chance of failure. Condom failure rates are usually calculated over the course of a year. For instance, a couple, using only condoms as their means of birth control over the course of a year, would have a 3% chance of getting pregnant. The failure rate for one-time condom use is significantly lower. This is at least the way that condom failure rates are calculated with regard to getting pregnant, but I suspect they calculate it similarly with regard to disease transmission.

[samtheman]

I seriously doubt condoms fail at a rate anywhere near 3%, except, as I stated earlier, for those failures classed as "operator error". I've used condoms for many years. Never had a single failure. Not one. In hundreds of times. Judging by pregancy prevention only, since I'm monogomous and don't have sex with anyone but my wife. We only have children when we want to have children. Otherwise, we use condoms and they work. 100% of the time in my case.

[Old Professer]

You know her First name?

[Spaulding]

Good Question, but here too the answer seems fairly obvious. Ask what it is that causes people to need transfusions? Transplant patients, hemophilliacs, commonly receive immune suppressants. Since these are aggregate data, is it any suprise that their immune systems are weakened. Immune suppressants define the disease, and may also open the host to the minor retrovirus we call the HIV virus.

[IronJack]

If heads is an infection then tossing the coin twice gives you a 75% probability of getting an infection.

If turning up heads in any combination is an infection, you're right. However, if turning up both heads or a given coin as heads constitutes the infection, then your odds are still only 1 in 2.

The more accurate way to view this would be (assuming sex with an infected person always results in transmission of the infection): How many persons are in my target group? (Not every person on earth is likely to have sex with you, so you can reduce the target population to a reasonable level). Out of that population, how many have the disease?

That will tell you your odds of contracting the disease. If you are homosexual, you increase your odds two ways: one, your target population is smaller. And two, the proportion of that population infected with the disease is much higher than the general population.

[usapatriot28]

Resulting in a 95.6% chance the initially uninfected person is infected at the end of the 520 events.

But that is making two huge assumptions.
1) That that 520 people you have sex with are chosen completly at random (or at least randomly distributed odds of being infected). If you have sex 520 times with the same person, who is uninfected, your chances are 0%.
2) That condom failure with an infected person always leads to infection. This is absolutly not true!! Just because you have sex (even unprotected) with an infected person, does not guarntee that you will get infected.

[DB]

If you have a 1 in 500,000,000 chance of winning the lottery then buying all 500,000,000 possible numbers guarantees you will win. There is no guarantee on how much you’ll win though… Multiple winners are the unknown wild cards…