How likely are you to have the disease?
|
| The following is an excerpt from "Tricks of the Mind" by Derren Brown.
"Imagine there is a terrible disease reported, and although it affects only one in ten thousand people, it is absolutely lethal. You are worried about it, so you decide to undergo a medical test to see if you have the disease. Now, no medical test is ever 100 percent accurate, but your doctor explains that this is known to be 99 percent accurate, regardless of whether or not you have the disease (in other words, it will deliver a correct positive or negative result 99 percent of the time). You decide to take the test. You're a little nervous , but you think it's a sensible thing to do. A blood sample is taken, and you're told the results will be sent to you in the post. A week later the envelope arrives from the testing centre. You open it up and read the contents. Staring you in the face is the answer you dreaded: the results are positive. The test has indicated that you have the lethal disease. You are devastated.
And you are right to be, aren't you?
Just for a moment, review the scenario above and ask yourself what may seem like a very easy question: how likely are you to have the disease?" |
| Votes | Answer |
|---|
| 9 | B - I'm 99% likely to have the disease | | 8 | E - None of the above | | 5 | D - I'm 0.9% likely to have the disease | | 2 | A - I'm 99.9% likely to have the disease | | 1 | C - I'm 9% likely to have the disease |
|
| User | Comment |
|---|
bill   | | posted 22-May-2008 7:16am |
Well, without giving this much thought, I would think 99%. But, clearly, I'm suppose to give it thought, so...
0.01% (1 in 10,000) of the time someone will have the disease. This seems like the tricky bit (you even wrote it without using numerals which seems sneaky and thus makes me assume it's very important).
99% of the time the test gives an accurate result.
Perhaps, it's better to say it as 1% of the time the test gives an inaccurate result.
So... that means the test more often gives a false-positive than a false-negative (because, the chance of someone having the disease is actually very small). So, perhaps, I don't have that much to worry about. Now, lets see if I can come up with my actual chances...
Out of 10,000 people, 1 will have the disease. If each of those 10,000 people takes the test, 1% will get an inaccurate result. Thus, 100 of them will get an inaccurate result. Since only 1 should have the disease, that inaccurate result would usually be a false-positive. So, I'm one of the 100, but only 1 of the 100 (there's even a 1% chance it's not one of the 100) has it. So, I'd say I'm close to a 1% chance of actually having the disease.
I'm picking 0.9% because, that's closest to my estimate and probably factors in the chance of it not even being in my 100 false-positives. | they    | | posted 22-May-2008 8:34am |
I believe I'm 9% likely to have it.
| Melf  | | posted 22-May-2008 8:50am |
I want to say 99%, but it's probably not. | | MagicalJamie | | posted 22-May-2008 9:10am |
Well I think it's a trick question.
If I had the lethal disease, wouldn't I die within the week that it took the results to get to me? Looks like I don't have it after all and the test was wrong =) | | icurok | | posted 22-May-2008 9:46am |
Bill got the answer right. It's an example of anchoring, whereby you instinctively rely on one piece of information (in this case, the 99% accuracy) when making a decision and find it hard to stray from your own instincts (even if they're massively wrong).
Let's say that a million people take the test.
Statistically, there will be 100 people with the disease and 999,900 people without it.
Statistically, 1 out of the 100 people with the disease will be told they don't have it, whereas 9,999 of the people who don't have it will be told that they do.
That means that out of the original million, 10,098 people will be told that they have the disease, but only 99 of them will. The exact chance of you having the disease is 0.980392156862745% (which I rounded down to 0.9%).
| | Zang | | posted 22-May-2008 10:09am |
Beats me. | | Enheduanna | | posted 22-May-2008 10:42am |
I imagine the statistics of the accuracy of the test aren't relevant; it's actually the odds that I'm really the one person in 10,000 who has it. Although .9% doesn't seem small enough for that. So maybe it's a combination of that and likelihood that the test is wrong. At this point I am way over my head in terms of math and logic skills, though, so the answer is that I have no idea. | | JessicaWoman99 | | posted 22-May-2008 2:10pm |
E none of the above because I am a healthy person and this is unlikely for me | | aquawolfy | | posted 22-May-2008 3:54pm |
99.9 % likely. | | RGirl | | posted 22-May-2008 5:37pm |
None of the above. I get tested a second time. | | Enigma | | posted 22-May-2008 11:50pm |
I'm 0.9% likely to have the disease | | LJD | | posted 23-May-2008 11:15am |
Fear, stress, negative thoughts can break down the body. The odds are against it, unless you have a hereditary propensity. | | moviesnob | | posted 23-May-2008 4:12pm |
MAH BRAIN!!! IT HURTZ!!! | | LindaH | | posted 23-May-2008 8:41pm |
Not any more likely than anyone else in the general population. | cloudhugger  | | posted 28-May-2008 11:17am |
None of the above.
Because....the chance that the disease was present in the first place is slimmer than before the test. Once the test was taken, the chances were higher. Seeing the poitive results made the chance 99% higher because the person due example is convinced, devastated...the emotional brain got involved and fudgeed it up for the rest of the body. Having a disease is different than seeing indications that there is something out of balance in the body and looking for healthier alternatives to health. Owning a disease is sad. Ignoring facts isn't good either. The mind will play tricks on the body if the two are not connected. | | Biggles | | posted 30-May-2008 9:18pm |
Before I was screened, I had a 0.01% chance of having the disease. Having received a positive result in a test where the sensitivity and specificity are both implied to be 99%, I would know that the positive predictive value of that result is 99% (so 99% of patients receiving that test result will actually have the disease). I've been trying to see where the trick is in this, but I'm clearly missing something unless it's somehow that individual risk remains the same after screening (i.e. 0.01%) as it was before, but that just seems silly. I can fiddle the numbers based on the information given so that the sensitivity is 100% and the specificity is 98%, but that still suggests I have a 98% chance of having the disease. The only other possble trick would be in the wording - that "accuracy" here is not quite what I am presuming...or that a "positive" result actually means "negative" for the disease.
Interested to read the answer... | | Biggles | | posted 30-May-2008 9:32pm |
This is why I shouldn't do maths at 2:30am after three days of exams... Using actual numbers to get my positive predictive value would have helped, rather than using percentages...I knew I was doing something silly because I'm well aware that the positive predictive value is low in rare diseases (which is why we don't bother screening for them). | | kcthedog |
 E - None of the above
Cute, but unconvincing.
Fun survey!
|
If you'd like to vote and/or comment on this survey, please Sign On.
Link this survey: http://surveycentral.org/survey/29273.html
Hits: 1 today (346 in the last 30 days)
Google Search:
|
