Disease Detectives B/C Question Marathon

[Flavorflav]

Um, no. I'm afraid my hint has confused the issue badly. Polio affects virtually no Americans under the age of sixty, and quite a few over that age. Does this indicate that polio risk is age-related?

[Crazy Puny Man]

Um, no. I'm afraid my hint has confused the issue badly. Polio affects virtually no Americans under the age of sixty, and quite a few over that age. Does this indicate that polio risk is age-related?

...

Sorry, I'm confused. I thought polio doesn't really exist in the U.S. anymore and that most of the few cases of polio would be in infants/young children

So...I don't know...

[Flavorflav]

Because it doesn't exist anymore, the only people who are affected by it are old. There are no new infections, but people paralyzed before the vaccine are still paralyzed. This has gone on so long that I am going to punt. My point was that this is cross-sectional data, representing prevalence data for different age groups at the same time. The big danger in cross-sectional studies is confusing cohort effects (effects of the time and place of birth) from the effects of aging. If one could show that historical rates of hearing loss were higher for some reason, then the higher prevalence among older people might not indicate an age-related increase in risk, but rather a secular decrease.

Your turn, I suppose.

[Crazy Puny Man]

Because it doesn't exist anymore, the only people who are affected by it are old. There are no new infections, but people paralyzed before the vaccine are still paralyzed. This has gone on so long that I am going to punt. My point was that this is cross-sectional data, representing prevalence data for different age groups at the same time. The big danger in cross-sectional studies is confusing cohort effects (effects of the time and place of birth) from the effects of aging. If one could show that historical rates of hearing loss were higher for some reason, then the higher prevalence among older people might not indicate an age-related increase in risk, but rather a secular decrease.

Your turn, I suppose.

So the reason older people have a higher prevalence of polio is due to the cohort effects (they were born a long time ago - time of birth)...

Oh. Now that makes sense

Next question...I suppose I'll be keeping this kinda simple: of Hill's criteria for causation, which of the criteria is the most important to confirm in order to establish a causal relationship between the risk factor under study and the effect?

[thedarknight]

Hey, would it be strength? Because the association between the probable cause and effect should be strong to continue one's investigation.

[Crazy Puny Man]

Hey, would it be strength? Because the association between the probable cause and effect should be strong to continue one's investigation.

Um...not quite what I'm looking for, no...

[Ploxytomatoes2016]

Specificity

[nomynameisnotkevin]

Hey, would it be strength? Because the association between the probable cause and effect should be strong to continue one's investigation.

Um...not quite what I'm looking for, no...

I got points for this one @ Palatine, so I'll say what I put down. Definition - if the definition of what you're looking for (disease, infection, illness, impairment whatever) is not consistent throughout or is too broad/specific, all of your data is useless.

[Crazy Puny Man]

No, and no. nomynameisnotkevin, I'm pretty sure that how you define a case is not one of Hill's criteria for causation...

Can you tell me what this question was?

As for specificity: not quite...I'm pretty sure there are certain risk factors that can produce multiple effects; for example, smoking leads to a variety of health problems (a weak example, but I think you know what I'm trying to say). Establishing a relationship like that would certainly help, but it's not the most important criterion

And to whoever suggested strength: a stronger association is indeed more likely to be causal than a weak association; however, strength is only a measure of how strong the correlation is; it doesn't necessarily indicate a causal relationship, and the criterion I have in mind is more important to establish than strength

Brain snack: a weak association does not mean that the relationship is not causal; for example, weak associations have been observed between a person's diet and the risk of coronary heart disease, but diet is still thought to be a major causative factor

[nomynameisnotkevin]

oops. I really need to read the question better. That was a diff. question. It was asking what was most important for a case-control study.

Here's my new answer
[hide]Is it consistency? So you can achieve the same results with diff samples, populations, @ diff. times?[/hide]