Reply
Wed 4 Apr, 2012 08:54 am
If a disease is undiagnosed, how can you know whether it is prevalent or not?
Context:
The inverse association was restricted to current and former smokers and was stronger after adjustment for smoking. Among smokers, theinverse association persisted even after finely stratifying on smoking status,time since quitting smoking, and number of cigarettes smoked per day.Sensitivity analyses did not support the possibility that the inverse association was due to prevalent undiagnosed disease.
@oristarA,
oristarA wrote:The inverse association was restricted to current and former smokers and was stronger after adjustment for smoking. Among smokers, theinverse association persisted even after finely stratifying on smoking status,time since quitting smoking, and number of cigarettes smoked per day.Sensitivity analyses did not support the possibility that the inverse association was due to prevalent undiagnosed disease.
What inverse association? Are we supposed to guess what this is about?
@oristarA,
prevalent in this case means "widespread". An example of a prevalent undiagnosed disease state is high blood pressure. Lots of folks have it but don't know it.
@oristarA,
Quote:If a disease is undiagnosed, how can you know whether it is prevalent or not?
That's true, but the researchers are looking at a hypothetical case (the supposed existence of a prevalent disease) and concluding that the results they showed were
not attributable to that.
@oristarA,
oristarA wrote:
If a disease is undiagnosed, how can you know whether it is prevalent or not?
Personally, I think this is a darn good question.
JPB is probably right, but it also kind of sounds like the author wasn't paying attention to what he was saying.
@roger,
You didn't read my post, then.
Every condition has a prevalence -- the rate it occurs in the population. In research lingo, a prevalent condition is one that is widespread - has a high prevalence. If something is known to be fairly prevalent it's because it's detected in a large number of people who are tested/treated for it. That observation is then projected onto the population at large (people who haven't been tested for it) and a rate of undiagnosed prevalent disease is calculated. McTag is right --- it's a hypothetical number that is derived from observation of a subset of the population.
There's insufficient info in the OP to know what disease we're talking about here, but the authors are stating that there's no evidence that there is a high undiagnosed rate in the general population that might explain their results. It gives their conclusions more weight because they're saying they've accounted for the possibility that their observation is due to "noise" amongst the subjects.
@JPB,
Thank you JPB.
And thanks to all others.
I think it is not necessary to upload the whole article to make the matter certain.
Because I think the importance of the article is not significant.