Bob Beall, CEO of CF Foundation

[just1more]

Ok, putting on my math teacher hat:

absolute mean = good old average = ie, if you add everyone's improvement/decreases and divide by the # of participants. This came out at 10.6%. NOTE: This can and is tweaked by odd results, such as someone(s) with a decrease in FEV1, or maybe someone with a great improvement.

relative mean = calculus to try and discount the impact of results the farther they are from the average. IE if the absolute mean was 100; then in relative mean results between 80-120 would have more impact on the reported # than a result of 38. The math is pretty nasty; though I will explain via PM if anyone really has a burning desire to know.

So ultimately, if you take everyone they saw at 10.6% increase. If you only counted the 'normal' patients results then it was closer to 17%.

 

[just1more]

Ok, putting on my math teacher hat:
<br />
<br />absolute mean = good old average = ie, if you add everyone's improvement/decreases and divide by the # of participants. This came out at 10.6%. NOTE: This can and is tweaked by odd results, such as someone(s) with a decrease in FEV1, or maybe someone with a great improvement.
<br />
<br />relative mean = calculus to try and discount the impact of results the farther they are from the average. IE if the absolute mean was 100; then in relative mean results between 80-120 would have more impact on the reported # than a result of 38. The math is pretty nasty; though I will explain via PM if anyone really has a burning desire to know.
<br />
<br />So ultimately, if you take everyone they saw at 10.6% increase. If you only counted the 'normal' patients results then it was closer to 17%.

 

[hmw]

Thanks Tom.

Hmmm. I like that and i don't like that. It sounds like for many people, then, results could be quite good, even better than the 'average' most widely published.

On the other hand, for those two numbers to differ by so much, there is clearly a lot of variability in response to this drug. In looking at a drug like this one, all responses are important to know about- an improvement of 2% or 20%. I think they all 'count.' We obviously need bigger populations of people on the drug, though, to get true perspective. When you have so few people taking it, it's hard to know what the true expectations will be.

 

[hmw]

Thanks Tom.

Hmmm. I like that and i don't like that. It sounds like for many people, then, results could be quite good, even better than the 'average' most widely published.

On the other hand, for those two numbers to differ by so much, there is clearly a lot of variability in response to this drug. In looking at a drug like this one, all responses are important to know about- an improvement of 2% or 20%. I think they all 'count.' We obviously need bigger populations of people on the drug, though, to get true perspective. When you have so few people taking it, it's hard to know what the true expectations will be.

 

[hmw]

Thanks Tom.
<br />
<br />Hmmm. I like that and i don't like that. It sounds like for many people, then, results could be quite good, even better than the 'average' most widely published.
<br />
<br />On the other hand, for those two numbers to differ by so much, there is clearly a lot of variability in response to this drug. In looking at a drug like this one, all responses are important to know about- an improvement of 2% or 20%. I think they all 'count.' We obviously need bigger populations of people on the drug, though, to get true perspective. When you have so few people taking it, it's hard to know what the true expectations will be.