PTC.S2.Q15 - Each of two drugs, S and T...

CEOofgettinga170

Can someone break down the stimulus and translate it for me? I feel like I don't understand what the stimulus is trying to say.

Why does the society has a policy decision to make? why pay $240,000? does that mean the society is paying for the experiment? why is it 120 * $2000? Also why is it only saving one more patient, if the drug goes onto the market wouldn't that benefit more people since its 1 more person per 120? And lastly why is the NA here is there is no other way of telling if S effective as T? Isn't the stimulus already established that trial shows that T would prevent 1 more death out of 120 than S would? So isn't that the effectiveness of the drug?

I'm confusion.

Lime Green Dot

Hi! I know this comes much later than your post - but I was just going through PTC myself again and looking for discussion on this Q, too.

I'll offer this to anyone who may still be searching for some perspective:

My understanding of the CONC is that it's saying we've got a choice to make -- either choose S and lose a life or choose T (coming from the "whether or not..." part at the end) and save a life... but pay big bucks for it. That's on the basis of the PREM of T saving "1 additional life out of 120" but costing much more than S.

I took the assumption to be that it's an all-or-nothing game. In other words, that we can't choose, on a case-by-case basis, which drug could be reliably administered to save a life.

But why does it have to be that way? What stops us from making an assessment about S's efficaciousness at the critical moment when we have to decide which of the 2 to give to the person who had the heart attack? And if we determine S won't work in that "1 in 120 cases," well then, we can opt for T and pay that +$2,000 for just that 1 case. Not every single case!

(D) poses this challenge to the ARG in more discreet wording; it asks, what if there is a somewhat less expensive way (i.e., less than dishing out the whole $240k for each set of 120 P needing a treatment), that's not too difficult to do and can be done pretty fast, to show us which of the 2 drugs, S or T, we can use in the moment? (D)'s like: HOLD. THE. BREAKS. We don't have to get rid of S entirely if we know which 1 case out of 120 we can't use it for.

If we use the negation test on (D), it becomes clearer why it must be that for the ARG to hold, there isn't this more sensible way available to us: What if "there is a quick, practical, and relatively [relative to $240k] inexpensive way of telling for any individual case whether S will work as well as T"? Well then, the silliness of the ARG's CONC becomes apparent!

And lastly why is the NA here is there is no other way of telling if S effective as T? Isn't the stimulus already established that trial shows that T would prevent 1 more death out of 120 than S would? So isn't that the effectiveness of the drug?

I think this confusion comes from missing the part in (D) that says "for any individual case"; just b/c it's more effective in 1 case out of 120 doesn't mean it's more effective for each case of the 120. Maybe that 1 additional case is a person who ate McDonald's morning, noon, and night for 50 years, and T would be powerful enough to just barely help that dude survive the night. But maybe the average person with fewer risk factors could take either S or T and be saved. Just one way to look at it. After all, if a drug prevents death, a drug prevents death, and that makes it sufficient to be "effective" in that particular case.

Why does the society has a policy decision to make?

B/c that's the conclusion that's drawn, and we have to roll with it. I think I understand where you're coming from -- like, why not the family members or the doctors or those witnessing the person having the heart attack? Remember though that we're very, very rarely trying to falsify the CONC. We are almost always asked to see how the PREM doesn't quite flow to the CONC and act, according to the Q-stem.

An additional point, I think by "society" the ARG just means the collective "we" -- try subbing that in and seeing if that helps a bit. But yeah, another "NA" could be that maybe it's not society that needs to make these cost decisions at all, but the families and doctors of the HA patient, for example, that ought to decide. On the face of things, though, that's not the weakest point in the ARG; so it's probably also not where we'd want to initially focus on as we scan A/C to fill in the NA gap.

why pay $240,000? does that mean the society is paying for the experiment? why is it 120 * $2000? Also why is it only saving one more patient, if the drug goes onto the market wouldn't that benefit more people since its 1 more person per 120?

I also questioned these parts and wondered why it had to be as presented. My best guess is that the conclusion is kind of giving us this generalizable example by showing us the minimum overall cost consequences in a hypothetical world where we could only choose treatment T or S 100% of the time.

In the "real world," there are obviously more than 120 people who might suffer a potentially fatal HA -- in that case, the ARG might have us infer that for however many cases there actually are, this all-or-nothing model holds and would multiply accordingly.

I think all of these uncertainties made this question pretty hard. My saving grace here was POE especially for the stickler of (C).