26 Apr
2006
26 Apr
'06
11:19 a.m.
Newer is not always better: re-evaluating the benefits of newer
pharmaceuticals
Whether newer pharmaceuticals justify their higher cost by saving other
types of health spending, so-called ?offset effects?, is an important
health policy question. We replicated the analysis of Lichtenberg
(1996), which suggested the effect of newer drugs substantially
outweighed the additional cost. We find Licthtenberg? results to be
highly dependent on both the specific model chosen and the dataset used:
substituting either a model that is less sensitive to expenditure
outliers or the newest data release results in the effect disappearing;
substituting both causes it to reverse in direction. In addition, we
estimate the effects of newer treatments separately for the 10 most
prevalent conditions using propensity score matching. We find that the
presence or absence of offset effects depends critically on the
condition in question; suggesting some classes of newer drugs may result
in lower costs, while others may not.
26 Apr
26 Apr
12:02 p.m.
This is a nice abstract, nuanced and precise. Once you have a firm
result, you might well include it, to lend precision to your finding--that
is, not just "reverse in direction" but "we demonstrate that new
pharmaceuticals cost XXX dollars more than their older counterparts, all
else equal." Also, if you are writing for a general audience, you might
use just a few words more to explain what exactly an "offset effect"
is--decreased hospitalizations, decreased doctors visits, what exactly?
Best,
Dan
On Wed, 26 Apr 2006, [windows-1252] Karen Gr?pin wrote:
Newer is not always better: re-evaluating the benefits
of newer
pharmaceuticals
Whether newer pharmaceuticals justify their higher cost by saving other
types of health spending, so-called ?offset effects?, is an important
health policy question. We replicated the analysis of Lichtenberg
(1996), which suggested the effect of newer drugs substantially
outweighed the additional cost. We find Licthtenberg? results to be
highly dependent on both the specific model chosen and the dataset used:
substituting either a model that is less sensitive to expenditure
outliers or the newest data release results in the effect disappearing;
substituting both causes it to reverse in direction. In addition, we
estimate the effects of newer treatments separately for the 10 most
prevalent conditions using propensity score matching. We find that the
presence or absence of offset effects depends critically on the
condition in question; suggesting some classes of newer drugs may result
in lower costs, while others may not.
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
12:27 p.m.
Good abstract and I agree with Dan's suggestions. One minor point - and
this goes generally - any time you refer to the author, refer to the paper instead.
So always Lichtenberg (1996) instead of Licthenberg personally.
Best,
Ian
On Wed, 26 Apr 2006, Dan Hopkins wrote:
This is a nice abstract, nuanced and precise. Once
you have a firm
result, you might well include it, to lend precision to your finding--that
is, not just "reverse in direction" but "we demonstrate that new
pharmaceuticals cost XXX dollars more than their older counterparts, all
else equal." Also, if you are writing for a general audience, you might
use just a few words more to explain what exactly an "offset effect"
is--decreased hospitalizations, decreased doctors visits, what exactly?
Best,
Dan
On Wed, 26 Apr 2006, [windows-1252] Karen Gr?pin wrote:
Newer is not always better: re-evaluating the
benefits of newer
pharmaceuticals
Whether newer pharmaceuticals justify their higher cost by saving other
types of health spending, so-called ?offset effects?, is an important
health policy question. We replicated the analysis of Lichtenberg
(1996), which suggested the effect of newer drugs substantially
outweighed the additional cost. We find Licthtenberg? results to be
highly dependent on both the specific model chosen and the dataset used:
substituting either a model that is less sensitive to expenditure
outliers or the newest data release results in the effect disappearing;
substituting both causes it to reverse in direction. In addition, we
estimate the effects of newer treatments separately for the 10 most
prevalent conditions using propensity score matching. We find that the
presence or absence of offset effects depends critically on the
condition in question; suggesting some classes of newer drugs may result
in lower costs, while others may not.
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
6:59 p.m.
New subject: [gov2001-l] Simulated Quantities Off?
Hey List,
I'm noticing that when I simulate quantities of interest for the ordered
probit, they differ from the zelig ones -slightly for the first and last
categories -quite a bit for categories 2 and 3 of the response variable.
Is this pattern related to the model or due to my incompetence?
Thanks,
Martin
7 p.m.
New subject: [gov2001-l] Simulated Quantities Off?
We think this might be a Zelig bug--check your predicted probability of
falling into response categories 2 and 3. If any of the predicted
probabilities are negative, that's an indication you hit the same bug we
did. Best, Dan
----
Ph.D. Student
Department of Government
Harvard University
Tutor, Currier House
dhopkins(a)fas.harvard.edu
http://www.danhopkins.org
On Wed, 26 Apr 2006, Martin Kanz wrote:
Hey List,
I'm noticing that when I simulate quantities of interest for the ordered
probit, they differ from the zelig ones -slightly for the first and last
categories -quite a bit for categories 2 and 3 of the response variable.
Is this pattern related to the model or due to my incompetence?
Thanks,
Martin
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
7:18 p.m.
New subject: [gov2001-l] Simulated Quantities Off?
Hi Martin,
We've noticed the same - there may be a Zelig issue here which we are
investigating. For now, I wouldn't worry too much about the discrepancies
with Zelig.
Best,
Ian
On Wed, 26 Apr 2006, Martin Kanz wrote:
Hey List,
I'm noticing that when I simulate quantities of interest for the ordered
probit, they differ from the zelig ones -slightly for the first and last
categories -quite a bit for categories 2 and 3 of the response variable.
Is this pattern related to the model or due to my incompetence?
Thanks,
Martin
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
7:25 p.m.
New subject: [gov2001-l] PS7 digital copy
Could someone email me a copy of PS7? my.harvard has been down for the
last hour or two, and came home to work but only brought my own code and
not the problem set because I figured I could download it again here...
Thanks!
-Dan
7:07 p.m.
I agree. I was also looking for a more _substantive_ conclusion at the
end of the abstract. the last third or so is heavily methodologically
oriented but by that point it would be good for you to be drawing
conclusions. nice job.
Gary
On Wed, 26 Apr 2006, Dan Hopkins wrote:
This is a nice abstract, nuanced and precise. Once
you have a firm
result, you might well include it, to lend precision to your finding--that
is, not just "reverse in direction" but "we demonstrate that new
pharmaceuticals cost XXX dollars more than their older counterparts, all
else equal." Also, if you are writing for a general audience, you might
use just a few words more to explain what exactly an "offset effect"
is--decreased hospitalizations, decreased doctors visits, what exactly?
Best,
Dan
On Wed, 26 Apr 2006, [windows-1252] Karen Gr?pin wrote:
Newer is not always better: re-evaluating the
benefits of newer
pharmaceuticals
Whether newer pharmaceuticals justify their higher cost by saving other
types of health spending, so-called ?offset effects?, is an important
health policy question. We replicated the analysis of Lichtenberg
(1996), which suggested the effect of newer drugs substantially
outweighed the additional cost. We find Licthtenberg? results to be
highly dependent on both the specific model chosen and the dataset used:
substituting either a model that is less sensitive to expenditure
outliers or the newest data release results in the effect disappearing;
substituting both causes it to reverse in direction. In addition, we
estimate the effects of newer treatments separately for the 10 most
prevalent conditions using propensity score matching. We find that the
presence or absence of offset effects depends critically on the
condition in question; suggesting some classes of newer drugs may result
in lower costs, while others may not.
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
6595
days inactive
6595
days old
gov2001@lists.gking.harvard.edu
7 comments
6 participants
participants (6)
-
dhopkins＠fas.harvard.edu -
dschrage＠fas.harvard.edu -
grepin＠fas.harvard.edu -
kanz＠fas.harvard.edu -
king＠harvard.edu -
yohai＠fas.harvard.edu