Re: Medical Research
- From: hrubin@xxxxxxxxxxxxxxxxxxxx (Herman Rubin)
- Date: 21 May 2006 18:09:08 -0400
In article <AWKbg.4910$y4.4755@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
george conklin <george@xxxxxxx> wrote:
"Herman Rubin" <hrubin@xxxxxxxxxxxxxxxxxxxx> wrote in message
news:e4n9jb$3bce@xxxxxxxxxxxxxxxxxxxxxxx
In article <3hqbg.4145$x4.3431@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
george conklin <george@xxxxxxx> wrote:
[Many cites back and forth deleted.]
With basic research, we are literally exploring in the dark;
if we were not, it would not be basic research. I suggest
that you look up the origin of the word, "serendipity".
At some point Herman many of the long-term diseases will be
understood
but not right now, for sure. So treatments seem marginal, like 3% over
10
years being counted as a 'victory.'
Some of the marginal treatments seem to be doing at least
a fair job. Type 2 diabetes is one of them, and a fair
proportion of the population has it for decades and does
not do that badly. The same is true for many of the
thyroid diseases. And ONE treatment has enabled most
Type 1 diabetics to live most of a full normal lifespan.
But to ever get that understanding of these diseases, it
will be necessary to do basic research, much of which will
not be related at all to those diseases or to any others.
True but then who is going to support basic research these days?
Everything is related to marginal improvements on existing technologies.
Can you imagine a vaccine being declared a victory if it prevented 3% of
deaths in 10 years?
It could be declared as something giving insight to be
pursued further, but nothing more. However, if it
prevented most of the deaths of a class which could be
distinguished, it would be a victory for those.
If something prevented all deaths from Type 1 diabetes,
it would be a victory, and that is less than 3%.
We have to get the government out of controlling
universities. It will have to be done carefully and
gradually, or the whole system will go bust.
Officially, they are doing no such thing. But they control
so much of the funding of the universities that it amounts
to control. Before WWII, the American research universities
supported research from their funds, either legislative
appropriations or foundational money. Also, those funds
went into supporting education; tuition is now a MUCH larger
proportion of the educational budget than it was before
federal support. About 25% of Purdue's budget comes from
designated federal funding.
This touches only a few disciples Herman. In the past, Chemists, for
example, had about 5% of majors but wanted huge amounts of the academic
budget which was otherwise generated by low-cost English and other similar
programs. When sociologists wanted computers to crunch census data,
administrations complained that we also were there to subsidize 'hard'
sciences and how dare we need computers? What about a piece of chalk every
5 years? They were pretty serious about that too.
By the time computers were available which could do that,
the universities were already relying on government funds.
It is true that universities did give some money to new
hires in chemistry, hoping that they would get more back
in grant overhead in a few years. Getting computers in
mathematics is harder than in sociology, as the social
scientists can make the point they have data to crunch.
Getting access to computers in statistics is not that
hard, but apart from the canned programs almost useless
in research, this does not help get good use from them.
Initially, the government supported research because
they had found out that researchers could do practical
things better than practitioners. Until about 1970,
they poured more money into the universities than the
good researchers could use. The political upheaval
cut this back.
In the past, if a noted Professor X wanted to get a
graduate student, or some equipment, the university
which did not heed the request was in danger of losing
the scholar. Now, the answer is to try to get the
money from the government granting agency; the feds
essentially control it. And politics gets in; if
one state gets "too much" of this federal money,
because it has the good universities, it is less
likely to get a grant than a department in a state
university where the state does little research, but
has one third-rater who can do some.
Some what? If you don't get results, you don't get renewed.
If you have a history of getting results, you are still
likely to be judged on what you are proposing. In basic
research, you are in no position to say what you will find,
and this leads to being turned down.
So we need to set up the concept of a research
university, allow them to set up research funds which
cannot be used for anything else, and which can keep
up the level of their courses even if the students
hate it and have a high failure rate.
Well, that happened anyway. That is why we have to import our
scientists. The research universities turn out few majors these days. Most
science majors seem to come from 1. foreign nations, 2. smaller
undergraduate insitutions where they don't get chased out.
Even this is running into problems. They are also
starting to neglect the basics.
We will have
to provide tax breaks for those who fund this, and
keep the "government watchdogs" out of it. Proxmire
gave his "golden fleece" awards to many important
research projects, and he did it under Congressional
immunity. I was berated by the granting officer for
the title of one of my papers, "Occam's Razor Needs
New Blades," which of course had noting to do with
razor blades of any kind.
Maybe he thought, "Schick Razor Needs New Blades" would have been
etter? -:)
BTW, Proxmire never fell for things like that. He or
his assistants always checked on terminology. While
I was on sabbatical at Madison, there was a conference
sponsored by the government in which some of the talks
were on "permanent waves", and this got no feedback.
I could go on; many basic courses in mathematics and
statistics are essentially not given, or not taken by
many, because they are not directly applicable. Even
if not applicable at all, they often clarify concepts.
Basic science will solve illness. But is not the human genome project
basic science Herman? We are well beyond what one person can do alone after
he/she finishes the chemistry lecture and 18 hours in the classroom.
Statistical decision theory can probably make major
improvements in the treating of disease NOW, if even
just the doctors understood it. And multivariate
statistical methods, not exactly new, but often
needing an expert in both theory and computing to
carry out, can do a great deal of improving medical
research. Problems with many parameters are not being
well treated from any approach, and not enough basic
research is being done on this. Unlike the physical
sciences, and even there this is showing up, one
cannot adequately control enough factors to avoid
this, but medical researchers limit themselves to
this, and not even using the quantitative variables.
The genome project was speeded up when a researcher was
turned down by the government; they did not think
investing in his method was worth it. He left the
academic field and got private funding, and the vast
speedup is largely due to those contributions, and the
followups from them.
In fact, medicine is probably the field least dependent
on government funding. Big Pharma is plowing money back
into research, including basic research, and there are
other private foundations.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
hrubin@xxxxxxxxxxxxxxx Phone: (765)494-6054 FAX: (765)494-0558
.
- Follow-Ups:
- Re: Medical Research
- From: george conklin
- Re: Medical Research
- References:
- Breast Cancer Research
- From: george conklin
- Re: Medical Research
- From: george conklin
- Re: Medical Research
- From: Herman Rubin
- Re: Medical Research
- From: george conklin
- Breast Cancer Research
- Prev by Date: Re: Medical Research
- Next by Date: Re: Medical Research
- Previous by thread: Re: Medical Research
- Next by thread: Re: Medical Research
- Index(es):
Relevant Pages
|
