Evidence-Based Funding Dr. Michael Lauer Keynote 2017 Spring NIH Regional Seminar

Evidence-Based Funding Dr. Michael Lauer Keynote  2017 Spring NIH Regional Seminar


Good morning, everybody. So, what I’d like to do is talk to you a little
bit about the landscape of biomedical research and NIH research in particular. And I want to start by telling you a story, something
that happened a few years ago which is that one of my colleagues showed me an article
that appeared in PNAS. It was written by two people from Penn State
University who lamented that the way universities advertise their research programs is by how
many grants or how much money they are receiving. So, they might say that we are a 20 million
dollar program or we are a 50 million dollar program or a 100 million dollar program. And they thought that this would be kind of
like an airline saying, “We burn more gasoline than any other airline to move passengers,”
or, “We spend more money than any other airline to move passengers.” So, this obviously raises the question as
to how do we know whether or not our research enterprise is performing well? So, in this article, they present this simplistic
view of the world. On the left is the grant application. It gets reviewed. Hopefully, your grant gets funded. Then, after the grant gets funded, you do
your research, you publish some papers, and then those papers may get cited. They say the problem is that right now, the
current finish line is where the grant gets awarded, and they suggest that perhaps an
optimal finish line might be later on down the line after the work has been done, the
papers have been published and cited. And maybe that’s not an optimal finish line,
but it does raise the question as to whether or not there ought to be a different finish
line. So, what goes into this? So, if we think about it, what goes into this
are people and money. So, what I’d like to do is walk through some
of the details on people and money first. Okay, so this is a plot that shows the trajectory
for NIH funding over time. The X axis is years. So, on the left is 1980, and on the right
is 2015. And the Y axis is the total amount of money. And this is inflation adjusted. So, back in 1980, NIH funding was about 10
billion dollars. If we go to 2000, it gets up to 20 billion
dollars. And then, it reaches a peak around 2003 of
about 30 billion dollars. So, we see that during the 1980s and the 1990s,
there was a dramatic increase in NIH spending. Then, in 2003, that suddenly stops. And from 2003 on, there has been a gradual,
steady decline in NIH spending with some bonds that are going up and down. One can see here that if one was working during
the 1980s and 1990s, one might imagine a world in which NIH spending was constantly going
up, that there was a period of continuous growth. But we see then that this came to a sudden
end. By the way, those of you who want to play
with this a little bit more, this comes from the AAAS dashboard. So, if you Google in “AAAS dashboard,” you
can go on it, you can play with it, and make all kinds of interesting pictures and get
a better sense as to what’s going on. So, this sudden stop in growth of NIH spending
has created a difficult situation. This was described in an article by four very
well-known thought leaders led by Bruce Alberts who previously was the editor-in-chief of
Science. And they say that this never-ending, erroneous
assumption of continuous growth has led to a state of hyper competition where we have
many, many more people applying for grants than we have grants available. And this state of hyper competition is something
that if it is not addressed, is a recipe for long-term decline. So, let’s think about this a little bit more. Judith Kimble, who’s at the University of
Wisconsin, organized a set of workshops where she brought together a group of people, and
she asked them about what they thought were the major problems that were besetting biomedical
research. And they said that there are two core problems. One is that there are too many scientists
vying after too few dollars. And the second is that there are too many
postdocs vying after too few faculty positions. And, essentially, everything else, like, for
example, the stresses on peer review are symptoms of these two core problems. So, if we ask the question about there being
too many scientists going after too few dollars, this raises the question as to whether or
not we’re funding the right number of investigators. John Lorsch, who is the director of the National
Institute of General Medical Sciences, NIGMS, wrote a very thoughtful paper a few years
ago in which he said that perhaps instead of focusing on the amount of dollars or the
amount of grants that we’re giving out, we should look at a metric which may seem, on
first glance, to be trivial, and that is the number of investigators that we actually support, then
harkens back to what we saw initially from the PNAS paper which suggested that we shouldn’t
just be looking at grants, but maybe we should be looking at something else. So, in this case, it’d be the number of investigators
whom we actually support. So, here are some data on the number of investigators
we actually support. This plot, on the X axis, shows fiscal year
going from 2003 to 2015. 2003, you may remember, that’s the top of
the cliff. That’s when the NIH doubling ended. And the Y axis shows number in thousands. What we see is that back in 2003, NIH was
supporting about 25 and a half thousand unique principal investigators. In other words, there were about 25 and a
half thousand unique people who could say that they were a principal investigator on
at least one NIH research project grant. We call those RPGs, but I’m avoiding acronyms. Now, if you look over time, you’ll see that
that number has not changed by very much. So, back in 2015, that number was a little
bit less than 28,000. From 2011 to 2015, the number really doesn’t
change at all. Now, of course, we ask the question, how many
people wanted to be funded by NIH at any given time? And that is showing a very different picture. So, that’s the blue line here. And what we do is we look over a period of
multiple years, how many people wanted to be funded in any given year. So, back in 2003, there were about 60,000
scientists who wanted to be a PI on at least one NIH research project grant. And by 2015, that number had increased to
nearly 90,000. This is the face of hyper competition. This is exactly what the folks at the University
of Wisconsin and the people in the PNAS article were talking about: a hyper competitive state
in which we have more and more scientists vying after a limited number of dollars. Okay. Now, let’s take a bit of a deeper look. What we’ve looked at so far is just simply
the number of people. So, we’ve looked at the number of scientists
whom we’re supporting, and we’ve looked at the number of scientists who want to be supported. Now, we’ll look at the distribution of scientists
according to a common demographic which is age. So, this is a plot that my colleagues in OER
made and was shown by my predecessor, Sally Rockey. On the X axis, we see age, and on the Y axis
is number. So, this is a histogram or a distribution
plot. The blue bars refer to PIs, principle investigators
on R01 grants. And what one sees here is that there is a
peak around age 40 or a little bit less. And the red line shows the distribution of
age across faculty members. And this is based on data from the Association
of American Medical Colleges, the AAMC. Now, this is 1980. You’ll see that these distributions look remarkably
similar. Here’s 1990. Here’s 1995. Here’s 2000. All right, that’s 2013. Okay, here’s what I want to do is I want
to go through these quickly. So, here’s 1990, ’95, 2013. Let’s take a look at this again. 1990, ’95, 2013. Now, what you’ll notice, in addition to the
fact that both distributions have shifted to the right, is that there is an increasing
gap between the distribution of PIs on NIH R01s and faculty at AAMC medical centers. Okay, so so far, what we’ve done is we’ve
looked at people. We talked about people and money. We looked at money on a macro level. We saw that NIH support increased dramatically
in the ’80s, ’90s, and early 2000s, and since then has been stagnant. We looked at people also on a macro level. We looked at the number of investigators being
supported. We saw that the number of investigators has
remained constant but the number of people who want to be supported has gone up dramatically. And then, we also looked at age distribution,
and we saw that there has been a marked shift in age distribution for faculty and for NIH-funded
researchers, but the shift for NIH-funded researchers has, if anything, been a bit more
dramatic. Now, we’re going to go back to money, but
instead of looking at money from a macro level, from the entire system, we’re now going to
look at money on a micro level. We’re gonna go back to this article that appeared
in PNAS, and the authors of that article said that in light of the hyper competitive system
that we’re in, we should pay particular attention to how much money is going to individual laboratories. Because they say although different research
activities incur different costs, at some point, returns per dollar diminish. So, we’re going to spend a fair amount of
time thinking about this sentence, going through this sentence and seeing what the implications
are. So, here’s a look at money across individual
investigators. What we do here is we take each investigator,
we see how much money that they’re getting, and then we rank them in order. So, we put, let’s say at the very bottom, or
the very top, we might put the person who’s getting the most amount of funding. And then, we take the person who has the next
most amount of funding and the next most amount of funding, and we go all the way down to
the person who’s getting the least amount of funding. Then, we split them up into ten equal stacks,
and we add up the amount of money that each of those stacks is getting. And this gives us a sense as to how money
is distributed across individual laboratories. So, if we look all the way on the right, there
we see these are the people here on the right. These are the people who are getting the least
amount of money. That’s decile one. And then, decile two are the people who are
getting the next most amount of money, and decile three are the people who are getting
the next most amount of money. And all the way on the left, these are the
folks who are getting the most amount of money. And what this plot shows is that 10% of principal
investigators are getting 40% of the money. So, there’s a skewed distribution in how money’s
being distributed. Now, of course, it begs a question which is
what’s the right distribution? I don’t think anybody would say that we should
be giving the exact same amount of money to every individual investigator. That doesn’t make much sense. And then, on the opposite extreme, nobody
would advocate that we should give all the money to one investigator and none of the
money to anybody else. And so it’s a subject of ongoing discussion. But in any case, what we do see here is that
there is a skewed distribution of funding. All right, now let’s look a little bit deeper
and see who’s getting that funding. This was an interesting paper that my colleagues
in OER and one of my colleagues from NHLBI wrote, and this was published a few months
ago in Plos One. And what they were interested in was looking
at the demographics of people who are receiving funding by NIH. Now, we’ve already looked at that little set
of slides showing that the age distribution has shifted to the right. One of the questions that has come up is whether
or not people at different stages of their career are feeling a different kind of pressure
given the hyper competitive environment. And so let me show you what this looks like. So, the X axis is, again, time. This goes from 1990 through 2015. And the Y axis shows the percent of all of
our funded investigators who fit within certain stages of their career. So, let’s start with the red line. The red line are those people who are early
in their career. And we define this using age as a proxy. So, we say that if age is 45 or less, that
somebody is relatively early in their career. So, back in 1990, a little bit more than half
of the investigators that we funded were in relatively early stages of their career. And that number steadily goes down. And then, around 2006, 2007, you see that
it stabilizes. Now, in 2007, 2008, what happened is that
NIH instituted a policy whereby we essentially gave a handicap to early stage investigators,
making it easier for them to get their applications funded. And so effectively, this appears to have arrested
the decline, the relative decline in early career investigators that NIH was funding. Okay, now, the blue line represents people
who are later in their career. In this case, we define that as being age
60 or higher. And what one sees is that beginning in the
mid-1990s, that number steadily rises. So, 1990 was maybe 5% of all the investigators
whom we supported. And by 2015, that number was more like 20%. So, that number has been steadily increasing. Now, what (Mark Scherett? @ 15:07) and his colleagues found particularly
interesting were the folks in the green lines. The green line are what we might call mid-career
people. Mid-career people would be people between
the ages of 45 and 60. So, they were doing reasonably well in the
1990s and then the early 2000s. They represented a growing proportion of the
NIH pie. But then, starting around 2005, 2006, they
actually we actually see a reversal in the trend, and we see a decrease, a relative decrease
in the number of mid-career investigators who are being funded. So, I think one of the key messages that we’re
seeing here is that within this environment of increasing hyper competition that we’re
seeing, people who are in their early and mid-career are having an increasingly difficult
time. Now, why might this be happening? So, this is an interesting paper that just
appeared about a month ago. It was written by Bruce Weinberg, who is an
economist at Ohio State University. He did this work funded off of a grant that
came from the National Institute on Aging. And what he did was he looked at the distribution
of age in the entire workforce and in the scientific workforce. So, let’s take a look at the dotted line here. So, this shows age on the X axis. So, going from 25 to 75. And the Y axis shows the share of the overall
workforce that’s made up by scientists and engineers. The proportion of the overall workforce made
up by scientists and engineers. So, if you look at the dotted line, that’s
the distribution back in 1993. And then, the circles, the open circles, that
represents the distribution in 2010. And so what one sees is is that there is a
marked over representation in 2010 compared to 1993 among scientists who are older. And they say that the reason for this, so
what did they find? They say that their major findings is that
the scientific workforce is aging rapidly, that it is aging more rapidly than the general
workforce as a whole. Now, of course, the whole workforce aging
because of Baby Boomers who are aging, decreasing birth rates, decreasing death rates. But nonetheless, the scientific workforce
is aging at an even greater rate than the general workforce. And they say that one of the primary reasons
for this may be that fewer older scientists are choosing to retire given the end of mandatory
retirement in the 1990s. And they suggest that this trend will continue
for a while, and why this may be a problem is that it may effectively crowd out scientists
who are in the early or middle part of their career. Now, why is this a concern? Well, you know, in part, this may be obvious. If you’re a new investigator, and you’re trying
to break into the system, if it is increasingly difficult to break into the system, this is
a concern. But it’s also a concern for the entire enterprise. This is written by Dr. Rockey. She says, “New investigators, they are the
innovation of the future. They pioneer new areas of investigation. Entry of new investigators into the ranks
of independent NIH-funded researchers is essential to the health of our biomedical research system.” So, we have an interest as a society to make
sure that the next generation of researchers will be able to get into the system and be
able to thrive. And there is growing recognition that people
who are in the early and middle stages of their career are having an increasingly difficult
time. So, last October, Nature ran a series of articles
about younger scientists, and this is taken from one of those articles. In the United States, funding successes for
all age brackets has decreased dramatically since the 1980s. So, researchers are spending more and more
time seeking funds. For how many of you does that resonate? That researchers are spending more and more
of their time worrying and figuring out ways to seek out funds. And they then go on and say that burden falls
most heavily on new faculty members, and that results in their being more conservative than
being ambitious. Okay, so so far now, what we’ve done is we’ve
looked at money. We looked at it at a macro level. We looked at it at a micro level. We saw on a macro level that there has been
a decrease, relative decrease in funding beginning in 2003 after a long period of growth. We saw on a micro level that there’s a skewed
distribution of funding across investigators. We looked at people. We saw that the number of investigators that
we are funding has remained constant, that the number of investigators who want to be
funded has been dramatically increasing, and that there have been shifts so that people
who are in the early middle parts of their career perhaps are feeling a greater degree
of stress within this hyper competitive environment. So, now, what I want to do is I want to switch
gears and talk about what happens then when the research is done. And how do we know whether or not the research
is being done well? So, this goes back to that initial schematic
of the way the world works. So, the research is funded, and then the work
gets done, is published, and there are citations that come out of it. And actually, this begs a much bigger question
which is how do we know whether or not we’re doing a good job? So, for example, we might know that a particular
baseball team is doing a good job because their won/loss record is very high. How do we know whether or not science is working
well? And how do we know whether or not a scientific
program is working well? And, unfortunately, we don’t have something
as simple as a won/loss record to be able to determine that. And therefore, a lot of people have been thinking
hard about ways in which we can talk about the value or measure of impact of research. This is a scheme that was put together by
Johnny Ioannidis from Stanford and Muin Khoury from the National Cancer Institute. And what they talk about is a PQRST framework. P is “productivity.” Productivity would be publications or publications
of highly cited papers. Q is “quality.” How good is the work? And there are ways of measuring quality. How many of you are in any way involved with
clinical trials? Okay. So, you may be familiar that there are ways
of measuring the quality of a clinical trial. There are objective ways of doing that. R is “replication” or “reproducibility.” We’ll talk a little bit about that later on. S is “sharing.” Sharing means that when you do your research,
you produce something that other people can use like data or resources or cell lines or
animal lines, and that then enables other people to do really good work. And the T is “translation”; the ability of
research to lead to something that will actually improve public health, and that’s something
that doesn’t always apply. For basic science projects, translation may
not be quite as relevant. But nonetheless, it is something important
to think about. Okay, so I’m gonna show you a little bit on
productivity and we’ll then think about what the potential implications of this would be. So, the question might be asked, if people
get research support from the NIH, does that increase their academic productivity? That may seem like a stupid question, but
nonetheless, it is one that requires a little bit of thought. So, here, what we do is we look at a group
of over 70,000 investigators, and we look at how much grant support they have over a
period of time. So, these investigators were supported between
1995 and 2014. For each investigator, we count out how much
grant support they’re getting, and we look at how many years they were getting some form
of grant support, and so that tells us what the average amount of grant support they were
getting in any given year during this time. So, these here on the X axis, this would be
the equivalent of getting, say, one R01 a year. And then, over here would be the equivalent
of getting two R01s a year, three R01s a year, and so forth. Now, the Y axis is a measure of productivity. And in this case, it’s a measure of publications
and how well cited those publications are. So, a paper that’s cited a lot counts for
more than a paper that doesn’t get cited at all. And so what we do is is we add up the total
amount of citation impact that a scientist would generate over time. We divide that by the number of years that
they were receiving support. And so that gives us a measure of their citation
impact per year. So, the X axis is input. It’s the amount of grant support they’re getting
per year. And the Y axis is output. It would be the citation impact per year. Okay, so what do you see here? As you go from getting just a little bit of
support, let’s say that might be the equivalent of an R03 a year or an R21 a year, up to
a full, one full R01 a year, we see a dramatic increase in productivity. As you go from having one grant per year to
have two grants per year, we see an increase in productivity. It’s not quite as marked as going from a little
bit to one grant per year. So, the incremental return is a little bit
less. As you go from two grants per year to three
grants per year, there is still an increase in productivity, but that increase, the slope
is not quite as great. And then, as you go from three to four, the
slope gets more and more shallow. So, what this suggests is that there is diminishing
incremental returns, that each additional grant adds not quite so much in terms of output. Another way of thinking about this is that
if you double input, that does not mean that you’re gonna double output. You’ll increase output, but you won’t increase
it by as much. And we’ll have to think about what that means. I mean it’s interesting, but what does that
mean? Well, what it might mean is that we want to
fund more investigators because by funding more investigators, we have an, for each
investigator, that new investigator that we bring into the system who is otherwise not
in the system, we increase the incremental productivity for the entire system. Now, on a more global level, there’s another
way of thinking about this which is that science itself is inherently unpredictable. We do not know what the next great discovery
will be, and we do not know which scientist is going to make the next great discovery. But what we do know is that the more scientists
we support, the more likely it is that somebody’s gonna come up with something really great. For those of you who’ve thought about stock
portfolio, it’s the same thing. Stock portfolio analysis, the more diversified
your portfolio is, the more likely it is that your portfolio will do well. So, a couple of years ago, FASEB put out a
report. And in the report, they talk about a lot of
the themes that we’ve alluded to, the increasing hyper competition within the research system. And so they argue here that research agencies
should monitor how much money individual laboratories are getting. And the reason is is that by limiting the
amount of money that any one laboratory gets, it would enable the agency to fund more laboratories. And by having more people engaged in research,
that might enhance the productivity of the overall enterprise. So, remember, we said the core problem is
that we have too many people vying after a limited number of dollars. And so as an attempt to mitigate that, what
FASEB was suggesting is that we should limit the amount of funding that any one investigator
gets so that this way, we can bring more people into the system. Okay, now we’re going to end by talking about
some other finish lines. So, you may remember that these authors suggested
that the optimal finish line is when a paper gets published and then gets cited. But clearly, this is not the reason why we’re
doing biomedical research. We are doing biomedical research because we
have other more global, more lofty goals in mind. So, let’s think about what they may be. Well, one is that we want to come up with
new treatments, new drugs, new cures that will improve public health. And the problem is that if we look at how
the overall system has been performing over the last few decades, it’s been performing
not so well. Now, this is a paper that came out in 2012
by Jack Scannell and colleagues, and what they did was they looked, so on the X axis
here is time. This is from 1950 to 2010. And the Y axis is the total amount of money
per new molecular entity. So, the total amount of monies being spent
on research per new molecular entity produced, and what one sees is that over this time,
it has been, the efficiency of the system has been steadily declining. In other words, it’s costing more and more
and more money to produce the next new molecular entity. So, the system is becoming increasingly inefficient. Why is this? So, the authors don’t know. They suggest a variety of reasons. One is that there may be too much regulation;
that people are being overly conservative. Perhaps certain paradigms of how research
is conducted may not be right. One particular one that I like is what they
refer to as the “Beatles Problem.” This is the “Beatles,” referring to the Beatles
of the 1960s. So, if you’re a producer and you produce The
Beatles, it’s going to be pretty hard to do better than that as you put and more resources,
let’s say, into your music empire. Okay, so one problem is that the system in
general doesn’t appear to be doing all that well. Now, another problem is that there are concerns
that some of the science that’s being done is not all that good, that it cannot be reproduced,
and that it’s resulting in the development of potential products, for example, that don’t
work. And this is, obviously, a clear concern. One of the key purposes of science is to produce
generalizable and verifiable knowledge. If the finding is something which is a quirk
due to statistical chance or due to the vagaries of a laboratory, that’s something which is
inherently not all that interesting and not all that useful. So, NIH, as many of you may know, is currently
implementing a number of policies to try to improve the rigor and reproducibility of the
science that we fund. And this is something that we can certainly
talk more about during the course of the seminar. Okay, now another way of thinking about an
optimal finish line is an optimal finish line is when you discover something really, really
great; something that is so great that it changes the world. So, here’s an example of an interesting analysis
that was done by people at the Gladstone Institute in San Francisco. And what they did was they looked at what
they referred to as a cure. So, in this case, the cure is Ivacaftor. Ivacaftor is a new agent for cystic fibrosis. It leads to a dramatic improvement in the
clinical state of a subset of patients with cystic fibrosis. So, here’s Ivacaftor. It’s at the very top there. And what they did was they looked at data
from a variety of sources, the FDA, the U.S. Patent Office, Pop Med, The Web of Science, there’s
a lot of data which is out there, and using these data, put together a story of how this
actually happened. So, there’s Ivacaftor at the very top over
here. There were a number of FDA trials that fed
into Ivacaftor being approved. And then, those FDA trials were then fed by
a large corpus of science. It was a large corpus of science that came
from thousands of investigators working in thousands of institutions who produced thousands
of documents, tens of thousands of documents or hundreds of thousands of documents, and
they did this over a period of 50 or 60 years. And using this kind of approach, one can dig
deeply into discovering what actually happened. How did this discovery come about? And you might ask, for example, what was the
role of NIH in enabling this discovery to come about? What was the role of different kinds of NIH
grants or different kinds of NIH grant mechanisms? What collaborations seemed to work well? Were there times when there was rapid progress
and times when progress seemed to slow down? So, we started to work with this kind of methodology. A number of my colleagues put together this
schematic. This is on a drug called a PCSK9 inhibitor. A PCSK9 inhibitor is a biologic that dramatically
reduces cholesterol levels substantially more than statins. And they produce a similar picture. There’s the PCSK9 inhibitor at the very top. There are some FDA trials that feed into it. Those trials are then fed by a number of key
publications, and some of those publications came from NIH-supported work, NIH-supported
scientists, and NIH-supported organizations. So, this is something that we’re now starting
to work on. Can we look at something like this and come
up with a better idea about what role NIH may have played and where NIH’s role and NIH-funded
scientists may have played a particularly crucial role in the development of this cure. Now, you could do this for any of a variety
of outcomes. So, for example, we’re also working a peanut
allergy guideline. And we’re looking at what was the role of
NIH-funded research in that guideline about how to help children with peanut allergies? And one could do this for hundreds or thousands
of outcomes that we might consider valuable, and in this way, we could learn much, much
more about how scientific discoveries come about, what the role of government funding
is in those great scientific discoveries and do this in a way which relies on large amounts
of data and presumably would be more rigorous and objective than simply telling stories
based on memory and anecdote. So, let me end by pointing you to the NIH
Strategic Plan. Congress, a few years ago, asked for the NIH
to submit a strategic plan. This was submitted about a year and a half
ago, and it was very well-received on Capitol Hill. And a critical point in the Strategic Plan
is that we not only aim to fund the best scientists and to fund the best science, but we also
have a responsibility to function as effective stewards of limited taxpayer money. We are very grateful for the appropriations
that we receive, and it is our responsibility to think hard about how we allocate those
appropriations. And so we want to function as effective stewards
and, effectively, our goal is to aim to be an excellent science agency that manages by
results. So, I hope you found this interesting and
gives you something to think about. There’s a lot going on here at this seminar. We hope that you will learn a lot, ask a lot,
talk to us a lot, talk to each other a lot, network, and we wish you, for those of you
who are already funded on NIH grants, wish you all the best for success, that you should
produce great work, that you should do great things for all of us, that you should continue
to be funded as long as you want to be funded for. For those of you who want to be funded, that
you should get funded and do great work. And for those of you who are helping those
people who are funded or who want to get funded, that you should do great work, and we should
all feel incredibly proud of the extraordinarily important work that we do. So, thank you very much, and I would be happy
to take questions. Okay, so the question here is, this gentleman
is a postdoc in a lab or that’s how your career got started, and the question is how do we
think about postdocs? So, that, of course, is a big topic of discussion. One of the points that the Kimble paper made
is that we have more and more postdocs vying after a limited number of faculty positions. There has been some work suggesting, for example,
that postdocs, people who do a postdoc after a PhD are not gaining an income advantage
compared to people who don’t do a postdoc after a PhD. There’s also been some work suggesting that
laboratories that have very large numbers of postdocs are more productive than labs
that don’t have quite so many. But it doesn’t scale. So, having twice as many postdocs in a laboratory
is not the same as twice as much productivity. The reality, of course, is that the majority
of people who do postdocs are not able to get into traditional faculty positions. That may have been the case 20 or 30 years
ago. That is not the case now. And so one thing we all have to do as a community
is think hard about the wide variety of careers that one could potentially pursue in science
other than being, let’s say, a PI in a typical academic laboratory. Another, of course, big issue with postdocs
has been their salaries. And some of you may have followed this. About a year or so ago, the Department of
Labor raised the floor for, the threshold for overtime pay, and that affected a number
of postdocs. And so back in November, December of last
year, we actually scaled up the NRSA pay scale in order to meet what the Department of Labor
was calling for. There are some folks who are from the outside
who are looking to biomedical research enterprise and are arguing that what we ought to be doing
is paying postdocs a lot more and eventually getting to a point where we are melding from
a postdoc model to a staff scientist model, that we would have a number, we would be supporting
a number of people who would work in a laboratory as a staff scientist but not necessarily as
a PI. My father was a scientist for his career,
and that’s how he worked. He had staff scientists who worked with him
for a very long time, and that worked out quite well. Yeah? Other thoughts? Yes! You’re gonna have to scream! Or maybe not. Audience: Thank you. Very interesting talk, indeed. One of the things I’m curious about is what
specific steps is NIH now taking to level the playing field, as you mentioned in your
initial slides, in terms of leveling out the age gap, in terms of funding, et cetera? Because the feedback that I’m getting, I’m
a new investigator, from some of the people who review NIH grants is that in grants that
are submitted by new investigators are typically, you know, before R01, R21, R03 type grants,
the typical, and more and more preliminary data is required. So, in other words, the feedback is that,
to young investigators, is to do incremental work, meaning what they have done in their
postdoc, to basically just advance it. So, you are right that if I were to establish
a new project or more innovative project, it will be very, very hard for me to get that
funded. So, my inclination would be to just extend
or just increment the previous work that I have done. So, what do you think NIH is doing to sort
of change this prevailing trend where reviewers are demanding sort of, in other words or
not, expecting incremental work rather than just new, innovative work? Michael S Lauer: So, that’s a very good question. Now, as I mentioned before, we’ve already
had some success in that back in 2008, when we instituted a new early stage investigator
policy where we basically said that early stage investigators have to have success rates
for their new grants that are equivalent to the success rates for new grants of established
investigators. And that clearly made a difference. If we look at the proportion of early stage
investigators and new investigators among all awardees, that number bumped up dramatically
in 2008. And it continues to be higher than it was
before that policy was implemented. And we’ve also seen, for example, that the
fall and the proportion of early career investigators supported by NIH has flattened. Nonetheless, it is not good enough. It clearly is lower than probably what it
should be, and it’s certainly a concern, exactly as you say, and exactly as others have said, that
it incentivizes people to be more conservative and less ambitious than they otherwise might
be. So, there are a couple things that we’re looking
at. One is that, as you mentioned, we’re trying
to figure out ways that with the limited amount of money that we have, we can free up more
money so that we can fund more investigators. A second is that we’re looking at a program
called the R35. How many of you are familiar with the R35? So, some of you are, and for those of you
who are not, look it up. So, the R35, the idea of an R35 is that we
don’t fund a project. We fund a program. So, you don’t submit specific aims. You talk more about what your program is about,
how your laboratory works, what kind of work your laboratory has done to date, and in general,
the kinds of things that you’re thinking about doing. It’s modeled in a way after the Howard Hughes
Medical Institute, the way they do their funding. Now, we have, not all institutes are participating
in the R35 program, but there are some that do, and some of the R35 programs, they are
specifically targeted towards younger investigators. So, I would encourage you to look it up, and
if your area of science happens to fit into one of those, then pay attention to it because
it may be something that would be interesting for you. All done? Okay. I want to thank you all very much again for
coming, and have a great time!

1 comment on “Evidence-Based Funding Dr. Michael Lauer Keynote 2017 Spring NIH Regional Seminar

  1. I have been working without support or funding on a project for decades. I have produced more than evidence, but proof of many things claimed impossible to demonstrate showing how they are done step by step.

    My work is on the mind, symbolic information processing. I have discovered a method of computing which is always exact and always gives immediate results, and every bit of it verified.

    I can not only not find support, I do not even get so much as a response from the work because it clearly disproves a great deal of human mythology in math and science.

    Where do people like me get funding and support?

    Here is a link to the work. It is a simple grammar book however, it has tons of supporting evidence and proof.

    Every possible grammar, from common grammar to geometry functions by complete induction and deduction of a unit. A mind processes all information exactly one way, if it is functional.

    Man is currently proto-linguistic.

    https://archive.org/details/AUniversalLanguage

Leave a Reply

Your email address will not be published. Required fields are marked *