REPOST: Structure Of An R01: Specific Aims
March 20, 2009
Reposted by request of a dear friend of the blog.
One of the most important skills a PI in the biomedical sciences must master is writing grant applications. As we allude to constantly, the basic grant award that is the sine qua non of a successful self-sustaining research program is the NIH R01. The R01 is generally awarded for 4 or 5 years, with an annual direct budget of ~$250,000. This is sufficient to support a small research program of about four or five people, including the PI.
In this series, we will discuss how to structure the Research Plan of a new R01 application (competitive renewals are a different beast), taking each section in turn: Specific Aims, Background and Significance, Preliminary Studies, and Research Design and Methods. (Note: NIH has asserted that in the near future, the R01 applications will be reduced in length from 25 single-spaced pages to 12. It remains to be seen how applicants and study sections will adjust their expectations to this new length.)
We start with the Specific Aims.
The Specific Aims section of the R01 Research Plan is the first page, and it should absolutely without exception fit on that one page. Not one page plus one or two lines, but one single page. This has to be the tightest, most structured part of the application, as many of the study section members who will be scoring your application will only read this section, if they read any of the application at all. The goal of this page is to encapsulate the basic gist of the proposed research and get the reviewer excited. If the reviewer is not excited after reading this page, you are toast.
The first paragraph of the Specific Aims page should, itself, be a self-contained complete overview of the project. The first sentence of this paragraph states in the broadest possible terms the long-term goal of the research project. The next sentence or two describes, again in very broad terms, the approach(es) to be deployed in service of that goal. The final sentence states the specific hypothesis to be tested or other particular purpose of the project that will be addressed using the approach(es).
The next paragraph or two summarizes the background/significance and preliminary studies. The goal here is to (1) justify the importance of the research question and the relevance/utility of the model system chosen to address that question and (2) demonstrate that the selected approach(es) are feasible and will definitely succeed in allowing the specific hypothesis to be tested or other particular purpose of the project to be achieved.
Next is a short transition paragraph that leads into the specific aims themselves. Mine are always two sentences. The first sentence restates the main technical and substantive conclusions of the preliminary studies. The next sentence introduces the specific aims. For example:
The following three Specific Aims are intended to test the hypothesis that blah, blah, blah.
Now comes the enumeration of the specific aims themselves. For an application for 4-5 years $250,000 budget, there should almost always be three specific aims. Each aim should be headed in bold type as follows:
Specific Aim #1: Identify blah basis for blah in the blah
The header for each aim should always fit on a single line of the page. I always present each aim as a verb phrase: identify blah, determine blah, test blah, etc.
After the header, the aim should be described in two or three sentences. The first sentence or two recites what experiments will be performed. The last sentence states what will be revealed by those experiments.
In relation to the nature of the aims themselves, one important consideration is that the aims should be posed in such a way that each aim is at least partially independent of the others. This is because a typical reviewer knee-capper is “If aim #1 is not successful, then none of the other aims can proceed”. My practice, when possible, is to have at least 75% of the experimental goals of aim #1 already complete and in the preliminary studies. Aim #2 can be not at all done, so long as the approaches are well-established and pretty-much guaranteed to work. Aim #3 is an opportunity to get a little more speculative and wide-ranging.
After enumerating and describing the specific aims themselves, the Specific Aims section ends with a single paragraph that reiterates the importance of the proposed research in light of what is already known in the field and what will be vua the proposed studies. The last sentence of this paragraph then pulls back into a very broad perspective on where the proposed research might lead in the long term, and why that will be very important.
Here is a redacted version of a Specific Aims page that embodies this structure:
DrugMonkey 
June 21, 2008 at 11:37 am
All you senior postdocs out there, if you are only going to take one piece of advice from this blog, take this one. This page is the single most important part of your grant and it is your primary opportunity to obtain advocates for your case on study section outside of your primary reviewers. Spend as much time as is needed on this section to make it as tight and compelling as possible.
In my dept we send these pages out to colleagues months prior (usually 2or3) to sending a grant. We go over them in excruciating detail paying attention to every word. After obtaining criticism we re-write and do it all over again, as many times as it takes to get it right. Your colleagues may not have time to go over your entire grant but they can easily go over your SA page with you, get as many eyes on it as you can.
Finally, that redacted page reminds me of many nightmares of the past 7.5 yrs. Can’t you use a pretty color like green or something?
LikeLike
June 21, 2008 at 12:36 pm
I echo Juniorprof. My postdoc advisor is an excellent grantwriter- as evidenced by a high rate of funding…. Without a doubt the single most important thing that I learned from him- were just these kinds of pieces of advice that PP is sharing with you.
Writing a grant and writing a paper are both writing- but grant writing is a special skill that has different goals from paper writing.
LikeLike
June 21, 2008 at 1:01 pm
Great outline and I will underline the importance of the single page of Aims as generating excitement.
I lean much more more translational in my work so I usually lead (first or second sentence) with the “significance” statement. Such as “eighty bajillion people are suffering and we lack knowledge/treatment in this area”. Other than that, what PP said. These are the most important 4 sentences (or so) in the entire grant.
Also, I like to end with the investigator/environment brag- tell ’em why you are the one who is absolutely the best possible laboratory/collaboration to work on this.
LikeLike
June 21, 2008 at 1:13 pm
The redacted Specific Aims page was generously provided to me by a colleague from another institution, whose research–as can be gleaned–is very focused on basic mechanistic studies.
LikeLike
June 21, 2008 at 2:08 pm
I will add my endorsement of the importance of Specific Aims. In fact I will say explicitly something explicit in PP’s post: the Specific Aims section is the most important part of your grant application. Spend a great deal of time crafting it, word by word, sentence by sentence. It will have the added bonus of clarifying your thinking for the rest of the proposal, which should follow the logic of the Specific Aims.
As PP said, this is the place where you hook the reviewer. You want him or her to say, “Yeah, this is interesting. I’d like to know the answer to this question.” If you can do that you more than halfway there. If you are unclear and muddy, you are more than halfway to rejection.
LikeLike
June 21, 2008 at 2:42 pm
Thanks for the helpful post!
I wrote up one of those fancy K99/R00 applications and spent what I felt like was an insane amount of time working on this very sheet, agonizing over what exact words were included or ditched. I think it helped though.
LikeLike
June 21, 2008 at 8:30 pm
I would also add that the Specific Aims serves as the spine/scaffold of the entire application — every single paragraph in the Research Plan should support/develop/refer back to at least one sentence of the specific aims.
LikeLike
June 22, 2008 at 12:05 am
Nice post and I absolutely agree with the “3” specific aims approach. With 4 (I’ve even seen 6!), the aims become too distilled. Often, the 4th aim is unrelated to the rest of the proposal and should be axed. That being said, I just submitted a proposal with 2 aims after fluctuating between 3 and 4, because that served to logically break up the proposal most effectively. Essentially an in vitro/in vivo dichotomy. However, this was the first time I submitted an NIH grant that was not 3 aims.
Also, its good to tell new grant writers to get a range of proposals and see what the “norm” is. Poor grant writers tend to forget that the human beings who are reviewing your proposal are tired, dealing with families, teaching, car repairs, etc. You do not want to make them work harder just to figure out how you are presenting your proposal. And by way of advice, just because the instructions say you can use 25 full pages with a 0.5″ margin on all sides and an 11 point font doesn’t mean you should. Give the reviewers a break, a happy reviewer is a happy scorer (psychology is a bitch, fight it at your peril).
LikeLike
June 22, 2008 at 6:52 pm
Nice post!
LikeLike
June 23, 2008 at 10:08 am
Thaks PhysioProf, spot on. Being a young investigator with a stubborn streak, my initial tendency when approaching the specific aims was to want to do the opposite. If I’m told that 3 aims is the best, I want to have 2 or 5. If one examines the way science actually gets done, always having 3 aims framed in the manner you describe for a 4-5 year project worth $1 million is preposterous.
I have now swallowed my foolish assistant professor pride, and all of my aims pages now have 3 aims, one page only. In fact, I’ve identified 13 elements common to every successful specific aims page I have read. All 13 occur in the redacted version you show.
Silly me: I was confusing good granstmanship with good science!
LikeLike
June 23, 2008 at 10:18 am
I am sure all of our readers would be fascinated to see your list.
LikeLike
June 23, 2008 at 2:06 pm
>I am sure all of our readers would be fascinated to see your list
Here’s my list. The disclaimer is that I am not funded, so please identify elements that can be improved or dispensed with. I’m looking forward to PhysioProf’s advice on the experimental design section, since Beaker definitely doesn’t have a 13-point plan for that part…
The 13 point aims page. These are partly cribbed from “The Grant Application Writer’s Workbook,” by Stephen W. Russell and David C. Morrison, July 2005 edition.
1. Opening sentence. Must mention the important problem in human health or fundamental biology that your proposal will address. Ideally, the problem will be fundamental and health-relevant, but either one of these two is sufficient.
2. Current knowledge statement. The “knowns” most relevant to this proposal (hat tip: Rummy)
3. The gap. “However, our understanding of blah remains poorly understood…”
4. Justification for why the gap is important enough to fill.
5. Our long-term goal is…. This should be a problem big enough to keep the investigator occupied for a decade, or even a whole career.
6. The objective of THIS application is…. This is the part of the long-term goal that you will be able to solve over the course of this grant’s funding.
7. Our central hypothesis is… This is the thing that all parts of the application must point back to–again and again and again.
8. Rationale why this project can and should be done NOW.
9. Rationale why this project can and should be done by YOU. Why you are “uniquely positioned” to do this work, rather than all of the others writing RO1s to do similar stuff. Better yet: be the only person equipped and positioned to do something really important
10. “Our specific aims are….” Each aim should begin with a good strong action verb and it should be short and sweet, like PhysioProf sez. The second part of each aim (sentence 2) should have a hypothesis that is the smaller part of the central hypothesis tested by that aim. The aims part is 3 X recursive. Ideally, as the Prof said, aim 1 is mostly accomplished already. Aim 2 might not be accomplished, but is a “slam dunk,” The 3rd aim shows your incredible vision and foresight in terms of pushing the envelope–a bit more risky, with potential for higher payoff. Alternatively, disease-based proposals can propose some stuff in aim 3 to move the basic discoveries of aims 1 and 2 towards translation: animal models, clinical studies, etc.
11. Innovation reminder: Toot your horn about how the research proposed is innovate and can “capitalize” on your unique advantage (training and past experience, innovative technique, great preliminary results). This is probably the most dispensable component of the 13, since innovation is “nice when you can justify it”, but few grants get slammed for being “not innovative enough, while plenty of innovative grants get slammed for not being logical or careful or detailed enough, especially if the applicant is a new investigator.
12. “At the end of this study, we will have…” The expected outcome, the payoff as a “deliverable.”
13. “Our studies will enrich science or lead to new understanding and identify novel drug targets, etc… This is the best possible human health benefit that might arise from this research in the LONG RUN. Your specific project endpoint does not need to produce a cure–it just has to be a new, strong brick in a big construction project.
LikeLike
June 23, 2008 at 2:37 pm
beaker, that is a most excellent list. my hat is off.
LikeLike
June 23, 2008 at 3:15 pm
Thank you beaker. That is very helpful.
LikeLike
June 23, 2008 at 3:24 pm
Beaker, that list is so good it’s scary.
LikeLike
June 23, 2008 at 3:30 pm
As an aside, I think Beaker’s list also exemplifies what is wrong with NIH funded basic research today. Like Zeno of Elea’s paradox of motion (http://en.wikipedia.org/wiki/Zeno%27s_paradoxes), heavy NIH dollars fund basic science studies that inch towards truly translational research without ever actually getting there.
LikeLike
June 23, 2008 at 3:49 pm
heavy NIH dollars fund basic science studies that inch towards truly translational research without ever actually getting there.
I hate this bullshit perspective. If you know what research will immediately pay-off contact Pfizer or Merck Im sure they would be interested. Knowing a goal does not tell you how to get there. Hey scientists, we need a cure/prevention for autism, GO! Personally, I find the whole “translational” (aka applied) research focus to be myopic at best. Essentially translational research is trying to put in practice what we may already know, without necessarily teaching us anything new. Basic research may not cure cancer next year, but it is guaranteed to generate new information that will help us understand ourselves.
The Noble prize in medicine in 2001 went to three yeast geneticists trying to understand how cells divide in the 70s. Their work provided seminal insights into how our cells divide and established numerous ways in which cancer cells need to develop. Penicillin made sure more soldiers died from battle field injuries in WWII not bacterial infections as in WWI. The X-ray was developed from studies in physics not to improve medicine and human health (same for the technology leading to MRIs). So, what waste-of-time basic research, like the ones that led to the aforementioned breakthroughs, should we be culling to focus on an immediate treatment?
LikeLike
June 23, 2008 at 4:37 pm
Beaker’s excellent list made me think of something I’d read at American Prospect about “Presidential Narratives.” The paragraph struck me at the time as also being about a great grant (if you substitute state of the country to state of the field, and candidate to grant writer).
“Successful presidential candidate stories have three parts. Part one of the story describes the state of the country and its government, clearly defining what is wrong. Part two describes the place the candidate wants to take us, the better day being promised. Part three explains why the candidate is the one and only person who can deliver us from where we are to that better day.”
http://www.prospect.org/cs/articles?article=the_power_of_the_campaign_narrative
LikeLike
June 23, 2008 at 5:01 pm
Basic research may not cure cancer next year, but it is guaranteed to generate new information that will help us understand ourselves.
Yes, but the NCI’s mandate is to cure cancer, not to help us understand ourselves. The track record on cancer treatment since Nixon declared War on Cancer sucks bad.
http://seer.cancer.gov/csr/1975_2005/results_single/sect_01_table.07_2pgs.pdf
From 1996 through 2005, US mortality from cancer dropped by 1.3% per year, but there was also an 0.8% drop in cancer incidence per year in that interval. That essentially means that all the basic research the NIH funded in those 10 years lead to a bottom-line 0.5% per year improvement in cancer survival.
My advice to you is essentially the current American Universal Health Care Plan: don’t get sick. Most of the people that have been waiting for taxpayer-funded basic cancer research to come through for them are dead now.
LikeLike
June 23, 2008 at 5:02 pm
I hate this bullshit perspective. If you know what research will immediately pay-off contact Pfizer or Merck Im sure they would be interested. Knowing a goal does not tell you how to get there. Hey scientists, we need a cure/prevention for autism, GO! Personally, I find the whole “translational” (aka applied) research focus to be myopic at best.
I think we need some new words or definitions or something. Me, I don’t view “translational” as necessarily being identical to “applied” or what BigPharma would be doing.
I certainly value basic research defined by investigators doing whatever in the heck occurs to them as being an interesting scientific question or problem. I grasp all the arguments about how you cannot predict what is going to be eventually of use.
I also think that the navel-inspecting, self-referential models (or theories or questions) can stand being brought to account for whether they are likely to be relevant to human health. Also to stand an evaluation as to whether those scientific traditions are actually getting in the way of more-relevant approaches and if more-relevant approaches can be added into the mix (instead of wholesale replacement).
As readers know, I am also a strong advocate of a two Aims for me, one Aim for Program (or Congressional) priorities approach to science. In the areas of basic research most relevant to me (and the study section on which I serve) I am frequently struck by a thought of “gee, if they’d just do this one simple manipulation or add these two experiments, voila! instant translational appeal”. In short, I see it as a very simple thing in many cases to do what I view as translational without wholesale re-orientation of the research approaches. If it isn’t going to really alter your research that much and it makes it easier to get funded, why would anyone turn up their nose? That I don’t get…
LikeLike
June 23, 2008 at 5:18 pm
I suspect that this is pretty specific to the approaches and subject matter of the basic science you are immersed in. It certainly is not the case for my own research, or the vast majority of what I see in the study sections I have served on.
Frankly, I have pretty much given up on the fake-ass “By studying this protein in C. elegans, we are like totally going to cure Alzheimer’s disease! For realz!!” I don’t give a flying fuck if I see this gibberish in the basic science grants I review, and I no longer waste my breath on it in my own.
LikeLike
June 23, 2008 at 5:46 pm
lorax @#8 Also, its good to tell new grant writers to get a range of proposals and see what the “norm” is. Poor grant writers tend to forget that the human beings who are reviewing your proposal are tired, dealing with families, teaching, car repairs, etc. You do not want to make them work harder just to figure out how you are presenting your proposal.
Beaker @#10 Being a young investigator with a stubborn streak, my initial tendency when approaching the specific aims was to want to do the opposite. If I’m told that 3 aims is the best, I want to have 2 or 5. If one examines the way science actually gets done, always having 3 aims framed in the manner you describe for a 4-5 year project worth $1 million is preposterous. I have now swallowed my foolish assistant professor pride,
These two comments really resonate for me and emphasize that a grant application is nothing more or less than a communication. And what is most important is that the applicant learn to speak the language as fluently as possible. It is a bit stupid to rail about the idiocy of the language itself- the rules of English are about as irrational and idiotic as you can get and yet if you want to communicate in English….
With respect to the “norm” the more specific you can get, the better. I see people writing in a style that may be appropriate for other funding agencies (?) but is totally foreign to NIH-land. This is not going to go well. If you usually write for other agencies, for goodness sake take some advice from the NIH-experienced! Ideally, you should see some examples and get some advice from applicants who have submitted to the specific study section you plan to direct your proposal towards. I have little doubt that the many CSR study sections have different dialects.
Three Aims is not as set in stone as PP and the chorus might be implying here. But I’d suggest you have good reasons for varying from the model. I’ve written 2 Aim R21s and 4 Aim R01s for example. Sometimes that’s the only way to structure things.
LikeLike
June 23, 2008 at 7:53 pm
I apologize for continuing to contribute to the tangential topic that has been raised in this thread: translational vs basic research, but here goes nonetheless..
Frankly, I have pretty much given up on the fake-ass “By studying this protein in C. elegans, we are like totally going to cure Alzheimer’s disease! For realz!!” I don’t give a flying fuck if I see this gibberish in the basic science grants I review, and I no longer waste my breath on it in my own.
I absolutely agree, in my own field its “potential drug target.” Total BS throw away sentences in papers and grants. Now admittedly I use the phrase too, however I only use it if I can actually envision how the protein I am studying could actually be used as a drug target, how the drugs might work, and where they could come from.
Basically, I think I agree with DM and PP in that applied/translational research is critical and the main thrust of NIH. I also think it is imperative for those of us not actually doing clinical research (which is trivial to justify IMO) to understand the potential application of our basic research. However, my problem is with the tone/mindset of whimple who seems to have a very shortsighted view. Based on their philosophy, penicillin may not have been identified because their view is so closemindedly goal oriented. That mindset makes SIWOTI reflex fire repeatedly.
How will studying this mold and its interactions with soil bacteria help? What, you say it might, might help?!?! Well people are dying of bacterial infections now, we don’t have time for might or maybe…get my leeches. and scene.
LikeLike
June 23, 2008 at 7:55 pm
re: 2 aims in the R21 application.
Much like R01s are generally 3 aims in scope, the R21 is about perfect for a 2 aim approach. 2 year project, limited scope of an exploratory project….it just reeks of 2 aims.
LikeLike
June 23, 2008 at 8:11 pm
(1) My personal position is that basic science is underemphasized/underfunded relative to translational/applied.
(2) R21s are almost always a total fucking fool’s errand! I would almost never recommend that someone write an R21. The circumstances under which they make practical sense are exceedingly circumscribed.
(BTW, don’t ever apologize for participating in, or even initiating, tangents. This is a fucking blog!)
LikeLike
June 23, 2008 at 8:46 pm
As I am right now today rewriting the specific aims for two of my NIH proposals… this advice is incredibly timely. Beaker- the 13 point list is awesome- but PP, with all those paragraphs I had to go down to
LikeLike
June 23, 2008 at 10:02 pm
Here’s a philosophical question: why is the RO1 page 1 so predictable that it can boiled down to 13 points in the first place? I was fortunate enough to train with some world-class scientists, who have literally thousands of publications between them. None of them EVER used a specific aims page or anything remotely resembling it as a roadmap for their science. Why is there such an obvious discrepancy between how we as scientists describe what we plan to do and how we actually go about doing the science once we get the money?
The answer lies in the review process. How to compare two grant applications fairly if one has 2 aims and another has 6? How to rapidly develop a logical, critical justification for picking one grant over the other if there is little continuity in the style among the grants written? If you were given a Harry Potter book, a Sidney Sheldon book, a Hemingway, and a Don Delillo–how would you possibly decide on the criteria for deciding which is better? Impossible! The study section would be in chaos.
The Specific Aims page and what comes after has evolved from a mixture of the rules laid out by the NIH, along with the realities of the reviewer, who has a big stack to go through, criticize, and defend. My initial mistake as a young investigator was to think that it was more than that. Its structure is simply the reality of where we are in terms of funding in the USA in 2008. The specific aims page is a school assignment. Learn to do it well, then get back to the science.
LikeLike
June 23, 2008 at 10:04 pm
This is a fucking blog!
Sweet, usually you have to pay to be on those kinds of blogs.
LikeLike
June 23, 2008 at 10:08 pm
Off topic, but as PP encourages: it’s a blog.
I’m all for funding more basic science, but not with NIH dollars. If you’re not even going to pretend that your work has any relevance to HEALTH, please go elsewhere (why do program officers even allow this?). The problem isn’t with funding applied science (like clinical trials), it is with funding translational science, where translational can be defined as taking observations made in model systems or in cell lines and determining whether those observations still hold true when applied to the entire human animal and whether if they do, some practical good could come of it. People who want to study mold and its interactions with soil bacteria: have I got some NSF dollars for you!
I don’t understand complaints about funding translational research (distinguished from clinical research)… we know (from NIH data) that clinical research today is mostly concerned with small gains and optimization and not large breakthroughs. You should gladly put some dollars towards translational research today; that way when your personal health needs the results of that research, it has a chance of being ready for you.
LikeLike
June 23, 2008 at 10:34 pm
Let me know when you come up with a way to figure out ahead of time whether a proposed research project is “relevant to HEALTH”. I’ll be waiting. That is, if I don’t die of old age first.
LikeLike
June 24, 2008 at 7:36 pm
PP-
What I was trying to say was with all those paragraphs I had to go down to 0.7 inch margins, and no breaks between paragraphs to get it on one page. I hate to do that.
Sorry I didn’t notice that my original comment had been cut off!
LikeLike
March 24, 2009 at 10:25 am
Thanks! This was really helpful! I ordered the Russell & Morrison workbook yesterday.
LikeLike
April 17, 2009 at 9:27 am
I had written an aims page, and then I remember this post….a bit of editing later, I am very happy with how it looks.
LikeLike
April 22, 2009 at 4:24 pm
I really like Beaker’s approach — I’m a professional grant editor, and it’s almost identical to mine (largely derived independently, and then honed by attending a Russell workshop). In fact, I think his/her suggested structure for the opening paragraph has now become even more important, given the revised review criteria that make “impact” the key scoring factor.
The one difference in my usual construction is moving sentence 9 (why this work should be done by YOU) to follow sentence 11. In position 9, that sentence disrupts the flow, because you’ve just established the need for doing the project NOW, so the attentive reviewer most wants to know WHAT you’re going to do. In my structure, position 9 holds my segue into the numbered Aims paragraphs — “We will test this hypothesis by pursuing three specific aims, to: ”
Finally, I (and Russell/Morrison) disagree about the absolute necessity of keeping the Aims section to just one page. I think it’s more important that all the points in Beaker’s list are well covered (with adequate white space between paragraphs) than to insist on meeting an arbitrary page recommendation — though that may become a requirement with the next version of the 398 package (with the 12-page research plan). Russell actually recommends 1 1/3 – 1 1/2 page as being ideal, with the rest of the second page being used for the Significance paragraph of the Background and Significance section, in hopes that some of the non-assigned Study Section members may keep on reading after the Aims section.
LikeLike
April 22, 2009 at 4:29 pm
drdrA, you don’t have to use a full line space to give a nice visual “break.” I find that 0.4 pts is the minimum; I prefer 0.6, while 12 (a full line space) is a waste. I find that the difference between using 0.6 pts and a full line in a 25-page grant is about three extra pages, if needed.
LikeLike