A note to new investigators

Academic scientists like myself have three major responsibilities: teaching, scholarship (i.e. research), and service to one’s discipline. As part of that service component, I review grants for the National Institutes of Health (NIH) through the Center for Scientific Review (CSR) several times a year.

I have a love-hate relationship with this aspect of my career. On the one hand, I get to see the science before it has been done. The ideas to be tested. Innovative new techniques. The potential not yet realized. I love this. On the other hand, most grants (more than 90% in my discipline) will fail. Most very good and even excellent proposals will not be funded. This is depressing. There are very few other funding sources which means that these scientists, especially new scientists, may end up not succeeding in their academic careers. I could write about what this will mean for the quality of higher education in the coming years but I have a different sort of rant in mind today.

This cycle I have several proposals from new investigators. Like everything else, grant writing is a SKILL. You must practice it to become proficient. Graduate education in the US generally provides only a small opportunity (e.g. 1 class with 1 or 2 mock proposals) to develop this expertise. Thus, most new investigators make many mistakes in grantsmanship when they first start submitting proposals. I have a few otherwise interesting proposals in my stack this time from new investigators. However, flaws in grant writing with the current funding environment, mean that it is very likely that these proposals will be seen again by this or another study section.

I will confess that my business education has shortened my attention span. I am looking for explicit, not implicit, explanations. If you think that your proposal may go to me or someone like me then please consider the following:

1. Use the spell check function of your word processor. If you cannot pay attention to this tiny detail why would I trust you with millions of our hard won tax dollars? The answer is, I would not.
2. Tell me explicitly why your work is important. How will your experiments make things better for many of us on the planet, or even a small group of us suffering from your favorite orphan disease?  Given where the funding line is, your work must be important and have “impact.” Remember, these are our tax dollars at work. What is the win?
3. Do not ask for 2 years of funding when you are proposing 4 years of work. One criterion that I must assess is feasibility. Is it feasible for you to do 4 years of work with 2 years of funding? That would be no.
4. Do not propose every possible experiment related to your disorder of interest. Yes these are all reasonable things to do but it is unlikely for one person to do all of them within the given grant period.
5. Do not propose to utilize a variety of technically challenging techniques that you have never done before as the basis for your proposal. Again, this is a feasibility issue. Is it feasible for you to develop these techniques within the proposed grant period without having lined up experts to help you with them? That would be NO. If you need a technique that you have not yet mastered, please do identify and get a letter from someone who can assist with the technique.
6. Please do tell me how this proposal will lead to the next proposal. This is related to the win. If all of your experiments succeed as proposed, how will science be different? What new avenues of research will open because of your results? Please do answer the question: and then you will what?
7. If you show me bar graphs and tell me some factor is reduced by 50%, please tell me whether this data has reached statistical or biological significance. I am not impressed by n = 1. Please do share your p values, statistical tests chosen and sample sizes. Did your experiment have sufficient power to detect a difference?
8. Are you using animal or human subjects? Select agents? Have you completed all of the regulatory sections appropriately? Have you justified your sample sizes (see #7)? Are the protections adequate? I will flag your proposal for administrative review if you have not addressed these sections adequately.
9. Can we talk about independence? If you have matured into an independent position but are still sharing space with a former mentor, please address this explicitly. We will talk about it at the meeting. We want you to succeed. We want you to have your own lab and your own office. We want to see letters of support from you mentor or department chair that should you win this proposal that these will be granted to you. A collaboration is great. Indentured servitude is not.
10. How about format? Grant proposals with all of their supporting documents can run more than 100 pages. Furthermore, you can submit up to 10 appendices. I would be very grateful if you did not. We get the proposals on disk & now need to print them ourselves. I do mine at home since it is not a grant appropriate expense for me. If you need to include one or two to prove you have the expertise great, please do, but otherwise, spare my toner cartridge.
11. What about white space? Please oh please do not overwhelm me with half-inch margins and no spaces between paragraphs. Please do break your large paragraphs into smaller ones. Please do include figures or tables or bulleted lists. It’s daunting when it is just page after page of the smallest font allowed.
12. I read the shortest grants first. The 150 page documents overwhelm me so I start with the shorter ones. Yes some proposal need all those pages (e.g. clinical trials, epidemiological studies, and multi-PI proposals) but most do not. Less is more. Truly.
13. Letters of collaboration. This may seem obvious but if you say in your proposal that Dr. X is going to do amazing technique Y for you, you need to have a letter from Dr. X that enthusiastically agrees to this. Also, be sure that the letter from Dr. X is current, not from 5 years ago.
14. Laboratory resources. If you tell me that you are going to do amazing technique Z, please have all of the required instrumentation needed for amazing technique Z. At the very least, have it in your budget to purchase it. If the nearest instrumentation is 3 states away, please line up a collaborator with access to that instrumentation.
15. Innovation. This is a tough one. We are asked to review and rank it yet at times this criterion competes with feasibility. It may be innovative but if no one has done this is it feasible? Yes I know this is not fair. Reviewers vary individually as to how they weigh these at times opposing criteria. At the very least, use some state of the art methodology perhaps to a problem that has not yet been approached this way. The creation of new models using old methods is still considered innovative.
16. Acronyms. Please define all of your acronyms at least once. Feeling especially kind? Put all of your abbreviations in a single list somewhere for me to refer to. Making lots of mutants in a gene family, please be kind and make me a table to summarize them all. Even better, add a comment as to why you chose these specific mutants. I am sure you had a reason so please do share it in a clear easy to find way.
17. Expertise. Do not assume that your reviewers are experts in every aspect of your proposal. Perhaps we use the same model system. Perhaps we use the same techniques. Perhaps we study the same pathway using different methods. It is unlikely that we are a 100% match since that would put us in a conflict of interest situation where we needed to recuse ourselves from reviewing your proposal. For example, if your proposal uses both mouse models and in vitro culture assays, one reviewer may have mouse experience, one reviewer may have the in vitro assay experience and a third reviewer may study your same biochemical pathway. Be kind to all of us and put your experiments in context in a general way at least in your specific aims.
18. Resubmissions. These days it is very likely you will end up resubmitting your grant. Our reviews will likely make you angry and upset since we will have missed some points or misunderstood others. However, you need to respond to us when you try again. BE POLITE. I do not care how angry you are, your introduction should begin with something appropriately obsequious like “I would like to thank the reviewers for their astute and insightful comments. I have greatly improved the proposal based on their comments. Reviewer 1 noticed X. Based on his/her comment, I have now revised specific aim 1 to address this concern.” Be sure that you somehow mark the changes in the revised document (e.g. thick line down the border, change of font, etc.). You may hate this advice.  But we reviewers spend hours reading your proposal and writing our reviews. We want to help you win your grant. We are trying to share with you, in a polite way, our concerns and recommendations so that should we see your grant again we can improve our enthusiasm (i.e. score). You may think we are idiots just please do not write this into your grant resubmission.
19. Sob stories. Please do not share these. Many of us are in the position where if we do not get our next grant we will not keep our jobs. Do not burden your reviewer with this. We are sorry. This is a hard path and many will not succeed. But, it will not help your score to whine about it. The major exception to this is an act of nature. Baylor flooded a few years back. New Orleans was decimated by Katrina. By all means if your laboratory tools were ruined by acts of nature and your resources lost, please do explain why your productivity is down.
20. Time lines. These are not required but they are incredibly helpful. Please do create some sort of GAANT chart that shows when each specific aim or sub aim will be worked on during the grant period. Often times we are asked to cut 5 year proposals into 4 years proposals and this will help us see what will be lost or whether the proposed request is reasonable.
21. Dependencies. This one is tricky. If you have a 5 year grant with 4 aims, you do not want all of the aims to be dependent on the success of a single aim. That is, do not have the majority of experiments dependent on the success of Aim #1 which is to be completed in year #1. Have a back up plan. Have alternative approaches. Be sure to share these with the reviewers. Be sure to explicitly tell us what the anticipated results are and how you will proceed if the experiment fails – either in outcome or technique. Have related parallel plans.
22. Review criteria. A final reminder. We review you (can you do it?), your institution (do you have everything you need?), your approach (is this the best way to do it?), the significance of your work (do we care?), the innovation of your work (is it shiny?), your budget (is it realistic?) and your compliance with regulatory issues (will you follow the rules?). Together these need to make a strong package. Failure to address any one of them can kill a proposal these days.