PREVIOUS ADVICE 
Thanks for answering my question about what to do if a Program Officer (PO) doesn't call back . What ended up happening was this:
After the PO didn't call me back the next day like he said he would, and before I saw your reply, I did essentially what you said to do in step 1. I sent an e-mail the following week asking him to get back to me at a time of his convenience to discuss the proposal and again gave my phone number. There was no response.
After that, I met with the chair of my department to discuss the proposal and go over the critiques. By the way, I'm a new investigator, and the proposal is a K08 career-development award. The chair said that the critiques were very benign, and all of the criticisms were easy to fix. (This makes me wonder why the application was triaged, but that's another story.) She wished that her critiques could be as benign.
When I mentioned that the PO would not return my phone calls or e-mails, she--the department chair--said to stop trying. Because the proposal was triaged and thus not discussed, the PO would have no further insights into how to help me beyond what was already in the written critiques, and that he would simply say to address the reviewers' comments and resubmit it. So, that's what I did, and now the revision is pending review.
Thanks again for your advice. If the revision is scored, then I know it will have been discussed at the study section, and then the PO may be in a better position to help me, or maybe also the scientific review administrator of the study section. I will follow your advice if I don't get called back.
Considering that continuing your efforts to make contact may well annoy the PO--which would not be to your advantage--your department chair's advice wasn't bad. But she missed a few key points.
"Sometimes PIs are less up-to-date or even less informed than the fellows when it comes to looking out for the fellows' interests," comments my NIH insider, himself a former PO. Contacting the PO has many potential advantages, including "beginning a relationship independent of the PI, but sometimes more." The insider continues:
In most reviews, the triage takes place in advance, and unanimous "lower half" applications are triaged automatically. However, if there is disagreement, there is a brief discussion. Or, an application may be initially in the upper half, but barely, and end up being triaged after discussion. In both of these cases, the PO will know that the application came close and was not way out of range and will have a little insight into why that was so. Finally, even if it was triaged without discussion, a good PO will share the insight of experience and provide what I diplomatically call "administrative guidance," which in the case of [a PO] who cares can go beyond the basics.
Although yours seems like a difficult case, generally it's a good idea to have a chat with your PO, even if your proposal was triaged. Your PO is behaving badly, and the advice of your department chair, although reasonable, does not represent the state of the art.
Best of luck,
I am currently a graduate student, and many of my friends have applied for Kirschstein National Research Service Award (NRSA) predoc funding. Two of them have received their reviews, but the scores do not seem to reflect the reviewers' comments. Specifically, the priority score was quite poor, but the comments were rather minor. One of the reviewers asked for pilot data--I know that requesting pilot data is typical of research grants, but is this a new trend for predoc NRSAs as well?
Any advice/insight you may have regarding the NRSA predoc proposal evaluation process will be much appreciated.
Whenever there is a disconnect between the comments and the priority score, it's very important to speak to someone who was in the room. Your friends should call the program contact listed on the review roster and get a firsthand report.
Although I can't of course speak about these particular proposals, it is not rare for priority scores and comments to be inconsistent, especially for predoctoral NRSAs. This is because these awards are based at least as much, and typically much more, on what an applicant has already achieved--his or her academic transcript and demonstrated research promise--than on the strength of the specific research plan. Comments, generally, are on the plan, yet for these awards priority scores are assigned mainly on the individual's qualifications. The best priority scores, in other words, typically go to students with the best GPAs, at the best colleges, working in the best labs, although very strong students at less-than-stellar schools often do just as well if they write an excellent proposal.
This typical focus on academic credentials makes it all the more surprising that one of the reviewers seeks preliminary data.
Then again, maybe it's not surprising. One circumstance in which a call for preliminary data can be legitimate in a fellowship application is if the reviewer is not convinced that the research program the applicant has become involved in is truly viable. Here's how NIH puts it. "Provide an account," reads page 11 of form PHS-411-1, "of any preliminary studies that might demonstrate the utility of the proposed project as a training experience" (italics are added for emphasis). This language applies to all individual NRSA applications--predoc and postdoc. As many embittered former scientists will tell you, nothing is worse than finding out when applying for postdocs or faculty positions that the lab you ended up doing your Ph.D. in was doing pointless or dated research, and that your years of very hard work were all in vain.
That's the charitable way of thinking of the request for preliminary data. The uncharitable explanation is that the reviewer is reviewing your research plan as if this were a standard research-grant proposal. A busy scientist might do this out of habit, or it might be a manifestation of the regrettably widespread view that a fellowship--even an individual fellowship--is just another way of facilitating the mentor's research program.
My advice for predoc NRSA applications? Attend the best department you can get into; notice I wrote "department," not grad school, because what really matters is the people you'll be working with directly and not the reputation of the college as a whole. Study hard. Impress your professors with your thoroughness, seriousness, and competence; notice that I did not write "brilliance," which, although it's a fine thing if you have it, tends to be innate, whereas other skills that are at least as important can readily be learned. Choose an important research topic that you are passionate about and a first-rate research adviser. Quickly make the work your own; your goal is to become an independent researcher, and the earlier your work becomes motivated by your own curiosity--and not by what your adviser or some senior postdoc tells you to do--the better. Tell the reviewers (in your application) where you expect to end up, and how you plan to get there; notice that I wrote "plan," not "hope." Write a research proposal that's unpretentious, short, literate, and sound.
When rejection comes your way, always remember that the reason you were rejected is simple: They just don't get it. In revising your rejected application, address the reviewer's comments thoroughly and tactfully. Acknowledge the faults and correct them, but never lose faith in yourself. And the next time around, make sure they can't possibly miss what a great scientist you are destined to become.