Hard charging early career Glam neuroscientist Kay Tye had an interesting claim on the twitters recently.
Wow, that sounds the like cheapest and shortest revision I've ever heard of. Ours are typically several (and I mean 2-4) orders of magnitude greater than that!
I can't tell if you are being sarcastic because 1 week of work sounds like a fantasy I have never experienced before. https://t.co/Qmmhy28ZYU
— Kay Tye (@kaymtye) September 10, 2018
The message she was replying to indicated that a recent request for manuscript revisions was going to amount to $1,000, making Kay's costs anywhere from $100,000 to $10,000,000. Big range. Luckily she got more specific.
Never hit 10M but have gone over 1M for sure. If you include salaries.
— Kay Tye (@kaymtye) September 11, 2018
For manuscript revisions.
Let us recap.
The bog standard NIH "major award" is the R01, offered most generically in the 5-year, $250,000 direct cost per year version. $1,250,000 for a five year major (whoa, congrats dude, you got an R01! You have it made!) award.
Dr. Tye has just informed us that it is routine for reviewers to ask for manuscript (one. single. manuscript.) revisions that amount to $1,000,000 in cost.
Ex-NIGMS Director Jeremy Berg cheer-led (and possibly initiated) a series of NIH analyses and data dumps showing that something on the order of 7 (+/- 2) published papers were expected from each R01 award's full interval of funding. This launched a thousand ships of opinionating on "efficiency" of NIH grant award and how it proves that one grant for everyone is the best use of NIH money. It isn't.
I have frequently hit the productivity zone identified in NIGMS data...and had my competing revisions criticized severely for lack of productivity. I have tripled this on at least one interval of R01 funding and received essentially no extra kudos for good productivity. I would be highly curious to hear from anyone who has had a 5 year interval of R01 support described as even reasonably productive with one paper published.
Because even if Dr. Tye is describing a situation in which you barely invest in the original submission (doubtful), it has to be at least $250,000, right? That plus $1,000,000 in revisions and you end up with at best 1 paper per interval of R01 funding. And it takes you five years to do it.
The Office of Extramural Research showed that the vast majority of NIH-funded PIs hold 1 (>70%) or at most 2 (cumulative >90%) major awards at a time.
NIGMS (and some of my fellow NIH-watchers) have been exceptionally dishonest about interpreting the the efficiency data they produce and slippery as otters about resulting policy on per-PI dollar limitations. Nevertheless, one interpretation of their data is that $750,000 in direct costs per year is maximally efficient. Merely mentioning that an honest interpretation of their data ends up here (and reminding that the NIGMS policy for greybeard insiders was in fact to be about $750,000 per year) usually results in the the sound of sharpening stone on steel farm implements and the smell of burning pitch.
Even that level of grant largesse ("largesse") does not pay for the single manuscript revisions that Professor Tye describes within a single year.
I have zero reason to doubt Professor Tye's characterization, I will note. I am familiar with how Glam labs operate. I am familiar with the circle jerk of escalating high-cost "necessary" experimental demands they gratify each other with in manuscript review. I am familiar with the way extremely well funded labs use this bullshit as a gatekeeping function to eliminate the intellectual competition. I am perhaps overly familiar with Glam science labs in which postdocs blowing $40,000 on single fucked up experiments (because they don't bother to think things through, are sloppy or are plain wasteful) is entirely routine.
The R01 does not pay for itself. It does not pay for the expected productivity necessary to look merely minimally productive, particularly when "high impact publications" are the standard.
But even that isn't the point.
We have this exact same problem, albeit at less cost, all down the biomedical NIH-funded research ranks.
I have noted more than once on this blog that I experience a complete disconnect between what is demanded in peer review of manuscripts at a very pedestrian level of journal, the costs involved and the way R01s that pay for those experiments are perceived come time for competitive renewal. Actually, we can generalize this to any new grant as well, because very often grant reviewers are looking at the productivity on entirely unrelated awards to determine the PI's fitness for the next proposal. There is a growing disconnect, I claim, between what is proposed in the average R01 these days and what it can actually pay to accomplish.
And this situation is being created by the exact same super-group of peers. The people who review my grants also review my papers. And each others'. And I review their grants and their manuscripts.
And we are being ridiculous.
We need to restore normalcy and decency in the conduct of this profession. We need to hold the NIH accountable for its fantasy policy that has reduced the spending capability of the normal average grant award to half of what it was a mere twenty years ago. And for policies that seek to limit productive labs so that we can have more and more funded labs who are crippled in what they can accomplish.
We need to hold each other accountable for fantasy thinking about how much science costs. R01 review should return to the days when "overambitious" meant something and was used to keep proposed scope of work minimally related to the necessary costs and the available funds. And we need to stop demanding an amount of work in each and every manuscript that is incompatible with the way the resulting productivity will be viewed in subsequent grant review.
We cannot do anything about the Glam folks, they are lost to all decency. But we can save the core of the NIH-funded biomedical research enterprise.
If you will only join me in a retreat from the abyss.