On 30 April 2025, the US National Institutes of Health (NIH) revised their 2024 Public Access Policy. Not only would the requirements of the Biden White House Nelson Memo still be going into effect, the timeline for their implementation has been accelerated from the beginning of 2026 to 1 July 2025. Jay Bhattacharya, Director of the NIH announced the policy change in a separate statement. To review, the Nelson Memo requires the immediate free availability, upon publication, of any papers detailing the results of federally-funded research, replacing the 12-month embargo previously allowed under the Obama era Holdren Memo. The policy also requires the research data behind those papers to be made publicly available at the time of publication, and all the data collected throughout the course of the funding award eventually be publicly released.
This announcement put to rest uncertainty, at least for the NIH, of whether the new Trump administration would support or reject this policy initiative from its predecessor. We reached out to the Scholarly Kitchen Chefs and asked their opinions on this development.
Rick Anderson
It has always been an open question how the Office of Science and Technology Policy (OSTP) in a Trump White House would position itself with regard to open (or public) access to published reports on federally funded research. Would they see the openness impulse as an expression of “power to the people” leftism, and therefore something to quash and eliminate from federal programming? Would they see it as a salutary expression of populist conservatism, opening government spending to more transparency and accountability? Would they ignore the issue entirely, seeing it as far down on their priority list of revolutionary actions and initiatives? (And if they ignored it, would federal agencies now headed by Trump appointees follow suit and ignore the provisions of the Nelson Memo, or would they continue to carry out the public-access guidelines begun under the Biden administration?) In the context of this general uncertainty, the NIH’s recent announcement that it will not only carry on with the Nelson Memo framework, but also aggressively accelerate its implementation timeline, is very interesting. Jay Bhattacharya, the recently installed head of the NIH, has a colorful and controversial history in many ways, but I would not have anticipated this particular move from him. I really have to wonder whether he had conversations with the scholarly publishing lobby before making this decision – and if so, whether it’s intended as a deliberate poke in the eye to them.
David Crotty
Who knew that the Nelson Memo would be the only Biden-era DEI regulation to survive in the Trump administration? The introduction to the Memo devoted as much text to discussing equity as it does the value of open science, faulted the previous Holdren Memo policy because it “has limited immediate access of federally funded research results to only those able to pay for it or who have privileged access,” and states that a goal of the Nelson policy is “To promote equity.” And yet while the NIH has banned any university with a DEI policy from receiving research funding, it seems willing to take a pass on following those same rules internally. The same sort of conflict can be seen in a policy requiring immediate unrestricted worldwide access to all funded research data from an administration that feels the country must “safeguard our intellectual capital, we must restrict foreign access to sensitive data and strengthen oversight of international collaborators.”
Asking for consistency and coherence from this administration is perhaps foolish, so let’s instead look at the practical aspects of implementing an expensive and time-consuming policy in a period where the agency is facing a 39.2% cut in overall funding. Reducing the agency’s budget from $48.7B to $29.6B means a lot less science is going to be done, and implementation of the Nelson Memo means each of those remaining projects just got a lot more expensive, which suggests even fewer will be funded overall. From the OSTP’s estimate, the average APC for a federally-funded paper is around $4,000. Multiply that by the 146,000 papers published last year with NIH funding (about half of all papers listing federal funding) and you get an annual APC spend (not accounting for inflation) of around $584M. But that’s only a small portion of the actual costs, which are largely found in the open data portion of the Memo. The first phase of the Realities of Academic Data Sharing (RADS) Initiative found (admittedly likely underestimations) a cost of $36,000 per NIH grant for the recipient’s data management and sharing. With the NIH issuing 60,000 grants per year, that’s $2.16B in spending. On top of that, the study saw annual institutional costs for data management and sharing of between $800,000 and $6M, averaging at $2.5M per campus, presumably funded from grant overheads, which are themselves in danger of being drastically reduced. Add in the new tariff surcharges coming from laboratory reagent suppliers and research budgets are starting to look extremely bleak.
And of course the policy is retroactive, so if you were awarded a grant before the Nelson Memo was issued, you had no chance to budget for these extra expenses and will have to find the money to pay them somewhere (presumably by letting go of postdocs, technicians, and graduate students).
Another group facing significant financial risk here are small and independent publishers, as well as research societies with journal publishing programs. To dispute Rick Anderson’s suggestion above, I’ve not heard of a single large commercial publisher doing any active lobbying against the Nelson Memo. There’s still a knee-jerk reaction out there, assuming that the largest of publishers are fighting against open access. But this is a vestige from a different era. Most of the large commercial publishers have fully embraced open access and are reaping the rewards in revenue and the market consolidation of power that comes with transformative agreements. This policy will only add to their profits and further cement their dominance. Expect the few remaining independent societies to seek out the shelter of publishing service agreements with the big commercial firms, completing the rush we saw when the policy was first announced.
Let’s also not ignore the costs of compliance. This is a complex policy, and to be effective, researcher behavior must be monitored and checked. Ensuring that a PMC ID accompanies each published paper is the easy part, checking to see whether the data collected in the course of the grant is complete and accessible will take significant effort. Given the massive layoffs of staff at federal agencies, who is going to perform this oversight and do all the work necessary to ensure compliance?
In the end, the benefits of open access and open science may prove valuable enough to justify the efforts involved. To achieve that outcome, the NIH (and all other federal agencies) must be willing to recognize the actual costs and efforts necessary to make these policies work and invest appropriately in the open science future they hope to achieve. If it’s worth doing, then it’s worth doing right.
Lisa Hinchliffe
Librarians play a key role in supporting federal public access policies. The news that NIH is accelerating the implementation of its 2024 Public Access Policy provides an opportune moment to highlight their contributions and consider the impacts of the acceleration.
While we are steeped in policy developments related to scholarly communications, most researchers are not. Without outreach and education from their librarians, it is unclear whether researchers will be aware of the early policy implementation and its implications. Librarians typically offer workshops, training sessions, and consultations to help researchers understand policy requirements and also develop user-friendly materials such as step-by-step guides, FAQs, and instructional videos that explain how to comply. Even librarians who welcomed the zero-embargo public access mandate are now scrambling to put such support in place immediately, and will be challenged to reach scholars at this point in the academic calendar. The coming finals, graduation, and summer break transition is not an ideal time to get researchers’ attention, particularly as labs will be experiencing a period of transition for their postdocs and graduate students.
In addition to general education, librarians often provide hands-on compliance support. They assist researchers in identifying which publications fall under the policy and guide them through the process of submitting manuscripts to the NIH Manuscript Submission (NIHMS) system. Librarians may also help track compliance status, troubleshoot submission issues, and communicate with NIH program officers if needed. Librarians also manage institutional repositories that can complement NIH’s public access efforts. By encouraging researchers to deposit their manuscripts into these repositories, librarians help improve discoverability, ensure proper versioning, and maintain accurate metadata. This can serve as an additional layer of open access support beyond PubMed Central.
Finally, libraries provide critical guidance on publisher policies and copyright. They help researchers understand journal-specific rules regarding public access, advise on how to retain rights to share manuscripts, and assist with negotiating or interpreting publishing agreements. This support is essential to ensuring that authors retain the necessary rights to comply with NIH’s requirements without infringing on publisher contracts.
With publishers now also pressured to implement their NIH-related practices by the accelerated deadline, librarians can anticipate additional confusion during the transition, as manuscripts that were accepted for review under one NIH timeline are now subject to another. Additionally, libraries may be seeking to amend timelines for recently negotiated transformative agreements to accelerate implementation so that manuscripts are covered by the July 1 date.
If one ranks the disruptions experienced in 2025, the acceleration of the implementation timeframe for the NIH Public Access Policy is among the least. But that is not because it is insignificant, but rather because others have been far more so. And, at least this change is in keeping with most libraries’ support for open access. There’s an immense amount of work to do in the coming weeks to prepare institutions and researchers. Nonetheless, what I can tell you from listserv discussions and the like is that librarians are gearing up and will be ready!
Haseeb Irfanullah
After reading new NIH Director Dr. Jay Bhattacharya’s April, 30’s statement on advancing the effective date of the 2022’s Nelson Memo by six months, I got confused.
Over the past few weeks, the Trump administration’s all out attacks on the US research system have been nothing but catastrophic. The House and Senate Appropriations Committees have already passed (updated on May 2, 2025) a 39.2 percent cut of the total NIH budget for 2026. How much of this will be from the research budget will soon be clearer, but, the budget cuts of science wings of other agencies, e.g., of the NASA (30-48.3 percent), of its Science Mission Directorate (30.9 percent), of the National Science Foundation (57 percent), and of the Department of Health and Human Services (35-38.9 percent) indicate that the NIH will face the same fate. And, in his first public speech on April, 21, Dr. Bhattacharya defended the possible NIH budget cuts.
And then, nine days later, in his April, 30 statement, the NIH Director dreamed about reinstalling the American people’s trust in research. But how does he plan to do that? By slashing the research budgets?
In the name of speedy access to NIH-funded research by the public, I am afraid, I anticipate partial or total destruction of scholarly communication in the US health sector, as we now know it.
- The peer-review system will be bent, stretched, and/or functionally dismantled for quicker public access.
- Since journal publishing is inherently slow, communication of important research will be encouraged through briefing notes, bulletins, preprints, and social media.
- Research funds will be allocated according to what politicians personally believe important.
- I expect to see unusual use of critical research findings led by politicians and politically appointed administrators; scientists will be sidelined.
- By underscoring the ‘Making America Healthy Again’ agenda, thus exclusively focusing on US citizens, the NIH will lose its international footing and leadership in groundbreaking work.
- In terms of the research-policy nexus, instead of evidence-guided policy influencing, we will see politician-guided research influencing.
The NIH is just the beginning; the R&D of other public sectors will follow. By submitting scholarly freedom to the politicians, are we going to see ‘de-growth’ of scholarly publishing in the US?
Todd Carpenter
The motivations of this administration can be difficult to decipher, and the policy decisions can often feel confusing. The case of the NIH Open Access (OA) policy, the rationale seems clear. Consistent with the Administration’s “free speech” ethos, the acceleration aligns broadly with its many priorities. However, there are many reasons for concern, beyond the near term implementation.
The justification presented in the acceleration memo appears well-grounded, is likely good for science, generally, and seemingly deeply held by the leadership. The near-term disruptions of advancing implementation to July 31, versus December 31, will likely be minimal and hardly a significant shift in direction by the NIH. More generally, signals have been emanating from Washington that the 2022 Nelson Memo, wasn’t as imperiled as one might have presumed in the rush to destroy all Biden-administration initiatives.
There are multiple motivations for open science. It demonstrably accelerates the distribution and improves the impact of research. It also supports the broad push for public knowledge and the right of the public, to see, access and consider the outcomes of publicly funded research. Less visibly—but just as crucial—is the critique that the existing research communications system is a dominated by powerful interests—be they corporate publishers, entrenched editorial hierarchies, the bibliometric approach that govern assessment and advancement, or the unsustainable cost models that support publishing.
This reinforcement of NIH policy also reflects a form of political triangulation. This administration has been extremely successful—much like the Clinton administration in the 1990s—of picking up causes with a kernel of sensibility and are broadly seen as beneficial policy. Public access mandates are one such cause. While acknowledging the benefits of open science, many also view rapid and open publication of research to circumvent entrenched ‘gatekeepers’ of science, who may suppress unorthodox or dissenting views.
The new NIH Director, Dr. Jay Bhattacharya, was a co-author of the controversial Great Barrington Declaration and his role adds another dimension to this discussion. This declaration—signed by more than 940,000, although only 64,000 of that number are scientists or medical practitioners—challenged prevailing strategies during the COVID-19 pandemic. In March 2020, Bhattacharya also co-authored an opinion piece in The Wall Street Journal entitled “Is the Coronavirus as Deadly as They Say?“, questioning the virus’s presumed mortality rate and the policies enacted to stem its spread. His argument at the time was that projections of 2–4% mortality rate was too high, in part due to limited testing of asymptomatic cases.
Time has borne out much of Bhattacharya’s early critique. Subsequent studies put the mortality rate in the range of 0.02-0.1%, depending on the methodology to calculate the number of COVID-19-related deaths. Regardless of the actual rate, it was certainly far less than was predicted in the earliest days of the pandemic and closer to Bhattacharya’s analysis than the prevailing wisdom presumed. Whether the “focused protection” approach he advocated would have been a better strategy remains a subject of ongoing debate.
However, five years ago, this was a decidedly unorthodox opinion and led to much criticism of the proponents. Eventually the controversy led to the creation of the Great Barrington Declaration. Core to that declaration is the premise that some information related to a science-based policy has been mischaracterized and dismissed. Inaccurate scientific assumptions, which drove policy decisions did have a negative impact on society generally.
One can see a direct connection between this and to the drive for greater transparency and public access to knowledge. The theory is that had ‘real’ science been the focus, better decisions could have been made. Removing the barriers to open science is also aligned with the administration’s broader mission to remove or delegitimize the traditional “gatekeepers”, who have dominated traditional media channels, to include scholarly publishers. However, scholarly publishing’s greatest value lies in its validation and vetting processes. We step away from that at our collective peril.
Rather than focusing on eye-catching news, it is critical that real attention be focused on the bigger issues, which will have broader and more significant impact. The larger goals of the administration are more concerning and are likely to have greater impact than this six-month acceleration of policy implementation. Last fall after the election, during discussions about whether the Nelson memo would be rescinded or not, my somewhat pessimistic response was, “What good will open sharing requirements be if there was no more—or far less—federal research funding?”
The release late last week of the administration’s “skinny budget” is an important political signal of the direction. One has to dig deeper at the budget details sent to Congress beyond the release to find that the budget included among other things a~ 50% cut to NSF, ~5% reduction in NIH science funding, a cut of 16% to OSTI, a 20% cut to NOAA research, a 53% cut to NASA, a 33% cut to NIST, eliminating various cultural and student exchange programs, and a the complete elimination of NEA, NEH, IMLS, and on and on. These cuts will have significant implications on research and education. A Science article from Friday quoted organizational leaders in the research community describing the cuts as “disastrous” and a “surrender of [the USA’s] leadership role”.
Reading through the budget, climate change and “woke” policies are the rationale for massive cuts across the research infrastructure supported by federal funding. Additionally, the political influence in determining where those funds get invested by this administration along with the dismantling of the independent decision-making systems that determine where funding goes should be especially concerning. One might not be surprised that the business connections or geolocation—and political leanings of districts—might have more of an impact on funding decisions than the soundness of a particular research topic.
While the presidential budget proposal is historically never enacted in full, it is a political statement of priorities. In the budget details one can see the scope of the initiative to reshape and cut federal spending. Climate change and “woke” activities are the rationale for massive cuts across the federally-funded research infrastructure far beyond the scope of modest programs that are the rationale for the cuts.
If one recognizes the vast societal benefits that result from investments in basic research and the secondary positive impacts of our research and higher educational systems, these cuts are penny wise but pound foolish. Hopefully, continuing historical precedent, Congressional representatives will reject the proposed cuts, restore the funding, and exert their power of the purse, to ensure that the administration doesn’t impound appropriated funding.
Discussion
10 Thoughts on "Ask The Chefs — The NIH Steps on the Open Access Accelerator"
Thank you for this interesting set of reactions.
A comment about David Crotty’s following quote:
“From the OSTP’s estimate, the average APC for a federally-funded paper is around $4,000. Multiply that by the 146,000 papers published last year with NIH funding (about half of all papers listing federal funding) and you get an annual APC spend (not accounting for inflation) of around $584M.”
This number was also included in the recent Clark and Esposito newsletter (https://www.ce-strategy.com/the-brief/zero-minus-six/).
Perhaps I am wrong, but this seems to be a misreading of the report. According to the report, “The average APC for Fully Open journals was $2,937.05 and for Hybrid journals was $3,999.23.” (p. 19). Moreover, the chart in the report located right above this statement notes, “APCs reflect fees posted in August 2023, which are likely higher than those charged from 2016-2021.” I take this to mean that APC *list* prices were consulted to arrive at this number.
If I am reading this correctly, then Crotty’s statement and extrapolation misapply the $4,000 estimate for hybrid APCs to all journals (the estimate for OA journals is $1,000 less); assume researchers receive no discounts (which are often available); and disregard the fact that publish-and-read contracts often secure de facto prices that are below the list price.
Yes, probably a lot of wiggle room in those numbers. First, it’s worth noting that prices have continued to rise for APCs (particularly for fully OA journals) since 2023. According to https://www.deltathink.com/news-views-open-access-charges-price-increases-back-on-trend, fully-OA APC prices went up 9.5% in 2024 and another 6.5% in 2025. So that would put the $3K low end figure from 2023 at $3498, and hybrid costs increasing 4.2% and 3% in those two years, bringing costs to $4293. And yes, you are right — there is no public information on what people are actually paying. This is also a big problem in studies where estimates are made about how much journals cost using their list prices, which no one pays because most customers gain access through package deals. So here, there’s a lack of information on Transformative Agreement discounts, although to be fair, TAs are still pretty scarce in the US, at least so far until this policy goes into effect. So maybe a better estimate is $510M to $627M, and as emphasized, this is only a small fraction of the costs of the policy, most of which will come from the open data requirements.
One aspect of this that I don’t understand: why does the NIH pay APCs when the article is going to be immediately and freely available on PubMed Central anyway?
What we’ve seen from the Rights Retention Strategy is that many, if not most, journals will not publish an article that is made freely available elsewhere immediately without some form of compensation. The bulk of the work and the expense spent on each paper by the publisher comes in that review process where it gets to the point of acceptance.
In the subscription world, the author pays for those services with a grant of rights — the journal is the only place where this paper can be read (and the journal can try to sell that access to earn money to pay off those expenses). In the OA world, largely the author pays for those services directly through an APC (or in the case of ACS, and ADC). If you posit a world where those expenses are incurred and then no one pays, it makes things unsustainable for the journal, as subscriptions will decline and no APC money is coming in to cover costs.
There are some journals, many of them very high end medical journals that publish a lot of material beyond just research papers (this allows them to keep selling subscriptions because much of their content will not be subject to these restrictions), some journals heavily reliant upon advertising revenue who are concerned that CC BY licensing would harm those channels, and some who ideologically support the idea of Green OA, which will allow immediate posting of the accepted manuscript in PubMed Central at no charge. But this is the minority of journals as far as I can tell, and forbidding the payment of an APC would limit author choice and academic freedom, not to mention deeply disadvantage born-OA journals and publishers.
It’s also pretty hard to publish in many fully OA journals without paying an APC.
You do not have to hand it to Jay Bhattacharya and the Great Barrington Declaration. COVID-19 mortality is not a natural, static number, but one that changes based on mitigation, treatment tools, and medical capacity. Lower than expected mortality isn’t a happy accident. Moreover, disability and worsened health outcomes resulting from COVID infection are ignored.
Let us remember that this is the same administration wanting to increase “measles treatments” instead of promoting sufficient vaccination and discourages reporting demographic differences in studies.
As I’ve noted elsewhere, I wonder how long it will take the rest of the world to start calling measles “the America Virus”.
Not sure whether to laugh or cry so I’ll content myself with staring into the middle distance.
Without getting into a long argument about the epidemiological containment strategies of the pandemic from 5 years ago, my point was simply to recognize that the core scientific point Bhattacharya was making at the time was that mortality wasn’t as high as the worst case projections at the time.
There certainly is evidence that would also support the counter-theory the removal of lock-down strategies did lead to a spike in COVID-related mortality in the summer of 2021. Furthermore, this death toll could have been significantly worse had there not been the protection provided by vaccination, which had been widely distributed in many countries by the time of the spike in cases. Also, there were other dubious claims made in Bhattacharya’s writing at the time, such as the claim that reinfection was very unlikely, which as many people who have contracted COVID several times can attest.
More important than the specifics about the policy choices, is the perception among Bhattacharya and his supporters that these ideas, which were unpopular at the time, turned out to be—in part—correct. This belief is driving a criticism and a distrust of the power structures, even more the notion expertise itself, It is reinforcing the goal of trying to advance a system that undermines those “elite gatekeepers” in scientific communciations.
Also, being correct about one thing, doesn’t translate to being correct about everything or even anything else. “Past performance is no guarantee of future outcomes.”
I appreciate your reply, Todd.
Better people than me (on this very blog) have written about how open science does not necessarily mean that science is properly communicated to a wider audience. I am unsure what, if any, measures taken by scientific and scholarly communities will satisfy Bhattacharya and his supporters or what could turn the tide for trust in expertise.