The NIH Public Access Policy: CHEST’s Position and Concerns Regarding the Responsibilities of the NIH Principal Investigator FREE TO VIEW

This editorial is intended to state the position of CHEST and its publisher/parent society, the American College of Chest Physicians (ACCP), in response to the NIH Public Access Policy. This NIH policy will affect authors whose NIH-funded research has been accepted for publication in CHEST.

On February 3, 2005, the NIH issued a new policy (NOT-OD-05-022) to enhance public access to publications resulting from NIH-funded research.¹² Beginning May 2, 2005, the policy requests that the principal investigators (PIs) publishing wholly or partially funded NIH research that has been accepted for publication in peer-review journals voluntarily submit an electronic form of the accepted manuscript to the NIH National Library of Medicine PubMed Central for archiving within 12 months of the date of journal acceptance.³ These articles will comprise an archive of publicly accessible NIH-funded research.

The on-line process for PIs to follow for depositing their manuscript has a number of steps; the process is still under revision at the date of this writing (April 20, 2005). As it stands, the NIH Public Access Policy implementation plan currently states on its Web site that “As additional details and instructions on the use of the PubMed Central manuscript submission system become available, these will be posted on the National Library of Medicine PubMed Central website.”³ Because the May 2, 2005, deadline is approaching fast, CHEST suggests that PIs of NIH-funded studies be sure that they thoroughly familiarize themselves with the NIH requirements and instructions.³

The Editors-in-Chief and the ACCP support the concept of public access; we already make all CHEST content freely available online 12 months after publication at www.chestjournal.org and participate in programs that give immediate free access to the world’s least financially endowed countries. We had hoped that the NIH policy would provide the publishers with the ability to complete the submission process on behalf of the PIs. Unfortunately, that is not the case. As a result, we believe the NIH policy in its current form has a number of pitfalls. We wish to note several key aspects of the NIH policy and procedures, share concerns we have about them, and provide suggestions on how PIs can avoid potential problems:

The NIH policy dictates that the PI is responsible for the final approval of the materials submitted to the PubMed Central repository. CHEST and other publishers have long believed that as the owners of the copyrighted material and holders of the final, peer-reviewed, copyedited, published article files, it would be easiest for all involved if the publishers could oversee the process from start to finish. Unfortunately, this is not the case, and the NIH policy firmly puts the onus on the PI. Although the system will allow for certain steps to be completed by an authorized third party, the final approval must be made by the PI. Therefore, we suggest that the author or PI should complete the entire process to avoid confusion and mistakes in the deposit that could be caused if one entity were to do part of the process, and another (the PI) finished it.

Because the PI, during the deposit process, must stipulate when the manuscript will be made freely available to the public on PubMed Central, there is the potential for violation of the copyright agreement between the PI and coauthors, and journals. CHEST wants to alert PIs to this potential pitfall.

Because different journals may have dissimilar time frames for making content available online for free (eg, CHEST makes its content freely available 12 months after publication), PIs may inadvertently choose a time frame that makes the article free on the NIH PubMed Central site sooner than a journal’s policy allows. This would be in violation of the journal’s copyright policies, putting the PI in an awkward position. To avoid such a situation, PIs must be knowledgeable about the journal’s copyright and free-access policies before performing the NIH deposit, and make themselves familiar with the CHEST copyright assignment form on the journal Web site. CHEST’s policies will be clearly communicated in its copyright assignment form on its Web site and in its Instructions to Authors. If authors abide by CHEST’s policy of making articles free online in the NIH repository 12 months after publication, this will not only satisfy CHEST’s copyright agreement but also the NIH policy. The question of why NIH does not simply link to the publishers’ Web sites where the content already exists in electronic form, rather than create a cumbersome and duplicative process is one that NIH has yet to answer satisfactorily.

Because the NIH policy instructs the PIs to deposit the accepted, peer-reviewed manuscript upon its acceptance by the journal, there is a risk that the PI will deposit an inaccurate version of the manuscript. As a clinical journal, CHEST publishes articles about therapeutic interventions that have a direct impact on patient care. Even though an article is accepted after having gone through a rigorous peer-review process, it still undergoes thorough copyediting procedures to correct errors and typographical mistakes, and clarify ambiguous language. Critics of the NIH plan argue that by urging PIs to submit the article immediately after acceptance but before copyediting is performed, the deposited NIH manuscripts might contain important inaccuracies. Because NIH is intending this content to be freely available to the general public, we are concerned that providing the scientific and general public with unedited article files may potentially lead to serious adverse consequences. Therefore, a number of journals, including CHEST, have modified their copyright statements and added disclaimers so that journals are not held responsible for uncorrected deposited content. Our revised statement and disclaimer appear in our Instructions to Authors as well as on the journal Web site. They will serve to make clear that the version submitted to NIH may not be the final version of the manuscript and that the author’s version of the manuscript has not gone through rigorous copyediting and should not be relied upon as being the definitive version of the article.

Because the NIH policy is voluntary, we believe that a better alternative would be for the PI to wait until the final, edited manuscript is published by the journal, and then submit that file to the NIH at that time. This brief delay will not prevent PIs from meeting the requirement of the NIH and CHEST of depositing the accepted manuscript within 12 months of its acceptance. CHEST is implementing a policy that we will provide the final, edited, PDF file to the PI if PIs request it. We hope PIs will consider this option because it is in the best interest of everyone. The most accurate version of the article is the one that should be made available to the public.

It is important to reiterate that the NIH Public Access Policy is a voluntary program that will only apply to manuscripts that are wholly or partially funded by NIH grants, and only to those research articles that have been accepted for publication in peer-reviewed journals such as CHEST. The on-line process has still not been completely finalized as of April 20, 2005. CHEST strongly suggests that authors of NIH-funded studies become very familiar with the NIH requirements and instructions.³CHEST understands that for many PIs, this process will be cumbersome and confusing. Although NIH has put the burden of responsibility on the PIs, we hope that at some point in the future the NIH will allow the publishers to facilitate the complete submission process, thereby ensuring accuracy and integrity of the data deposited. In the meantime, we hope PIs who publish in CHEST will consider the option to deposit the final, edited version of the manuscript rather than the initial accepted author’s manuscript.