The Ad Hoc Committee for Advocating Scholarly Communication sponsored an Open Forum to discuss the issues that have surfaced in the past year regarding the NIH Public Access Policy. Since April 7, 2008 all investigators funded by the NIH have been required to submit their final, peer-reviewed manuscripts to PubMed Central (PMC) and these submissions must be publicly available no later than 12 months after the official date of publication. In addition, anyone submitting an application, proposal, or progress report to the NIH must include the PMC or NIH Manuscript Submission reference number when citing applicable articles that arise from their NIH funded research. (NOT-OD-08-033 <http://grants.nih.gov/grants/guide/notice-files/NOT-OD-08-033.html>) Principal investigators and institutions are responsible for full compliance of this law. Libraries and research offices across the country have been establishing new procedures to enable the research community to meet the terms of both the government requirement for public access as well copyright compliance with the journal publishers.
The Open Forum began with a brief introduction given by Hope Barton, Chair of MLA’s Governmental Relations Committee. She announced that the Federal Research Public Access Act (FRPAA, sponsored by Representatives Cornyn and Lieberman in 2006) may be re-introduced to Congress. This bill will extend public access to research from funding from the departments of Agriculture, Commerce, Defense, Education, Energy, Health and Human Services, Homeland Security, and Transportation. <http://www.arl.org/sparc/advocacy/frpaa/index.shtml>
David Gillikin, Chief of the Bibliographic Services Division of the NLM, gave a brief update on the NIH Public Access Policy and indicated that NIH views this as an educational year, when authors and institutions are learning about the process. He discussed some statistics for the PubMed Central Manuscript submission process, including: 56 percent of the submissions are completed within one month of article publication and 93 percent of the submissions are completed within three months. He also shared this graph, indicating the steady increase in the number of submissions to PubMedCentral.
Monthly Aggregate Submission Statistics
Dr. Neil Thakur, point person from the NIH for the Public Access Policy, then spoke by videocast <http://publicaccess.nih.gov/PublicAccess_MLA_Q&A_5-09_small.wmv>.
He made a number of important points that helped clarify issues that have arisen within the manuscript submission process. First, institutions and Principal Investigators (PIs) are responsible for manuscript submission. Second, if errors are found in the article, there are two ways to handle them:
- for the published version of the journal article, the PI should work with the publisher to make corrections–these are usually published as errata
- for errors with the PMC manuscript version, contact the NIHMS Help Desk <http://nihms.nih.gov/db/sub.cgi?page=email&from=home>.
In both cases, the original PMCID# will be retained after corrections.
There are 4 methods for submitting an article to PMC, depending on the specific journal or publisher policy:
A. The published version is submitted by the publisher automatically without author involvement.
B. The published version is submitted by the publisher by arrangement with the author – there is often a cost associated with this method.
C. Author (or designee) deposits manuscript.
D. Publisher deposits manuscript but author must approve version and format it to finish the process.
If a manuscript is submitted through the NIH Manuscript Submission System (NIHMS) and a PMC reference number is not yet available, use the NIH Manuscript Submission System reference number (NIHMSID) instead for progress reports, applications, or proposals. If an article is published in a journal whose publisher deposits all NIH-funded final published articles in PMC without author involvement, or if the author makes arrangements to have a publisher deposit the final, published article in PMC (Methods A or B, respectively), a PMCID might not be assigned until several weeks after publication. During this time, signify compliance with the policy by indicating “PMC Journal – In Process” at the end of a citation to the article. <http://publicaccess.nih.gov/FAQ.htm> Authors are not permitted to state “PMC Journal – In Process” for Methods C or D.
Issues seem to be developing with option D. If an author does not have the NIHMSID, the paper is most likely out of compliance and the worst thing to do in this situation is to say that it is in compliance. This is a serious issue. It is the responsibility of the institution and the PI to get the number. If you are not submitting the manuscript yourself (for example, if you are a co-author affected by methods C or D), you can obtain the NIHMSID by: asking the corresponding author, the publisher, or the NIHMS Help desk via their website link: http://publicaccess.nih.gov/contact.htm. Problems with Method D can be avoided by a discussion among the authors and perhaps also the publisher before the paper is submitted. One of the central pieces of advice given was for NIH-funded authors to consider the following with regard to this new policy when submitting a work to a journal for peer review:
- What submission method will be used?
- What version of the work will be posted to PMC?
- Who will submit the work?
- When will it be submitted?
- Who will approve the submission?
- When can the paper be made available for public viewing on PMC?
NIH reports that often the publisher has indeed submitted the manuscript and an email has been sent to the corresponding author; however, that author has not responded by completing the submission process. NIH cannot and will not negotiate with the publishers. The author entered into an agreement with a third party and NIH is not party to those agreements. The NIH suggests talking to the corresponding authors, talking to the publisher about options, or talking with your institution’s office of sponsored research and say, “I am having trouble meeting the terms and conditions of my award and I need your help to resolve this issue.”
There have been instances when the publisher did not submit the manuscript in a timely fashion. What is the PI to do under those circumstances? The NIH recommends contacting the publisher. In some cases in the past, the solution was to cancel Method D and instead use Method C, in which the author or designee handles the submission completely. In this case, the publisher had to grant the author permission to make the deposit. This situation is occurring less often.
Another misunderstanding, about identifying PMCIDs on grant applications and reports, was discussed. You do not have to include the PMCID for papers that are not authored by one of the PIs or arise from one of their awards. PIs must identify all applicable articles that arose from their NIH award and must identify them with their PMCIDs . An application with multiple PIs must identify PMCIDs from all PIs. However, authors do not have to identify PMCIDs for other articles that they cite in a bibliography or in references.
Following the videocast by Dr. Thakur, Ms. Barton and Mr. Gillikin were joined by others: Karen Albert, Chair, MLA Ad Hoc Committee for Advocating Scholarly Communication (and a Chapter member); Senior Director for Education and Information Services at Fox Chase Cancer Center; Gretchen Arnold, Director and Associate Dean, University of Virginia, Claude Moore Health Sciences Library; and Pat Thibodeau, Associate Dean for Library Services, Duke Medical Center Library. The panel answered questions from the audience:
- Pat Thibodeau tried to express the frustration of PIs and librarians with publishers not submitting in a timely fashion. However, David Gillikin was adamant that NIH could not intercede. He suggested that most of the problems encountered during the submission process are due to authors not following up with the two-step author approval process after publisher submission associated with Method D.
- One of the panelists reminded researchers to submit non-Medline indexed journal articles to PMC. An example was given of an NIH-funded, peer reviewed research article published in an Irish folklore publication.
- The PMID-to-PMCID converter functionality has been expanded: <http://www.ncbi.nlm.nih.gov/sites/pmctopmid>
- There was a discussion about comparing NIH-funded articles from an institution as identified in a Medline search with PubMed Central submissions from the same institution to determine compliance. David Gillikin indicated that since not all NIH-funded papers are included in PubMed, there would be no way to accurately determine compliance. Also, it would not be possible to identify those manuscripts “in process” and those embargoed.
Related Resource:
NLM Theater presentations offered at the Exhibit Booth at MLA 2009
”PubMed Central and NIH Public Access Update”
http://www.nlm.nih.gov/bsd/dist_edu.html#m
Marie FitzSimmons, MS, AHIP
Assistant Librarian, Reference
George T. Harrell Health Sciences Library
Hershey Medical Center
Penn State University College of Medicine
mfitzsimmons@hmc.psu.edu
Thank you so much for this summary. I have just been helping with citations returned from a progress report and this helps with all the issues that came up.