Terry J. DuBose, M.S., RDMS, FAIUM, University of Arkansas for Medical Sciences, June 19, 1999

Dear Dr. Harold Varmus, NIH Director;

In response to your proposal for Electronic Publications (http://www.nih.gov/welcome/director/ebiomed/ebiomed.htm) and after reading the cautions from the BMJ editors (http://www.bmj.com/cgi/content/full/318/7199/1637), I would like to offer the following thoughts and encouragement.

We in the Ultrasound section of OBGYN.net have been publishing peer reviewed case studies since January, 1998: http://www.obgyn.net/us/us.htm. We have recognized a benefit which your proposal appears to overlook. You state that "These boards could be identical to those that represent current print journals...." and that ",... report would need to be approved by two individuals with appropriate credentials."

An advantage we found is that we are not limited to 2 or 3 peers for review. This limit was convenient for print journals because of the sheer cost and complexity of having more that a few manuscript copies for the reviewers. We post our formal "Case Studies" to a "dead" page (URL) that has no links from anywhere. The URL for this page is emailed to the Editorial Advisory Board (EAB), and every member of the Board has the opportunity to review the case and discuss it in the background via a closed EAB listserv. We are finding that some EAB members have a particular interest or insight that I, as Chair of the EAB, would not have known. In this way I believe we have the potential for a much more rigorous and fair peer review than those of the paper journals were the Editor selects 2-3 "experts" on the subject. This latter method, it is acknowledged, is not always fair and may lead to considerable bias as to what or who gets published... or not.

We have also published "Informal Cases" (http://www.obgyn.net/us/cotm.htm#Informal_Cases). These are similar to your proposed "general repository". Our Informal Cases are simply representative sonographs, with a bit of text describing the circumstances. These images and text are uploaded to the OBGYN.net server via a FTP button and form. All uploads are automatically placed on the server and the message with the URL to the image(s) is forwarded, also automatically, to the EAB only. Once the EAB agrees that the case is appropriate to our site, the Webmaster is notified to post the case to the Informal Cases area for public access. This process usually only takes 24-48 hours or less, if there are no questions about the case. We have had very good cases up in miniutes!

This process has been successful, and I invite you to look at this site and then view some of the discussion about the cases that takes place on our Ultrasound Forum http://forums.obgyn.net/forums/ultrasound/. We have assisted more than a few physicians in remote areas of the world with difficult cases. I know you are busy, so for your convenience, I will include the URL to our first and one of my favorite informal case and the resulting message from the posting physician.

Case: http://www.obgyn.net/us/present/9801/mamula.htm Response: "Yes,it's just as simple as that. Later I attached them to the E-mail and sent them to Terry and Kevin. Now they are there, on the Obgyn-net ultrasound site. I couldn't belive my eyes. I'm still very excited. Can You imagine, I'm sitting here, in my home in Crotia in Europe, several thousands miles away from Terry, and God knows how many miles from all of You on THIS list and we are able to communicate with a speed of light....... amazing technology ..... "Ozren Mamula M.D. Obgyn Rijeka, Croatia. http://forums.obgyn.net/forums/ultrasound/ULTRASOUND.9712/0042.html

If you go to the Forum site and do a search on "Ozren Mamula" you will find that many people all over the world were involved in this case. Which is one of the problems.... perhaps too many opinions. Fortunately, in this case the physician's fears of Amniotic Band Syndrome were quickly alleviated. The cautions of the BMJ editors about a site silting up with bad science is a very real one. However, it is an opportunity to improve scientific communications and world health. We at OBGYN.net are still discussing and tweaking how to best do electronic peer review... do not let my optimism give you the idea that we have not had any problems. But we are not afraid to go there!

Please accept my encouragement to proceed. We must figure out how to best utilize this wonderful media, it is too good to be avoided or suppressed.

Thanks for your time and interest.

Peace, Terry J. DuBose, M.S., RDMS, FAIUM
Chair, Ultrasound Section OBGYN.net
Director, Diagnostic Medical Sonography Program
University of Arkansas for Medical Sciences
Little Rock, AR USA
http://www.uams.edu/chrp/dmshome.htm
http://www.io.com/~dubose/
http://www.obgyn.net/women/advisors/tdubose.htm

William C. Topp, June 18, 1999

as a lifelong member of the scientific establishment you will not realize the simple fact that the substantial majority of effective life sciences and medical libraries are not open to the "public". among those scientists excluded from these libraries are the unemployed and the entrepreneurial individuals who have chosen to leave organized science to start small technology based businesses. these people are totally disenfranchised in an intellectual sense. they have a need to keep current that is just as urgent as that of a practicing traditional research scientist but no economically viable way to do so.

your proposal for e-biomed would, overnight, empower these scientific minorities and allow them to become, once again, productive scientists in a non-traditional career path. i supp

ort your initiative 100% for simply this reason. not every qualified and potentially productive scientist has access to a print library. without such access these scientists are, literally, doomed.

bill topp

Peter H. Byers, M.D., Department of Pathology, University of Washington, June 18, 1999

Dear Dr. Varmus,

Although I am the editor of The American Journal of Human Genetics (for the next few months at least) I am writing in an unofficial capacity and not as representative of the American Society of Human Genetics. You have already heard from Uta Francke in that regard.

When I first heard from Pat Brown last fall about the proposal I was wondering what he (and the rest of the group) was trying to fix and just what was broken. Clearly there are/were a couple of things. The price of some journals is clearly off scale. In our field, The American Journal of Medical Genetics, published by Wiley, is probably the best/worst example. A library subscription to AJMG runs a little over $3000 per year and my back of the envelope calculations (with a little help from Bill Curtis when he was there) suggest that they net close to $2M per year from that journal alone. If that went to subsidize other biomedical publications, I could see the justification, but I doubt that it does. I gather that there are some other journals--some in Chemistry, no doubt others in other fields--that are priced at even higher levels and, no doubt, provide even richer rewards to their private publishers. AJMG has a different rate for the individual subscriber which more nearly mirrors costs--about $450 per year--although even that probably allows profit to the company. Clearly, one solution to these egregious charges is for all libraries to boycott the journals. We have often suggested that individuals at the institutions donate their copies. This may work although I am sure all hell would break loose once it was recognized. Perhaps one benefit of the E-Biomed proposal is to see that it is targetted at such pricing and force a lowering before the journal disappears altogether. Perhaps. In any event, there are clearly some consumer options to dealing with high journal prices other than a large scale reallocation of publication.

A second issue that the proposal may be pointed at is accessibility to the published record for all investigators. At this point I have a little trouble seeing that as an issue. Most papers that we turn down at AJHG do have a way of finding their way into print (I usually hear about it from the author, just to show me that it was a good paper after all) and with the current search an indexing systems they are immediately available. Delay in publication rarely affects their impact or deprives their authors of recognition. I think it is the rare instance in which truly landmark papers are denied either publication or recognition although I think there are few data on the subject. Anecdote soon become anecdata in an equally uncritical fashion, so it is hard to judge.

Then, I guess, the underlying issue is how are we to deal with biomedical publishing in the future and how can we best take advantage of the opportunities of electronic publication. Electronic publications offer a broader venue for data presentation in a "non-linear' format that is not possible in standard print media. It will require some training of the older generation of scientists (like me, I must now regret to add) to provide the "literacy" and comfort with those presentations and to adapt to a new way of looking at data. The linking that can go from text to text quickly is an advantage that means much can be accomplished at the desk without venturing to the library. No doubt there are many more advantages.

These advantages should not come at the price of diminishing the level of review that science gets, and perhaps more importantly that medicine gets, prior to publication. Like democracy, peer review is not perfect, but it really does seem to be the best we have. It provides education both to the reviewer and to the writer. Many are the times that reviewers for AJHG have helped the scientist to clarify thinking, to add the critical experiments, and to prevent publication of major errors that were just not seen. This educational process should not be short-changed simply to allow more to be published. It may be costly in terms of time and personnel if poorly reviewed pieces are brought to the public eye (perhaps no more costly than at present). But the cost could be considerable if medical publications receive less than stringent review. I know that I am not alone with these concerns, and Uta echoed them in her note to you.

I do think that economic factors get short shrift in the proposal, at least in the iterations I have seen. Indeed, they are curiously absent. Electronic publication is not cost free, if papers are to be posted in a fashion consistent with current electronic publication standards. Manuscripts will have to be typeset, the pages formatted, the tables and figures inserted, and the electronic links to the rest of the published world will have to be made. We can provide you with some estimate of these costs as they are done through the University of Chicago Press for us. I suspect that Steve Weiss or Dave Ginsburg from JCI can provide you with similar cost estimates. AJHG is created as an electronic journal at the outset and then a PDF is used to construct the print version. This contrasts with how most journals are run in that the print edition is primary and the electronic version is secondary. Our costs to create the journal (this is copy editing, type setting, figure creation and incorporation, correspondence with authors and corrections of proofs, and creation, posting, compilation of the electronic version costs us about $80 per "print" page. We will publish about 3900 pages this year so the costs will be just under $320,000. In addition, we pay UC Press $70,000 for other costs, some of which would probably not be part of an all electronic package. The paper version, because of paper and mailing costs add about 1.5 times the amount or about $450,000. Our editorial office costs are about $200,000 per year.

AJHG is published on a "daily" basis until the issue if filled at which point it is compiled into the same, or similar format as the print copy. The time to electronic publication from acceptance can be as short as 3 weeks. A slightly more efficient system could bring it down to about 10 working days. We can send you the password (although Francis Collins would be happy to lend his, I am sure) if you want to look. Nature Cell Biology appears to be following this pathway to daily publication and I suspect that most journals will be there within a couple of years.

I can imagine a system in which the author becomes the "publisher" by using templates, but I cannot see that the "cost" disappears--it simply is reallocated. This becomes a "hidden" cost in NIH grants (of course now everything is hidden with the new "modular" grant) but it does need to be taken into account. I have a hard time imagining the NIH assuming the cost for electronic publication because I think the expense allocation would take away from what the NIH does best--funding basic and clinical research--with little to show for it.

I do think that electronic publication will rule the day. We will lose the serendipity of propinquity that one enjoys in reading a written text and experiencing the "aha" of seeing the article right next door. But we will gain a lot, too. Much of this is being implemented by society based and private publications and they provide excellent models for E-biomed. They retain the educational aspect of review, and they vet potentially harmful, dangerous, and uncertain medical issues, both of which are key issues in biomedical/scientific publication. I think that the review process saves us money in the long run to assuring a certain quality of the work. Peer review is quite different than a cursory review for "authentication", but perhaps this is not quite what is meant.

I see no justification to transfer the costs of publication, without income benefit, to the NIH or other governmental source, especially if it is done at the cost of support for investigator initiated research (you don't get to publish more but not support it!).

Clearly, we are well down the road, a road that is being constructed by the extant journals. There seems to be little need to offer more sites, less critically reviewed, for publication. There is a need to provide "connectivity" within the published literature that is more open, and I think this, too is happening. Currently the "literature" starts in 1967, the first electronic data set available in abstract, but not full text, version. It is surprising how much we have had to re-do because of what we don't know from the literature prior to that time. It might be better to work back (I know it is being done) and to work on the linking of present, past, and future publications than to create an even more massive "current and future" literature.

Sincerely yours,
Peter H. Byers, M.D.
Department of Pathology
University of Washington

Eileen Pritchard, Librarian, California Polytechnic State University, June 18, 1999

First let me say that I congratulate you on your proposal. The medical world will welcome it as well as any field allied with it.

I have one additional proposal which has possibly been already made.

In addition, I hope this proposal would suggest an archive of some kind to eventually be availabe to establish "reference links" out to older articles as well as the PubMed abstracts. If the database were available for links, then this would be a great advantage in time and convenience.

Or perhaps I should put this in the form of a question. Does NLM plainly state and encourage authors to make reference links to their abstracts in the literature cited of their recent papers?

Eileen Pritchard (Librarian, California Polytechnic State University)

Steven Tracy, Ph.D., Professor, University of Nebraska Medical Center, June 16, 1999

Re the electronic publishing idea:

Wonderful concept, great idea, something administrations everywhere would think a great idea to bolster their cv's, but in terms of the market, not a great idea.

Relying on the government primarily to fund all the publishing, review, and other costs to maintain such a site depends completely upon a Congress and Senate willing to use public monies for this. We all remember the deleterious effects of the "golden fleece" award by one of the dimmer bulbs in public service some years back and we still have such individuals running loose in both houses, many in leadership positions as well as running for high office. When money gets tight, as it surely will again (then again), this sort of thing will be that which will be curtailed by Congress. One outcome could be increasingly difficult publishing venues (say, one e-journal for any biochemistry, one for microbiology, one for immunology, etc) or a partial/complete loss of places to publish.

The current system, driven by the marketplace, serves us well; journals that do not receive manuscripts go out of business and we are all aware of the best/better/good/not so good journals in our own fields of interest. However, the NIH might wish instead to study the advisability of making available free to subscribers (say, students, faculty, staff of any institution in the US who provide rationale...say, no grant funding, no departmental funds, etc.) any scientific journal that is online, thus permitting strapped individuals fast access to published materials without having to subscribe. Such a privilege could be reviewed annually for appropriateness. This would also have the advantage of keeping some drones at work at the NIH who might otherwise be productive elsewhere....

I would prefer the funds that might be devoted even to the study of this proposal be directed toward the support of basic science, a far more productive use in the short and longer terms.

Steven Tracy, Ph.D.
Professor
Department of Pathology and Microbiology
Enterovirus Research Laboratory
University of Nebraska Medical Center
Omaha, NE

Bill Klemm, Professor, Texas A&M University, June 16, 1999

Dear Dr. Varmus:

We have some software that might fit in with your e-biomed idea. We have a new, simple software system for annotating Web pages. The writer of a document imbeds one line of code and saves the document as an html file and posts it to his Web site. The Note-it! server software on that same site then allows anyone, anywhere in the world to make in-context Web link annotations in that document. The pop-up windows are actually html readers, so the "note" can even be a full-fledged web document.

Thus, a manuscript could be posted on the Web (with the server code in it) and reviewers or other scientists could annotate and debate selected items of the text. This is like the Web discussion groups you suggested should accompany Web journals, but here the discussion is IN CONTEXT and in a fully active browser window.

Check it out at http://www.cvm.tamu.edu/wklemm/contents.htm (follow the Note-it! link).

Please spread the word. Also, if you see a way that NIH could support enhancements, let me know. For one thing, we need a way to accommodate dozens, maybe hundreds, of attached notes anchored to the same point in a document. We also need an addition that lets us display a outline tree of attached notes along with topic statements.

Bill Klemm
Professor, Texas A&M University

Michael Borowitzka, Murdoch University, June 16, 1999

One question I have relating to electronic publishing is - Who is responsible for archiving the paper so that they are available in many (100+) years time? Paper, despite all of tits problems is still a lot cheaper and easier to keep and retrieve for long periods.

If there is no structure for maintaining a true long-term electronic 'library' which can be asily accessed even in 50 years time then the electronic publications must be seen as ephemerata and this would limit their relevance significantly.

Michael A. Borowitzka
School of Biological Sciences & Biotechnology
Murdoch University
Perth, W.A. Australia

Jim Cummins, June 15, 1999

I believe the concept is excellent and timely. I am already heavily involved in on-line editing, proofreading and refereeing for a variety of journals and it is largely foolproof. Of course, the anonymous peer review system should be kept ,and this has privacy and confidentiality implications, but they are not differnt in principle than the use of snailmail and paper.

Olivier C. Wenker, M.D., DEAA, University of Texas, MD Anderson Cancer Center, June 15, 1999

Dear Harold Varmus:

I did read your draft for electronic publications. I think it is very well done and addresses most of the current issues regarding this topic. I assume that E-BIOMED would be indexed by PubMed/IndexMedicus. This is a pressing issue. No quality authors means no quality articles. Most of the quality authors want to see themselves listed in the most important index in biomedicine. This listing will finally help the authors to get their academic recognition and subsequently their promotions. I like the alternative publication routes (without peer-review, lower recognition level). This is quite a new concept giving everyone the chance to be published without damaging the high quality of the peer-reviewed section.

I am producing fully electronic journals since 1996. I went through all the pain of growing such a venture. I am trying to conduct research in this field in order to promote academic recognition for electronic work. I would be very interested to participate in your "Board of Governors".

Let me answer a couple of questions about the Governing Board:

1. How much authority over editorial boards: The Governing Board should provide guidelines about the peer-review process. It should not be directly involved in peer-review. It should make sure that the guidelines are followed and implemented. The editors-in-chief of the e-journals should be part of a subcommittee of the Governing Board and report to the Governing Board.
2. Other functions: The Governing Board should be responsible for acquiring programs that facilitate electronic publishing. It should set rules for the layout and functionality of the e-journals. It should set the guidelines for enhancing the visibility of the web site.

Olivier C. Wenker, M.D., DEAA
Associate Professor of Anesthesiology and Critical Care
The University of Texas, MD Anderson Cancer Center
Houston, Texas
Editor-in-chief, The Internet Journal of Anesthesiology
http://www.ispub.com/journals/ija.htm
President, Internet Scientific Publications LLC
http://www.ispub.com

Bradley Hertz, June 15, 1999

Dear Professor Varmus,

I think the E Biomed program would stimulate discussion of health topics by both professionals and lay people. I would fit in the category of the lay public and professional researchers, as well as possessing a treatise on Neural Computation that I would like to submit to E-Biomed(if the treatise were not considered restricted material).

Thank you.

Sincerely,

Bradley Hertz

Michael D. Lockshin, M.D., Hospital for Special Surgery, June 15, 1999

Dear Dr. Varmus:

The May 5, 1999, Draft Proposal, E-biomed, is an exciting idea, but it is not perfect. Problems lie in the inherent differences of desires and priorities that distinguish private sector from government. Problems also reflect differences between “medicine” and “science” in what is perceived to be valuable science. I write to endorse the concept of E-biomed but to argue against specifics of its proposed implementation. I also argue that E-biomed, purporting to advance “science” publication, threatens medical specialty publication and thus may constrain rather than assist the betterment of medical science. My qualification to speak to this issue is a 20+ year national career in the private sector and 7+ year career in the public sector, including one year as Acting Director of an NIH Institute.

• The first problem is that E-biomed proposes oversight by an NIH-designed or controlled committee (“boards…composed of members of scientific societies or other groups approved by the E-biomed Governing Board”). (Italics in quotations are emphasis added by me.) Currently published journals neither seek nor accept approval of anyone but their own constituencies and are not likely to welcome oversight by a larger body. Priorities of specialty societies, which publish high quality journals, differ from and are more diverse than priorities of a national government agency. Such diversity is well known to NIH: priority scores of grant applications, rated by different study sections, diverge widely because the priorities of study sections differ, not because one study section has a better understanding of science than does another. Priorities of private organizations are likely to diverge even more. The difference between public and private priorities is valuable but is threatened if review committees are answerable to only single governance.

• Second, any connection-or perceived connection-between review for publication and review for funding (an inevitable association if NIH participates in either the selection of reviewers or acceptance for publication in E-biomed) is an inescapable conflict of interest. The fact that NIH study section reviewers currently consider an applicant’s publication record to be an independent validation of his or her worth, and that reviewers for publication consider NIH funding to be independent of their decisions, is a strength of American science that ought not be squandered.

• Third, reviewers serving NIH, whether instructed or not, are conscious of federal policy. Such policy may be explicit, such as that regarding abortion or animal rights, or implicit, such as a given Institute’s interests implied in its issuance of a Request for Applications. It is inevitable that thoughts of NIH policy will influence reviewers acting on behalf of E-biomed.

• Fourth, private sector journals frequently publish articles to promote organization, specialist, or patient-relevant policy. These articles may directly oppose NIH policy; the debate on silicone breast implants is an example. Would the E-biomed Board of Governors respond proactively or politically to submissions that, scientifically strong and pertinent, are also political? Private sector journals have this freedom. Such freedom adds to, and does not detract from, the vigor of scientific debate.

• Fifth, the language of the draft proposal implies preference for “science” over “medicine,” an elitist position that does not fulfill NIH’s mission as an agency for health rather than for science. The language in the draft proposal-“limited number of prime listings…[in contrast to specialty listings] which, while less prestigious”-is an example of the elitism. “…[C]redentials might be…receipt of a research grant from reputable funding source…” is another. Not all elegant science is funded from a “reputable” source, and excellent contributions are found as often in the pages of specialty journals as in the pages of Nature, Science, and Cell. Submission of a proposed publication through “two individuals with appropriate credentials…to be established by the E-biomed Governing Board,” also an elitist policy, may be appropriate for a quasi-governmental institution such as the National Academy of Sciences, but it is inappropriate policy for a specialty publication. The latter is potentially more attentive to public importance of a medical problem. It is more likely than is a general science journal to value an imperfect study, performed under difficult circumstances, that advances knowledge in a medical field. Establishment of reviewers’ credentials by a private sector publication’s board rather than by E-biomed’s Board would be reasonable policy.

Stand, June 15, 1999

Hello,

I have read your proposal for the E-Biomed research, and I find that it would be a very useful tool to disseminate information abroad. I am a person living with AIDS in Jacksonville, Florida. I am hopelessly awaiting a cure that may never come in my remaining life. Not too long ago I lived in the DC area, and worked as a courier for multiple companies. Most of my work led me to the National Library of Medicine to pick up, and return books that some very important people would borrow from time to time. I also did a lot of work in the main NIH buildings, delivering and picking up test results and things of that nature.

The internet has really grown over the years, and it has become a great medium for exchanging information and ideas. Hopefully, with the use of this medium, it will also speed the process of finding a cure before I am counted as one of the many lost to this disease that I have. Many of us are still waiting, and are finding it hard to believe that there will ever be a cure. But it is this type of endeavor that gives us all hope, and gives us the courage to carry on in our daily lives.

Don't get me wrong; I am as healthy as a horse due to the advent of the current drug coctails. But in the 2 years that I have had full-blown AIDS I have been through 4 different coctails, 2 of which failed for me, and the other, besides the Viramune that's working for me now, I could not take (which was the Sustiva) being as I had a very bad reaction to it. With the way it's going now, due to the fact that my disease was not caught during the early stages, I will be running out of options within a few years.

I wish you the best of luck on this project, and hope it runs smoothly for you. I really admire what this project will mean for the many of us that are out here just waiting, and it should also speed up the process in which information about cancers is disseminated.

Thanks,
Stand

Gregory A. Petsko, Brandeis University, June 14, 1999

I need to think about this proposal a bit longer before reaching a conclusion, but my immediate reaction is that not enough attention has been paid to the Law of Unintended Consequences. Science is under siege by those who would lump it together with religion, new-age mysticism, and similar belief systems as just another politically-driven aparatus that serves the interest of its members. These post-modern critics hold that all "truths", including scientific "truth", are equal and that the basis for choosing truth is what social agenda it advances.

Against this tide we have as our best defense what Henry Bauer in his book "Scientific Literacy and the Myth of the Scientific Method" calls the filter of science. An essential step in that filter is getting an idea published in the top-ranked peer-reviewed scientific literature. This process is supposed to be slow and difficult so that bias, egregious errors and so forth are filtered out. Not a perfect process to be sure, but it helps to separate science from those other "belief systems".

I worry that the new system as stated in the proposal may have the unintended consequence of making it appear that science regards all ideas and data as equally valid or valuable; that peer-review may lose its effectiveness as a filter, and that we will each have to do, in fields we are not qualified to do, what referees in the best journals are supposed to do for us.

This does not mean I oppose the proposal - not yet, anyway. It does mean that whenever a sea change of this magnitude is proposed in a working process, I believe it behooves all of us, especially those who propose it, to consider the Law of Unintended Consequences, lest we place ourselves in the position of the rock star who sang, "I Fought the Law and the Law Won."

Greg Petsko
Brandeis University