• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

About FDA

  • Print
  • Share
  • E-mail

FDA and the Internet: Advertising and Promotion of Medical Products - Discussion Group 4

   

Contents

VOLUME I: Remarks   WWW 101  Group 1 Group 2   Presentations    Group 3
VOLUME II: Remarks Group 4   Presentations   Group 5
 

 

October 17, 1996

DISCUSSION GROUP 4

Website Links

Moderator: George Lundberg, M.D.

 

DISCUSSANTS

Tom Buford, Intermedics Orthopedics, Inc.
Jack Calfee, American Enterprise Institute
Dave Vance, Glaxo Wellcome, Inc.
Becky Cleavenger, AMGEN
Mark Deitch, Bioscience Communications
Craig Douglas, Frontier Media Group
Allen Douma, Health Response Ability
Edward Fotsch, Net Source Communications
Lawrence Ganslaw, American College of Cardiology
Linda Golodner, National Consumer League
Mark Grebenau, Sandoz
Susan Hamet, Knoll Pharmaceutical
Tom Jones, Family Internet
Kiyoshi Kuromiya, Critical Path AIDS Project
Karen Lancaster, PR Newswire
Paul Leuhr, Federal Trade Commission
John Mack, VirSci Corp
Jamie Marks, Body Health Resources Corporation
Matthew Naythons, NetHealth
Sara Stein, Stanford Department of Psychiatry
 
   FDA
Ilisa Bernstein, Pharm.D., J.D.
Melissa M. Moncavage, M.P.H.
Patrick O'Brien, Pharm.D., J.D.
William V. Purvis
Edward L. Spenser, M.S., D.V.M.
Robert Temple, M.D.
Deborah Wolf

 

Discussion Group 4

Website Links

DR. NIGHTINGALE: Now, I would like to turn to Dr. Lundberg to moderate the session. Let me first introduce him and then we will have the FDA panel introduce themselves.

Dr. Lundberg has worked in tropical medicine in Central America and forensic medicine in New York, Sweden and England. His major professional interests are toxicology, violence, communication, physician behavior, strategic management and health-system reform.

He is a past President of the American Society of Clinical Pathologists. He has been at the American Medical Association since 1982 where he is Editor in Chief of Scientific Information and Multimedia with editorial responsibility for its 54 medical journals, the American Medical News, and various electronic and consumer products, and the Editor of the Journal of the American Medical Association.

As a frequent lecturer, radio and television host and guest and a member of the Institute of Medicine of the National Academy of Sciences, Dr. Lundberg holds academic appointments as a Professor at Northwestern and Harvard.

We are very fortunate that he has agreed to be the moderator today. I know that he has been personally and deeply very much involved in this issue and is very enthusiastic about this technology. We are very much looking forward to this discussion group, Dr. Lundberg.

While that is occurring, why don't we start with the FDA introductions starting with Dr. Wolf.

DR. WOLF: I am Deborah Wolf. I am Regulatory Counsel in the Promotion and Advertising Policy in the Center for Devices and Radiological Health.

DR. TEMPLE: I am Bob Temple. I am Associate Director for Medical Policy in CDER and DDMAC reports to my office.

DR. LUNDBERG: Before we go further, I want to ask the people in the back of room if they can hear clearly. Is the sound system working? They say they can. Very good. Next?

DR. SPENSER: I am Ed Spenser with the Division of Surveillance, Center for Vet Medical.

MR. PURVIS: Good morning. I am Bill Purvis, the Director of the CBER Advertising and Promotional Labeling Staff.

DR. O'BRIEN: I am Patrick O'Brien and I am Regulatory Counsel with CDER's Division of Drug Marketing.

MS. MONCAVAGE: I am Melissa Moncavage. I am a public-health advisor in CDER's Division of Drug Marketing and Advertising.

DR. BERNSTEIN: I am Ilisa Bernstein. I am a Senior Science Policy Advisor in the Office of Policy and the Office of the Commissioner.

DR. NIGHTINGALE: Stuart Nightingale, Office of Health Affairs.

DR. LUNDBERG: Thank you very much, FDA panelists. We will come to our other panelists momentarily.

Dr. Nightingale, Dr. Bernstein, fellow panelists, fellow journalists, fellow physicians, fellow communicators, fellow people interested in this new technology, good morning.

Damn it, I said, "Good morning."

AUDIENCE: "Good morning."

DR. LUNDBERG: Interactive. That is what the day is supposed to be all about. I'm sure it will be, but I just wanted to get it started right.

I consider it an honor and a privilege today to have the opportunity to moderate this important panel on Internet website linkages. I want to thank the Food and Drug Administration for calling this meeting. I think we are all going to have a lot of fun.

Many of you were here yesterday. I wasn't. I was in Chicago working, but I gather that the day was extremely interesting and sets a wonderful tone for today and, perhaps, a hard act for this group to follow. But I am confident that we will exceed yesterday's discussion as we go on through the morning.

We have from now until 10:40 to address the question of Internet website linkages. We will be guided by a set of questions that was developed by the FDA staff. I am told that you all have copies of the questions so you will have them in front of you.

Everyone in the front of the room will be given an opportunity to speak. Some may speak longer and more frequently than others. That is the way of the world. But it is one of my jobs to try to keep it as balanced as reasonable.

I will attempt to keep it orderly and to ask people to limit their comments to reasonable ones that are not particularly lengthy. I will cut you off if you get into major speechmaking and it goes beyond time.

My main job is to be a combination of time keeper--although I understand we have another time keeper over there who has a little box that will give us lights and stuff in case anything goes on too long. My job is also to be stimulant, referee and traffic cop.

Just as in the National Football League, we will not have the benefit of instant replay, but the entire program is on the record. Just as in Baltimore baseball, spitting on the umpire will not be tolerated. We have a penalty box, if I can mix the metaphors of sports. It is behind the screen over there. For anyone who spits on the umpire or otherwise behaves badly, you have to stand behind the screen.

The basic temporal structure is as follows; self-introduction of all participants which we will do shortly. I have given myself three minutes for an introductory framing. Do you call a three-minute speech a speech? Whatever. Three minutes of introduction.

We will go over the questions and one panelist has already asked that we invert the questions and start with the last one first, which I like because I normally read an article from the back to the front. So I resonate with that person. We probably will do the last question first.

Each panelist will have an opportunity to respond to the questions and, again, the aggressive ones who raise up their hand and get the attention of the moderator will get first shot. Then somewhere halfway in the middle of this thing, they are going to turn on a red light and we are going to have an opportunity for FDA individuals to ask questions and interact and we can go back and forth for a while.

Then we will get on with the rest of the panel as we understand it. At the end of that part, we will also have an opportunity to interact back and forth across this more than a semicircle, this three-quarter circle, saving the last 15 minutes for statements and questions from anyone in the assembled group. There are now two microphones there which is even better than one.

We will ask people to address the mikes, to get in line and then they will be given an opportunity following the same kinds of rules we have to make their statements and to ask questions after their affiliations and names have been entered into the record.

Now, proceeding clockwise from the moderator sees it, why don't we start right there. There are microphones. You must speak into them. There is not one for each person, but there are enough to share nicely. Who are you? Who do you work for? Why are you here?

MR. BUFORD: Tom Buford, Intermedics Orthopedics. I am in marketing. We are creating a site and we are here to learn with everybody else.

DR. LUNDBERG: I thought you were going to be here to teach.

MR. BUFORD: We are teaching and learning.

MR. CALFEE: I am Jack Calfee. I am a resident scholar at the American Enterprise Institute. I am here because I love to watch the FDA struggle with these problems. I think this is the neatest problem that has come up yet.

MR. VANCE: Dave Vance from Glaxo Wellcome, attorney in the HIV-oncology area. I am here as a kind of patient activist inside the company trying to work to make sure we legally get the most information out to the people who need it.

MS. CLEAVENGER: Becky Cleavenger, AMGEN, Regulatory Affairs.

DR. LUNDBERG: Why are you here, Becky?

MS. CLEAVENGER: Just to work with FDA, I guess, in developing this guideline.

MR. DEITCH: Mark Deitch, Bioscience Communications. We develop professional education programs, and weblinks are the limiting factor in most programs that we develop for the Internet.

MR. DOUGLAS: I am Craig Douglas, Creative Director, Frontier Media Group. I am here because we specialize in creating interaction. Obviously, the rules are going to be set and we have to live by those rules.

DR. DOUMA: My name is Allen Douma. I am Medical Director for Health Response Ability Systems. About three-and-a-half, four, years ago, we actually launched the Health and Medical Channel, not then a channel, on America Online and we are still managing about 70 percent of the activity of America Online's health channel.

DR. FOTSCH: Good morning. My name is Ed Fotsch. I am a physician and the Director of Interactive Health at Net Source where we provide Internet and Intranet systems for The College of Surgeons, The College of Cardiology, a number of manufacturers. I am here because of my intense personal interest in the medium and also because several of our clients have an equally intense interest.

MR. GANSLAW: My name is Lawrence Ganslaw. I am Associate Director of Marketing for the American College of Cardiology and am really here to learn about how subspecialty societies can collaborate with industry without obviously breaking any rules.

MS. GOLODNER: I am Linda Golodner. I am President of the National Consumers League. We run the Internet Fraudwatch and I am here because we are trying to represent the consumer interest on advertising and promotion issues.

DR. GREBENAU: I am Mark Grebenau. I am with Medical Services at Sandoz Pharmaceuticals. I am here because the Internet is here and it is not going to go away.

MS. HAMET: I am Susan Hamet with Knoll Pharmaceuticals in Regulatory Affairs. We are here to learn and to see how we can use this medium to the best of our ability and to try to see where we are at policywise.

MR. JONES: I am Tom Jones, Vice President of Marketing and Health directing the content for Family Health which is a domain on Family Internet. Family Internet is a huge webmall that now has 70,000 pages up. We are here because we want to learn about the regulations because we already have clients that are coming to us and asking these questions.

MR. KUROMIYA: I am Kiyoshi Kuromiya of Critical Path AIDS Project. I am here wearing three hats. I am a consumer of Internet treatment information as a person with AIDS. I am also a content publisher on the Internet. I publish a newsletter, Critical Path AIDS Project. And we have a website, critpath.org, which reaches approximately 10,000 individuals with AIDS each day.

I am also an Internet service provider. We host about 30 websites for grassroots AIDS organizations and other community organizations.

MS. LANCASTER: Good morning. I am Karen Lancaster. I am with the Washington Bureau of PR Newswire. Nationally, we work to distribute full-text news releases from the vast majority of pharmaceutical companies. We are here to share the needs that they have expressed to us for their communications needs and, hopefully, share some knowledge and also learn.

MR. LEUHR: My name is Paul Leuhr. I am an attorney with the Federal Trade Commission, specifically in the area of consumer protection. I work in issues related to cross-border fraud and Internet fraud. I am here because the FDA wanted another government official here to take some heat.

MR. MACK: I am John Mack, President of VirSci Corporation and Webmaster of Pharm Infonet, an independent drug information site on the World Wide Web. I am here to learn more about the issues facing pharmaceutical advertisers on independent sites such as ours.

MR. MARKS: I am Jamie Marks, President of Body Health Resources Corporation. We publish a major HIV/AIDS website known as The Body at thebody.com which carries materials from 35 of America's top AIDS organizations, universities, professional societies, all related to various dimensions of HIV/AIDS. We are here because it is integrally important to our financial model to know what the pharmaceutical industry may or may not be able to do with us.

DR. NAYTHONS: My name is Matthew Naythons. I am President of NetHealth which is a San-Francisco-based Internet developer and publisher. I am here because we will have to live by the rules that come out of this meeting and we want to have a hand in forming them.

DR. STEIN: My name is Sara Stein. I am a clinical-research psychiatrist at Stanford University. At Stanford, I am also a contributor to the Medical Website and also am a member of the Library Information Systems Committee. I am also a consultant to ChemTrack, Incorporated, where I developed the Telephone-Counselor Hotline and Training Program for delivering HIV results by telephone.

DR. LUNDBERG: Thank you all very much. I wonder if the assembled group was as impressed as I was with the comments of people saying they are here to learn as opposed to here to influence or here to change or here to teach or whatever. I think that is wonderful. I think basically all of us are here to learn. I am here basically to learn and I appreciate that.

Also, I find it interesting that we have Austin, Texas on the far end and we have two people from the Bay Area who happen to be next to each other. Everybody else, I could tell, or virtually everybody else, is from within 100 miles of where we are sitting, or standing, today which I think is interesting, also.

Introductory comments. The Internet and the World Wide Web are among the most recent iterations of that great old beast of burden, the computer; I say "old" beast of burden, the computer.

The computer and all of its permutations are tools, mechanical and electrical agents to be used by humans to accomplish tasks. The Internet is the medium; it is not the message. The message is the message. The Internet is a means to an end. It is not an end.

The computer was made for man, not man or woman for the computer. In this new electronic age of communication, we are no longer in the publishing business; we are in the information business. The Internet has the potential of becoming the world's largest garbage dump of misinformation. Some believe that it is rapidly realizing that potential.

I see the Internet and the World Wide Web as a massive combination of numerous old methods of communication; the ethics, laws and expectations of all of the old methods have been worked out over centuries, even millennia. We are now trying to work out the ethics, law, expectations and good moral practices of communication on the Internet as a crash course at breakneck speed.

Some of the old rules will apply, to be sure. Some will not. Breakneck speed may, in fact, break some necks in the process. I see use of the Internet as a combination of personal conversation, the telephone, conference calls, sewing circles, bar-room boasting, still photographs, the telegraph, the newspaper, television, films, book publishing and even others.

Some of each of these media communication rules apply and some of them don't in an amazingly complex gemisch. On the Internet, anyone can be a publisher. It is the ultimate vanity press. Be your own writer, editor and publisher, and fake all the information including your identification, if you want. Who is to know? Who owns copyright? Who owns the intellectual property? Who knows? Who cares? A lot of people probably care.

In the electronic information age, we think content will be king and trusted brand-name identification will carry over into this medium for those who don't want to waste their time or to be fed uncertain information.

If you value your site, will you voluntarily link to another. Does that linkage convey equal value? Are you stating by that linkage that you create that the viewer may trust the new site as if it were your own? Is that the image conveyed to the user even if you don't intend it?

Who carries whom on which bootstraps; which linker, or shall we say, linkee? How about disclosures, disclaimers or information firewalls? Who is responsible for what? Are the names even there? Is the owner's name there? Is there an address other than an Internet address and is it fake or real?

Caveat emptor, "let the buyer beware," has become caveat lector, "let the reader beware," or caveat viewor, "let the viewer beware." When we are dealing with medical information, as we are today, primarily, it becomes caveat aeger, "let the patient beware," because, of course, it is mostly patients who use this new permutation of the beast of burden called the Internet.

You all have the questions. We will read them prior to discussing them but, as requested by a panelist, we will invert them. We will start with Question No. 3. "What is the state of technology today with respect to website linkages; e.g., ability to link, ability to identify who is linking to your website, ability to prevent others from linking to your website, ability to control where a link is to your website, et cetera." The control issue; not the volition of it, but the technology of it.

Who on our stellar and interesting panel wishes to start by answering that fairly simple question? Who has the right answer? Are there any other panelist in the audience who would like to replace this panel?

DR. NAYTHONS: I was hoping that someone from the technical side would answer this one. I have asked my technical people and they tell me that essentially you can stop people coming in and you can obviously link where you want to go yourself.

There are three levels of linking in from the most difficult which is encrypting your site, to password protection to identifying and blocking by something called an IP protocol which is the address of the people coming in who comes in to you. If you police your site on a regular basis, you can block people from coming in.

I don't have the technical answer to control where a link is to your website.

DR. LUNDBERG: We have our opening statement. Silence means complicity. Does anyone have another view?

DR. GREBENAU: I would also like to hear a technical person, or at least a more technical person, answer this question. I have been given to understand that whenever someone visits a website, not only can that person's reverse domain be determined, knowing where that person came from originally, but, if I am not mistaken, and I may very well be, it is possible to determine the path that that person followed in order to get to your website, and that one can just go backwards along the stepping stone, so to speak.

One can find out where linkages have been created since these linkages could conceivably be created without knowledge of the person being linked.

DR. LUNDBERG: Remember, we are talking technology and possibilities here but as you talk technology and possibilities, also think practicality and what is sensible and feasible to do in addition to what is technically possible.

DR. DOUMA: Just to respond to that thought. What the technical people tell me is, in fact, you can't trace somebody's path. HTML, the Internet programming system, basically is an event-driven thing and it is a single event. Every time you go out there, you create a new event and that erases everything that came before except if you are what is called cashing it on your own computer or you are cashing it in a bigger computer in the sky through your on-ramp, whether that is a commercial service or not.

So, no; you can't trace back.

With regard to identification of who is coming in, you certainly can use the IP protocols and block a number of things. Typically, it would be a domain name or a person that is persona non grata for a variety of reasons.

The other thing is you could block using other pieces of information. There are typically about five or six technical pieces of information that are standardly driven or delivered through the browser software that you are using. And this is sort of universal stuff. It has to do with hardware that you are using. It has to do with browser using, et cetera.

There is no personal information that that carries. I am not sure who makes up these words, but that packet of information is now getting the nomenclature of a cookie. The cookie carries basically some standard computer language stuff but also the new browsers are looking at the ability for your cookie to carry personal information.

However, that personal information is only what you choose to put in the cookie. And that is the marketer's dream, that everyone who comes to visit their site has a 25-question questionnaire filled out so they know what the demographics are. If you would like to do that, you are able to.

MR. LEUHR: I have had to do a fair amount of training of our attorneys and investigators and so I get many of these technical questions. By default, I have had to become a quasi-techie. On the issue of cookies, talking with some of the people from Netscape and IBM, they tell me that there is the capacity, many times, to trace back to the most recent link that a person came from, even if it is not from your own home site.

In other words, they can trace back one link. But they said, in practical terms, what they think is happening out in the market is not that people are gathering what they call clickstreams where you have been surfing the entire time but, instead, there are some consortia that are being developed where a whole bunch of different website developers, perhaps even in the medical community, will exchange information and tell each other, "Yes; Paul Leuhr from the Federal Trade Commission visited my site three times last week."

By putting all their information together, they may be able to put together a profile of an individual-- i.e., me--from the FTC and where I have been surfing, at least among the consortia members.

DR. LUNDBERG: This is the right and this is the left. The FDA is on the left, unless you are in the audience and I guess that is the right and that is the left.

MR. MARKS: At The Body, we are very concerned with protecting the privacy of those who use our site and we have been extremely careful about not gathering and recording their Internet protocol addresses which could reveal their identifies.

A subscriber to America Online, for example, arrives at our site and it is clear that someone has hit from America Online, and that doesn't reveal the individual's identity. But if someone comes from a smaller domain, or a corporate domain, or a domain that would come closer to disclosing who that individual is, raises very serious questions about someone who thinks he or she may be getting private information about HIV risks and transmission and treatment, who might unwarily be divulging that identity and marketers could then pursue that person if they were unscrupulous.

So we have been extremely careful in our policies of not forcing people to register and not logging IP addresses.

DR. LUNDBERG: But we have heard here--I think you said that even if you haven't registered or logged an ID of any kind, you can be tracked as to who you are. You left a fingerprint of some kind; is that what you said?

MR. LEUHR: Most of the time, I think there has to be an active effort by the website manager to what they call "drop a cookie" on a person. For those of you who look at the Internet screen and want to know what a cookie looks like, oftentimes, it will show up as cgi-bin somewhere in the computer language.

I think, for the most part, people use cookies to identify if a person has been to their site before, once before, twice before. But there usually does have to be an active step taken to drop that cookie, so to speak.

MR. MARKS: Also, the savvy browser can knock out the personal information in his or her version of Netscape et cetera so that that is not captured.

MR. VANCE: As we said yesterday, the technology is changing so quickly to give any hacker, let's call it, the challenge of becoming invisible--there was reference yesterday to the site in Finland that, apparently, as I call it, is a cling-on cloaking device where you can go through there and come out anonymous.

There will be people who will figure out ways to make themselves invisible. Obviously, they have to have either a privacy interest which is one consideration we have to balance or they have got some possible malevolent intent and they don't want people to trace back.

I think what we have got to balance is how do you protect the privacy interest of the AIDS patient, the herpes patient, versus the polluters. We haven't talked yet about polluters.

On the linking issue, we linked to a medical vendor with a very reputable site that had a chat room totally independent of us and discovered, on looking at it last week, that they had been polluted and had not been doing maintenance, at least on that day. They have pornography. One site we went to had everything from raising proper children of a certain religious persuasion, information on the Simpsons, information on various sexual activities and information on aliens.

Some of that disappeared on that site from Friday to Monday because they went in and cleaned it up. But actually the Simpsons from October 6th was replaced by the Simpsons on October 14th. So part of this blocking that is being talked about is this person whose name is on it and very easy to identify because there is not anything bad on the Simpsons could be blocked easily.

Some of the other ones, it wasn't clear you could block as easily. But this pollution is a factor.

I think we need to look at public parks as kind of a metaphor for this. You may have the soccer team meet in a public park in the morning, followed by a family reunion, followed by the KKK meeting which is perfectly fine and lawful but the police would watch it, and god knows what goes on in the bushes at night. You may want to watch that differently, but we are not going to close public parks because some people may misuse them.

DR. LUNDBERG: There is a lot of stuff in that comment.

MR. VANCE: I am trying to raise interest.

DR. LUNDBERG: All good and all interesting.

DR. FOTSCH: Two quick comments. One was that yesterday there was quite a bit of discussion vis-a-vis patient security or medical-record security, and also discussion with regard to anonymity.

I just want to make the point that from and end-user/patient standpoint, those were frequently one and the same. The anonymity that sort of got a bad rap yesterday from the patient standpoint is, in fact, protection of his privacy.

Second, as a follow up, we are really looking at the technology in a snapshot kind of a way. It is moving very quickly. There are technical forces that are being asked to devise a mechanism by which you can identify visitors. Frequently, that push is coming from marketing departments, as an example.

There are also people that are saying, "We don't want to be identified." I think that from a regulatory standpoint, you have to say, "Is it going to be consistently possible to identify users?" I think the answer is no, not consistently. You will not consistently be able to do that. Whether you can in a snapshot technology or not, it seems to me, begs the point of what you can assume from a regulatory standpoint.

DR. LUNDBERG: We are a little off the subject. We are into users here. We are theoretically supposed to be talking about you having a website linking to another website, another website linking to your website with or without your permission, with or without your knowledge.

I would like to focus the discussion back a little more on that although the user part is enormously interesting as well.

DR. GREBENAU: Dr. Lundberg, I think you just said exactly what I was going to say. There are two issues here; identifying users and identifying the link from which they came. If we regard a website as the electronic equivalent of a mall, we are not so interested as to who it is coming through the door as through which door that person came and from where, because of possible linkages, let's say, to sites which talk about unapproved indications for a drug which we manufacture.

DR. LUNDBERG: So you are saying that if somebody got there, they came through a link and that is the way to trace it down?

DR. GREBENAU: I am saying that that is the issue which we should be addressing, whether it is possible or practical. And it may come down to an issue of practicality, of identifying through which door or through which portal people are coming and whether the linkage or the association between the portal through which they came and the site that one is maintaining constitutes any kind of a regulatory problem.

DR. LUNDBERG: Before we go further, and I am not forgetting you, how many on the panel believe that you, in control of your website, can determine with whom you link, that you send people to other websites from your website and you can control that? How many of you believe that?

DR. DOUMA: I can control where they are going from me?

[Show of hands.]

DR. LUNDBERG: You all believe that. I believe that. I wondered if anybody disbelieved that.

How many of you believe that you, on your website, can control and prevent or enhance other websites linking to your website?

[Show of hands.]

DR. LUNDBERG: Only a few hands. Dr. Douma, explain that.

DR. DOUMA: What I can do is, in fact, block out anybody coming from a particular domain.

DR. LUNDBERG: But you have to know that domain to block them out specifically.

DR. DOUMA: That's correct.

DR. LUNDBERG: What if you don't know that and haven't instructed your computer?

DR. DOUMA: If I don't know the domain, and, in fact, right now, and this goes back to what Paul was saying, the cookies that are being developed, some of them will, in fact, grab where you last were and make that part of the information package that you send.

DR. LUNDBERG: This is coming. This is not today.

DR. DOUMA: No; it is today.

DR. LUNDBERG: This is today.

DR. DOUMA: It is today. It is doable today. Now, the other thing is that every cookie developer that I have ever talked to says that the only way that is going to be a marketable tool or a marketable piece of software is if the user has the ability to change whether or not it grabs where you last were.

It is not going to be built into the system because most users are not going to want to give up that kind of information. Therefore, even though it is technically doable today, it is practical to do, it is not a big deal, the vast majority of time, you will not know where somebody has linked from.

If you do, then you can obviously block them. But I think it is not even worth discussing because 90 percent of the time, people aren't going to want to tell you.

DR. LUNDBERG: I don't know where you get your 90 percent but, from my personal view, privacy is one of the major marketing tools of the Internet, the ability for somebody to go somewhere, listen, talk, whatever, without easily being discovered as to where they were, who they talked to, whatever, just as if they went into a bookstore and browsed wearing a mask, or whatever.

MR. KUROMIYA: These are very difficult issues. During five days of testimony and deliberations during the Communications Decency Act, we went over many of these issues. I was a litigant in those proceedings.

Let me give you just one example. The question of liability for cashing services, for information on the Internet coming from abroad--and I realize there is another panel on international issues, but is there liability for cashing services where information is just passing through?

These were seriously deliberated over, but it is impracticable to control everything on the Internet and it is going to be very difficult to regulate. I think the resolution on many of these issues was that it was an open marketplace.

On the issue of non-approved indications, that is what many persons with AIDS are looking for when they are searching the Internet, information they can't get from their clinicians or from the feds or from, perhaps, other individuals. Maybe a few people with experiences in these areas, both individual physicians and patients who have gone a little afield in terms of using--

DR. LUNDBERG: Sir, I'm sorry; there is a red light that has come in here for the moderator that controls me. Before I respond to the red light, there are two people here who have not spoken yet at all who are going to be given a chance to speak.

You can turn the red light off. I am going to come back to you shortly.

DR. NIGHTINGALE: Actually, we don't have any questions right now.

DR. LUNDBERG: Then why was the red light on? The red light was on to tell us that we don't have to stop. Thank you very much. That is what you call regulation, I guess.

MR. MACK: I suppose, in terms of technology, the government could develop a v-chip for webservers and require that other servers, when they send out requests, have information about what kind of site they are. If they have unapproved indications, they have a special code and the chip won't allow them in or something like that is probably possible.

DR. LUNDBERG: From Stanford? Stanford Psychiatry, yet.

DR. STEIN: As a psychiatrist, I detect a bit of paranoia here. I'm sorry; I couldn't help it. I think what we are doing is focussing on a technology which is very cumbersome, very expensive and not easily accessible to the majority of website maintainers or website owners.

I doubt that someone who has a website that gets 20,000 or 30,000 hits a week is going to want to trace back and see who is linking to that. I have to agree with Mr. Vance from Glaxo, even though it hurts me to say that, that what we are talking about is looking at the spirit of the Internet here.

I would like to give you an anecdote as an example of the state of links right now. We maintain a site on breast cancer and survival which is very widely accessed by many patients. I got a call from an insurance buy-out company where they buy insurance policies at face value from terminally ill patients requesting whether or not they could link to our site.

That would offend many of the people that I work with and we asked them not to. On the other hand, the larger question is, "But there are breast-cancer patients or HIV patients or others who are linking to their site who would ultimately benefit from gaining access to my site."

So is it really the path of how they got into the park or the fact that they got into the park and now they are able to use the things that they want to use.

MR. JONES: In the year that we have been up, I don't know anyone who was illegally linked with us and our webmasters have not reported anything.

DR. LUNDBERG: What do you mean by "illegally linked with you?"

MR. JONES: In other words, somebody, without asking us to do that for them.

DR. LUNDBERG: By law, do they have to ask you?

MR. JONES: I am not saying it is illegal or legal, it is just has not been done by our permission. We have no evidence that anyone has done this and, as I said, we get over 100,000 hits a day on this thing and we have no evidence of it. We have five or six people that are working all night on this thing.

DR. LUNDBERG: How do you know where they come from? How do you know that those are not coming from all kinds of other website linkages to you?

MR. JONES: Because we are looking at our stuff constantly. If some strange thing came on board, you would be able to see it on your own site, wouldn't you?

DR. LUNDBERG: Dr. Douma, do you agree with that?

DR. DOUMA: Actually, I don't, unless you are constantly monitoring to see the domain of where people are coming from, and most of the people on the Internet are coming from main domains anyway. What you will see is there are a lot of folks from Compuserve and AOL and MSN and URLs. And that is all you will know. You won't know that they actually "linked," meaning what happens is they went into Dr. Douma's "Sell all the vitamins you can sell" nook corner.

Dr. Douma, in order to make his site robust said, "Oh; those family-health folks; they have got a great site. I am just going to stick a link over there." What happens is somebody comes to me, I sell them a million dollars worth of vitamins and then they link to you, but all you know is they came from wherever they came from. You don't know that they came to me in between.

So your being unaware of it is simply the technology isn't letting you know what is going on.

DR. LUNDBERG: I feel like we have spent enough time on No. 3, the last part. Having solved all the problems, I am going to now open it to the far left over here to respond or ask questions.

DR. TEMPLE: Let me see if I understand this. The first question that you skipped was about what limits there might be on your sending someone who comes to your site to someone else. But we haven't gotten to that yet. This one is about whether there is any realistic way, if I understand it, to control who has access to a particular site.

In the morning session yesterday, I think there was fair agreement that the idea that there is going to be successful limitation of access to people's sites is, on the whole, unrealistic; not that you couldn't do it, not that you couldn't encrypt it or stuff like that.

But I think there was a fair assumption that anything that is on your site is going to be available to essentially any party, patients, physicians, and that trying to make those distinctions was not realistic.

I wonder if this group feels roughly the same way about that. Is that the general thrust of this discussion so far?

DR. LUNDBERG: I think when we did the vote, everybody agreed you can limit and determine where you send people, but most people did not believe you could limit or control in any significant practical way who comes to your site. At least, that is what most of the hands said.

DR. FOTSCH: There are multiple different types of security protocols that are out there and hundreds of millions of dollars being spent on these kinds of security issues that basically fall into two categories. One is access and the other is encryption.

I think we have used them interchangeably here today. They are clearly different. But blocking access to your site is something that financial institutions and various kinds or organizations are chasing. If you are unable to do that, the expansion of the intranet type applications that allow access from the Internet which, in fact, is occurring all the time, even today, would be stopped completely.

So whether you can guarantee that you are keeping someone off your site, I think everyone would be in agreement on. But whether you have any ability to limit whether there is access to your site from the outside, I believe, in fact, that you can.

DR. DOUMA: It is also to distinguish limiting access to your site and limiting access to things that are at your site. You can certainly create a password system to have half of your site, or whatever part of your site you want, which is only accessible by those who have a password that get to your site and then they fill out whatever form you want them to fill out that will let in--

DR. LUNDBERG: Nobody disagrees with that. You can limit parts of your site as you wish with a password and all that kind of stuff. Agreed. That is from the linkage or from individuals; right?

DR. DOUMA: Yes. But I think that is probably the most important thing to understand is that getting to your site probably doesn't make any difference one way or the other. It is what they get when they get there that is important.

DR. LUNDBERG: You can control that.

DR. DOUMA: Yes.

MR. LEUHR: I should say that I do not represent the views of the FTC or any particular Commissioner and that I am just speaking for myself. But to clarify one point, I think, especially on the issue of unapproved uses, if you are concerned about somebody else who is advocating an unapproved use and then says, "Link to this site," I think that is something that will be difficult to control as well from a practical and technical point of view.

It is not just an issue of whether a person will have access to a site on an individual basis. I think it will be very difficult for any web designer to control who is pointing at their website. So there are two issues; one is who is pointing at you and the other one is who is coming through the front door.

DR. O'BRIEN: The regulations provide that sponsors have certain labeling requirements for intended uses of their products when they create the intended uses and, in some instances, when they are aware of the intended uses.

I think part of the reason for asking Question No. 3 had to do with the surveillance if you become aware of an intended use, or should you do something or can you do something to detect what kind of intended uses may flow for your product based on web linkages that are pointed in the direction of your site.

I guess it seems to me that sponsors are probably looking at this not just because of FDA's intended use regulations but also because of product liability concerns. I am wondering what technology is out there to determine why people are coming to your sites, or is there technology to do that.

DR. NAYTHONS: There is no technology to determine why people are coming to your site. I think it is important to understand that it is really a shotgun approach. You may discover a bad domain, and I will explain "bad," pointing people to your site and, if so, then it is your choice to shut it down or not.

But there is no way to monitor that on a constant basis. I am a physician and a journalist. I did a Guide to the Internet. I spent more time than I should reviewing reviews of sites on the Internet. I was shocked by some of the things that were out there.

For instance, there was a site in South Africa that promoted an instant saliva test for AIDS. We stumbled upon it through the strangeness of the Net. If we were discover that that site was linked to ours, we would shut it out. But there is no way to know. We would have to literally stumble upon it, stumble upon the linkage on their site and shut it out.

There is no way to regulate that that I can imagine.

DR. O'BRIEN: I guess my question, though, is can you shut the link out or can you only shut that provider?

DR. NAYTHONS: If the link is coming from a domain, and I know the address of that domain, I can instruct my server to shut it out; yes. In this case, it was a single domain. There is no reason we shouldn't be able to shut that domain out.

DR. STEIN: I want to play devil's advocate for a moment which is that if somebody is seeking some information about HIV or HIV testing and they come in through a back door which isn't something that your company supports, that there are a number of ways including disclaimers and the kind of what we were talking about, watermark logo, so that if they come in through, say, a secondary page on your website, your company's name is still there.

I think that people who are looking for information look to many sites. I would hate to shut out--I mean, that is kind of self defeating. You are shutting out the people that you want to come to your site because they are serious consumers of information and need help.

DR. NAYTHONS: Very quickly responding to that. I believe when you look at all the snake-oil salesmen which can exist out there, I believe there is a responsibility to be a guide as well as have an open site. Now, this doesn't take away from what I believe is the open nature of the Internet or the open nature of the park.

But to carry the park metaphor, and I promise not to use the superhighway metaphor, if you stumble upon someone committing an act that you don't personally approve of in the bushes, and that truly offends you, perhaps you ask them to leave or you avoid that. But you don't just nod at it.

I agree that information should flow freely, but I was personally offended by what seemed to me pure snake-oil salesmanship.

DR. DOUMA: Two responses. One is technical. You can shut out that domain if the person has that domain as their address. However, there are very few people who are going to carry that domain as their address. They are going to come from some other domain, visit that site, get a link and link to your site. So you really can't effectively control it.

The second thing is I want to build on what Sara said. Let me just give an offline metaphor. There is a lot of tabloid journalism about health and there are a lot of advertisements in the main media about health that I consider dangerous, quite frankly dangerous, if people follow it.

I would certainly not want to prevent them from reading JAMA simply because the minute before they read the tabloid. I think the analogy here that Sara is trying to put forward I agree with totally.

MR. GANSLAW: I think that that question really headed us toward Question 2 here. And I guess my point is can you stop an independent party pointing to you. I think that it seems to miss the point a little bit because if the company isn't actually sponsoring this other page--I mean, anybody else can talk about off-label uses.

There are free-speech issues there and I am not sure that if they point them to the place where there is correct information about a particular pharmaceutical agent that that is, necessarily, a bad thing. I think that hopefully people would get to the place where there is correct information on something.

It seems to me that, again, if the company is not involved that is a completely different question. It seems to me more within sort of the public eye.

DR. LUNDBERG: The American College of Cardiology has provided us with a natural segue to go back to Question No. 1 which starts with ADF. Now, I don't know if it was determined yesterday what that meant or not. I wasn't provided information. I tried to figure out what ADF was. I figured, based upon what has been in the newspapers recently, it must be Archer Daniels Finland. Or perhaps it is Americans for Democratic Fiddling. Or maybe Acute Drug Functions. I am not sure.

MR. VANCE: It is the American Dance Festival based in Durham.

DR. BERNSTEIN: Point of information. It is a very simple thing. It is FDA backwards.

DR. LUNDBERG: That is a disclosure; the FDA is backwards.

DR. TEMPLE: No; the drug company is backwards.

DR. LUNDBERG: "ADF is a leader in clinical research, development and marketing of human and animal drugs, biologics and medical devices. ADF knows that federal law and regulations prohibit the promotion of approved human and animal drugs, biologics and medical devices for unapproved uses."

Editorial note; I must rise to, once again, correct the FDA for its incorrect use of the word "unapproved uses." As JAMA pointed out in an editorial more than ten years ago written by John Archer and never disagreed with by the FDA, there is no such thing as an unapproved use. There are only uses that are not approved for labeling.

There are approved products and devices. Once they are approved, physicians can prescribe them for whatever they want, for better or worse. Labeling is the issue. Approval of labeling is the issue. I will be very happy to be argued down if anybody wants to but I am quite sure I am correct.

Meanwhile, we will go ahead.

"They want to incorporate several hypertext links from their website that contains information about their products to other websites. Given the current law," and, of course, there is one with all of its explanations about that thick, I understand, "what types of links, if any, would be appropriate? What parameters, if any, should be considered for links from ADF to other websites without violating the law?

"ADF notices that on some websites, before leaving the website to link to another website, the user is automatically presented with a screen that indicates that the user is leaving the website to go to another one. Join ADF in their discussion of whether there is any benefit to this type of information," that kind of a disclaimer.

So let's go back to (a). You have got the case study. Given the current law, what types of links, if any, would be appropriate? Who wants to start? The gentleman from Glaxo who has a simpler comment this time, I'm sure.

MR. VANCE: No. We met with the FDA and we told FDA that we were doing links. Links are totally appropriate and they should be allowed. Links to sites that don't contain off-label information and are otherwise in compliance with FDA regulations shouldn't be a concern.

Links to sites that contain independent information, whether it is on or off-label, complying with FDA regulations or not, should also be permitted. Again, I will emphasize that the technology is new but we do these things already. As I said yesterday, we give out patient information, we give out professional information, that links to things like AMA.

We link to the American Cardiology Association, all these other professional groups for physicians. If they contact those groups through that link of the address, they often get on-label information.

Likewise, for AIDS patients for years, cancer patients for years; we give them links to the various cancer groups, the various AIDS groups. If they contact those groups, they get on- and off-label information.

The key should be exactly what we do for continuing medical education or as was recently announced for the reference text policy, it should be if you go to another site, is it independent, although, perhaps, more important, is it scientifically credible. Is it scientifically rigorous, objective and balanced? If it is, then what should FDA's concern be?

DR. LUNDBERG: I presume you are going to indicate how you determine scientific credibility?

MR. VANCE: The one standard FDA has suggested is independence.

DR. LUNDBERG: Independence means credibility? Come on.

MR. VANCE: Supposedly under the current policies it has. I think there can be some judgment. It may well be that as for scientific meetings, one of the indicia of credibility was whether you had enough discussants, whether there was the ability to have different points of view.

I would suggest on links, if you had one link among the many that are out there, that might not be credible. But if, in fact, you were to link to the AMA AIDS site, to AIDS Treatment News, to some other patient groups, to the CDC then a variety of responsible patient and government links would be an appropriate way of doing it.

If this is a new medical organization that was created to deal with something like, let's say, ulcers, that had no history, that would be somewhat more questionable. On the other hand, I picked up a wonderful piece at the AIDS conference from, again, Pharmacia Upjohn that listed, I think, 75 AIDS links.

Our searches have indicated that there are something like 348 associations that give out AIDS information on the Internet. We have not chosen to put on 348 links and I don't think should be required to link to everybody. But to link to just one entity that is not representative might suggest that something was skewed.

But if you give a balanced set of links, whether they are on-label or off-label should not be an issue.

DR. LUNDBERG: If I may try to summarize what you said. I think you are saying the old rules are good and this just happens to be a new medium. Let's stay with the old rules and it would be all right. Is that about right?

MR. VANCE: That's correct.

MR. GANSLAW: I think, though, there can be a different interpretation of the current rules. When somebody does either come to your page or then goes to another, it is an affirmative act. People search you out. It is not as though this is broadcast media. I think that this is just as if anybody had asked a company for a paper on an off-label use or they are searching information out.

This is an affirmative thing that you are doing.

It is not like somebody is leading people along on the Internet, dragging them to the places they would necessarily want them to go.

DR. FOTSCH: I would agree that the links are an absolutely critical aspect, both internally for navigation and also, obviously, externally. Without links, it is the telephone without the telephone numbers.

I think that the medium at this point assumes a link is an endorsement of some sort. And I think that it is probably important to acknowledge that. I think it is also important that there be some disclosure as to when you are, in fact, leaving a site.

I am a little bit concerned with the concept I think I heard that a site that an FDA-regulated company would link to would be held to the same standard as the site that is sponsored by the FDA. If that is the case, the pharmaceutical companies could only link to each other, which I doubt would happen.

DR. LUNDBERG: You are paraphrasing an FDA statement or you are making a proposal or you are making a joke?

DR. FOTSCH: I was hoping for the latter.

DR. LUNDBERG: You got the latter. How about the former?

DR. FOTSCH: It seemed to me that there are FDA guidelines with regard to fairness, balance and disclosure and that those need to be specifically applied to the, in this case, ADF site and not, necessarily, to the sites that ADF would link to.

MS. HAMET: I have a question. To some extent, I understand that linking would be an endorsement. But one of the concerns that I would have is if I were to link to, say, an AHCPR guideline, is there a liability issue there or is there a copyright concern to be able to link to a publication?

DR. LUNDBERG: The question is is it an infringement of copyright of some kind if you, on your site, link to some other site that has print copyright. I think the answer is no. The answer is no on that one.

We got one answer. How about that?

DR. DOUMA: The analogy I see is--and let me use JAMA as an example because it is actually my favorite medical journal. JAMA, because it is so good, resides in every public library in the United States. We are having advertising sponsored by pharmaceuticals bringing unlabeled or non-labeled information to the universe.

Now we are talking about it is slightly different here and that we make it, perhaps, a little bit easier and we are, therefore, worried about the linkage whereas before--the linkage is already there. It is already there in the print media.

DR. LUNDBERG: Excuse me; would you explain a little better how you think the public library brings unlabeled information to the public? Did you really mean that the advertisement was regulated for physician advertisement but available to the public?

DR. DOUMA: What I mean is that JAMA carries information about off-label uses.

DR. LUNDBERG: You mean in its articles, not in its advertisements.

DR. DOUMA: In its articles; correct.

DR. LUNDBERG: Of course it is true, as do all other medical journals.

DR. DOUMA: Absolutely correct. Here we are talking about a similar kind of link. Since I represent the media, myself, as a the media--I don't represent AOL particularly--but we want everybody to know everything about everything there is to know.

So on- and off-label information to us is of value both ways and the question I want to ask us all, perhaps, is, stepping back just a second, why do we, in fact, want to prevent the distribution of off-label information by a certain subset of our society?

In this case, it is people, perhaps, who have a vested interest and that is our concern. But they also have the best information. So we have to look at what is the give and take on that regardless of whether it is the Internet.

DR. LUNDBERG: I take that as a question of the Food and Drug Administration. Is there anyone from this panel who wishes to respond directly to the question why, it is alleged, do we, through this medium, wish to prevent the availability of information now available in every public library in the United States to people who might be interested in it if it happens to be for something that is off-label?

DR. TEMPLE: I think we are following the Schultz-Nightingale Rule which is assume the present rules, present law, which is that you are not allowed to do that.

DR. LUNDBERG: Because we are a government of laws not of whims.

DR. TEMPLE: As everybody knows, there have been challenges to what companies ought to be able to distribute.

DR. LUNDBERG: Are there challenges as to whether JAMA can be in the public library?

DR. TEMPLE: No; of course not.

DR. LUNDBERG: Then what is the difference?

DR. TEMPLE: What is the difference between targeting particular articles to distribute and referring someone to the public library? Is that the question?

DR. LUNDBERG: I don't want to phrase his question. Let him phrase his question himself.

DR. DOUMA: For example, let's assume--well, I know JAMA is up on the Internet and they are getting advertisers who are advertising on JAMA in same way--it is totally analogous to the print. JAMA now is able to be accessed, or they can link to or be linked from, either way, that creates access to the general public about off-label uses of medications.

So the analogy, I think, holds 100 percent. But here we are now trying to wrestle with are we concerned about it here using this electronic media when, in fact, it is already happening and has been for, I don't know, 50 years--

DR. LUNDBERG: 113 years.

DR. DOUMA: 113 years--pardon my ignorance--with print media.

DR. TEMPLE: Is this FDA question time because I have a question in that area. If this seems like the right time, I can launch it now.

DR. LUNDBERG: Go right ahead.

DR. TEMPLE: It really relates to the sorts of things that David Vance was dancing around. There is an FDA policy that says general things like texts and stuff can be referred to. And even though they may contain information about off-label use, it is a general thing prepared independently. We are not that worried about that. I don't know if that policy is now final--yes; okay.

DR. LUNDBERG: For those of you who can't see, Dr. Temple has been validated by the galleries.

DR. TEMPLE: It had been proposed. But, at the same time, there was a distinction made between targeting particular articles or things like that. So someone mentioned an ACPAR, a position paper. Just to put a point on, because this has come up before.

ACPAR often says that drugs of a class are all good for the same thing. Sometimes, individual companies have a claim for an ace inhibitor or a calcium channel blocker or something that other people don't have. The industry is, in my experience, of two minds about whether a company that doesn't have a claim ought to be referring to ACPAR so they can get all the claims.

FDA has an interest in whether you claim what you have actually done or have a more general claim and ride along on someone else's claims. Those areas are of some interest and we have been distinguishing between referring someone to a text or to a whole journal or to the whole index to JAMA and pointing to one or two articles that may be of particular interest.

I wanted David to talk a little more about that distinction which I think was a little blurry as it finally came out. We, historically, if you analogize our concerns, would say if someone comes to your site and you want to direct them to a particular claim, that is a potential problem. If you want to tell them where they can find a text or have a whole text they can search, or something, that would be less of a point.

Where do you come down on that sort of question?

MR. VANCE: Let me first go back to this other issue we were talking about about there are metaphors out there. I think I may have mentioned yesterday, the New England Journal, JAMA, are available in libraries. We have never considered those to be direct-to-consumer even though consumers can see them.

Likewise, AMA T.V.. and Lifetime T.V., the same metaphor was accepted. Again, I think we have lots of good metaphors. Glaxo Wellcome's policy on the Internet is what we put on the net has to be on-label, comply with the rules unless it is something like an annual report, the financial documents or it is something that falls under 312.7 as scientific information such as press releases.

But everything that we link to in our links cannot be off-label. If we have something that is in an on-label AIDS program, whether it is direct-to-consumer which we voluntarily preclear with FDA or whether it is professional, which we still have mostly to clear for accelerated approval for Epivir.

That link says it is only to the AMA netsite for AIDS. It doesn't say, "Click this spot to see the latest off-label information on our AIDS drugs." Likewise, in the Pharma discussions we had the other day, there was a discussion about whether SSRI antidepressants could have a link to a patient group on premenstrual syndrome.

Our recommendation from a Glaxo side would be no, that if you had an antidepressant, you could well have a link to OCD, if that was an indication you had, to a depression group of patients. And you also could have a link that might say, "Other patient groups." If you went to this other patient group list, it might have cancer groups, it might have groups for ear infections, it might have a whole bunch of groups.

It might, in that group have PMS but there would be, then, nothing that clearly tied it back to an off-label promotion. So the link, in itself, should not offer clearly off-label information. But if this independent link happens to have off-label with the on-label and is a credible, independent group, there is a liability issue as raised by Knoll.

We are not going to put you out to sleazebag company and get our drug misused. We are not going to have something that has a risk of killing somebody misused in a way if they are not going to put the appropriate balance on and our patients are going to be hurt, the doctor is going to misuse it. We would be stupid to link there; just credible sites and offered in a non-off-label way.

DR. STEIN: Three points. First of all, that is very admirable on the part of Glaxo, but there are other sites that will not abide by that sense of morality. So the question becomes should there be a disclaimer or a disclosure that you are now going to independent sites, you are leaving this site and going to an independent site. Then they can list whatever it is they want because they may link to a subpage which has a piece of text which validates some point they are trying to make and doesn't offer the whole overview to the consumer.

I want to raise a second correlatory regulatory issue for the FDA and that is purchasing links. As I mentioned yesterday, the search engines have begun to sell a place in the list. Open-place index is the first to do so. The first ten hits you get on any search there have been paid for and is determined by how much they paid.

The question is companies that sell links so that if I wanted to be on the Glaxo site and I am whatever company, how much did I pay. I think that that is another disclosure that is required which is that these are paid links.

DR. LUNDBERG: I would be interested--the word "disclosure" just came up. It has been raised two or three other times before. Would any of the panelists like to either indicate how they, on their websites, face an issue of honesty and disclosure about conflicts of interest that might be present, about financial backing that might be present, about hanky-panky that they want to confess to the people who would want to watch it, who might want to watch it anyway because they enjoy watching confessed hanky-panky? I don't know.

Does anybody want to talk about how they handle disclosures if they handle that at all?

MR. GANSLAW: I think we are just starting to explore this because our page is a bit in its infancy. But I think we are going to basically follow the model in other media where we work with industry; that is, we are going to try to maintain a ratio between content in advertising and we are going to clearly delineate what is content in advertising.

When people leave our page, we want to make it plain that they are leaving the American College of Cardiology. We don't want people thinking when they are now at some other site that they are still looking at our content.

So I think quality educational providers have a self interest in making sure that people know where they are and that they are identifying with their content when they are there and not when they are not.

DR. LUNDBERG: Okay. But when they leave their site, they are going somewhere that you have guided them to, or that you have made available by a linkage on your site?

MR. GANSLAW: If we provide links, in the case, for instance, of a company sponsored maybe part of the package for helping an educational program is a link to a particular company site, that is obviously appropriate.

DR. LUNDBERG: Are you telling somebody that, "They paid us for this linkage?"

MR. GANSLAW: I think that is one form of--basically, it is another form of advertising, if you will. That could be one of the various forms. That happens every day.

DR. LUNDBERG: The linkage happens. Does disclosure of payment happen every day or happen at all? If some company pays the American College of Cardiology in order to have a link from your site to theirs and you receive money because of that, if that happens and you choose to disclose here, have you disclosed that to the viewers or not? And do you think you should?

MR. GANSLAW: With regard to the American College of Cardiology, any time that a company has provided support for a program, it is clearly acknowledged as such.

DR. LUNDBERG: I am not talking about support for a program. I am talking about a specific linkage, to link to another site and they paid you to provide that link. Do you disclose that and how do you disclose that?

MR. GANSLAW: I am not exactly sure that if there wasn't support of an education program, that would be available through us.

MR. MACK: The use of banners, for example, on the Internet now has become a common feature, and the banner, itself, indicates, perhaps, that you are leaving the site and going to another site. For example, we run banner ads to a site and it says, "Visit the Allerdays site," and it gives the URL. Now, that is clearly a paid-for link and I think it is recognized as such by people visiting your site because this is so commonly used.

There are other types of links. For example, we have some sponsored information and we indicate that this information was provided by an unrestricted educational grant by Company X. Now, Company X has told us specifically not to have a link from the word "Company X" to their site. Now, I am not exactly sure why not.

They also told us not to use their company logo to identify themselves. Again, I am not exactly sure why not. I would like to find out more about that.

DR. LUNDBERG: Any other confessions?

MR. JONES: Because of the size of our site, we are now being approached by different companies who want to buy clickthroughs. Clickthroughs are then placed in various areas on the site because, even in family health, we have so many different health areas in there that we can place a banner where a company can click through.

We do charge for that. We have a rate that we charge for them to click through. We have a measuring system that can tell them when we have a stat package that can help them in that. We also get companies that come to us and want to develop a site within our domain and then we work on that.

We have a medical group that approves that. And then, if it is a drug company, then they are responsible for the information that is talking about that product. Most of these, though, are disease-management types of things because just to put products on is going to be very, very boring.

DR. STEIN: Could I ask Tom a question?

DR. LUNDBERG: Yes.

DR. STEIN: Can you explain the clickthrough a little better? Is it just a word that is highlighted and there is no indication that you are leaving your site to go to another site?

MR. JONES: This is very new with us because we have just got two clients doing this. It seems to be a trend. A lot of them are using it to test your site for, perhaps, later to come on and have us develop a full site around a particular disease or something on that site.

It is a good method for testing. We are offering this now for the balance of the year for people who want to test it. We don't particularly say to them, "You are leaving the site." We have a banner up there. It tells them that you can click here and you can go to this disease information source, and they click and they go off.

DR. LUNDBERG: Do you believe that the site to which they are referred from your site is offered a degree of warranty by you that what they get on that site and any other sites they might be sent to by that site, is validated and vouched for by you?

MR. JONES: The only thing that we do when they offer a site that they want to clickthrough, we do have a complete medical staff who reviews it. Most of this today is from drug companies, and they are very, very high-caliber type drug companies. Their sites have already been looked at by their people and supposedly by FDA people.

So we don't feel that we are responsible that much for what the content is except that our doctors want to look at it and make sure that it is--

DR. LUNDBERG: So you don't feel responsible for what is on their site. Do you feel responsible for where they will be referred from those sites?

MR. JONES: No.

DR. LUNDBERG: Not at all?

MR. JONES: No.

DR. DOUMA: I think there is a standard that is developing on the Internet I thought I would briefly mention. It has to do with your banner advertising which is, basically, carrying--either it can be a logo or it can carry a promo which has more information which is better for the advertiser.

People who are on the Internet at all for any length of time know that that is an ad and that is where they are going. So it is standard fare. I think we don't have to worry about that. In fact, if you looked at our sort of common broadcast media analogy, there is nothing between programming and--when the ad comes up, it doesn't say, "Hear ye, hear ye; this is an ad."

DR. LUNDBERG: An ad is an ad is an ad and a reasonable viewer will be able to tell it is an ad, you are saying?

DR. DOUMA: Typically, with the standard way of doing it. It is possible, and this is where we need to be rethinking about this--I can write an article and, in that article, I can create hypertext links from that article to another place where it can be on another website or it can be on my own website.

In doing that, there is greater danger that I can fool somebody into thinking something is an ad or not an ad. There, I think there is a moral, ethical demand on all of us to think what is reasonable so that the person going knows that they are getting information that, perhaps, may be biased.

But I do want to reinforce what Tom has said which is that, for me, I would rather have pharmaceutical and medical-product information up online because it has been reviewed by so many people, it is certainly more reviewed than the typical media reviews information today.

DR. LUNDBERG: Following that, we will go right back to 1(a), then; "Given the current law, what types of links, if any, would be appropriate?" Do you want to say? Anybody? Given the current law, what types of links, if any, would be appropriate? Does anybody want to say that distinctly, succinctly, for consensus for the group?

We have a volunteer from the FTC.

MR. LEUHR: This is related to that precise question but it is not what types of private links would be appropriate, but I think this would be an appropriate time to talk about what types of public links would be appropriate.

I think those of us on the public-sector side have a responsibility to get as much information out there as possible. I know we at the FTC are in the process of reorganizing our page precisely because we know that there is not an ability to control access, not an ability to control, necessarily, linkage.

So, for the first time, we are really kind of in the business of self-promotion. Most government agencies try to keep a low profile in many instances. But I think in instances like this, there is an opportunity for us to join the educational forum.

I have clicked through the FDA's site and I know it is organized much like ours is in terms of offices and around certain topics. I think it is a very good-looking site and very easy to read and use, but I would also like to suggest focusing on a little broader category.

For example, patients. For those patients who are looking for guidance from the FDA, what are your guidelines as to on-label versus off-label use? Information and health providers who are thinking about linking. Once you do come up with some type of policy plan, perhaps put a policy statement up there and suggestions about how they can provide good information which you think is acceptable.

MS. GOLODNER: I just wanted to respond to the question you raised. I think what Glaxo mentioned is their guidelines on their site were very good. I would hope that other companies would follow that. However, I know that other companies won't follow that. So I think it is very important to have disclosure.

When someone does enter a site of a company, obviously, you know it is for promotion of their products. So there should be disclosure if you are going into another site that is not promotional.

I was disturbed about paying for information that would be on a site that a consumer might consider an independent source of information. And I was wondering in the family-health site whether or not there is disclosure that someone has paid to be on that site.

MR. JONES: No; there isn't. It is just like if you saw an ad in a magazine, you are going to assume that somebody paid for it.

MS. GOLODNER: But you know it is an ad. You might know it is an ad.

MR. JONES: We are not clicking through to an ad at this point. When they are clicking through, they are going to a health-information site that is sponsored by a major drug company, or, in one other instance, where a drug company is using the clickthrough to recruit people for clinical studies which is becoming a big part of this.

But, no; I haven't seen anything on the Web. I know we don't. We don't say, "This is a paid-for thing." We are a commercial site to begin with, and we just don't deal with health. There are many ways and many reasons why people come to the site. They can come because of travel, through a cooking, finance, sports, you name it.

But when they get onto our sites, we have links within our site because it is so large. For us to say that everyone that clicks through to something, that it is a paid thing--I don't know how we would do it, frankly.

DR. LUNDBERG: Maybe some of the older ones in the group remember the disc-jockey scandal with payola. This is flashing in my memory as a radio kind of thing. A bunch of money passes behind the door, chooses a certain thing, there it is. People listen and have no idea and that goes to Number One in the charts. That smacks of payola.

MR. VANCE: Succinctly, links should be allowed to to do anything that complies with FDA regulations. That is a no-brainer. Links should also be allowed to sites that are independent and have some reasonable scientific rigor, objectivity and balance. And that is going to be a challenge to work with that issue.

Disclosure, in my mind, has always been a no-brainer. There is nothing more stupid you can do with any activity from a pharmaceutical company than not disclose because, no matter how good the program is, if someone comes up and says, "Oh; they hid the fact it was some slimeball pharmaceutical company," you have just done so much damage to yourself.

Disclosure is a no-brainer. It has got to occur. When you are putting a hotlink up, it should say, as it does, it is a JAMA site. We shouldn't have to say anything more than that. Or it is a CDC. Likewise, our vendors--at least we have got a couple of attorneys working on this. I don't control everything. But the ones I am working with, the vendors do disclose that they are getting an unrestricted grant from Glaxo Wellcome.

In one case, in a link I saw from a Canadian subsidiary I knew nothing about, 30 subsidiaries of various pharmaceutical companies around the world all supported this entity in Canada that posts everybody's press releases on a regular basis, to my surprise.

DR. STEIN: I would like to see links that are clearly identified that you are going to an independent site, perhaps leaving this site. I would like to see links that are clearly identified if they are paid for. One of the things that I am worried about is that you will be on a site and, for example, there will be a word called "depression," underlined, and you click on it and you wind up in the middle of someone else's site who has paid you, for example, and you are now reading about a single antidepressant, for example.

Unless you are watching the address, you don't realize that you have left the site. So that is one of the things I am worried about. Just a sort of reminder to the pharmaceutical companies, one of the things that I think academicians would like to see which we rarely do is the difference between efficacy and effectiveness trials.

For those of you who are not clear, efficacy trials are generally whether a drug works against a disease. It is usually tested against a placebo. But effectiveness trials are drugs one against another. Sometimes the differences are very, very significant.

For example, a chemotherapy that I am thinking of for advanced ovarian and breast cancer which is very effective against placebo but when compared with another chemotherapy gives only 8 more days of survival. I think that pharmaceutical companies can, at times, be very misleading with their information.

MR. VANCE: If I could respond to that. That was the subject two years ago of another FDA hearing on quality of life, cost-effectiveness, outcomes research. It is very complex and because it is comparative, and creates many problems for FDA, that it still under negotiation.

So many of those things we can't do under FDA regulations but if it were available in an independent site where we could link to a JAMA or to another organization, perhaps we should be able to do that.

DR. LUNDBERG: We have segued to 1(c), "ADF notices that on some websites, before leaving the website to link to another, the user is automatically presented with a screen that indicates that the user is leaving the website to go to another one," quite different from the depression click that Dr. Stein just spoke about.

"Join ADF in their discussion of whether there is any benefit to this type of information," the disclosure that you have left here. You have gone somewhere else. I would add, relationships between the there and the somewhere else.

DR. DOUMA: I do want to comment on (a). Given the current law, as far as I can tell, particularly we need to look at--there is the media and there are, perhaps, other uses of the online technology in the media. Whether it is consumer or professional media, it seems to me there is no doubt that it is appropriate to link to whatever is in the best interest of the consumer, professional or otherwise.

Certainly, we need to start from that point. I also, personally, think it makes a lot of sense for good sites regardless of whether, in the standard media, to be able to link to appropriate information meaning information that is helpful, not necessarily label versus off-label.

With regard to this particular question, the answer is that if you do that, you will probably prevent half the people--I don't know where I got that number either, George--you would prevent a huge number of people from actually going to whatever the site is.

Now, the solution to that is to try to work out so that when we are linking to another website, you are linking to a place that it is pretty obvious when you get there that you have gone there and what it is.

DR. LUNDBERG: You think it is a technical difficulty that prevents them from just clicking and going that blocks people out?

DR. DOUMA: It is a time difficulty. If you are on the web every time you bring another screen, you have got to download it and look at it and then you have got to click again and go to the website you actually wanted to go to.

MR. GANSLAW: Just to go back to Dr. Stein's comment. I think that if the sort of intrapage link isn't from a pharmaceutical company on their particular page, there is simply nothing wrong with that. That seems to be outside the sphere of regulation because that is just people interlinking if you are not actually going from a pharmaceutical company.

On (c), I think that there is a lot of merit to knowing where you are. I think the key to allowing people to understand what is and what is not good information is to know exactly where you are. I think that good content providers, companies who want to protect their own interest in understanding that their content is quality when you are leaving that side; buyer beware.

I think that that is positive for quality providers. It is probably not as good in the interest of people who are not quality providers.

MR. MACK: I think there is a benefit to having an intrapage. We are trying to do that with banner ads to at least tell people what they might be expecting from the site that they are going to. We are beginning to review some of the sites that we have links to, and we have physicians rating these sites.

Then we have one of these intrapages that after you have clicked on the link, it says, "Now you are going to the gridsite, for example, and you will find some nice things here. It was rated," blah, blah, blah, and so on, "by our staff."

For a pharmaceutical company, I think you might want to have one of these pages because then you can track which links people are actually clicking on from your site. That might be a benefit to you to know a little bit more about what your readers are interested in.

MR. MARKS: I think this is one method, having the intermittent screen to disclose you are now leaving the site. But I would agree that there are even additional and, perhaps, better ways before you click on the link going elsewhere to make it very plain that you are linking to something outside the content of the site that you are currently at.

I think that that disclosure only enhances the credibility of the site that you are publishing.

DR. NAYTHONS: I don't think that we should forget that the goal is the better-informed consumer. That is the goal of all of this. With a technology called Frames, you can actually link to your site and make it appear that the information is within your site.

So I think it is very important that there be an exit page that lets you know you are leaving a site. I have been told that you don't have to click through it, that it can appear and disappear. It obviously takes time to load but if loading one exit page is going to slow down your site, then I think you are in trouble anyway.

The goal is to make the consumer better informed and it can be done with a splash exit page.

DR. LUNDBERG: We all have our vested interests. We all have our conflicts of interest. Many of us are providers. We are also consumers. We are the public and we are all patients. Patients is really where it comes down to.

I want to thank this panel for doing an awfully good job with a lot of interesting and difficult questions. We now come to that time in the program when officially, formally, we have questions, discussion, challenges, et cetera, from our distinguished FDA panelists.

And, at 10:25, we go to public comments, public discussion and questions.

Have at it. Who is first?

DR. BERNSTEIN: David and Larry and I guess a lot of you said that disclosure is good, and that is something that patients or consumers have a right to know. But what I am wondering, and I guess this is more directed to the companies or those of you who develop websites for companies, can you comment a little bit on whether--you can put up a statement and say, "You are leaving. We are not responsible for where you are going."

But what is the continuing responsibility that you have, or do you have any continuing responsibility, to monitor those sites where you are linking to determine what information is being said there, and how it affects your products?

DR. LUNDBERG: In an ongoing way, for quality control, you are linking to somewhere else, should you monitor those somewhere else and, if so, what is due-diligence in that regard?

MR. VANCE: So far, we have limited the number of links we have done probably for those reasons. We have tried to pick the most reputable places and, as I kind of suggested earlier, this issue of pollution was one that just arose last Friday where some people were checking some links which I think still are reputable links. We just didn't understand the magnitude of this pollution issue and we have not agreed inside the company about how you deal with a very reputable link that has very good information on it when, in fact, somebody can dump crap on it.

I would hope that the company will go back and decide that, "We will want to make sure there is adequate policing and cleansing done of that site, but if it is inherently good, we will allow that to go."

But, as I said, there are 383 AIDS sites and although we are very committed to being one of the leading companies in HIV, we decided that connecting to 383 sites blind would be crazy and monitoring those probably also would be crazy and irresponsible, something we just couldn't manage.

DR. LUNDBERG: Other responses?

DR. DOUMA: I think it is absolutely critical if anybody is providing any site that they have any pride of ownership that they monitor where they are linking directly to because that is your consumer's or customer's experience. And unless you are concerned about that, you are not going to survive very long anyway.

One of the problems is not where you link to but where that link links to and links to and links to and ten generations down. So you do the best you can. I don't think it is regulatable only in the sense that I don't know where you can draw the line.

In fact, the federal government has some linkage to some sites. If you keep following it down, you will be amazed where you can get to.

DR. FOTSCH: I would agree with those comments. I think that one of the concerns from a regulatory standpoint is that you so handcuff the drug companies that they, in fact, are not and do not become the resource that they can be at a time when consumers really are looking for information.

Six times, yesterday, I heard the standard default, "Consult your physician." I have practiced medicine for ten years. I have been on the other side of it. I don't know how many people have tried to consult their physician lately, but it is not that easy. And managed care is not making it any easier.

I think that while consumers have a right to know where they are vis-a-vis the Internet, whether they are still on an FDA-regulated site, for example. I think that the company should be held to some level of reasonableness as opposed to technical standards.

If you put them in a position in which they have to continuously police the sites that they link to or the sites that link from there, I think that they are going to become superfluous to the equation of healthcare information on the Internet and that will be a sad thing.

DR. LUNDBERG: Other questions?

DR. TEMPLE: I guess I hear some potential disagreement here. David is describing a fairly tight system, roughly running according to present rules where you can send people off to general independent sites pretty freely in accordance with current policy, but you have to watch it.

You can't send them off to specific targeted things directed at off-label uses. I guess I would like to hear a little bit more discussion of whether the current arrangements still seem reasonable in this new era. Can Glaxo do it because they are very big and massive and have capabilities that other people don't have? What about all that?

DR. LUNDBERG: Responses from others, especially people who haven't yet spoken. I want to give them a chance, for sure.

DR. FOTSCH: I would like to comment that I think you have hit it on the head. They have a very tight policy and a tight structure. Unfortunately, the Internet does not respect that. The Internet is not tight. The Internet is all about connections and it is very loose. I think that it is fine, and I think that it is an admirable policy.

But making that regulation, it seems to me, is really scary because it is handcuffing to other organizations. While I would applaud that particular policy, I would wonder how deep it goes because my guess is the site that they may link to may link to other sites and other sites and other sites.

How deep could even Glaxo go in terms of the veracity of all the information. Essentially, I would argue, you could probably get anywhere on the Internet from those sites.

MR. MARKS: I actually think it is very important that the drug companies police the sites that they link to on a regular basis to make sure that they continue to comply with independence and balance, rigor and so forth.

I think for independent sites like ours, it is equally important to have a huge latitude in determining what organizations it is possible to link through. I think that we are going to see, over the next year or two, as more and more independent sites have sponsors that may conflict, the goals of which conflict with other companies that were linking to independent sites and now that the independent site is sponsored, there may be a different decision in terms of what to continue to link to.

I think drug companies are going to be policing that anyway, and it is valuable if they are going to be policing who is sponsoring the independent sites information, to have them continue in their monitoring role for the other tests of rigor, balance, objectivity.

DR. LUNDBERG: Interesting use of the word "police." We have the Internet police. Posses? Or the Texas Rangers? Or 100,000 uniformed officers on the street? Or just what in the Internet police? Citizens arrest, perhaps, in the Internet police?

DR. TEMPLE: I just wanted to make one distinction about my question clear. We don't, I don't think, have any fantasy about trying to control independent sources. As we like to say, independent sources can be biased, stupid, anything they want. They are independent. We don't regulate them.

But drug companies can't necessarily do all those things. So it was particularly the responsibility of a drug company for its linkages that I was interesting in hearing more comment on, not what those independent people may do on their own. That is really not our business.

DR. LUNDBERG: Mr. Vance, you have spoken often and well. Are there any others representing other companies who would like to enter in this fray a bit more than they have?

MR. DOUGLAS: We work for a number of pharmaceutical companies so it is something of a composite opinion, I suppose, but I do think it is a question of linking to services as opposed to linking to content within the site. Within a regulated site, you can't put, "Here is a link for the off-label information you may be looking for."

But I don't think you can police all of these sites. The linking is a service. It is actually part of the attraction. It is why somebody might come to your site. The service that a pharmaceutical company can provide to somebody coming to their site with quality links--and it is a quality service, so it is sort of regulated by the user in that if they deem no quality, they don't come back, so it doesn't much matter what you do.

So it is incumbent upon the site sponsors to make sure they are quality links and the service that you are providing and the evolution of promotion as a result. The links are the thing people come for, not the promotion. So I think they, in a sense, regulate themselves in that context.

People don't channel surf to the commercials. The surf away for content and these links really do represent content.

DR. LUNDBERG: Other questions from the FDA panel? We are winding down in timing for this group. If not, I will go the this response.

MR. MACK: I still want to make a distinction between a company linking from its site to, say, an article that discusses unapproved uses and linking from that article to the company site, for example. We have a newsletter and we often publish reviews of peer-reviewed literature. We are a secondary publisher. In those reviews, we mention drug names.

We like to know if it is permissible to carry ads within those articles linking, perhaps, to a drug page within the pharmaceutical company's site, itself. Obviously, a lot of our articles do discuss unapproved uses but these have been already reported on in the peer-reviewed literature.

DR. LUNDBERG: That is a good practical question. Anybody from the FDA panel want to respond to it? You are on the record, of course.

DR. TEMPLE: Again, we are mostly now responding. We don't tell you what you can click to. But a serious question would be whether a company could place an ad with you referring specifically to an off-label use. That is the sort of question I was raising.

I hear from David Vance basically saying probably not. Not too many other people have spoken on that question, but that would be the very sort of thing we are interested in.

DR. LUNDBERG: In order to sense the group and our timing, I would appreciate a show of hands of all people in the audience who would like an opportunity at the microphone if there is time to ask questions or make a statement. Would you please raise your hand if you wish such.

[Show of hands.]

DR. LUNDBERG: There are only five or six. I still think we better go out there and start because it might proliferate.

It is 10:25. We have the last 15 minutes allocated for discussion and questions from the audience to anyone. Please approach the microphone, identify yourself and your organization and then make your statement or ask your question.

MR. CAVALLINI: My name is Mario Cavallini with Slack, Incorporated. I am part of the Webmaster team and I am here to teach, particularly to clear up some of the misinformation that came up during the discussion of Question No. 3.

Consider a typical Internet user, Nell Pureheart. One night, her husband, Fang, decides to use the computer, goes to visit several sites, winds up at the home page for disreputabledrug.com. At dub.dub.dub.disreputable.com, they find a link to goodguys.org which happens to be one of the sites that I serve on my server.

I operated a plain vanilla server equal to any other baseline type of service. I can tell, no matter what browser that person is using, the following things; I can identify Nell Pureheart's computer as the computer that is requesting information for my server.

I can tell that the last page that Nell Pureheart's computer visited was at disreputable.com. I can tell what type of computer Nell Pureheart's computer is. I can tell what type of operating system it is using. I can tell what version of browser softwear is being used. Plain vanilla server I cannot trace back in the history of the quickstream.

The sorts of things I can do with this information. I can, in the future, look at my logs and see the footprint of that event. If I decide that I don't want that sort of traffic again, I can block Nell Pureheart's computer. I can't necessarily block Fang. I can block the computer which will also block Nell.

Also, I hesitate to say anything is impossible but so far as I am aware, with plain vanilla servers, I cannot block out accesses that are coming from the link at disreputable.com.

Now, again, using plain vanilla server and doing routine statistics logs, I can get a fairly good sense of who is linking into my sites. If you are operating a site and you don't know where the links are coming from, you have got a webmaster who is grossly negligent.

I can give you two examples where this is important and I will make this brief. One of the sites that I do operate on, my real site, is a set of various topic-specific Internet directories including one for orthopedics. In doing routine log analysis, I happen to notice I was getting a lot of traffic from an orthopedic surgeon in Taiwan.

I went to take a look at it in a reverse lookup and found out that this person had pirated, had whacked, our entire orthopedic Internet directory and had put it in his home page including, nicely enough, the links to our copyright disclosure. I send an IC mail message to him saying that this didn't make sense, that if he wanted to do the work of maintaining the links and preventing link rot and all that, well, fine, but we do that anyway.

So I mentioned to him to instead put a link to our orthopedic Internet directory.

Another example is that I found a lot of traffic coming from a page that I am not going to identify in any way except to note that this was going into our orthopedic Internet directory from a page that is for people who find braces and casts and other manners of restraint erotic, including such things as plaster of paradise.

Again, I could, if I wanted to, send a letter to that person and say, "Hey; knock it off. We are not that kind of site." But the point of the fact is that anybody coming to us looking for hot babes in plaster is going to be grossly disappointed.

As, as the point that Dr. Stein and others made, there are going to be people who will look for real information or what we consider worthy information such as the efficacy of functional bracing. And that is worth not breaking that link.

DR. LUNDBERG: Thank you very much for your comment.

MR. BOULDING: Mark Boulding, Fox, Bennett and Turner. I have a question for Jamie Marks and, perhaps, for other patient-information organizations that use the sponsorship model. Apparently, there is a really a lack of any specific FDA guidance on the Internet. Have you found that this uncertainty has caused any reluctance on the part of sponsors?

What is your experience? Are they waiting for FDA guidance before they move to work with organizations such as your own?

MR. MARKS: The answer is most definitely they are waiting for guidance. They are being extremely cautious and scrupulous and don't want to risk violating the law in the slightest way. The sponsorship that we have received has been in the form of independent educational grants, and the sponsors stop at that, don't put up promotional materials, don't put links to their own sites.

So I think that there is tremendous interest, that our resource which is award winning and has served, I think, 200,000 people so far and potentially can grow even more and can grow better but we are definitely waiting as sponsors wait to see what they may do and how they may move forward.

DR. LUNDBERG: Thank you for the question. One other response?

MR. KUROMIYA: Critical Path Project provides or hosts webspace for many, many organizations in the Philadelphia area and free Internet access for both individuals, the staffs of these organizations. We do not actively look for any sponsorship. We prefer no conflicts of interest at all.

MR. McKILLEN: Dan McKillen from Soundside Consulting. In February, I moderated a discussion group of healthcare information services on the Internet where we discussed linking. I moderated another group in June where also FDA officials attended and 30 pharmaceutical manufacturers.

We came to some of the same conclusions that you have come to here today when we discussed linking, particularly some of the comments made by Dr. Stein and others that, in terms of linking into your site, there is a lot you can do to control them because people purchase links, they barter links back and forth, they purchase key words.

You can go to Infoseek and buy the word "cancer." Any time somebody types in the word "cancer," you can buy that word and have a banner link right to your site. So there is quite a bit that you can do to arrange links. Now, some of the search engines are actually letting you pop in the first ten search results if you pay them enough money.

There are unrestricted educational grants and other ways to arrange links to your site. An article that was published this year in Wired Magazine suggests that there is a new website that goes up every three seconds on the Web. To think that you can have enough financial resources and personnel resources to monitor every time somebody links to your site is utterly ridiculous.

That was the consensus we reached and I hear that coming from here as well. Linking out is another area, and you can use already established standards that are out there for linking out--scientific accuracy, independent, balance, rigor, et cetera--which I am hearing from this group as well, that if you just exercise reasonable efforts to monitor these because a lot of the sites that you link to on Tuesday can change on Wednesday and there can be other information.

So using that standard that is already established in other media I think makes sense. However, many people may not realize that about three years ago on the Internet, there were some discussions about treating ulcers not as an acid problem but as an infectious-disease problem.

DR. LUNDBERG: What a crazy idea.

MR. McKILLEN: And using aspirin to prevent heart attacks long before this became vogue in medicine. A lot of those sites were not necessarily put up by the AMA or the New England Journal or other pillars of medicine. So it is not really cut and dried in terms of where the good information may be.

There may be some excellent information out there that does not come from a good peer-reviewed site out there.

DR. LUNDBERG: What I suggest is there may be and how does the viewer who is not sophisticated have any idea?

MR. McKILLEN: The last point is that the exit-door strategy, as I call it, saying, "You are now leaving the site--" a lot of times when you don't realize you are linking within the site or somewhere else. I think the exit-door strategy is a good one. I have seen the people at Tiazac--tiazac.com--that do a good job of that.

DR. LUNDBERG: Thank you very much. Good comments.

MR. ARMSTRONG: Neil Armstrong, Teleflex Medical Group. Once or twice this morning, we have heard the word "reasonable." Let's not forget that. Particularly a lot of medical device companies out there are very small firms. They don't have the in-house computer technical expertise to do, perhaps, everything that a large pharmaceutical company can.

We don't know who is looking at our site. We struggle to get something on the Web by using subcontracted services. We need to be out there but we are never going to know who is on our site. However, from the people we've met through our site, we do believe people who get there are not computer-illiterate. They can't be to get to our site.

If our site has a crimson background and they suddenly move onto one with a yellow background, they know they have got onto a different site. We don't always have to tell people the obvious. There is a reasonable viewer, a reasonable browser out there.

DR. LUNDBERG: Thank you. Very nice comments.

MS. TODD: Leslie Todd, Vivus, with another opinion from a small company of an economic nature. In terms of monitoring our links to other sites, if we were to have them, it would be financially unreasonable to monitor every day, every week, every hour, a lot of other sites.

Does that limit us to just a few sites and if it is just a few sites, does that make it not so much of a fair balance?

The other economic point I wanted to make is in technology. While you may be able to have someone develop a webpage for you for $1,000 and then maybe pay $100 a month to have that site hosted, to add a password protection somewhere may be another $5,000. To add a graphic somewhere may be another $10,000.

To go on with monitoring sites, telling what kinds of servers someone has, where they are coming from, is it an icky site, that may be another $20,000. So it is real quick to get up in the tens of thousands of dollars. One example is we got a quote from one company for a full-fledged site that might do all of that for $230,000.

Thanks.

DR. LUNDBERG: Thank you very much. I have the yellow light. I have about 30 seconds of closing comments. I'm sorry we were not able to recognize the other people at the microphones.

In summary, the Internet is a wonderful new technology. If used properly, it can benefit people greatly. Being a combination of nearly all forms of human communication, including speech, in a country with constitutional guarantees of free speech, the Internet will be very difficult to regulate.

The sheer volume of websites, a new one, by our understanding, each four seconds plus the international nature of the Web and the potential anonymity of the authors renders regulation extremely difficult.

President Clinton says that government's job is to give people the tools to do things better. I agree. I also agree with the First Lady that it takes a village. We have had a village here today to help each other in this difficult area.

The most effective regulation on the Internet, as in all other activities, will be professionals with the public interest in mind regulating themselves. The absence of professionalism is self-governance. We have heard a lot of good self-governance here today.

I want to thank all of our presenters, thank the FDA, again, for sponsoring this. I believe we recess now and reconvene at 11 o'clock here.

DR. NIGHTINGALE: I want to thank you very much for your excellent moderating of the session. I want to thank the discussion panelists, the FDA and the audience. I want to say that we are going to come back at 11:00. There are going to be presentations at that time. There will be no discussion topic, members seated or panelists seated. We are just going to have the presentations from the international group.

[Break.]