Steven L. Dubovsky, MD

InterPsych  (IP),  an  international  organization of  electronic
mental  health forums  for professionals  and patients, is one of
the  best organized  and most influential  of a growing number of
Internet-based  groups in  which clinicians  and  patients  share
information.  One  of the  most important functions  of IP forums
is  to provide  a medium for  consultation about difficult cases.
On lists  restricted  to mental  health professionals, clinicians
regularly  provide  brief  case  vignettes  and  ask  for  advice
about  clinical management.  These  kinds  of discussions  can be
very useful, but they can also be risky. 

One obvious issue is that no Internet discussion is confidential.
Even if  access to a closed list  could truly be  restricted, and
even if  patient material  is disguised,  the possibility  exists
that  an individual could  be identified.  To  protect the confi-
dentiality of the clinician-patient relationship, many therapists
feel that it is necessary  to ask a patient's permission  to dis-
cuss any information with a third party for any reason, and there
is no  reason to think that the same should not be true of Inter-
net  discussions.   Yet  one  rarely  sees  acknowledgement  that
clinicians  have obtained  permission from their patients  before
describing them on an electronic forum. 

Research on the process  of clinical supervision  is beginning to
suggest  that the impression  that is gained  by hearing about  a
patient  does not correspond well  with the  impression gained by
actually examining the patient.  Internet consultations obviously
do not involve direct observation of  the patient,  and generally
time and space considerations  as well as difficulty reading long
paragraphs on a  computer make the  information that is presented
fairly sparse.  However, the marginal amount  of patient material
does not seem to deter "consultants"  from offering advice,  some
of it cautious  and some of  it rather bold.  When  one considers
that many of the patients about whom advice is asked have complex
illnesses,  the certainty  with which  suggestions are offered is
interesting. If predictable follow-up were available, it might at
least be possible for  readers to know whether the advice worked.
However,   much  of  the  time we  can  only  guess  whether  the
suggestion was followed and if so, whether it helped or hurt.

Since  people don't  ask for  advice  when things are going well,
most  recommendations  offered  on the  Internet  do not  involve
standard approaches, which have already failed.  There is nothing
wrong  with offering  untested  treatments  so  long as  everyone
understands the nature  of the suggestions.   But to a surprising
degree,  suggestions are offered with a greater sense of authori-
ty than would be justified by data about the patient or, for that
matter, about the treatment.  Recent discussions on several lists
devoted to psychiatric medications illustrate this issue.

Last year,  many Internet psychopharmacologists were recommending
both  dexfenfluramine  and  phentermine/fenfluramine   (phen/fen)
combinations  for  weight loss.  Since weight  gain is  a  common
problem  with antidepressants  and a  number of other psychiatric
medications,  this is  an important issue.   A number  of experts
reported using these medications frequently based on a conviction
that  they were safe  and  effective.   What was missing from the
discussion  was  the  proviso  that  solid data  supporting their
assertions did  not  exist.  In fact,   a randomized study of 822
patients  (average weight=97 kg) found a mean weight loss of only
9.8 kg with  dexfenfluramine compared to 7.2 kg with placebo (1).
Another study showed sustained average weight loss of 9.4 kg over
3 years  with  dexfenfluaramine  -  generally 10% or less of body
weight (2). 

How safe were the widely recommended medications?  Shortly before
people on the Internet  began singing the praises of dexfenflura-
mine,  a large case controlled  study  in four European countries
showed  that the risk  of primary  pulmonary  hypertension (PPH),
which cannot be detected until it is irreversible and potentially
fatal,  was increased tenfold in obese  patients who took dexfen-
fluramine,  with a  somewhat smaller  increased risk  in patients
treated   with    fenfluramine (3).    Patients  who  took  these
medications  for more than 3  months were  a  little  more   than
23 times as likely as untreated obese controls to develop PPH. In
a  study  of 32 Swiss patients who  developed PPH on the appetite
suppressant aminorex during the 1960s  and 1970s,  four had taken
the drug for a month or less,  and 7 patients did not develop PPH
for 3 months to a year after treatment was discontinued (4). 

I was interested in the reaction on the Internet to a report from
the Mayo Clinic  and the  MeritCare Medical Center in Fargo, N.D.
of 25 obese women with no previous history of cardiac disease who
had  been treated for  about a year with phentermine/fenfluramine
combinations  (phen/fen)  and   who  were  noted  during  routine
clinical  evaluations  to  have heart  murmurs.   These  patients
received  echocardiography (ultrasound of the heart)  that showed
damage to the aortic, tricuspid, and/or mitral valves (5).  Eight
of the patients were also found to have PPH,  and 5 of the entire
sample required surgery  on their heart valves.  Debate continues
about whether high blood serotonin levels  caused by the phen/fen
combination  could account for the damage,  but at about the same
time  a public health  advisory from  the U.S. Food  and Drug Ad-
ministration noted that another 9 cases of heart valve damage had
been reported since the Mayo/Fargo study (6). In response to this
news,  which was of  course rapidly disseminated on the Internet,
the  same psychopharmacologists  who had recommended weight  loss
drugs unequivocally began to advise against them, as if they were
right on the cusp of the latest breaking news. I recall seeing no
expressions  of  concern that  international leaders  in  psycho-
pharmacology had missed the boat and  no fears that they may have
been  a little too  quick to expose their patients to unnecessary

In  the case  of  phen/fen  and  dexfenfluramine,   the  greatest
strength of Internet  consulting was  also its greatest weakness.
It is tremendously appealing to hear about the latest advance and
the newest treatment,  and it is more than a little gratifying to
be able to tell others something new.  However, in our rush to be
on the cutting edge and try the latest and greatest,  we may be a
little too quick to accept opinions as authority, failing to make
the distinction between an observation in one or two patients,  a
few clinical trials,   and a prospective controlled study  -  the
"gold standard" of proof in clinical science. 

A few weeks ago, a psychiatrist I know asked me what I knew about
lamotrigine,  a new  anticonvulsant  medication that  has  gained
rapid popularity  as mood stabilizer in  bipolar mood disorder. I
said  that  there were  not yet  any controlled studies  of  this
application,  so it could only be considered experimental. He re-
plied proudly that he knew of such a study,  and he was surprised
that I wasn't familiar with it. I asked to see it, and he sent me
a list of "lamotrigine FAQ's" based on the experience of a single
expert.   To this consumer of Internet data, the boundary between
fact  and opinion was blurred in  the same manner as the boundary
between intuition and knowledge.

This  tendency to  take the  first new thing  one hears  about as
gospel may  not  be scientific,  but it  is hardly limited to the
Internet. Lamotrigine and gabapentin, another new anticonvulsant,
have come into  widespread use,  sometimes  as initial treatments
for  bipolar  disorder,  in  the  absence of  a single  published
controlled  study  of a  short-term  antimanic  effect.   Neither
medication has been available for a long enough period of time to
be  subjected  to any kind of study of their ability  to  prevent
recurrences of mania  or depression, but many patients are taking
them as mood stabilizers.  Do they work?  Are they safe?  In most
other  clinical specialties,  these questions  would be as much a
subject of discussion as the treatments themselves. 

Why  do we  seem so  quick to  accept as  established  treatments
approaches are supported only by  the conviction  that they help?
Perhaps  our  credulity is being strained by our need for answers
to the difficult clinical  questions that plague  us in an era in
which  all of the easy disorders are being treated by specialists
outside the  mental health field.   Perhaps we  are so anxious to
feel that we are at the  cutting edge  that we ingest information
without chewing,  failing to distinguish between intriguing ideas
worthy of formal study and the latest and greatest treatment. But
in our rush to be in the know,  we may lose that critical faculty
that is essential to our performance as clinicians.

The  warp speed with which information circulates on the Internet
is a blessing, but uncritical acceptance of the published word of
opinion leaders has  the potential to  be a curse.   No clinician
would  accept  the initial  assessment of  a patient  as the last
word. If this did occur, the patient would end up being forced to
conform  to the  therapist's first  opinion,  and the opportunity
would be missed to  revise the initial hypothesis in the light of
further data and find the best approach for that patient.  As the
Internet  matures,  discussions will  hopefully  contain the same
level of data reporting and  critical appraisal of  one's own and
others'  statements that  would be  expected  in  any  scientific


1. Guy-Grand B, Apfelbaum M, Crepaldi G, Gries A, Lefebvre P, Turner
   P. International trial of long-term dexfenfluramine in obesity. 
   Lancet 1989;2:1142-1145. 
2. Rosenbaum M, Leibel RL, Hersch J. Obesity.  N.Engl.J.Med. 
3. Abenhaim L, Moride Y, Brenot F, et.al. Appetite- suppressant drugs 
   and the risk of primary pulmonary hypertension. International Primary 
   Pulmonary Hypertension Study Group.  N.Engl.J.Med. 1996;335:609-616. 
4. Mark EJ, et al. Fatal pulmonary hypertension associated with 
   short-term use of fenfluramine and phentermine.  N.Engl.J.Med. 
5. Connolly HM, Crary JL, McGoon MD, et al. Valvular heart disease 
   associated with fenfluramine-phentermine.  N.Engl.J.Med. 
6. Nightingale SL. Health advisory on concomitant fenfluramine and 
   phentermine use.  J.A.M.A. 1997;278:379-379.

Dr Dubovsky is professor of psychiatry and medicine at the University 
of Colorado School of Medicine and vice chair of the Department of 
Psychiatry. He has been president of InterPsych in 1997. His research is 
in mood, psychotic and psychosomatic disorders. He consults on difficult 
patients around the United States and specializes in treating mental 
health professionals and physicians. His latest book is "Mind-Body 
Deceptions" (WW Norton, 1997).