Hanky-Panky in the Pharmaceutical Industry

March 5, 2002

Volume 1, Number 1 --- Summer 1996

INTERNATIONAL JOURNAL OF PSYCHOPATHOLOGY,

PSYCHOPHARMACOLOGY AND PSYCHOTHERAPY

ISSN: 1088-6710

----------------------------------------------------------------------------

----

Hanky-Panky in the Pharmaceutical Industry.

Seymour Fisher, Ph.D.

Center for Medication Monitoring

Department of Psychiatry & Behavioral Sciences

University of Texas Medical Branch

Galveston TX 77555-0441

----------------------------------------------------------------------------

----

ABSTRACT

This paper reprints three essays that were originally distributed on the

Internet, to the Psychopharmacology Forum of InterPsych, in the latter half

of 1995. Part I is a factually based account of how a pharmaceutical company

unethically intervened in the publication of a scientific article reporting

clinical results inimical to that company's interests. A follow-up Part II

offered additional documentation of drug companies' influence on academic

research, and further raised questions as to the industry's pervasive

influence on the training of young physicians. Part III revealed some of my

thinking behind the hope that the public can be aroused to bring moral

pressure on the pharmaceutical industry.

Cite as: Fisher S Hanky-Panky in the Pharmaceutical Industry. Int J

Psychopath Psychopharmacol Psychother 1996, 1 (1). URL

http://www.psycom.net/ijppp.v1n1.html

PART I (First posted on the Internet in late July 1995)

Does the pharmaceutical industry want clinicians and patients to learn more

about possible side effects of newly marketed drugs? I think the time has

come to make public just one egregious example of how individual drug

companies can influence the publication of clinical research results that

are not in their best financial interests. (I also have documented instances

of how insidiously the pharmaceutical industry can influence publication of

other manuscripts and even NIH support of research projects dealing with

adverse drug reactions of newly marketed drugs. But that's another story

for, perhaps, another time.)

First, however, for those who do not know me, I'd like to point out that I

have nothing to gain personally by going public with this issue. I'm just

about 70 years of age, and my academic credentials and career don't need any

embellishing (a brief resume can be found in " Who's Who in America " ).

Next, I urge you to read the article on " Postmarketing Surveillance by

Patient Self-Monitoring: Preliminary Data for Sertraline versus Fluoxetine "

in the July, 1995 issue of the Journal of Clinical Psychiatry

(1995;56:288-296). This paper is based on large-scale data indicating that

many adverse reactions known to be induced by fluoxetine (Prozac) were being

reported with even greater frequency by sertraline (Zoloft) patients; the

tables also include suggestions to the clinician for age and gender patient

types most at risk. Zoloft is manufactured by Pfizer Incorporated (Roerig

Division).

The manuscript was accepted for publication on May 12, 1994.

On December 8, 1994 the Editor wrote me to say that he had become " concerned

that our largely clinician readership might interpret the results more

literally than our investigator colleagues. This apprehension led me to

draft the accompanying commentary, which I would like to publish along with

your article. " Although none of the journal's three reviewers who had

originally recommended publication voiced this apprehension, the Editor's

proposed commentary was entitled " What will this drug do to me, doctor? " ,

and tacitly implied that our results and conclusions might be spurious.

I replied to this letter on December 20, showing that most of the

substantive criticisms he raised in his proposed commentary were simply not

valid, suggesting instead that the research results along with the article's

carefully qualified discussion of the results should be able to speak for

themselves.

Letter from the Editor dated December 30: " I have revised and (I hope you

will agree) 'softened' some of my comments. I hope you will be more

comfortable with the current draft. " His revised commentary included

sentences such as " It would be simplistic and premature, however, to treat

this report as gospel and conclude that in reality sertraline produces a

higher frequency of unwanted reactions than does fluoxetine. " And the final

paragraph was to be: " The report by Fisher et al. is thought- provoking and

can frame hypotheses for additional testing. The actual incidence of side

effects of these two SSRIs will become clearer with time and additional

study. " (Similar caveats were actually included in the discussion section of

the article, but without the pejorative flavor of the proposed Editorial.)

By February of 1995, when we had not yet received page proof nine months

after acceptance of the article, I phoned the editorial office for

information. I was told it was scheduled for the May issue. However, in

April when we had still not received either page or galley proof, and when a

follow-up phone call elicited the information that the publication date was

now postponed until July, I undertook a quickie " research project. " This led

to a letter I wrote to the Editor on May 1, in which I expressed the view

that publication of his proposed Commentary would be grossly unfair unless I

was also given the opportunity to respond to the Editorial. What follows was

my proposed rebuttal:

COMMENTARY ON " What will this drug do to me, doctor? "

In this issue, an article by Fisher (Fisher S, Kent TA, and SG, 1995)

presents data from more than 2,700 fluoxetine and sertraline patients using

a well-validated postmarketing surveillance method developed to signal

possible adverse drug reactions (ADRs). The preliminary results indicated

that many adverse reactions known to be induced by fluoxetine were being

reported with substantially greater frequency by sertraline patients. The

article is accompanied by an Editorial Commentary (Gelenberg AJ, 1995),

admonishing readers not to " conclude that in reality sertraline produces a

higher frequency of unwanted reactions than does fluoxetine. " Certainly this

could be a premature conclusion to draw. But a legitimate question can be

raised as to why this particular paper is being singled out when the implied

" conclusions " in more than 90% of the papers published in this Journal and

in other psychiatric journals are also generally subject to alternative

interpretations, not all of which may be equally plausible.

The Editor notes that a bias could have been introduced because we relied

" on a comparatively small percentage of volunteers [almost 20%] out of an

approached population. " But all postmarketing surveillance studies use only

a minute sample of the total population of interest (Baum C and Anello C,

1989). The more salient question is whether there is reason based on

empirical evidence to believe that the final selected samples favor one drug

group over the other. If selection causes a bias in our method, we should

not have been able to detect in our validation studies so many of the

commonly accepted ADRs for various drugs (Fisher S, 1995; Fisher S,

SG and Kent TA, 1993). However, it is always possible in any postmarketing

surveillance method that volunteer subjects (including physicians who are

urged to report possible ADRs to the FDA) or even medical record samples

could introduce a bias. Similarly, although the Editor questions " whether

this technique is well suited for comparing incidences of adverse events

between a newer and an older agent, " he also acknowledges that results from

our past studies along with the statistical controls used in the data

analyses suggest that what we were seeing in this sertraline study is not

simply a " newer drug " phenomenon.

So, again, why the red-flag editorial? A review of 119 articles published in

this Journal from July 1993 through April 1995 (excluding supplements,

monographs, and the October 1994 issue, which was unavailable) offers some

clues. The mean article length was slightly less than six pages (skewed

upward by a few longer papers); the mean publication lag, defined as the

number of months between the date of acceptance and the published issue

date, was eight months -- for which most authors are grateful to the Editor.

Only two of the 119 articles were not published until 11 months after

acceptance, and none had a lag of one year or more. There was no

relationship between the length of an article and the publication lag.

While some issues of the Journal included a " commentary " on a specific

paper, none of them were signed by the Editor. In fact, a cursory search

through issues dating back to 1990 found only one previous Editorial, which

also focused on adverse drug reactions (Gelenberg AJ, 1992). Yet, our

sertraline paper not only prompted an Editorial, but publication was delayed

more than a year after it was formally accepted on May 12, 1994.

During the past four years of the 10-year development of our postmarketing

method [continuously supported by the National Institute of Mental Health

along with other nonpharmaceutical funding sources], we have became acutely

aware of the fact that, once a new drug has been marketed, many

pharmaceutical companies clearly do not want their drugs to be carefully

monitored for possible ADRs -- in particular, not by any method that can

systematically and sensitively compare possible ADR profiles. The Editor of

this Journal is to be commended for having the courage to publish our

sertraline/fluoxetine paper, but one cannot help wonder to what degree

external pressures may have contributed to both the publication delay and

the need for a cautionary Editorial.

Presently, the ultimate clinical preference for one psychopharmacological

agent over another is mainly determined not so much by true differences in

therapeutic efficacy (most antidepressants in most situations are about

equally effective) but by presumed differences in their ADR profiles

(Gelenberg AJ and Schoonover SC, 1991). Systematic health services research

carried out in the real world of postmarketing pharmacotherapy is of

paramount importance for clinicians to be competent to practice

empirically-based, rational patient care. The real bottom line here is that,

although publication of our paper could have an adverse effect on company

sales, sertraline seems to have a more troublesome ADR profile than

fluoxetine, particularly in respect to those known ADRs that appear to be

common to the SSRI class. But only continued astute clinical observations

and systematic research will judge whether these preliminary results based

on patient self-monitoring indeed help provide more accurate answers to the

patient's question, " What will this drug do to me, doctor? "

[End of rebuttal]

About two weeks after I had sent the above proposed Editorial Reply, I

received a phone call from the Editor while he was attending the American

Psychiatric Association meeting in Miami, saying he could not possibly

publish my commentary in its proposed form. He agreed that I was entitled to

space for rebuttal, but informed me that I would have to modify its

contents. I said I would consider this. Then, just a few days later, he

phoned again to say that he had decided to drop his proposed editorial, and

that therefore I could forget about the rebuttal. In our discussion, he

admitted that this decision was made after conferring with Pfizer

representatives at the APA meeting.

So you will not find in the July, 1995 issue of the Journal of Clinical

Psychiatry any Editorial Commentary or reply to accompany the Fisher et al.

article -- finally made available to clinicians (and their patients) a full

14 months after acceptance.

Some obvious questions arise from this sequence of events:

1. Why did the Editor wait a full seven months after the paper's acceptance

to decide that he should write an editorial to accompany publication of the

article? Did it take a few months for word to get back to Pfizer, who

heavily subsidizes the Journal, that the paper was in press?

2. Why was final publication of the article delayed for 14 months, when most

articles were being published in about eight months and no other article in

that Journal between 1993 and 1994 had had a lag more than 11 months? Did

Pfizer want to see publication put off as long as possible while its Zoloft

sales were going strong? (Odd coincidence: note that the lead article in

this July issue of the Journal is not only authored by the Editor but claims

to have been accepted way back in April 1994.)

3. And why did the Editor decide at the last moment, after consulting with

Pfizer representatives, to drop the idea of writing the Editorial? After

reading the proposed rebuttal commentary, was Pfizer now concerned that

publishing the editorial and the rebuttal, in addition to emphasizing the

apparently unfavorable side-effect profile for Zoloft, might actually serve

to increase the readers' awareness of Pfizer's role in the publication

process?

PART II (First posted in August 1995)

About one month ago I posted to a number of different Internet mailing lists

and newsgroups a fact-based chronology involving publication of an article

on postmarketing surveillance, showing how one drug company (Pfizer

Incorporated) blatantly stretched the ethics envelope when its profits were

threatened. The article, published in the July issue of the Journal of

Clinical Psychiatry (1995;56:288-296), was based on large-scale data

indicating that many adverse reactions known to be induced by fluoxetine

(Prozac) were being reported with even greater frequency by sertraline

(Zoloft) patients. Zoloft is manufactured by Pfizer. So far I've received

about 100 replies (many more from mailing lists than from newsgroups), all

highly supportive.

Some readers interpreted my original post as appropriately taking aim at the

Editor of the Journal of Clinical Psychiatry. I had hoped it was clear that

my frustration was directed not so much at the Journal as at drug company

advertisers, who can make life very uncomfortable for even the most

conscientious and scrupulous journal editor.

Now I'd like to offer as Part II the following additional related facts, in

the hope that more of you whether clinicians, scientists, ethicists, or

simply concerned consumers of prescription drugs will become aware of the

many ways the pharmaceutical industry attempts to unethically influence the

acquisition and dissemination of knowledge about their drugs to medical

practitioners and their patients. Colleagues who know me will, I believe,

agree that I'm not prone to either " sour grapes " reactions or paranoid

ideation, but what has gradually emerged over the past four years is a

not-too-pretty picture of concerted unethical efforts being made to

influence psychopharmacological publications, education, and even research

support.

In our systematic research on measuring the frequency of adverse reactions

for newly marketed drugs, this is not the first time my colleagues and I

have obtained evidence of drug companies exerting pressure upon journal

reviewers and editors. And it wasn't just one company, Pfizer. Lilly did

essentially the same thing when we first attempted to publish our fluoxetine

vs trazodone paper, which was subsequently published more than two years

after we had first written it (Fisher S, SG, and Kent TA, 1993).

Nor is the NIH research grant review process immune from similar viruses.

Recently, after 10 years of continuous funding from the National Institute

of Mental Health (NIMH), further support was abruptly terminated, when at

least two members of the original study section -- one being the chairman

(!), another being the primary reviewer -- had been conducting numerous

company-supported drug studies, substantially contributing either directly

or indirectly to the reviewers' overall income. Incredibly, NIH claims there

was no conflict-of-interest in the review because " Research that focuses on

other than assessing the efficacy of a particular pharmacologic agent(s) do

not represent financial conflict of interest situations " (grammar exactly as

written).

Having personally served on NIH study sections, I know that some committee

members who review grants do let drug companies know (unethically) about

research grant applications that might be inimical to the industry's best

interests (e.g., postmarketing surveillance of new drugs). Now, if a member

of the committee is also receiving part of his/her total income by

conducting clinical trials and/or extensive consulting for that company,

then I submit that s/he has a potential conflict-of-interest when it comes

to discussing and voting upon the grant application. There are always leaks

about who voted for or against a particular application, and a committee

member would know that an " approval " vote for a grant that the

pharmaceutical industry would like to see killed could lead to loss of

income.

The bottom line here seems to be that, in NIH's view, a committee member can

be in conflict-of-interest if personal income GAIN might result from his/her

vote. But if an external source should exert subtle intimidation, NIH says

there is no conflict-of- interest even though a vote in one direction might

result in a LOSS of personal income. Does this make sense to you?

Want more food for thought? Are so many of our hospitals and medical school

departments so broke that their residents have to get free lunches

accompanied by drug company representative sales pitches rivaling some of

the best infomercials seen on TV? Of course, this ensures that " truthful "

information about the efficacy and side effects of their product(s) compared

to other similar drugs will be imparted to the residents without any need

for those hard-working physicians-in-training to bother to consult the

published literature.

I've also been struck by the ubiquitous but unobtrusive presence of drug

company representatives at Grand Rounds and other lectures where the

speaker's large honorarium comes from the company. Does Big Daddy/Mommy

watch and listen to make sure that the home office will know if the speaker

says anything bad about its products?

One reply I received to my original post suggested that groups like the APA

(American Psychiatric Association) and NCDEU (NIMH's annual New Clinical

Drug Evaluation Units meeting) should publicly air the issue.

To this, Dr. Ivan Goldberg commented online:

>Maybe I am more cynical than I like to think, but with

>the important role played by " industry " at both the APA

>and NCDEU, I'll bet you $100 that no matter how hard

>any of us push, that no symposium on drug company

>influence over the funding and publication of post-

>marketing studies of psychopharmacologic agents will

>be scheduled at APA or NCDEU.

Most of the key psychopharmacology players at APA and NCDEU also have power

roles in the American College of Neuropsychopharmacology (ACNP). Since I'm a

past president of the ACNP, I have some first-hand knowledge of how

courageously the College (whose members comprise drug company

representatives as well as academicians and government representatives) acts

when the pharmaceutical industry is unhappy about something:

At its 1993 meeting the ACNP had scheduled a luncheon meeting on the very

subject mentioned above in Dr. Goldberg's post. However, at the last minute

the topic was changed to something much more important, like " Can SSRI

antidepressants be used to make basset hounds act like German shepherds? "

For its 1994 meeting the ACNP Council was considering having the editors of

a number of psychiatric and psychopharmacological journals participate in a

study group session open to all members of the College. The topic was to be

a discussion of the ethics of grant and journal reviewers informing drug

companies of submitted grants and manuscripts with content inimical to a

company's interests. As a corollary, the study group would also be expected

to discuss drug company advertisers' pressures upon journal editors. This

proposed study group was never held.

We're talking here about a third-rail issue that no one officially wants to

touch. Aren't there more people out there willing to get involved by

contacting their colleagues, elected representatives, and any media friends

they might have?

Somehow we ought to be able to loudly and clearly warn the " ethical "

pharmaceutical industry that, while we admire and are grateful for their

many therapeutic accomplishments, we will not tolerate their placing dollars

before truth. Excelsior!!

If you've read this far, many thanks for your interest. I hope you'll take

some action.

PART III (First posted in September 1995)

During the past month an assortment of Internet replies have been pouring in

following the Part II posting, some simply requesting copies of Part I, some

offering advice, and others expressing reactions. The many threads under the

headings of Hanky-Panky " and Unethical Practices - Drug Industry " in

different mailing lists and newsgroups (e.g., sci.med.,

sci.med.psychobiology, sci.med.pharmacy, psycho-pharm@...) make for

fascinating reading. A few posts came from loonies (some with doctorate

degrees!) with knee-jerk reactions to any suggestion that there might be

some problems with ethics within the ethical " pharmaceutical industry. One

particular reply, however, came from an obviously well-read and thoughtful

individual. Since his position was so clearly stated and probably represents

the view of many others, I believe it warranted a separate response. The

following is essentially what I sent out on the Internet between September

10-12: * * * * * At 09:33 PM 9/9/95 -0400, (Bill) Boyer, M.D.,wrote:

>Dr. Fisher raises important points in his well-written

>update. However I wish to take issue with (what I see

>as the implication of his article: that there is, some-

>where, somehow, some highly ethical yet practical

>solution to the influence of pharmaceutical companies.

>As Dr. Fisher implies, research costs money and as he

>shows, there are two sources of money, pharmaceutical

>companies and the ever-dwindling resources of the

>government. Without funding, little research of any

>merit will get done. Period.

>

>The best, though not perfect, solution resides in the

>free enterprise system and competetion. This is how

>undue influence is prevented in other areas of the

>economy. As long as there several pharmaceutical

>companies competing with each other to obtain profit

>and avoid loss, as long as there are several scientific

>publications competing with each other for prestige as

>well as advertising, there will be built-in limits to

>how much influence any one of them can exert. This

>will happen not only because their power is diluted,

>but because undue actions by any one of them may open

>up an opportunity for their competitors. I also firmly

>believe that anyone who graduates from a professional

>school is (or was) smart enough to appropriately

>evaluate material presented over a " free lunch. "

>

>I see Dr.Fisher's call to " do something " , including

>contacting our legislators, as a de-facto call for

>increased governmental control, which history shows

>is the surest way to stifle science.

>

>Again, the status quo is not necessarily good,

>but all of the alternatives look worse.

>

> (Bill) Boyer, M.D.

>----------------------------------------------------------

> If you prefer Cogito ergo sum to Non sum qualis eram

> you are putting Descartes before Horace - J. Thurber

>----------------------------------------------------------

I sincerely wish it were true that, as Dr. Boyer views it, the competition

among pharmaceutical companies is sufficient to solve some of the serious

ethical problems that have been facing the industry for many years. Because

he says he sees no " practical solution to the influence of pharmaceutical

companies, " he is content with the status quo. Well, the present status

isn't exactly anything to quo about; therefore, I believe (a) some action is

called for, and ( there ARE viable alternatives.

I too am strongly in favor of competition -- which in most areas can result

in economic, technological, and material benefits for the consumer. However,

history shows that unrestricted competition always leads to disaster for the

consumer. I don't think anyone would argue that the FDA should be abolished

and that individual companies should be left to decide which drugs to

market: Do we expect that competition among drug companies would ensure that

only safe drugs were marketed? And where were the beneficial effects of

competition to the consumer in the days before the Kefauver-

amendments, when a drug didn't even have to be shown to be effective to be

sold in the market place? It is simply untrue that ethical behavior will

prevail " because undue actions by any one [company] may open up an

opportunity for their competitors. " Letting drug companies (or other

companies) compete among themselves unfettered is roughly comparable to

having Mark Fuhrman appointed head of L.A.'s Internal Affairs division. To

save space, I won't elaborate on this point right now, especially since it

doesn't directly bear on the question of " What should we be doing now? "

So let's cut to the chase. As I noted above, there are indeed alternatives

to the status quo, but Dr. Boyer has badly misread my hope. Contrary to his

fear that my " call to 'do something' " calls for increased governmental

control, increased governmental controls are certainly NOT what I'm either

advocating or hoping to achieve (heaven forbid!). The optimal level of

federal regulations on business should be exactly the same as for good

pharmacotherapy: the smallest dosage necessary to achieve the desired effect

(i.e., maximum benefit for the patient/consumer).

I believe Dr. Boyer grossly underestimates the power of the " people " to

effect behavioral change -- even in the highly competitive world of

business. I could list a long string of events where pressure from the

people (possibly including perceived threats from legislators who, after

all, do speak for the people) has led to changes in corporate policies. To

mention just one, look how the drug companies have cut back on their

exotically located " seminars " after the " 60 Minutes " segment on paid

vacations for physicians and their wives. (And please, let's not have

semantic arguments here over the use of the term " vacations. " If you think

the underlying major purpose of those " seminars " was genuine continuing

medical education, then I suppose you're similarly susceptible to all the

tobacco industry's arguments.)

Yes, Dr. Boyer, I believe that if enough people raise their voices, the

resulting publicity can bring enough pressure on the pharmaceutical industry

to take a higher moral road -- and we WILL see some changes.

By the way, why are we all so prone to believe that WE, the fortunately

better educated, the logical thinkers, obviously can't be bought so easily?

This doesn't stand up too well under close examination, when one " free

lunch " after another, whether small or large, is piled up over time and

friendships are developed. Furthermore, we shouldn't be taken in by the

argument that since we're exposed to reps from ALL the different companies,

we're not going to favor just one or two -- isn't there a strong need for

lobbying reform in Washington despite the fact that our elected

representatives are continually meeting with so many varied and competing

special interest groups?

I suspect the only real difference between our views concerns optimism vs

pessimism about corporate nature. I believe that if enough of us speak with

a forceful voice, the pharmaceutical industry will elevate their ethical

standards. Dr. Boyer, why not join me in seeing whether a bit more optimism

is justified? Can't hurt.

REFERENCES

Baum C, Anello C. (1989) The spontaneous reporting system in the United

States. In Strom BL (Ed) Pharmacoepidemiology. New York: Churchill

Livingstone. Pp. 107-118.

Fisher S. (1995) Patient self-monitoring: A challenging approach to

pharmacoepidemiology. Pharmacoepidemiology and Drug Safety. 4; 359-378.

Fisher S, SG, Kent TA. (1993) Postmarketing surveillance by patient

self-monitoring: trazodone versus fluoxetine. Journal of Clinical

Psychopharmacology. 13; 235-242.

Fisher S, Kent TA, SG. (1995) Postmarketing surveillance by patient

self-monitoring: Preliminary data for sertraline versus fluoxetine. Journal

of Clinical Psychiatry. 56; 288-296.

Gelenberg AJ. (1992) Imperfect drugs in an imperfect world. Journal of

Clinical Psychiatry. 53; 39-40.

Gelenberg AJ. (1995) " What will this drug do to me, doctor? " Journal of

Clinical Psychiatry. 56; 00-00 [Personal communication, subsequently never

published].

Gelenberg AJ, Schoonover SC. (1991) Depression. In Gelenberg AJ, Bassuk EL,

Schoonover SC (Eds) The Practitioner's Guide to Psychoactive Drugs. (3rd

edition) New York: Plenum. Pp.23-89.

ACKNOWLEDGMENTS

The Center for Medication Monitoring's postmarketing surveillance studies

have been continuously supported by grants from the National Institute of

Mental Health, the Hogg Foundation for Mental Health, and the University of

Texas Medical Branch's institutional sources.

March 5, 2002