Turning Medicine

Into Snake Oil HOW PHARMACEUTICAL MARKETERS PUT PATIENTS AT RISK

MAY 2006

APRIL 2006 NEW JERSEY PUBLIC INTEREST RESEARCH GROUP i Turning Medicine Into Snake Oil HOW PHARMACEUTICAL MARKETERS PUT PATIENTS AT RISK

Authored by Abigail Caplovitz ©2006 The State PIRGs

The State PIRGs

ACKNOWLEDGEMENTS

Th e State Public Interest Research Groups (PIRGs) © 2006, NJPIRG Law and Policy Center and the gratefully acknowledge Dr. Bob Goodman, State PIRGs. Assistant Professor of Clinical Medicine, Division of General Medicine, Columbia University College Th e State PIRGs are a network of independent, of Physicians and Surgeons, and Dr. Ronald state-based, citizen-funded organizations that Caplan, Associate Professor of Public Health, Th e advocate for the public interest. Since 1970, we have Richard Stockton College of New Jersey, for peer been delivering results-oriented citizen activism review. to protect our environment, encourage a fair and sustainable economy, and foster a responsive Authored by Abigail Caplovitz, Consumer democratic government. We uncover threats to Advocate at NJPIRG Law and Policy Center. public health and well-being and fi ght to end them, using the time-tested tools of investigative research, Th e author also wishes to thank Dena Mottola, media exposes, grassroots organizing, advocacy and executive director of the NJPIRG Law and litigation. Policy Center, and Alison Cassady, researcher for the State PIRGs, for their tremendous assistance. For additional copies of this report, send $10 Th anks also to Emily Clayton of CALPIRG (shipping included) to: Education Fund for her review. Special thanks to Alec Meltzer for layout. NJPIRG Law and Policy Center 11 North Willow Street Th e author alone bears responsibility for any factual Trenton, NJ 08608 errors. Th e recommendations are those of the www.njpirg.org State PIRGs. Th e views expressed in this report are those of the author and do not necessarily For more information about the State PIRGs, visit express the views of funders of the State PIRGs or www.pirg.org. NJPIRG Law & Policy Center. ii TURNING MEDICINE INTO SNAKEOIL Contents

EXECUTIVE SUMMARY ...... 1

SECTION 1: INTRODUCTION ...... 5

SECTION 2: DECEPTIVE ADVERTISING FINDINGS ...... 7 Deceptive Drug Advertisements Are Widespread ...... 7 False Advertisements Often Contained Multiple False Messages ...... 8 Deceptive Advertising is Dangerous ...... 10 The FDA Letter Data Grossly Understate the False Advertising Problem ...... 13

SECTION 3: THE FDA’S FAILURES ...... 15 FDA Letters Go Out Too Late to Be Effective...... 15 Less Than A Quarter Of The Letters Call For Corrective Advertising...... 15 The FDA Letters Do Not Deter Future Deceptive Advertising ...... 15

SECTION 4: FALSE MARKETING BEYOND ROUTINE FDA REVIEW ...... 19 Sales Representatives: What They Say Matters ...... 19 Clinical Trials ...... 21

SECTION 5: SOLUTIONS ...... 25 Steps the States Can Take ...... 25 FDA Approaches ...... 26 Improving FDA Enforcement ...... 26 The Medical Profession’s Role ...... 27

SECTION 6: CONCLUSION ...... 28

APPENDIX A: CASE STUDIES OF DECEPTIVE MARKETING ...... Center Spread Introduction ...... A1 Case Study 1: Vioxx, Misrepresenting Potentially Fatal Risks ...... A1 Case Study 2: OxyContin, Misrepresenting Addiction Risks and Broadening Its Use ...... A4 Case Study 3: Paxil CR, Concealing Suicide Risks and Promoting Unnecessary Medication ...... A6 Case Study 4: Neurontin, Using Off-Label Promotion as the Official Marketing Strategy ...... A8 Case Study 5: Accutane, Concealing Risks and a Weak Regulator ...... A9 Case Study 6: Tindamax, Disregarding the FDA ...... A10

ENDNOTES ...... 29 M A Y 2 0 0 6 THE STATE PIRGS iii Index of Tables

TABLE 1 Types of Deceptive Marketing Aimed at Doctors ...... 7

TABLE 2 Types of Deceptive Marketing Aimed at the Public ...... 8

TABLE 3 Number of Ways FDA Identified Ads as False or Misleading, 2001-2005 ...... 9

TABLE 4 General Recidivism: Companies Receiving More Than One Letter in the Study Period ...... 16

TABLE 5 Specific Recidivism: Companies Receiving More Than One Letter On the Same Drug and Same Issue ...... 17-18

TABLE 6 Citing Studies for Propositions They Do Not Support ...... 22 1 TURNING MEDICINE INTO SNAKEOIL Executive Summary

False and misleading prescription drug advertising Recidivism is rampant. is common and dangerous. Prescription drug • 28 companies—approximately 1/3 of the total— marketers are inundating doctors, and to a received more than one letter declaring their ads lesser extent, the public, with marketing that false or misleading in the fi ve years we examined. misrepresents risks, promotes unproven uses, In fact, these companies accounted for two-thirds and makes unsubstantiated claims. Th e false and of all the letters received. misleading messages are communicated through conventional advertising, sales representatives, • 26 companies received more than one letter doctors speaking on behalf of drug marketers, and relating to advertising for the same drug that was through clinical trial suppression, manipulation deemed false or misleading in the same way. and misrepresentation. Sadly, the Food and Drug Administration (FDA) is ineff ective at addressing Deceptive marketing includes sales the problems. Th is report takes a comprehensive representatives. look at all of these facets of the prescription • Sales representatives, as a group, form long and drug marketing problem and suggests eff ective deep relationships with doctors, beginning in solutions. medical school. Research suggests those early relationships increase doctors’ receptiveness to sales representatives once they are in practice. FINDINGS IN BRIEF • Perhaps refl ecting those relationships, other We looked at enforcement letters FDA sent to drug research has shown that sales representatives have marketers from 2001-2005. Our research reveals: a profound infl uence on prescribing decisions. Deceptive drug marketing is pervasive, dangerous, and primarily aimed at doctors. • Sales representative statements accounted for 30 • From 2001-2005, 85 companies received 170 of the 869 deceptive messages in the FDA letters, notices from the FDA explaining that the an amount that is enormous given the very small marketing for 150 diff erent drugs was false and/or percentage chance that the FDA will detect such misleading. statements. Other research suggests that as much as 11% of sales representative statements are false • 62% of the false or misleading messages targeted and favorable to the product they pitch. doctors, and those messages were expressed by 38 diff erent types of advertising. By contrast, the Deceptive marketing includes clinical trials. public was exposed to 17 diff erent types of false • In the letters identifying advertising as false or or misleading ads. misleading because it contained unsupported claims, FDA highlighted at least 82 times that • Th e false messages were serious: 35% the advertising cited clinical trials for propositions misrepresented risk; 22% promoted unproven they did not support. In some instances, the cited uses; and 38% made unsupported or misleading trials even contradicted the claims. claims. For deceptive messages targeting doctors, 37% misrepresented risk; 24% promoted • Drug marketers turn clinical trials into marketing unproven uses; and 36% made unsupported or tools by suppressing some unfavorable data; by misleading claims. using PR fi rms to write favorable reports (the PR fi rm does not appear as an author of the report, M A Y 2 0 0 6 THE STATE PIRGS 2

instead a doctor is retained to be the named trials. Drug marketers would register every author); by misrepresenting unfavorable data clinical trial done in humans for every drug they that is published; and, most subtly, by designing sell in the state. To be successful, the clinical trial studies to get only the results they want. registry must include all the clinically signifi cant aspects of the trial design and trial results. Such a Our numbers dramatically registry would be placed in the state’s department understate the problem. of health, and could be fi nanced with registration Th e FDA letters we examined do not address fees from the drug marketers. anywhere near the full universe of prescription drug marketing. • To address the problem of deceptive classic advertising, deceptive sales representative • Th e FDA routinely reviews only “classic” statements and deceptive doctor-to-doctor advertising and does not comprehensively marketing, states can create a new type of monitor sales representatives, doctors acting as citizen lawsuit. Th is would allow citizens pitchmen, or clinical trial data manipulation. to sue for injunctive relief—stopping the Moreover, the FDA’s review of classic advertising false advertising and conducting corrective is not complete; not all ads are submitted to it, advertising—reasonable attorney’s fees, and, and of those that are, the FDA only reviews some. at the judge’s discretion depending on the circumstances of the case, civil penalties payable • Th e FDA letters rarely identify how many times, only to the state. Suits could only be won if the or where, an ad was used. A deceptive print deceptive advertising created a public health ad may have run in several newspapers and risk; deceptive advertising that misleadingly, but magazines. Each of those print runs would be not dangerously, hypes a drug’s properties would another dissemination of the deceptive messages not qualify. Doctors, their patients, attorneys in the ads. general, and in certain instances, the public, would have standing to sue, depending on the • Th e FDA reviews advertising after it has been type of marketing. disseminated and only requires corrective measures a quarter of the time. Examples of suffi ciently dangerous advertising might include promoting a drug for illnesses for • Th e best measure is how many people internalized which the company knows it’s not eff ective, or the deceptive measure, an impossible fi gure to denying or consistently minimizing serious risks. determine. Th e 869 disseminations of deceptive Th e advantage of this approach is it enables the messages that we were able to count from 2001- recipients of deceptive advertising—the people 2005 included TV ads, print ads, and other mass who can most easily detect it—a way to address media. How many people are deceived by a single the problem but it avoids creating fi nancial deceptive TV ad watched by a million viewers? incentives that would distort enforcement. Similarly, a single sales representative may convey deceptive messages to hundreds or thousands of Increasing Enforcement at FDA doctors in a year. To make the FDA a potent regulator able to prevent and correct deceptive advertising, it needs more power and fi nancial resources to: RECOMMENDATIONS States Can Solve the Problem • Review all advertising submitted to it before • To address the scientifi c misconduct that is the it is disseminated, in a commercially relevant suppression, manipulation and misrepresentation timeframe, so that deceptive classic advertising is of clinical trial data, states should establish a not used; comprehensive, searchable database of clinical 3 TURNING MEDICINE INTO SNAKEOIL

• Review sales representative training materials • Medical schools and teaching hospitals and make unannounced inspections of training should heavily invest in training students and sessions; residents to be skeptical of pharmaceutical sales representatives and to rely on independent sources • Review the presentation materials for talks given of information. by doctors on behalf of drug marketers and make unannounced visits to the talks; REPORT ROADMAP After introducing the problem and laying out the • Require and oversee corrective advertising in regulatory context, the report presents the results of every situation where deceptive marketing occurs; our analysis of the most comprehensive database on false and misleading advertising available: FDA’s • Require drug marketers to get the FDA’s approval enforcement letters to pharmaceutical companies before citing any study as support for any claim; engaging in deceptive marketing practices. We and fi nally, look at fi ve years of letters to see what kinds of false messages pharmaceutical companies are directing • Levy signifi cant fi nes against drug marketers, fi nes toward whom and how. We also explain why those that escalate to truly punitive levels, to serve as a numbers are grotesque understatements of the deterrent and eliminate today’s rampant recidivism. problem. One reason they are understatements is The Medical Profession’s Role: that they mostly address conventional advertising, Improve Prescriber Education such as ads in professional journals or on TV; they and Information Resources rarely address sales representative statements or the presentations made by doctors consulting for the Th e medical profession and the independent drug marketer. Th e latter activities are currently organizations and academic institutions that service beyond the FDA’s resources to monitor. it can help.

• Doctors need better access to independent, Th en we look at the ways the FDA currently fails to accurate, digested information about drugs. Th e address even the classic advertising slice of the false information produced by the clinical trial registry marketing problem, the one it monitors as closely as should be packaged by an independent group or it can. As part of our evidence of the FDA’s failure, agency into a form easily useable by prescribers we describe the high rates of “general recidivism,” who want information about treatment options. that is, drug marketers that have received multiple Th e information provided should include not only letters from the FDA about their false or misleading the clinically important information about each marketing, and “specifi c recidivism,” that is, drug drug, but also how the drug compares to other marketers who have received multiple letters about treatments in terms of safety, effi cacy, and cost. their advertisements for a single drug, advertisements Th e Drug Eff ectiveness Review Project that are all false or misleading in the same way. (DERP) generates this information, but it is aimed more at policy makers than prescribers. We complete our analysis of the deceptive marketing Similarly, Consumers Union takes DERP’s problem by focusing on the marketing outside of data and packages it for patients, as part of its the FDA’s routine review. Specifi cally, we focus on BestBuyDrugs project. To the extent that the prescription drug sales representatives and clinical information is already accessible (for example, Th e trials. Sales representatives are powerful marketing Medical Letter), the profession must fi nd a way to forces because they have many opportunities to ensure that doctors use it. Only by breaking their interact with physicians, and the evidence shows reliance on sales representatives and other sources that they give false and misleading information far of promotional information can doctors ensure too often. As disturbing as our fi ndings in this area they are getting unbiased information. are, they may be mitigated to some extent, given that M A Y 2 0 0 6 THE STATE PIRGS 4

doctors may expect sales representatives to present THE APPENDIX—CASE STUDIES misleading information. After all, their job is to To fi ll in the big picture of deceptive marketing sell drugs, not educate physicians. Clinical trials, we sketch, we present six case studies of deceptive however, are the cornerstone of prescription drug marketing of prescription drugs in the appendix, science, and few physicians let alone patients would located in the center spread. Four—Vioxx, anticipate the extent to which drug marketers shape OxyContin, Paxil, and Neurontin—are off ered and control them. primarily to illustrate diff erent features of the problem and to convey how deceptive messages can We conclude with concrete solutions that states permeate drug marketing. Two other case studies, can take now and off er recommendations for Accutane and Tindamax, are included to highlight addressing FDA’s problems. Fortunately, steps the the FDA’s inability to police drug marketers. states can take are powerful enough to rein in the drug marketers to the point where the public can again be confi dent that they and their doctors are consistently receiving accurate information. Best of all, the state steps are inexpensive. 5 TURNING MEDICINE INTO SNAKEOIL Introduction

Consumers and doctors are inundated with messages Deceptive Advertising Distorts about which drug to use. Ubiquitous advertising is the Crucial Risk/Benefi t Analysis not necessarily bad; doctors need to learn the new Th e Food and Drug Administration (FDA) approves treatments available, and consumers can be inspired drugs for very particular uses, based on the data to see their doctors for necessary treatment, as well showing safety and effi cacy and the balance between as empowered to be active participants in their them. After negotiation with the drug’s maker, the health care. But when drug marketers give false FDA codifi es its risk-benefi t judgment as the drug’s and misleading information about their products to indication in the drug’s label. As Dr. Henry I. Miller, doctors, patients are placed at risk. former head of the FDA’s Offi ce of Biotechnology recently explained in the Wall Street Journal: On purely medical criteria, the decision to prescribe a certain drug for a certain person is very complex; Th e ‘safety’ of a drug is a relative thing. each drug’s eff ects are diff erent, and each patient has Safety and effi cacy, the two criteria unique issues that must be considered, including their required for marketing approval of a drug, age, gender, disease state, other medications they are inextricably linked. Th e judgments are taking, and other medical conditions they have. of regulators (and practicing physicians) Moreover, non-medical factors also can infl uence require a global and often diffi cult prescribing decisions. Research reveals patient calculation of risk and benefi t, including demand, driven by direct-to-consumer advertising, consideration of what alternative therapies aff ects prescribing,1 as does contact between doctors are available. For a given drug, we are and sales representatives,2 advice from other doctors,3 willing to tolerate greater uncertainty and gifts to doctors,4 the availability of free samples,5 and more severe side eff ects for a potential cure the insurance coverage the patient has.6 for pancreatic cancer or AIDS, for example, than a new drug that treats heartburn. Nonetheless, our society has so far maintained When FDA grants marketing approval, the bedrock faith in one idea: doctors can sort through drug is deemed safe and eff ected to be used all the issues and pressures and consistently make for the conditions on the label.8 appropriate prescribing decisions. A 2003 survey showed broad trust in doctors’ prescribing decisions, Th us, for the uses the FDA has said a drug maker but also the corrosive eff ect of drug marketing.7 can market a drug—and only for those uses—the Roughly two-thirds of the respondents “trust[ed] public should have confi dence a drug is eff ective [their physician] to choose the drug that is best,” enough to be worth the risk. FDA’s critics, however, while 23 percent worried their doctors’ judgment argue such confi dence is misplaced.9 If the critics might be impaired by drug marketing and another are right, some drugs approved uses, however ten percent were undecided. limited, are already broader than justifi able.

Th e public’s trust in their doctors’ prescribing Th e delicate risk-benefi t analysis that underlies decisions might plummet if the public understood prescription drug approvals is what makes ensuring just how often drug marketers conceal risks from doctors have full and accurate information about a doctors, urge doctors to prescribe drugs for uses drug so important. Th e marketing regulations are that have not been shown to be safe or eff ective, intended to ensure that is what doctors get. and make misleading claims to doctors about the drugs they promote. Doctors are free to prescribe “off -label,” meaning M A Y 2 0 0 6 THE STATE PIRGS 6

for uses that the FDA has not approved. Th is How the FDA Polices Deceptive Advertising practice makes sense as long as the underlying Under current regulations, drug marketers submit premise is true: motivated by their patients’ needs their promotional material to the FDA before using and aware of their unique situations, doctors are it, but are then free to use the material until the FDA making thoughtful, informed choices to use drugs tells them otherwise. Eventually, the FDA reviews experimentally. Doctors are capable of making the some of the material submitted to it; if the FDA necessary judgments. Recognition of that capability deems an ad false or misleading, it writes the drug justifi es the right to prescribe off -label. marketer an “Untitled Letter” explaining its decision and telling the marketer to stop using the ads. If In contrast, in recognition of drug companies’ desire the FDA is very concerned about the false and to sell as many drugs as possible, the law does not misleading messages in the ad, it issues a “Warning allow drug companies to promote drugs for off - Letter” instead of an Untitled Letter, and demands label uses. If drug marketers can promote drugs that not only the marketer stop the false advertising, for off -label uses, the concept of “on-label” use is but also that it send out corrective messages. empty; the drug labeling is irrelevant. On occasion, the FDA attends medical conferences or Defi ning Deceptive Advertising lectures at which drugs are promoted and determines FDA regulations identify two diff erent kinds of that statements made by sales representatives or prescription drug advertising: product specifi c ads doctors consulting for the drug marketers are false and reminder ads.10 Product specifi c ads promote a or misleading. Th e FDA also learns of these types certain drug and are subject to the most extensive of violations from upset doctors or competing drug disclosure requirements. Risks must be disclosed with marketers. Th e FDA responds in the same way: it equal prominence as benefi ts, the drug’s approved writes Untitled and Warning Letters. uses must be clear, and every claim about the drug (such as “it’s cheaper,” “patients like it better,” or “it Th e FDA letters are clear and detailed, and most improves patients’ quality of life”) must be supported are posted on the FDA’s website, along with the by substantial evidence. Furthermore, the drug promotional material to which they refer. While marketer can promote only approved uses and must these letters are not the same as a conclusive fi nding disclose any limits on a drug’s approved use—such of falsity because the drug marketer has the right to as being appropriate only after other treatments have respond and appeal, they are nonetheless the best failed or only for the most severe forms of the illness. available database to indicate what the scope of the deceptive advertising problem is. Reminder ads build brand awareness by “reminding” people of the drug’s name. Th e opposite extreme of Deceptive Advertising’s Role in Maximizing Profi ts product specifi c ads, these ads cannot include any In the face of strong pressure from Wall Street information about the drug. Th ese ads help build to produce “blockbuster” drugs, drug marketers brand awareness and are generally aimed at doctors have a signifi cant incentive to engage in deceptive in the form of gifts emblazoned with a drug’s logo. advertising. FDA limits on how drug marketers can promote a drug’s uses and strict risk disclosure Disease awareness ads are a third type of ad. Th e requirements pose a tremendous sales challenge to ads help consumers understand what the symptoms drug marketers. Approved uses defi ne the size of the they experience might mean and inspire them to legitimate market for a drug, and full awareness of go to the doctor for diagnosis. Drug marketers risks can deter some doctors and patients within that generally run these ads about diseases that their market.11 In many cases, marketing solely within the drugs treat, so the increased diagnoses lead to FDA’s prescribed limits caps a drug’s potential sales increased sales. Disease awareness ads cannot have at a relatively low number. Persuading doctors that a any drug related information at all. drug is safer than it is, or that it is useful beyond its indication, is a shortcut to blockbuster status. Th e FDA deems any deviation from these requirements “false or misleading advertising.” 7 TURNING MEDICINE INTO SNAKEOIL Deceptive Advertising Findings

Table 1 Types of Deceptive Marketing Aimed at Doctors

Aimed at Doctors No. of Ads % of Ads Detail aid/Sales aid 43 20% DECEPTIVE DRUG ADVERTISEMENTS Journal ads 36 17% ARE WIDESPREAD Brochure 19 9% To assess the scale of the deceptive advertising Oral statements by sales representatives 19 9% problem, we studied all 17012 of the FDA Warning Convention panel/Exhibit panel/Booth panel 15 7% and Untitled Letters from 2001-2005 posted Audio conferences 11 5% at http://www.fda.gov/cder/warn/warn2005. Direct mailers 10 5% htm and tallied how many times drug marketers Case study 8 4% disseminated misleading or false messages to Visual aid 8 4% doctors and the public. We also reviewed the Posters 6 3% existing literature to contextualize the results. We Dear Doctor letter 3 1% found false advertising is common, varied, and targeted at doctors more than at the public. Many Abstract distributed at conference 2 1% of the false messages minimized or omitted risks or Flash card 2 1% promoted unproven uses for the drug. Medical information packs 2 1% Reprint carrier 2 1% From 2001-2005, 85 companies deceptively marketed 150 drugs. Video 2 1% Booklet 1 0.4% From 2001-2005, the FDA wrote at least 170 letters Bottle holders 1 0.4% to 85 diff erent companies for false and misleading advertising.13 Th e letters involved the promotion of Calendar 1 0.4% 150 diff erent drugs, drugs for conditions as mild as Dear Director of Nursing letter 1 0.4% allergies to as serious as cancer. Fact card 1 0.4% File card 1 0.4% False advertising aimed at doctors was more common and varied than false advertising Four sided card 1 0.4% aimed at consumers. Handout 1 0.4% Jar 1 0.4% In the period we studied, drug marketers disseminated a total of 358 false or misleading ads, Material distributed at a conference 1 0.4% 210 (59%) to doctors and 148 (41%) to consumers. Model 1 0.4% As Tables 1 and 2 show, the types of deceptive Notepad 1 0.4% marketing are varied. In total, 38 diff erent types Packers 1 0.4% of false and misleading marketing pieces targeted Physician starter packs 1 0.4% doctors, and no one kind dominated; in fact, seven Pill wall chart 1 0.4% diff erent kinds of advertising occurred with double Professional formulary switch aid 1 0.4% digit frequency, and combined, these accounted for only 73% of the deceptive ads aimed at doctors. Professional promotional labeling 1 0.4% Promotional banner 1 0.4% As Table 1 shows, doctors are bombarded with false Puzzle 1 0.4% messages in all areas of their professional life, whether Sell sheet 1 0.4% when reading journals, attending conferences, meeting Wall chart 1 0.4% with sales representatives, reviewing materials given Wholesaler fact sheet 1 0.4% to them by sales representatives, listening to talks by other doctors or reading their mail. Focusing M A Y 2 0 0 6 THE STATE PIRGS 8

misleading messages on doctors makes sense from a Table 2 marketing perspective because doctors are the ones Types of Deceptive Marketing Aimed at the Public who produce sales by writing prescriptions. Types of Ads No. of Ads % of Ads Statements made by sales representatives and Print ads 48 32% audio conferences given by drug marketer- Tv ads 38 26% sponsored doctors to other doctors were the Websites 31 21% 4th and 6th most common sources of false and Radio ads 9 6% misleading messages identifi ed by the FDA letters, Videos 5 3% respectively, accounting for 14% of the total combined. Both of these types of marketing occur Billboards 3 2% largely outside the FDA’s monitoring; that the Press release 3 2% FDA detected them frequently enough for them Fulfi llment letter 2 1% th th to rank as the 4 and 6 most common forms of Answers at toll free # 1 1% false advertising targeting doctors suggests that Infomercial 1 1% the false advertising problem in those two settings Brochure 1 1% could be enormous. Puzzle 1 1% In contrast, only three types of ads account for 79% Direct mail 1 1% of the deceptive marketing aimed at consumers— Starter kit 1 1% print ads (36%), TV ads (26%) and websites Tear sheet 1 1% (21%)—and drug marketers are required to submit Restroom poster 1 1% these ads to FDA for review. Nonetheless, these Special magazine 1 1% advertisements, like all direct-to-consumer (DTC) ads, have the potential to mislead millions of people, far more than the marketing aimed solely at doctors. Moreover, doctors are members of the public, so these deceptive ads can reinforce the misleading messages doctors are already receiving. As Table 3 shows, the ads discussed by the FDA letters contained a total of 869 false or misleading FALSE ADVERTISEMENTS OFTEN messages. CONTAINED MULTIPLE FALSE Doctors received more false messages MESSAGES than the public. Th e 358 ads addressed by the FDA letters often Of the 869 deceptive messages, 535 (62%) were contained multiple false or misleading messages. A aimed at doctors, and 334 (38%) were aimed at single ad might omit some critical risks, minimize the public. To some extent, Table 3 shows the others and promote for unproven uses. Nearly explosion in direct-to-consumer advertising; in three-fourths of the letters involved fi ve or fewer 2001, drug marketers targeted doctors with more deceptive messages (72 %), while 20% had between than three times as many misleading messages 6 and 10 deceptive messages, and 8% had more as they did the public; in 2002, drug marketers than 10. Th ree of the FDA letters discussed the targeted doctors with more than twice as many dissemination of more than 40 deceptive messages.14 misleading messages; in 2003 and 2004, marketers Th e letters that involved high numbers of deceptive targeted the public for more misleading messages, messages either involved the most diff erent pieces but the gap narrowed. In 2005, the trend of marketing (e.g., multiple sales aids, journal ads, reversed and doctors again were targeted for more etc.) or had the most complete information on misleading messages than the public, although the dissemination of the false ads, or both. gap was again narrow. 9 TURNING MEDICINE INTO SNAKEOIL

Table 3 Number of Ways FDA Identifi ed Ads as False or Misleading, 2001-2005

Misrepresented Promoted for Uses Unsubstantiated or risk and safety not proven safe and Other Total misleading claims Year information eff ective For For For For For For For For For For All All All All All Dr’s Public Dr’s Public Dr’s Public Dr’s Public Dr’s Public 2001 93 28 121 80 23 103 79 27 106 11 8 19 263 86 349 2002 39 17 56 22 5 27 43 18 61 0 7 7 104 47 151 2003 27 22 49 14 21 35 24 37 61 0 2 2 65 82 147 2004 18 23 41 5 11 16 16 25 41 0 6 6 39 65 104 2005 20 15 35 9 5 14 31 30 61 4 4 8 64 54 118 2001-2005 197 105 302 130 65 195 193 137 330 15 27 42 535 334 869

Note that the 2001 numbers are signifi cantly higher than the other years in most categories; that is because FDA issued nearly twice as many letters in 2001. Why is not clear. It may refl ect a shift in enforcement emphasis; the FDA issued several letters relating to promotion at medical conferences in 2001, but very few in the other years. The change may refl ect the change of administration in Washington D.C. and a resulting shift in resources and priorities. The sharp decrease between 2001 and 2002 may also refl ect a decrease in false advertising by drug marketers, however, the year to year changes among 2002-2005 are nowhere near as great, so it is hard to imagine that a change in marketing practices represents the whole shift between those or any other years.

METHODOLOGY FOR FALSE MESSAGES: WHAT WE COUNTED

To the extent possible, we counted how many false and misleading such ads, ten aimed at the public and fi ve aimed at doctors. Placing these messages were being disseminated to doctors and/or the public. Thus, ads in the “other” category reduced the other totals. a print ad that the FDA said omitted some serious risk information and minimized other risk information counted as two violations; if the letter The category “Misrepresented Risk and Safety Information” includes these indicated the ad ran in both the New York Times and the Washington Post, FDA-identifi ed violations: omitting risks; minimizing risks; inadequately that was counted as disseminating four false or misleading messages. In communicating risk; misrepresenting safety profi le; misleading safety claim; one case, a letter criticized fi ve print ads aimed at the public and mentioned omitting safety information; using outdated labeling (new version included 11 different publications the ads ran in, and indicated that the ads ran in more risk information); and lack of “fair balance.” Fair balance is the FDA’s other outlets too. The letter did not clarify which ran where, however, so requirement that risk information be displayed as prominently as a drug’s we conservatively counted 11 disseminations of each violation in the ads, benefi ts and is a type of minimizing or omitting risk information. rather than assuming all fi ve ads ran in all 11 publications, which would have produced 55 disseminations of each violation. However, most of the The category “Promoted for uses not proven safe or effective” includes time the FDA letter did not indicate where an ad was used, so we primarily the FDA-identifi ed violations of: promotion of unapproved use; promotion counted how many times drug marketers created false or misleading of an investigational drug (i.e. one not yet approved for sale in the U.S.); messages for dissemination. broadening a drug’s indication; inadequate communication of a drug’s indication; omission of material facts relating to a drug’s indication We counted only what the FDA characterized as the violation of the drug or use; promotion of an unapproved dosing regimen; and inadequate marketing regulations, rather than each example of the violation the FDA communication of contextual information, in one case regarding the included in the letter. Thus when the FDA stated that an ad made misleading appropriate population for a drug, and in the other, regarding health risks. claims about the drug’s impact on a patient’s quality of life and gave four examples from the ad, we counted it as one violation. But if the FDA The category “Unsubstantiated or misleading claims” includes the FDA- indicated the ad made several types of misleading claims, each type of identifi ed violations of: unsubstantiated claims; unsubstantiated superiority misleading claim was counted separately. claims; unsubstantiated effi cacy claims; misleading superiority claims; misleading comparative claims; misleading claims; misleading effi cacy Some letters addressed ads that the drug marketer had designed as either claims; overstatements of effi cacy; and failure to disclose a material fact a disease state ad or as a “reminder ad”, but which made product-specifi c (unrelated to safety or type of use). claims and so were really product-specifi c ads in disguise. These ads violated the rules in a number of ways, always including the omission of risk The category “Other” includes the 15 ads that were identifi ed as product- information and what the drug’s approved uses are. However, the FDA was specifi c ads in disguise and violations such as failing to make adequate not consistent in describing these violations, so rather than create our own provision for consumers to get full information about the drug, failing method of counting violations or simply add them to our totals relating to to disclose the drug’s prescription status or minimizing the role of the risk or promotion of unapproved uses, we placed them as single violations prescriber, failing to include a brief summary of a drug’s prescribing in the “other” category. In the fi ve years we reviewed, the FDA identifi ed 15 information, and failing to disclose the drug’s generic name. M A Y 2 0 0 6 THE STATE PIRGS 10

To some extent, that change in distribution may treatment with these agents and the also refl ect FDA reducing its investigation of appropriate conditions for their use sales representatives’ statements at conferences— raises signifi cant public health and safety certainly there were far fewer letters for those concerns.” (11-14-02 Untitled Letter to in the later years—and a change in how FDA Xanodyne) enforcement resources are allocated. According to the Government Accountability Offi ce, the FDA Our review of the deceptive messages identifi ed cannot review all advertising submitted to it and by the FDA found that many of them, particularly prioritizes reviewing broadcast advertisements.15 those targeting doctors, either misrepresented drugs’ Th us as direct-to-consumer advertising grew, the risks or promoted drugs for unproven uses. Both of FDA had less resources to review other forms of those types of deceptive messages are particularly advertising, and the relative decrease in detecting dangerous for patients because they can directly false ads aimed at doctors may only refl ect a result in inappropriate prescriptions. reduction in detection, not in occurrence. 37% of the deceptive messages targeting doctors endangered patients by omitting, DECEPTIVE ADVERTISING IS minimizing, or misrepresenting drugs’ risks. DANGEROUS While any false and misleading advertisement is bad, those teaching doctors that a drug is safer Deceptive advertising puts patients at risk. Th e than it really is are particularly troubling. Doctors FDA’s May 29, 2003 Warning Letter to Hoff man- relying on that information will unintentionally LaRoche explains the stakes: endanger their patients.

You have disseminated a professional Screening patients appropriately and closely sales aid and a patient video that omit following warnings, precautions and contra- or minimize material facts regarding the indications ensures that each prescribing decision is safety profi le of Xeloda, make misleading as accurate a risk-benefi t analysis as possible, which effi cacy claims for Xeloda, make is what the public wants. When drug marketers unsubstantiated superiority claims over misrepresent risks, this calculation is off and other cancer therapies and omit material patients are unjustifi ably endangered. facts about the approved indications for Xeloda. Th ese claims have the potential to In the period we studied, 37% of the false misguide physicians in making prescribing or misleading messages targeting doctors and treatment decisions, and, therefore, involved omitting, minimizing, or otherwise jeopardize patient safety. (Emphasis misrepresenting drugs’ risks. (Table 3, fi rst set added; at p. 8) of columns, 197/535.) Th e public fared only slightly better: 31% (105/334) of the false or Sometimes the FDA was even more explicit: misleading messages aimed at the public omitted or minimized risks. “Aventis’ false or misleading promotion in the sales aid may compromise patient Again, if we had treated the 15 ads that pretended survival and safety.” (Emphasis added; not to be product-specifi c ads diff erently, the 12-15-02 Untitled Letter to Aventis at p.2) proportion of risk violations would be even higher for both groups, because none of those ads Or contained any risk information for the drugs they promoted. “Your failure to disclose the serious, sometimes fatal, risks associated with 11 TURNING MEDICINE INTO SNAKEOIL

Examples of Marketing That Dangerously To get a fuller understanding of how drug Misrepresents Risks: marketers can misrepresent the risks of their • Pharmacia promoted Celebrex for use in drugs, please see the case studies in Appendix A, patients at risk of serious bleeding. particularly Vioxx, OxyContin, and Accutane. Pharmacia promoted Celebrex as safe for patients taking the drug Coumadin, although the 24% of the Deceptive Messages Targeting prescribing materials state that patients taking Doctors Endangered Patients by Promoting both may experience serious bleeding. Th e Unproven and/or Unjustifi ed (“Off -Label”) Uses. FDA explained: “minimization of this risk raises As the data in the second set of columns in Table signifi cant public health and safety concerns.” 3 shows, 24% (130/535) of the deceptive messages (FDA Warning Letter to Pharmacia 2-1-01) targeting doctors promoted drugs for uses for which FDA had not approved them as safe and • A sales representative denied Sporonox’s serious eff ective. Th e public fared somewhat better; 19% cardiac and liver risks. (65/334) of the deceptive messages targeting them A sales representative for the Janssen Research promoted drugs this way. Again, if we had treated Foundation told visitors to Janssen’s booth in the 15 ads that pretended not to be product- the commercial exhibit hall of the American specifi c ads diff erently, the proportion of unproven Society of Health-System Pharmacists’ Annual use violations would be even higher for both Meeting that Sporonox does not aff ect the liver, groups, because none of those ads contained any that liver monitoring is not necessary when information about the approved indication of the using it, and further, that he did not know of drugs they promoted. any cardiac risks associated with it. Despite the sales representative’s assurances, Sporonox’s Promoting off -label can be more dangerous than prescribing information has a Black Box Warning misrepresenting risks. Our analysis incorporates about cardiac risks and also warns of rare but fatal two basic types of off -label promotion; promoting liver risks, even in people who had no previous unproven uses and promoting unjustifi ably risky uses. liver problems. Black Box Warnings are risks considered so serious and important that the By defi nition, when drug marketers are promoting FDA visually isolates them and puts them in bold uses that are not in a drug’s label because the drug text in a black box on the label. (FDA Untitled marketer has not proven safety or effi cacy, the Letter 6-27-01) drug marketer does not know the drug will work nor if such use will harm the patient. Th us the • Sales representatives denied warnings that patient gets all the drug’s risks and quite possibly Avandia could cause congestive heart failure. no benefi t. Worse, if eff ective treatments exist, Two sales representatives for GlaxoSmithKline the patient’s condition may continue to worsen (GSK) told visitors to the GSK exhibit booth at needlessly. An example of this type of off -label the 10th Annual American Association of Clinical promotion is Neurontin, discussed in Appendix A. Endocrinologists meeting that no new warnings had been added to the prescribing information for When drug marketers promote off -label by Avandia and that Avandia could be used safely broadening the drug’s indication—meaning they with insulin. In fact, the Avandia information urge doctors to ignore the safety-based limitations had been revised to warn that Avandia could the FDA imposed on a drug’s use—they are cause congestive heart failure when used with promoting the drug for uses it is eff ective for, but insulin, even in patients without a history of which the FDA decided are not justifi ed, given the congestive heart failure or pre-existing cardiac drug’s risks. Patients given these prescriptions are failure. Th e FDA had previously asked GSK to by defi nition exposed to unnecessary, excessive risks. stop misrepresenting the congestive heart failure An example of this type of off -label promotion is risk. (FDA Warning Letter 7-17-01) OxyContin, discussed in Appendix A. M A Y 2 0 0 6 THE STATE PIRGS 12

Despite the risks, indirect evidence suggests that incentive—the need to sell as many pills as possible— off -label promotion by drug companies is common to urge prescriptions for as many ‘uses’ as possible. and successful. In 2002, 115 million off -label prescriptions were written.16 In 2003 Knight- Th e widespread off -label prescribing identifi ed Ridder did an investigativeinvestigative rreporteport oonn tthehe dandangersgers of by Knight-Ridder and quoted above is hard to off -label prescribing and explained the scope of the understand absent a concerted off -label promotion problem: eff ort. Interestingly, one drug class that Knight- Ridder found to be rarely given off -label are Doctors are giving their patients epilepsy cholesterol medicines, a type of medicine that has an drugs for depression and hot fl ashes enormous on-label market, and one of the drug types and to help them lose weight. Th ey most commonly prescribed off -label, anti-seizure use antidepressants to treat premature medicine, has a limited on-label market. Could drug ejaculation and pain, and powerful marketers’ ‘need’ for a blockbuster-sized market for antipsychotics for insomnia and attention their drugs—and therefore off -label promotion— defi cit disorder. High blood-pressure pills explain that diff erence in off -label usage? are prescribed for headaches and anxiety; antibiotics are used to treat viruses. Examples of Promoting Off -Label Use • Cubist Pharmaceuticals promoted Cubicin for Some drugs, in fact, are sold mostly off -label use as a pneumonia treatment despite for unapproved purposes. Eight out of data showing it is ineff ective. 10 prescriptions for the epilepsy drug Cubist Pharmaceuticals used a website to Topamax aren’t for epilepsy. promote its antibiotic Cubicin for use against the bacteria that causes community-acquired Th alidomide, the notorious morning- pneumonia. In fact, the prescribing information sickness drug that caused horrible birth specifi cally says Cubicin should not be used for defects and ushered in today’s FDA drug- pneumonia because clinical trial results showed safety rules, even is on the market, and 99 that Cubicin was not eff ective, leading to patient percent of its prescriptions are off -label. deaths. Th e letter noted the promotion for pneumonia “is misleading and poses a signifi cant [Of the]15 top-selling classes of drugs… public health risk because such practice could lead some, such as cholesterol medicines, rarely to therapeutic failure and death.” (FDA Warning are given as unapproved treatments. But Letter 8-17-04) three-quarters of anti-seizure medications are prescribed off -label, as are nearly two- • A sales representative promoted Tracleer for off - thirds of antipsychotics and about one- label congestive heart failure treatment, despite quarter of antidepressants[.]17 the data showing no benefi t and increased risks to patients. While off -label uses, by defi nition, lack suffi cient A territory manager for Actilion promoted evidence to demonstrate safety and effi cacy, not all Tracleer as useful for congestive heart failure to off -label uses are equally risky. Some have some members of a congestive heart failure unit. Th e support while others have none, and in some manager claimed that studies had shown that instances, patients are benefi ting from off -label uses. Tracleer had shown neither a positive or negative Acknowledging that off -label use can benefi t some eff ect on congestive heart failure and suggested and sometimes has some justifi cation, however, does that the doctors might have a use for it. However, not change the imperative against off -label promotion. data that Actilion itself helped create showed no Off -label use, at doctors’ discretion, is allowed precisely benefi t and in fact some worsening of congestive because of the potential benefi ts; off -label promotion heart failure patients. Furthermore, even if Tracleer is forbidden because drug marketers have a strong posed no risk of worsening the patients’ congestive 13 TURNING MEDICINE INTO SNAKEOIL

heart failure—even though ineff ective treatment risks has developed.18 Th e FDA cannot dictate intrinsically poses such a risk—Tracleer is an the label; it and the drug manufacturer must agree intrinsically dangerous drug. It poses serious risks on the language, a negotiation process that can to patients’ livers and to fetuses, risks high enough enable marketing concerns to shape a drug label’s that it is not supposed to be used except through a content.19 Perhaps that infl uence explains why, specifi c risk minimization program. Th e territory when signifi cant evidence of Vioxx’s heart risks manager mentioned neither the risks nor the risk persuaded the FDA that Vioxx’s label should program. (FDA Untitled Letter 10-30-02) contain some information about heart risks, that information was not placed in the “Warnings” • Berlex promoted Climara for off -label uses that section.20 Th e FDA recognizes the diffi culty it has are likely to endanger patients. in making drug labeling refl ect drugs’ risks and has Berlex Laboratories promoted its estrogen patch asked for the authority to dictate the content of Climara for the treatment of patients with prescribing information.21 gallstones and hypertrigliceridemia [elevated triglicerides] in a journal ad and a panel for an Despite the Regulations, Not All Ads Are exhibit booth at conventions. Climara is not only Submitted to the FDA not approved for those uses, but may endanger Drug marketers do not always submit patients if used that way; Climara’s prescribing advertisements to the FDA for review, despite information warns that Climara use can cause the regulations. Th e FDA is well aware of the gallbladder disease and elevate triglicerides problem.22 Indeed, 32 of the false and misleading leading to pancreitis and other complications. ads addressed in the FDA letters we reviewed (FDA Untitled Letter 1-6-03) had not been submitted to the FDA; the FDA found them by other means. Th us the universe of To get a fuller understanding of how drug potentially false conventional advertising is larger marketers promote their drugs off -label, please than the one the FDA is given the opportunity to see the case studies in Appendix A, particularly review. Neurontin, Tindamax, and Vioxx. The FDA Does Not Review All Ads Submitted To It Not all ads submitted to the FDA for review THE FDA LETTER DATA GROSSLY may in fact be reviewed. As the Government UNDERSTATE THE FALSE Accountability Offi ce (GAO) explained: ADVERTISING PROBLEM FDA said that it receives numerous While the data from the FDA letters demonstrate marketing and promotional materials for that the deceptive advertising problem is pervasive promoted prescription drugs and that and severe, we must emphasize that the data and while every eff ort is made to review the case studies dramatically understate the problem. materials, it cannot guarantee that all materials are reviewed because of limited Drug Marketers Help Shape Drugs’ Labels—the resources and competing priorities.23 Baseline for Judging Ads Th e most subtle reason the numbers are an Th us, the universe of ads the FDA reviews for understatement is the fact that the baseline against accuracy is smaller than the one it is theoretically which ads are judged is not as strict as the public given the opportunity to review, which as noted and doctors might imagine. is already smaller than the full universe of conventional prescription drug advertisements. Drug marketers have profound infl uence on the Th us the data discussed above refl ect the false and content of a drug’s label, particularly after the misleading ads detected in a subset of a subset of drug is on the market and new evidence of serious the true universe of ads. M A Y 2 0 0 6 THE STATE PIRGS 14

The Data Mostly Refl ect Ads, drug marketers attempt a recall. a false journal Not Placement of Ads advertisement could not, nor could a television ad By the time the letter is issued, a print ad, for that has already completed its run. Th us a false ad example, may have run in several professional may continue disseminating false messages long after journals, newspapers or magazines. Very few of the company agrees to stop using it. the FDA letters identifi ed the outlets in which the advertisements appeared; when FDA identifi ed The Best Measure Is How Many People the outlets, it often was by example rather than Internalized a False Message comprehensive. Th us our fi ndings understate the Finally, even if we could count how many times number of times the drug marketers disseminated drug marketers communicated false or misleading the deceptive messages identifi ed by the letters we advertising, that total would still grossly understate reviewed. the problem of false and misleading drug advertising. In the end, the critical number is not how many Deceptive Ads Are Not (and often false ads were disseminated, but how many people cannot be) Recalled received and internalized them. A single false TV While the FDA may declare a “sales aid” ad seen by millions of people may have prompted advertisement false and forbid its further use, it may thousands of them to make a decision harmful to already be in thousands of doctors’ offi ces, distributed their health. Th ere is no way to estimate how many during meetings between sales representatives people are being aff ected by the false messages, other and the doctors. While the sales aid in theory than to say it is orders of magnitude greater than the could be recalled, no FDA letter suggested that number of messages being put out. 15 TURNING MEDICINE INTO SNAKEOIL The FDA’s Failures

Th e data in the tables, illustrated by the case studies LESS THAN A QUARTER OF THE in Appendix A, reveal a false and misleading LETTERS CALL FOR CORRECTIVE prescription drug marketing problem that is ADVERTISING. endangering patients. Th e size and depth of the problem prove that current methods of addressing Only 23% of the FDA letters call for corrective false advertising are failing. advertising, allowing whatever misperceptions the false ads created to remain unchallenged in three Setting aside the fact that the FDA does not review quarters of the cases. Moreover, corrective measures much of drug marketers’ activity for accuracy and such as “Dear Doctor” letters and labeling changes fairness, the FDA letters fail to eff ectively police may not correct doctors’ understanding. what marketing the FDA does review in at least three ways. Serious risks with Propulsid, Rezulin, Posicor and Duract came to light after they were on the market and doctors had formed habits for prescribing them. FDA LETTERS GO OUT TOO LATE TO Th e FDA addressed the risks by changing the drugs’ BE EFFECTIVE. labeling and requiring the drug marketers to send Dear Doctor letters announcing and explaining To prevent the dissemination of deceptive messages, the labeling changes. Th e letters and labeling the FDA would have to review the materials in changes failed to change doctors’ prescribing habits, advance and release only accurate advertisements for and patients died. All four drugs are now off the distribution. Sending letters after deceptive material market.25 Since Dear Doctor letters failed in these is distributed might prevent further distribution of cases—when drug marketers had every incentive to the deceptive ads, but it does nothing to prevent ensure they succeeded—the letters are unlikely to doctors or the public from internalizing the already- correct a false or misleading advertising message. distributed deceptive messages, nor does it ensure that doctors or the public do not continue to view the already-distributed deceptive materials. For example, a doctor might hold onto an issue of a THE FDA LETTERS DO NOT DETER professional journal for years and see a deceptive ad FUTURE DECEPTIVE ADVERTISING any time he returns to it to review an article. Only by preventing the dissemination of deceptive messages We looked at two types of recidivism, general could the FDA eliminate the risk that prescriptions and specifi c: how many companies advertised are written based on inaccurate information about falsely more than once, and how many companies risk or eff ectiveness. Unfortunately, at present the advertised falsely about the same drug in the same FDA does not have the necessary resources to review way? In both cases, recidivism rates were high. materials before they go out. General Recidivism While reviewing the materials before dissemination Over One-Th ird of the Companies Received More Th an would be best, sending letters at the beginning of an One Letter. advertising campaign would be more constructive Approximately one-third of the companies than present practice. At present letters can take accounted for two-thirds of the letters sent in weeks to months to be sent out, allowing the false 2001-2005. In all, 28 companies accounted for messages to persist, even to fully run their course.24 113 of the letters. However, even within the group M A Y 2 0 0 6 THE STATE PIRGS 16

Table 4: General Recidivism Companies Receiving More Than One Letter in the Study Period # Letters Companies received Total by each letters Pfi zer, Inc. 15 15 Novartis Pharmaceuticals Corp. 9 9 AstraZeneca, Aventis, GlaxoSmithKline 7 21 Hoff man-LaRoche 6 6 Berlex Laboratories, Boehringer Ingelheim, Bristol-Myers Squibb Company. 5 15 Abbot Laboratories, Alcon Research, Ltd. (including one received by Alcon Laboratories for the 4 8 same drug at issue in one of Alcon Research, Ltd.’s letters) Allergan, Inc., Alza Corporation, Bayer Corporation, Fujisawa Healthcare, Inc., SuperGen, Inc. 3 15 Actelion Pharmaceuticals US, Inc., Connetics Corporation, Elan Pharmaceuticals, Eli Lilly, 2 24 Fujisawa Healthcare, Inc., Galderma Laboratories, Gilead Sciences, Inc., Janssen Pharmaceutica, Janssen Research Foundation, Pharmacia Corporation, Sanofi -Synthelabo, Inc., Schering Corporation. Total number of letters received by general recidivists: 113

of recidivists, some companies were much more this risk are particularly troublesome frequent disseminators of deceptive advertising because we have previously objected in than others. Pfi zer, Inc. received 15 letters; the next two untitled letters to your promotional closest company was Novartis with nine, and three materials for Celebrex that among other companies tied for third with seven letters each. violations minimized the Celebrex/ Table 4 shows the breakdown: Coumadin drug interaction. Based on your assurances that corrective steps had been We also found several examples of specifi c taken in order to prevent future violative recidivism; that is, marketing the same drug in practices of this type, we considered these the same deceptive manner after the FDA has matters closed. Despite your assurances, demanded it stop. however, your violative promotion of Celebrex has continued. (Emphasis(Emphasis aadded;dded; Specifi c Recidivism Pharmacia Warning Letter, 2/1/01 at p. 1) One example of specifi c recidivism and the FDA’s frustration with it is the conduct discussed in a Of the 170 letters we reviewed, 27 (16%) discussed February 1, 2001 letter to Pharmacia: past letters that the FDA had sent regarding similarly false advertising. You have engaged in repeated promotional activities that minimize the potentially An enforcement system that responds late, requires serious risk of using Celebrex and corrective advertising a quarter of the time, and Coumadin (warfarin) concomitantly. Your experiences signifi cant recidivism is a failure. minimization of this risk raises signifi cant public health and safety concerns because it To get a fuller sense of the FDA’s failures, please minimizes the risk of signifi cant bleeding. review the case studies in Appendix A, particularly Your promotional activities that minimize Accutane and Tindamax. A1 TURNING MEDICINE INTO SNAKEOIL

Case Studies of Deceptive Marketing INTRODUCTION studies describe drug marketers misrepresenting risks (Vioxx, Case studies are the easiest, most intuitive way to convey how OxyContin, Accutane), suppressing or misrepresenting drug marketers disseminate false and misleading messages clinical data (Vioxx, Paxil, Accutane, Neurontin), promoting about risks and unproven uses to doctors and how ineff ective unproven uses (Vioxx, Neurontin), promoting uses where the FDA is at addressing the problem. the risks outweigh the benefi ts (OxyContin, Paxil), and disregarding the FDA’s weak enforcement eff orts (Accutane, We present six case studies: Vioxx, OxyContin, Paxil, Tindamax). Neurontin, Accutane and Tindamax. Each one highlights a diff erent aspect of the deceptive marketing problem, although In all of these cases, the deceptive advertising had a single the stories have some overlap as well, which emphasizes goal: persuade doctors to write more prescriptions than sound that the issues are not unique to any one drug. Th e case medicine would dictate, putting patients at risk.

CASE STUDY 1: VIOXX particularly troublesome because we have previously, TARGETING DOCTORS WITH DECEPTIVE in an untitled letter, objected to promotional materials MESSAGES ABOUT POTENTIALLY FATAL that also minimized Vioxx’s safety profi le.” (FDA RISKS letter, 9/17/01, at p. 2)

Vioxx is an anti-infl ammatory pain reliever approved in 1999 Th e promotional activities the Warning Letter addressed were for osteoarthritis, acute pain, and menstrual pain. Eventually six audio conferences, events in which a doctor gave a Merck- its approved uses expanded to include rheumatoid arthritis and initiated and supported presentation to colleagues; a press acute migraine pain. It quickly became one of Merck, Inc.’s release; and oral statements by Merck sales representatives at most important drugs, with $2.5 billion in sales in 2003 alone.72 two professional conferences. Merck withdrew Vioxx from the market in September 2004 after clinical data unequivocally demonstrated serious heart Th e Vioxx letter and surrounding story involves several key risks with its use. During its fi ve years on the market, millions features of the deceptive marketing problem, but it makes a of people took Vioxx, resulting in an estimated 88,000-139,000 good showcase for one in particular: doctors are being heavily heart attacks of which some 26,000-55,000 were likely fatal.73 targeted with potentially deadly false messages about risks. Another estimate claims Vioxx caused some 160,000 heart attacks and strokes.74 Deceptive Doctor-to-Doctor Marketing As the Warning Letter indicates, Merck was misrepresenting Merck is now battling mass litigation alleging Vioxx the risks of Vioxx to both the public (the press release) and caused heart attack deaths and injuries. Regardless of the doctors. Th e press release claimed that Vioxx had “a favorable outcome of those cases, the record shows that Merck hid and cardiovascular safety profi le,” a claim the FDA found “simply misrepresented the risks of Vioxx and the results of a landmark incomprehensible.”75 clinical trial, as well as promoted Vioxx for unproven uses. Nonetheless, the misleading marketing aimed at doctors was Targeting Doctors worse. With regard to doctors, Merck was spreading the false In September, 2001, the FDA wrote to Merck regarding its messages on risk in two ways, via sales representatives and via promotion of Vioxx. Th e Warning Letter described several peer-to-peer (doctor-to-doctor) presentations. diff erent types of promotional activities that minimized the serious risks associated with Vioxx use—the same heart attack Doctor-to-doctor marketing is a common strategy. Large risks that Merck is now being sued for failing to warn about— drug companies, including Merck, commonly retain hundreds and stated: of doctors as speakers and consultants to give promotional presentations on their drugs, each of whom may speak at several “Your minimizing these potential risks and such events each year, and the events are becoming ever more misrepresenting the safety profi le of Vioxx raise common.76 In 2004, doctors spoke at 237,000 meetings and signifi cant public health and safety concerns. Your talks sponsored by pharmaceutical companies, up from 66,000 misrepresentation of the safety profi le of Vioxx is in 1999.77 M A Y 2 0 0 6 THE STATE PIRGS A2

Doctor-to-doctor marketing is particularly eff ective, because presentation. As moderators of the presentations, they had the doctors see each other as credible sources of information, opportunity to correct the information; their silence amounts to more so than pharmaceutical sales representatives. Merck a tacit endorsement of the doctor’s statements, as does the fact understands the unique impact of doctor-to-doctor marketing that six equally misleading presentations occurred. Indeed, the on sales, according to company documents reported by the Wall FDA viewed Merck as responsible for the doctor’s statements, Street Journal: which is why it discussed them in the Warning Letter to Merck. In a Merck slide presentation dated December 2001, two Merck employees observed that doctors who Deceptive Sales Representative-to-Doctor Marketing attended lectures or more intimate roundtable-type Merck has a large staff of sales representatives who meet with discussions were much more likely to increase their doctors at their offi ces, over meals, at continuing education prescribing of certain medications than those who events and conferences on a regular basis. According to the spent time with a Merck sales representative. Warning Letter, Merck sales representatives misrepresented the risks of Vioxx to people attending the 119th Annual According to the document, doctors who attended a Meeting of the Maryland Pharmacists Association and the lecture by another doctor wrote an additional $623.55 Annual Meeting of the American Society of Health-Systems worth of prescriptions for the painkiller Vioxx over a Pharmacists. Th e sales representatives, like the doctor in the 12-month period compared with doctors who didn’t audio conferences, claimed that the clinical data showing attend. Doctors who participated in the more intimate high risks of heart attacks with Vioxx use meant only that discussions wrote an additional $717.53 worth of naproxyn protected hearts, not that Vioxx hurt them.83 Th e fact prescriptions for Vioxx...Th at compared to an increase that diff erent sales representatives made similarly misleading of only $165.87 in Vioxx prescriptions by doctors who statements at diff erent conferences in diff erent regions of the attended a meeting with a salesperson.78 country (the fi rst was in Maryland, the second in California) suggests Merck trained all its sales representatives to off er this Th e slide presentation described by the Wall Street Journal, explanation of the heart attack data. dated only three months after the FDA sent Merck the Warning Letter, suggests that such tactics were an integral part Sales representative training is the drug marketers’ opportunity of Vioxx’s marketing. to control and shape the messages that doctors receive from the sales force. Documents released because of the Vioxx litigation Th e six doctor-to-doctor audio conferences addressed in the confi rm Merck specifi cally trained sales representatives to Warning Letter were misleading in fi ve ways. mislead doctors about Vioxx’s risks by teaching them to play “dodge ball.” Th e sixteen page training handout had twelve First, the doctor giving the presentations misrepresented tough questions about Vioxx’s cardiovascular risks that doctors Vioxx’s heart attack risks. He gave inaccurate information might ask, one per page, followed by four pages with only one about the heart attack rate in the Vioxx patients in the leading word: “DODGE!”84 clinical trial. He then explained away that understated data by claiming as fact a wholly untested hypothesis: Vioxx did Clinical Trials and Data Manipulation as Promotional Tools not hurt hearts; the comparator drug in the trial, naproxyn, Merck’s desire to sell more Vioxx and to capture a larger share protected hearts.79 Second, the doctor suggested Vioxx was safe of the painkiller market led it to manipulate the structure of to use with patients taking another drug, Coumadin, when in its trials and ignore results from its own clinical trials that fact in combination the drugs pose a serious risk of dangerous suggested Vioxx presented a cardiac risk. bleeding.80 Th ird, the doctor omitted a large range of other risk information, including certain types of patients who should Choosing Not to Investigate Heart Attack Risks not be given Vioxx and what the most common adverse events Pre-approval, Merck had data and theory to suggest that Vioxx with Vioxx use are.81 Fourth, the doctor made a number of harmed hearts and protected stomachs, but they did not have unsubstantiated claims about Vioxx’s purported superiority to enough information about either to defi nitively say what the naproxyn and similar drugs and Celebrex. Finally, the doctor risks or benefi ts were.85 Faced with these two possible areas for claimed Vioxx was safe and eff ective for use in rheumatoid further investigation, Merck decided to see if Vioxx protected arthritis, cancer prevention and treatment, Alzheimer’s disease, stomachs, but chose not to do a heart risk study. and gout, all of which were “off -label” uses, that is, ones for which Vioxx had not been proven safe or eff ective.82 Merck made two informative decisions related to the design of VIGOR, the “does-Vioxx-protect-stomachs?” study that Th e Warning Letter does not suggest that Merck’s compared Vioxx to naproxyn, a generic drug with similar pain representatives tried to correct the doctor’s misleading relieving eff ectiveness but known risks to the stomach. First, A3 TURNING MEDICINE INTO SNAKEOIL

Merck decided that comparing cardiovascular events would Misstating Clinical Data to Claim Reduced Risk not be a “pre-specifi ed analysis,” meaning determining the Th e Warning Letter also explains that Merck’s heart risks posed by Vioxx versus naproxyn was not an original misrepresentations about the VIGOR data included intention of the study.86 Second, Merck excluded patients at falsehoods, namely claiming that the Vioxx patients were only high risk of heart attack from the study. Th at decision lowered four times as likely to have a heart attack as the naproxyn the chance that Vioxx’s potential heart risks would appear in patients, and after accounting for relatively high risk patients, the trial’s results. the Vioxx patients were only twice as likely to have a heart attack as the naproxyn users were. In fact, the Vioxx users were Nonetheless, the VIGOR trial showed patients taking Vioxx fi ve times more likely to have a heart attack over all and three had fi ve times as many cardiovascular events such as heart times as likely once relatively high risk patients were excluded.92 attacks as the patients taking the comparator drug did, Merck told doctors and the public that Vioxx does not hurt hearts, Merck again gave doctors false information about the VIGOR the other drug protects them.87 Only when another clinical data in a letter that Merck sent to thousands of doctors in 2001. trial—conducted fi ve years and several billion dollars in sales Th at letter purported to describe the results from the VIGOR after Vioxx came on the market—found that Vioxx patients trial and asserted that only 0.5 percent of Vioxx participants had a signifi cantly increased risk of heart attacks versus placebo, had incurred “cardiovascular events,” or heart and circulation did Merck acknowledge the danger and remove Vioxx from the problems. Th at would be only 20 of the 4,000 patients who market.88 Th at defi nitive trial, incidentally, was not intended took Vioxx during VIGOR. In truth, 590 Vioxx patients to investigate heart risks. It was intended to determine Vioxx’s suff ered cardiovascular adverse events while taking the drug, eff ectiveness in preventing colon polyps and in the treatment of a fact Merck reported to the FDA, and 101 of the 590 people colon cancer.89 had serious problems, including heart attacks.93

Heart attacks are potentially fatal, millions of Americans are In addition, the New England Journal of Medicine, the already at elevated risk of heart attacks, and Vioxx was intended prestigious journal that published the VIGOR study, for a broad market: the millions of Americans needing pain accused Merck of suppressing key VIGOR data and thereby relief. For those reasons, doctors needed to understand what misrepresenting the study’s heart risk results.94 Merck strongly risk, if any, Vioxx posed to hearts. By choosing to do two large contests the Journal’s critique by emphasizing that the withheld studies designed to provide commercial advantage, designing at data had arisen after a “pre-specifi ed cut-off date.”95 Th e least one of them to reduce the chance of heart risks showing, non-Merck authors of the VIGOR report similarly defend and never choosing to do a large study designed to investigate the report and add that they do not believe including the data heart risks, Merck revealed how deeply marketing concerns, not would have led to diff erent conclusions about the safety of patient concerns, often shape which studies are done. Vioxx.96 Th e New England Journal of Medicine rejectedrejected ttheirheir explanations and reiterated its concerns, emphasizing that the Misleadingly Explaining Away Clinical Data Demonstrating Risks “pre-specifi ed” date was decided near the end of the study, Merck completed the VIGOR study in March 2000. Even that it was inexplicably a month earlier than the cut-off date with the structural advantages given by excluding high risk used for the stomach-protecting data, and that it believed the patients, Vioxx’s heart risks surfaced too strongly to be ignored: omitted results were signifi cant.97 the Vioxx patients were fi ve times as likely as the naproxyn patients to have heart attacks. Not an Isolated Incident While the Warning Letter only addressed six audio Merck decided to explain the VIGOR results in a way that conferences, two sales representatives’ statements and a press minimized their negative impact: Vioxx does not endanger release, it is unlikely that all of Merck’s other promotional hearts, Merck asserted, naproxyn protects them, the way aspirin activities accurately portrayed Vioxx’s risks. In fact, FDA had does.90 Th e FDA’s 2001 Warning Letter criticized Merck for previously objected to promotional materials, in 1999, for promoting this idea, in particular for presenting it as truth similar reasons, suggesting a concerted eff ort by Merck to rather than as hypothesis, since Merck had no evidence that mislead doctors about Vioxx’s potentially fatal risks. naproxyn protected hearts.91 M A Y 2 0 0 6 THE STATE PIRGS A4

CASE STUDY 2: OXYCONTIN Minimizing Addiction Risks to Ease Acceptance TARGETING DOCTORS WITH MESSAGES of a Highly Addictive Drug THAT MINIMIZED RISKS AND BROADENED Purdue Pharma promoted OxyContin for relatively mild THE INDICATION pain relief, positioning it to replace take-as-needed drugs like Percoset.104 Th e promotion involved minimizing the risks of OxyContin, like Vioxx, is a “blockbuster” drug that became the addiction by suggesting that opioid addiction rarely occurred; focus of mass litigation for its maker’s alleged failure to warn of Purdue claimed it happened less than 1% of the time, a fi gure its risks—in this case, risks of addiction and death by overdose. FDA rejected as unsupported.105 As early as 1998, Purdue Pharma distributed to doctors 15,000 copies of a video that OxyContin is a controlled-release opioid designed to provide contained the less than 1% claim106 and “presented the pain pain relief for 12 hours. Like nearly all opioids, OxyContin is relief experiences of various patients and the pain medications, a Schedule II controlled substance, refl ecting its morphine-like including OxyContin, they had been prescribed.”107 Purdue addictiveness. FDA approved it in 1995 for the treatment of distributed a revised version of this video in 2000, which “moderate-to-severe pain where use of an opioid analgesic is the FDA said also appeared to minimize OxyContin’s appropriate for more than a few days;” 98 it came on the market risks.108 Doctors otherwise leery of morphine because of its at a time when treating pain became a medical priority.99 addictiveness often turned to OxyContin, which they thought off ered morphine’s proven pain relief without its proven risks.109 Unfortunately, crushing an OxyContin tablet destroys its time-release mechanism, releasing the large, 12 hour dose of Targeting Primary Care Physicians Instead oxycodone in the tablet all at once. By 2000, large numbers of of Just Cancer or Pain Specialists people, particularly in Appalachia, fi gured this out and started While cancer doctors might seem to be the natural market for crushing and snorting OxyContin or intravenously injecting such a potent narcotic, from the outset Purdue Pharma ran a it.100 By March 2001, the media began to report OxyContin’s massive marketing campaign aimed at primary care physicians new nickname: “Hillbilly Heroin.”101 “Hot spots” of abuse and other doctors.110 By 2003, nearly half of all OxyContin developed in many areas, and the Drug Enforcement Agency prescribers were primary care physicians. Th e DEA feared (DEA) began concerted eff orts to combat it and the associated these doctors lacked adequate training in pain management and crime wave.102 were likely to prescribe OxyContin inappropriately.111

Th e OxyContin story as related below is an important Purdue Focused On Marketing to Doctors, illustration of two parts of the false and misleading marketing Not Patient Safety problem: how drug marketers aggressively target doctors with Using Data to Target Doctors More Eff ectively But Not To Identify misleading messages and increase the size of their potential Problems market by “broadening the indication” of a drug. Th at is, Over time, Purdue dramatically increased its sales force promoting it for its approved use—pain relief—without regard for OxyContin from 618 representatives in 1996 to 1,066 to the safety-based limitations on that use. Th is type of off - representatives in 2001. Th ese representatives were calling on label promotion is more subtle than recommending Vioxx as a some 100,000 doctors a year.112 cancer treatment, but is no less eff ective at increasing the drug’s potential market while endangering patients. Purdue Pharma maximized the eff ectiveness of its sales representatives by using extensive data about doctors’ OxyContin’s original approved use had three signifi cant prescribing habits to determine which doctors to target. Such limitations: “moderate to severe pain,” “opioid analgesic is targeting-by-prescribing-data-analysis is common practice for appropriate,” and “for more than a few days.” Th e record pharmaceutical marketers; they buy data from pharmacies, the strongly suggests Purdue Pharma encouraged doctors to federal government and the American Medical Association and prescribe OxyContin much more frequently than those use it to create detailed prescriber profi les. 113 Trade magazines restrictions would dictate. Moreover, even after the FDA such as Pharmaceutical Executive andand Pharma Marketing News narrowed OxyContin’s approved use to state “moderate to off er extensive advice on how to exploit the data for maximum severe pain when a continuous, around-the-clock analgesic is benefi t.114 Nonetheless, not until the fall of 2002 did Purdue needed for an extended period of time,” the record suggests start using that vast trove of data to search for evidence that Purdue continued promoting OxyContin for broader uses that individual doctors were contributing to the OxyContin abuse FDA had judged do not justify the risks. problem.115

Purdue’s marketing eff orts worked. Th e drug quickly became a Beyond Sales Representatives cash cow, bringing Purdue more than $1 billion/year by 2001, In addition to its sales force, Purdue’s marketing eff orts comprising approximately 90% of Purdue Pharma’s revenues.103 included: A5 TURNING MEDICINE INTO SNAKEOIL

expanding its physician speaker bureau and The FDA Warning Letter conducting speaker training conferences, sponsoring In January 2003, two years after the prescribing information pain-related educational programs, issuing OxyContin was updated and three years after serious abuse problems had starter coupons for patients’ initial prescriptions, surfaced, the FDA wrote Purdue Pharma a Warning Letter sponsoring pain-related Web sites, advertising regarding two print advertisements that Purdue ran in the OxyContin in medical journals, and distributing Journal of the American Medical Association. OxyContin marketing items to health care professionals.116 In the Warning Letter, the FDA castigates Purdue and emphasizes the possible consequences of false or misleading Notably, other than the websites and starter coupons, the advertising, particularly broadening the indication and marketing was not aimed at consumers; doctors were the minimizing risks: primary focus. Your advertisements thus grossly overstate the Purdue Pharma’s OxyContin marketing items, such as safety profi le of OxyContin by not referring in the “OxyContin fi shing hats, stuff ed plush toys, coff ee mugs with body of the advertisements to serious, potentially heat-activated messages, music compact discs, [and] luggage fatal risks associated with OxyContin, thereby tags,”117 made it much easier for doctors to remember that potentially leading to prescribing of the product based OxyContin was an option but did nothing to convey its true on inadequate consideration of risk. In addition, your abuse potential. Indeed, according to the DEA, the scale of journal advertisements fail to present in the body of the OxyContin marketing eff ort was unprecedented for a drug the advertisements critical information regarding with OxyContin’s abuse potential.118 Th e DEA’s concern with limitations on the indicated use of OxyContin, these items is clearer if one imagines Purdue distributing—and thereby promoting OxyContin for a much broader doctors using—fi shing hats, stuff ed plus toys, coff ee mugs with range of patients with pain than are appropriate for heat-activated messages, music compact discs, and luggage tags the drug. Th e combination in these advertisements of labeled “morphine.” suggesting such a broad use of this drug to treat pain without disclosing the potential for abuse with the drug Abuse Rose as OxyContin Became a and the serious, potentially fatal risks associated with its Mainstream Treatment use, is especially egregious and alarming in its potential Despite OxyContin’s potency and abuse potential, the impact on the public health. (emphasis added; FDA marketing campaign made it a mainstream treatment. letter to Purdue Pharma, 1/17/03) According to the Government Accountability Offi ce: Th e FDA required Purdue to not only stop using such ads, In 2001 and 2002 combined, sales of OxyContin but to disseminate corrective information to the audience that approached $3 billion, and over 14 million saw the ads. According to a December 2003 GAO report, prescriptions for the drug were dispensed. Purdue’s corrective ad ran for three months and appeared OxyContin also became the topselling brand-name in approximately 30 medical journals.124 While signifi cant narcotic pain reliever in 2001 and was ranked 15th progress has been made in reducing and limiting OxyContin on a list of the nation’s top 50 prescription drugs by abuse, it more likely results from extensive eff orts made by retail sales.119 the DEA, the hardest hit states, and Purdue, in contexts other than its advertising. Th ese eff orts are discussed in the GAO In 2001, concerned by escalating abuse of OxyContin, the report. FDA required Purdue to change the prescribing information to narrow its approved use to “moderate to severe pain when a It is impossible to say how many patients became addicted continuous, around-the-clock analgesic is needed for an extended to opioids or died from an opioid overdose because of their period of time” and added new warnings.120 Th e FDA followed unnecessary treatment with OxyContin, how many patients’ up the labeling change with an August 2001 letter to Purdue existing opioid addictions were worsened or became fatal stating that thenceforth, all OxyContin promotional materials by moving to the more easily abused OxyContin, or how should prominently disclose the new warnings, precautions, many drug abusers who were not patients were able to get and narrowed indication for OxyContin; Purdue agreed.121 the drug illicitly simply because it was so widely used and Th ese changes were in part “intended to change prescription available. Nonetheless, even Purdue concedes that OxyContin’s practices.”122 It is unclear what eff ect if any these changes had on widespread use and availability may have contributed to its prescriptions, however, as OxyContin sales increased another 7% abuse.125 in 2002 to $1.5 billion and 7.2 million prescriptions.123 M A Y 2 0 0 6 THE STATE PIRGS A6

CASE STUDY 3: PAXIL CR disorder to people experiencing more ordinary degrees CONCEALING SUICIDE RISK DATA AND of anxiety, fear or self-consciousness in social or PROMOTING UNNECESSARY MEDICATION work situations, while also minimizing the serious risks associated with the drug.” (FDA Letter to GlaxoSmithKline made an aggressive eff ort to dramatically GlaxoSmithKline on June 9, 2004 at p. 1, emphasis expand the size of Paxil CR’s potential market. Glaxo did added.) this by concealing data of suicide risks, ineff ectiveness and withdrawal problems and by persuading doctors to prescribe, “Th e TV ad suggests that anyone experiencing anxiety, and patients to demand, the medication for treatment of the fear, or self-consciousness in social or work situations ordinary anxieties of life. is an appropriate candidate for Paxil CR. … Overall, the TV ad suggests that Paxil CR therapy is appropriate Paxil CR is an “SSRI” drug, a type of anti-depressant that for patients with lesser degrees of performance anxiety includes Prozac. In addition to being an anti-depressant, or shyness, which do not generally require drug Paxil won approval for the treatment of a then little known treatment.” (Id. at pp. 2-3; emphasis added.) psychological syndrome known as Generalized Anxiety Disorder. By heavily promoting both the disorder and Paxil, Others besides the FDA were critical of GlaxoSmithKline’s GlaxoSmithKline was able to persuade doctors and consumers messaging. One commentator suggested that GlaxoSmithKline that Paxil was appropriate for the off -label use of the anxieties essentially manufactured a disease in order to maximize sales.127 of every day life. Going from being a preferred treatment for Th e incentive for a drug marketer to broaden its consumers an extreme condition to the preferred treatment for a relatively from those specifi cally identifi ed by the FDA’s narrow normal one transformed Paxil from a potentially niche drug to indications for a drug is obvious; the broader a drug’s use, the a blockbuster. In 2001 alone, GlaxoSmithKline sold $2.1 billion more drugs sold. Furthermore, pharmaceutical companies of Paxil.126 are driven by Wall Street’s expectations to manufacture “blockbuster” drugs.128 As with Hollywood, manufacturing Promoting Paxil for the anxieties of everyday life is precisely the a blockbuster involves a slick marketing campaign aimed at same kind of indication-broadening as promoting OxyContin creating the biggest audience for the drug possible. for ordinary pain. Th e three major diff erences between the OxyContin story and the Paxil story are: “Every marketer’s dream is to fi nd an unidentifi ed or unknown market and develop it,” Barry Brand, • GlaxoSmithKline targeted its Paxil ads at the public as well as SmithKline’s product director for Paxil, told Advertising doctors; Age. “Th at’s what we were able to do with Paxil.”129

• Paxil is not proven to be eff ective in treating ordinary anxiety; Ad-Driven Consumer Demand Results in Unnecessary and Paxil Prescriptions How many thousands of people seeing the ad thought they • OxyContin’s basic risk—drug abuse—was arguably obvious should ask their doctor for a Paxil prescription and face the despite being consistently misrepresented. Paxil’s most risks of taking it, when treatment was unnecessary? A recent famous risk—causing adolescents to commit suicide study130 from the Journal of the American Medical Association or become psychotic—was deliberately concealed by ( JAMA) suggests many thousands or more. GlaxoSmithKline for years. Th e authors of the JAMA study trained actors to portray either Broadening the Indication—Medicating Unnecessarily “major depression” or “adjustment disorder with depressed and Dangerously mood” involving mild and short-term symptoms, and sent Dramatically Expanding Paxil’s Market by Medicating the them to primary care physicians to seek treatment. Ostensibly, Ordinary doctors would prescribe anti-depressant medication to patients Unlike the other FDA letters discussed so far, the Paxil letter with major depression relatively often and rarely, if ever, for focuses on a “direct-to-consumer” advertisement that told the adjustment disorder patients. As JAMA noted, public to take Paxil to cope with everyday anxieties. Th e June 2004 “Untitled Letter” to GlaxoSmithKline discussed a TV Although several small trials suggest that advertisement called “Hello, my name is…”, stating it was: antidepressants confer modest benefi ts on patients with minor depression, there are no data to support “concerning from a public health perspective because their use in adjustment disorder, especially when it broadens the use of Paxil CR beyond the narrowly- characterized (as in our study) by a clear precipitant, defi ned and more serious condition of social anxiety mild symptoms, and short duration.131 A7 TURNING MEDICINE INTO SNAKEOIL

When the actors visited the doctors, they took one of three Putting Profi ts Before People: Suppressing the Risks of actions: one third of the time the actors requested Paxil, Withdrawal mentioning that they had seen an ad for it; one third of the In addition to criticism for the direct-to-consumer advertising time they made a general request for medication; and one addressed by the FDA in the 2004 letter, GlaxoSmithKline has third of the time they did not request medication from the come under fi re for what it failed to say to doctors. doctors they visited. Th e researchers found that while major depression patients were prescribed drugs more often, doctors According to ABC News, a 1997 GlaxoSmithKline safety gave drugs to a signifi cant number of the adjustment disorder review noted that withdrawal eff ects were showing up in Paxil patients. Moreover, requesting Paxil specifi cally aff ected what clinical studies in substantial numbers. 138 One study discussed prescriptions were written: in the safety review showed 25% of Paxil patients experienced withdrawal symptoms, versus only 5.9% of placebo patients. Among [patients] portraying major depression, Another study, involving patients with major depression, [Paxil] was rarely prescribed (approximately 3%) found that 42% of Paxil patients had at least one withdrawal unless the [patient] specifi cally requested Paxil; symptom. Nonetheless, a December 1, 1997-May 31, 1998 if Paxil was requested by name, 14 (27%) of 51 “business plan guide” reported by ABC News instructed received Paxil/paroxetine, 13 (26%) received an sales representatives to “minimize concerns surrounding alternative antidepressant, and 24 (47%) received no discontinuation symptoms” and tell doctors that they occurred antidepressant. in two of 1,000 patients.139

As expected, antidepressant prescribing was less Nor could prescribers debunk these claims by turning to Paxil’s common in adjustment disorder. …Th ere was a offi cial package insert; until December, 2001 it merely included strong prescribing gradient according to request the vague warning that a “withdrawal syndrome” was a rare type: 55% of [patients requesting Paxil] received potential side eff ect, and then GlaxoSmithKline simply added a an antidepressant compared with 39% of [patients] “precaution” to Paxil’s prescribing information to inform doctors making a general request and 10% of those making “discontinuation” symptoms could occur in 2% or greater of no request.132 patients.140 (Th e company preferred the term “discontinuation” because it does not connote addiction as strongly as withdrawal Th e study did not conclude that direct-to-consumer does.)141 advertising for anti-depressants or any other drugs was necessarily bad, suggesting that it can guard against under- Why did GlaxoSmithKline misrepresent the risks of treatment for major depression as well as lead to over- withdrawal? As a company memo explained: treatment for lesser conditions.133 Nonetheless, in light of this study, it is worth reiterating what the FDA letter said about Discontinuation: why this is an issue the Paxil CR TV ad: ’97 Seroxat/Paxil sales to end Sept already exceed $1 Billion Overall, the TV ad suggests that Paxil CR therapy is appropriate for patients with lesser degrees of Th is heading was followed by a picture of a big money bag.142 performance anxiety or shyness, which do not generally require drug treatment. Putting Profi ts Before Children: Suppressing Data Showing Paxil Ineff ective for Adolescents and Suggesting Th e FDA’s Untitled Letter criticizing the Paxil TV did It Created Suicide Risks for Them not require corrective advertising and was received by In a 1998 memo fi rst uncovered by the Newark Star Ledger,143 GlaxoSmithKline fully a month after the ad had stopped GlaxoSmithKline discusses two clinical trials that showed Paxil running.134 was not eff ective, particularly in adolescents; placebo worked just as well. Th e memo states that the company would not Devastating Risks Without Benefi t submit the data—including safety data—to regulators because Prescribing Paxil when it is not needed is dangerous and of the potential negative impact on sales and that it would exposes patients to unnecessary risks. Paxil can cause birth withhold the data until the company could submit it without defects,135 suicide,136 and severe withdrawal.137 Although all of that negative impact: these side eff ects may occur rarely, when the number of people taking a drug is large, a rare risk still hits a signifi cant number “…no regulatory submissions will be made to of people. obtain either effi cacy or safety statements relating to treatment of adolescent depression at this time. However, data (especially safety data) from these studies M A Y 2 0 0 6 THE STATE PIRGS A8

may be included in any future regulatory submissions, the risk of suicide in adolescents, and include those discussed provided we are able to able to go on and generate in the memo above. With the support of a 2001 internal robust, approvable effi cacy data. Th e rationale for company memo to sales representatives that claimed: ‘’Paxil not attempting to get a safety statement at this time demonstrates REMARKABLE Effi cacy and Safety in the is as follows…it would be commercially unacceptable treatment of adolescent depression,’’146 Spitzer also alleged that to include a statement that effi cacy had not been GlaxoSmithKline sales representatives told doctors Paxil was demonstrated as that would undermine the profi le of appropriate for adolescents.147 [Paxil].”144 (emphasis added.) Suppressing data showing that Paxil was not eff ective in New York Attorney General Eliot Spitzer made headlines adolescents and, worse, may cause them to commit suicide or in 2004 when he sued GlaxoSmithKline for consumer become psychotic, boosted Paxil’s sales. According to Spitzer, fraud for its failure to disclose clinical trial data.145 Spitzer’s GlaxoSmithKline reaped $55 million from Paxil prescriptions allegations were based on the fact that GlaxoSmithKline to adolescents and children in 2002 alone.148 sponsored fi ve trials of Paxil in adolescents suff ering from major depression, but published only one of the trials, which Without admitting wrongdoing, GlaxoSmithKline settled had mixed results. Th e four unpublished trials failed to show Spitzer’s suit two months after it was fi led for $2.5 million and any benefi t for the drug and suggested that it might increase agreed to post its clinical data on line.149

CASE STUDY 4: NEURONTIN Other reports noted that tactics marketing fi rms developed for OFF-LABEL PROMOTION AS OFFICIAL Warner-Lambert’s off -label campaign included “continuing MARKETING STRATEGY medical education classes for physicians, “home study kits” for doctors who couldn’t attend meetings, prepaid calling cards that Neurontin, an epilepsy medicine created by Warner-Lambert, would trigger recorded messages about Neurontin, Web sites, is the archetype of off -label promotion. Unlike OxyContin and special supplements to medical journals.”152 and Paxil, Warner-Lambert could not broaden Neurontin’s indication to include everyday ailments. Most people don’t According to Dr. Franklin, a senior Warner-Lambert marketing experience a mild form of epilepsy on a regular basis. executive told his staff of doctors, whose job it was to market to other doctors, that: Th e Neurontin story came out because a whistleblower, Dr. Franklin, fi led suit against Warner-Lambert. Dr. Franklin we need to be holding their hand and whispering contended that Warner-Lambert was not satisfi ed with the in their ear, “Neurontin for pain, Neurontin for limited on-label potential sales for Neurontin and hired an monotherapy, Neurontin for bipolar, Neurontin for advertising agency to design an aggressive, comprehensive everything.”153 strategy to persuade doctors to prescribe it off -label.150 Th at strategy, the New York Times reported,reported, Dr. Franklin was among the doctors being given the instruction. “included paying doctors to appear as authors of journal articles on off -label uses of Neurontin, In these ways, Warner-Lambert promoted the drug to “doctors articles that were actually written by nonphysicians for more than a dozen medical conditions for which it was working under the direction of the company’s not approved, conditions like attention defi cit disorder in marketers. Th e company then paid hundreds of children, neurological pain and bipolar disorder.”154 For at least doctors to attend expensive dinners and weekend one off -label use, migraine pain, Warner-Lambert had data retreats, where they were urged to prescribe showing Neurontin was ineff ective, but suppressed the data and Neurontin. continued promoting for it.155

Other doctors, often frequent prescribers of Driven by the concerted marketing campaign, sales for Neurontin, were paid to speak to other physicians Neurontin soared, reaching over a billion dollars a year. about Neurontin’s benefi ts. Finally, the company paid Some 90% of sales were for off -label uses.156 In 2004, Pfi zer, doctors to prescribe Neurontin and include those which had acquired Warner-Lambert, pled guilty to criminal patients in clinical trials, which Dr. Franklin contends charges stemming from the suit and agreed to pay $430 were designed mainly for marketing purposes.”151 million.157 A9 TURNING MEDICINE INTO SNAKEOIL

CASE STUDY 5: ACCUTANE and brochures and possibly other materials that suggested CONCEALED RISKS AND A WEAK Accutane was an eff ective treatment for depression, while failing to disclose that it had been associated with increased REGULATOR 164 depression. In February 1998 the FDA announced165 it Accutane is an eff ective but dangerous acne drug. It was was adding a suicide warning to Accutane’s label. Th e developed in 1971 but not initially marketed because it causes very next month FDA sent a Warning Letter to Hoff man- birth defects.158 It was approved in 1982 as a treatment of last LaRoche criticizing the Accutane-is-good-for-depression ad resort for “severe, recalcitrant, nodular acne.”159 Few treatments campaign.166 Hoff man-LaRoche took the letter in stride, are helpful for that kind of serious acne. For that reason, with a company spokeswoman admitting: “Once the label Accutane’s known serious birth defect risks were outweighed, changed, we knew we’d have to change our promotion.”167 and the drug was approved. Overtime, evidence has emerged that Accutane can cause suicide as well as severe gastrointestinal An Ineff ective FDA problems. In the drug’s 23 years on the market, the FDA FDA Failure to Add a Warning has revised Accutane’s labeling 29 times without expanding Despite the FDA’s 1998 announcement of a labeling change its “severe, recalcitrant, nodular acne” indication, presumably to warn of suicide risks and the company’s acknowledgement because the risks do not justify any other use.160 Indeed, the of it, the change itself did not occur until May 2000.168 Why birth defect risk is so high and has proved so unmanageable is unclear, although it may have been because of the FDA’s over the years that now patients and their doctors must register inability to dictate labeling changes, as discussed earlier. As a and patients must promise to use two forms of birth control result, Accutane’s prescribing information inaccurately omitted and take regular pregnancy tests. the risk information for nearly two more years without penalty.

Nonetheless, Accutane has become commonly prescribed. Th e Th e FDA Untitled Letter on Misrepresenting Risks FDA estimates 100,000 Accutane prescriptions are fi lled each As a 2002 FDA Untitled Letter describes, at a conference in month.161 Much of that use is likely off -label-by-broadened- March 2002, FDA representatives asked a Hoff man-LaRoche indication, treatment of relatively mild acne. How much of sales representative working an exhibit booth for Accutane if the decision to use Accutane so lightly is a direct result of the package insert contained any information about psychosis Hoff man-LaRoche’s marketing of Accutane is not clear. and suicide. Not only did the sales representative fail to tell the FDA representatives the information that had been in the What is clear, however, is the evidence suggesting that package insert for nearly two years, the representative stated: Hoff man-LaRoche has consistently hidden and downplayed the suicide risk of Accutane. As the case study will show, that “We don’t feel its an issue” behavior is an important illustration of two parts of the false and misleading marketing problem: the ability of companies to “News has hyped it up” hide risk information for commercial advantage and the failure of the FDA to ensure accurate and timely warnings about risk. “Like any drug used in patients with depression, such as penicillin, it could bring it out.”169 Hiding and Misrepresenting Suicide Risks Hiding Information from the FDA Given the consistency and tenor of the comments, the sales By 1997, French regulators had seen enough data on representative could have made similar comments to doctors Accutane’s ability to cause suicide that they required a during visits to their offi ces and at any other professional warning on the French packaging of Accutane. Hoff man- conferences the sales representative attended. LaRoche did not tell the FDA of the French regulators’ decision.162 Also in 1997, a Hoff man-LaRoche doctor Reassuring and misleading statements like those made by studied the suicide data and recommended in an internal the sales representative at the conference are are intended report that a warning be added to Accutane’s U.S. label. to persuade doctors to prescribe Accutane without regard to However, as Hoff man-LaRoche’s Global Head of Regulatory the serious risk of patient psychosis and suicide. Any doctor Aff airs testifi ed, marketing concerns kept the company from who internalizes statements like these could endanger their acting on that recommendation. Instead they sent a copy patients unnecessarily. Indeed, the July 2002 letter describing of the report to the FDA that was edited to remove the the incident notes that the statements “raise signifi cant public recommendation about a warning.163 health and safety concerns.”

Disingenuous Advertising FDA Letter Was Late and Did Not Require Corrective Action Also in 1997, Hoff man-LaRoche ran a yearlong marketing Despite acknowledging signifi cant public health and safety campaign aimed at doctors, using ads in professional journals concerns, and despite the fact that the representative was M A Y 2 0 0 6 THE STATE PIRGS A10

making such misleading statements two years after the labeling stop such advertising but not requiring any corrective action. had been revised, despite the company’s history of disingenuous Moreover, by sending the letter a full four months after advertising and downplaying of risks, FDA sent Hoff man- the conference, the FDA’s silence allowed the misleading LaRoche only an “Untitled Letter,” telling the company to promotional activities to continue for that time.

CASE STUDY 6: TINDAMAX Tindamax is an antibiotic approved to treat specifi c diseases. DISREGARDING THE FDA As the March 3, 2005 letter to Presutti Laboratories, Inc. notes, Presutti sought to add three new uses to its approved Tindamax is not as well known as Vioxx, OxyContin, or indication and submitted data in support of its request. Accutane, at least in part because it has not been the subject Th e FDA denied the new uses because the data did not of mass litigation. Nonetheless, the FDA letter regarding prove Tindamax safe or eff ective for them. Despite being it is noteworthy because it shows the extent to which specifi cally told no, Presutti promoted for those uses anyway companies will disregard the FDA’s instructions and promote by including them in a “Dear Doctor” letter it sent to their drugs off -label. Simply put, Presutti Laboratories prescribers. promoted Tindamax for conditions that the FDA, looking at Presutti’s data, specifi cally said Tindamax could not be As the FDA pointed out to Presutti: “you have encouraged the approved to treat. potentially unsafe use of Tindamax.” 17 TURNING MEDICINE INTO SNAKEOIL

Table 5: Specifi c Recidivism Companies Receiving More Than One Letter On the Same Drug and Same Issue

Date Company of FDA Drug Repeated Deceptive Issue Letters Abbott Laboratories 6/10/04 Norvir Omitted material facts. The 4/27/01 letter was not posted on the 4/27/01 FDA website and is otherwise omitted from this report. Actelion Pharmaceuticals US, Inc. 7/20/05 Tracleer Omitted risk information. 10/30/02 Alcon Laboratories 4/27/05 Cipro HC Otic Made unsupported superiority claims. 7/18/03 Allergan, Inc. 9/6/05 Lumigan Made unsupported superiority claims. Allergan also received a 3/26/01 letter on 6/8/01 regarding a diff erent Lumigan false/misleading promotion issue. Aventis Pharmaceuticals 11/12/03 Taxotere Omitted and misrepresented risk; overstated effi cacy. Also, 12/18/02 Aventis received a letter on 7/26/01 regarding a diff erent Taxotere false/misleading promotion issue. Baxter Pharmaceutical Products 3/26/01 Brevibloc Minimized risks. 10/1/99 Berlex Laboratories 11/20/01 Magnevist Promoted off -label. 4/15/99 Bristol-Myers Squibb Company 8/7/03 Pravachol Broadened indication, overstated effi cacy. 8/13/01 3/29/01 Cubist Pharmaceuticals 8/17/04 Cubicin Omitted risks. 11/22/00 Cytogen Corporation 7/18/05 Quadramet Made unsubstantiated effi cacy claims. The 11/9/01 letter was 11/9/01 not on the FDA website and is otherwise omitted from this report. Elan Pharmaceuticals 8/5/2002 Zanafl ex Made unsubstantiated quality of life claims. 1/16/02 Endo Pharmaceuticals 6/28/05 Lidoderm Omitted and minimized risks. 11/24/99 Genpharm, Inc. 6/18/03 Amnesteem Made unsubstantiated claims and minimized risk. The 2/20/03 2/20/03 letter was not a warning or untitled letter; instead, it proactively forbade precisely the advertising at issue in the 6/18/03 letter. Gilead Sciences, Inc. 7/29/03 Viread Tablets Omitted and minimized risk information. 3/14/02 GlaxoSmithKline, Inc. 1/31/05 Coreg Omitted and misrepresented risk information. 2/9/99 10/15/97 M A Y 2 0 0 6 THE STATE PIRGS 18

Date Company of FDA Drug Repeated Deceptive Issue Letters GlaxoSmithKline, Inc. 7/17/01 Avandia Tablets Minimized risks. 6/28/01 10/20/00 6/29/99 Hoff man-LaRoche Inc. 5/29/03 Xeloda Minimized risks, overstated effi cacy, broadened indication. 1/9/02 Maxim Pharmaceuticals, Inc. 5/31/01 Ceplene Promoted an investigational (not yet approved for anything) 12/27/00 drug. Merck, Inc. 9/17/01 Vioxx Misrepresented risks. 12/16/99 Novartis Pharmaceuticals 4/21/04 Diovan Promoted off -label use. Corporation 11/4/99 9/23/99 Organon Inc. 12/12/03 Follistim Minimized risks. 6/6/98 Pfi zer, Inc. 4/13/05 Zyrtec Made unsubstantiated superiority claims. Pfi zer also received a 7/8/03 letter on 4/30/02 regarding a diff erent Zyrtec false/misleading 12/21/98 promotion issue. Pharmacia Corporation 2/1/2001 Celebrex Misrepresented risks, made unsubstantiated superiority claims. 4/6/00 capsules 10/6/99 Shire Pharmaceutical 5/7/01 ProAmatine Overstated effi cacy, minimized risks Development, Inc. 3/12/01 The 1/23/01 & 3/12/01 letters were not posted on FDA website 1/23/01 and are not otherwise included in this report. 7/19/00 7/6/00 Supergen, Inc. 8/18/05 Nipent Misrepresented risks. 5/10/01 1/6/97 TAP Pharmaceutical Products, Inc. 8/2/02 Prevacid Misleadingly communicated indication. 3/15/00 VIVUS Inc. 5/24/04 MUSE Failed to disclose a material fact. 4/1/98 19 TURNING MEDICINE INTO SNAKEOIL False Marketing Beyond Routine FDA Review

Classic advertising—printed materials, commercials, every two doctors.29 According to a 2005 report branded offi ce supplies and other “tchochkes”—is in the Journal of the American Medical Association just one portion of drug marketing. As discussed ( JAMA), sales representatives have a major impact above, the FDA has a system in place to address on doctors: false classic advertising, albeit an ineff ective and overburdened one. However, a huge portion of “Interactions with pharmaceutical drug marketing is not classic advertising, and representatives increase the likelihood it occurs with little chance of FDA oversight. of physicians making formulary requests Two components are relatively visible—sales for drugs with no clear advantage over representatives and doctors hired has consultants existing ones, prescribing nonrationally, or speakers.26 A third component is largely prescribing costlier drugs, and prescribing invisible, although it has gained increasing and very fewer generic drugs.”30 important attention recently: the manipulation and suppression of clinical trial results. As a Group, Sales Representatives Form Long and Deep Relationships With Doctors Sales representatives and sponsored doctors making Sales representatives may have such a great false or misleading statements about drugs are impact on doctors’ prescribing habits because dangerous because of their many opportunities they begin relationships with doctors while the to instill those messages in doctors. Presenting doctors are still in medical school, relationships clinical trial results to maximize commercial, rather that may condition doctors to be more receptive than clinical, benefi t or simply hiding negative data to sales representatives once they are in practice.31 is scientifi c misconduct and is perhaps the most While in school, students interact with sales dangerous false marketing of all, as it is among the representatives in many ways, participating in most diffi cult to detect and undermines the entire drug company-sponsored meals, receiving small premise of pharmaceutical medicine. gifts, having conference admission paid and other items.32 Th e 2005 JAMA sstudytudy qquoteduoted All three of these types of marketing are currently above surveyed students at eight medical schools outside FDA’s ability to routinely monitor. around the country to assess student attitudes toward pharmaceutical sales representatives and these gifts. Th e survey found that students felt SALES REPRESENTATIVES: WHAT “entitled” to the various sponsorships and gifts and THEY SAY MATTERS commented that the results “suggest[] that as a group they are at risk for unrecognized infl uence Sales representatives are a cornerstone of by marketing eff orts.”33 pharmaceutical marketing. In 2004, more than 90,000 sales representatives called on doctors, more Once doctors have begun their practices, sales than double the amount from a decade earlier.27 representatives continue to interact with them in Drug marketers use prescribing data to focus their many ways. Perhaps one of the most unsettling, sales representatives eff orts and measure their from a patient’s perspective, is the practice of success at increasing prescriptions.28 According “shadowing” doctors while they examine patients, to IMS Health, a business that consults with the access for which they pay the doctors hundreds industry, those representatives focus on the 175,000 of dollars a day. 34 Once in the exam room physicians who write 80 percent of prescriptions, with the doctor, the sales representative can ask resulting in a ratio of one sales representative for patients about their experiences with medications. M A Y 2 0 0 6 THE STATE PIRGS 20

As reported regarding Neurontin’s marketing, Sales Representatives Make False and shadowing enabled sales representatives to meet Misleading Statements with patients, review charts, and recommend Considering how much physician contact sales medications.35 Not until 2003 did the American representatives have, and how strong their Medical Association vote to prohibit this practice relationships can become, the accuracy of their without the patient’s informed consent.36 statements is critically important. Regrettably, the FDA letters discussed above prove that Over time, individual sales representatives develop pharmaceutical sales representatives sometimes say a relationship with individual doctors they see that things to doctors that are dangerously false about can aff ect prescribing decisions in and of itself. the drugs they are promoting. Consider this comment from a sales representative: In all, 30 of the 539 false or misleading messages If I’m close to getting a bonus or making aimed at doctors were statements by pharmaceutical President’s Club, I’ll use the data,” she sales representatives, a number which seems says. “I can go to the doctors who I have a enormous when you consider how small the good relationship with and ask them why possibility is that such statements will come to the they’re prescribing the competition. I’ll tell attention of the FDA. Almost all of the detected them that I’m not far away from a bonus statements were discovered by FDA representatives and ask them to write six prescriptions to asking questions of sales representatives at get me there.37 conferences, an activity the FDA lacks the resources to do comprehensively. Th ese relationships are often cemented with free lunches for the doctor and the offi ce staff and small A 1995 study published in JAMA suggests that as gifts. As one former sales representative recently much as 11% of sales representatives’ statements explained to Th e Atlantic Monthly: are inaccurate.41 Th e authors analyzed 106 statements made by sales representatives during Bribes that aren’t considered bribes… 13 presentations. Statements were considered Th is, my friend, is the essence of inaccurate if they contradicted the 1993 Physician’s pharmaceutical gifting. …Ideally, a rep Desk Reference or materials the sales representative fi nds a way to get into a scriptwriter’s were distributing. Th e representatives made 12 psyche…You need to have talked enough inaccurate statements (11%), all of which were with a script-writer—or done enough favorable to the drug they were promoting. Th e 27 recon with gatekeepers—that you know physicians attending the presentations “generally what to give.38 failed to recognize the inaccurate statements.”42 Given that pharmaceutical marketing is a much In the same article, another sales representative bigger business now than it was in 1995, it is hard explained the purpose of all the free lunches, small to say if the percentage of false statements remains gifts, and attention he lavishes on the doctors he accurate. visits as: “You’re absolutely buying love.”39 A number of doctors who prescribed themselves Sometimes, the gifts aren’t small, and scandals Vioxx are among the plaintiff s now suing Merck, about junkets and expensive presents prompted the Inc. for its failure to warn of Vioxx’s cardiovascular pharmaceutical industry to adopt a voluntary code risk. A critical feature of their allegations is that of conduct regarding gift giving, a code that may Merck sales representatives lied to them about not be eff ective.40 Vioxx’s risks.43

Most troubling of all about the evidence of false or misleading sales representative statements is that 21 TURNING MEDICINE INTO SNAKEOIL

nearly all of the promotional activity between sales Th ese six examples of drug marketers hiding or representatives and doctors occurs without scrutiny misrepresenting trial data are not the only ones of the FDA or any other regulator. uncovered in recent years. A report in the New England Journal of Medicine cited ninenine eexamplesxamples of To learn more about sales representatives’ role in trials where data unfavorable to the drug company deceptive marketing, please see the case studies sponsor were either suppressed or, the investigator in Appendix A, particularly Vioxx, OxyContin, believed, misrepresented in the article that was Accutane, and Paxil. published.46

While suppressing or misrepresenting studies CLINICAL TRIALS demonstrating serious risks or ineff ectiveness are obviously the worst types of marketing-induced Data Suppression and Misrepresentation clinical trial fraud, the FDA letters reveal a diff erent Th e case studies in Appendix A illustrate four and likely more pervasive type: citing studies for examples of how marketing shapes clinical trial product claims they do not support. Fully a third research and data presentation. In the Vioxx of the false and misleading messages in the FDA case study, Merck, Inc. repeatedly chose not to letters related to unsupported or overstated claims do a study that would settle the Vioxx heart (third set of columns in Table 3); many of those risk question. Moreover, when confronted with were unsupported because, as the FDA explained, unfavorable data, the company tried to spin the the studies cited to support them did not in fact data favorably by promoting an unsupported provide support. hypothesis as well as fl atly misstating the numbers. In the Accutane case study, Hoff man-LaRoche In the fi ve years we studied, the FDA sent 25 analyzed clinical trial data, concluded that companies a total of 38 Untitled and Warning adding a suicide-risk warning to its prescribing Letters that involved more than 82 mis-cited information was appropriate, and then, concerned studies.47 Th e FDA noted some studies were too about its impact on marketing, omitted that small for statistical signifi cance or were poorly recommendation when submitting the report to designed. In other cases, the company selectively the FDA. As described in the Paxil case study, presented only the favorable data from the GlaxoSmithKline went further, choosing not to studies. Some times the claims—particularly for report four trials showing complete ineff ectiveness eff ectiveness—were inconsistent with the data in and serious potential risks. Similarly, the the package insert, and no other data was cited. In Neurontin case study notes that Warner-Lambert a few cases the cited studies were not relevant or reportedly suppressed clinical data showing actually contradicted the claims. Neurontin ineff ective for migraine pain. Table 6 gives several examples. More recently, a scientist who conducted a clinical trial for Proctor & Gamble using the drug Actonel Ghostwriters Allow Drug Marketers to Spin has accused the company of misrepresenting Data in Clinical Trial Reports his data, withholding other data from him, Suppressing or misrepresenting data that are and writing abstracts under his name that were commercially damaging, or citing studies for impossible for him to appropriately review.44 commercially favorable propositions they do not Finally, Northfi eld Laboratories, Inc. is currently support, are unfortunately not the only forms of conducting a clinical trial for its blood substitute false and misleading marketing involving clinical PolyHeme without revealing the data from a past trials. Sometimes drug marketers have marketing clinical trial that showed a much higher death professionals—not medical professionals-- rate for people taking PolyHeme instead of real ghostwrite clinical trial reports. Alternatively, blood.45 drug marketers hire medical writers and have their M A Y 2 0 0 6 THE STATE PIRGS 22

Table 6: Selected Studies Drug Marketers Cited for Product Claims They Do Not Support

# Letter Why the study Failed to Support Company Drug, Where Mis-Cited Unsupported Claim Date & the Claim Studies Type SuperGen, Inc. Nipent Exhibit 1 Nipent attacks only 8/18/05 Cited study involves preliminary in panel & leukemia, not stem cells Warning vitro data, and contradicts package handout insert. Bristol Ifex Journal Ad 7 Ifex “minimizes 3/13/01 “none of the 7 articles and abstracts Meyers disruption of a patient’s Untitled discusses the use of Ifex and its Squibb daily activities or eff ect on daily activities” lifestyle” Sanofi -Synthelabo, Inc. Visual Ad 1 Plavix works better than 5/9/01 Selectively presented data and aspirin Untitled misstated data. GelTex, Inc. Renegal Press 1 Renegal reduces cardiac 8/17/01 Interim data; study still on-going. Release calicifi cation; it’s more Untitled eff ective and less risky than other treatments AstraZeneca LP Zomig Detail 1 Zomig is eff ective 10/9/01 Cites a two part study. The fi rst part aid treatment for migraines Untitled is blinded, placebo controlled and randomized, and showed no benefi t from using Zomig for migraines. Part 2 was open-label, uncontrolled and non-comparative. The detail aid cited part 2 to claim Zomig was eff ective for migraines. Alcon Cipro HC Otic 3 (a) Cipro HC Otic 7/18/03 (a) Cited study involved in vitro is safer than other Untitled data; (b) cited study was open label treatments; (b) Cipro (allowing bias) and too small for HC Otic is better than statistical signifi cance; (c) cited data another treatment; (c) do not involve Cipro HC Otic. in vitro, Cipro HC Otic kills 99.9% of certain bacteria XCel Migranal More than Migranal has 12/19/03 Cited studies do not compare Detail Aid 1 comparable safety to Untitled Migranal and triptans. Triptans, another type of drug, and also is better than Triptans Janssen Risperdal, 8 Risperdal does not 9/04/04 Risperdal’s label has a warning Pharmaceutica “Dear Health increase risk of diabetes Warning on the risk of diabetes, the cited Care Provider studies “do not represent the Letter” weight of the pertinent evidence” and two of the cited studies show Risperdal increasing the risk of diabetes. 23 TURNING MEDICINE INTO SNAKEOIL

marketers supervise their work. In both cases the Two years later, the New York Times reportedreported tthathat reports are later published under doctors’ names advertising agencies had purchased or signifi cantly without revealing the marketers’ input. Th is type invested in the companies that conduct clinical of ghostwriting ensures that the commercially trials for pharmaceutical companies. With favorable results are highlighted and the marketing fi rms profi ting from and controlling unfavorable ones are downplayed. these clinical trial companies, it seems likely that the prevalence of the tactic may have increased.51 Having marketers prepare/oversee preparation of articles for publication in scientifi c journals is Marketers ghostwriting for doctors is not limited particularly chilling because the published studies, to clinical trial results. A public relations fi rm granted the approval and gravitas of the journal, are working for GlaxoSmithKline ghostwrote a letter what passes for scientifi c proof. Warner-Lambert minimizing Paxil’s withdrawal symptoms that was hired an advertising fi rm to prepare 20 such articles eventually edited and signed by Dr. Bruce Pollack for Neurontin.48 Marketing ghostwriters were of the University of Pittsburg and published in also hired to help promote fen-phen, the notorious the Journal of Clinical Psychiatry.52 Paxil sales diet drugs that have since been removed from the representatives found the letter very helpful in market.49 responding to doctors’ concerns about Paxil and withdrawal. While Dr. Pollack denies the letter Th e New England Journal of Medicine ppublishedublished a was ghostwritten, GlaxoSmithKline documents report in 2000 that describes ghostwriting: reveal that a public relations fi rm hired by the company sought to have such a letter published, “More recently, a practice that one might identifi ed three doctors who would make good call the nonwriting author–nonauthor ‘authors,’ one of whom was Dr. Pollack, and writer syndrome has developed. Many prepared a draft that made the same points, in the interviews conducted for this report same order, that Dr. Pollack’s eventual letter did.53 confi rmed the wide prevalence of this syndrome in publications of drug-trial Drug Companies Shape the Studies for reports, editorials, and review articles. Th e Marketing Purposes syndrome has two features: a professional Embedding Marketing Concerns in the medical writer (“ghostwriter”) employed Clinical Trial Design by a drug company, CRO, or medical Even further in the shadows than the ghostwriters communications company, who is paid are the ways marketing drives the design of clinical to write an article but is not named as an trial studies themselves. Th e New England Journal author; and a clinical investigator (“guest of Medicine articlearticle quoted aboaboveve rreviewedeviewed tthehe wwaysays author”), who appears as an author but clinical trials can be designed to reliably produce does not analyze the data or write the results favorable to drug companies: manuscript. “If a drug is tested in a healthier population …In one study, 19 percent of the articles (younger, with fewer coexisting conditions surveyed had named authors who did and with milder disease) than the not contribute suffi ciently to the articles population that will actually receive the to meet the criteria for authorship of the drug, a trial may fi nd that the drug relieves International Committee of Medical symptoms and creates fewer adverse eff ects Journal Editors. Eleven percent had than will actually be the case….If a new ghostwriters who contributed to the drug is compared with an insuffi cient dose work but were not named as authors.”50 of a competing product, the new drug will (Internal citations omitted.) appear more effi cacious….Clinical trials often use surrogate end points that may M A Y 2 0 0 6 THE STATE PIRGS 24

not correlate with more important clinical case study in Appendix A. After choosing not to end points. Companies may study many investigate heart risks, Merck, Inc. tried to not to surrogate end points and publish results fi nd them in its VIGOR trial, which it conducted only for those that favor their product.”54 to see if Vioxx protected stomachs. In designing (internal citations omitted.) VIGOR, Merck chose to exclude people at high risk of heart attack, and at the outset did not intend A similar critique was off ered by Dr. Richard to analyze any cardiovascular results. Smith, editor of the British Medical Journal fforor 25 years.55 His 2005 essay entitled, “Medical Journals Using Clinical Trials to Build Prescribing Habits Are an Extension of the Marketing Arm of In trials conducted after a drug has been approved, Pharmaceutical Companies,” has this sidebar: the conduct of the trial itself is can be a marketing device to build prescribing habits in doctors. “Examples of Methods for Pharmaceutical Consider what an executive of a company that Companies to Get the Results Th ey Want conducts these post-approval trials said to the from Clinical Trials Boston Globe: • Conduct a trial of your drug against a treatment known to be inferior. “In addition, the studies are designed to • Trial your drugs against too low a dose get hundreds of doctors in the habit of of a competitor drug. prescribing a certain drug,” said another • Conduct a trial of your drug against panelist, Robert Deininger, executive vice too high a dose of a competitor drug president of AAI International of Natick, (making your drug seem less toxic). which also conducts these trials for drug • Conduct trials that are too small to companies. “It’s also a platform for direct- show diff erences from competitor drugs. to-consumer advertising.”” 58 • Use multiple endpoints in the trial and select for publication those that give Drug Companies Sponsor Most Research and favourable results. Their Tactics Produce Self-Serving Outcomes • Do multicentre trials and select for By 2001, more than 80 percent of all clinical publication results from centres that are trails were funded by the pharmaceutical favourable. industry.59 Not surprisingly, given the infl uence • Conduct subgroup analyses and drug companies have on the design and select for publication those that are publication of studies, scholars have found that favourable. sponsorship by pharmaceutical companies is • Present results that are most likely to a strong predictor of results: trials of cancer impress—for example, reduction in drugs sponsored by the industry were eight relative rather than absolute risk.”56 times more likely to reach positive conclusions, and 98 percent of papers based on industry- Dr. Smith also points out that drug marketers will sponsored research favored the drugs being publish the same data multiple times to magnify the examined.60 In contrast, 79 percent of studies eff ect of the positive results and that journal editors based on independent research were favorable. 61 have a diffi cult time identifying the problem.57 Dr. Richard Smith, in his essay describing trial design manipulation by drug companies, stated An example of trying to design a study to get that company sponsored trials were four times as the desired results and no others is in the Vioxx likely to be favorable.62 25 TURNING MEDICINE INTO SNAKEOIL Solutions

Th e FDA letter system, as discussed above, is result, many medical researchers and organizations ineff ective at correcting or deterring deceptive recognize that new laws are needed to ensure that classic advertising and does almost nothing to registration of all trials is done consistently and watchdog non-classic advertising. Whistleblowers with clinically meaningful information.67 and crusading attorney generals have been eff ective in isolated instances but cannot be counted upon to To eff ectively address all of the methods of appear every time. Nor are personal injury lawsuits, distorting clinical trials discussed above, a registry the engine that eventually brings much of the truth must include for each trial: to light and triggers other enforcement action, an effi cient way to prevent and correct false advertising. • Th e key features of a trial’s protocol, with any amendments, including the trial’s purpose, drugs Th e most eff ective and possible changes that can be involved, outcomes to be measured, when they are made are at the state level. to be measured, etc.;

• an explanation for key features, such as the STEPS THE STATES CAN TAKE criteria used to include or exclude patients, and the dosages of drugs chosen for comparison Creating an Independent Information purposes; Resource for Prescribers and Patients-- A Publicly Accessible Clinical Trial Database • the results of the trial, including the adverse Clinical trial registries, i.e. searchable databases events, and if the drug is already marketed in the of trials and results, are one answer to clinical U.S., where the adverse events are refl ected in the trial data manipulation and suppression.63 drug’s package insert, including a direct quotation 68 Medical journals have made important headway of the relevant part of the package insert; and in increasing the registration of clinical trials by refusing to accept reports from trials that are not • whether any of the data were published, and registered.64 Most researchers comply with this where; the authors of any publication, the requirement by registering their trials with www. company who employs them and their role clinicaltrials.gov, a largelylargely vvoluntaryoluntary ffederalederal rregistryegistry in the publication; trial funding sources; that does not serve the core purpose of ensuring fi nancial relationships between principal that doctors and patients have the information they investigators and the pharmaceutical sponsor; need to make sophisticated and tailored prescribing fi nancial relationships between authors and the decisions. Drug marketers and researchers often pharmaceutical sponsor; and any other factor a take advantage of the registry’s lax and minimal competent regulator deems necessary. registration requirements by using vague and clinically useless language.65 Private Attorneys General It would take vast resources to enable the FDA to Recognizing this problem—drug marketers can eff ectively monitor sales representatives’ promotional meet the registration requirement for publication activities, ensure that drug marketers submit all and still not provide needed clinical data—the materials for review, and ensure that all submitted editors of the New England Journal of Medicine materials are reviewed in a timely way. One solution have urged researchers to register comprehensive, is to deputize prescribers, the public, and their specifi c, clinically important information.66 As a associations to bring enforcement lawsuits. M A Y 2 0 0 6 THE STATE PIRGS 26

In order to ensure that a plaintiff will step forward, representative say. While the Pennsylvania eff ort was standing must be broadly available and attorneys’ initiated as a cost containment eff ort, the project also fees must be awarded to victorious plaintiff s. To should improve doctors’ information about risks and guard against nuisance lawsuits, only injunctive benefi ts. Th e “unsales” representatives go to doctors’ relief rather than monetary damages should be offi ces much like drug company sales representatives available, and the threshold of actionable false do, but they practice “academic detailing”—eff orts advertising should be relatively high: to win, a to educate doctors about the evidence relating to plaintiff should show the advertising created a all treatments, with at least one goal being shifting public health risk. (It’s worth noting that in several prescribing patterns to cheaper drugs that are just of the letters discussed in this report, the FDA eff ective as the new ones being pushed by the drug charged that the advertising posed signifi cant company sales representatives. Th at eff ort is too public health risks.) In addition to injunctions, new to see results; however, it is an experience states judges should be empowered to levy civil fi nes that should monitor, and if eff ective, mimic.69 range up to punitive levels, based on factors such as severity of the public health risk, whether the drug marketer had disseminated similar advertising FDA APPROACHES before, etc. Th e fi nes would be payable to the state. IMPROVING FDA ENFORCEMENT More Sunshine To improve patient safety, we must restructure the States must require better disclosure of the FDA FDA, particularly to separate drug safety from 70 Untitled and Warning Letters and any other drug approval. Such reform would undoubtedly evidence of deceptive marketing, including improve oversight of drug marketing. Short company documents that were made public as of a complete overhaul, however, we could take a result of litigation. Drug marketers should be signifi cant albeit expensive steps to improve the required to prominently display them on their federal regulation of deceptive drug marketing. websites and disseminate them to the press. Similarly, the state’s Director of Consumer Aff airs Fixing the FDA Letter System or Department of Health should issue an annual To make the FDA letter system an eff ective report highlighting the FDA letters and any other enforcement system, we recommend several evidence of deceptive drug marketing relating changes, nearly all of which require substantial to any drug sold in the state. In addition, the investments of new resources. department should post that information on its website, along with plain English summaries as • FDA must review every piece of ‘classic’ necessary to be sure the public can understand how advertising before it is disseminated. Given and why the messages were deceptive. Finally, the FDA’s current inability to review all pieces of state should require drug marketers to disseminate advertising even after the fact, and industry’s the FDA letters, in envelopes easily recognized as need for quick, commercially relevant review, containing such letters, to every person licensed by preventative screening would require a sharp the state to prescribe drugs. increase in staffi ng.

Th e comprehensive clinical trial registry, the new • FDA must review all the training materials enforcement action and the improved publicity of for sales representatives and observe training evidence of deceptive marketing would both be sessions unannounced. Without such powers corrective and deterrent. and resources, FDA cannot eff ectively police what sales representatives say when they are meeting Counter-Advertising with doctors. Given the number of sales forces Pennsylvania’s Department of Aging has hired an and trainings, serious monitoring would require a “unsales” force to counter the eff ects of what drug sharp increase in FDA staff . 27 TURNING MEDICINE INTO SNAKEOIL

• FDA must send personnel to a substantial the attention of doctors otherwise inundated number of conferences where sales with unsolicited mail. Dissemination of the representatives are staffi ng exhibit booths and regulatory letters is a critical component of check their promotional activities. Again, corrective advertising. without these resources it is impossible to ensure that large-scale deceptive advertising does not THE MEDICAL PROFESSION’S ROLE happen at these events. Because of the large number of events each year, and the large number Improve Prescriber Education and Information of exhibit booths at each, implementing this Resources recommendation also requires a sharp increase in Th e medical profession and the independent FDA staff . organizations and academic institutions that service it can help. • For deterrence purposes, FDA must have, and use, punishments short of removing a drug from • Doctors need better access to independent, the market. If false advertising recidivism quickly accurate, digested information about drugs. Th e and consistently resulted in punitively large fi nes, information produced by the clinical trial registry FDA could deter drug companies from repeating should be packaged by an independent group or the off ense. agency into a form easily useable by prescribers who want information about treatment options. • FDA must make the letters more useful to Th e information provided should include not only doctors and the public. FDA should post all the clinically important information about each FDA Warning and Untitled Letters for false drug, but also how the drug compares to other and/or misleading marketing on a single website, treatments in terms of safety, effi cacy, and cost. indexed and searchable by type of violation, 71 company, drug and date. Such organization Th e Drug Eff ectiveness Review Project would make it much easier for prescribers and (DERP) generates this information, but it is patients alike to see who is committing the most aimed more at policy makers than prescribers. dangerous sorts of violations and for which Similarly, Consumers Union takes DERP’s drugs. Similarly, every Warning and Untitled data and packages it for patients, as part of its Letter should be accompanied by a plain English BestBuyDrugs project. To the extent that the summary so that non-medical professionals can information is already accessible (for example, Th e understand what was wrong with the marketing. Medical Letter), the profession must fi nd a way to ensure that doctors use it. Only by breaking their • FDA must require corrective action in every reliance on sales representatives and other sources case. Because false and misleading drug of promotional information can doctors ensure advertising poses a serious public health risk, they are getting unbiased information. FDA should require the drug marketers to correct their misleading claims. FDA should • Medical schools and teaching hospitals also require drug marketers to send to all should heavily invest in training students and relevant prescribers a copy of the Untitled or residents to be skeptical of pharmaceutical sales Warning Letter in a uniquely and prominently representatives and to rely on independent sources marked envelope so that the letters catch of information. M A Y 2 0 0 6 THE STATE PIRGS 28 Conclusion

Deceptive marketing of prescription drugs is Doctors must recognize that drug marketers are pervasive and dangerous. Most Americans take regularly targeting them with deceptive messages prescription drugs at some point in their lives, about risk and eff ectiveness, and increase their and many take medications every day. When skepticism of sales representatives, promotional the doctors prescribing those medicines have materials, and even drug company-sponsored internalized the deceptive messages about drugs’ clinical trials by an order of magnitude. risks and eff ectiveness, those prescribing decisions may harm patients, not heal them. Taken by the Patients must understand that the deceptive wrong patient, a medicine is a poison. marketing sometimes works. 29 TURNING MEDICINE INTO SNAKEOIL Endnotes

1 See, “FDA Oversight of Direct-to- Jr. Pharmaceutical Executive Eugene: Mar Maxim Pharmaceuticals, MedImmune, Consumer Advertising Has Limitations” 2003. Vol. 23, Iss. 3, p. 102 (discussing Merck & Co., Nephrx, Novartis United States General Accounting Offi ce, a Wall Street Journal Online/Harris Pharmaceuticals Corporation, Novo Nordisk October 2002, available at http://www. Interactive Healthcare poll) Pharmaceuticals, OraPharma, Inc., Organon, gao.gov/new.items/d03177.pdf, hereinafterhereinafter 8 Ortho-McNeil Pharmaceutical, Inc., Pfi zer, “GAO DTC Report”; “Infl uence of “Paternalism Costs Lives” H. I. Miller Pharmacia Corporation, Pharmacia & Patients’ Requests for Direct-to- Wall Street Journal Mar. 2, 2006 at p. A14. Upjohn, Pharmacyclics, Inc., PharmaMar, Consumer Advertised Antidepressants: 9 Praceis Pharmaceuticals, Presutti A Randomized Controlled Trial”, R. L. “Reform of Drug Regulation — Beyond Laboratories, Inc., Procter & Gamble Kravitz; R. M. Epstein; M. D. Feldman; C. an Independent Drug-Safety Board,” W. Pharmaceuticals, Purdue Pharma L.P., E. Franz; R. Azari; M. S. Wilkes; L. Hinton; A. Ray and C. M. Stein, N. Engl. J Med. Roche Laboratories, Sanofi -Synthelabo, P. Franks. JAMA. 2005;293:1995-2002, at 354;2 at 194. (Describes how limited risk Inc., Santen, Inc., Schering Corporation, 2000. (JAMA Paxil Study). “Do Ads Really and effi cacy information is under the current Shire Pharmaceutical Development, Inc., Drive Pharmaceutical Sales?” S. Findlay. FDA process, with examples.) SkyePharma, Supergen, Inc., Takeda Marketing Health Services Chicago :Spring 10 Pharmaceuticals North America, Inc., 2002. Vol. 22, Iss. 1, p. 20-25; “DTC Ads 21CFR202.1 TAP Pharmaceutical Products, Inc., Infl uence Majority Of Consumers, Say 11 United Th erapeutics Corporation, VIVUS, Doctors” R. Th omaselli, Advertising Age Re accurate understanding of risks Inc., Watson Pharmaceuticals, Women’s Chicago: Jan 20, 2003. Vol. 74, Iss. 3, p. 6. deterring both doctors and patients from a Capital Corporation, Wyeth-Ayerst, Wyeth 2 treatment, albeit in the context of surgery, Pharmaceuticals, Xanodyne Pharmacal, Inc., “Medical Students’ Exposure to not prescription drugs, see “Medicare Says It Xcel Pharmaceuticals and Attitudes About Drug Company Will Pay, but Patients Say ‘No Th anks’,” G. 14 Interactions A National Survey” F. S. Kolata New York Times, New York, Mar. 3, Warning Letter to Pharmacia regarding Sierles, A. C. Brodkey, L. M. Cleary, MD, F. A. 2006 at C.1, C.3. Celebrex marketing on 2/1/01, total McCurdy, M. Mintz, J. Frank, D. J. Lynn, J. 12 deceptive messages disseminated: 43 (5 Chao, B. Z. Morgenstern, W. Shore, J. Woodard We reviewed all the letters at this website: audio conferences by a Pharmacia-sponsored JAMA, September 7, 2005—Vol 294, No. http://www.fda.gov/cder/warn/warn2005. doctor promoted 2 unapproved uses, 9 pp. 1034-1042; “Infl uences on GPs htm. Four of the letters made reference claimed superiority to three treatments Decisions to Prescribe New Drugs—the to Untitled Letters from the time period without support, omitted risks, minimized Importance of Who Says What” H. Prosser, we studied but which were not posted at a dangerous drug-drug interaction, S. Almond, and T. Walley, Family Practice the site. See Table 5. As a result, it is not and minimized a contra-indication (5 2003, 20:61-68. possible to know how many letters were sent x 8 = 40); a sales aid, a card, and a wall 3 out in the time frame we studied, and from chart all minimized a contra-indication “Infl uences on GPs Decisions to Prescribe the very beginning our numbers understate (3 x 1 = 3)). Warning Letter to Merck New Drugs—the Importance of Who Says the true problem. regarding Vioxx marketing on 9/17/01, What” H. Prosser, S. Almond, and T. Walley, 13 total deceptive messages disseminated: Family Practice 2003, 20:61-68 Abbot Laboratories, Actelion 45 (6 audio conferences by a Merck- 4 Pharmaceuticals US, Inc., Alcon Research, sponsored doctor minimized heart attack “Gifts to Doctors Is Eff ective Marketing, Ltd., Allergran, Inc., Alza Corporation, risks, omitted other risks, minimized a Some Drug Firm Employees Say” S. Amgen, AstraZeneca, Aventis, Barr dangerous drug/drug interaction, promoted Murphy. Knight Ridder Tribune Business Research, Inc., Baxter Pharmaceuticals, at least 3 unapproved uses and made an News Washington: Nov 17, 2002. p. Bayer Corporation, Berlex Laboratories, unsubstantiated claim versus other NSAIDs 1 (company memos released as part of Biovail Pharmaceuticals, Boehringer (6 x 7 = 42); 2 sales reps each minimized litigation reveal eff ectiveness of tactics.) Ingelheim, Bracco Diagnostics, Bradley risks at a conference (2 x 1 = 2); and one “’Tis Always the Season for Giving” E. Pharmaceuticals, Braintree Laboratories, press release misrepresented safety (1 x 1 = Clayton, CALPIRG, Sept. 2004. Bristol-Myers Squibb, Centocor, Inc., 1)) Untitled Letter to Cephalon regarding 5 Cephalon Inc., Connetics Corporation, Provigil marketing on 1/3/02, total deceptive “Do Drug Samples Infl uence Resident Corixa Corporation, Critical Th erapeutics, messages disseminated: 44 (8 journal ads Prescribing Behavior? A Randomized Inc., Cubist Pharmaceuticals, minimized abuse potential, promoted for Trial” R. F. Adair, L. R. Holmgren. The unapproved use of daytime sleepiness and American Journal of Medicine. New York: Cytogen, Dutch Ophthalmic USA, misleadingly claimed the drug’s mechanism Aug 2005.Vol.118, Iss. 8; pg. 881; “Drug Elan Pharmaceuticals, Eli Lilly, Endo of action was known (8 x 3 = 24); 4 sales Samples Sway Doctors” J. Olson. Knight Pharmaceuticals, Forest Laboratories, aids promoted unapproved use of daytime Ridder Tribune Business News. Washington: Fujisawa Healthcare, Galderma sleepiness (4 x 1 = 4); 5 sales aids promoted a Jul 28, 2005. pg. 1 Laboratories, GelTex Pharmaceuticals, misleading switch protocol and misleadingly 6 Inc., Genentech, Inc., Genzyme Corp, claimed the drug’s mechanism of action “Kaiser goes generic / HMO limits Gilead Sciences, Inc., GlaxoSmithKline, was known (5 x 2 = 10); 3 sales aids made senior coverage, raises co- payments” Guilford Pharmaceuticals Inc., Hoff man- unsubstantiated claims of superiority to V. Colliver. San Francisco Chronicle. San LaRoche, ICN Pharmaceuticals, Inc., ISTA other treatments (3 x 1 = 3); 1 website Francisco, Calif.: Oct 25, 2003. pg. B.1 In Pharmaceuticals, Inc., Janssen Research promoted for unapproved use of daytime addition, patients without insurance are Foundation, Janssen Pharmaceutica, Johnson sleepiness, promoted a misleading switch more likely to be infl uenced by a drug’s cost. & Johnson, King and Spaulding/Genfarm, protocol and misleadingly claimed the drug’s 7 Kos Pharmaceuticals, Lifecyle Ventures, mechanism of action was known (1 x 3 = 3). “Patients trust their doctors” A. Branch Lilly ICOS LLC, Mallinckrodt Inc., M A Y 2 0 0 6 THE STATE PIRGS 30

15 24 GAO DTC Report at 17. For example, the Warning Letter sent to Zaneski. Knight Ridder Tribune Business 16 GlaxoSmithKline for its Paxil advertising, News Washington: Jun 17, 2004. p. 1 “‘Off -label’ drugs take their toll” discussed in Appendix A, was received a 30 A. Young and C. Adams Knight Ridder month after the deceptive TV ad stopped Quote is from: “Medical Students’ Newspapers Nov. 02, 2003 (http://www. running. Th e problem is systemic, and Exposure to and Attitudes About Drug realcities.com/mld/krwashington/news/ refl ects a procedural change instituted Company Interactions A National Survey” special_packages/riskyrx/7146578.htm?temp in 2002. GAO DTC Report at 17 and F. S. Sierles, A. C. Brodkey, L. M. Cleary, MD, late=contentModules/printstory.jsp) 24. (“FDA’s oversight has been adversely F. A. McCurdy, M. Mintz, J. Frank, D. J. 17 aff ected by a January 2002 change in its Lynn, J. Chao, B. Z. Morgenstern, W. Shore, J. Ibid. procedures for reviewing draft regulatory Woodard JAMA, Sept. 7, 2005—Vol 294, No. 18 letters that was directed by the Department 9 1034-1042 at 1034-35. A comprehensive Th e FDA’s diffi culty in getting warnings of Health and Human Services (HHS). review of the studies documenting the power and other labeling changes made post- Th is change has signifi cantly increased the of sales representatives is “Physicians and marketing is not the only evidence of drug time between DDMAC’s identifi cation the Pharmaceutical Industry, Is a Gift marketers’ post-approval power. Recent of a misleading advertisement and FDA’s Ever Just a Gift?” A. Wazana, JAMA 2000; reports indicate that the clinical trials request to remove it from dissemination, 283:373-380. See Also, “Infl uences on GPs companies promise to do after a drug is with the result that some regulatory letters Decisions to Prescribe New Drugs—the approved—studies that are part of the reason may not be issued until after the advertising Importance of Who Says What” Prosser the FDA agrees to approve the drug— campaign has run its course.”) H., Almond S. and Walley T., Family Practice usually are not done. Prior to approval, the 25 2003 20:61-68 (fi nding sales representatives FDA can force a company to do any study See, “Complex Drug Labels Bury were one of the largest infl uences on English the FDA wants, provided the drug marketer Safety Message Warnings Elude Patients, doctors’ prescribing decisions.) Despite is suffi ciently interested in getting the right Doctors Alike” R. Rubin USA Today this demonstrated power, pharmaceutical to market. See “New Drugs Hit the Market, McLean, Va.: May 3, 2000. p. 01A; “Th e marketers are increasingly disenchanted with but Promised Trials Go Undone,” G. Harris Fine Print: Are Too Many Doctors traditional sales representative strategies; New York Times New York, Mar. 3, 2006 Missing Safety Alerts on Drugs?” C. Adams some advocate a shift to more on-line at B. 1. Wall Street Journal New York, N.Y.:Mar promotion. See, e.g., “Pharma Marketing 19 24, 2000. p. B1. “Th e Weak Link in the News Special Supplement, eDetailing” “Better Warning Labels For Drugs Drug-Safety Chain: Doctors” P. Raeburn. published by VeriSci Newton, PA 2005. Urged by FDA” G. Hess. Chemical Market Business Week NewNew YYork:Aprilork:April 10, 2000. 31 Reporter NewNew YYork:Marork:Mar 7, 2005. VVol.ol. 267, Iss. 3676, p. 50 See Also, FDA talk paper “Eff ect of Restricting Contact Between Iss. 10, p. 24 (citing testimony of FDA Jan 2000 (Th e FDA found that 229 adverse Pharmaceutical Company Representatives before a Senate Committee), “FDA Seeks events from Propulsid, including 70 deaths, and Internal Medicine Representatives More Say On Drug Labels; Regulators involved patients with risk factors listed on Post-Training Attitudes and Behavior” Want Greater Authority To Dictate in Propulsid’s label) and “AdherenceAdherence to B. B. McCormick, G. Tomlinson, P. Brill- Warnings” C. Rowland Boston Globe Black Box Warnings for Prescription Edwards, A. S. Detsky, JAMA, October Boston, Mass.:Mar 2, 2005. p. C.1 (same) Medications in Outpatients,” K. E. Lasser, 24/31, 2001 Vol 286, No. 16 (med students 20 D. L. Seger, D. T. Yu, A. S. Karson, J. M. at school that forbade interactions with “Testimony Before the Senate Finance Fiskio, A. C. Seger, N. R. Shah, T. K. Gandhi, J. representatives during school were more Committee” Dr. David Graham, Nov. 18, M. Rothschild, D. W. Bates, Arch Intern Med. skeptical of the sales representatives, and 2004. 2006;166:338-344 (showing that doctors saw them less often, once in practice than 21 prescribe drugs with black box warnings to students from a school that allowed sales “Better Warning Labels For Drugs precisely the patients the warning counsels representatives to interact with students Urged by FDA” G. Hess. Chemical Market against). and sponsor events.) “Medical Students’ Reporter NewNew YYork:Marork:Mar 7, 2005. VVol.ol. 267, 26 Exposure to and Attitudes About Drug Iss. 10, p. 24 (citing testimony of FDA Doctor-to-doctor deceptive marketing is Company Interactions A National Survey” before a Senate Committee), “FDA Seeks not a focus of this report, but to understand F. S. Sierles, A. C. Brodkey, L. M. Cleary, More Say On Drug Labels; Regulators the role it can play, please see the case MD, F. A. McCurdy, M. Mintz, J. Frank, D. Want Greater Authority To Dictate studies in Appendix A, particularly Vioxx J. Lynn, J. Chao, B. Z. Morgenstern, W. Shore, Warnings” C. Rowland Boston Globe and Neurontin. J. Woodard JAMA, September 7, 2005—Vol Boston, Mass.:Mar 2, 2005. p. C.1 (same) 27 294, No. 9 1034-1042 (citing one other 22 “Marketing to Doctors a Key Approach study). “OxyContin Abuse and Diversion and for Drug Sales Representatives” C. T. 32 Eff orts to Address the Problem” United Zaneski. Knight Ridder Tribune Business “Medical Students’ Exposure to States General Accounting Offi ce Report, News Washington: Jun 17, 2004. p. 1 and Attitudes About Drug Company December 2003, at p. 28, (Hereinafter 28 Interactions A National Survey” F. S. “GAO OxyContin Report”). See the discussion in the OxyContin Sierles, A. C. Brodkey, L. M. Cleary, MD, F. A. 23 case study in Appendix A at FNs 115 and McCurdy, M. Mintz, J. Frank, D. J. Lynn, J. Id. See Also, GAO DTC Report at 17, 116 infra, and the accompanying text, for Chao, B. Z. Morgenstern, W. Shore, J. Woodard 21-22. (Discusses FDA’s inability to review more information on how pharmaceutical JAMA, September 7, 2005—Vol 294, No. 9 all direct-to-consumer advertising and companies target doctors. 1034-1042 at 1036 its methods of prioritizing what does get 29 33 reviewed.) “Marketing to Doctors a Key Approach Ibid., at 1040. for Drug Sales Representatives” C. T. 31 TURNING MEDICINE INTO SNAKEOIL

34 53 “Marketing to Doctors a Key Approach SuperGen Inc. 5/10/01 Untitled Letter; Ibid. Th ee marmarketers’keters’ letterletter andand DDr.r. for Drug Sales Representatives” C. T. Allergan, Inc. 6/8/01 Untitled Letter; Pollack’s letter are available at http:// Zaneski. Knight Ridder Tribune Business Pfi zer 6/29/01 Untitled Letter; Connectics, abcnews.go.com/images/Primetime/paxil_ News Washington: Jun 17, 2004. p. 1 See Inc. 8/13/01 Untitled Letter; GelTex, Inc. pollock.pdf Also, “Suit Says Company Promoted Drug 8/17/01 Untitled Letter; Merck, Inc. 9/17/01 54 in Exam Rooms” M. Petersen New York Warning Letter; AstraZeneca LP 10/9/01 “Uneasy Alliance — Clinical Times New York, N.Y.: May 15, 2002. p. Untitled Letter; Fujisawa 11/13/01 Untitled Investigators and the Pharmaceutical C.1 (discusses allegations about shadowing Letter; Connectics 11/15/01 Untitled Industry,” T. Bodenheimer NEJM Vol. 342 program of Warner-Lambert, also mentions Letter; Berlex 11/19/01 Untitled Letter; May 18, 2000 No. 20 1539-1544 at 1541. shadowing by Alza Corporation.) and “Th e Forest Laboratories 9/13/02 Untitled Letter; Th e problem of industry’s profi t motive Drug Pushers,” C. Elliott, Th e Atlantic Janssen Research Foundation 11/20/02 distorting scientifi c research is not limited Monthly, Vol.Vol. 297, No.No. 3, pp.pp. 82-93. Untitled Letter; Aventis 12/15/02 Untitled to prescription drugs. A stark example is 35 Letter; Pfi zer, Inc. 10/24/03 Untitled tobacco industry research; see ““ResearchResearch “Suit Says Company Promoted Drug in Letter; Organon, Inc. 12/12/03 Untitled from tobacco industry affi liated authors: Exam Rooms” M. Petersen. New York Times Letter; XCel 12/19/03 Untitled Letter; need for particular vigilance,” S. Chapman New York, N.Y.: May 15, 2002. p. C.1 Wyeth-Ayerst Laboratories 3/18/04 Untitled Tob. Control 2005;14;217-219. 36 Letter; Novartis 6/27/03 Untitled Letter; 55 “Medical Group Votes to Bar Drug Alcon 7/18/03 Untitled Letter; Novartis “Medical Journals Are an Extension Salespeople from Witnessing Patient 8/22/01 Untitled Letter; Janssen 9/3/04 of the Marketing Arm of Pharmaceutical Exams” B. Japsen. Knight Ridder Tribune Warning Letter; Janssen Pharmaceutica Companies” R. Smith PLoS Med 2(5): Business News Washington: Jun 18, 2003. 9/04/04 Warning Letter; GlaxoSmithKline e138, published May 17, 2005 (available at p. 1 1/31/05 Warning Letter; Centocor, Inc. http://medicine.plosjournals.org/perlserv/ 37 2/11/05 Untitled Letter; AstraZeneca 3/8/05 ?request=get-document&doi=10.1371/ “Doctoring sales” E. Strout. Sales and Untitled Letter; Boehringer 3/22/05 Untitled journal.pmed.0020138) Marketing Management NewNew YYork:ork: MaMayy Letter; United Th erapeutics Corp. 4/13/05 56 2001. Vol. 153, Iss. 5, p. 52-60. Warning Letter; Alcon Laboratories 4/27/05 Ibid. at p. 0365. 38 Warning Letter; Endo 6/28/05 Warning 57 “Th e Drug Pushers,” C. Elliott, Th e Letter; Abbott Laboratories 7/15/05 Untitled Id. and at FN 15 and 16. Atlantic Monthly, Vol.Vol. 297, No.No. 3, pp.pp. 82-93 Letter; Pfi zer, Inc. 7/20/05 Warning Letter; 58 at p. 90. SuperGen, Inc. 8/18/05 Warning Letter; “Critics Say Drug Trials Often A 39 Allergan 9/6/05 Warning Letter; Alcon Marketing Tool” A. Dembner, Boston Ibid. 9/29/05 Untitled Letter Globe Boston, Mass.: Jun 25, 2002. p. A.1 40 48 59 Ibid. at p.p. 92 “Whistle-Blower Says Marketers Broke “Undue infl uence” J. Washburn. Th e 41 the Rules To Push a Drug” M. Petersen. American Prospect Princeton:AugPrinceton:Aug 13, 2001. “Th e accuracy of drug information from New York Times. New York, N.Y.: Mar 14, Vol. 12, Iss. 14, p. 16-22. pharmaceutical sales representatives” M. G. 2002. p. C.1 60 Ziegler, P. Lew and B. C. Singer JAMA Vol. 49 Ibid. 273 No. 16, Apr. 26, 1995 “Madison Ave. Has Growing Role 61 42 In the Business of Drug Research” M. Id. Ibid. Petersen. New York Times. New York, N.Y.: 62 43 Nov 22, 2002. p. A.1 “Medical Journals Are an Extension “Vioxx Doctors Wooed by Merck Are 50 of the Marketing Arm of Pharmaceutical Now Its Foes” H. Won Tesoriero Wall Street “Uneasy Alliance — Clinical Companies” R. Smith PLoS Med 2(5): Journal, Mar. 10, 2006 at B1. Investigators and the Pharmaceutical e138, published May 17, 2005 (available at Industry,” T. Bodenheimer NEJM Vol. http://medicine.plosjournals.org/perlserv/ 44 No. 20 1539-1544 at “Scientist Presses Claim P&G 342 May 18, 2000 ?request=get-document&doi=10.1371/ 1542. See Also, “ Misrepresented Actonel Data,” S. Ellison, Developing an open journal.pmed.0020138) Wall Street Journal, Feb. 23, 2006 at B6. relationship with the drug industry” P. 63 45 Vallance Th e Lancet London:London: SSepep 24-S24-Sepep See, “Next steps in trial registration,” K. “Amid AlarmAlarm BelBells,ls, A Blood SSubstituteubstitute 30, 2005. Vol. 366, Iss. 9491, p. 1062-1064. Abbasi and F. Godlee BMJ 2005;330;1222- Keeps Pumping” T. M. Burton, Wall Street (discusses ghostwriting at FN 3 and 19 and 1223 (at FNs 1-6 and accompanying Journal, Feb. 22, 2006 at A.1. the accompanying text.) text) and “Registries and Registration of 46 51 Clinical Trials” C. Haug, P. Gøtzsche, and T. “Uneasy Alliance — Clinical “Madison Ave. Has Growing Role V. Schroeder, N Engl J Med 353;26 2811-13 Investigators and the Pharmaceutical In the Business of Drug Research” M. (at FNs 1-7 and accompanying text). Industry,” T. Bodenheimer NEJM Vol. 342 Petersen. New York Times. New York: Nov 64 May 18, 2000 No. 20 1539-1544 at 1541-42. 22, 2002. p. A.1 “Uniform Requirements for Manuscripts 47 52 Submitted to Biomedical Journals: Aventis 3/8/01 Untitled Letter; Bristol “Drug Maker Withheld Paxil Study Writing and Editing for Biomedical Meyers Squibb 3/13/01 Untitled Letter; Data” Primetime Live ABC News December Publication Updated February 2006” Bristol Meyers Squibb 3/29/01 Untitled 9, 2004, available at http://abcnews.go.com/ International Committee of Medical Journal Letter; Pfi zer, Inc. 4/1/01 Untitled Letter; Health/story?id=311956&page=1 at page 4, Editors at III.J.III.J. available at htthttp://www.p://www. Sanofi -Synthelabo 5/9/01 Untitled Letter; viewed March 31, 2006. icmje.org/ (requirement of registration) M A Y 2 0 0 6 THE STATE PIRGS 32

71 “Trial registration at ClinicalTrials.gov While all the letters reviewed for this L. Laine, R. Burgos-Vargas, B. Davis, R. Day, between May and October 2005.” Zarin report were from a single site, some of them M. Bosi Ferraz, C. J. Hawkey, M. C. Hochberg, DA, Tse T, Ide NC. N Engl J Med Dec. 29, referred to letters that were not posted. See T. K. Kvien, T. J. Schnitzer, A. Weaver N 2005;353:2779-87. (evidence of eff ectiveness supra at FN 12 and Table 5. Engl J Med 2006 (March 16 edition; posted of registration requirement for publication 72 early at http://content.nejm.org/cgi/reprint/ on registration rates; also evidence for “Struggling Merck Names New Boss,” NEJMc066096.pdf (viewed(viewed FFeb.eb. 24, 2006)) clinical insignifi cance of many compliant BBC World News May 5, 2005, at http:// at p. 1 registrations.) news.bbc.co.uk/2/hi/business/4517351.stm 87 65 (viewed February 16, 2005) VIoxx Warning Letter “Trial registration at ClinicalTrials.gov 73 88 between May and October 2005.” Zarin “Testimony Before the Senate Finance See, e.g., “Th e $20bn drugs disaster” T. DA, Tse T, Ide NC. N Engl J Med Dec. 29, Committee” Dr. David Graham, Nov. 18, Boles. Knight Ridder Tribune Business 2005;353:2779-87; “Trial Registration 2004. News Washington:Aug 28, 2005. p. 1. Report Card.” J. M. Drazen, and A. J.J. Wood 74 89 N Engl J Med Dec. 29, 2005;353:2809-11. Estimate by cardiologist Eric Topol, cited See “Warning Signs: E-Mails Suggest 66 by Dr. Graham in his testimony. Ibid. Merck Knew Vioxx’s Dangers at Early “Trial Registration Report Card.” J. 75 Stage; As Heart-Risk Evidence Rose, M. Drazen, and A. J.J. Wood N Engl J Med FDA Warning Letter to Merck, 9/17/01 Offi cials Played Hardball; Internal Dec. 29, 2005;353:2809-11. See Also, “Th e at p. 6. (Herein after “Vioxx Warning Message: ‘Dodge!’; Company Says ‘Out of Ottawa Statement, Part One: Principles Letter) Context’” A. W. Mathews and B. Martinez. for international registration of protocol 76 Wall Street Journal. New York, N.Y.: Nov 1, information and results from human trials “New Treatment: To Sell Th eir Drugs, 2004. pg. A.1 (study was to test effi cacy of of health-related interventions” Edited Companies Increasingly Rely on Doctors; preventing colon polyps.) “Th e $20bn drugs by K. KrKrleža-Jerić,leža-Jerić, A. Chan, K. DickeDickersin,rsin, For $750 and Up, Physicians Tell Peers disaster” T. Boles. Knight Ridder Tribune I. Sim, J. Grimshaw, and C. Gluud, for the About Products” S. Hensley and B. Martinez Business News Washington:Aug 28, 2005. Ottawa group, at p.10, E.5 available at www. Wall Street Journal. New York, N.Y.: Jul 15, p. 1. (Treatment of colon cancer.) ottawagroup.ohri.ca 2005. pg. A.1 90 67 77 Vioxx Warning Letter at p. 1 “Th e Ottawa Statement, Part One: Ibid. 91 Principles for international registration 78 Vioxx Warning Letter of protocol information and results Id. 92 from human trials of health-related 79 Vioxx Warning Letter at p. 4 interventions” Edited by K. KrKrleža-Jerić,leža-Jerić, Vioxx Warning Letter at pp. 3-4 93 A. Chan, K. Dickersin, I. Sim, J. Grimshaw, 80 “At VVioxxioxx TTrial,rial, a DiscrDiscrepancyepancy Appears and C. Gluud, for the Ottawa group, at p.10, Vioxx Warning Letter at pp. 4-5 to Undercut Merck’s Defense” A. Berenson E.5 (statement registration must be an 81 New York Times Jul. 20, 2005. enforceable legal requirement) available Vioxx Warning Letter at p. 5 94 at www.ottawagroup.ohri.ca and “Trial 82 “Expression of Concern: Bombardier et Registration Report Card.” J. M. Drazen, Vioxx Warning Letter at p. 6. al., “Comparison of Upper Gastrointestinal and A. J.J. Wood N Engl J Med Dec. 29, Importantly, Vioxx was eventually proven Toxicity of Rofecoxib and Naproxen in 2005;353:2809-11, at 2810. to be eff ective for rheumatoid arthritis and Patients with Rheumatoid Arthritis,”” G. 68 such use was added to its label. However, D. Curfman, S. Morrissey, and J. M. Drazen Th is component is necessary because the other uses were never demonstrated. N Engl J Med 2000;343:1520-8. drug marketers have the right to negotiate 83 95 the language of the package insert with Vioxx Warning Letter at p. 7. “Response to Expression of Concern the FDA, and they exercise that right to 84 Regarding VIGOR Study,” A. Reicin, D. minimize any negative commercial impact “Warning Signs: E-Mails Suggest Shapiro N Engl J Med 2006 (March 16 of the label and maximize any positive Merck Knew Vioxx’s Dangers at Early edition; posted early at http://content.nejm. commercial impact. Th e FDA has asked Stage; As Heart-Risk Evidence Rose, org/cgi/reprint/NEJMc066096.pdf (viewed(viewed Congress for the authority to dictate drugs’ Offi cials Played Hardball; Internal Feb. 24, 2006)). labeling but to date has not received it. See Message: ‘Dodge!’; Company Says ‘Out of 96 FN 19 above. Context’” A. W. Mathews and B. Martinez. Ibid. 69 Wall Street Journal. New York, N.Y.: Nov 1, 97 “Negative Advertising As Drug Bill 2004. pg. A.1. “Expression of Concern Reaffi rmed” G. Soars, Some Doctors Get an ‘Unsales’ 85 D. Curfman, S. Morrissey, and J. M. Drazen Pitch” S. Hensley Wall Street Journal New Ibid. Dr. David Graham, an FDA N Engl J Med 2006 (March 16 edition; York NY: Mar. 13, 2006 p. A1. offi cial, testifi ed that pre-approval Merck posted early at http://content.nejm.org/cgi/ 70 Inc. did a study named 090 that showed a reprint/NEJMc066096.pdf (viewed(viewed FFeb.eb. 24, “Testimony Before the Senate Finance 7-fold risk of heart attacks. “Testimony 2006)). Committee” Dr. David Graham, Nov. Before the Senate Finance Committee” Dr. 98 18, 2004; “Reform of Drug Regulation David Graham, Nov. 18, 2004. GAO OxyContin Report at p. 35. — Beyond an Independent Drug-Safety 86 99 Board,” W. A. Ray and C. M. Stein, N. Engl. “Response to Expression of Concern GAO OxyContin Report at p.p. 8 J Med. 354;2 at 194. Regarding VIGOR Study,” C. Bombardier, 33 TURNING MEDICINE INTO SNAKEOIL

100 112 Ibid. at pp.pp. 9-10. Th ee prproblemoblem araroseose Id. at p.p. 20 ((ThTh ee numbersnumbers incincludelude 300300 a blockbuster” Business Week Online at fi rst and primarily in rural areas of Maine, sales representatives provided by Abbott http://www.businessweek.com/magazine/ Kentucky, Ohio, Pennsylvania, Virginia, and pursuant to a co-promotion agreement. content/06_10/b3974082.htm (visited Mar. West Virginia. Th e report notes that in 2000, 671 Purdue 1, 2006) (discusses Wall Street expectations 101 reps were calling on 70-94,000 doctors; and need for Merck to develop a new Th e fi rst news reference to the phrase no estimate is given for the 300 Abbott blockbuster) or “AntibioticAntibiotic wworkork affl icted found in Lexis-Nexis was “Th is painkiller representatives at the time. “Some 100,000” by the bottom line blues” L. Beil Th e Dallas can kill OxyContin abuse makes attacks on is our conservative estimate of how many Morning News, Mar. 1, 2006 (noting impact pot seem laughable OxyContin Mountain doctors were being called on after the of Wall Street expectations on what types of curse or miracle drug ?” G. Stone Charleston Purdue sales force expanded to 766 and drugs get researched). Gazette (West Virginia) at p. 1A. Abbott’s 300 reps are considered.) 129 102 113 “Paxil is Forever,” B. Hawkins CityCity Beginning almost immediately after See, “Doctoring sales” E. Strout. Sales Pages, Vol. 23, Iss. 1141 Oct. 16, 2002 OxyContin was approved, in 1996, the DEA and Marketing Management NewNew YYork:ork: 130 began addressing abuse problems. However, May 2001. Vol. 153, Iss. 5, p. 52-60, and “Infl uence of Patients’ Requests the DEA’s eff orts were signifi cantly scaled “Th e Drug Pushers,” C. Elliott, Th e Atlantic for Direct-to-Consumer Advertised up in 2001, when it developed an “Action Monthly, Vol.Vol. 297, No.No. 3, pp.pp. 82-93 at Antidepressants: A Randomized Plan” to deter OxyContin abuse and joined pp.88-89. Controlled Trial”, R. L. Kravitz; R. M. national education eff orts. Th e DEA’s 114 Epstein; M. D. Feldman; C. E. Franz; R. eff orts continued at a high level from then “No Margin for Error” M. Tenaglia, Azari; M. S. Wilkes; L. Hinton; P. Franks. on. GAO OxyContin Report at pp.pp. 36-38. P. Angelastro. Pharmaceutical Executive JAMA. 2005;293:1995-2002. (JAMA Paxil 103 Eugene:Sep 2005. Vol. 25, Iss. 9, p. 90- Study). Ibid. at 14. 92,94,96,98; “Pharma Marketing News 131 104 Special Supplement, Increase Physician JAMA Paxil Study at p.p. 2000. Originally OxyContin was marketed Access and Eff ectiveness” published by 132 as “Th e One to Start with and Stay with,” a VeriSci Newton, PA 2005. (same). Ibid. at p.p. 1998. slogan promoting using OxyContin as a pre- 115 133 emptive replacement for weaker and lower GAO OxyContin Report at 34. Id. at p.p. 2000 dose opioids on the market. Id. at 17. 116 134 105 Ibid, at 21. “Glaxo Is Rebuked By FDA Over Ad Id. at 28. PPurdueurdue distrdistributedibuted a rrevisedevised 117 For Paxil CR Drug” J. Whalen Wall Street version of this video in 2000, which the Id. at 25. Journal New York, N.Y.:Jun 14, 2004. p. B.5 FDA said also appeared to minimize 118 135 OxyContin’s risks. Purdue distributed Id. “AntidepressantAntidepressant Risk fforor NNewbornsewborns” 12,000 copies of another video that, 119 J. C. Dooren. Wall Street Journal New although submitted to the FDA, was Id. at p.p. 30. York, N.Y.: Feb 9, 2006. p. D.3 (increased not reviewed by it. Th ese videos became 120 risk of infants with persistent pulmonary obsolete with the 2001 labeling change but New label available at http://www.fda. hypertension when drug is taken in second were not recalled or replaced. Id. gov/cder/foi/label/2001/20553s022lbl.htm half of pregnancy); “U.S. Food & Drug 106 121 Administration; Agency advises of risk of Id. at 27-28. PPurdueurdue distrdistributedibuted 12,000 GAO OxyContin Report at 36. birth defects with Paxil” Women’s Health copies of another video that, although 122 Law Weekly Atlanta:Atlanta: Jan 15, 2006. pp.. 4 submitted to the FDA, was not reviewed by FDA Talk Paper, 7/25/01, available (increased risk of heart defects in infants it. Th ese videos became obsolete with the at http://www.fda.gov/bbs/topics/ when mothers take Paxil in fi rst trimester) 2001 labeling change but were not recalled ANSWERS/2001/ANS01091.html 136 or replaced. Id. 123 “Suicide Warning Ordered On Drugs; 107 GAO OxyContin Report TTableable 2 at Antidepressant Risk Seen In Youth” C. Y. Id. at 27. p. 31. Johnson, Boston Globe Boston, Mass.: Oct 108 124 16, 2004. p. A.1 Id. Ibid. at 26. 137 109 125 “Trouble In Prozac Nation,” D. Stipp, J. Doctors’ aversion is so commonplace Id. at p.p. 30. L. Yang Fortune New York: Nov 28, 2005. as to have a name, “morphine stigma.” 126 Vol. 152, Iss. 11, p. 154-169. See, “What’s the Fuss Over OxyContin “Paxil is Forever,” B. Hawkins CityCity 138 and Other Long-Acting Opioids?” Pages, Vol. 23, Iss. 1141 Oct. 16, 2002 “Drug Maker Withheld Paxil Study L. Broadman, J. P. Rathmell American 127 Data” Primetime Live ABC News December Society of Anesthesiologists Newsletter, “Health: First, you market the disease 9, 2004, available at http://abcnews.go.com/ Nov., 2001 at http://www.asahq.org/ then you push the pills to treat it” B. I Health/story?id=311956&page=1 at page 3, Newsletters/2001/11_01/broadman.htm Koerner. Th e Guardian Manchester (UK):Jul viewed March 31, 2006. 110 30, 2002. p. 2-8 139 GAO OxyContin Report at p.p. 17 128 Ibid. Business plan guide available at 111 See, e.g., “Merck: Out Of Th e Ivory http://abcnews.go.com/images/Primetime/ Ibid at 17-18. Tower: Its pragmatic strategy includes paxil_bpg.pdf (minimize(minimize at pp.. 2 and 2 in tweaking the vitamin niacin to make it 1,000 at p. 3) Viewed March 31, 2006. M A Y 2 0 0 6 THE STATE PIRGS 34

160 SmithKline Beecham, GlaxoSmithKline’s at http://www.msnbc.msn.com/id/5120989/ http://www.accessdata.fda. predecessor, appears to have been somewhat , viewed March 31, 2006.) gov/scripts/cder/drugsatfda/index. fl ip about the issue. In a May 1997 memo 149 cfm?fuseaction=Search.Label_ApprovalHis to the “Paxil Sales Team”, the company “Glaxo Settles New York Suit Over tory#apphist explained how to respond to concerns about Unpublished Clinical Data; Drug Maker 161 Paxil and withdrawal and concluded: “Let’s Agrees to Make A Payment of $2.5 “FDA tightens Accutane rules,” USA B. face it in the end. Th e only thing an anxious Million, Create Registry of All Trials” Today, AugustAugust 12, 2005. Martinez Wall Street Journal and agitated patient [anxiety and agitation New York, 162 are common Paxil withdrawal symptoms] N.Y.: Aug 27, 2004. p. B.3 http://www.house.gov/stupak/accutane_ will be saying is: Where’s My Paxil!!!!!!” See 150 chronology.shtml (U.S. Rep. Stupak’s site http://abcnews.go.com/images/Primetime/ “Leading the News: Ad Agency Helped giving an Accutane chronology.) paxil_wheresmy.pdf at p.2p.2 Push Neurontin, Documents Show” R. 163 140 Zimmerman Wall Street Journal New York, “Special report: Drugmaker Ibid. N.Y.: Nov 8, 2002. p. A.3 (campaign plan rebuff ed call to monitor users; Lawsuits 141 was designed by Cline Davis & Mann Inc., say Accutane caused birth defects, “Drug Maker Withheld Paxil Study the Manhattan advertising fi rm. depression and deaths” K. McCoy. USA Data” Primetime Live ABC News December 151 TODAY McLean, Va.:Dec 7, 2004. p. A.1 9, 2004, available at http://abcnews.go.com/ “Whistle-Blower Says Marketers 164 Health/story?id=311956&page=1 at page 3, Broke the Rules To Push a Drug” M. FDA Warning Letter to Hoff man- viewed March 31, 2006. Petersen. New York Times. New York, N.Y.: LaRoche, released 3/5/98, available at http:// 142 Mar 14, 2002. p. C.1 www.fda.gov/cder/warn/warn1998.htm http://abcnews.go.com/images/ 152 (description of marketing material); “Roche Primetime/paxil_moneybag.pdf viewedviewed “Parke-Davis Hired Firms to Unit’s Ads On Acne Medicine Draw FDA March 31, 2006. Th e font size and bolding Strategize Drug Promotions for Fire” R. Sharpe. Wall Street Journal. New approximates the original. Unapproved Uses” L. Kowalczyk Knight York, N.Y.: Mar 18, 1998. pg. 1 (campaign 143 Ridder Tribune Business News Washington: ran for over a year) “Paxil Memo Raises Questions” E. Nov 8, 2002. p. 1 165 Silverman Th e Star Ledger, Newark,Newark, NJ FFeb.eb. 153 “Side eff ects prompt stronger 18, 2004 at p. 21. “Whistle-Blower Says Marketers warning on acne drug” Anonymous. FDA 144 Broke the Rules To Push a Drug” M. Consumer. Rockville: May/Jun 1998. October, 1998 company memo, “ http:// Petersen. New York Times. New York, N.Y.: Vol.32, Iss. 3; pg. 3. www.ahrp.org/risks/SSRI0204/GSKpaxil/ Mar 14, 2002. p. C.1 166 pg1.html at p. 5. 154 FDA Warning Letter to Hoff man- 145 Ibid. LaRoche, released 3/5/98, available at http:// “Spitzer Sues a Drug Maker, Saying It 155 www.fda.gov/cder/warn/warn1998.htm Hid Negative Data” G. Harris. New York “Pfi zer to Settle Medicaid-Fraud 167 Times New York, N.Y.: Jun 3, 2004. p. A.1 Case; Drug Maker Agrees to Pay About “Roche Unit’s Ads On Acne Medicine 146 $430 Million in Fines Over Neurontin Draw FDA Fire” R. Sharpe. Wall Street Th e memo is available at http:// Marketing,” D. Armstrong and R. Journal New York, N.Y.: Mar 18, 1998. pg. 1 abcnews.go.com/images/Primetime/paxil_ Zimmerman Wall Street Journal New York: 168 remarkable.pdf, viewedviewed MarMarchch 31, 2006. May 13, 2004. p. A.3 http://www.accessdata.fda. 147 156 gov/scripts/cder/drugsatfda/index. “Spitzer Sues a Drug Maker, Saying Ibid. cfm?fuseaction=Search.Label_ It Hid Negative Data” G. Harris. New 157 ApprovalHistory#apphist (FDA(FDA historhistoryy of York Times New York, N.Y.: Jun 3, 2004. “Pfi zer pays $430m to settle off label labeling changes for Accutane); http://www. p. A.1 and “Sales Reps Told Not to Divulge drug case” D. Teather. Th e Guardian house.gov/stupak/accutane_chronology. Paxil Data, Drug Maker Memo Cited Manchester (UK):May 14, 2004. p. 25 shtml (U.S. Rep. Stupak’s site giving an Risks to Youths”, E. Silverman, Th e Star 158 Accutane chronology.) Ledger, Newark,Newark, NJ SSeptembereptember 29, 2004 http://www.house.gov/stupak/accutane_ 169 at p. 1. chronology.shtml (U.S. Rep. Stupak’s site FDA 7/25/02 Letter to Hoff man-LaR 148 giving an Accutane chronology.) “Spitzer Sues GlaxoSmithKline Over 159 Paxil” Associated Press Jun. 2, 2004 (available Ibid.