Skip to Content
No Fees Unless We Win 855-948-5098

A Cure Worse Than the Disease


Antidepressant Black Box Warning

TRIAL | Drugs & Devices | March 2005 / Volume 41, Issue 3

Recent FDA actions and newly disclosed research on the link between antidepressants and suicide-especially among young people-are strengthening plaintiffs’ claims against drug makers.

A 13-year-old boy hangs himself after taking Paxil for nine days. A 19-year-old man on Zoloft for 11 days throws himself off a cliff and dies. A 13-year-old boy on Prozac stabs his aunt to death. These are but a few cases of violent acts by children and adults taking drugs in a class of antidepressants called selective serotonin reuptake inhibitors (SSRIs).

On October 15, 2004, the FDA announced in a public health advisory that it had directed SSRI manufacturers to include a black-box warning on their products’ labels to alert health care providers of an increased risk of suicide in children and adolescents taking the drugs.1 The agency also will require that manufacturers provide a “patient medication guide” with each prescription to advise users of the risk.2 Significantly, the FDA stated: “A causal role for antidepressants in inducing suicidality has been established in pediatric patients.”3

On December 6, 2004, the British Medicines and Healthcare Products Regulatory Agency (MHRA) warned that although in adults the benefits appear to outweigh the risks, adults-particularly young adults-also should be closely monitored, especially during early treatment.

This was the culmination of mounting public awareness of the risk of antidepressant-induced suicide, beginning in the United Kingdom with an explosive expos³ by the British Broadcasting Corp. (BBC) in October 2002.4 The program told the story of the hidden risks of … suicide related to Paxil. The broadcaster received unprecedented response: over 65,000 telephone calls and 1,500 e-mails, largely from SSRI users who had suffered adverse reactions to the drugs.5

This public outcry prompted the MHRA to review the Paxil clinical trial database. On June 10, 2003, the agency announced:

New data from clinical trials in children and adolescents . . . do not demonstrate efficacy in depressive illness . . . and show an increase in the risk of harmful outcomes, including episodes of self-harm and potentially suicidal behavior in the [Paxil] group compared to placebo. Various analyses suggest that the risk of these outcomes is between 1.5 and 3.2 times greater with [Paxil] compared to placebo.6

On December 10, 2003, the MHRA came to the same conclusion regarding other similar antidepressants, including Celexa, Effexor, Lexapro, Luvox, and Zoloft.7

Here at home, the FDA dawdled. The agency finally held a public advisory committee hearing on February 2, 2004, to discuss reports of suicide in pediatric clinical trials of antidepressants and whether these events could be attributed to the drugs.

Andrew Mosholder, the senior FDA epidemiologist who reviewed the data, reportedly came to the same conclusion as the British regulators: that the use of SSRIs generally should be “discouraged” in children and adolescents due to a significant risk of suicidal behavior.8 However, on February 1, the day before the hearing, the media reported an FDA insider’s claim that senior agency officials were trying to suppress Mosholder’s findings.9

Mosholder was not permitted to present his conclusions at the hearing. But after listening to drug safety advocates and dozens of families with tragic stories to tell, the advisory committee urged the FDA to issue warnings about the suicide risk.10 Agency officials instead pushed to have a group at Columbia University reanalyze the data, which ultimately led to an eight-month delay before the FDA acknowledged the risk.

In the meantime, on March 22, 2004, the FDA required manufacturers to provide stronger warnings about the risk of suicide for both kids and adults.11

In September, the advisory committee met again to discuss the results of the Columbia University review, which largely confirmed the findings of Mosholder and the British MHRA.12 At this hearing’s end, the committee recommended that the agency require a black-box warning for all antidepressants.

The circumstances surrounding the FDA’s alleged suppression of Mosholder’s report led to two congressional investigations: one by Sen. Chuck Grassley (R-Iowa)13 and the other by the House Committee on Energy and Commerce.14 Rep. Joe Barton (R-Tex.), chair of that committee, held two public hearings last September, during which lawmakers interrogated and criticized antidepressant manufacturers and FDA officials for failing to protect the public health concerning the suicide risk.15 Additional hearings are expected this year.

In congressional testimony related to Vioxx, an FDA epidemiologist, David Graham, said that “as currently configured, [the FDA] is incapable of protecting America against another Vioxx. We are virtually defenseless.”16

Unreliable Literature

The antidepressant controversy has highlighted the unreliability of scientific literature about the drugs-on which manufacturers rely to promote and defend their products and to oppose plaintiffs’ claims that antidepressants cause suicidal and violent behavior. Emerging evidence, in particular from editors of medical journals that publish the research, has revealed that the pharmaceutical industry has manipulated the results of research it funds, so much so that some reports published in medical journals barely resemble the underlying studies on which they are based.

In addition, studies that show a drug is effective are three times more likely to be published than those showing a drug doesn’t work or does more harm than good.17 Negative studies simply do not reach the journals. The studies that are published often are ghostwritten by the drug companies, listing the names of scientists considered leaders in their fields as the authors. Some contain marketing messages designed to maximize sales.18

An example is an article about a highly publicized Zoloft study by Karen Dineen Wagner, a psychiatrist with the University of Texas Medical Branch in Galveston.19 The article, which touted Zoloft as “safe and effective” for treating pediatric depression, was published shortly after British regulators announced that Paxil would be contraindicated for use in children and while they were examining Zoloft clinical trials. Letters to the editor from a number of physicians criticized the article, saying that the actual data suggested no benefit from Zoloft (or barely more than placebo), and that Pfizer, the drug’s manufacturer, improperly influenced the presentation of the data and the interpretation of the results.20 Discovery in a pending Zoloft suicide case revealed that the author of the article was a Pfizer employee, not Wagner.21

Ghostwrittten and unpublished studies are an increasing concern in medicine and academia, and the antidepressant controversy provides a case in point. An analysis in The Lancet, which compared the published and unpublished results of pediatric antidepressant clinical trials, found that the published data showed largely positive results; however, when coupled with the unpublished data, the results showed the benefits did not outweigh the risks.22

The journal’s editors wrote a scathing editorial, saying, “The story of research into [SSRI] use in childhood depression is one of confusion, manipulation, and institutional failure.” They called these failings a “disaster” and suggested that “changes are required at every level of the global health-care infrastructure.”23

Another study, published in the British Medical Journal, reached similar conclusions regarding SSRI research and said, “A more critical approach to ensuring the validity of published data is needed.”24

The two congressional SSRI investigations are examining these issues. And manipulation of the literature was, in part, the subject of New York Attorney General Eliot Spitzer’s lawsuit charging GlaxoSmithKline with fraudulent promotion of Paxil.25

Long Road of Litigation

Although much has happened since 2002 to increase public awareness of the risks of antidepressants, they have been the subject of lawsuits for more than a decade. The litigation began in 1990 with cases involving Prozac, the first SSRI antidepressant. More than 150 Prozac-suicide cases were filed in federal courts in the early 1990s; most of these were consolidated in a multidistrict litigation.26

Only three SSRI cases have made it to trial. Most cases have been quietly “resolved.” An unknown number were dropped following the first Prozac verdict, in favor of the manufacturer. Other cases have fallen prey to summary judgment motions on issues such as actual causation, Daubert, and most recently, federal preemption.

The first to go to trial, Fentress v. Shea Communications, involved a man, Joseph Wesbecker, who, while on Prozac, went to his workplace and opened fire with an AK-47 assault rifle and other weapons. He killed 8 people and injured 12 more before turning a gun on himself. The jury returned a 9-3 verdict in favor of Prozac’s manufacturer, Eli Lilly.27

Lilly touted the verdict as the “death knell” of Prozac-suicide litigation and proclaimed the drug vindicated. However, months after the verdict, rumors began to spread that during the trial, Lilly had entered into a secret settlement (a high-low agreement) allegedly requiring plaintiff counsel to not present to the jury evidence that the company had previously faced criminal charges for failing to report to the FDA deaths related to its drug Oraflex.28

The trial judge found out about the secret agreement and was furious. He later complained to a BBC reporter:

After the verdict came in, Eli Lilly gave it a great deal of publicity and various people went on television and on the radio and in the newspapers proclaiming that this was a vindication of Prozac. . . . I think the public has a right to expect that a trial is a bona fide contest and not some sort of show that one side puts on with the consent of the other to influence public opinion. Because it was done to discourage other plaintiffs and to help settle the pending lawsuits for less money than they might have been settled for otherwise. Between these two parties, each side got what they wanted. But I think a bigger issue is whether the system was somehow corrupted a little bit, and I believe it was.29

The judge took the matter to the Kentucky Supreme Court, which found that “there was a serious lack of candor with the trial court and there may have been deception, bad-faith conduct, abuse of the judicial process, and perhaps even fraud.” The defense verdict was revoked, and the case was recorded as settled.30

The second case to go to trial was Forsyth v. Lilly, which involved a man who had been taking Prozac for 11 days when he stabbed his wife 15 times and then impaled himself on a butcher knife. His two adult children sued, and following a three-week trial in federal court and two days of deliberations, the jury returned a verdict for Lilly.31

In an appeal to the Ninth Circuit, the plaintiffs claimed the judge excluded crucial evidence during the trial, including the fact that in 1985 German regulators rejected Lilly’s application to market Prozac in that country because of significantly higher rates of suicidal behavior compared to placebo.32 The judge also excluded a report by FDA epidemiologist David Graham-the same expert who blew the whistle on Rezulin and most recently Vioxx-that was critical of Prozac and Lilly’s denial of a suicide link.

While that appeal was pending, the Forsyth children filed a derivative lawsuit against Lilly alleging fraud on the court, claiming that Lilly withheld evidence from discovery.33 Before the appeals court ruled, the parties settled the case.

In the last case to go to trial, a man who had been on Paxil killed his wife, daughter, infant granddaughter, and then himself.34 The daughter’s husband-father of the infant-sued, and the jury found for the plaintiff. GlaxoSmithKline appealed, and the parties later settled.

Common Defenses

Federal preemption

Although several cases have been filed against Pfizer for suicides allegedly induced by Zoloft, none has reached trial. Pfizer has used the doctrine of federal preemption to dismiss the cases, arguing that the FDA would not allow it to issue a suicide warning. In two recent cases in Texas, the company succeeded, at least in part, on these grounds, with the court granting summary judgment.35

Although case law, the case records, and recent events undercut the preemption argument, Pfizer used it successfully in the Texas lawsuits by procuring an FDA amicus brief originally submitted to the Ninth Circuit in a 2001 case, Motus v. Pfizer. The brief, written by then-FDA Chief Counsel Daniel Troy-a former attorney for Pfizer working at the firm Wiley, Rein & Felding-supported preemption, saying that “any warning that suggested a causal relationship between Zoloft and suicide would have been false or misleading and would have misbranded the drug in violation of federal law.”36

Pfizer had never asked the FDA to allow it to place such a warning on the label, nor had it submitted the evidence in its possession that would have justified the warning.37 However, the brief speculated that the FDA would not have allowed such a warning even if the company had sought one-despite the fact that the agency’s own regulations expressly provide that a drug company can add or strengthen a warning without prior FDA approval.38 Indeed, on August 22, 2003, Wyeth, on its own initiative, issued a warning about Effexor-induced suicidality for children and adolescents, using this very regulation as authority.

Pfizer has submitted Troy’s brief in virtually every Zoloft-suicide case filed since 2001. Other SSRI manufacturers have done the same, arguing that the Motus brief also applies to their drug. The FDA has not submitted new briefs in these cases. The preemption defense has been a significant hurdle over the past few years; however, the new FDA warnings should help considerably.

Actual Causation

Defendants in antidepressant cases invariably argue that the plaintiff cannot prove the drug actually caused the harm claimed. Careful case screening and preparation are key to challenging this defense.

In evaluating whether to accept a case, find out whether the decedent ever attempted suicide before he or she began taking the antidepressant. If so, you should usually reject the case, but not always: For example, the case may be viable if the person was taking an antidepressant at the time of the previous suicide attempt.

The timing of the suicide is also important. Causation is stronger if it occurred in the first weeks of taking the drug or after an increase in dosage.

Inquire about any noticeable adverse reactions to the drug or statements the decedent made about possible side effects. Find out whether the decedent ever experienced

  • symptoms of akathisia, a neurological phenomenon, usually drug-induced, that creates a state of extreme restlessness, often resulting in agitated physical movement such as pacing
  • a feeling of “ants” crawling on or under the skin
  • a feeling of wanting to jump out of his or her skin
  • agitation, hostility, or restlessness
  • a feeling of existing outside of reality, as if the person were watching television rather than being a participant in life
  • a feeling of being outside his or her body
  • manic or psychotic reactions.

Did the decedent ever make statements about how the drug was making him or her feel, such as “I’m losing my mind” or “the drug is making me crazy”?

Also determine whether the decedent was taking any other medication at the time of the suicide. If so, the defendant may blame the other drug for the person’s death or argue that it is not possible to discern the cause. If the decedent was taking another medication, you will need to research the prescribing information, including interactions listed on the second drug’s label, contraindications, precautions, warnings, and adverse reactions. You will also need to obtain an expert opinion on its significance to the case.

The drug company will do all it can to uncover evidence that could attribute the suicide to something other than its product. First it will blame the decedent’s underlying mental condition. It will also look for circumstances in his or her life that might have contributed to a vulnerable state of mind, such as divorce, a death in the family, or job difficulties.

The company will leave no stone unturned, so you must do the same. Obtain the decedent’s medical and pharmacy records, education and employment records, calendars, journals or diaries, e-mail exchanges, and other writings. Interview family members, friends, and anyone who observed the decedent while he or she was taking the drug.

No life is perfect. No doubt you will find circumstances in the decedent’s life that could have contributed to his or her vulnerable state, and the defense most certainly will argue that they did. You must weigh the decedent’s personal and medical history, the facts and circumstances leading up to the suicide, and the evidence of adverse reactions he or she experienced. This analysis will help you judge whether the case is strong enough to counter the actual-causation defense.

Daubert and Frye

You will also certainly face general causation defenses under Daubert v. Merrell Dow Pharmaceuticals, Inc.,39 or Frye v. United States.40 Because the FDA has affirmed that general causation has been established with respect to suicides among children and adolescents, companies may be more hesitant to file Daubert motions in these cases. However, unless and until the FDA comes to the same conclusion for adults, you can expect drug companies to continue to fight this battle in cases involving adults.

Increasing concerns about ghostwritten and unpublished studies may benefit plaintiffs facing a drug company’s argument that the published literature is the “gold standard,” on which all experts are bound to rely in reaching a general causation opinion.

Learned Intermediary

In a failure-to-warn case involving a pharmaceutical product, the learned intermediary doctrine “simply substitutes the physician for the consumer as the person to receive those warnings.”41 To defeat this defense, as in any pharmaceutical case, you must establish that the prescribing doctor did not have “independent knowledge” of the risk, that the risk was not “universally known,” and that it was not “unknowable” at the time of the injury or death. Carefully examine the substantive state law on this issue.42

Until recently, most physicians were completely unaware of the suicide risk associated with antidepressants. Because the drug companies have denied the risk-indeed, many still deny it-they cannot argue persuasively that doctors had independent knowledge of the risk. In addition, a physician’s negligence, if any, “generally is not an intervening cause exonerating a drug company, if the doctor’s act was reasonably foreseeable or if the inadequacy of the drug company’s warning may have contributed to the plaintiff’s injury.”43

The FDA’s recent mandate to place black-box warnings on SSRI labels about the risk of drug-induced violence and suicide in children was an important milestone that would not have been reached without the efforts of courageous victims, scientists with unshakable integrity, the media, lawmakers, and attorneys representing families who lost loved ones. This combined effort has finally forced the FDA to do what it should have done years ago.

The manufacturers, no doubt, will continue the cover-up, blaming the violence and suicides on anything but their drugs, despite mounting evidence to the contrary. Litigation continues to play its significant and appropriate role of advocating for those harmed by SSRIs, increasing public awareness, inspiring social change, and contributing to public safety.



3. FDA, LABELING CHANGE REQUEST LETTER FOR ANTIDEPRESSANT MEDICATIONS (emphasis added) (undated), available at (last visited Jan. 26, 2005).

4. Panorama Program: The Secrets of Seroxat, (BBC television broadcast, Oct. 13, 2002), transcript available at (last visited Jan. 26, 2005).

5. Charles Medawar et al., Paroxetine, Panorama, and User Reporting of ADRs: Consumer Intelligence Matters in Clinical Practice and Post Marketing Drug Surveillance, 15 INT’L J. RISK & SAFETY IN MED. 161 (2002), available at (last visited Jan. 26, 2005).

6. Gordon Duff, Chairman, Comm. on Safety of Medicines, Message: Safety of Seroxat (Paroxetine) in Children and Adolescents Under 18 Years-Contraindication in the Treatment of Depressive Illness (June 10, 2003), available at (last visited Jan. 26, 2005).

7. Gordon Duff, Comm. on Safety of Medicines, Memorandum: Selective Serotonin Reuptake Inhibitors-Use in Children and Adolescents with Major Depressive Disorder (Dec. 10, 2003), available at (last visited Jan. 26, 2005).

8. Andrew D. Mosholder, FDA, Memorandum: Suicidality in Pediatric Clinical Trials with Paroxetine and Other Antidepressant Drugs: Follow- up to 9-4-03 Consult (Feb. 18, 2004), available at (last visited Jan. 26, 2005).

9. Rob Waters, FDA Under Fire over Barred Antidepressant Report, SAN FRANCISCO CHRON., Feb. 1, 2004, at A1.

10. FDA, PSYCHOPHARMACOLOGIC DRUGS ADVISORY COMM. MEETING (Feb. 2, 2004), transcript available at (last visited Jan. 26, 2005).


12. FDA, PSYCHOPHARMACOLOGIC DRUGS ADVISORY COMM. & PEDIATRIC ADVISORY COMM. MEETING (Sept. 13 & 14, 2004), transcript available at and (last visited Jan. 26, 2005).

13. Press Release, Sen. Comm. on Finance, Grassley Questions FDA’s Handling of Research on Antidepressants, Suicide (Mar. 25, 2004), available at (last visited Jan. 26, 2005).

14. Letter from the House Comm. on Energy and Commerce to Mark B. McClellan, Comm’r, FDA (Mar. 24, 2004), available at (last visited Jan. 26, 2005).

15. Publication and Disclosure Issues in Antidepressant Pediatric Clinical Trials: Hearing Before the Comm. on Energy and Commerce, 108th Cong. (Sept. 9, 2004), transcript available at (last visited Jan. 26, 2005); FDA’s Role in Protecting the Public Health: Examining FDA’s Review of Safety and Efficacy Concerns in Anti-depressant Use in Children: Hearing before the House Comm. on Energy and Commerce, 108th Cong. (Sept. 23, 2004), transcript available at (last visited Jan. 26, 2005).

16. FDA, Merck, and Vioxx: Putting Patient Safety First? Hearing Before Senate Comm. on Finance, 108th Cong., 4 (Nov. 18, 2004) (statement of David Graham, M.D., M.P.H.), available at (last visited Jan. 26, 2005).

17. Faye Flam, A Prescription for Full Disclosure, PHILADELPHIA INQUIRER, Aug. 15, 2004, at C01.

18. David Healy & Dinah Cattell, Interface Between Authorship, Industry, and Science in the Domain of Therapeutics, 183 BRIT. J. PSYCH. 22 (2003).

19. Karen Dineen Wagner et al., Efficacy of Sertraline in the Treatment of Children and Adolescents with Major Depressive Disorder, 290 JAMA 1033 (2003); see also Shari Roan, Zoloft Appears to Work for Children, with No Major Side Effects, LOS ANGELES TIMES, Sept. 1, 2003, at F3.

20. Matthews et al., Letters to the Editor, Efficacy of Sertraline in the Treatment of Children and Adolescents with Major Depressive Disorder, 291 JAMA 40 (2004).

21. Pfizer Inc.’s Response to Plaintiff’s Amended First Set of Interrogatories (Aug. 16, 2004) & Pfizer Inc.’s Response to Specially Prepared Interrogatory No. 4, Szybinski v. Pfizer, No. YC 047439 (Cal., Los Angeles County Super. Ct. filed Sept. 8, 2003).

22. Craig J. Whittington et al., Selective Serotonin Reuptake Inhibitors in Childhood Depression: Systematic Review of Published Versus Unpublished Data, 363 LANCET 1341 (2004).

23. Editorial, Depressing Research, 363 LANCET 1335 (2004).

24. Doecke Jureidini et al., Efficacy and Safety of Antidepressants for Children and Adolescents, 328 BRIT. MED. J. 879 (2004).

25. State v. GlaxoSmithKline, No. 905-03 (N.Y., Albany County Sup. Ct. filed June 2, 2004).

26. In re Eli Lilly & Co. Prozac Prod. Liab. Litig., MDL No. 907 (S.D. Ind. Dec. 10, 1991).

27. Fentress v. Shea Communications, No. 90-CI-6033 (Ky., Jefferson County Cir. Ct. Dec. 12, 1994); Michael Jonathan Grinfeld & Michael Welner, Pill Poisoned? The Seasoning of Medication Defenses, FORENSIC ECHO, Feb. 1, 1998, at (last visited Jan. 26, 2005).

28. Grinfeld & Welner, supra note 27.

29. File on 4 (BBC radio broadcast, Gerry Norham reporting, May 30, 2000) at 14-17 (transcript on file with author).

30. Potter v. Eli Lilly & Co., 926 S.W.2d 449 (Ky. 1996), abrogated by Hoskins v. Maricle, 150 S.W.3d 1 (Ky. 2004); Grinfeld & Welner, supra note 27.

31. 904 F. Supp. 1153 (D. Haw. 1995).

32. Appellant’s Brief, Forsyth v. Eli Lilly, No. 99-116821 (9th Cir. filed Mar. 31, 2000).

33. Forsyth v. Eli Lilly, No. 00-000401 ACK LEK (D. Haw. filed June 2, 2000).

34. Tobin v. SmithKline Beecham Pharm., No. 00-CV-0025-Bea (D. Wyo. June 6, 2001), motion for new trial denied, 164 F. Supp. 2d 1278 (D. Wyo. 2001).

35. Dusek v. Pfizer, No. H-02-3559, 2004 WL 2191804 (S.D. Tex. Feb. 20, 2004); Needleman v. Pfizer, No. 3:03-CV3074-N, 2004 WL 1773697 (N.D. Tex. Aug. 6, 2004).

36. Brief for Amicus Curiae the United States, Motus v. Pfizer, 358 F.3d 659 (9th Cir. 2004).

37. See 21 C.F.R. §201.57(e) (2003), which mandates changes in warnings “as soon as there is reasonable evidence of an association of a serious hazard with a drug” and states that “a causal relationship need not have been proved.”

38. See 21 C.F.R. §314.8(d) (2003).

39. 509 U.S. 579 (1993).

40. 293 F. 1013 (D.C. Cir. 1923).

41. 63 AM. JUR. 2D Products Liability §1200 (2002)(“Learned-Intermediary Doctrine; Products Provided by Physicians”); see also Davis v. Wyeth Lab., 399 F.2d 121, 130 (9th Cir. 1968).

42. See 63 AM. JUR. 2D Products Liability §1200, supra note 41; id. §1174 (1997) (“Causation Requirements in Warnings Cases; Physician Conduct”).

43. 63 AM. JUR. 2D Products Liability §1174, supra note 42 (emphasis added).



  • Please enter your first name.
  • Please enter your last name.
  • Please enter your phone number.
    This isn't a valid phone number.
  • Please enter your email address.
    This isn't a valid email address.
  • Please enter your city.
  • Please enter your state.
  • Please make a selection.
  • Please make a selection.
  • Please enter a message.