In my wildest dreams, I could never have imagined being drawn into a story of intrigue involving my own government’s efforts to hide, from the public, reports of psychiatric drugs associated with cases of murder, including homicides committed by youth on the drugs. But that is precisely the intrigue I now find myself enmeshed in.

The saga began several years ago. My child had the misfortune of being born during the last month of eligibility for kindergarten, and was subsequently labeled with A.D.H.D. – which stands for August Date Hikes Diagnosis. While other Americans with the same chronological impairment such as Man Ray and Robert Ringling managed to make something of themselves despite being born in the month of August, it seemed my child was doomed to failure from the get-go, unless provided lifesaving stimulant medication.

With an abiding uneasiness about both the alleged disorder and its miracle remedy, as they were presented to me, I set out to understand as much as I could about stimulant medications, prescribed disproportionately to the youngest children in the class.

It wasn’t long before I stumbled upon the FDA Adverse Event Reporting System (FAERS), also referred to as MedWatch. The FDA publishes quarterly FAERS data files on its website containing hundreds of thousands of reports of various drug adverse events. Though unencrypted, the FAERS files might as well be, as the data appear hieroglyphic to the average person who is not a database expert, including myself at the time.

With the aid of the internet and with a lot of trial and error, I taught myself to write Structured Query Language (SQL) code to decipher the FAERS files. I first plumbed the depths of the adverse event data searching for reports of pediatric fatalities associated with stimulant medications, and found hundreds of them. Expanding my queries to include all psychotropic medications, I eventually identified nearly 2,000 pediatric fatalities. (More on the pediatric psychotropic fatalities in a future post.)

As I contemplated the gravity and the scale of the human tragedy, I began to wonder what drug side effects these children experienced at the time of their deaths. The FAERS data files yielded answers: cardiac arrest, respiratory arrest, hepatoxicity, multi-organ failure, Stevens-Johnson Syndrome, Toxic Epidermal Necrolysis, Neuroleptic Malignant Syndrome, completed suicide, homicide . . . Wait, what?  Homicide, as a drug side effect? Then I saw it again: Murder.

The light bulb went off. I performed a query for homicide or murder as a drug side effect. To my astonishment, there were over 700 reports in FAERS of homicides linked to psychotropic medications.

In October 2014, I submitted a Freedom of Information Act (FOIA) request to the FDA to obtain copies of the FAERS homicide reports. After ten months of “Foot Dragging and Alibis,” which is what Rep. Joe Barton (R-TX) once suggested the agency’s acronym stood for, the FDA still had not produced a single report. Frustrated with the FDA’s stonewalling, I filed a federal lawsuit in August 2015. Within three months of filing the FOIA lawsuit, the FDA coughed up over 3,000 pages of FAERS reports.

However, hundreds of pages were completely redacted, while many hundreds more were heavily redacted. A letter from Deputy Director of Information Disclosure Policy Howard Philips attested that the records were redacted in accordance with the FOIA statute, and other applicable laws. The FDA claimed that the redaction of information was justified under FOIA’s privacy exemption.

To add some perspective, according to the HHS Freedom of Information Annual Report, among 10,145 FOIA requests the FDA processed during fiscal year 2015, the privacy exemption was applied only 24 times. Federal regulations require the FDA to “make the fullest possible disclosure of records to the public” in response to FOIA requests.

All but the case numbers were redacted in forty-seven of the FAERS homicide reports that the FDA released. The FDA had suppressed all of the report information for these cases: age, gender, drug name(s), reported drug reaction(s), case narrative, etc. The wholesale censorship of entire FAERS reports turned out to be an untenable action on the part of the FDA.

Pursuant to 21 CFR 20.81, the FDA cannot properly withhold any record that contains data or information that have been previously disclosed in a lawful manner to a member of the public. The age, gender, drug name(s), and reported drug reaction(s) had already been publicly disclosed in FAERS data files available on the FDA’s website. It was based on this publicly available information that I was able to ascertain and request the case reports involving homicide in the first place.

I fired off a letter to then-Acting Commissioner Stephen Ostroff, protesting the FDA’s improper withholding of public information, citing the federal regulation prohibiting such conduct. A month later, I received a second production of records totaling over 3,000 pages. The agency did not acknowledge any wrongdoing, or even explain what was different about the new document dump. This time around, though, the FDA had not redacted age, gender, drug name(s), and reported drug reaction(s), but the case narratives remained entirely redacted.

Insofar as the FDA had released the case narratives of hundreds of the other FAERS homicide reports, albeit heavily redacted at times, I surmised that the case narratives of these 47 reports in particular must contain information that was damaging either to the pharmaceutical companies, the FDA, or both.

Here are four examples of medication-linked homicide case narratives that were being withheld.

Case 10213469

The narrative in this case report, when it was initially sent to me, was completely redacted by the FDA. This is what I received:

However,  I knew from the FAERS data files that the case involved a 10-year-old taking Vyvanse (lisdexamfetamine), a stimulant drug prescribed to a million children in the U.S for ADHD. The girl had reportedly experienced a drug reaction that led her to commit homicide. The FDA would eventually send me a lesser-redacted version of the report, confirming the information in the FAERS data files, but the case narrative was still completely redacted.

Then, on April 12th of this year, the FDA presented a Vyvanse pediatric safety review to the Pediatric Advisory Committee. In advance of the meeting, the FDA’s “Pediatric Postmarketing Pharmacovigilance and Drug Utilization Review” was made public as part of the briefing materials posted on the agency’s website. The safety review contained a bombshell case narrative summary.

Homicide (n=1) 

Case # 10213468, USA, 2014: A 3-month-old female infant was left alone with a babysitter’s 10-year-old daughter. Lisdexamfetamine was prescribed to the 10-year-old daughter of the babysitter; the 10-year old girl had ADHD, ODD, and attachment disorder. The infant sustained various injuries. The autopsy reported the cause of death was “asphyxia and suffocation,” as the result of “homicide.” Additionally, the infant’s blood contained traces of amphetamine (lisdexamfetamine).

In an appendix, I noticed that one of  the FAERS’ reports that I had requested (10213469), which had been completely redacted, was only one digit off from the FAERS report quoted in the Vyvanse pediatric safety review, and was also listed as a duplicate report.  And then I put one and one together:  The FDA had represented to the Court handling my FOIA lawsuit that the case narrative of the FAERS homicide report I had requested—number 10213469—was exempt from disclosure under a FOI request for privacy reasons, yet now the FDA had publicly disclosed the case!

On April 13^th, the day after the Pediatric Advisory Committee met and considered the FDA’s Vyvanse pediatric safety and drug utilization review, I fired off an email to the Assistant U.S. Attorney on the case, copying the FDA’s Assistant Chief Counsel, demanding a lesser-redacted version of FAERS report 10213469, pursuant to 21 CFR 20.81, since the agency had publicly released a summary of the case narrative of a duplicate report.

I figured that the FDA would now have to cough up the report.

The next day, on April 14^th, Shire submitted a New Drug Application (NDA) for a chewable formulation of Vyvanse, as if following a script written long before the Pediatric Advisory Committee meeting. The company wrote in its press release that the Vyvanse chewable tablets are intended for patients “who may have difficulty swallowing or opening a capsule,” which is likely targeting very young children.

A day later, on April 15^th, the Assistant US attorney sent me an email, indicating that rather than provide me a lesser-redacted version of FAERS report 10213469, the FDA had instead decided to remove the case summary detailing the homicide from its drug safety review on its website.   Just like that, Uncle Sam had covered up the homicide of a 3-month-old infant girl, by an amphetamine-addled child, as if the baby had never existed. Now you see the homicide, now you don’t.

Apparently, we can’t have a story made public about a 10-year-old girl on Vyvanse who forced the ADHD drug down a baby girl’s throat before suffocating her to death. That would be bad for business. Especially even as Vyvanse chewable tablets are being approved for the market.

There is one more part to this story of homicide linked to ADHD drugs. Earlier, after I had filed my FOIA lawsuit concerning the FAERS homicide cases, the FDA approved Adzenys XR-ODT, the first orally disintegrating amphetamine tablet approved for kids with ADHD.  Without fanfare, a homicidal ideation warning was added to the label of Adzenys XR-ODT: “Anxiety, psychosis, hostility, aggression, suicidal or homicidal ideation have also been observed.”

Adzenys XR-ODT was approved as a bioequivalent of Adderall XR, another Shire-manufactured amphetamine drug that was formerly the most prescribed drug for ADHD prior to Vyvanse. Oddly, at this writing, Adderall XR does not have a homicidal ideation warning on its label, whereas its bioequivalent Adzenys XR-ODT does. I’ve emailed the FDA Division of Drug Information for an explanation of the inconsistent homicidal ideation label warnings for these bioequivalent drugs, and was told that a Subject Matter Expert (SME) had to be consulted before the agency could respond. Of similar interest, a homicidal ideation warning was added to the Vyvanse label as well.

Case 7979016

This case involves a 16-year-old male from Canada taking Prozac, who experienced the reported drug reaction of “homicide.”  The FDA initially released a completely redacted version of this report, claiming in effect that public disclosure of any information whatsoever would constitute an unwarranted invasion of personal privacy. Once I reminded the FDA of 21 CFR 20.81, the agency produced a lesser-redacted version. This time, the FDA did not redact age, gender, nationality, drug name, or drug reactions disclosed in the FAERS data files, yet persisted with the redaction of the entire case narrative.

As it so happens, I had requested FAERS 7979016 twice, so the FDA produced yet another version of the report. This time, much of the case narrative could be read, and it contained a bombshell:

The reporting psychiatrist assessed the homicide, self-injurious behavior, manic symptoms, and worsening of his condition as related to fluoxetine, it drove him over the edge and it contributed to his actions.

Under the pretext of a phony privacy claim, the FDA had, in its previous redactions,  deliberately kept hidden a psychiatrist’s damaging causality assessment linking a popular antidepressant to homicide.

Case 8464514

This case involves a 35-year-old female from Australia, who took the antidepressant nortriptyline and killed her daughter. In one version of the report, the FDA as I subsequently learned, had completely redacted the following narrative:

My husband was drinking. I took small doses of valerian for a month and had weird dreams and premonitions. When I took nortriptyline, I immediately wanted to kill myself, talked myself out of it. I’d never had thoughts like that before. My husband was angry, shouting. I walked outside a lot, with palpitations, trouble breathing, and became more depressed. My smoking went up to 25 a day, no alcohol. I didn’t sleep for two nights, dreamt, then slept maybe three hours, felt awful. I dreamt that my daughter had dark teeth and I saw a black halo around her head, a spear hanging over it. I felt like a zombie. I believed I had to help my daughter, that a bad spirit possessed her. I picked up a knife and stabbed her and woke up. I was not myself. I was looking on from the outside, controlled by dark forces. She said, “Mum, what are you doing here?” I realized what I’d done. I asked my husband to kill me. He called the police. I felt better in the police cells without the pills, but the pills started again, and thoughts of killing myself returned.

The FDA–and this is almost hard to believe–had redacted signs or symptoms of medication-induced suicidal ideation (“When I took nortriptyline, I immediately wanted to kill myself. I’d never had thoughts like that before” and “I asked my husband to kill me”); parasomnia or hallucinations (“I dreamt that my daughter had dark teeth and I saw a black halo around her head, a spear hanging over it”); delusions (“I believed I had to help my daughter, that a bad spirit possessed her); automatism: (“I felt like a zombie”); homicide, somnambulism, and parasomnia (“I picked up a knife and stabbed her and woke up”); dissociation (“I was not myself”); depersonalization (“I was looking on from the outside”); paranoia (“controlled by dark forces”); as well as positive dechallenge (“I felt better in the police cells without the pills”); and positive rechallenge – considered the gold standard with regard to causality (“but the pills started again and thoughts of killing myself returned.”)

The FDA then provided another version of the report to me, this time with bits and pieces of the above testimonial unredacted, yet with much of the passage still missing. However, this version contained another gem:

Ranbaxy medical reviewers comment: The case is deemed serious. Medical Reviewer considered the case to be possibly related to suspect drug due to its temporal association as per WHO UMC system for standardized causality assessment.

The FDA understood that there was likely a causal link to homicide. This was the finding that the FDA did not want to make public.

Case 6179785

This case report describes a 47-year-old male prescribed Prozac (fluoxetine), lithium, temazepam, and trazodone who committed homicide. The FDA redacted the type of place he entered, as well as whom or what he shot, but did reveal that the subject ultimately shot himself.  Only the report cited a BMJ article  entitled “FDA to review ‘missing’ drug company documents,” which contained the following passage:

The documents received by the BMJ reportedly went missing during the 1994 Wesbecker case that grew out of a lawsuit filed on behalf of victims of a work-place shooting in 1989. Joseph Wesbecker, armed with an AK-47, shot eight people dead and wounded another 12. He then shot and killed himself. Mr Wesbecker, who had a long history of depression, had been placed on fluoxetine one month before the shootings.

This is a well-known case, dating back to 1989, which has been the subject of a book, The Power to Harm, by John Cornwall. The survivors and relatives of the dead sued Eli Lilly, the manufacturer of Prozac. The jury ruled in favor of Eli Lilly, which–as this was the first such case to be tried in court over whether an SSRI could stir homicidal actions–proved to be a boon to the company. Its drug had been cleared, and Lilly’s stock price soared. However, as Cornwall later revealed, Eli Lilly had made a secret deal with the plaintiffs during the trial, paying them a huge sum of money to deliberately lose the case.

So here it is twenty-five years later, and the FDA, in its case report of this fluoxetine-related homicide, which was the subject of a book, redacted some of the pertinent information. And this leads to the obvious question: has the FDA attempted to hide, from the public, links between psychotropic medications and mass shootings? More on that subject in a future post.

Now the FDA Wants Some Case Reports Back

Besides antidepressants and homicide, the three preceding FAERS reports and many others like them share two additional commonalities: 1) The cases were widely publicized in news and scholarly publications; and 2) The FDA now wants the versions that spilled some of the narrative details back from me. Much like the FDA removed the evidence of a 3-month-old girl murdered by a ten-year-old on Vyvanse from its Vyvanse pediatric safety review, the FDA, in its efforts to get these documents back, apparently wants to conceal the details about other homicides linked to psychotropics.

It is doubtful that the legislators who passed the Freedom of Information Act intended for government redactors to be censoring media reports and scholarly publications.

Last week, consumer advocacy group Public Citizen also sued the FDA, alleging the agency has arbitrarily and capriciously redacted public information from the curricula vitae of advisory committee members, thus obscuring their ties to pharmaceutical companies. It seems as though the FDA views the redaction process as thwarting the intentions of the FOIA act, and keeping secret information that might damage commercial interests.

I’ll end with a prediction: more homicidal ideation warnings are coming to psychotropic drug labels. Pharmaceutical companies will need to protect themselves from failure-to-warn lawsuits, and the FDA will no longer stand in their way from doing so, like when they wouldn’t allow Wyeth to place a suicidal ideation warning on Effexor. Tellingly, a homicidal ideation warning was also added to the Effexor label in the premarketing evaluation adverse events section, i.e.,  the company received reports of homicidal ideation before the drug was even approved.

More reports linking psychotropic drugs to homicides can be found here.