No More Tears: The Dark Secrets of Johnson & Johnson

In the 1980s, scientists began to investigate how to synthesize a naturally-occurring protein called erythropoietin, which stimulates red blood cell growth. Amgen, a medical company, was doing trials on a synthetic form of erythropoietin called EPO, but it needed a partner to get FDA approvals. Amgen partnered with J&J. They theorized that EPO could be used to treat patients receiving dialysis, who tend to have low red blood cell counts. 

In 1988, the FDA approved EPO and J&J began marketing it as Eprex and Procrit. J&J began to market Procrit to cancer patients, who tend to have low red blood cell counts. However, EPO also carries a risk of “heart attacks, strokes, and embolisms” (124). Further, researchers associated with J&J had begun to realize that EPO could “supercharg[e]” cancer growth.

In order to avoid manufactured outrage that the FDA was unnecessarily holding back advancements in cancer treatment, the FDA quickly approved EPO for the market with a small clinical trial of 131 patients. However, in 1989, the year it was released, two healthy young professional cyclists died of heart attacks linked to EPO use. In 1993, J&J began to study the effects of EPO on cervical cancer patients, but they never released the results.

In the 1980s, drug companies got access to a computer directory that detailed exactly what prescriptions doctors were writing. They realized that doctors given cash and cash-equivalent incentives were much more likely to prescribe their medications. Drug companies entered into an “arms race of cash, freebies, and pseudo-jobs for doctors” (131) to increase drug sales. Further, J&J created a web of ghostwritten journals and corporate-funded conferences to give their sales pitches a veneer of scientific rigor.

Meanwhile, the FDA approvals process and oversight slowed due to lack of funding and resources. In 1992, the FDA commissioner, David Kessler, made a deal with the drug companies that they would fund the FDA. In 1997, the FDA agreed to reduce trial burdens on drug companies, reduce advertising restrictions, and reduce criminal oversight, leaving enforcement up to federal prosecutors.

In the 1989, the federal Medicare program implemented a “fee schedule” for doctors, reducing their take-home pay. At the same time, insurance companies began to more closely scrutinize doctors’ practices. This meant doctors were particularly eager for the perks offered by drug reps and drug companies that offered “part-time marketing gigs” (139) to supplement their income and practice.

In 1998, Procrit, the synthetic EPO, was “J&J’s most profitable product ever” (141) despite findings that EPO was not beneficial to dialysis patients. That year, J&J launched an advertising campaign to market Procrit to cancer patients. The FDA admonished J&J for its misleading claims, and the ad was temporarily pulled. In 2001, the FDA commissioner Daniel Troy reduced the FDA’s role in monitoring false advertising. In 2004, he left the role to become a lawyer for J&J.



In 2001, J&J began enrolling cervical cancer patients for a study on the impacts of EPO. However, they ended the study in 2003 because “patients given EPO were dying at a more rapid clip than those not given EPO” (143). They did not report the findings to the FDA until 2007 and never published the study.

In October 2003, the Lancet, a leading medical journal, published a study that found EPO “may be killing cancer patients” (145). The New York Times published a follow-up article that framed the study as an outlier, but the author of the study, Dr. Michael Henke, boldly stated he would no longer prescribe EPO for cancer-related anemia. His findings were reiterated soon after by Paul Goldberg, the editor of The Cancer Letter, who pointed to a study of Procrit that showed a higher fatality rate in breast cancer patients who took Procrit. 

Further, Goldberg asserted multiple clinical trials of Procrit in cancer patients had been cancelled due to risks. Generally, doctors did not report when their studies of EPO had poor findings, but Henke insisted on sharing the results, even though his German pharmaceutical funder hesitated about doing so. In response to Henke’s study, in 2004 the FDA held a public meeting of cancer experts in which J&J admitted to stopping five studies of patients getting EPO. J&J agreed to put a warning label on Procrit that stated it “might promote tumor growth and increase the risks of death” (150).

Due to the way that the Medicare program reimburses for drug payments, it is in the drug companies’ interests to create a large gap between what the hospital pays and the price reported to the government, the average wholesale price (AWP). J&J made money on Procrit by, in part, providing it to the hospital or doctors cheaply or even for free, making it attractive to providers, while reporting high prices to the government for reimbursement. They also provided financial incentives for doctors to prescribe Procrit. The FDA and Department of Justice considered prosecution for the misconduct, but they ultimately decided against it because of the complexity and lack of deterrence value.

Harris argues that EPO “increases death rates by more than a third” (154) while making companies like J&J billions.

After the 2004 FDA meeting, J&J and Amgen had to consent to clinical trials of EPO. In 2006, Amgen tried to hide the results of a Danish Head and Neck Cancer Study that had to be ended after it was found EPO caused tumor growth in cancer patients. In 2007, Paul Goldberg, editor of The Cancer Letter, released the results. A Canadian study confirmed the findings shortly after. In 2007, following these findings, the FDA chastised J&J and Amgen for hiding their study results on EPO. Soon after, the FDA put a black box warning—the most serious form of warning—on EPOs.

However, doctors continue to prescribe EPOs for cancer patients with anemia. Doctors justify this decision by arguing that “it’s helpful for their patients, many of whom are going to die anyway” (157). In 2007, the Centers for Medicare and Medicaid Services (CMS) issued guidance to doctors counseling against Procrit use, but they were formally reprimanded by the entire Senate and many members of the House of Representatives for doing so.

In 2011, CMS researchers released a study finding that EPO was not effective in kidney patients as well as cancer patients, and should be used only rarely. They also restructured Medicare payments to crack down on overbilling and abuse, resulting in fewer EPO prescriptions in most patients. However, cancer patients were exempt from the new guidance and EPO continued to be prescribed by oncologists. J&J published a study showing EPO caused cancer growth in women with breast cancer. From 2011, Procrit sales began to decline. However, oncologists in “small towns, small markets, and small cities” continue to prescribe Procrit (160). Harris argues they do so because they are more reliant on the financial perks for doing so.

Harris gives a history of the development of Risperdal, a treatment for recurrent psychosis, a symptom typical of schizophrenia but which can occur in other mental disorders like “bipolar disease.” Harris notes that Risperdal can also have severe side effects such as weight gain, diabetes, and stroke. It is sometimes used to control the behavior of children and the elderly.

In the 1980s, J&J began developing a replacement for their antipsychotic drug Haldol as the patent was about to run out. They claimed Risperdal caused fewer tremors and tics than “typical” antipsychotics and that it could be used to treat dementia. However, a 2005 study comparing Risperdal to Haldol found it caused more side effects, including tics and tremors. The FDA refused to claim Risperdal was an improvement over Haldol, but it approved the drug in 1993 for schizophrenia. 

J&J wanted to market it to a larger population, so they decided, “since Risperdal was approved to treat schizophrenia, and since schizophrenics suffered almost every psychiatric illness and symptom imaginable, Risperdal could be sold as a treatment for everything. Thus was born ‘Sell to the Symptoms’” (165).

J&J directed its sales reps to push doctors to prescribe Risperdal for any symptom of schizophrenia, including anxiety and agitation. J&J did this even though they were aware this practice was “illegal.” Other drug companies soon followed suit.

J&J executive Alex Gorsky was at the forefront of the marketing push. Gorsky and J&J had to convince state Medicaid directors that the higher cost for Risperdal compared to Haldol was worthwhile, as most patients with schizophrenia receive care through Medicaid. J&J used its foundation, the Robert Wood Johnson Foundation, as part of this push. In 1997, in Texas, Dr. Steven Shon, a Texas medical director, “consulted” with the foundation to create an algorithm called TMAP, adopted by Texas state mental health providers, that would promote Risperdal over older antipsychotics. Then J&J lobbied for Texas to mandate the algorithm be used by all mental health providers in the state. This was used as a model J&J exported to other states. In 2004, The New York Times reported on the arrangement. This encouraged other state medical providers to seek payments from J&J. The model was implemented in 16 other states.

Meanwhile, J&J funded the National Alliance on Mental Illness (NAMI), an advocacy group for parents of children who suffer psychotic breaks, and encouraged them to lobby for TMAP implementation. This is common; 83% of health advocacy groups receive money from medical companies (174).

The high price of mandated Risperdal pushed states to restrict access to their state mental health services.

By 1997, Risperdal sales had soared. However, studies on its effects on children and the elderly were concerning. In 1999, the FDA rejected a request to approve Risperdal for use in dementia and Alzheimer’s patients due to excess deaths in trials. Nevertheless, Gorsky pushed to market it to the elderly through J&J’s “ElderCare sales force” (177). In 2010, prosecutors alleged that J&J knew this practice was illegal.

In 1997, J&J partnered with Omnicare, “the nation’s largest pharmacy benefit manager (PBM) for nursing homes” (178). This helped J&J reach a larger share of Medicaid patients, those in care homes. Omnicare got illegal “kickbacks” in exchange for switching patients from older antipsychotic medications to Risperdal.

Risperdal soon was being prescribed to many elderly patients without schizophrenia. Despite the FDA’s admonishments about false advertising and refusal to authorize it for dementia, Gorsky expanded the ElderCare sales force and efforts to market the drug to children.

In clinical trials, children prescribed Risperdal for ADHD or conduct disorder were effectively tranquilized, reducing their problematic behavior, but they suffered from severe side effects of weight gain that could lead to heart attacks and stroke.

Between 1994 and 2003, many American children were medicated for “bipolar disorder.” This effort was led by Dr. Joseph Biederman of Harvard, a psychiatrist who partnered with J&J and advocated that bipolar disorder could appear in children as young as one or two. He conducted a series of trials that purported to show Risperdal was safe in children, some of which were ghostwritten by J&J. Biederman also advocated for J&J in the Continuing Medical Education classes (CMEs) physicians are required to take. Other psychiatrists criticized Biederman’s approach. Senator Chuck Grassley investigated Biederman and J&J’s relationship to reveal the history of payments, but Harvard did not punish the behavior.

Risperdal can result in the development of gynecomastia, or breast growth, in children, including boys. In 2000, J&J studied this side effect and found as many as 13% of patients developed the disorder before ending the study. They hid the results. In 2001, an Australian and New Zealand study into the use of Risperdal in patients with dementia found it caused “a sharp rise in heart attacks” (194). Nevertheless, Gorsky pushed the sales team to continue to market to children and the elderly. 

J&J conducted its own internal testing and found similar results, but “suppressed this information for years” (197). In 2003, the FDA mandated J&J include a warning on its label about risks of heart attack and stroke. A recent analysis found Risperdal contributed to the death of 1.2 million elderly people between 2000 and 2019.

In 2003, J&J launched a “Back-to-School” marketing campaign for its M-tab version of Risperdal for children. They held ice cream and popcorn parties at the offices of child psychologists and neurologists. They also provided free samples. Meanwhile, J&J sought to hide evidence from its studies into gynecomastia caused by Risperdal. They decided to publish the study by excluding children over the age of 10.

That same year, the FDA mandated J&J include a warning that the drug might cause diabetes. Blood sugar deregulation is a common side effect of antipsychotic drugs. J&J refused until the FDA threatened legal action. However, they continued to claim Risperdal did not cause diabetes in marketing messages.

In 2001, Vicki Starr joined J&J as a drug sales rep who pushed Risperdal as part of the “Sell to the Symptoms” campaign. A former pharmacist, she grew increasingly uneasy with J&J’s messaging and approach pushing Risperdal, especially as her nephew was taking the drug. A former colleague at Eli Lilly, another drug company, encouraged her to become a whistleblower. She was the first of five J&J employees who blew the whistle in 2004 and 2005, accusing J&J of “illegal promoting [of] Risperdal for off-label uses” (213). In the wake of these cases, Gorsky left J&J for four years.

In 2004, J&J pulled back on marketing to children to minimize legal exposure.

In 2005, the FDA mandated that Risperdal would have to carry a warning of “increased mortality in elderly patients with dementia-related psychosis” (215). Nevertheless, J&J continued marketing to the elderly. They advised practitioners to diagnose their elderly patients with “late-onset schizophrenia,” a “virtually nonexistent” illness (216), to mitigate legal liability for the prescriptions. Risperdal continues to be used to control patient behavior in understaffed nursing homes. In 2023, the Biden administration passed legislation to improve nursing home staffing levels and reduce the use of antipsychotics.

After the 2005 FDA mandate, many sales reps grew increasingly uneasy with the messages they were pushing about Risperdal use in elderly patients. One rep stated that “nothing else has ever come close to the shady tactics of J&J” (220) in his career.

In November 2005, J&J disbanded its ElderCare sales force. Its competitors pled guilty to similar charges of inappropriately marketing antipsychotic medication to the elderly. J&J, by contrast, insisted it had a First Amendment right to “sell products they believed in” (221). In 2009, J&J met with DOJ lawyers and argued it should not be charged and the FDA, dependent on J&J’s contributions, refused to take action. In 2012, Gorsky was selected to be J&J’s next chief executive following his oversight of the acquisition of Synthes, a marker of “surgical power tools and orthopedic implants.”

In 2013, J&J finally agreed to plead guilty for its illegal marketing of Risperdal to the elderly and children and to pay a $2.2 billion fine. No doctors or executives were charged, although the DOJ did consider charging Gorsky. Harris speculates that federal prosecutors may have worried about pursuing charges against J&J because they did not wish to harm their future careers at major law firms, of whom J&J is a top client. For instance, Dave Hoffman, the first federal prosecutor of the Risperdal case, went on to work at Sidley Austin, a firm of which J&J was a client.

The typical narrative of the opioid epidemic in the United States begins with the FDA approval of Purdue Pharma’s OxyContin in 1995. Harris argues that the story actually begins with the FDA’s approval of J&J’s long-release fentanyl patch, Duragesic, in 1990. 

Both OxyContin and Duragesic claimed to be able to provide long-term pain relief through slow-release mechanisms; however, these mechanisms could be easily bypassed. Although the FDA raised concerns about Duragesic’s patch technology and its inability to deliver low doses consistently, it was approved. By 1993, 52 patients using Duragesic patches as prescribed had died of accidental overdose. This resulted in a black box warning on Duragesic, the strongest FDA warning.

Purdue Pharma marketed OxyContin as a product that could be used “broadly and widely for everyday chronic pain” (232). Around 1997, J&J began to take a similar approach to marketing Duragesic patches. J&J began to donate to advocacy groups like the American Pain Society and the American Academy of Pain Medicine, which issued a statement arguing for greater access to opioids outside of end-of-life or critical care settings.

In 1998, the FDA warned J&J that Duragesic resulted in worse side effects like abdominal pain than older medications like morphine. It also warned that J&J’s marketing of the medication falsely made it seem like it would not cause impairment.

In the 1960s, medical companies began growing poppies in Tasmania, an Australian island, for opioid production. In 1982, J&J bought Tasmanian Alkaloids, an opioid producer, and began making codeine for its Tylenol plus codeine pills. However, it had more opioids than it could use, so J&J began to provide wholesale distribution of the active ingredient for opioids to other companies, including Purdue Pharma. 



When Purdue requested more of the ingredient as OxyContin sales soared, J&J worked to develop a high-yield poppy, named “Norman,” which debuted in 1996. As demand increased through the 1990s, J&J provided incentives to Tasmanian farmers like luxury cars to encourage them to plant Norman poppies. J&J ultimately provided a majority of all opioid raw ingredients in the United States. J&J and Purdue struck a four-year agreement that J&J would be the sole provider of its raw ingredients for the manufacture of OxyContin.

In 2000, stories about the ravages of the prescription opioid epidemic began to appear in American newspapers. The focus of many of those stories, then and now, is FDA failures, Purdue Pharma and the Sackler families, and payments to doctors. However, they do not often acknowledge J&J’s role in the crisis.

Around 2001, OxyContin’s prescriptions began to fall. Seeing room in the market, J&J began falsely marketing Duragesic, its fentanyl patches, as non-addictive. At the time, the dangers of fentanyl were not as widely known by the public as they are now. Since fentanyl does not appear in urine tests like other opioids, it did not appear in analyses of overdoses or deaths captured in the DAWN system, allowing J&J to claim it was safer and non-addictive compared to other opioids. J&J sales reps used DAWN results to market Duragesic. In 2002, McKinsey counseled J&J to “target high abuse-risk patients” (242) to further improve sales of Duragesic. J&J also focused marketing of Duragesic to so-called “pill mill” doctors, a small group of doctors with much higher-than-average opioid prescriptions.

In 2003, Duragesic’s patent ran out, meaning competitors could make generic versions. A company called Mylan Labs developed a generic fentanyl patch that “used ‘matrix’ technology, wherein fentanyl is embedded in an adhesive layer, while J&J’s patch used a ‘reservoir’ method” (245). J&J argued to the FDA that the “matrix” technology was unsafe and ripe for abuse. Meanwhile, J&J launched a “matrix patch” in Europe.

In 2008, the FDA made J&J recall Duragesic reservoir patches in the US because of “manufacturing defects.” At the same time, J&J was secretly recalling (buying back) Tylenol and other medications because of contamination. In 2009, in response to FDA pressure, J&J stopped selling reservoir patches in the US and began to sell “matrix” Duragesic patches.

Although the FDA punished executives at Purdue Pharma for their role in the opioid epidemic, J&J largely avoided any charges or fines for its contributions. In addition to its Duragesic patch, in 2008, J&J launched Nucynta, another so-called “nonaddictive,” extended-release opioid. Although the FDA reviewers noted Nucynta had the same problems as OxyContin in terms of potential for dependency and overdose, it was nevertheless approved. In 2011, the FDA chastised J&J for making misleading claims that Nucynta did not provoke the same abdominal problems as other opioids.

The same year they launched Nucynta, J&J changed how sales reps recorded marketing pitches to reduce liability in the event of litigation. They also funded a network of speeches, seminars, patient groups, and similar to spread the message that the epidemic of untreated pain is more serious than the opioid epidemic. They used this network to “influence in some other way almost every source of information doctors relied upon in treating their patients” (250).

Expert Andrew Kolodny argues that J&J was the “kingpin of the opioid epidemic” (251).

Harris summarizes how J&J’s approach to introducing new products shifted over the years. Early on in its history, J&J introduced new products with the assumption that they were safe, even if it later hid evidence that they were not. However, as the FDA’s power waned, J&J began bringing products to market that they knew were unsafe before launch, as they knew they would face few consequences. This approach made the company billions of dollars while maintaining the company’s branding as one that did good.

Harris provides a brief history of the development of hormonal birth control, noting especially that in 1988 levels of estrogen in hormonal birth control were lowered substantially to reduce risk of blood clots. In 1996, J&J announced it would be developing a birth control patch called Ortho Evra. However, in 1999 a J&J trial of the product found it resulted in higher levels of estrogen doses than typical birth control pills because of the difficulty of dosing via patch. Further, the doses were inconsistent, meaning more people got pregnant while on the patch than with conventional hormonal birth control. J&J massaged the results of the trial to make Ortho Evra seem safe and reported those altered results to the FDA. It won FDA approval in 2001. In 2004, regulators in New Zealand were so concerned about excess deaths and strokes due to Ortho Evra that they reviewed the evidence carefully and refused its approval.

In April 2002, J&J launched Ortho Evra in the US. The FDA began receiving reports of severe side effects and accidental pregnancies. J&J continued to obfuscate its concerning internal findings about side effects from the FDA and European regulators. Nevertheless, in 2005, the FDA issued a warning label for Ortho Evra alerting users to the high estrogen levels in the patch, and sales fell. However, the FDA did not pull Ortho Evra from the market “because women need as many contraceptive options as possible” (261).