President John F. Kennedy addresses Congress in 1962 about the need to protect the public from prescription drugs and other products. (Photo: John F. Kennedy Presidential Library and Museum)
In 1962, President John F. Kennedy addressed Congress about the need to protect the public from a variety of consumer products, especially prescription drugs.
While Kennedy hailed the lifesaving benefits of thousands of new drugs that had come on the market, he issued a prescient warning: “These new drugs present greater hazards as well as greater potential benefits than ever before, for they are widely used; they often are very potent; and they are promoted by aggressive sales campaigns that tend to overstate their merits and fail to indicate the risks involved in their use.”
That year, public health officials reported 1.4 million Americans had been the victims of a “therapeutic misadventure,” or serious drug reaction. The U.S. Public Health Service said there probably were more.
Since the birth of the modern pharmaceutical industry after World War II, many drugs that were deemed safe and effective when brought to market were later linked to serious side effects and deaths.
And as millions of these adverse events piled up over the decades, federal auditors would find time and again that the original Public Health Service warning was right on the money: Whatever the number of events, it was many more than was being reported.
Knowing the true safety of prescription drugs has been a holy grail of modern medicine because nearly half of Americans take at least one drug and a quarter take at least three. But it has been a failed quest, one sorely exposed during the coronavirus pandemic as thousands of serious adverse events have been linked to unproven drugs that were repurposed as COVID-19 treatments.
Yet the root of the problem has been known for decades, and potential solutions have failed, a Milwaukee Journal Sentinel investigation has found. Due largely to its voluntary structure — only drug companies are required to report adverse events — the primary system for monitoring those incidents has never fulfilled its promise.
Research indicates that fewer than 10% of adverse drug events actually are reported to the U.S. Food and Drug Administration.
Requiring hospitals to report cases of patients who may have been seriously harmed by a drug would dramatically increase both the number of reports and the quality of the data, said Norman Marks, a physician who oversaw the FDA’s adverse events system from 2000 to 2014.
“Absolutely,” Marks said. “The hospitals are probably the way to do it.”
Marks said that during his years with the FDA several pilot programs were designed to simplify and encourage more reporting by health care professionals, including at hospitals. But the agency, he said, did not provide support for the programs.
During the pandemic, nearly 7,000 serious adverse events, including 1,900 deaths, were filed with the FDA for drugs used to treat COVID-19, a Journal Sentinel analysis of the agency’s database found.
Track COVID-19 in Wisconsin:See the latest numbers and trends
How to interpret COVID-19 data:What experts say about positive cases, deaths and hospitalizations
Most of those reports, filed through Sept. 30, were for drugs that had been approved for other conditions and were repurposed for COVID-19. They include an antimalarial drug, an antibiotic, an HIV drug and a rheumatoid arthritis drug, none of which have proven to be effective in preventing deaths in COVID-19 patients.
What is unknown is whether those 7,000 reports represent a large portion of all serious adverse events linked to COVID-19 drugs or just the tip of the iceberg.
Journal Sentinel investigations have shown that use of unproven and potentially dangerous drugs, such as the anti-malaria drug hydroxychloroquine and the antibiotic azithromycin, have led to a surge in reports of serious adverse events and deaths. Both drugs were promoted as COVID-19 cures by President Donald Trump and others.
Likewise, a Journal Sentinel investigation showed that after doctors promoted the powerful rheumatoid arthritis drug Actemra as a possible lifesaver for COVID-19 patients, there was an increase in its use and then a spike in serious adverse events and deaths.
Futile efforts at reform
Federal officials have long known that more cooperation from hospitals and doctors could dramatically improve the adverse events system. But for a variety of reasons, ranging from ignorance about filing reports to a lack of time to a fear of malpractice lawsuits, regulators have failed to enlist the medical community’s support.
Audits of the program and other efforts to fix it show futile attempts to enlist the backing of powerful groups such as the American Hospital Association and the American Medical Association, according to reports and records reviewed by the Journal Sentinel.
When the U.S. General Accounting Office, now the Government Accountability Office, audited the program in 1974, it said that each year drugs adversely affect 6 million people in the U.S. and that the FDA’s reporting system was not being adequately used.
“The FDA’s effort to use hospitals has met with little success,” said the GAO, which serves as an independent watchdog agency for Congress.
Even the use of bounties in the late 1960s that paid doctors $5 for each report and paid hospitals a $50 monthly administration fee failed to work.
By the time of a 1982 GAO audit, things only had gotten worse.
“The number of reports from private and federally operated hospitals has decreased dramatically since 1970,” the audit said.
It said more cooperation was needed from groups such as the American Hospital Association and the American Medical Association, both of which told the auditors that fear of legal action was a barrier to reporting adverse events to the FDA.
In 2000, another GAO report said that only 1% to 10% of adverse events were being sent to the system.
“This has been an intractable problem for a long time,” said Paul Beninger, an associate professor of public health and community medicine at Tufts University School of Medicine and former FDA official.
Requiring hospitals to report serious adverse events would improve the system, he said, but “the pushback would be incredible.”
After GAO audits in the 1980s showed that hospitals were reporting less than 1% of adverse events caused by medical devices, the law was changed and, beginning in the 1990s, hospitals and other medical facilities were required to report serious injuries and deaths related to medical devices.
Likewise, since 1986, health care providers have been required to report certain adverse events suspected to have been caused by vaccines.
The American Hospital Association declined to comment on whether hospitals should be required to report serious adverse events from drugs to the FDA.
In an email, spokeswoman Nancy Foster said hospitals use a variety of tools to prevent medication errors, such as bar codes and prescribing software that help prevent unwanted drug interactions and dosing mistakes.
In 2016, researchers at Tufts Medical School and the drugmaker Amgen surveyed health care professionals, many of whom worked at hospitals, and found that 51% had not reported an adverse drug event to the FDA in the previous five years. One of the reasons was a common, long-standing complaint about using the system: a lack of time.
However, in 2010 researchers at Harvard Medical School and drugmaker Pfizer came up with what appeared to be a relatively painless solution.
In a pilot program, they found that electronic health records like those used today by many health care systems easily could be adapted to trigger detailed adverse drug reports that would be acceptable to doctors.
It took just 53 seconds for a doctor to complete the form.
It also seemed to work. While the 26 doctors who took part in the program submitted no adverse event reports to the FDA in the previous year, they sent in 217 reports during the five months they were in the program.
The program was so promising that it was presented to the FDA, said Jeffrey Linder, lead author of the study.
“The FDA wasn’t that interested in it,” said Linder, now chief of general internal medicine and geriatrics at the Northwestern Feinberg School of Medicine. “It kind of died on the vine.”
But Linder said he remains convinced that such an approach could dramatically increase the number of adverse event reports from hospitals to the FDA and help identify potential safety signals.
While the program showed the feasibility of using electronic health records to generate reports of adverse events, there were questions about the quality of the data from those automated reports, said Chanapa Tantibanchachai, a spokeswoman for the FDA. For instance, she said, the reports often did not contain a narrative section, which limited their value.
The Harvard-Pfizer program “represents an important first step in bringing spontaneous reporting from the point of care into the electronic age,” she said. “Nonetheless, the value of the system will depend on the data it provides to drug safety scientists.”
She also noted that the adverse events reporting system is just one database used by the agency to monitor and evaluate the safety of medicines.
Yet the system has been the primary method for collecting reports of adverse events for drugs once they get on the market.
Requiring hospitals to report serious adverse drug events would substantially increase reporting, but funding for additional hiring might need to be provided to hospitals, said Aaron Kesselheim, a professor of medicine at Harvard Medical School who studies prescription drugs and the FDA.
“It would make the adverse events reporting system more rigorous,” he said.
An elusive goal
A robust system for monitoring the safety of drugs after they get on the market has long been a goal of drug safety experts. Part of the reason has to do with the way drugs are tested and approved.
The clinical trials that lead to the approval of new drugs are of limited duration and often exclude people with the kinds of health conditions that could make them more susceptible to a serious drug side effect. But once a drug gets on the market, those same kinds of real-world patients may be given the drug.
A system that quickly captures large numbers of those side effects is more likely to detect problems so that doctors and patients can be warned earlier.
In addition to substantial underreporting, the FDA’s adverse events system has limitations. Individual reports aren’t verified, and the mere filing of a report does not prove causation. And the system only provides a numerator. It does not say how many people are using a drug.
But once a safety signal is detected, researchers can use other measures, including the FDA’s national system known as Sentinel that can tap into and analyze large databases of health insurance organizations and health care providers to determine if there is a real problem.
Between 1969 and 2002, 75 drugs were removed from the market at least partly because of adverse event reporting, according to a 2005 study by FDA researchers.
Those include drugs that caused hepatitis, abuse and dependence, serious skin disorders, bleeding, cancer, a variety of heart problems, a rare brain disease and death.
The system also helped provide an endless stream of red flags and precautions, including dozens of black box warnings, the most stringent kind issued by the FDA.
But in 2004, adverse event reports were not being filed often enough to have made a difference in the life of Helen Tschannen.
Bev Webber, left, with her mother, Helen Tschannen, in this family photo at Tschannen’s 77th birthday celebration on Feb. 12, 2004. (Photo: Courtesy of Bev Webber)
Tschannen died in an Illinois hospital that year of a fungal infection known as histoplasmosis, a condition that for most people causes only mild symptoms. But she was at higher risk because she just had started using the powerful rheumatoid arthritis drug Remicade.
Remicade was first approved in 1998, and by 2001 there already were enough reports of invasive histoplasmosis cases that the FDA issued a black box warning for it and other drugs in its class.
But it was not until 2008, four years after Tschannen died, that the FDA issued yet another black box warning, the most serious level, telling doctors that histoplasmosis cases were not being consistently recognized in patients using it and three other related drugs.
Those delays in treatment led to the deaths of patients, the FDA said.
Tschannen’s daughter, Bev Webber, a retired nurse who lives in Mukwonago, said she believes that more vigilant reporting of adverse events linked to Remicade might have allowed doctors to recognize her mother’s infection sooner and start treating her earlier.
“I do think it would have made an impact on her treatment,” Webber said. “If the targeted treatment for histoplasmosis had been started when she started feeling ill, I think her chances of recovery would have greatly improved.”
Daphne Chen of the Milwaukee Journal Sentinel staff contributed to this report.
Read or Share this story: https://www.jsonline.com/story/news/investigations/2020/11/30/even-before-covid-19-fda-has-done-poor-job-tracking-drug-side-effects/3778215001/