The short answer is that there is no single drug the FDA pulled — there are dozens, and the exposed pattern is more alarming than any individual case. The most infamous example is Vioxx, a prescription arthritis painkiller approved in 1999 and voluntarily withdrawn by Merck on September 30, 2004, after clinical trials revealed it significantly increased the risk of heart attacks and strokes. By the time it was pulled, Vioxx had been linked to between 88,000 and 140,000 cases of serious heart disease in the United States alone. Merck ultimately paid $4.85 billion in 2007 to settle roughly 47,000 injury claims, then another $950 million to the Department of Justice in 2011 on criminal charges, and $830 million more to settle an investor class action in 2016.
But Vioxx was not an isolated failure. During an 18-year stretch from 1993 to 2010, 17 FDA-approved drugs were withdrawn from the market after being prescribed at 112 million physician office visits across the country. Nine of those 17 drugs had been prescribed more than one million times before anyone pulled them. For anyone caring for a loved one with dementia — a population that often takes multiple medications daily and is especially vulnerable to adverse drug effects — these numbers should command serious attention. This article walks through the worst cases, explains what went wrong in each, examines why the system keeps failing, and offers practical guidance for protecting yourself and the people you care for.
Table of Contents
- Why Did the FDA Pull Drugs It Had Already Approved — And What Actually Went Wrong?
- The Drugs That Stayed on Shelves for Decades Before Anyone Acted
- Zantac and the Problem of Contamination After Approval
- How to Protect a Loved One With Dementia From Unsafe Medications
- The Manufacturing and Compounding Failures You Do Not Hear About
- Why Foreign Manufacturing Raises Additional Concerns
- What Is Changing and What Still Is Not
- Conclusion
- Frequently Asked Questions
Why Did the FDA Pull Drugs It Had Already Approved — And What Actually Went Wrong?
The FDA approval process, for all its rigor, has a fundamental limitation: clinical trials are conducted on relatively small populations over relatively short periods. A drug tested on 5,000 people for 12 months may look safe, but once it reaches millions of patients taking it for years, rare but devastating side effects can surface. That is precisely what happened with Vioxx. Merck’s own clinical data showed elevated cardiovascular risk after 18 months of use, but the company allegedly withheld that information from doctors and patients for more than five years. The gap between what the manufacturer knew and what the public was told is the core of what went wrong — not just with Vioxx, but with a recurring pattern of post-approval failures.
The problem intensifies with the FDA’s accelerated approval pathway, which was designed to get promising drugs to patients with serious conditions faster. Of 328 drugs approved through this pathway between 1992 and 2024, 49 — nearly 15 percent — were subsequently withdrawn. And when a drug approved through accelerated approval does fail, it takes the FDA an average of 46 months to actually pull it from the market. That is almost four years of continued prescriptions after the evidence has turned against the drug. For older adults, particularly those managing cognitive decline who may not be positioned to advocate for themselves, those four years represent a dangerously long window of exposure.
The Drugs That Stayed on Shelves for Decades Before Anyone Acted
Some of the worst cases are not about drugs that were pulled quickly once problems emerged — they are about drugs that lingered on the market for staggering lengths of time despite mounting evidence of harm. Darvon and Darvocet, both containing the painkiller propoxyphene, were available from 1955 until 2010. That is 55 years on the market before withdrawal. During just the period from 1981 to 1999, more than 2,110 deaths were reported in connection with these drugs, and the cause of withdrawal was serious cardiac toxicity. The question is not just why these drugs were approved — it is why they were allowed to remain available for more than half a century while people died. However, it is important to understand that “the FDA pulled it” does not always mean the FDA acted decisively.
In many cases, manufacturers voluntarily withdrew their products only after litigation, public pressure, or independent research made continued sales untenable. Merck pulled Vioxx voluntarily. Bayer withdrew its cholesterol drug Baycol (cerivastatin) in 2001 after it was linked to rhabdomyolysis, a fatal degenerative muscle disease, at rates far exceeding those of other statins. Nearly 10 million prescriptions for Baycol had been written before it came off the market. The distinction matters: if you are waiting for the FDA to protect you or your family member, you may be waiting longer than is safe. Independent vigilance — from patients, caregivers, and physicians — fills a gap that the regulatory system has repeatedly failed to close.
Zantac and the Problem of Contamination After Approval
Not every withdrawal involves a drug that was inherently dangerous at the time of approval. Zantac, the widely used heartburn medication containing ranitidine, was pulled on April 1, 2020, after the FDA requested removal of all ranitidine products from the market. The issue was contamination with NDMA, or N-nitrosodimethylamine, a probable carcinogen. What made this case particularly troubling was the discovery that NDMA levels in ranitidine increased over time and at higher storage temperatures, meaning that older bottles sitting in medicine cabinets or on pharmacy shelves contained dangerously elevated levels of a cancer-causing compound.
The contamination was first flagged by an independent laboratory in the summer of 2019 — not by the manufacturer and not by the FDA. This is a critical point for dementia caregivers to understand. Many older adults keep medications for extended periods, sometimes well past expiration dates, and may not follow recall notices. Ranitidine was available over the counter, meaning millions of people were taking it without physician oversight. The Zantac case illustrates a different kind of failure: a drug that may have been safe as originally formulated but became dangerous through degradation, and a system that relied on outside testing to catch the problem.
How to Protect a Loved One With Dementia From Unsafe Medications
For caregivers managing medications for someone with cognitive decline, the practical challenge is real: how do you stay ahead of problems the FDA itself takes years to identify? The first step is maintaining an active, current medication list and reviewing it regularly with a pharmacist — not just a physician. Pharmacists are often more current on drug recalls and safety warnings than prescribing doctors, and a thorough medication review can flag interactions, duplications, and drugs that have come under scrutiny. The second step is weighing the tradeoff between newer drugs and established ones.
Newer medications approved through accelerated pathways carry a roughly 1-in-7 chance of eventual withdrawal, based on the historical data. Older, well-established drugs have a longer safety track record but, as the Darvon case shows, longevity on the market is not a guarantee of safety either. The best approach is to ask direct questions: What is the evidence base for this drug in older adults specifically? Has it been tested in populations with cognitive impairment? Are there safer alternatives with longer post-market track records? A drug that works well in a 45-year-old clinical trial participant may behave very differently in an 82-year-old with Alzheimer’s disease who is also taking four other medications.
The Manufacturing and Compounding Failures You Do Not Hear About
Drug withdrawals get headlines, but a quieter and arguably more frightening problem is the surge in manufacturing and compounding failures that put patients at immediate risk. In April 2024, ReviveRX recalled 751 vials of compounded tirzepatide — a GLP-1 drug used for diabetes and weight loss — because the vials contained testosterone instead of tirzepatide. This was classified as a Class I recall, the FDA’s most serious category, meaning the product could cause serious injury or death. Imagine a dementia patient receiving testosterone injections when their caregiver believed they were administering a diabetes medication. The consequences could be severe and baffling to diagnose.
The problem extends beyond compounding pharmacies. The FDA recalled more than 233,000 bottles of duloxetine, a widely prescribed antidepressant also used for nerve pain in older adults, due to contamination with a carcinogenic nitrosamine impurity — the same class of contaminant that brought down Zantac. Meanwhile, the FDA found that Novo Nordisk failed to promptly report cases of suicide and suicidal ideation among patients taking its blockbuster drugs Ozempic and Wegovy. For caregivers, the warning here is that even currently approved, widely prescribed medications can carry risks that are not being fully communicated. Staying informed through FDA safety announcements and asking pharmacists about recent recalls is not optional diligence — it is necessary protection.
Why Foreign Manufacturing Raises Additional Concerns
A growing share of medications consumed in the United States is manufactured overseas, and FDA oversight of foreign facilities has not kept pace. The FDA found deficiencies in 66 percent of foreign facility inspections — twice the rate of deficiencies found in domestic inspections. This means that drugs produced abroad are failing quality checks at double the rate of domestically manufactured products, and many foreign facilities go years between inspections.
For caregivers selecting medications, this is difficult information to act on because the country of manufacture is not always clearly labeled on a prescription bottle. However, asking your pharmacist whether a generic medication is domestically produced, or requesting a specific manufacturer known for quality, is a reasonable step. It is also worth noting that many of the contamination-related recalls — including nitrosamine issues — have been traced to manufacturing processes at overseas facilities.
What Is Changing and What Still Is Not
There are signs of incremental improvement. The FDA has faced increasing congressional pressure to shorten the timeline for withdrawing failed drugs, and public awareness of the accelerated approval pathway’s limitations has grown. Independent laboratories and patient advocacy groups now play a more active role in post-market surveillance than they did during the Vioxx era. The Zantac contamination, for instance, was caught by outside testing, which suggests that the ecosystem of safety monitoring is broadening beyond the FDA and manufacturers alone.
But structural problems remain. The FDA still relies heavily on manufacturers to self-report safety data, and the average 46-month withdrawal timeline for failed accelerated-approval drugs has not meaningfully shortened. For families navigating dementia care, the takeaway is unsentimental: the system will not reliably protect your loved one in real time. Medication vigilance — regular reviews, awareness of recalls, skepticism toward newly approved drugs without long track records, and open communication with pharmacists — is a caregiving responsibility that cannot be delegated to regulators alone.
Conclusion
The history of FDA drug withdrawals is not a story about one bad drug or one reckless company. It is a systemic pattern in which approval processes that are necessarily limited in scope collide with commercial pressures that discourage transparency about emerging risks. From Vioxx’s cardiovascular devastation to Zantac’s carcinogenic contamination to compounding pharmacies shipping vials of the wrong drug entirely, the failures span decades and show no sign of disappearing. Seventeen drugs withdrawn over 18 years, 49 accelerated approvals pulled, and millions of prescriptions written for medications that turned out to be dangerous — these are the numbers that define the problem. For dementia caregivers, the practical imperative is clear.
Maintain current medication lists. Schedule regular pharmacist reviews. Ask hard questions about new prescriptions, especially drugs approved through accelerated pathways. Monitor FDA recall announcements. And never assume that because a drug is FDA-approved, it is guaranteed to be safe for an elderly person with cognitive impairment and multiple co-existing conditions. The approval is the beginning of the safety story, not the end of it.
Frequently Asked Questions
How often does the FDA pull a drug it already approved?
Between 1993 and 2010, 17 drugs were withdrawn after approval, and they had been prescribed at 112 million physician office visits. Of drugs approved through the accelerated approval pathway specifically, about 15 percent — 49 out of 328 between 1992 and 2024 — were eventually withdrawn.
How long does it take the FDA to withdraw a drug once problems are found?
On average, it takes the FDA 46 months to withdraw a drug that received accelerated approval and subsequently failed to demonstrate benefit or safety. That is nearly four years of continued patient exposure.
Are older, long-established drugs safer than newer ones?
Not necessarily. Darvon and Darvocet remained on the market for 55 years before being pulled for cardiac toxicity, with over 2,110 deaths reported in just an 18-year window. A long market history does not guarantee safety, though newer drugs approved via accelerated pathways do carry a statistically higher withdrawal rate.
What was the most harmful drug the FDA approved and later pulled?
By sheer scale of injury, Vioxx is the most significant. It was linked to between 88,000 and 140,000 cases of serious heart disease in the U.S. and resulted in over $6.6 billion in settlements and penalties paid by Merck.
Should dementia patients avoid all newly approved medications?
Not categorically, but caregivers should exercise heightened caution. Ask whether the drug has been tested in elderly populations or people with cognitive impairment. Request information about how long the drug has been on the market post-approval, and whether any safety signals have emerged. Weigh the benefits against the roughly 1-in-7 historical withdrawal rate for accelerated-approval drugs.
How can I check if a medication has been recalled?
The FDA maintains a searchable database of drug recalls at fda.gov. You can also sign up for FDA safety alerts by email. Your pharmacist is another reliable source — they receive recall notices directly and can flag affected medications during routine reviews.
