The most widely cited estimate is that roughly 250,000 people die from medical errors each year in the United States, a figure that would make it the third leading cause of death behind heart disease and cancer. That number comes from a 2016 Johns Hopkins study published in The BMJ, and while it remains controversial, no comprehensive national tracking system exists to confirm or replace it.
Where the 250,000 Figure Comes From
In 2016, researchers at Johns Hopkins University analyzed the existing scientific literature on medical errors dating back to a landmark 1999 report by the Institute of Medicine (IOM). That earlier report had estimated between 44,000 and 98,000 Americans died each year from preventable medical mistakes. The Johns Hopkins team took data from studies published since then, calculated an average error rate, and extrapolated it across all U.S. hospital admissions in 2013. They arrived at 251,454 deaths per year.
The researchers themselves acknowledged the estimate’s limitations. Extrapolating from smaller studies to the entire hospital population requires assumptions that may not hold across every type of facility and patient. But they argued the figure highlights a critical gap: the U.S. has no systematic way to measure deaths caused by medical error. Death certificates use a coding system that captures the disease a patient had, not whether a mistake contributed to the death. Errors simply don’t appear in the official mortality data.
Why the Number Is Hard to Pin Down
Medical errors are genuinely difficult to count. A patient with advanced heart failure who receives the wrong medication and dies had multiple factors contributing to their death. Was the error the primary cause, or did it accelerate an outcome that was already likely? Different studies draw that line in different places, which is one reason estimates range so widely, from the IOM’s original 98,000 ceiling to the Johns Hopkins figure of 251,000.
There’s also no universal definition of “medical error.” Some analyses include only clear-cut mistakes like wrong-site surgery or dangerous drug interactions. Others fold in failures of communication, delayed treatments, and systemic breakdowns where no single person made an obvious mistake but the patient was still harmed. The broader the definition, the higher the count.
Diagnostic Errors Are the Biggest Contributor
A separate Johns Hopkins analysis estimated that 795,000 Americans are permanently disabled or die each year because of misdiagnosis. Not all of those cases are fatal, but the scale is striking. The study identified 15 dangerous diseases that accounted for about half of all serious harm from diagnostic error. The top five were stroke, sepsis, pneumonia, blood clots, and lung cancer.
These conditions share a common thread: they can mimic less serious problems early on, and delays in recognizing them shrink the window for effective treatment. A stroke misread as a migraine, sepsis dismissed as a routine infection, or lung cancer initially treated as chronic bronchitis can each become fatal when the correct diagnosis comes too late.
Medication Errors and Surgical Complications
Beyond misdiagnosis, medication mistakes and post-surgical complications are major sources of preventable harm. The World Health Organization estimates that 1 in every 20 hospital patients experiences preventable harm, and roughly half of that harm involves medications: wrong doses, dangerous drug combinations, or drugs given to patients with known allergies.
Surgical complications, hospital-acquired infections, and pressure injuries also carry significant costs. One actuarial study estimated that measurable preventable errors in hospitals cost $17.1 billion per year, with post-surgical complications and healthcare-associated infections making up the bulk of that figure. Events that should never happen in a hospital, sometimes literally called “never events” (such as operating on the wrong body part), added another $3.7 billion in excess costs.
Why Staffing Levels Matter
Research consistently links nurse-to-patient ratios with patient safety outcomes. When nurses are responsible for more patients, the risk of errors, complications, and death goes up. The Agency for Healthcare Research and Quality notes that multiple studies have documented this relationship, and the mechanism is straightforward: higher workloads mean more tasks competing for attention, more interruptions, greater fatigue, and higher rates of burnout.
A British study found that when nurses had more patients to manage, essential care steps were more frequently missed. These aren’t dramatic errors. They’re things like delayed medication administration, missed changes in a patient’s vital signs, or incomplete handoffs between shifts. Individually, each lapse seems small. Collectively, they create the conditions where serious harm becomes more likely.
Progress Has Been Slow
The 1999 IOM report, titled “To Err Is Human,” was supposed to be a turning point. It generated enormous public attention and spurred hospitals, regulators, and professional organizations to invest in patient safety. More than two decades later, the results are mixed. According to a review published through the AHRQ Patient Safety Network, effective interventions have been developed for specific problems like hospital-acquired infections and certain types of medication errors. Implementation, however, remains inconsistent, and progress in other areas has been uneven. The review’s conclusion is blunt: preventable harm remains significant.
Part of the challenge is that hospitals are inherently complex systems. Fixing one type of error doesn’t prevent others, and new technologies introduce new failure points even as they solve old ones. Electronic health records, for instance, reduced some prescribing errors but created new ones related to alert fatigue, where clinicians are bombarded with so many automated warnings that they start dismissing them reflexively.
What These Numbers Mean in Context
If the 250,000 estimate is roughly accurate, medical errors kill more Americans each year than chronic lower respiratory diseases, accidents, or stroke, each of which typically ranks among the top five causes of death. Even the more conservative 1999 estimate of 44,000 to 98,000 deaths would place medical errors among the top ten.
Globally, the picture is harder to quantify because most countries face the same data collection problems the U.S. does. The WHO has made patient safety a priority, noting that the burden of harm from unsafe care is one of the leading causes of death and disability worldwide, but precise global death tolls don’t exist.
The honest answer to “how many people die from medical errors” is that nobody knows the exact number. The best available evidence puts the U.S. figure somewhere between 100,000 and 250,000 per year, a range wide enough to reflect genuine scientific uncertainty but narrow enough to confirm that the problem is enormous by any standard.