In the United States, estimates of annual deaths from medical errors range widely, from roughly 40,000 to over 370,000, depending on how researchers define and count them. Globally, the World Health Organization estimates that more than 3 million people die each year from unsafe medical care. These numbers make medical error one of the most significant yet least visible causes of death worldwide.
The wide range in estimates reflects genuine disagreement among researchers about what counts as a preventable medical death and how to measure it. Understanding the different figures, and why they vary so much, gives a clearer picture of the real risk.
The Major U.S. Estimates
The most frequently cited figure comes from a 2016 analysis by researchers at Johns Hopkins, who argued that medical errors cause roughly 250,000 deaths per year in the United States, making them the third leading cause of death behind heart disease and cancer. That estimate drew enormous attention, but it also drew serious criticism. The Agency for Healthcare Research and Quality noted that the Johns Hopkins figure was generated by combining error rates from earlier studies without following standard guidelines for that kind of statistical synthesis. The original data also couldn’t clearly determine whether adverse events actually caused a patient’s death or whether the patient died from their underlying illness after experiencing an error during treatment.
More recent research from Johns Hopkins, published in 2023, focused specifically on diagnostic errors and arrived at even larger numbers. That analysis estimated 371,000 Americans die and another 424,000 are permanently disabled each year from misdiagnosis alone, totaling roughly 795,000 serious harms. A follow-up calculation using more recent patient volume data pushed that estimate to around 909,000 serious harms annually. At the other end of the spectrum, older studies based on hospital autopsy data placed misdiagnosis-related deaths at 40,000 to 80,000 per year.
The honest answer is that no one knows the precise number. The true figure almost certainly falls somewhere between the conservative estimates of tens of thousands and the higher estimates of hundreds of thousands. What every study agrees on is that the number is large enough to represent a major public health problem.
Why the Numbers Vary So Much
Counting deaths from medical error is fundamentally different from counting deaths from, say, car accidents. When someone dies after a medication mix-up in an otherwise healthy patient, the connection is clear. But most patients who experience medical errors are already seriously ill. A 70-year-old in the ICU with organ failure who also receives the wrong medication presents a much harder question: did the error cause the death, or was the outcome inevitable?
Studies that use broad definitions of “harm” and count any error in the care of a patient who later died will produce high numbers. Studies that require a clear causal link between the error and the death produce much lower numbers. Neither approach is wrong exactly, but they’re measuring different things. The broader estimates capture the full scope of the problem, while the narrower ones identify cases where the error was clearly the deciding factor.
There’s also a reporting problem. Medical errors don’t appear on death certificates. The U.S. uses a coding system for causes of death that has no category for “medical error,” so these deaths get recorded under whatever condition the patient had. This means national mortality statistics systematically undercount the problem.
Diagnostic Errors Are the Biggest Category
Misdiagnosis is the single largest contributor to serious harm from medical error. This includes missed diagnoses, delayed diagnoses, and wrong diagnoses. A 2023 analysis in BMJ Quality & Safety estimated that U.S. hospitals see roughly 2.4 million harmful diagnostic errors per year, with about 376,000 of those resulting in death or permanent disability comparable to losing a limb, an eye, or a kidney. Roughly 319,000 of those cases were considered preventable.
These errors happen across all care settings, not just hospitals. Missed diagnoses in outpatient clinics, where a doctor fails to order the right test or misreads symptoms during a routine visit, account for a significant share of the total. Cancer, infections, and vascular events like strokes are among the most commonly missed conditions.
Surgical Errors and “Never Events”
Some medical errors are so clearly preventable they’re called “never events,” things that should never happen under any circumstances. These include operating on the wrong body part, leaving a surgical instrument inside a patient, or performing the wrong procedure entirely. An estimated 4,000 or more surgical never events occur in the U.S. each year, though a typical hospital might see a wrong-site surgery only once every 5 to 10 years. While dramatic, these events represent a small fraction of overall medical error deaths. The larger surgical risks come from more routine complications: infections acquired during hospital stays, reactions to anesthesia, and errors in post-operative care.
The Global Picture
The World Health Organization estimates that about 1 in every 10 patients is harmed during medical care worldwide, with more than 3 million deaths annually attributed to unsafe care. The risk is not distributed equally. In low- and middle-income countries, as many as 4 in every 100 patients die from unsafe care, a rate driven by limited resources, understaffing, and gaps in training and infrastructure.
Globally, the WHO estimates 134 million adverse events occur each year in hospitals. These rates have remained stubbornly consistent from 2015 through 2024, with no significant reduction despite growing awareness of the problem. The COVID-19 pandemic made things worse, straining hospital systems and increasing rates of hospital-acquired infections, medication errors, and other preventable harms.
Has the Problem Gotten Better?
Not meaningfully. Data from high-income countries shows that roughly 10% to 12% of hospitalized patients experienced adverse events annually between 2015 and 2024. That rate held essentially flat across the entire decade. While individual hospitals and health systems have made progress with safety checklists, electronic prescribing systems, and better protocols, the overall numbers haven’t budged at a population level.
The pandemic set things back further. Overwhelmed hospitals saw spikes in preventable errors as staff managed impossible patient volumes with depleted resources. Projections for 2024 and beyond suggest that without major systemic improvements, adverse event rates and associated deaths will stay at their current levels.
The Financial Scale of Malpractice
Medical liability costs the U.S. healthcare system an estimated $55.6 billion per year, or about 2.4% of total healthcare spending. Only a fraction of that goes to malpractice claims and settlements. The vast majority, roughly $45.6 billion, is spent on defensive medicine: extra tests, procedures, and referrals that doctors order primarily to protect themselves from lawsuits rather than because the patient needs them. This means the ripple effects of medical error extend far beyond the patients directly harmed, shaping how medicine is practiced for everyone.
It’s worth noting that malpractice claims capture only a tiny slice of actual errors. Most medical errors never result in a lawsuit. Studies have consistently found that fewer than 2% of patients harmed by negligent care ever file a claim, meaning the legal system reflects the tip of a much larger iceberg.

