The Autopsy / Case File · Medical Error & Preventable Care
The Closed Claims Project
There is no CDC surveillance system for anesthesia deaths. No National Vital Statistics category tracks them as their own cause. No federal registry logs which drug, which monitor gap, which airway complication actually killed the patient. The most detailed record of how modern anesthesia kills the people it's supposed to protect was built by the industry's own malpractice insurers — because nobody else was going to build it.
The American Society of Anesthesiologists' Closed Claims Project has spent four decades pulling case files from 35 professional liability insurers covering roughly half of America's practicing anesthesiologists, reconstructing thousands of anesthesia-related deaths and severe injuries one closed lawsuit at a time. It has found that claims for death or brain damage fell steadily as monitoring technology spread through the 1980s and '90s — and that today, the leading mechanism behind the worst outcomes isn't the operating room at all. It's a patient breathing too slowly, hours after surgery, with nobody watching closely enough to notice.
Join the discussion →
Anesthesia is one of the few areas of medicine where death is an accepted, if vanishingly small, possibility built into an otherwise elective procedure. Measuring exactly how small, and exactly what causes it when it happens, turns out to be a surprisingly incomplete project for American medicine. Deaths that occur under or shortly after anesthesia get coded, on a death certificate, to whatever underlying condition or surgical complication seems most relevant — not to "anesthesia" as a discrete, trackable cause. The result is a real and measurable hazard with no dedicated national dataset behind it, tracked instead by the one institution with both the case files and the financial incentive to understand exactly how it happens: the malpractice insurance industry.
From One in a Thousand to Two in a Hundred Thousand
The scale of the historical improvement is not in dispute, even without a federal tracking system. In the 1940s, roughly 1 in every 1,000 anesthetic administrations resulted in a patient's death. A widely cited 1954 study put the rate at 64 deaths per 100,000 procedures. By the 1980s, the rate had fallen to roughly 1 in 10,000 — still, by modern standards, a startling number of avoidable deaths, and the backdrop against which the ASA Closed Claims Project was founded in 1985. More recent data, covering 2011 through 2016, estimated mortality attributable to anesthesia administration at approximately 2.3 per 100,000 patients, with the risk concentrated heavily among patients carrying more significant underlying health conditions rather than spread evenly across the surgical population.
What drove that decline is, by contrast, a matter of real dispute even among the researchers who study it closely. The conventional story credits the rise of pulse oximetry and end-tidal capnography — continuous, automated monitoring of a patient's blood oxygen and exhaled carbon dioxide — as basic, mandated standards of care starting in the 1980s. The Closed Claims Project's own data complicates that story: an analysis of claims from 1975 to 2000 found that death and brain-damage claims decreased steadily over the period, but that the decline did not track cleanly with the spread of pulse-oximetry and capnography monitoring in the same claims. The truer answer is probably not one intervention but several arriving together — monitoring technology, but also better-designed anesthesia machines, revised drug protocols, structured crisis-response training, and the founding of the Anesthesia Patient Safety Foundation itself in 1985, the same year the Closed Claims Project began pulling case files.
What the Case Files Actually Show
The Closed Claims Project doesn't just count deaths — it reconstructs, claim by claim, what specifically went wrong. Reviewers with clinical anesthesia training examine each closed malpractice file and code the "damaging event": the specific mechanical, pharmacological, or judgment failure that produced the injury. Over the database's first quarter-century, the balance between two damaging-event categories shifted in a way that itself tells a story. Respiratory events — a blocked or misplaced airway, an unrecognized esophageal intubation, inadequate ventilation — dominated the early claims data. Cardiovascular events, by contrast, climbed steadily as a share of the total. By 1992, the two categories had essentially converged, each accounting for roughly 28 percent of claims, evidence that better airway management and monitoring had closed some of the gap even as anesthesia was being used on older, sicker patients whose hearts carried more of the underlying risk.
Nobody files a death certificate that says "anesthesia." The only institution that has spent forty years reconstructing exactly how these deaths happen is the industry's own malpractice insurers, working backward from a lawsuit.
One of the database's more specific and unsettling findings concerns neuraxial anesthesia — epidurals and spinals — when it triggers a cardiac arrest. Cases coded as neuraxial cardiac arrest in the Closed Claims database are overwhelmingly high-severity: more than 90 percent end in the patient's death or permanent brain damage. That single statistic has done real work in clinical anesthesia training, reinforcing that a cardiac arrest during a neuraxial block is treated as a full-scale emergency requiring immediate, aggressive resuscitation rather than a routine complication to be managed conservatively — a distinction the closed-claims data made starkly clear in a way no single case report could.
The Hazard Moved Out of the Operating Room
The database's most consequential recent finding is about where the danger has shifted, not just how often it occurs. A closed-claims analysis of postoperative opioid-induced respiratory depression — reviewing 357 acute pain claims and identifying 92 cases of likely opioid-related respiratory depression among them — found that 77 percent of those 92 cases ended in severe brain damage or death. Eighty-eight percent of the respiratory-depression events occurred within 24 hours of surgery, meaning the anesthesiologist's direct involvement had typically already ended; the patient was recovering on a hospital floor, on standard postoperative opioid dosing, when their breathing slowed and nobody caught it in time. Ninety-seven percent of the reviewed cases were judged preventable with better monitoring and a faster response — not an unavoidable complication of pain management, in the reviewers' own clinical judgment, but a systems failure that better postoperative surveillance would have caught.
A parallel pattern shows up in ambulatory and office-based anesthesia, a fast-growing segment of American surgical and procedural care that happens outside a traditional hospital operating room — endoscopy suites, dental offices, outpatient procedure centers. Closed-claims analyses of monitored anesthesia care in these settings have found that respiratory depression following sedatives or opioids is the single most common damaging mechanism, and that oversedation paired with inadequate monitoring makes up the majority of claims arising from anesthesia delivered outside the operating room altogether. As more procedures move to lower-acuity, lower-oversight settings — often explicitly to save cost — the closed-claims data suggests the same respiratory-depression mechanism that anesthesiologists spent decades engineering out of the operating room has simply resurfaced somewhere the monitoring infrastructure hasn't caught up yet.
The database has done more than describe the problem after the fact — it has directly shaped how anesthesia is taught and practiced. The ASA's Practice Guidelines for Management of the Difficult Airway, first published in 1993 and revised repeatedly since, drew explicitly on Closed Claims Project airway-injury data to build a structured, step-by-step algorithm anesthesiologists now train on before they ever face a real difficult intubation. The ASA's Basic Standards for Anesthetic Monitoring, adopted in 1986 and expanded since, emerged from the same era of claims review that first identified how often a preventable monitoring gap, not an unavoidable complication, was the actual mechanism behind a bad outcome. Neither guideline exists because a federal regulator required it. Both exist because a specialty society's own malpractice data made the pattern impossible to keep ignoring.
A Registry Built Because No One Else Would
What makes the Closed Claims Project unusual in the landscape of American patient-safety data is its origin. It wasn't built by a federal health agency responding to a public health mandate; it was built by a specialty medical society, using access only a malpractice insurer's cooperation could provide, specifically because no government surveillance system existed to answer the question the specialty itself needed answered: what is actually killing our patients, and how do we stop it? Four decades in, it remains the most detailed source of mechanism-level data on anesthesia mortality in the country — not despite being built outside the federal public-health system, but because nothing inside it ever stepped up to do the job.
And Then You're Dead
The operating room got safer because forty years of lawsuits got read, one by one, by people trying to find the pattern. The hazard didn't disappear. It moved down the hall, onto the recovery floor, to the hours after the anesthesiologist stopped watching — the one place the same discipline hasn't fully caught up yet.
That's it. That's the whole thing.
Read about the alarm that gets lost in the noise · or see every case file
Sources
- Historical anesthesia mortality rates (1 in 1,000 in the 1940s; 64 per 100,000 in a 1954 study; roughly 1 in 10,000 by the 1980s; 2.3 per 100,000 for 2011-2016) — "Progress in Patient Safety in Anesthesia," PMC/NIH, and cited historical anesthesia mortality literature synthesized therein
- The ASA Closed Claims Project's founding in 1985, its structure (35 participating professional liability insurers, covering roughly 50% of practicing U.S. anesthesiologists), and the 1975-2000 finding that death/brain-damage claims declined without a matching rise in documented pulse-oximetry/capnography use — American Society of Anesthesiologists Closed Claims Project database documentation; "The Closed Claims Project: Has It Influenced Anesthetic Practice and Outcome?"
- The convergence of respiratory and cardiovascular damaging-event claims to roughly 28% each by 1992, and the 90%+ death/brain-damage rate in neuraxial cardiac arrest claims — Anesthesia Patient Safety Foundation, "Closed Claims Project Shows Safety Evolution"; ASA Closed Claims Project topic analyses on neuraxial complications
- The 357-claim postoperative opioid-induced respiratory depression analysis (92 likely cases, 77% severe brain damage or death, 88% within 24 hours of surgery, 97% judged preventable) — "Postoperative Opioid-induced Respiratory Depression: A Closed Claims Analysis," Anesthesiology, ASA Closed Claims Project
- Respiratory depression as the leading damaging mechanism in monitored anesthesia care and non-operating-room anesthesia (NORA) claims, tied to oversedation and inadequate monitoring outside the traditional operating room — closed-claims analyses of ambulatory/office-based and NORA anesthesia claims, PMC/NIH literature review
- The ASA Standards for Basic Anesthetic Monitoring (adopted 1986) and the ASA Practice Guidelines for Management of the Difficult Airway (first published 1993, developed citing Closed Claims Project airway-injury data) — American Society of Anesthesiologists Standards, Guidelines and Statements
- Photo: U.S. Navy, "Moustaoha Abdilldhi, a local anesthesia nurse, and Lt. Heather Yurka, Camp Lemonnier anesthesiologist, set up an anesthesiology machine," Peltier General Hospital, Djibouti, February 2012, via Wikimedia Commons (public domain).
Comments
Loading…