You’re three weeks out from surgery. You looked up where your surgeon went to medical school, read a page of patient reviews, and confirmed the hospital has a decent reputation around town. That felt like doing your homework.
The question you probably didn’t ask: how many of these operations does this surgeon do in a year, and how many does this hospital do?
That one question has more research behind it than almost anything else you can ask before a procedure. Researchers call it the volume-outcome relationship. In plain English: hospitals and clinicians who perform a procedure often tend to get better results than those who perform it rarely.
It’s also easy to oversell. The link is strong for some operations, thin for others, and missing for a few. And “experience” turns out to mean something narrower than most people assume.
What the volume research shows
Start at the strong end of the evidence, which is complex abdominal surgery.
A 2023 systematic review and meta-analysis in the journal HPB pooled studies on pancreatic surgery published from 2000 to 2018. Patients treated at high-volume hospitals had roughly a third the odds of dying after the operation compared with patients at low-volume hospitals, an odds ratio of 0.35. Major complications were lower too, but by a much smaller margin, an odds ratio of 0.87. For individual surgeons, the mortality odds ratio was 0.29.
Numbers like that get misread constantly, so this is the translation. An odds ratio of 0.35 does not mean you have a two-thirds chance of dying at the other hospital. It means whatever your personal baseline risk is, it drops by roughly that proportion at the busier place. For an operation with a 1-in-20 death rate, that shift is large and worth driving for. For a procedure with a 1-in-5,000 death rate, the same ratio barely moves your odds at all.
A 2021 meta-analysis in the ANZ Journal of Surgery narrowed in on pancreatoduodenectomy, the operation most people know as the Whipple. Across 17 studies, in-hospital mortality was 63% lower at hospitals above the high-volume line, and the finding held no matter where researchers drew that line or which country supplied the data. The median cutoff used to define “high volume” across those studies was 25 of these operations a year. That’s one roughly every other week.
A 2026 dose-response meta-analysis in Annals of Vascular Surgery tried a smarter design on elective abdominal aortic aneurysm repair, pooling 11 observational studies covering more than 200,000 patients. Instead of sorting hospitals into two buckets, the authors modeled volume as a continuous curve. Hospitals at the 90th percentile of volume had 35% lower odds of death than hospitals at the 10th percentile.
That analysis also surfaced something more revealing than the mortality figure. Higher-volume hospitals did better on failure to rescue after open repair. Failure to rescue is the share of patients who develop a serious complication and then die from it. It isn’t about avoiding complications. It’s about catching them. A busy team spots a slow internal bleed at 3 a.m. faster, has the right specialist already in the building, and has pulled off that particular save before. That mechanism, more than any surgical flourish, is probably what a large share of the volume effect is made of.
Where the effect gets thin, and where it vanishes
Now the other half of the picture, which rarely makes it into hospital marketing.
An Italian national evidence overview published in Epidemiologia e Prevenzione in 2013 did something unusually useful: it reviewed 47 systematic reviews covering 38 clinical areas and sorted them by what the evidence supported. For 26 conditions and procedures, including coronary bypass, carotid surgery, hip fracture, neonatal intensive care, and most cancer operations, the reviews showed a positive volume-outcome association for in-hospital or 30-day death. For two, hip replacement and rectal cancer surgery, the reviews showed no association. For ten more, including appendectomy, hysterectomy, inguinal hernia repair, and trauma, the evidence simply wasn’t good enough to say.
That last group matters. “No proven volume effect” is not the same as “volume doesn’t matter here.” It usually means nobody has studied it well.
Cardiology shows how the picture can shift over time. A 2021 systematic review in HRB Open Research pulled 22 observational studies on percutaneous coronary intervention, the stent procedure used to open blocked heart arteries, covering more than 6.4 million patients. Total hospital PCI volume showed no significant association with mortality. Operator volume did, and so did hospital volume for emergency PCI during a heart attack. The authors also noted a trend from significant to non-significant results as studies got more recent, and they graded the overall certainty of the evidence as very low.
Emergency general surgery splits the same way. A 2024 meta-analysis in the journal Surgery looked at urgent intra-abdominal operations like cholecystectomy, colectomy, and bowel resection across more than 2.3 million patients. Low hospital volume was not significantly tied to higher complication rates in the main analysis. Low surgeon volume was, and that finding held up when the authors tested it.
And sometimes the question hasn’t been asked at all. A 2022 systematic review in the Journal of Pediatric Surgery on gastroschisis, a birth defect where an infant’s intestines develop outside the abdominal wall, found 12 studies on hospital volume and zero on surgeon volume. The evidence on in-hospital death was rated low certainty. Everything else was very low.
Hospital volume and surgeon volume are two different things
Notice what kept happening in those studies. Sometimes the hospital number mattered and the surgeon number didn’t. Sometimes it was the reverse. For pancreatic surgery, both did.
They are separate variables because they describe separate things. Surgeon volume is about hands and judgment inside the operating room. Hospital volume is about everything around it: the anesthesia team, the ICU nurses, the blood bank, how fast interventional radiology answers a page at midnight, whether anyone has seen this complication before.
So a high-volume surgeon at a hospital that almost never does your operation is a real situation, and it’s not obviously safe. Neither is a famous hospital where your specific surgeon has done the procedure eleven times. Ask about both. They’re different answers.
Why “how long have you been doing this” is the wrong question
Here is the common belief worth correcting. Most people treat years in practice as a proxy for skill. The evidence runs the other way.
A systematic review in Annals of Internal Medicine in 2005 collected 62 separate evaluations linking physician age or years since graduation to knowledge and quality of care. Thirty-two of them, 52%, found performance declined as years in practice increased, across every outcome measured. Another 13 found declines on some measures and no relationship on others. Thirteen found no association at all. Exactly one found consistent improvement with experience. The authors concluded that physicians who have been in practice longer “may be at risk for providing lower-quality care.”
That sounds harsh, and it needs context. The finding is about adherence to current standards, not about wisdom. Medicine changes. A clinician trained in 1994 learned a set of practices, and keeping up with twenty-five years of revisions is real work that not everyone does. Judgment built over decades genuinely helps with rare presentations and messy cases where guidelines run out.
The honest version is this: seniority is not a quality signal on its own, and recency is. A surgeon who did 80 of your operation last year is a better bet than one who did 400 of them across a career that peaked in 2011.
Does doing more make you better, or do the better ones get sent more?
This is the caveat that should sit next to every number in this article.
Almost all volume-outcome research is observational. Nobody randomly assigns patients to a low-volume surgeon, and nobody ever will. So when busier surgeons get better results, there are two competing explanations. Practice makes perfect: repetition builds skill and builds a team around it. Or selective referral: skilled clinicians get a reputation, referrals pile up, and their volume is a result of being good rather than a cause of it.
Both are probably true at once, in proportions that differ by procedure. But a 2026 study in the journal Health Policy found a clever way to pry them apart. The researchers studied hip fracture surgery across public hospitals in Italy’s Emilia-Romagna region from 2017 to 2019. Hip fracture is an emergency, and in that system patients go to the nearest hospital, so reputation-driven referral barely operates. Higher surgical volume in prior periods was still linked to lower 30-day mortality, and the association got stronger the further back the volume was measured. That pattern fits learning by doing, not reputation.
One useful piece of history: the volume threshold idea entered American hospital policy through the Leapfrog Group, a coalition of large employers and purchasers that proposed steering complex procedures to high-volume hospitals and set volume minimums for five of them. A 2006 analysis in the journal Biometrics examined the statistical logic underneath that hypothesis using esophagectomy data, and concluded the relationship is nonlinear, which is a reminder that a hard cutoff is a policy convenience, not a biological fact.
Where you can look up numbers, and where you can’t
This is where most advice articles get vague. This is the real answer.
The federal government publishes some of this. According to the Centers for Medicare & Medicaid Services, clinician profile pages in Medicare’s Care Compare tool include use data for 19 procedures, among them hip replacement, cataract surgery, and coronary artery bypass. Those counts come from Original Medicare and Medicare Advantage claims over a recent 12-month period. The same profiles list quality program scores, hospital and facility affiliations, and whether the clinician offers telehealth.
Now the limits, stated plainly. Nineteen procedures is a short list. The counts reflect Medicare patients only, so a surgeon whose practice skews younger will look far less busy than they are. There is no national public registry of procedure volume for most operations, for most specialties, or for non-surgical care at all. For the overwhelming majority of what clinicians do, the number exists only inside the practice, and the only way to get it is to ask.
Two other federal lookups are worth knowing. The first is Open Payments, run by the Centers for Medicare & Medicaid Services. Drug and device companies report to it what they pay physicians and certain other clinicians, plus any ownership interests, and the public can read all of it. Just the 2025 program year accounts for $14.67 billion spread across 17.07 million records. What you see there is what the companies themselves attested to, and CMS makes a point of not calling any single relationship a conflict. Read it as context, not a verdict. And the Federation of State Medical Boards maintains a public directory of every state medical board, which is where license status and disciplinary history live. State licensure and credential decoding are their own subjects, covered separately in this library.
Three questions that get a real answer
Most patients never ask about volume because they’re afraid of sounding rude or distrustful. The reframe that fixes it: you’re not auditing anyone. You’re gathering the same information the surgeon would gather if the patient were their own parent.
Ask these three, in this order, and use the actual wording.
1. “How many of these do you do in a typical year, and how many does this hospital do?”
Two numbers, one question. A good answer is specific and comes quickly, because clinicians who do a lot of something know their own numbers. A vague answer, a pivot to total career cases, or visible irritation is itself information. Watch for the career-total dodge. “I’ve done hundreds over the years” answers a different question than the one you asked.
2. “What’s your complication rate for this, and how do you track it?”
The second half matters more than the first. Surgeons in registry-participating programs can usually describe how their outcomes are measured and compared. Someone who has never looked is telling you something. You’re not expecting a perfect record. You’re checking whether anyone is counting.
3. “If you couldn’t do this operation, who would you send me to?”
This is the one that works. It’s not adversarial, it gives the clinician an easy out, and the answer is genuinely useful either way. Clinicians know who in their region does the hard cases. If they name someone instantly and without ego, you’ve learned something good about them too.
For non-surgical care, swap in the right denominator: “How many patients with this specific condition are you treating right now?” A rheumatologist managing nine people with your rare autoimmune disease is a different proposition from one who sees it twice a year.
When experience is not the thing to optimize
Chasing volume has a cost, and it is not always worth paying.
For time-critical conditions, distance kills faster than inexperience does. Stroke, heart attack, sepsis, major trauma, a bowel obstruction with signs of strangulation, a ruptured aneurysm: in those situations the nearest capable facility is the right facility. The research on emergency PCI supports going fast to a hospital that does it, not far to a hospital that does it more. Do not drive past an emergency room to reach a better one.
Travel has other real costs too. Recovery far from home means your family can’t visit, your follow-up gets fragmented, and complications land in the lap of a local clinician who wasn’t in the operating room. For a high-risk, elective, complex procedure, that trade is usually worth making. For a common operation with good outcomes everywhere, it usually isn’t.
Some signs mean the issue isn’t volume at all, and you should get a licensed clinician involved promptly rather than shopping for a busier one. You’re getting worse on a treatment plan and no one has changed it. A diagnosis doesn’t explain your symptoms. You’ve had the same procedure twice with the same result. Someone outside your prescribing team has told you to stop a prescribed medication, which is never advice to follow without calling the prescriber first.
The one thing to take away
If you remember a single sentence from this article, make it this: ask about recent, repeated experience with your specific problem, not about years in practice and not about reputation.
Volume raises the floor. It does not guarantee the ceiling, it matters far more for complex procedures than common ones, and the evidence behind it is observational, which means it describes a pattern rather than a promise. A busy surgeon can still have a bad day, and a modest-volume surgeon at a well-run hospital can be the right choice.
But you will learn more in ninety seconds of asking those three questions than in an hour of reading star ratings. Ask them out loud, write down the answers, and notice how the person responds. The reaction is part of the data.
Keep exploring: abnormal lab result, doctor not listening and lower medical bills.
Sources
- Centers for Medicare & Medicaid Services. “Physician Compare Initiative.” cms.gov
- Centers for Medicare & Medicaid Services. “Medicare Care Compare.” medicare.gov
- Centers for Medicare & Medicaid Services. “Open Payments.” cms.gov
- Federation of State Medical Boards. “About Physician Licensure.” fsmb.org
- Federation of State Medical Boards. “Contact a State Medical Board.” fsmb.org
- National Center for Complementary and Integrative Health. “6 Things To Know When Selecting a Complementary Health Practitioner.” nccih.nih.gov
- HPB (via PubMed). “Fischer C, et al. "The relationship of hospital and surgeon volume indicators and post-operative outcomes in pancreatic surgery: a systematic literature review, meta-analysis and guidance for valid outcome assessment." HPB (Oxford), 2023. PubMed. DOI: 10.1016/j.hpb.2023.01.008.” doi.org
- ANZ Journal of Surgery (via PubMed). “Kovoor JG, et al. "In-hospital survival after pancreatoduodenectomy is greater in high-volume hospitals versus lower-volume hospitals: a meta-analysis." ANZ Journal of Surgery, 2021. PubMed. DOI: 10.1111/ans.17293.” doi.org
- Annals of Vascular Surgery (via PubMed). “Coutinho ML, et al. "Hospital Volume and Outcomes after Elective Abdominal Aortic Aneurysm Repair: A Systematic Review and Dose-Response Meta-Analysis." Annals of Vascular Surgery, 2026. PubMed. DOI: 10.1016/j.avsg.2026.06.025.” doi.org
- HRB Open Research (via PubMed). “Walsh KA, et al. "The relationship between procedural volume and patient outcomes for percutaneous coronary interventions: a systematic review and meta-analysis." HRB Open Research, 2021. PubMed. DOI: 10.12688/hrbopenres.13203.1.” doi.org
- Surgery (via PubMed). “Rafaqat W, et al. "Does practice make perfect? The impact of hospital and surgeon volume on complications after intra-abdominal procedures." Surgery, 2024. PubMed. DOI: 10.1016/j.surg.2024.01.011.” doi.org
- Journal of Pediatric Surgery (via PubMed). “Morche J, et al. "Relationship between volume and outcome for gastroschisis: A systematic review." Journal of Pediatric Surgery, 2022. PubMed. DOI: 10.1016/j.jpedsurg.2022.03.022.” doi.org
- Annals of Internal Medicine (via PubMed). “Choudhry NK, Fletcher RH, Soumerai SB. "Systematic review: the relationship between clinical experience and quality of health care." Annals of Internal Medicine, 2005. PubMed. DOI: 10.7326/0003-4819-142-4-200502150-00008.” doi.org
- Health Policy (via PubMed). “Paktinat S, Ugolini C, Verzulli R. "Does practice make perfect? Panel evidence from hip fracture surgery in Italy." Health Policy, 2026. PubMed. DOI: 10.1016/j.healthpol.2026.105703.” doi.org
- Epidemiologia e Prevenzione (via PubMed). “Amato L, et al. "Volume and health outcomes: evidence from systematic reviews and from evaluation of Italian hospital data." Epidemiologia e Prevenzione, 2013. PubMed. PMID: 23851286 (no DOI in PubMed record).” pubmed.ncbi.nlm.nih.gov
- Biometrics (via PubMed). “Betensky RA, et al. "Hospital volume versus outcome: an unusual example of bivariate association." Biometrics, 2006. PubMed. DOI: 10.1111/j.1541-0420.2005.00449.x.” doi.org
- Interactive Cardiovascular and Thoracic Surgery (via PubMed). “Sepehripour AH, Athanasiou T. "Is there a surgeon or hospital volume-outcome relationship in off-pump coronary artery bypass surgery?" Interactive Cardiovascular and Thoracic Surgery, 2012. PubMed. DOI: 10.1093/icvts/ivs448.” doi.org