Anesthesia quality metrics hospital leaders should track

A subsidy negotiation tends to start with a spreadsheet: units, collections, staffed hours, the gap the hospital is asked to fill. Quality rarely gets the same discipline. It shows up as a sentence in the contract, "the group shall maintain high standards of care," with no measure attached and no data behind it.

That gap is a liability, not a convenience. A board that only sees cost numbers, and a group that only reports cost numbers, will optimize for cost. If your anesthesiologists, CRNAs and CAAs feel that pressure without a matching safety floor, the savings show up first in the budget and later in an incident report. A scorecard fixes the sequence: agree on what you are measuring, agree on a baseline, then talk about money.

What belongs on a scorecard

No single instrument covers anesthesia quality end to end. Several established sources supply pieces of it. Medicare's Merit-based Incentive Payment System (MIPS) includes an anesthesiology specialty measure set that most anesthesiology practices already report for their own payment.¹ The Anesthesia Quality Institute's National Anesthesia Clinical Outcomes Registry (NACOR) is the largest voluntary case-level registry in the specialty, now holding data from more than 100 million anesthesia cases submitted by participating practices and hospitals.² The American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) tracks 30-day outcomes, including airway complications, with risk and case-mix adjustment built into its comparisons.³ A hospital-side scorecard borrows from all three, plus the operational and staffing measures a perioperative team already tracks.

The table below groups a working set by what each one is, what it tells you, who actually controls the result, and whether it holds up as a payment term. That last column matters most. A measure that depends on surgeons, nursing and sterile processing as much as on anesthesia should inform a conversation. It should not set a bonus or a penalty for the anesthesia group alone.

Measure	What it tells you	Who controls it	In payment terms?
Perioperative hypothermia prevention	Whether a temperature of at least 35.5°C is documented near the end of surgeries under general or neuraxial anesthesia lasting 60 minutes or more, a MIPS outcome measure tied to infection, bleeding and cardiac risk4	Anesthesia clinician, intraoperatively	Yes
PONV prophylaxis for high-risk patients	Whether patients with three or more risk factors receive combination anti-emetic therapy before or during surgery, per SAMBA consensus guidance built into the MIPS measure5	Anesthesia clinician	Yes
Unplanned postoperative reintubation	A rare but serious airway event, roughly 0.14% to 0.19% of cases in one large cohort, and linked there to higher 30-day mortality6	Shared: anesthesia judgment, patient factors, case complexity	Only with case-mix adjustment
Patient-reported anesthesia experience	Survey questions on how the anesthesia team communicated and how well pain and anxiety were handled	Anesthesia clinician and system access to pre-op counseling	With enough case volume to be reliable
First-case on-time starts and turnover time	Whether staffed minutes convert into scheduled cases, using standard definitions	Shared: anesthesia, surgeon, nursing, sterile processing	Only the anesthesia-attributable share
Day-of-surgery cancellations for anesthesia reasons	Cases stopped for a medical or anesthetic finding; in one orthopedic service's review, medical and anesthesia factors accounted for about a fifth of same-day cancellations, behind patient no-shows but ahead of surgeon and clerical causes7	Mostly anesthesia, through the pre-op assessment process	Yes, if cause is documented case by case
PACU length of stay and OR holds	Whether recovery capacity keeps pace with the schedule; an OR hold occurs when a full PACU keeps the anesthesia team caring for a patient in the operating room, blocking the next case	Shared: nursing staffing, bed availability, discharge criteria	Rarely alone; watch alongside bed data
Add-on case response time	How fast an unscheduled case gets a room and a team	Shared, but staffing flexibility (a floater clinician, open capacity) is the group's lever	With a defined urgency scale
Vacancy rate by clinician role	Whether the group can staff its own coverage grid without emergency help	The group's recruiting and retention	No, use as a data request
Locum-covered hours as a share of scheduled hours	Dependency on temporary staffing, an early sign of the next subsidy request	The group	No, use as a data request
Voluntary turnover among group clinicians	Whether the staffing model is holding, across physicians, CRNAs and CAAs alike	The group	No, use as a data request
Registry participation and data completeness	Whether the group is measuring itself against peers at all	The group	Yes, as a condition of the contract
Timely, complete monthly reporting to the hospital	Whether the joint review actually has numbers to work from	The group	Yes, as a condition of the contract

A few of these need more than the table gives them.

The reintubation rate is the clearest case for caution. It is rare enough that a single bad month at a smaller hospital can look alarming and mean nothing, and common enough at a busy academic center that a raw count says little without adjusting for how sick and how complex the surgical population is. ACS NSQIP handles this with hierarchical modeling that adjusts for patient risk and procedure mix before comparing one hospital to the pooled average.³ Your own comparison does not need that machinery, but it needs the same instinct: read the trend, not the single point, and ask what changed in the patient population before you ask what changed in the anesthesia care.

The shared operational measures deserve the same instinct in the other direction. First-case starts, turnover, PACU holds and add-on response times move with staffing and scheduling decisions made well outside the anesthesia group's control. Our questions to ask before renewing an anesthesia contract covers the standard definitions for on-time starts and turnover, and how to divide accountability before you put either one in a contract. The rule that carries over here: track the shared measures jointly, and pay only for the portion the group actually controls.

Setting targets without cherry-picking

A target set from a national benchmark on day one invites an argument about whose benchmark is right. A target set from a baseline invites a conversation about whether performance improved. Pull two to four quarters of the hospital's own numbers, across the whole scorecard, before anyone proposes a number to hit. A group that offers to accept a target on its best measure while staying silent on the other twelve is not offering much.

Case mix will move some of these figures for reasons that have nothing to do with the anesthesia team. An older, sicker surgical population will show more PONV risk factors, more hypothermia risk and, at the margins, more airway events. None of the standard tools eliminate that effect entirely, and the physical status classification most anesthesia records already use as a shorthand for surgical risk is not, on its own, a predictor of operative risk: the same ASA class covers a skin lesion removed under local anesthetic and a pancreatectomy.⁸ Ask what adjustment, if any, sits behind a number before you compare it across services, sites or years.

Guardrails when you change cost inputs

A scorecard matters most at the moment a contract changes the ratio of anesthesiologists to CRNAs or CAAs, or narrows the coverage grid to cut cost. Both moves can be sound. Both need a matching safety check.

Ratio changes affect how many rooms one anesthesiologist can supervise at once, and supervision does not scale evenly. A review of one year of anesthesia records at a tertiary hospital found lapses in required physician presence on 35% of days even at a 1:2 supervision ratio, concentrated before 8 a.m., and found that tightening from 1:2 to 1:3 had a large effect on those lapses unless the schedule added a floater clinician or staggered the first-case start times.⁹ Our explainer on care team ratios covers how Medicare pays each model. Before locking in a ratio change, add a 90-day watch on first-case supervision lapses and on-time starts, and set the threshold that triggers a review before volume grows into the gap.

Coverage credits carry a subtler risk. A bonus tied purely to turnover speed, room utilization or add-on response time rewards moving faster, and moving faster is not free. Pair any efficiency incentive with a floor on the safety measures in the table above: a turnover bonus should not pay out in a quarter where hypothermia prevention, PONV prophylaxis or reintubation rates moved the wrong way. It is the same principle as the joint operating rhythm in our guide to working with a difficult anesthesia group: one set of numbers, reviewed together, before either side asks the other for more.

What the scorecard is for

A scorecard is not a cost report with extra rows. It is a quality record that happens to have a cost attached, built from measures the group can see coming, updated on a schedule both sides agree to, and reviewed before either side proposes a change to ratios, coverage or pay. That order, quality first, cost second, is the difference between a contract renewal grounded in evidence and one that reads as an exercise in the same subsidy math with a new coat of paint. Our anesthesia performance monitoring work builds this kind of scorecard from a hospital's own case data, before the next renewal puts it to the test.

Sources

1. American Society of Anesthesiologists, "MIPS Quality Performance Category," ASA Quality Payment Program resources, January 2026. https://www.asahq.org/advocating-for-you/qpp/quality ↩

2. American Society of Anesthesiologists, "NACOR," Anesthesia Quality Institute Registries, September 2026. https://www.asahq.org/aqi/registries/nacor ↩

3. American College of Surgeons, "ACS NSQIP," ACS Quality Programs, undated. https://www.facs.org/quality-programs/data-and-registries/acs-nsqip/ ↩ ↩²

4. Centers for Medicare & Medicaid Services and American Society of Anesthesiologists, "Quality ID #424: Perioperative Temperature Management," MIPS Clinical Quality Measure specification, version 9.0, December 2024. https://qpp.cms.gov/docs/QPP_quality_measure_specifications/CQM-Measures/2025_Measure_424_MIPSCQM.pdf ↩

5. Centers for Medicare & Medicaid Services and American Society of Anesthesiologists, "Quality ID #430: Prevention of Post-Operative Nausea and Vomiting (PONV) – Combination Therapy," MIPS Clinical Quality Measure specification, version 9.0, December 2024. https://qpp.cms.gov/docs/QPP_quality_measure_specifications/CQM-Measures/2025_Measure_430_MIPSCQM.pdf ↩

6. Sofjan I, et al., "Risk Factors for Postoperative Unplanned Reintubation in a Cohort of Patients Undergoing General Anesthesia," Cureus, May 2023. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10257981/ ↩

7. Bernstein M, Hall L, Jebeles G, Kelly T, Shah A, "Trends in Patient-Related Factors Associated with Same-Day Surgical Cancellations at an Urban Medical Center," Foot & Ankle Orthopaedics, December 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12744115/ ↩

8. Fitz-Henry J, "The ASA classification and peri-operative risk," Annals of The Royal College of Surgeons of England, April 2011. https://pmc.ncbi.nlm.nih.gov/articles/PMC3348554/ ↩

9. Epstein RH, Dexter F, "Influence of Supervision Ratios by Anesthesiologists on First-Case Starts and Critical Portions of Anesthetics," Anesthesiology, March 2012. https://pubmed.ncbi.nlm.nih.gov/22297567/ ↩