Guide · Choosing a hospital

How to Choose a Hospital with Data

A practical framework for evaluating hospitals using CMS quality data.

The short answer

For an emergency, go to the nearest hospital; for planned care, compare the measures that fit your procedure, infection rates for surgery, mortality for cardiac care, maternal measures for childbirth, and favor hospitals that do more of it.

By the numbers

What the CMS data shows

5,426
Hospitals tracked
3.1 / 5
Avg CMS star rating
288 · 10%
Five-star hospitals

National CMS star-rating distribution

How all 2,866 rated Medicare hospitals break down across the 1–5 star scale

hospitals

What this shows Most hospitals cluster in the 3–4 star range; only 10% reach five stars and a similar share sit at one star, the headline rating is a relative ranking, not a pass/fail mark.

Source CMS Hospital Compare, overall hospital rating As of 2026

According to the Centers for Medicare & Medicaid Services (CMS), these figures cover 5,426 Medicare-certified hospitals as of March 2026, read our methodology for how the data is compiled.

Hospital selection here draws on Centers for Medicare & Medicaid Services (CMS) Hospital Compare data for 5,426 Medicare-certified hospitals, refreshed March 2026, see our methodology.

Emergency vs. Planned Care

The hospital selection process differs completely based on urgency:

  • Emergency care: Go to the nearest hospital. For heart attacks, strokes, and trauma, time is the most critical factor. Paramedics will transport you to the nearest appropriate facility. There is no time for quality comparison.
  • Planned procedures: You have days, weeks, or months to research. Use PlainHospital's quality data to compare options. For elective surgery, childbirth, and scheduled procedures, choosing a higher-quality hospital is worth the extra effort.

Step 1: Identify Your Options

Start by listing hospitals that meet your practical requirements:

  1. In-network: Check your insurance plan's hospital network. Out-of-network care can cost 2-3x more.
  2. Within reasonable distance: How far are you willing to travel? For routine surgery, 30-60 minutes is common. For complex/rare procedures, patients often travel hours or across state lines.
  3. Offers your procedure: Not all hospitals perform all procedures. Verify the hospital has the department and specialists you need.

Browse hospitals by state on PlainHospital to see what's in your area.

Step 2: Compare Relevant Quality Measures

Once you have 2-4 hospital options, compare the quality measures most relevant to your situation:

  • Any surgery: Infection rates (CLABSI, SSI), patient safety indicators, and procedure-specific complication rates.
  • Heart conditions: Heart attack mortality, heart failure readmission, and timely treatment measures.
  • Childbirth: C-section rates, maternal health complications, and newborn outcomes. Patient experience in maternity is also valuable.
  • Joint replacement: Complication rates, readmission rates, and patient experience.
  • General hospitalization: Overall safety scores, readmission rates, and patient experience.

Each hospital page on PlainHospital shows all available quality measures in one view.

Step 3: Check Volume

Hospital volume, how many of your specific procedure they perform, is one of the strongest predictors of outcomes. Research shows:

  • High-volume hospitals have 25-50% lower mortality rates for complex surgeries compared to low-volume hospitals.
  • The volume-outcome relationship is strongest for complex procedures: cardiac surgery, cancer operations, organ transplants.
  • For common procedures (knee replacement, appendectomy), the relationship is weaker but still exists.

CMS doesn't always publish procedure volume publicly, so ask the hospital directly: "How many [your procedure] do you perform per year?"

How procedure volume changes outcomes

Volume is one of the most consistent predictors of hospital outcomes, high-volume centers learn from repetition, refine workflows, and build specialized teams. Studies of cardiac surgery show 30-day mortality dropping from roughly 4.5% at low-volume hospitals to around 2.1% at high-volume centers performing 200+ cases per year.

Pancreatic surgery shows an even sharper gap: complication rates can fall from 25% at low-volume facilities to 12% at high-volume cancer centers. The pattern repeats across complex surgical specialties.

Volume vs. Outcomes, Selected Procedures
Procedure Low-Volume High-Volume Outcome Measure
Coronary Artery Bypass ~4.5% ~2.1% 30-day mortality
Pancreatic Resection ~25% ~12% Major complications
Total Hip Replacement ~3.2% ~1.1% 90-day readmission
Bariatric Surgery ~6.0% ~2.8% Serious complications

Star ratings vs. specific quality measures

The CMS overall star rating compresses dozens of measures into a single number. That makes it easy to scan but obscures procedure-specific performance. A 3-star hospital may have excellent cardiac care but weaker maternity outcomes, and vice versa. For planned procedures, drill into the specific quality measures that match your situation rather than relying on the overall star.

Reading infection benchmarks correctly

Infection measures use a Standardized Infection Ratio (SIR). A SIR of 1.0 means infections occurred at the predicted national rate. A SIR of 0.5 means roughly half the predicted rate (better). A SIR of 1.5 means 50% above predicted (worse). Confidence intervals matter, small hospitals with wide intervals may not actually differ from average even if the point estimate looks alarming.

When patient experience scores diverge from clinical scores

HCAHPS patient-experience scores measure communication, responsiveness, and environment. They correlate weakly with mortality and infection rates. A hospital may score high on patient satisfaction yet have worse-than-average safety outcomes, or vice versa. Both matter, but they answer different questions: experience reflects how the stay feels; clinical scores reflect whether you survive and recover.

Worked example: comparing two hospitals for joint replacement

Suppose you're choosing between Hospital A (community, 4-star overall, performs ~150 joint replacements/year) and Hospital B (academic, 3-star overall, performs ~850 joint replacements/year). Hospital A has a 4.2% complication rate vs Hospital B's 2.3%. Hospital A's median Medicare payment is $18,500 vs Hospital B's $22,400.

The headline star rating favors A, but the procedure-specific data favors B: the volume difference (~150 vs ~850) drives a complication-rate gap of 4.2% vs 2.3% - nearly double the risk at the higher-rated facility for this specific procedure. The $3,900 cost gap is real but small relative to the outcome gap. Unless travel logistics or insurance dictate otherwise, the data favors Hospital B.

Step 4: Ask Your Doctor

Data provides the framework; your doctor provides clinical judgment. Ask your physician:

  • Which hospital do you recommend for my condition, and why?
  • Do you have privileges at the hospitals I'm considering?
  • How many times have you performed this procedure, and what are your outcomes?
  • Are there specialist referrals that would change the hospital choice?

A doctor's recommendation combined with PlainHospital's quality data gives you the most complete picture for an informed decision.

Frequently Asked Questions

Does my insurance limit which hospitals I can use?

Yes. Most insurance plans have a network of preferred hospitals. Using an out-of-network hospital can cost significantly more, sometimes 2-3x. Check your insurance plan's provider directory before evaluating hospitals on quality. The best hospital in the data may not be the best choice if it's out of network.

Should I choose the closest hospital or the best-rated one?

For emergencies, go to the nearest hospital, minutes matter for heart attacks, strokes, and trauma. For planned procedures (elective surgery, childbirth), you have time to compare quality data and choose a better-rated facility even if it's farther away. The extra drive time is worth it for a hospital with significantly better outcomes.

How important is hospital volume for my procedure?

Very. Research consistently shows that hospitals (and surgeons) performing more of a given procedure have better outcomes. This is especially true for complex surgeries, cardiac surgery, cancer operations, and organ transplants. Ask the hospital how many of your specific procedure they perform annually. Higher volume generally means better results.

What about teaching hospitals vs. community hospitals?

Teaching hospitals (affiliated with medical schools) typically have more specialists, more advanced technology, and handle more complex cases. Community hospitals may offer a more personalized experience, shorter wait times, and are often closer to home. For routine care, community hospitals are usually fine. For complex or rare conditions, teaching hospitals often have better expertise.

Can I check a specific doctor's quality at a hospital?

CMS Hospital Compare data is at the hospital level, not the individual physician level. For surgeon-specific outcomes, ask the surgeon directly about their volume and complication rates. Some states publish surgeon-level data for cardiac surgery and other high-risk procedures. PlainHospital focuses on facility-level data.

What if my area only has one hospital?

Many rural areas have a single hospital option. In this case, use PlainHospital data to understand its strengths and weaknesses, not to choose an alternative, but to be an informed patient. Know which quality measures your hospital excels at and where it falls short. For planned procedures where quality matters and you can travel, consider hospitals in the nearest metro area.

Sources

  • CMS, Hospital Compare
  • Agency for Healthcare Research and Quality, Guide to Choosing a Hospital

This content is for informational purposes only and does not constitute medical advice. Always consult your healthcare provider and insurance plan when choosing a hospital.

Every figure on PlainHospital is rendered directly from federal source data, no number is typed in by an editor. This page draws directly on federal source data, no figure is typed in by an editor. See our editorial standards & corrections policy, the methodology behind these numbers, or report a data error.