Search any surgeon's name. You'll find a star rating. You won't find what it actually means.
That's the problem with the existing surgeon review ecosystem — Healthgrades, Vitals, RateMDs, Zocdoc, Google Reviews. The stars look authoritative. The underlying signal is something different than what most patients assume they're reading.
What patient star ratings actually measure
A patient who sees a surgeon for a hernia repair has direct experience with roughly six things:
• How long they waited in the office
• Whether the front-desk staff was polite
• How long the consult lasted
• Whether the surgeon explained the procedure clearly
• How they felt during recovery (pain, mobility, follow-up)
• Whether the outcome matched what they were told to expect
Patient-review platforms turn those six observations into a five-star rating. The platforms aggregate those ratings, weight them in opaque ways, and present a number that — for most patients — is the strongest available signal about a surgeon they're considering.
That's a useful signal for the things it captures. It's not a useful signal for the things it can't.
What patient stars can't see
A patient on the table is unconscious. The 90 minutes that determine the actual outcome of the surgery happen during that window, and the patient has no access to any of it.
The things that drive complication rates, readmissions, and long-term recovery quality are mostly invisible to the patient:
Intraop decision-making: When the anatomy isn't what the imaging suggested, what does the surgeon do? Improvise? Pause and consult? Push through?
Closure technique: How careful is the surgeon with hemostasis, irrigation, layer-by-layer closure? These small choices drive infection and bleed rates downstream.
Communication with the team: Does the surgeon call out next steps? Acknowledge the count? Listen when the scrub tech notices something off?
Response to surprises: A great surgeon's average day looks like a fine surgeon's average day. The difference is what happens on the hard day.
The pattern across cases: One patient sees one surgery. The circulating nurse who's done forty cases with this surgeon sees the pattern.
Patient stars can't see any of this. Not because the platforms are doing something wrong — but because the patient, by definition, wasn't in the room.
The downstream problem
The result is a review ecosystem that's well-calibrated for office experience and poorly calibrated for clinical quality.
Two surgeons can have identical 4.7-star ratings:
• Surgeon A: warm, runs on time, careful explainer. Complication rate 1.2 percentage points above national average. OR team describes their intraop decision-making as "rushed when the case gets hard."
• Surgeon B: blunt, runs behind, doesn't make small talk. Complication rate 0.8 points below average. OR team describes them as "the person you want in the room when something goes wrong."
A patient choosing between them based on stars is being told the wrong thing. Not maliciously. Structurally.
What peer observation captures
Anonymous, verified review from OR staff — circulators, scrubs, anesthesiologists, device reps — measures exactly the layer patient reviews can't reach:
• Technical care (closure, hemostasis, instrument handling)
• Surgical judgment (case selection, intraop decisions)
• Communication and team culture
• Behavior on hard days, not just average ones
• The pattern across cases
This isn't a replacement for patient reviews. The patient experience is real and worth measuring. But pretending it captures clinical quality is a category error.
The Bostrom argument — that Americans have better data for picking a restaurant than a surgeon — is partly about volume. It's also about whether the data being collected matches the question being asked.
How the layers fit together
The honest answer is that no single source tells you what you need to know about a surgeon. The useful comparison stack looks something like:
Hospital-level outcome data (Leapfrog, CMS Star Ratings, Joint Commission certifications). What the institution does well, on average.
Surgeon-specific outcome data (where available — some specialties more transparent than others). The lagging indicator.
Patient experience reviews (Healthgrades, Vitals, RateMDs, Google). The bedside, office, and recovery experience.
Peer observation from the OR team (Physician Signal). The clinical layer patient stars can't see.
Each of these answers a different question. None of them is the full picture. The mistake is treating one of them — usually the most accessible — as if it were.
The shift
Patient star ratings filled a real gap. Before Healthgrades existed, patients had nothing. Five-star aggregation of bedside manner is better than no data at all, and the platforms that built that ecosystem deserve credit for normalizing the idea of evaluating doctors publicly.
What's changed is that the data layer below patient experience — the part that happens during the procedure — is now collectible too. Not from patients. From the people standing across the table.
That's the part the existing ecosystem can't reach. And it's the part that, for most surgical decisions, actually predicts what happens.
If you're researching a surgeon today, read the patient reviews. Then look for the peer review. The combination is what the old ecosystem was always missing.
Search any surgeon's Insight Score
Not medical advice. Reviews are professional opinions only.