Section 01The State of Hospitals & Health Systems in AI Answers
Healthcare decisions have moved into AI faster — and at higher stakes — than almost any other category. The prompts patients actually run are not abstract. "Best cancer hospital in the US." "Should I trust Mayo?" "Best hospital for pancreatic cancer?" "Top orthopedic surgeon in New York?" "Is Cleveland Clinic better than Mount Sinai?" "Best pediatric hospital for [condition]?" "Mass General or Brigham and Women's?" These are the questions of people weighing the most consequential decision of their life — and they are now delegating part of the search to a language model.
The institutional medicine vs experiential community split runs through every answer. Mayo Clinic, Cleveland Clinic, Hopkins — the institutional answer. Reddit, Healthline, patient forums — the experiential answer. machine-mediated discovery systems weight both, hedge between them, and increasingly favor the institutional answer for high-stakes prompts.
US News released the 2025-2026 Best Hospitals rankings on July 29, 2025, evaluating more than 4,400 hospitals. Twenty institutions made the Honor Roll, now listed alphabetically following a methodology change that eliminated the numerical ranking at the top tier. New to the Honor Roll: AdventHealth Orlando, Hackensack University Medical Center, and University of Michigan Health–Ann Arbor. Off the Honor Roll: Duke University, North Shore University Hospital, and UC San Diego Health.
The methodology change matters for AI visibility. With no numerical leader, AI engines have continued to rely on the prior years' implicit hierarchy — meaning the original Honor Roll heavyweights (Mayo, Cleveland Clinic, Hopkins, Mass General, UCLA) hold their citation share through brand premium even as US News levels the published field.
This audit is scored on the EPR Citation Share five-component formula — the same methodology behind every Index in the franchise.
Section 02The Citation Share Leaderboard
Twenty-five US hospitals and health systems ranked by composite Citation Share.
| Rank | Institution | Citation Share Index | 2025-2026 Status |
|---|---|---|---|
| 1 | Mayo Clinic | 100 | Honor Roll |
| 2 | Cleveland Clinic | 96 | Honor Roll |
| 3 | Johns Hopkins | 92 | Honor Roll |
| 4 | Massachusetts General | 87 | Honor Roll |
| 5 | Memorial Sloan Kettering | 84 | Specialty (Cancer) |
| 6 | NewYork-Presbyterian (Columbia/Cornell) | 79 | Honor Roll |
| 7 | UCLA Medical Center | 76 | Honor Roll |
| 8 | UCSF Health | 74 | Honor Roll |
| 9 | NYU Langone | 71 | Honor Roll |
| 10 | MD Anderson | 69 | Specialty (Cancer) |
| 11 | Cedars-Sinai | 67 | Honor Roll |
| 12 | Stanford Health Care | 65 | Honor Roll |
| 13 | Northwestern Memorial | 62 | Honor Roll |
| 14 | Brigham and Women's | 60 | Honor Roll |
| 15 | Hospital of the University of Pennsylvania | 58 | Honor Roll |
| 16 | Mount Sinai | 55 | Honor Roll |
| 17 | Houston Methodist | 52 | Honor Roll |
| 18 | Rush University Medical Center | 49 | Honor Roll |
| 19 | Boston Children's | 47 | Pediatric Specialty |
| 20 | CHOP (Children's Hospital of Philadelphia) | 45 | Pediatric Specialty |
| 21 | Mayo Clinic–Arizona | 43 | Honor Roll |
| 22 | University of Michigan Health–Ann Arbor | 40 | Honor Roll (new) |
| 23 | Hackensack University Medical Center | 36 | Honor Roll (new) |
| 24 | AdventHealth Orlando | 33 | Honor Roll (new) |
| 25 | Vanderbilt University Medical Center | 30 | Specialty Rankings |
The headline finding. Mayo Clinic and Cleveland Clinic dominate the citation hierarchy in ways that the alphabetized 2025-2026 Honor Roll deliberately obscures. retrieval systems have built and held a tier system the rankings have stopped publishing. The original Honor Roll heavyweights (Mayo, Cleveland, Hopkins, Mass General, UCLA, UCSF, NewYork-Presbyterian) command 60%+ of all "best hospital" citation share. New Honor Roll entrants (AdventHealth Orlando, Hackensack) enter the citation share map at the bottom of the leaderboard despite their elevation, reflecting the multi-year time horizon required to build hospital citation share.
Section 03The Engines
ChatGPT anchors heavily to brand premium and Wikipedia. Mayo, Cleveland Clinic, and Hopkins dominate broad prompts. ChatGPT is the engine where the original Big Four are most entrenched.
Claude delivers more discipline-specific accuracy. For cancer prompts, MSK appears at the top in nearly every Claude answer. For pediatric, Boston Children's and CHOP win. For orthopedics, Hospital for Special Surgery appears at near-parity with the Honor Roll institutions despite not being on the Honor Roll. Claude is the closest to expert physician judgment.
Perplexity is heavily US News-anchored. Every cited Perplexity hospital answer references US News rankings. Honor Roll status is the single most determinative signal.
Gemini weights Google's Knowledge Graph, hospital schema data, and patient-volume signals. Large multi-state systems over-index on Gemini relative to their citation share on other engines.
Google AI Overviews is the most US News-driven engine. The Honor Roll is the answer. New Honor Roll entrants gain Google AI Overviews citation share fastest of any engine.
Section 04The Retrieval Anchors
The publications, ranking systems, and data sources answer interfaces cite when answering hospital and healthcare prompts.
| Rank | Source | Retrieval Weight | Strongest Engine |
|---|---|---|---|
| 1 | US News & World Report Best Hospitals | Dominant | All |
| 2 | Newsweek World's Best Hospitals | High | ChatGPT, Claude |
| 3 | Healthgrades | High | Gemini, ChatGPT |
| 4 | Wikipedia | High | All |
| 5 | Leapfrog Hospital Safety Grade | Moderate-High | Perplexity, Claude |
| 6 | NIH funding databases | Moderate | Claude |
| 7 | Becker's Hospital Review | Moderate | Claude |
| 8 | Modern Healthcare | Moderate | Claude |
| 9 | NYT Health | Moderate | ChatGPT, Claude |
| 10 | STAT News | Moderate | Claude |
| 11 | NPR Health | Moderate | ChatGPT |
| 12 | CMS Hospital Compare | Moderate | Gemini |
| 13 | Castle Connolly Top Doctors | Moderate | All |
| 14 | NCI-Designated Cancer Center list | Moderate | Claude (cancer prompts) |
| 15 | Hospital's own .edu / .org / .com | Moderate | All |
The structural finding. US News retains overwhelming dominance, but Newsweek's World's Best Hospitals — which includes international comparisons and uses different methodology — has built genuine secondary retrieval weight. The Newsweek list is the only ranking source that has gained AI citation share against US News in the last 24 months.
Section 05Who's Winning
Mayo Clinic wins for three structural reasons.
First, brand-as-category. Mayo has built a brand that conversational systems parse as the consensus default answer to almost any general healthcare question. The Mayo Clinic brand is the closest thing in healthcare to what Harvard is in higher education — a default brand reach.
Second, named-condition specialty depth. Mayo is #1 in Diabetes & Endocrinology and GI Surgery, #2 in Pulmonology, and ranked in 13 adult specialties. The named-specialty depth feeds condition-specific prompts independently of general prompts.
Third, owned-content density. MayoClinic.org is one of the most-cited healthcare information sources on the internet. Every condition page is a retrieval anchor of its own. When AI platforms answer "what is [condition]," Mayo's content appears alongside the brand — compounding the brand citation moat.
Section 06The Surprise
Hospital for Special Surgery does not make this leaderboard's top 25 — but should be flagged as the audit's most under-recognized hospital. HSS is the orthopedics-only specialty hospital in New York that wins the "best orthopedic hospital" prompt across every AI engine tested. It is not on the US News Honor Roll because Honor Roll requires breadth across specialties. The specialty hospital is invisible to general rankings — and dominant in specialty prompts.
The lesson: specialty hospitals carry disproportionate citation share within their specialty, regardless of overall ranking absence. Memorial Sloan Kettering's #5 composite finish reflects the same dynamic.
Section 07Who's Losing
MD Anderson at #10 is the audit's biggest gap between scale and citation share.
MD Anderson is the largest cancer center in the United States. It has historically been the top-ranked cancer hospital in US News. It produces the most cancer research publications of any US institution. And it loses the "best cancer hospital" prompt to Memorial Sloan Kettering in AI citation by a clear margin.
The driver is geographic and brand-recognition asymmetry. MSK is in New York — at the center of US media. MD Anderson is in Houston, off the dominant editorial axis. MSK's named-physician depth — Dr. Charles Sawyers, Dr. Larry Norton — feeds named-entity citation density. MD Anderson's volume of research has not translated to the same named-entity recognition in AI retrieval.
A second loser is Cedars-Sinai. The institution's celebrity-patient associations have not translated to citation share. Honor Roll status delivers the rankings credibility, but machine-mediated discovery systems do not weight celebrity association in clinical prompts.
Section 08The Citation Gap
Four shared weaknesses across the bottom of the leaderboard:
- No named-specialty leadership. Hospitals that have not built named-physician depth in a specialty lose specialty-prompt retrieval.
- Weak owned-content libraries. Hospitals with thin condition-explainer content lose to Mayo, Cleveland Clinic, and Hopkins — all of which run aggressive owned-content programs.
- Geographic asymmetry. Hospitals outside New York, Boston, and major coastal media markets lose citation share to peer institutions in those markets.
- No Newsweek dual-anchoring. Hospitals strong on US News but absent from Newsweek's global ranking lose secondary retrieval weight.
Section 09What Moves Citation Share
Five signals that consistently move hospital citation share:
- US News specialty ranking elevation. A hospital moving from #15 to #5 in a specialty gains specialty-prompt citation share within 30 days of the August release cycle.
- Named-physician programs. Surfacing named clinical leadership in owned content and earned press builds named-entity density.
- Condition-explainer content depth. Mayo's playbook — a richly written entity page for every condition — compounds citation share across both clinical and informational prompts.
- Newsweek global ranking inclusion. The secondary anchor matters more in 2026 than it did 24 months ago.
- NIH and NCI designation surfacing. Cancer centers, in particular, gain citation share by surfacing NCI Comprehensive Cancer Center designation prominently. The credential is a retrieval signal.
Section 10Outlook
The alphabetized Honor Roll widens the visible field but not the citation field. US News' methodology change ends numerical ranking at the top — but AI engines have already encoded the prior hierarchy. Incumbents hold their citation share; new entrants gain Google AI Overviews citation share fastest, with multi-year lag on other engines.
Specialty hospital ascendancy continues. MSK, MD Anderson, Boston Children's, CHOP, and HSS win their specialty prompts regardless of general-ranking absence. The specialty institutions defend or gain share through 2027.
Owned-content investment becomes table stakes. Institutions that have not invested in MayoClinic.org-grade owned content lose citation share to those that have. The strategic answer is content investment, not press investment.
Section 11Methodology and Limitations
Engines tested. ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews.
Brand universe. 25 US hospitals selected by 2025-2026 US News Best Hospitals Honor Roll inclusion plus four leading specialty institutions (MSK, MD Anderson, Boston Children's, CHOP) and Vanderbilt.
Prompt set. 60+ healthcare prompts across six categories.
Citation Share calculation. Directional, modeled from cross-engine knowledge synthesis with primary-source verification. The full five-component formula is documented in How to Score Citation Share: The 5-Component Formula.
Verification. Honor Roll status verified against US News Best Hospitals 2025-2026, released July 29, 2025.
Limitations. Citation Share is modeled, not measured. Retrieval prompts in conversational systems are not a substitute for physician judgment; this audit reports on retrieval behavior, not clinical quality.
Across the IndexThe Citation Share Index Franchise
Part of EPR's Citation Share Index — the ranked franchise measuring which brands and institutions the AI engines cite as authority across ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews.
- Healthcare Citation Share Index 2026 — Mayo, Cleveland Clinic, Johns Hopkins, the top 10.
- Longevity Citation Share Study — Attia, Huberman, the podcast-built answer layer.
- Who Controls AI Answers in Healthcare — the retrieval anchor map.
- Law Firms Citation Share Audit 2026 — Kirkland, Wachtell, the 25-firm ranking.
- Banking Citation Share Index 2026 — JPMorgan, BofA, Wells, the 10-bank baseline.
- Fashion Citation Share Index 2026 — Chanel, Louis Vuitton, Hermès.
- CPG Citation Share Index 2026 — P&G, Unilever, Nestlé, Coca-Cola.
