CME in South Carolina

Four research lanes — one of which failed and whose ground was covered directly — plus direct extraction of all three platform tenants. ⭐ SOUTH CAROLINA HAS NO INTRASTATE ACCREDITOR, AND GEORGIA'S FILLS THE VACUUM. The South Carolina Medical Association is an accredited PROVIDER (24 activities) but is not among ACCME's 28 Recognized Accreditors — the list was read live in a browser, because a plain fetch of that page returns a client-rendered shell. Five of the state's twelve registry providers, 62 activities, are accredited by the MEDICAL ASSOCIATION OF GEORGIA: McLeod Regional 30, Roper St. Francis 15, Grand Strand 9, Self Regional 6 and The Hawkins Foundation 2. This is the Rhode Island shape with Georgia playing Massachusetts, and it explains a zero that an earlier pass measured without an explanation — the Southern States CME Collaborative has no South Carolina providers because Georgia is already here. ⭐ AND THE ACCREDITOR'S OWN SCOPE STATEMENT CLOSES EXACTLY WITH THE REGISTRY. MAG's accreditation page says verbatim that it accredits organizations authorized to offer credit 'that meet Georgia, South Carolina and Tennessee license renewal requirements'. Sweeping provider_accredited_by across eleven states finds MAG providers in exactly those three — Georgia 15, South Carolina 5, Tennessee 1 — and ZERO in North Carolina, Florida, Alabama, Virginia, Mississippi, Louisiana, Arkansas and Kentucky. Two independent sources agree, making this the second state after Alabama whose intrastate layer is closed rather than inferred. The difference is worth keeping: Alabama's accreditor publishes a ROSTER, Georgia's publishes only a SCOPE SENTENCE, and the registry supplies the roster itself. Note also that MAG's declared Tennessee scope is real but DORMANT at 1 provider and 0 activities, which is the explanation the Tennessee pass measured and could not account for. Every state MAG does not reach has an accreditor of its own. ⚠️⚠️ NEITHER OF THIS STATE'S PLATFORM TENANTS IS GUESSABLE FROM THE ORGANIZATION'S CURRENT NAME, AND FIFTEEN NAMESPACE GUESSES FOUND NEITHER. USC School of Medicine Columbia runs CMETracker as `USCSOMPHR` — an initialism ending in the hospital's name — and Prisma Health Upstate runs EthosCE as `ghscme`, from Greenville Health System, the organization's FORMER name. Both were found from the provider's own CME page. That is now two for two here and a repeat of the Tennessee finding, so the provider's own site is the PRIMARY discovery route and the namespace probe is the fallback, not the reverse. MUSC's CloudCME tenant, by contrast, is simply `musc` and yielded 189 unique activities of which only seven are in-person South Carolina conferences, every one in Charleston. ⚠️⚠️ THE ETHOSCE .ics FEED IGNORES ITS OWN MONTH PATH, AND ITS CALENDAR CONTAINS NO CONFERENCES AT ALL. calendar.ics/<YYYY-MM> returns the identical 485 events for every month asked for, so eleven monthly fetches produce 5,335 VEVENTs and 485 unique UIDs — a loop that concatenates without de-duplicating on UID inflates the catalog elevenfold and looks thorough doing it. Worse, all 485 are sessions of 234 departmental series: the six real conferences live only in the separate /courses view. A pass that read the calendar would have concluded this tenant runs no conferences while it runs six, one of them a 24-credit course. ⚠️ AN ACCREDITATION STATEMENT IS NOT A DESIGNATION, AND THE PER-ROW FETCH DOES NOT ALWAYS RESCUE ONE. The standing advice for a CloudCME row whose credit is a bare tag is to fetch its courseoverview page, which states the figure. For two MUSC rows it does not: the overview carries only 'The Medical University of South Carolina is accredited by the ACCME to provide continuing medical education for physicians' — the provider's standing permission to award credit, not this activity's allocation. Both are recorded as credit unknown. ⚠️ AND TWO FIELDS CARRIED CONTENT THE PARSER DOES NOT READ. A CMETracker row whose CreditElig is null, and which the extractor therefore reports as credit-missing, says in its Notes field that 'AMA credit and hours of attendance will be awarded; the exact amount will be determined and posted once the agenda is finalized' — credit affirmed but not quantified, a third state alongside a missing designation and a published 0.00. And a row with FacCity, FacState, FacAddress and FacZip ALL null publishes its city in the `Room` field instead: the state trauma symposium is in Myrtle Beach while every other row on that tenant is in Columbia, so a default to the tenant's home city would have mis-sited it by 150 miles. A null venue field is not an unpublished venue. ⚠️ AAFP PRESCRIBED CREDIT IS NOT AN AMA PRA CATEGORY 1 DESIGNATION, AND THE PUBLISHER SAYS SO OUTRIGHT. Three activities here carry the sentence 'AAFP Prescribed credit is accepted by the American Medical Association as equivalent to AMA PRA Category 1 Credits™' — an equivalence claim, not a designation — and the AHEC pages add the decisive line: 'When applying for the AMA PRA, Prescribed credit earned must be reported as Prescribed, not as Category 1.' A page containing the string 'AMA PRA Category 1' is not thereby an AMA PRA Category 1 activity. The same discipline keeps the osteopathic society's 20 AOA Category 1-A credits and the PA academy's pending AAPA Category 1 out of the physician-credit count. TWO OF THE STATE'S SPECIALTY SOCIETIES HOLD THEIR ANNUAL MEETING IN ASHEVILLE, NORTH CAROLINA — the Neurological Association at the Crowne Plaza and the Society of Pathologists at the Omni Grove Park Inn — so both are filed as North Carolina rows with South Carolina organizations. In the other direction the coast pulls meetings in: Kiawah Island, Hilton Head Island, Charleston, Myrtle Beach and Isle of Palms host conferences run from Illinois, Georgia, Florida, North Carolina, California and Ottawa. Nine South Carolina-sited rows already existed here before this pass began, banked by the North Carolina and Tennessee passes. ⚠️ AND THIS PASS CORRECTED ONE OF THEM, IN A WAY WORTH REUSING. The Carolinas Cancer Conference was seeded from North Carolina as 'credit unknown'. Its registration page says verbatim 'This program will not offer continuing education credits'. The standing 'unknown, not zero' rule protects a page that is SILENT about credit; this page makes an affirmative negative statement, and recording that as an unknown tells the reader credit might yet appear when the publisher has said it will not. THE SOCIETY LAYER IS HOSTED RATHER THAN SELF-PUBLISHED, WHICH IS WHAT A STATE WITHOUT ITS OWN ACCREDITOR LOOKS LIKE FROM THE INSIDE. The Medical Association staffs and publishes several specialty societies' meetings on its own calendar; the Rheumatism Society's annual meeting is a course on MUSC's tenant; the Neurological Association's is a course on Greenville's; and the Psychiatric Association's own page advertises 'more than 15 hours of lectures', which is not a credit designation, while the real 12.50 AMA PRA designation for that same meeting sits on the Greenville tenant. A society with a thin website can still have a fully designated meeting — on the accrediting provider's platform. SOUTH CAROLINA AHEC SITS AT THE NORTH CAROLINA POLE BUT DOES NOT PAY OFF LIKE IT. It runs one statewide Learning Portal across all four regional centres, and its Continuing Professional Development catalog is a single server-rendered page of 149 courses — but the catalog is overwhelmingly nursing certification review, SANE, fetal monitoring, social work, counselling and physical therapy. Only 17 of the 149 detail pages mention AMA PRA at all, most of those enduring online modules, and the only live physician-facing activities are two ALSO courses. The listing page itself contains zero credit strings, which was checked against a positive before being believed: all 149 detail pages were fetched and 17 came back positive, so the null is a property of the listing view and not of the reader. The health department is at the North Carolina pole too — its provider training offers only 'free continued education credits' with no designation named — and note the agency has been renamed, so a pass looking for SC DHEC finds a redirect rather than the Department of Public Health. CREDIT COMPOSITION, COUNTED AS A PINNED SET RATHER THAN A SUBSTRING MATCH: NINETEEN of the sixty-one rows added carry a verified AMA PRA Category 1 figure, eighteen of them South Carolina-sited — fourteen dated conferences and webinars with a designation statement, three recurring series with a published per-session figure, and one enduring activity. The count is a literal list of row ids, not a search for the phrase in prose, because counting rows whose prose MENTIONS 'AMA PRA Category 1' is the defect that inflated Alabama's figure to seventeen and Illinois's to 120 against a true 59. WHAT IS NOT HERE. Four registry providers publish nothing learner-facing at all — McLeod Regional (30 activities, confirmed by full sitemap enumeration), Spartanburg Regional (21, and its public CE department is explicitly nursing and allied health), AnMed (11, whose eight CME subpages are all planner-facing) and Self Regional (6) — which is 68 registry activities yielding no rows. The Greenville County Medical Society's calendar is login-walled, so it is UNREAD rather than empty. MUSC's CMETracker namespace returns a persistent 503 across four attempts and two product paths while known-live tenants return 200 and five plausible fake tenants return 404, so it is a routed namespace that is erroring rather than an absent one; its CloudCME tenant bounds the exposure. Eleven societies are dates-TBA with a known season rather than inactive, and the Symposia Medicus high-risk-pregnancy conference at Hilton Head in November 2027 is deliberately unseeded because its own page publishes neither venue nor credit. ⚠️ AN ADVERSARIAL REVIEW OF THIS PASS CORRECTED TWENTY-NINE ROWS' AUDIENCES, IN BOTH DIRECTIONS, AND THE CREDIT FIGURES SURVIVED IT UNCHANGED. Twenty-seven rows claimed nurse practitioners or physician assistants on pages that name neither — the Alabama over-claim recurring, and this time the row's own note had recorded the weakness ('PA is claimed on the strength of the CloudCME audience list') while citing a field that does not exist on those pages. A hedge written into the prose is not a substitute for not making the claim. Two rows were corrected the OTHER way and are the more instructive half: the osteopathic society's conference was shipped DO-only because its credit is AOA Category 1-A, while its page says it is 'designed for Doctors of Osteopathic Medicine, MD Physicians … as well as Advance Practice Providers' — THE CREDIT SYSTEM IS NOT THE AUDIENCE — and a hospital's series row said no audience was published when the page carries a heading reading 'Target audience' that names nurse practitioners and physician assistants outright. The review also found two quotations attributed to the wrong place: an accreditation sentence quoted with an office name spliced into it, and an audience line that exists on its page only INSIDE AN HTML COMMENT. Both are fabricated quotations even though every word of each exists somewhere. And it found an UNDER-claim, which is the same defect facing the other way: both ALSO courses publish 'up to 8.0 AAFP Prescribed credit' and were recorded as having no published hour count.