FDA’s Coordinated Outbreak Response and Evaluation (CORE) Network has posted Executive Incident Summary abstracts for five 2026 outbreak investigations: Salmonella in Mogo moringa capsules (ref #1377) and Rosabella moringa capsules (#1365), E. coli O157:H7 in Raw Farm raw cheddar and raw milk (#1369), Salmonella Newport in Guatemalan cantaloupe (#1366), and Listeria monocytogenes in Clover Hill Dairy requesón (#1380). They are the closest thing the public gets to FDA’s own account of an investigation, and they are worth reading for what they say and for what FDA blacked out.
The short version
The five outbreaks sickened 138 people. FDA named the food in all five, reported the outbreak strain in product in three, and issued inspectional observations in four. One firm refused to let FDA in, and its name is redacted.
| Ref | Outbreak | Sick | Key finding in the abstract |
| #1377 | Salmonella Typhimurium, Mogo moringa capsules | 34* | Outbreak strain in three opened containers; two FDA-483s, one issued by FDA’s India Office |
| #1365 | Salmonella Newport and Kentucky, Rosabella moringa capsules | 10 | Supplier refused FDA access; no source of contamination identified |
| #1369 | E. coli O157:H7, Raw Farm raw cheddar and raw milk | 9 | FDA-483s at all three Raw Farm locations |
| #1366 | Salmonella Newport, cantaloupe from Guatemala | 70 | Two Guatemalan firms on Import Alert 99-35; patients’ states redacted |
| #1380 | Listeria monocytogenes, Clover Hill Dairy requesón | 15 | Outbreak strain in nine samples; listed state counts total 12 |
*From FDA’s closing advisory; the #1377 abstract gives no final count.
Moringa, twice
CDC told CORE on February 6 about a new extensively drug-resistant Salmonella Newport cluster, with moringa leaf powder the leading hypothesis that day, according to the #1365 abstract. Traceback followed Rosabella capsules bought online by five Newport and two Kentucky patients. One of 22 samples, from a patient’s home, matched the outbreak strain. Ambrosia Brands, LLC recalled 52 lots on February 13. On February 23, FDA inspected a redacted firm and issued a ten-item FDA-483 covering specifications, ingredient testing, recordkeeping, sanitation and contamination.
Then comes the sentence that should stop anyone. On March 2, FDA’s Office of Global Policy and Strategy tried to inspect and sample a redacted firm, which refused access and would not provide records. FDA tied the capsules to a single powder supplier but identified no source of contamination. Ten people in eight states were sick, three hospitalized.
The abstract does not say where that firm is, though the attempt ran through FDA’s international office. If it is abroad, Congress already answered the question. Section 807(b) of the Food, Drug, and Cosmetic Act says food from a foreign establishment that refuses FDA inspectors shall be refused admission, and FDA has guidance on what counts as a refusal. The abstract does not say whether FDA used that authority, and with the name redacted, nobody outside FDA can check.
The Mogo outbreak followed. CDC notified CORE on May 13 of a Salmonella Typhimurium cluster in 12 states, per the #1377 abstract. Traceback covered Mogo capsules eaten by five people in Florida, Oregon, Tennessee and Washington, and TNVitamins capsules eaten by one person in North Carolina, with patients who had purchase information reporting Amazon. Mogo’s powder came from a redacted supplier in India that sourced from two small farms. FDA issued two FDA-483s, one on May 22 to a redacted firm and one on June 5 from its India Office, the second for missing records and procedures, sanitation, lack of hazard controls, personnel practices and facility design. Three opened containers from patients’ homes carried the outbreak strain.
FDA’s closing advisory puts the outbreak at 34 sick and 11 hospitalized, plus a breastfed infant sickened with the outbreak strain, though it says 14 states in its text and lists 22 in its case-count box. FDA says it has increased screening of moringa powder imported from India. TNVitamins capsules were also recalled in the January 2026 moringa outbreak tied to Live it Up Super Greens, which makes three moringa outbreaks this year.
Raw Farm, again
The #1369 abstract is the only one of the five with no redacted firm names. CDC told CORE on March 10 about seven E. coli O157:H7 cases in California, Florida and Texas, with Raw Farm raw cheddar as the suspect. FDA says epidemiologic, limited traceback and laboratory data confirmed Raw Farm raw cheddar and raw milk as the vehicle. For-cause inspections at three Raw Farm locations each ended with an FDA-483, for poor equipment maintenance, lack of preventive controls and unsanitary handling. The final count was nine.
What the abstract leaves out: most patients were children under five, and Raw Farm recalled some cheddar on April 2 only after initially refusing, while disputing that it caused the outbreak, according to the Congressional Research Service. FDA also found E. coli O157:H7 in a Raw Farm sample matching a different, closed 2025 outbreak. This was Raw Farm’s second cheddar outbreak in about two years; the 2024 outbreak sickened 11 people in five states. I am preparing a citizen petition asking FDA to require a warning statement on cheese made from unpasteurized milk.
Cantaloupe from Guatemala, and a strain with a five-year trail
The #1366 abstract opens with CDC’s February 17 notice of a Salmonella Newport cluster whose early signal pointed to a general healthy diet and cruise ship travel. The states are redacted. Traceback of cantaloupe bought by 10 patients led to Guatemala, with Ayco Farms, Inc. of Pompano Beach, Florida, as importer on eight of nine legs, sourcing from Ayco Farms San Jorge and Agrobassy farms. Inspections there produced FDA Form 4056 observations on dump tank water and, at Ayco, a fraying food-contact conveyor belt. Of 23 samples taken in Guatemala, five yielded 73 isolates of four non-outbreak serotypes, and one environmental isolate matched five clinical isolates collected between 2021 and 2026.
Read that again. The outbreak strain was never found, but FDA found Salmonella in Guatemala that genetically matches people who got sick at various points over five years. Ayco recalled a lot on March 24, FDA placed Agrobassy and Ayco San Jorge on Import Alert 99-35 on April 1, and the outbreak closed at 70 cases. Food Safety News reported the recall covered 8,302 cartons sold in California, Florida, New York and Pennsylvania and was made Class I on April 20; Ayco has said the melons shipped between December 12, 2025 and January 16, 2026. I could find no FDA outbreak advisory, and the abstract still does not say where the 70 people lived.
Requesón, and numbers that do not add up
The #1380 abstract says CDC told CORE on May 27 of eight cases in Maryland, New York and Virginia. Traceback for four patients led to one manufacturer, Clover Hill Dairy, LLC of Mechanicsville, Maryland. Maryland suspended the dairy’s milk processing permit, recalls followed, and the outbreak strain turned up in nine product and environmental samples.
The abstract then reports fifteen cases in four states and lists them as Illinois 1, Maryland 4, New York 6 and Virginia 1. That adds to twelve, the count FDA published June 24. FDA’s closing advisory confirms 15 sick, 14 hospitalized and one death, with illnesses dating to March 6, 2023. The abstract mentions neither the hospitalizations nor the death, nor that the outbreak stretched over three years. A summary of record should add up, and it should say that someone died.
What the black ink hides
The redactions are marked (b)(4) and (b)(5), the Freedom of Information Act exemptions for confidential commercial information and internal deliberative material. Here they cover the firm that refused FDA, the Indian firms in the Mogo chain, a Guatemalan firm cited for dump tank water, and the states where the cantaloupe patients lived. Which firm sold contaminated food, or turned away a federal inspector, is not a trade secret in any sense a sick consumer would recognize.
Stop Foodborne Illness has a petition pending asking FDA to name implicated firms (Docket FDA-2026-P-1448). I am preparing a companion petition that asks, among other things, for these abstracts to be published without those redactions.
Questions FDA can answer
Is the firm that refused inspection a foreign establishment, and if so, has its food been refused admission under Section 807(b)? Is the Guatemalan strain matching patients from 2021 to 2026 being tracked as a recurring strain? Will FDA correct the requesón abstract and add the hospitalizations and the death? Each has a factual answer, and FDA’s abstract page is public for anyone who wants to ask.
Recent Comments