Three Federal Outbreak Pages, One Saturday: CDC Is Working 30, FDA Lists 15 With 14,168 Sick and Nine Naming No Food, and USDA Lists One and Does Not Yet Know Which Ingredient Was Contaminated or Whether It Went Into Anything Else.

Three federal pages carry the running account of what is making Americans sick. As of October 15, 2026, FDA’s Investigations of Foodborne Illness Outbreaks table carries fifteen active investigations with 14,168 people sick, 652 hospitalized where FDA publishes a number, and two dead. CDC’s Current Outbreaks page says the agency is coordinating thirty multistate investigations: fifteen Salmonella, seven E. coli, six Listeria and two Campylobacter. USDA’s Food Safety and Inspection Service lists four investigations for all of 2026 on its Outbreak Investigations: Response table and marks one of them active. I walked FDA’s table on September 29and October 1; here is where all three pages stand now.

What moved on FDA’s page this week. The joint FDA and USDA E. coli O157:H7 file, #1421 at FDA and 2026-04 at FSIS, went from 38 sick to 50. I wrote about the first lawsuit in that outbreak on Thursday and about why the file is the case for both of my citizen petitions on Friday. Listeria file #1417, posted September 23 with no food named, went from 44 to 46, still traceback only, no inspection or sampling. Listeria file #1415, posted September 30 with eleven sick, added sampling. The Graziers raw milk cheese advisory was re-stamped October 9 with one new line, distribution to the Republic of Korea; the counts have not moved since September 25: 13 sick, 8 hospitalized, 3 with HUS, nine states. The broccoli sprout advisory now lists three retail seed recalls, EAK Distribution, Eden Brothers and Gardeners Basics, all on supplier lot W4105, the lot FDA’s own seed sample matched by whole genome sequencing; the supplier is not named, and the count has sat at 32 since September 24. Cyclospora file #1392 went from 235 to 240. And the Salmonella Oranienburg file, #1387, posted July 8 with no food named, is now marked ended at 99 sick, with traceback, inspection and sampling checked. Fourteen weeks, three boxes, no food.

FDA’s active investigations as updated October 15, 2026. Counts for the five rows with an advisory are from the advisories. Inspected means FDA’s on-site inspection box is checked.

Nine files name no food, and they hold 626 people. Four of the nine are ongoing and hold 184: the Salmonella Newport file, #1403, posted August 26, at 116 sick with no inspection, unchanged since September 23; the 46-person Listeria file with traceback only; and two Listeria files of eleven each. The other five are marked ended and hold 442: three Cyclospora files of 240, 38 and 22, and two Salmonella files of 43 and 99. In those five the outbreak is over by FDA’s own column, the investigation is still active, and the food line reads Not Yet Identified. For the 442 people in them, ended means nobody is going to tell them what they ate.

Seven rows say the outbreak ended, and two of them say the investigation did too. Seven of the fifteen rows carry Ended in the outbreak column and Active in the investigation column. Two of the seven are the files FDA announced it had closed on October 2: the iceberg lettuce Cyclospora outbreak, 12,883 sick, 570 hospitalized and two dead in 21 states, and the Sinaloa jalapeño Salmonella outbreak, 488 sick and 65 hospitalized in 34 states. Both advisories say FDA’s outbreak investigation has closed. Thirteen days later, both rows are still in the Active section. Those two files are 13,371 of the table’s 14,168 people, 94 percent. Take them out and FDA’s open table is thirteen files and 797 people.

CDC is working thirty, and you cannot tell which thirty. CDC’s dashboard gives four numbers and nothing else; FDA’s table gives reference numbers and no CDC cluster codes. Of the eight ongoing rows on FDA’s table, three have a CDC outbreak notice: the Graziers cheese, the broccoli sprouts and the alfalfa sprouts. The wrap file, the Newport file and the three Listeria files have none. CDC says six Listeriainvestigations; FDA has three Listeria rows and FSIS none, so at least three exist on no public page. CDC says two Campylobacter; neither regulator lists one. CDC says fifteen Salmonella; FDA’s table has six Salmonellarows, three of them ended, so at least nine have no page. CDC says seven E. coli; FDA has three E. coli files, one of them carrying three serotypes, so at least two more. And the dashboard counts none of FDA’s four Cyclospora rows, because it tracks four germs and Cyclospora is not one of them. Some of this is architecture: CDC counts clusters that may never become outbreaks, and the regulators post only what they manage. A reader still cannot say from the three pages how many outbreaks the government is working, or which.

USDA lists one, and its only sentence is that it does not know yet. FSIS’s table has four entries for 2026. Three, a May headcheese Listeria file, the May Kebab Shop beef kofta E. coli O157:H7 file and a June Salmonella Enteritidis file at the Broome County, New York jail, are marked No in the Active column. The fourth is the wrap file, the only one marked Yes, and FSIS’s page carries no case count for it; the 50 lives on FDA’s table. What FSIS’s page does carry is a sentence of its own: the wrap identified so far is past shelf life and no longer for sale, and FSIS is working with FDA to identify the contaminated ingredient and whether it was used in other products. More than three weeks after the file opened, with 50 people sick, the agency that regulates the chicken says it does not yet know which ingredient was contaminated or whether it went into anything else. There is no public health alert because the one product identified is gone; whether its ingredient is in something else is the open question.

Three closed files got summaries this week, and two of them black out the names. FDA posted Executive Incident Summary Abstracts for the egg, moringa and infant formula outbreaks. I covered the egg summary on Friday: a laying house tested positive and was not cleaned before new hens went in, 134 sick, 34 hospitalized, 18 states. The moringa summary closes that outbreak at 131 sick, 127 of them SalmonellaTyphimurium, 36 hospitalized in 38 states. Traceback converged on a contract manufacturer and on the importer of the moringa powder behind all three products; FDA inspected a foreign supplier on March 2 and issued a Form 483 with five observations; thirteen product samples were positive, four different strains; and a firm was placed on Import Alerts 99-43 and 99-19. The contract manufacturer and the supplier are redacted as (b)(4), confidential commercial information. The formula summary adds three things to what FDA’s advisory already said about the Nara and ByHeart link. The two California infants’ isolates and the open Nara can from Washington were not closely related to anything in the national database, and the open can and the Washington infant were not closely related to each other either; only that infant’s isolate sat within five SNPs of the 2025 ByHeart isolates, including four unopened cans. On June 25, a firm whose name is redacted was placed on Import Alert 40-05, FDA’s detention list for adulterated and misbranded infant formula. And the closing line: not enough evidence to confirm sources and routes of contamination.

Fifteen open at FDA, 14,168 sick, nine files with no food and 626 people in them, seven marked ended and still active. Thirty at CDC, none named. One at USDA, with an ingredient it cannot name and a distribution it cannot rule out. The pages are public. Nobody is assigned to make them agree, and the two petitions I am finishing ask for the parts that would.