All posts by admin

E. coli O157:H7 Hazelnuts

Outbreak: E. coli O157:H7 Hazelnuts 
Product: HazelnutsInvestigation Start Date: 02/07/2011
Location: Multi-StateEtiology: E. coli (STEC) O157:H7
Earliest known case onset date: 12/20/2010Latest case onset date: 01/28/2011
Confirmed / Presumptive Case Count: 8 / 0Positive Samples (Food): 22

Eight outbreak cases of lab-confirmed E. coli O157:H7 were identified in Michigan, Minnesota, and Wisconsin. Initial, hypothesis generating interviews identified in-shell mixed nuts as a common exposure, specifically in-shell hazelnuts. A traceback investigation confirmed that the in-shell hazelnuts consumed by cases came from a common distributor which resulted in a press release and recall. Ultimately, the close collaboration between public health and agriculture agencies in multiple states, Centers for Disease Control and Prevention (CDC), and the United States Food and Drug Administration (FDA) allowed the identification of a novel vehicle for an O157 outbreak, with a very small number of detected cases.


A multi-state outbreak of E. coli O157:H7 infections associated with hazelnuts.


On February 7, 2011, the Minnesota Department of Health (MDH) Public Health Laboratory (PHL) determined that two human clinical E. coli O157:H7 isolates submitted through routine surveillance had indistinguishable PFGE patterns. A review of the national PulseNet database revealed four additional human E. coli O157:H7 isolates with the outbreak PFGE pattern in two states (three in Wisconsin and one in Michigan). A multi-state investigation was initiated.


Eight cases from three states were ultimately identified in this outbreak Minnesota (3), Wisconsin (4), and Michigan (1). All three Minnesota cases were male and had a median age of 62 years (range, 55 to 64 years). All three cases reported experiencing bloody diarrhea and cramps, two (66%) reported fever, one (33%) reported vomiting, and none reported fever. Two cases were hospitalized, each for 3 days. No cases developed hemolytic uremic syndrome and none died.
Upon initial interview, the first two Minnesota cases both reported consuming ground beef, sausage, lettuce, and nuts during the week prior to illness onset. Specific exposure information (i.e., brand and purchase location) collected on the ground beef, sausage, and lettuce consumed by the cases indicated these items were not from a common source. Upon re-interviews, all eight cases in the three states reported consuming in-shell hazelnuts also called filberts. Four case reported consuming hazelnuts as part of mixed nuts, and seven case reported purchased hazelnuts from bulk bins at grocery stores. One Wisconsin case reported purchasing packaged in-shell hazelnuts. However, further investigation at the grocery store where this product was purchased revealed that these hazelnuts were re-packaged at the store after originally being sold from a bulk bin.
A traceback investigation conducted by the Minnesota Department of Agriculture (MDA), in conjunction with the Michigan Department of Agriculture, California Food Emergency Response Team (CAL-FERT), and Wisconsin Department of Agriculture, Trade and Consumer Protection (WDATCP) found that the mixed nuts and in-shell hazelnuts purchased by cases originated from a single distributor, DeFranco and Son’s of California. On March 4, DeFranco and Sons issued a voluntary recall of all hazelnuts and mixed nut products distributed from November 2 through December 22, 2010. Recalled product was distributed to stores in seven states (Minnesota, Iowa, Michigan, Montana, North Dakota, South Dakota, and Wisconsin). A press release was issued on March 4, 2011 to inform the public.
In-shell hazelnuts collected by MDA from a case patient’s home tested positive for the outbreak PFGE subtype of E. coli O157:H7 on March 3, 2011. Additional mixed nut samples that included hazelnuts collected from recalled retail product by WDATCP and collected from DeFranco and Son’s by CAL-FERT also tested positive for the outbreak PFGE subtype of E. coli O157:H7. DeFranco and Son’s received hazelnuts from two companies in Oregon but did not maintain internal product traceability. The FDA conducted inspections of the two Oregon companies.


This was a multi-state outbreak of E. coli O157:H7 infections associated with eating in-shell hazelnuts grown in Oregon. Rapid collaboration between multiple state health departments and state departments of agriculture were crucial in identifying in-shell hazelnuts as the vehicle. This is the first documented outbreak of E. coli O157:H7 infections associated with nuts.


Hip-hop Measles

Outbreak: Hip-hop Measles 
Product: N/AInvestigation Start Date: 5/27/2007
Location: Lane County, OregonEtiology: Measles
Earliest known case onset date: 5/20/2007Latest case onset date: 6/5/2007
Confirmed / Presumptive Case Count: 2 / 1Positive Environmental Samples: N/A
Hospitalizations: 2Deaths: 0

In 2007, two cases of measles detected in Oregon (both unvaccinated) led to an investigation of potential exposures within a hospital and the community at large. This investigation cost approximately $170,000 across local and state health departments, and the related medical system, highlighting the costs associated with measles contact investigation.



On May 27, 2007, Lane County Health and Human Services (LCHHS) received a report of a possible measles case admitted to a local hospital. The index case was in his twenties, unimmunized, and had been in Japan during his putative incubation period. A second case was identified later. The cases lived in a mid-sized urban community (pop. 200,000), and, as we were later to find out, had active social lives.


Hospital staff reviewed their employees’ immunization status and the airflow system. Measles cases and their exposed contacts were interviewed using Oregon’s standard measles case-report form. For those contacts lacking documentation of immunity, vaccine or immunoglobulin (IG) was offered. Instructions for voluntary quarantine were given to exposed non-immune contacts. The costs of containing the two measles cases was estimated for the hospital and local and state public health departments.
Public health recommendations included three tiers of contact investigation:
(1) Health-care workers (HCWs) in direct contact; patients in the waiting room and emergency department (ED); household contacts and close friends.
(2) HCWs in units potentially exposed via air flow.
(3) High-risk patients (pregnant moms, babies, immunocompromised) potentially exposed via air flow.


On May 31, 2007, Lane County officials confirmed the diagnosis of measles in the index case by polymerase chain reaction testing. His prodrome began on May 20. He flew on May 21 from Tokyo to San Francisco, and thence on May 22 to Eugene. His rash was first noted on May 25. He spent time at a local hospital ED and visited a health food store, naturopath and Japanese restaurant during his communicable period.
The patient was not given a mask while in the ED waiting for his initial evaluation; rather, he was placed in a regular airflow room and then wheeled through the hospital without a mask and ultimately put in a taxi for the ride home. Review of the hospital’s airflow system revealed that air from the ER (where case was housed but not isolated) was shared with the Coronary Care unit and Mother Baby Unit. The circulated air had a mixture of about 20% outside air and 80% recycled indoor air with 90%–95 % effective filtration and no HEPA filter.
During the investigation, the index patient refused to identify household contacts and did not respond to LCHHS phone calls, making contact investigation difficult. An unannounced home visit helped to clarify the situation and obtain new information.

Information regarding 4 persons exposed on airline flights was not received until two weeks after the likely exposure. A week later, health officials were informed of two additional persons considered exposed, having sat next to or in front of the case, but phone numbers were not provided, and they had common last names. It also transpired that the case provided an incorrect seat number, and the model of the one of the airplanes was different from that listed on the airline’s website, further confusing attempts to identify exposed persons.
A second, unimmunized case, who had socialized with the index patient the night he arrived home from Japan, developed a febrile prodrome on May 30 and a rash consistent with measles on June 1. Koplik spots were visible. He declined lab testing.
Although nurses advised case #2 to stay home to avoid spreading the disease, he went to public places. On May 29, the case caught a hip-hop show at a local concert hall, then to a downtown bar. The next night, he went out for sushi.
Three bands that played at the concert were on a national tour. During these shows, the attendees typically stand, dance and mingle, the band is on a stage just above the floor and the band members often venture into the audience. The band members were in Utah when they were notified about their possible exposure, and specimens to verify immunity were collected in Colorado. The testing was performed at CDC in Georgia, and after the tests proved negative they were vaccinated while performing in Iowa.


This investigation presented numerous challenges. Both measles patients came from families that did not believe in measles vaccination, and who “don’t think measles are a big deal.” Local public health officials’ recommendations for isolation and quarantine didn’t impede their pursuit of an active social life—which greatly increased the work of contact tracing. The following factors also complicated the response:

  1. 1. Exposures during multiple airplane flights
  2. 2. Delay in receipt of information about people exposed during travel
  3. 3. Exposures among a community of unimmunized peers
  4. 4. Delayed isolation of the case at the hospital
  5. 5. Shared ventilation between the case’s room and other hospital units
  6. 6. Lack of airborne precautions during transit through hospital at time of discharge
  7. 7. Withholding of information and non-compliance with voluntary home isolation
  8. 8. Limited documentation of measles immunity among healthcare workers

Impact of Two Measles Cases

As a result, the investigation of these two measles cases and containment of the outbreak entailed substantial amounts of personnel time and money, as detailed below:

  1. • Hospital:
  2.   – Incident Command System (ICS) activated
  3.   – 1600 titers in a 2-week period
  4.   – 97% of HCWs were immune
  5.   – Cost of titers $40,000
  6.   – 600 fit tested for N95 masks
  7.   – 10 HCWs placed on furlough for several days
  8.   – 3 HCWs furloughed for 21 days
  9.   – 63 shots given
  10.   – New policy requiring proof of measles immunity
  11.   – Infection education module updated
  12.   – Isolation & transferring process reviewed
  13.   – $100,000 (estimated cost)
  14. • Local Health Department:
  15.   – ICS activated
  16.   – 2 cases
  17.   – 168 contacts investigated
  18.   – 90% were immune
  19.   – 4 shots given
  20.   – 4 people were placed in voluntary quarantine
  21.   – $50,000 (estimated cost)
  22. • State Health Department:
  23.   – ICS activated
  24.   – $20,000 (estimated cost)

Conclusion and Highlights

This outbreak was successfully controlled, despite the potential for spread. The limited extent of this outbreak, even in the setting of broad exposure, highlights the high level of population immunity achieved in Oregon and in other states.
This outbreak and its burden on clinical and public health resources could have been limited by adherence to recommendations of the Advisory Committee on Immunization Practices (ACIP) for vaccination of high-risk adults against measles.

  1. • Consider using quarantine orders and the quarantine process as outlined in statute to minimize the risk of spreading the virus
  2. • Consider taking legal action when cases are do not comply with public health investigation and control efforts
  3. • Develop educational materials with clear, relevant messages targeting vaccine-hesitant communities affected by the outbreak
  4. • Continue efforts to ensure networking with the alternative medical community
  5. • Expand use of digital communications for public information
  6. • Ensure airborne infection control precautions in healthcare settings
  7. • Promote measles vaccination and documented evidence of immunity among healthcare workers in Oregon

Museum Exhibit Contest


Welcome, citizen, to the Museum’s first outbreak submission contest!

Submit a significant or interesting outbreak investigation and you may win official International Outbreak Museum T-shirts! The winning outbreak investigation submission will also become an official online exhibit right here on the Museum website.

   To download these rules as a PDF, Click Here
   To download these rules as a Microsoft Office Word Document, Click Here

Promotion Period:The International Outbreak Museum (IOM) Outbreak Submission T-shirt Contest begins March 1st, 2018 and ends March 31st, 2018. A winner will be chosen within 30 days of the promotion end date. We will contact the winner if additional online exhibit materials are needed.

Prize: The winner (or each member of the winning team) will receive a T-shirt in (choose one) small, medium, large, or extra large (subject to availability). In addition, your winning outbreak investigation will become an official IOM website online exhibit.

Special Requirements: Since the chosen winner’s exhibit will become an official online exhibit at, the aforementioned must be willing to work in a limited fashion with museum staff to finalize the online outbreak exhibit. This may include a request for additional information or materials (such as pictures, video, an epi curve, outbreak questionnaire, etc).

Eligibility: Anyone can submit an outbreak investigation. There are no restrictions for eligibility based upon geographic location or whether or not you participated in the submitted outbreak investigation. T-shirts will be shipped to the winners within 2-8 weeks (to the best of our ability).

How To Enter: You may enter this contest as an individual or a small team. To enter the contest, a historically significant outbreak investigation report must be submitted to International Outbreak Museum staff.

To submit your exhibit, please fill out the online exhibit submission form (below) and submit it along with any supporting materials (images, captions for images, publications, presentations, epi curve, etc) to the following email address:


NOTE: For examples of current online outbreak exhibits, please click “Outbreak Exhibits” at the top of this page or click here.

   To download this form as a PDF, Click Here
   To download this form as a Microsoft Office Word Document, Click Here


  • 1) Investigation jurisdiction (local Health Department, state Health Department, Federal, Other)
  • 2) Pathogen
  • 3) Vehicle/venue
  • 4) Case counts (confirmed/presumptive/suspect)
  • 5) Geographic distribution of cases (single county/state-wide/multi-state/International)
  • 6) First and last onset dates
  • 7) Pictures! (please include captions for each photo submitted as part of exhibit)
  • 8) Number of positive non-human specimens (water, environment, food, etc.)


  • 9) Publication citation (optional)
  • 10) Presentations (optional)
  • 11) Investigation tools (e.g., questionnaire) (optional)


Please limit the 3 sections below to a total of ~1000 words

  1. 1) Abstract: Background, Methods, Results, Conclusions
  2. 2) Highlights: unique features or reasons why the outbreak is important
  3. 3) Lessons learned
  4. Please send this completed form to


Raw Flour E. coli

Outbreak: Raw Flour E. Coli 
Product: Raw FlourInvestigation Start Date: 6/1/2016
Location: MultistateEtiology: E. coli O121
Earliest known case onset date: 12/21/2015Latest case onset date: 9/5/2016
Confirmed Case Count: 63Positive Environmental Samples: 39
Hospitalizations: 17Deaths: 0

CDC worked with public health and regulatory officials in many states and the U.S. Food and Drug Administration (FDA) to investigate a multistate outbreak of Shiga toxin-producing Escherichia coli (STEC) infections.

Public health investigators used the PulseNet system to identify illnesses that may be part of this outbreak. PulseNet, coordinated by CDC, is the national subtyping network of public health and food regulatory agency laboratories. PulseNet performs DNA fingerprinting on STEC bacteria isolated from ill people by using techniques called pulsed-field gel electrophoresis (PFGE) and whole genome sequencing (WGS). CDC PulseNet manages a national database of these DNA fingerprints to identify possible outbreaks.

Sixty-three people infected with the outbreak strains of STEC O121 or STEC O26 were reported from 24 states. A list of the states and the number of cases in each can be found on the Case Count Map page. WGS showed that isolates from ill people were closely related genetically. This close genetic relationship means that people in this outbreak were more likely to share a common source of infection.

Illnesses started on dates ranging from December 21, 2015 to September 5, 2016. Ill people range in age from 1 year to 95, with a median age of 18. Seventy-six percent of ill people were female. Seventeen ill people were hospitalized. One person developed hemolytic uremic syndrome, a type of kidney failure, and no deaths were reported.


Flour is a raw, minimally processed product intended to be cooked before consumption. Although several previous Shiga-toxin-producing E. coli (STEC) outbreak investigations in the United States suspected contaminated flour as the source, none had been proven. In February 2016, PulseNet, the laboratory network for foodborne disease surveillance, detected a 12-state cluster of STEC O121 infections having the same rare genetic fingerprint. A multistate outbreak investigation was initiated. An additional STEC O26 strain was linked to the outbreak after testing of implicated flour.

A case was defined as infection with an outbreak strain of STEC O121 or O26 occurring between December 21, 2015, and September 5, 2016. Case-patients were interviewed about foods and other exposures in the week before illness onset. We performed univariable matched exact conditional logistic regression to identify exposures associated with illness, comparing them to exposures among people with reportable non-STEC enteric infections (primarily salmonellosis and campylobacteriosis), matched on age, gender, and state of residence. Four controls were sought for each case and interviewed by state and local health officials. Samples of suspected products were collected and cultured for STEC. A common point of contamination was sought through traceback. Whole genome sequencing (WGS) was performed on selected clinical and food isolates.

Fifty-six cases of STEC O121 and one case of STEC O26 infection were identified in 24 states; seventeen people were hospitalized; none died. Using General Mills flour (OR 21.0, 95% CI 4.7‒94.4) and tasting unbaked homemade dough or batter (OR 36.0, 95% CI 4.6‒280.2) were both significantly associated with illness. Traceback identified a common flour production facility. Three illnesses were in children exposed to raw dough for playing at several locations of a single restaurant chain. Leftover flour samples collected from cases’ homes and additional samples collected from the flour producer were tested, and five STEC strains were isolated (one STEC O26, three STEC O121, and one STEC O103). All isolates tested were closely related genetically.

This is the first investigation to link definitively an outbreak of STEC infections to raw flour. Nearly 250 products containing the implicated flour were recalled by the flour producer as well as by several companies that used recalled flour. Consumers should not eat products containing uncooked flour. Using uncooked dough for play should also be discouraged at restaurants and home. Flour producers should consider adding prominent packaging labels to warn consumer not to eat undercooked or raw flour. Foodborne illnesses associated with raw flour are likely preventable if appropriate control measures are taken from grain fields and production facilities to restaurants and consumers.

Publication citation: Crowe SJ., et al. Shiga toxin–producing E. coli infections associated with flour. New Engl J Med 2017; 377: 2036n43.


Listeria Soft Cheese

Outbreak: Listeria Soft Cheese 
Product: Soft CheeseInvestigation Start Date: 06/27/2013
Location: Multi-StateEtiology: Listeria monocytogenes
Earliest known case onset date: 5/14/2013Latest case onset date: 6/16/2013
Confirmed / Presumptive Case Count: 5 / 0Positive Samples (Food / Environmental): 18 / 1

Five outbreak cases of listeriosis were identified in Minnesota, Indiana, Illinois, and Ohio. The use of the CDC Listeria Initiative questionnaire to identify a soft cheese signal, along with re-interviewing cases (or their proxies) and calling restaurants to get specific details on cheese brand and cheese type were critical in identifying the outbreak vehicle. Environmental isolates in the PulseNet national database that were collected from the Crave Brothers cheese production plant (and indistinguishable from outbreak case isolates) were also a key supporting clue. These clues led to the testing of Crave Brothers cheese samples and detection of the outbreak stain in this product.

This investigation resulted in perhaps the fastest identification of a commercially distributed food vehicle of a listeriosis outbreak; the cluster was detected in Minnesota on June 27 and public health interventions were implemented on July 3 (6 days later). This was a remarkable accomplishment, especially with only 5 cases to work with. The rapidity of the investigation can be attributed primarily to the urgency displayed by the lead investigator and the aggressive acquisition of details on brand and type of soft cheese consumed, in collaboration with epidemiologists in the other affected states.

The MN CoE has created a detailed case study about this investigation that can be used for training public health practitioners in outbreak investigation. Public Health students love it too!


A multi-state outbreak of listerosis associated with soft cheese from Crave Brothers.


On June 27, 2013, the Minnesota Department of Health (MDH) Public Health Laboratory (PHL) determined that two clinical Listeria monocytogenes (LM) isolates submitted through routine surveillance had indistinguishable PFGE patterns. A review of the national PulseNet database revealed three additional LM isolates with the outbreak PFGE pattern in Indiana, Illinois, and Ohio. A multi-state investigation was initiated.


Five cases from three states were ultimately identified in this outbreak (Minnesota, 2; Indiana, 1; Illinois, 1; and Ohio, 1). In addition, subsequent whole genome sequencing efforts identified another likely case in Texas, but exposure information was not available. Four (80%) cases were female and the median age of the cases was 55 years (range, 31 to 67 years). All five cases were hospitalized and one died. Listeria Initiative case report forms were completed for all five cases and all five reported consuming a variety of soft cheeses. Potato salad was reported by both Minnesota cases; however, no other cases reported consuming it. Epidemiologists conducted a case-case comparison study to compare outbreak case food consumption rates against estimated background consumption rates using non-outbreak associated LM cases. LM cases with the outbreak PFGE pattern were significantly more likely than sporadic LM cases to have consumed any soft cheese (odds ratio, 16.8; 95% confidence interval, 1.2–∞) and yogurt (odds ratio, 7.5; 95% confidence interval, 1.3–∞). Review of the initial interviews revealed that cases reported different brands of yogurt, making this an unlikely source of illness.

The Minnesota cases were re-interviewed to get additional specific product details on the soft cheese exposures. The first Minnesota case reported consuming a cheese plate at a restaurant in North Dakota. The investigator called the restaurant which reported that the cheese plate included Crave Brothers Les Frères cheese. The Illinois case reported consuming Crave Brothers Les Frères cheese purchased from a grocery store. The Indiana case reported consuming a cheese plate at a restaurant, and a review for the restaurant menu indicated that Crave Brothers Les Frères was served on the cheese plate. The Ohio case was re-interviewed and also reported consuming a cheese plate at a restaurant which included Crave Brothers Petit Frère cheese. PulseNet identified non-human isolates in PulseNet with the outbreak PFGE pattern that had been collected from the Crave Brothers cheese production facility in 2010 and 2011.

On July 1, Crave Brothers cheese was collected for testing from two locations of the grocery chain where MN case #2 had purchased the product, and submitted to Minnesota Department of Agriculture (MDA) for LM testing. In addition, leftover blue cheese, Irish cheddar, parmesan cheese, and an unknown hard cheese were collected for testing from MN case #2’s home. By July 3, Enzyme-linked Fluorescent Assay results indicated that a wedge of Crave Brothers Petite Frère with Truffles from one grocery store location in Minnesota, two wedges of Crave Brothers Les Frères from another location of the same grocery store chain in Minnesota, and a wedge of blue cheese collected from MN Case #2’s home all tested positive for LM. A traceback of the Crave Brothers cheese consumed by the cases revealed that the cheeses had different distributors, which indicated that the source of the contamination was the Crave Brothers plant.
A press release was issued warning consumers not to eat Crave Brothers Les Frères, Petit Frère, and Petit Frère with truffles and Crave Brothers issued a voluntary nationwide recall of these cheeses.
The spouse of the second Minnesota case called investigators to report that he thought they had purchased Crave Brothers Petit Frère. He did not have a receipt but had used a credit card to make the purchase. The investigator contacted the grocery store and gave them the credit card transaction numbers, the date of the transaction, and the dollar amount of the transaction. Using this information the store was able to reprint the original receipt with included Crave Brothers Petit Frère.


This was a multi-state outbreak of LM infections associated with Crave Brothers cheese. The use of the Listeria initiative questionnaire along with re-interviewing cases (or their proxies) and calling restaurants to get specific details on cheese brand/type were critical in identifying the outbreak vehicle.
Environmental isolates in the PulseNet national database that were collected from the Crave Brothers cheese production plant were also a key supporting clue. This investigation resulted in perhaps the fastest identification of a commercially distributed food vehicle of a listeriosis outbreak.