I love Mondays in Australia because it’s Sunday in the U.S., football and hockey are on TV for background, the kid is at school when not in France, and I write (Sorenne painting in France).
Fourteen years ago, me and Chapman went on a road trip to Prince George (where Ben thought he would be eaten by bears) to Seattle, then to Manhattan, Kansas, where in the first week I met a girl, got a job, and then spinach happened.
Leafy greens are still covered in shit.
I am drowning in nostalgia, but things haven’t changed, and, as John Prine wrote, all the news just repeats itself.
Same with relationships.
Former U.S. Food and Drug Administration food safety chief, David Acheson, writes that on October 31, 2019, FDA announced a romaine lettuce E. coli O157:H7 outbreak for which the active investigation had ended and the outbreak appeared to be over. As such FDA stated there was no “current or ongoing risk to the public” and no avoidance of the produce was recommended.
Since that announcement, however, I have seen a number of articles condemning FDA and CDC. Why? Because the traceback investigation of the outbreak began in mid-September when CDC notified FDA of an illness cluster that had sickened 23 people across 12 states. So why the delay in announcing it to the public?
Despite the critical (and rather self-serving; always self-serving) stance on the “inexcusable” delay taken by a prominent foodborne illness attorney and his Food Safety “News” publication – which blasted a headline FDA “hid” the outbreak – my stance, having been an FDA official myself involved in outbreak investigations, is that the delay was practical and sensible.
Why? As FDA states right in its announcement:
When romaine lettuce was identified as the likely source, the available data indicated that the outbreak was not ongoing and romaine lettuce eaten by sick people was past its shelf life and no longer available for sale.
Even once romaine was identified as the likely cause, no common source or point of contamination was identified that could be used to further protect the public.
During the traceback investigation, the outbreak strain was not detected in any of the samples collected from farms, and there were no new cases.
Thus, neither FDA nor CDC identified any actionable information for consumers.
So, if it is not in consumers’ best interest to publicize an issue that no longer exists, why should they be driven away from a healthy food alternative? Why should unfounded unease be generated that will damage the industry, providing no benefit for consumers but ultimately impacting their pockets? There is just no upside to making an allegation without information. We’ve seen the impact on consumers and the industry when an announcement of a suspected food turns out to be incorrect; specifically “don’t eat the tomatoes” when it turned out to be jalapeno and serrano peppers. Having learned from such incidents, FDA’s approach is: If we don’t have a message that will help protect the public, then there is no message to be imparted.
So, rather than condemn FDA and CDC, I would commend them for getting the balance correct. And, perhaps, instead of any condemning, we should be working together to get the answers faster, to get outbreak data through better, faster, more efficient and coordinated traceability. Our entire system is too slow – a topic we have discussed many times in these newsletters.
The public and the scientific community need to be informed to prevent additional people from barfing.
I also rarely eat lettuce of any sort because it is overrated and the hygiene controls are not adequate.
Greek salad without lettuce is my fave.
Going public: Early disclosure of food risks for the benefit of public health
NEHA, Volume 79.7, Pages 8-14
Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell
Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.
I’ll leave the summary of two antimicrobial resistance reports to my friend and hockey colleague (and he’s a professor/veterinarian) Scott Weese of the Worms & Germs Blog (he’s the semi-bald dude behind me in this 15-year-old pic; I’m the goalie; too many pucks to the head):
Two reports came out this week, both detailing the scourge of antibiotic resistance.
They’re both comprehensive, with a combined >400 pages explaining that this is a big problem.
I’m not going try to summarize the reports. I’ll just pick out a few interesting tidbits.
From the CCA report (Canada):
According to their modelling, first-line antimicrobials (those most commonly used to treat routine infections) helped save at least 17,000 lives in 2018 while generating $6.1 billion in economic activity in Canada. “This contribution is at risk because the number of effective antimicrobials are running out.”
Antimicrobial resistance was estimated to reduce Canada’s GDP by $2 billion in 2018. That’s only going to get worse unless we get our act together. It’s estimated that by 2050, if resistance rates remain unchanged, the impact will be $13 billion per year. If rates continue to increase, that stretches to $21 billion. Remember, that’s just for Canada, a relatively small country from a population standpoint.
Healthcare costs due to resistance (e.g. drugs, increased length of stay in hospital) accounted for $1.4 billion in 2018. But remember that people who die from resistant infections can actually cost less. If I get a serious resistant infection and die quickly, my healthcare costs are pretty low since I didn’t get prolonged care. All that to say that dollar costs alone don’t capture all the human aspects. Regardless, this cost will likely increase to $20-40 billion per year by 2050.
In terms of human health, resistant infections were estimated to contribute to 14,000 deaths in Canada in 2018, with 5,400 of those directly attributable to the resistant infection (i.e. those deaths would not have occurred if the bug was susceptible to first line drugs). That makes resistance a leading killer, and it’s only going to get worse.
The document’s dedication says a lot. “This report is dedicated to the 48,700 families who lose a loved one each year to antibiotic resistance or Clostridioides difficile, and the countless healthcare providers, public health experts, innovators, and others who are fighting back with everything they have.”
The forward has some great messages too:
To stop antibiotic resistance, our nation must:
Stop referring to a coming post-antibiotic era—it’s already here. You and I are living in a time when some miracle drugs no longer perform miracles and families are being ripped apart by a microscopic enemy. The time for action is now and we can be part of the solution.
Stop playing the blame game. Each person, industry, and country can affect the development of antibiotic resistance. We each have a role to play and should be held accountable to make meaningful progress against this threat.
Stop relying only on new antibiotics that are slow getting to market and that, sadly, these germs will one day render ineffective. We need to adopt aggressive strategies that keep the germs away and infections from occurring in the first place.
Stop believing that antibiotic resistance is a problem “over there” in someone else’s hospital, state, or country—and not in our own backyard. Antibiotic resistance has been found in every U.S. state and in every country across the globe. There is no safe place from antibiotic resistance, but everyone can take action against it. Take action where you can, from handwashing to improving antibiotic use.
Some might say it’s alarmist. However, I don’t think it’s alarmist when someone really should be raising the alarm. We need to talk about it more, not less. We need to get people (including the general public, healthcare workers, farmers, veterinarians, policymakers) on board, to realize it’s a big issue that needs to be addressed now. “Short term pain for long-term gain” certainly applies here. We can keep delaying and the numbers will keep going up, or we can invest in solutions.
The numbers are scary but specific numbers don’t really matter in many ways. “Lots” is all we should have to know to get motivated. However, decision-makers like numbers, so these numbers hopefully will be useful to show the impact and potential benefits of investing in this problem, and motivate them to put money into antimicrobial stewardship. Saving lives should be enough, but that often doesn’t cut it. Antibiotic resistance doesn’t have a good marketing campaign. Everyone knows why people were wearing pink last month and why there are some pretty dodgy moustaches this month. Those are important issues, for sure. However, considering the overall impact, antibiotic stewardship needs to get more people behind it if we’re going to effect change.
In an ongoing effort to understand sources of foodborne illness in the United States, the Interagency Food Safety Analytics Collaboration (IFSAC) collects and analyzes outbreak data to produce an annual report with estimates of foods responsible for foodborne illnesses caused by pathogens. The report estimates the degree to which four pathogens – Salmonella, E. coli O157, Listeria monocytogenes, and Campylobacter – and specific foods and food categories are responsible for foodborne illnesses.
The Centers for Disease Control and Prevention (CDC) estimates that, together, these four pathogens cause 1.9 million foodborne illnesses in the United States each year. The newest report (PDF), entitled “Foodborne illness source attribution estimates for 2017 for Salmonella, Escherichia coli O157, Listeria monocytogenes, and Campylobacter using multi-year outbreak surveillance data, United States,” can be found on the IFSAC website.
The updated estimates, combined with other data, may help shape agency priorities and inform the creation of targeted interventions that can help to reduce foodborne illnesses caused by these pathogens. As more data become available and methods evolve, attribution estimates may improve. These estimates are intended to inform and engage stakeholders and to improve federal agencies’ abilities to assess whether prevention measures are working.
Foodborne illness source attribution estimates for 2017 for salmonella, Escherichia coli O157, listeria monocytogenes, and campylobacter using multi-year outbreak surveillance data, United States, Sept.2019
In interviews, 12 (71%) of 17 ill people reported contact with a turtle.
This investigation is ongoing and CDC will provide updates when more information is available.
Turtles can carry Salmonella germs in their droppings while appearing healthy and clean. These germs can easily spread to their bodies, tank water, and habitats. People can get sick after they touch a turtle or anything in their habitats.
Always wash hands thoroughly with soap and water right after touching, feeding, or caring for a turtle or cleaning its habitat.
Adults should supervise handwashing for young children.
Don’t kiss or snuggle turtles, because this can spread Salmonella germs to your face and mouth and make you sick.
Don’t let turtles roam freely in areas where food is prepared or stored, such as kitchens.
Clean habitats, toys, and pet supplies outside the house when possible.
Avoid cleaning these items in the kitchen or any other location where food is prepared, served, or stored.
Pick the right pet for your family.
CDC and public health officials in several states are investigating a multistate outbreak of human Salmonella Oranienburg infections linked to contact with pet turtles.
Public health investigators are using the PulseNet system to identify illnesses that may be part of this outbreak. PulseNet is the national subtyping network of public health and food regulatory agency laboratories coordinated by CDC. DNA fingerprinting is performed on Salmonella bacteria isolated from ill people by using a standardized laboratory and data analysis method called whole genome sequencing (WGS). CDC PulseNet manages a national database of these sequences that are used to identify possible outbreaks. WGS gives investigators detailed information about the bacteria causing illness. In this investigation, WGS showed that bacteria isolated from ill people were closely related genetically. This means that people in this outbreak are more likely to share a common source of infection.
Ill people reported contact with red-eared sliders and other turtles that were larger than four inches in length. Previous Salmonella outbreaks have been linked to turtles with a shell length less than four inches. Due to the amount of Salmonella illnesses related to these small turtles, the U.S. Food and Drug Administration banned the sale and distributionexternal icon of turtles with shells less than four inches long as pets.
Regardless of where turtles are purchased or their size, turtles can carry Salmonella germs that can make people sick. Pet owners should always follow steps to stay healthy around their pet.
This investigation is ongoing, and CDC will provide updates when more information becomes available.
This is the CIDRAP summary of the latest CDC number crunching on microorganisms that lead to barfing.
The Centers for Disease Control and Prevention (CDC) late last week released a summary of foodborne illnesses in 2017 based on an annual analysis of data from the Foodborne Disease Outbreak Surveillance System, and norovirus was the most common pathogen reported, responsible for 46% of illnesses. Salmonella and Shiga toxin–producing Escherichia coli were also linked to a substantial number of outbreaks.
In 2017, the CDC tracked 841 foodborne outbreaks, which included 14,481 illnesses, 827 hospitalizations, 20 deaths, and 14 food product recalls. A single etiologic agent was confirmed in 395 outbreaks (47%), which are defined as two or more related cases.
Tainted seafood and poultry were tied with causing the most outbreaks, with mollusks (41 outbreaks), fish (37), and chicken (23) the specific food items most often implicated. The most outbreak-associated illnesses were from turkey (609 illnesses), fruits (521), and chicken (487), the CDC said.
California had the most outbreaks (107), followed by Ohio (69), and Washington state (67).
As in past years, restaurants with sit-down dining were the most commonly reported locations for food preparation associated with outbreaks (366).
For a country that still proclaims, we “enjoy the safest food supply in the world” in U.S. Department of Agriculture missives, when we’ve been arguing reduced risk is a better message for 25 years and that there are so many countries with the self-proclaimed title of safest food in the world they can’t all be right – it’s alarming that Mycobacterium bovis has been transmitted from deer to a human.
My dad went a few times but I’m not sure if he enjoyed it or not.
The U.S. Centers for Disease Control reports that in May 2017, the Michigan Department of Health and Human Services was notified of a case of pulmonary tuberculosis caused by Mycobacterium bovis in a man aged 77 years. The patient had rheumatoid arthritis and was taking 5 mg prednisone daily; he had no history of travel to countries with endemic tuberculosis, no known exposure to persons with tuberculosis, and no history of consumption of unpasteurized milk. He resided in the northeastern Lower Peninsula of Michigan, which has a low incidence of human tuberculosis but does have an enzootic focus of M. bovis in free-ranging deer (Odocoileus virginianus). The area includes a four-county region where the majority of M. bovis–positive deer in Michigan have been found.
Statewide surveillance for M. bovis via hunter-harvested deer head submission has been ongoing since 1995; in 2017, 1.4% of deer tested from this four-county region were culture-positive for M. bovis, compared with 0.05% of deer tested elsewhere in Michigan. The patient had regularly hunted and field-dressed deer in the area during the past 20 years. Two earlier hunting-related human infections with M. bovis were reported in Michigan in 2002 and 2004. In each case, the patients had signs and symptoms of active disease and required medical treatment.
Whole-genome sequencing of the patient’s respiratory isolate was performed at the National Veterinary Services Laboratories in Ames, Iowa. The isolate was compared against an extensive M. bovis library, including approximately 900 wildlife and cattle isolates obtained since 1993 and human isolates from the state health department. This 2017 isolate had accumulated one single nucleotide polymorphism compared with a 2007 deer isolate, suggesting that the patient was exposed to a circulating strain of M. bovis at some point through his hunting activities and had reactivation of infection as pulmonary disease in 2017.
Whole-genome sequencing also was performed on archived specimens from two hunting-related human M. bovis infections diagnosed in 2002 (pulmonary) and 2004 (cutaneous) that were epidemiologically and genotypically linked to deer (3). The 2002 human isolate had accumulated one single nucleotide polymorphism since sharing an ancestral genotype isolated from several deer in Alpena County, Michigan, as early as 1997; the 2004 human isolate shared an identical genotype with a grossly lesioned deer harvested by the patient in Alcona County, Michigan, confirming that his infection resulted from a finger injury sustained during field-dressing. The 2002 and 2017 cases of pulmonary disease might have occurred following those patients’ inhalation of aerosols during removal of diseased viscera while field-dressing deer carcasses.
In Michigan, deer serve as maintenance and reservoir hosts for M. bovis, and transmission to other species has been documented. Since 1998, 73 infected cattle herds have been identified in Michigan, resulting in increased testing and restricted movement of cattle outside the four-county zone. Transmission to humans also occurs, as demonstrated by the three cases described in this report; however, the risk for transmission is understudied.
Similar to Mycobacterium tuberculosis, exposure to M. bovis can lead to latent or active infection, with risk for eventual reactivation of latent disease, especially in immunocompromised hosts. To prevent exposure to M. bovis and other diseases, hunters are encouraged to use personal protective equipment while field-dressing deer. In addition, hunters in Michigan who submit deer heads that test positive for M. bovis might be at higher risk for infection, and targeted screening for tuberculosis could be performed. Close collaboration between human and animal health sectors is essential for containing this zoonotic infection.
Notes from the Field: Zoonotic mycobacterium bovis disease in deer hunters—Michigan, 2002-2017
James Sunstrum, MD1; Adenike S hoyinka, MD2; Laura E. Power, MD2,3; Daniel Maxwell, DO4; Mary Grace Stobierski, DVM5; Kim Signs, DVM5; Jennifer L. Sidge, DVM, PhD5; Daniel J. O’Brien, DVM, PhD6; Suelee Robbe-Austerman, DVM, PhD7; Peter Davidson, PhD5
According to the U.S. Centers for Disease Control (CDC) an outbreak of Shiga toxin-producing Escherichia coli (STEC) — E. coli O103 and E. coli O121 — linked to ground bison appears to be over.
CDC, several states, the U.S. Food and Drug Administration, and the Canadian Food Inspection Agency investigated a multistate outbreak of infections.
On July 16, 2019, Northfork Bison Distributions, Inc., in Saint-Leonard, Quebec, Canada, recalled external icon ground bison produced between February 22, 2019, and April 30, 2019. Recalled ground bison was sold to distributors as ground bison and bison patties, referred to as Bison Burgers and/or Buffalo Burgers. Recalled ground bison was also sold to retailers in 4-ounce burger patties.
Do not eat, sell, or serve recalled Northfork Bison products.
As of September 13, 2019, this outbreak appears to be over.
A total of 33 people infected with the outbreak strain of STEC O103 and STEC O121 were reported from eight states.
Eighteen people were hospitalized. No cases of hemolytic uremic syndrome, a type of kidney failure, were reported. No deaths were reported.
When ordering at a restaurant, ask that ground bison burgers be cooked to an internal temperature of at least 160°F.
The U.S. Centers for Disease Control reports that waterborne hepatitis A outbreaks have been reported to CDC. Person-to-person transmission of hepatitis A has increased in recent years.
Reported drinking water–associated hepatitis A outbreaks have declined since introduction of universal childhood vaccination recommendations and public drinking water regulations. However, unvaccinated persons who use water from untreated private wells remain at risk.
Public health officials should raise awareness of risks associated with untreated ground water among users of private wells and of options for private well testing and treatment. Water testing and treatment are important considerations to protect persons who use these unregulated systems from HAV infection.
Hepatitis A virus (HAV) is an RNA virus primarily transmitted via the fecal-oral route and, in rare cases, causes liver failure and death in infected persons. Although drinking water–associated hepatitis A outbreaks in the United States are rarely reported (1), HAV was the most commonly reported etiology for outbreaks associated with untreated ground water during 1971–2008 (2), and HAV can remain infectious in water for months (3). This report analyzes drinking water–associated hepatitis A outbreaks reported to the Waterborne Disease and Outbreak Surveillance System (WBDOSS) during 1971–2017. During that period, 32 outbreaks resulting in 857 cases were reported, all before 2010. Untreated ground water was associated with 23 (72%) outbreaks, resulting in 585 (68.3%) reported cases. Reported outbreaks significantly decreased after introduction of Advisory Committee on Immunization Practices (ACIP) hepatitis A vaccination recommendations* and U.S. Environmental Protection Agency’s (USEPA) public ground water system regulations.† Individual water systems, which are not required to meet national drinking water standards,§ were the only contaminated drinking water systems to cause the last four reported hepatitis A outbreaks during 1995–2009. No waterborne outbreaks were reported during 2009–2017. Water testing and treatment are important considerations to protect persons who use these unregulated systems from HAV infection.
U.S. states and territories have voluntarily reported waterborne disease outbreaks to WBDOSS since 1971.¶ Waterborne hepatitis A outbreaks (1971–2017) reported as of March 13, 2018, were reviewed. An outbreak of hepatitis A was defined as two or more cases of HAV infection epidemiologically linked by time and location of water exposure. To compare occurrence with other waterborne exposure pathways, outbreaks reviewed included those caused by drinking, recreational, environmental (i.e., nondrinking, nonrecreational water), or undetermined water exposures.** As described previously (1), data reviewed included location; date of first illness; estimated number of primary cases, hospitalizations, and deaths; water system type according to USEPA Safe Drinking Water Act definitions (i.e., community, noncommunity, and individual); setting of exposure; drinking water sources (i.e., ground water, surface water, and unknown); and water system characteristics.†† Community and noncommunity water systems are public water systems that have 15 or more service connections or serve an average of 25 or more residents for ≥60 days per year.§§ A community water system serves year-round residents of a community, subdivision, or mobile home park. A noncommunity water system serves an institution, industry, camp, park, hotel, or business. Individual water systems are small systems (e.g., private wells and springs) not owned or operated by a water utility that have fewer than 15 connections or serve fewer than 25 persons. The number of outbreaks before and after public health interventions were compared; chi-squared tests were used to identify significant (p-value<0.05) differences. Data were analyzed using SAS software (version 9.4; SAS Institute) and visualized in ArcGIS (version 10.6.1; Environmental Systems Research Institute).
Thirty-two drinking water–associated hepatitis A outbreaks were reported to CDC during 1971–2017; the last one occurred in 2009 (Table). These drinking water–associated outbreaks accounted for 857 cases (range = 2–50), with no reported deaths. Data on number of deaths were unavailable for three outbreaks. Data on hospitalizations were unavailable for all outbreaks. Outbreaks occurred in 18 states, all in the lower continental United States (Figure 1). One environmental outbreak (1975) and one recreational water outbreak (1989) were reported during this period, but were excluded from this analysis.
The most commonly reported water system type associated with an outbreak was individual, accounting for 13 of 32 (41%) outbreaks and 257 of 857 (30.0%) cases, followed by community (10 [31%] outbreaks; 241 [28.1%] cases) and noncommunity (9 [28%] outbreaks; 359 [41.9%] cases). All individual water systems with outbreaks were supplied by private wells or springs. The majority of all drinking water outbreaks and cases were associated with systems supplied by ground water (30 [94%] outbreaks; 804 [93.8%] cases) and with an absence of water treatment (23 [72%] outbreaks; 585 [68.3%] cases).
The incidence of reported drinking water–associated hepatitis A outbreaks significantly decreased after introduction of the 1989 USEPA Total Coliform and Surface Water Treatment Rules (77% decline from 1971–1989 [24 outbreaks] to 1990–2017 [eight]; p = 0.003), the 1996 ACIP hepatitis A vaccination recommendations (87% decline from 1971–1996  to 1997–2017 [three]; p<0.001), and the 2006 Ground Water Rule and expanded ACIP vaccine recommendations (78% decline from 1971–2006  to 2007–2017 [two]; p = 0.038) (Figure 2). From 1995 through 2009, all four hepatitis A drinking water–associated outbreaks, resulting in 35 cases, were attributed to individual water systems using untreated ground water sources. No water-associated hepatitis A outbreaks have been reported since July 2009.
Reported drinking water–associated hepatitis A outbreaks have declined since reporting began in 1971, and none have been reported since 2009, mirroring the overall decline in U.S. cases (4,5). Vaccination for hepatitis A, combined with USEPA regulations that require testing and, where necessary, corrective actions or treatment for drinking water supplies, likely played a role in reducing reported hepatitis A drinking water–associated outbreaks.
Vaccination efforts have led to significant changes in hepatitis A epidemiology (4,6,7). HAV infection rates in the United States have decreased since the introduction of hepatitis A vaccine in 1995 (4,5). Vaccine recommendations were originally targeted to children in communities with high rates of hepatitis A infections west of the Mississippi and other groups at risk (e.g., international travelers, men who have sex with men, illicit drug users, persons with clotting factor disorders, and persons with occupational risk). By 2006, routine hepatitis A vaccination was recommended for all children aged ≥l year regardless of geographic area of residence (5). Although vaccination was never recommended for users of individual ground water systems, this group likely benefited from the recommendations targeting children and other groups at risk. Incidence of HAV infection is now lowest among persons aged 0–19 years (4). However, the proportion of HAV-associated hospitalizations steadily increased during 1999–2011, likely because of more severe disease in older adults, with persons aged ≥80 years experiencing the highest rates of infection (6). The number of hepatitis A cases in the United States reported to CDC increased by 294% during 2016–2018, compared with the period 2013–2015 (8), primarily because of community-wide outbreaks in persons reporting homelessness or drug use (7). ACIP recommends vaccination to persons who use drugs and recently expanded recommendations to persons experiencing homelessness.¶¶
Reported drinking water–associated hepatitis A outbreaks were most commonly linked to individual water systems that used wells with untreated ground water. Recreational and environmental outbreaks were only reported twice, suggesting that drinking water is a more common waterborne exposure pathway for hepatitis A. Nearly 43 million U.S. residents, or 13% of the population, are served by individual water systems, primarily from ground water sources (https://pubs.er.usgs.gov/publication/cir1441external icon). Untreated ground water sources were associated with 30% of all drinking water–associated outbreaks reported to CDC during 1971–2008 (1). The USEPA Total Coliform and Surface Water Treatment Rules of 1989 and Ground Water Rule of 2006 provide enhanced safety measures for public water systems using ground water sources and might have contributed to the absence of reported hepatitis A outbreaks linked to community water sources since 1990. However, federal regulations do not apply to individual water systems, which often have inadequate or no water treatment (9). Private wells or springs were the only contaminated drinking water systems to cause the last four reported hepatitis A outbreaks during 1995–2009. CDC recommends that owners of private wells test their water annually for indicators of fecal contamination (https://www.cdc.gov/healthywater/drinking/private/wells/testing.html). Factors contributing to fecal contamination of ground water include nearby septic systems or sewage, weather patterns (e.g., heavy rainfall), improper well construction and maintenance, surface water seepage, and hydrogeologic formations (e.g., karst limestone) that allow for rapid pathogen transport (2,9).
The findings in this report are subject to at least three limitations. First, waterborne hepatitis A outbreak reporting is through a passive, voluntary surveillance system; health departments have varying capacity to detect, investigate, and report outbreaks, which might result in incomplete data on outbreak occurrence and characteristics within and across jurisdictions. Thus, outbreak surveillance data might underestimate the actual number of drinking water–associated hepatitis A outbreaks and might underreport information regarding health outcomes such as cases of illness. Second, attributing the source of an outbreak to individual water systems can be particularly difficult because hepatitis A can also be spread through person-to-person transmission within a household. Finally, outbreak data before 2009 did not include case-specific information; thus, demographic factors, including age, could not be assessed.
Drinking water–associated hepatitis A outbreaks have declined and essentially stopped, likely in large part because of the introduction of an efficacious vaccine as part of the routine childhood immunization program and microbial drinking water regulations for public water systems. The degree to which these interventions have contributed to the decline in outbreaks is uncertain. However, waterborne outbreak surveillance data is not yet finalized for 2018, and the recent increase in person-to-person transmission of hepatitis A (7,8) has the potential to cause a resurgence in waterborne outbreaks through increased fecal HAV contamination of private ground water supplies. Outbreak data suggest that individual water systems, primarily those systems drawing untreated ground water from wells, pose the highest risk for causing drinking water–associated hepatitis A outbreaks. These systems are not regulated by USEPA; CDC recommends that owners evaluate their well water quality at least yearly. If indicators of fecal contamination are detected, remediation and treatment of private well water is recommended. Guidance on private well testing and treatment solutions for microbial contamination is provided by USEPA (https://www.epa.gov/privatewells/protect-your-homes-waterexternal icon) and CDC (https://www.cdc.gov/healthywater/drinking/private/wells/index.html). Although the current nationwide outbreak of hepatitis A is not water-associated, considering ground water as a possible transmission route is warranted during community-wide outbreaks of hepatitis A. Ground water can be contaminated with HAV during community transmission of hepatitis A, increasing the risk for persons using untreated water. Public health education about the risks associated with drinking untreated ground water from individual systems, as well as relevant safety measures (i.e., water testing, water treatment, and vaccination), is needed to prevent future drinking water–associated hepatitis A outbreaks.
US: Impact of public health interventions on drinking water-associated outbreaks of hepatitis A-United States, 1971-2017
The U.S. Centers for Disease Control reports in August 2018, two Oregon patients with diagnosed Salmonella infection were interviewed using a standard enteric illness questionnaire; both patients reported having eaten raw cake mix.
Standardized interview questionnaire data collected from 207 Oregon patients with salmonellosis in 2017 indicated a 5% rate of consumption of raw “cake mix or cornbread mix” (Oregon Health Authority, unpublished data, 2017). The binomial probability that both 2018 patients were exposed to raw cake mix by chance was determined to be 0.003, prompting the Oregon Health Authority (OHA) to collect and test the contents of 43 boxes of unopened cake mix of various brands from six retail locations. OHA sent samples to the Institute for Environmental Health Laboratories in Lake Forest Park, Washington, for pathogen testing. Salmonella Agbeni was isolated from an unopened box of white cake mix from manufacturer A, and whole genome sequencing (WGS) data describing the isolate were uploaded to the U.S. National Library of Medicine’s National Center for Biotechnology Information (NCBI) website (https://www.ncbi.nlm.nih.gov/pathogensexternal icon). OHA used the NCBI database to compare sequence data with the cake mix isolate (PNUSAS056022) and then consulted CDC’s System for Enteric Disease Response, Investigation, and Coordination (SEDRIC), a web-based, outbreak investigation tool designed for collaborative, multistate investigations of enteric disease outbreaks.* On October 19, OHA determined that clinical isolates from four patients from Maryland, Ohio, and Wisconsin, with specimen isolation dates ranging from June to September 2018, were genetically related to the Salmonella Agbeni isolate from the unopened box of white cake mix, within four single nucleotide polymorphisms (SNPs).
On October 22, 2018, OHA notified state public health counterparts in the three states of this finding and inquired about raw cake mix exposures among their patients. The Wisconsin patient reported having consumed an entire box of raw white cake mix over several days during the likely exposure period. In addition, WGS analysis indicated that this clinical isolate was closely related genetically (within one SNP) to the isolate cultured from the Oregon white cake mix. On October 25, CDC requested officials in Maryland, Ohio, and Wisconsin to interview patients using a questionnaire with specific questions about baking exposures.
On October 31, the Food and Drug Administration (FDA) initiated an investigation of manufacturer A with regard to the Salmonella-positive white cake mix. In addition to the investigation and document collection, FDA collected samples including an ingredient (flour), finished cake mix, and environmental samples. All collected samples tested negative for Salmonella. On November 5, a voluntary recall of manufacturer A’s classic white, classic butter golden, signature confetti, and classic yellow cake mixes was announced because they might be contaminated with Salmonella bacteria.
On January 14, 2019, CDC declared this outbreak, which totaled seven cases in five states,† to be over (1). This is the first time that OHA used WGS data on the publicly available NCBI website to detect a multistate outbreak associated with a widely distributed consumer product, which resulted in product action. WGS of food and environmental isolates and subsequent analysis on the NCBI and SEDRIC platforms are emerging as useful tools in identifying outbreaks associated with widely distributed products with long shelf lives and low background rates of consumption, such as raw cake mix. Detection of these outbreaks is typically difficult and relies mainly upon epidemiologic evidence from investigation of a larger number of cases (2–4). These efforts also highlight the value of collaboration between public health epidemiologists and laboratorians as well as the use of new technological tools for outbreak detection. During outbreak or cluster investigations, food and environmental samples should be collected as quickly as possible whenever practical, particularly when epidemiologic data suggest an association. WGS, in conjunction with the NCBI website and SEDRIC, can be used to identify genetically related isolates quickly.
US: Notes from the field: Multistate outbreak of salmonella Agbeni associated with consumption of raw cake mix – five states, 2018