SYNOPSIS

Escherichia coli O157 Outbreaks

in the United States, 20032012
Katherine E. Heiman, Rajal K. Mody, Shacara D. Johnson, Patricia M. Griffin, L. Hannah Gould

Medscape, LLC is pleased to provide online continuing medical education (CME) for this journal article, allowing
clinicians the opportunity to earn CME credit.
This activity has been planned and implemented in accordance with the Essential Areas and policies of the
Accreditation Council for Continuing Medical Education through the joint providership of Medscape, LLC and
Emerging Infectious Diseases. Medscape, LLC is accredited by the ACCME to provide continuing medical education
for physicians.
Medscape, LLC designates this Journal-based CME activity for a maximum of 1.0 AMA PRA Category 1
Credit(s)TM. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
All other clinicians completing this activity will be issued a certificate of participation. To participate in this journal
CME activity: (1) review the learning objectives and author disclosures; (2) study the education content; (3) take the
post-test with a 75% minimum passing score and complete the evaluation at http://www.medscape.org/journal/eid; (4)
view/print certificate.
Release date: July 15, 2015; Expiration date: July 15, 2016
Learning Objectives
Upon completion of this activity, participants will be able to:
Identify the most common source of outbreaks of Escherichia coli O157 in the United States
Identify the most common specific food source associated with foodborne outbreaks of E. coli O157 in
the United States
Evaluate epidemiologic variables associated with the severity of E. coli O157 outbreaks
Assess other epidemiologic data from outbreaks of E. coli O157 in the United States
CME Editor
Jean Michaels Jones, BSN, Technical Writer/Editor, Emerging Infectious Diseases. Disclosure: Jean Michaels
Jones, BSN, has disclosed no relevant financial relationships.
CME Author
Charles P. Vega, MD, Clinical Professor of Family Medicine, University of California, Irvine. Disclosure:
Charles P. Vega, MD, has disclosed the following financial relationships: served as an advisor or consultant for
McNeil Pharmaceuticals.
Authors
Disclosures: Katherine E. Heiman, MPH; Rajal K. Mody, MD, MPH; Shacara D. Johnson, MSPH; Patricia M.
Griffin, MD; and L. Hannah Gould, PhD, MS, MBA, have disclosed no relevant financial relationships.

Infections with the Shiga toxinproducing bacterium Esch- 1,272 hospitalizations, and 33 deaths. Transmission was
erichia coli O157 can cause severe illness and death. We through food (255 outbreaks, 65%), person-to-person con-
summarized reported outbreaks of E. coli O157 infections tact (39, 10%), indirect or direct contact with animals (39,
in the United States during 20032012, including demo- 10%), and water (15, 4%); 42 (11%) had a different or un-
graphic characteristics of patients and epidemiologic known mode of transmission. Beef and leafy vegetables,
findings by transmission mode and food category. We combined, were the source of >25% of all reported E. coli
identified 390 outbreaks, which included 4,928 illnesses, outbreaks and of >40% of related illnesses. Outbreaks at-
tributed to foods generally consumed raw caused higher
Author affiliation: Centers for Disease Control and Prevention, hospitalization rates than those attributed to foods gener-
Atlanta, Georgia, USA ally consumed cooked (35% vs. 28%). Most (87%) water-
borne E. coli outbreaks occurred in states bordering the
DOI: http://dx.doi.org/10.3201/eid2108.141364
Mississippi River.

Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015 1293
SYNOPSIS

S igns and symptoms of infection with Shiga toxin

producing Escherichia coli O157 can include diarrhea
that is often bloody, severe stomach cramps, and vomit-
or O157:NM infection resulting from a common expo-
sure. Epidemiologically linked, clinically compatible cases
without laboratory confirmation were included in the case
ing; infection can progress to hemolytic uremic syndrome count. States within the contiguous United States were cat-
(HUS) and death (1). In the United States, these infec- egorized as north if they were located entirely north of
tions and related illnesses are estimated to cost >$405 parallel 37 north latitude and south if entirely below this
million annually (2). latitude; states crossing this line (California and Nevada)
E. coli O157 can be transmitted to humans through were excluded from some analyses.
contaminated food and water, directly between persons, Implicated foods were classified into categories by
and through contact with animals or their environment. using an existing classification scheme (5) with 2 modi-
The most common reservoir is cattle, and ground beef is fications: fruits and nuts were classified separately and
the most frequently identified vehicle of transmission to meat acquired by hunting was classified with other meat.
humans. E. coli O157 was first recognized as a foodborne Implicated foods were further classified by whether they
pathogen after outbreaks during 1982 were linked to ground were usually consumed cooked (beef, other meats, poul-
beef consumption (1). Since then, many other sources have try) or raw (leafy vegetables, fruits, unpasteurized dairy,
been identified (3), mostly through outbreak investigations. sprouts). Steak was considered tenderized if described as
We describe the epidemiology of E. coli O157 outbreaks tenderized or injected. Settings were classified as home;
during 20032012. restaurant (e.g., restaurant, delicatessen, or banquet facil-
ity); institution (e.g., childcare center, school, or prison);
Methods fair; petting zoo; recreational area (e.g., park or beach);
and other (e.g., grocery store, commercial production fa-
Surveillance cility, or farm animal show).
We identified E. coli O157 outbreaks in the United States
using these sources: the Foodborne Disease Outbreak Analysis
Surveillance System (FDOSS) for outbreaks during We summarized demographic and epidemiologic find-
20032012; the Waterborne Disease Outbreak Surveil- ings by transmission mode and food categories for which
lance System (WBDOSS) for outbreaks during 20032010 >3 outbreaks were reported. Two periods, 20032007 and
(data were not available after 2010); the National Outbreak 20082012, were evaluated to assess changes over time.
Reporting System (NORS) for outbreaks facilitated by Severity was assessed by calculating hospitalization (no.
transmission between persons, animal contact, environ- hospitalized/total no. ill) and HUS rates (no. physician
mental contamination, or with an unknown transmission diagnosed HUS cases/total no. ill). Culture-confirmation
mode during 20092012 (4); reports to the Centers for Dis- rates (no. culture-confirmed/total no. ill) were also calcu-
ease Control and Prevention of person-to-person and ani- lated. Analyses of settings were limited to outbreaks that
mal contactassociated outbreaks during 20032008; and occurred in a single setting type. For transmission modes
PubMed using these search terms: Escherichia coli O157 other than food, data for age group, sex, and culture confir-
outbreak, STEC O157 outbreak, Escherichia coli O157, mation were not available before 2009. We compared the
STEC O157, and O157. Data reported for each outbreak, median outbreak size and state outbreak rates (outbreaks
with varying degrees of completeness, included primary with exposure occurring in a single state/sum of state popu-
transmission mode; states where exposure occurred; num- lations during 20032012, from U.S. Census Bureau inter-
ber of illnesses (culture-confirmed and epidemiologically censal estimates) using the Kruskal-Wallis test, and per-
linked); hospitalizations; HUS cases (defined by physi- formed categorical analyses using 2 and Fisher exact tests;
cian diagnosis; no laboratory data were collected); deaths, 2-tailed p values <0.05 were considered significant.
age, and sex distribution of patients; month the outbreak
began; and setting where the outbreak occurred or, for Results
foodborne disease outbreaks, the place of food preparation. During 20032012, a total of 390 E. coli O157 outbreaks
For foodborne disease outbreaks, the implicated food (if (353 O157:H7, 15 O157:NM, 1 O157:H7 and O157:NM,
identified) was reported. For waterborne disease outbreaks, and 22 O157 with an unknown H antigen) were reported.
water source type (e.g., recreational, drinking, waste water) These outbreaks resulted in 4,928 illnesses, 1,272 (26% of
was reported. illnesses) hospitalizations, 299 (6%) physician-diagnosed
HUS cases, and 33 (0.7%) deaths (Table 1). The median out-
Definitions break size was 6 illnesses (range 2238). Primary transmis-
An E. coli O157 outbreak was defined as an event during sion modes were foodborne (255 outbreaks, 65%), animal
which 2 persons had culture-confirmed E. coli O157:H7 contact (39, 10%), person-to-person (39, 10%), waterborne

1294 Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015
 E. coli O157, United States, 20032012

Table 1. Characteristics of Escherichia coli O157 outbreaks by transmission mode and food category, United States, 20032012*
Outbreaks (% of Illnesses (median Hospitalizations Physician-diagnosed HUS Deaths (% of
Transmission source all outbreaks) outbreak size) (% of all illnesses) (% of all illnesses) all illnesses)
Food 255 (65) 3,667 (6) 1,035 (29) 209 (6) 25 (0.7)
Beef 78 (20) 1,144 (7) 316 (28) 67 (6) 5 (0.4)
Poultry 1 (0) 60 5 (8) 0 0
Other meat 7 (2) 39 (4) 12 (31) 4 (10) 0
Dairy 16 (4) 140 (6) 52 (37) 22 (16) 0
Leafy vegetables 29 (7) 922 (16) 321 (35) 53 (6) 7 (0.8)
Fruits 6 (2) 57 (8) 20 (35) 5 (9) 6 (10.5)
Sprouts 3 (1) 35 (13) 4 (11) 0 0
Nuts 1 (0) 8 3 (38) 0 0
Other foods 29 (7) 580 (11) 123 (21) 24 (4) 0
Food unknown 85 (22) 682 (5) 179 (26) 32 (5) 10 (1.5)
Animal contact 39 (10) 552 (6) 127 (23) 51 (9) 2 (0.4)
Person-to-person 39 (10) 322 (5) 45 (14) 24 (7) 2 (0.6)
Water 15 (4) 154 (6) NA NA 1 (0.6)
Other or unknown 42 (10) 233 (4) 65 (28) 15 (6) 3 (1.3)
Total 390 (100) 4,928 (6) 1,272 (27) 299 (6) 33 (0.7)
*HUS, hemolytic uremic syndrome; NA, not available.
Poultry was thought to be cross-contaminated by ground beef (http://www.ct.gov/dph/lib/dph/infectious_diseases/ctepinews/vol29no5.pdf).
Outbreaks transmitted from other foods were 2 each of: guacamole, pico de gallo, salsa, and potato salad; and 1 each of: cookie dough; alfalfa sprouts
and iceberg lettuce; baked beans and unknown fruit; meatballs, steak and green salad; cantaloupe and hamburger; lamb and beef; lettuce and green
grapes; lime and bean dip; macaroni; Mexican wheat snack; sandwich; seafood; vegetable-based salad; pepperoni; jerky; multiple foods. Agencies
sometimes report >1 food vehicle when epidemiologic evidence cannot distinguish between them even when it is likely that only one was the source.
Excludes 154 illnesses from the denominator that were associated with waterborne disease outbreaks because no data was collected for hospitalization
and physician-diagnosed HUS cases for this transmission mode.

(15, 4%), and other or unknown (42, 11%). Foodborne dis- Vehicles
ease outbreaks caused the most illnesses (3,667, 74%), hos- Of the 255 foodborne disease outbreak reports, 170 (67%)
pitalizations (1,035, 81%), physician-diagnosed HUS cas- implicated a specific food, of which 141 (83%) could be
es (209, 70%), and deaths (25, 70%). During 20092012, classified into a single category. The implicated categories
when data for culture-confirmed infections were collected were beef (78 outbreaks, 55%), leafy vegetables (29, 21%),
for all outbreaks, the proportions of transmission were un- dairy (16, 11%), fruits (6, 4%), other meats (7, 5%), sprouts
known modes (76%), foodborne (69%), person-to-person (3, 2%), nuts (1, 1%), and poultry (1, 1%) (Table 1). The
(62%), and animal contact (40%). types of beef implicated were ground beef (54 outbreaks,
The median annual number of outbreaks reported 69% of beef-associated outbreaks) and steak (10, 14%).
during 20082012 was lower than during 20032007 (45 The types of steak, reported in 5 outbreaks, were sirloin
vs. 33, p = 0.12) (Figure 1). The median annual number (4 outbreaks) and filet mignon (1). Steaks were mechani-
of foodborne disease outbreaks did not change, although cally tenderized in 5 outbreaks (2 sirloin, 1 filet mignon,
the number attributed to dairy (11 vs. 5) and fruits (5 vs. 2 unknown). Other implicated meats were venison (3 out-
2) more than doubled. The median annual number of breaks), lamb (2), and bison/buffalo (2).
outbreaks attributed to other modes of transmission did In all 16 outbreaks attributed to dairy, unpasteurized
not change. products were implicated; 13 (81%) to unpasteurized milk

Figure 1. Number of
Escherichia coli O157
outbreaks by transmission
mode and year (n = 390),
United States, 20032012.

Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015 1295
SYNOPSIS

and 3 (9%) to cheese made from unpasteurized milk. foods generally consumed raw than for cooked foods (35%
Most outbreaks attributed to leafy vegetables implicated vs. 28%, p = 0.0001). Deaths occurred most commonly in
lettuce (22 outbreaks, 76%), including romaine (3), ice- outbreaks attributed to leafy vegetables (7 deaths, 0.8% of
berg (1), and mesclun mix (1); others in this category illnesses in leafy vegetable outbreaks), fruits (6, 11%), and
were spinach (4, 13%) and salads with unspecified types beef (5, 0.4%).
of greens (3,10%). Outbreaks attributed to fruit impli-
cated unpasteurized apple cider (4 outbreaks), fruit salad Demographic Characteristics
(1), and strawberries (1), a newly identified food vehicle Age group distribution of cases varied by transmission
for this pathogen. Other newly identified food vehicles mode and food category (Figure 2). Patient age group
were hazelnuts sold in the shell (1 outbreak), packaged was available for 2,884 (81%) illnesses in foodborne out-
cookie dough (1), salsa (2), and pico de gallo (2). A to- breaks, 63 (90%) illnesses in waterborne outbreaks, and,
tal of 39 outbreaks were transmitted by contact with ani- during 20092012, 121 (96%) illnesses in person-to-person
mals; >1 animal type was reported for 24 (62%), includ- outbreaks and 189 (73%) illnesses in animal contact out-
ing cattle (15), goats (12), sheep (8), pigs (3), rabbits (3), breaks. Most of the illnesses in person-to-person outbreaks
chickens (2), and 1 each of elk, llama, alpaca, hedgehog, (73, 60%) were among children <5 years old. The highest
and dog. Every outbreak that identified a reported animal proportions of illnesses in waterborne and animal contact
type included >1 ruminant animal. Although no specific outbreaks (67% and 43%, respectively) were among per-
animal types were listed for the remaining 14 outbreaks, sons 519 years of age. Foodborne disease occurred among
livestock were reported for 2 and petting zoos or county persons in all age groups; only 8% of foodborne illnesses
fairs for 11. occurred among children <5 years of age, but >25% of
Of the 15 waterborne disease outbreaks, 10 were at- dairy-associated illnesses were among children of this age
tributed to recreational water (3 treated, 7 untreated), 3 to group. The largest proportions of illnesses in outbreaks at-
drinking water, and 1 possibly to wastewater. The water tributed to beef and dairy were among persons 519 years
source was unknown for the remaining outbreak. old (38% and 47%, respectively), whereas the largest pro-
portion reported in outbreaks attributed to sprouts and leafy
Severity vegetables was among persons 2049 years old (71% and
The severity of outbreaks differed by transmission mode 45%, respectively).
and among foodborne disease outbreaks by food category Overall, 55% of patients were women; the percent-
(Table 1). Hospitalization rates were lower for person-to- age did not vary significantly by transmission mode. Pa-
person outbreaks compared with all other transmission tient sex was available for 3,385 (92%) illnesses in food-
modes (14% vs. 28%, p<0.0001). Physician-diagnosed borne disease outbreaks, 63 (90%) illnesses in waterborne
HUS rates were higher for animal contact outbreaks than outbreaks, and, from 20092012, 126 (97%) illnesses
for all other modes (9% vs. 6%, p = 0.0005). The 33 deaths in person-to-person outbreaks and 193 (80%) illnesses
occurred in foodborne (25 deaths), person-to-person (2), in animal contact outbreaks. In foodborne disease out-
animal contact (2), and waterborne (1) disease outbreaks; breaks, the proportion of female patients was highest in
transmission mode was unknown for 3 deaths. outbreaks attributed to fruits (67%) and leafy vegetables
Among foodborne disease outbreaks, hospitalization (65%) and lowest in outbreaks attributed to meats other
rates were lower for outbreaks attributed to beef than for than beef (31%) (p<0.0001); cases were evenly distrib-
all other food vehicles (28% vs. 35%, p<0.001). Hospi- uted among women and men in outbreaks attributed to
talization rates were higher among outbreaks attributed to beef and dairy.

Figure 2. Percentage of Escherichia coli O157 outbreak illnesses by age group and A) transmission mode (n = 3,417) and B) selected
food categories (n = 1,574), United States, 20032012.

1296 Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015
 E. coli O157, United States, 20032012

Seasonality Most (17, 89%) outbreaks with person-to-person

E. coli O157 outbreaks occurred year-round; however, near- transmission and a reported setting occurred in institutions;
ly half occurred during JulySeptember (175, 46%) (Figure all except 1 institutional outbreak occurred in a childcare
3). The seasonality of foodborne outbreaks varied by food center. Petting zoos and fairs were the most common set-
category. Beef-associated outbreaks occurred most often in ting for outbreaks attributed to animal contact (17, 71%).
July (21 outbreaks, 27%), and leafy vegetable-associated out- Waterborne outbreaks occurred most often in recreational
breaks in September (8, 28%). Waterborne disease outbreaks settings (7, 47%), including lakes (4), a pool (1), a sports
occurred most often during June (6, 40%) and outbreaks as- complex (1), and an interactive fountain (1).
sociated with animal contact during August (13, 37%).
Discussion
Geographic Distribution of Outbreaks and As in previous reports, our analysis found that food was the
Multistate Outbreaks major mode of transmission for E. coli O157 outbreaks (3).
Outbreaks were reported by 43 states (Figure 4). The me- Although beef, especially ground beef, remained the most
dian outbreak rate in northern states was >2 times the rate common source (3), 14% of beef outbreaks were attributed
in southern states (0.015 vs. 0.006/100,000 population, p to steak. Outbreaks attributed to foods generally consumed
= 0.005), which was also true for foodborne disease out- raw had higher hospitalization rates than other outbreaks,
breaks (0.007 vs. 0.002, p = 0.02). Most waterborne disease and leafy vegetables were the major source of outbreaks in
outbreaks were reported in states bordering the Mississippi the fall. Foodborne disease outbreak rates were highest in
River (13, 87%), including all 3 drinking water outbreaks northern states, and most waterborne disease outbreaks oc-
and all 7 recreational water outbreaks associated with lakes curred in states bordering the Mississippi River. New food
or other natural, untreated bodies of water. Waterborne vehicles were identified, including raw, prepackaged cook-
disease outbreaks reported by California (1) and Idaho (1) ie dough (6), hazelnuts (7), strawberries (8), and salsa (9).
were associated with treated recreational water. More outbreaks were reported during 20032012 than
Of the 390 outbreaks reported, 44 (17%), all foodborne, during the previous 20 years (3), although outbreaks were
were multistate. Multistate outbreaks were larger than sin- smaller (median 6 illnesses vs. 8). The hospitalization rate
gle state outbreaks (median 22 cases vs. 5, p<0.0001). The was higher than during the previous 20 years (27% vs.
food categories implicated in multistate outbreaks were 17%), and the HUS rate was slightly higher (6% vs. 4%).
beef (22 outbreaks, 17 attributed to ground beef), leafy veg- Contaminated food was responsible for a larger percent of
etables (12), unpasteurized dairy products (3), sprouts (1), outbreaks during our study (65% vs. 52%), and the percent
nuts (1), and other meats (bison, 1). attributed to another or unknown mode decreased (10% vs.
21%). Among foodborne disease outbreaks, the percentage
Settings caused by E. coli O157 in beef decreased (31% vs. 47%),
Among 222 foodborne outbreaks for which food prepara- and among outbreaks associated with beef, the percentage
tion settings were reported, a single setting was reported linked to ground beef decreased (21% vs. 41%), whereas
for 183 (84%). Restaurants were the most common settings the percentage associated with other types of beef increased
(73 outbreaks, 40%), followed by private homes (60, 33%) (9% vs. 6%). These differences might be partly attributable
(Table 2). Beef was most frequently prepared in a home to continued improvements in surveillance, including the
(34, 56%), followed by restaurants (15, 25%). Leafy veg- maturation of the national molecular subtyping network,
etables were most often prepared in restaurants (13, 59%), PulseNet, which supported earlier detection of more out-
followed by homes (4, 18%). breaks (10), as well as improved outbreak investigations

Figure 3. Number of Escherichia coli O157 outbreaks by month and by A) transmission mode (n = 390) and B) selected food categories
(n = 255), United States, 20032012.

Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015 1297
SYNOPSIS

Figure 4. Single-state Escherichia coli O157 outbreaks (n = 346) by state and transmission mode, United States, 20032012.
A) Foodborne transmission (n = 211); B) animal contact transmission (n = 39); C) person-to-person transmission (n = 39);
D) waterborne transmission (n = 15). Curved line denotes 37N latitude.

and systematic, electronic reporting of outbreaks of all virulence factors. Outbreaks attributed to foods generally
transmission modes (4). consumed raw had higher hospitalization rates than those
Beef, particularly ground beef, continues to be the ma- attributed to cooked foods. Outbreaks with high propor-
jor source of E. coli O157 outbreaks, likely because cattle tions of illnesses among young children, especially those
are the main reservoir for E. coli O157. Contamination of attributed to animal contact, dairy products, other meats,
raw beef usually occurs during slaughter and meat process- and fruits, also had the highest rates of physician-diag-
ing by contact with hides contaminated with feces (11,12). nosed HUS, which might relate to the higher risk for HUS
Grinding can spread contamination through vast amounts among children (18). However, this finding was not true for
of ground beef (11). Mechanically tenderizing steaks inter- person-to-person outbreaks, in which the proportion of ill-
nalizes surface contamination (13,14), so steaks not cooked nesses among young children was highest. The reasons for
thoroughly could harbor E. coli O157. The risk for con- this finding are unknown, but intensive case finding efforts
tamination varies for different locations on the carcass (15). during suspected person-to-person outbreak investigations
Cattle carcasses sampled for microbial contamination had might identify more mild illnesses and asymptomatic pa-
higher bacteria counts on sites on the rear (16,17). tients, decreasing the overall hospitalization rate.
Our finding that outbreaks attributed to leafy veg- Higher hospitalization rates in outbreaks attributed to
etables, dairy products, fruits, and other meats were more certain foods might be influenced by patient sex. For ex-
severe than outbreaks attributed to beef could have several ample, hospitalization rates were 35% for outbreaks asso-
explanations, including strain virulence and patient age and ciated with leafy vegetables, and 66% of illnesses were in
sex. Outbreak reports did not provide information on strain female patients. Studies have found that women were at

1298 Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015
 E. coli O157, United States, 20032012

Table 2. Escherichia coli O157 outbreaks by setting (n = 357), United States, 20032012
Setting/no. outbreaks
Restaurant/ Fair/petting Recreational
Transmission source banquet facility Home Institutional* zoo area/facility Other Unknown Total
Food 73 60 16 2 0 32 39 222
Beef 15 34 3 0 0 9 8 69
Poultry 0 0 1 0 0 0 0 1
Other meat 1 5 0 0 0 0 0 6
Dairy 0 2 0 0 0 11 0 13
Leafy vegetables 13 4 3 0 0 2 4 26
Fruits 0 2 0 1 0 2 0 5
Sprouts 1 0 0 0 0 0 0 1
Nuts 0 1 0 0 0 0 0 1
Other foods 13 6 0 0 0 2 3 24
Food unknown 30 6 9 1 0 6 24 76
Animal contact 0 1 2 17 0 4 15 39
Person-to-person 0 2 17 0 0 0 20 39
Water 0 2 2 0 7 4 0 15
Other or unknown 0 0 0 0 0 0 42 42
All outbreaks 73 65 37 19 7 38 116 357
*Institutional setting comprises camp, daycare, hospital, nursing home, prison, school, church, and workplace cafeteria.
Recreational area/facility comprises county and state park, beach, and a sports complex.
Other settings comprise a grocery store, commercial products, a 4H steer competition, communities, a farm, a National Western Stock Show, a public
outdoor area, and a club.
Setting was not available for 3 outbreaks involving leafy vegetables.

increased risk for HUS after E. coli O157 infection (18 feces and longer day length than with higher ambient tem-
20), although others have not found an association (21,22). perature, although both were statistically significant (30).
Additionally, the proportion of illnesses in female patients However, several studies of prevalence of E. coli O157 in
might relate to gender-specific food preferences. cattle feedlots have found either no variation by geography
Most outbreaks occurred during the summer. Cattle (31) or higher prevalence in herds in the south (32), where
shed the largest number of E. coli O157 organisms in their summertime daylight hours are shorter but ambient tem-
feces during summer months (23,24), coinciding with a peratures are higher. Geographic distribution might also
higher prevalence of E. coli O157 on hides in processing be influenced by presence of cattle called super shedders
plants (25). However, leafy vegetable-associated outbreaks that shed higher numbers of E. coli O157 bacteria (33). Dif-
exhibited distinct fall seasonality, which could be the re- ferences in reporting by state could also affect geographic
sult of summertime application to seedlings of irrigation patterns (4).
water, soil amendments, or fertilizers that might contain Several factors might have contributed to our finding
more E. coli O157 organisms than other seasons. There- that waterborne outbreaks were reported primarily among
fore, leafy vegetables harvested during fall might be more states bordering the Mississippi River. Cattle density is
likely to become contaminated than those grown at other high in the Midwest (25). E. coli O157-containing feces
times. Seasonality of leafy vegetableassociated outbreaks from cattle and other animals can be washed into nearby
could also relate to harvest location. Nearly three quar- bodies of water, especially during heavy rainfall (34). Karst
ters of leafy vegetable-associated outbreaks occurred dur- formations, which are rock formations that occur in several
ing AprilOctober; during this same time each year, most Midwest states, develop when acidic water begins to break
US-produced lettuce is harvested from the Salinas Valley down bedrock surfaces, allowing surface water to enter
(California) (26). For 4 of 5 recent outbreaks (20112012), fractures in limestone and contaminate ground water; stud-
detailed traceback information implicated lettuce harvested ies have shown that E. coli can survive in karst streams for
from farms in Salinas Valley (K.E. Heiman, unpub. data). prolonged periods (35). Karst formations exist in the coun-
The cause of higher infection rates in northern states ties that reported 2 of the 3 E. coli O157 outbreaks in which
than southern states is unknown (27). Cattle density might drinking water was the source; all were associated with
play a role. In other countries, areas with the highest cat- well water. The remaining well waterassociated outbreak
tle density had some of the highest rates of human E. coli occurred on a farm after heavy rain and flooding. However,
O157 infection (28). Cattle density does not appear to be karst formations are unlikely to explain outbreaks associ-
higher in northern states compared with southern states ated with untreated recreational water. Only 15 waterborne
(25), although it is very high in counties in California outbreaks were reported; therefore, interpretation of these
where most lettuce is produced (29). Other environmental findings is limited.
factors might have a role. Edrington et al. found a stronger Our study had several limitations. Reported out-
correlation between prevalence of E. coli O157 in cattle breaks likely underrepresent the number that occur (36).

Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015 1299
SYNOPSIS

Many go unrecognized or unreported because they are not Acknowledgments

detected, no common source is found, or resources are not We thank Kristin Holt for input into the analysis and state and lo-
available for investigation (37). Sources of outbreak-re- cal public health officials for submitting outbreak reports to CDC.
lated illnesses may differ from those of sporadic illnesses,
Ms. Heiman is an epidemiologist in the Outbreak Response
although our findings were similar to those of another US
and Prevention Branch at the Centers for Disease Control and
study that examined risk factors for sporadic E. coli O157
Prevention. Her primary research interests are in enteric disease
infections (38). We analyzed outbreaks by their primary
surveillance, and outbreak investigation.
mode of transmission; however, outbreak-related infec-
tions could be transmitted by a variety of modes (e.g., References
several illnesses were transmitted from person-to-person 1 Riley LW, Remis RS, Helgerson SD, McGee HB, Wells JG,
Davis BR, et al. Hemorrhagic colitis associated with a rare
in an outbreak linked to spinach). State and local health Escherichia coli serotype. N Engl J Med. 1983;308:6815.
departments investigated and reported outbreaks; we did http://dx.doi.org/10.1056/NEJM198303243081203
not verify transmission mode. Our findings might be in- 2. Frenzen PD, Drake A, Angulo FJ. Economic cost of illness due to
fluenced by changes in surveillance: outbreaks transmit- Escherichia coli O157 infections in the United States. J Food Prot.
2005;68:262330.
ted person-to-person, by animal contact, and by an un- 3. Rangel JM, Sparling PH, Crowe C, Griffin PM, Swerdlow DL.
known mode were not systematically reported until 2009 Epidemiology of Escherichia coli O157:H7 outbreaks,
(4). We used physician-diagnosed HUS, which might United States, 19822002. Emerg Infect Dis. 2005;11:6039.
overestimate the rate of HUS (39); however, the rate of http://dx.doi.org/10.3201/eid1104.040739
4. Hall AJ, Wikswo ME, Manikonda K, Roberts VA, Yoder JS,
physician-diagnosed HUS for all outbreak-related ill- Gould LH. Acute gastroenteritis surveillance through the National
nesses was 6%, the same as was reported in a study that Outbreak Reporting System, United States. Emerg Infect Dis.
used defined laboratory criteria (2). We found one third of 2013;19:13059. http://dx.doi.org/10.3201/eid1908.130482
foodborne disease outbreak reports did not have an iden- 5. Gould LH, Walsh KA, Vieira AR, Herman K, Williams IT, Hall AJ,
et al. Surveillance for foodborne disease outbreaksUnited States,
tified implicated food item; therefore, the foods in this 19982008. MMWR Surveill Summ. 2013;62:134.
analysis might not be representative of all foodborne E. 6. Neil KP, Biggerstaff G, MacDonald JK, Trees E, Medus C,
coli O157 outbreaks. Musser KA, et al. A novel vehicle for transmission of Escherichia
coli O157:H7 to humans: multistate outbreak of E. coli O157:H7
infections associated with consumption of ready-to-bake
Conclusions commercial prepackaged cookie doughUnited States, 2009.
More E. coli O157 outbreaks were reported during 2003 Clin Infect Dis. 2012;54:5118. http://dx.doi.org/10.1093/
2012 than during the previous 20 years (3), probably be- cid/cir831
cause of continued improvements in surveillance. Out- 7. Miller BD, Rigdon CE, Ball J, Rounds JM, Klos RF, Brennan BM,
et al. Use of traceback methods to confirm the source of a
breaks in the United States continue to be caused mostly multistate Escherichia coli O157:H7 outbreak due to in-shell
by contaminated food, especially beef. People can protect hazelnuts. J Food Prot. 2012;75:3207. http://dx.doi.org/
themselves through proper handling of raw beef, by cook- 10.4315/0362-028X.JFP-11-309
ing nonintact beef products such as ground beef and me- 8. Laidler MR, Tourdjman M, Buser GL, Hostetler T, Repp KK,
Leman R, et al. Escherichia coli O157: H7 infections associated
chanically tenderized steak to internal temperatures of at with consumption of locally grown strawberries contaminated by
least 160F, and by thorough handwashing after contact deer. Clin Infect Dis. 2013;57:112934. http://dx.doi.org/10.1093/
with animals or their environment. Clearly labeling me- cid/cit468
chanically tenderized beef might help consumers make 9. Kendall ME, Mody RK, Mahon BE, Doyle MP, Herman KM,
Tauxe RV. Emergence of salsa and guacamole as frequent vehicles
safer choices. Pasteurizing dairy products and fruit juices of foodborne disease outbreaks in the United States, 19732008.
is recommended. Further research is needed to identify Foodborne Pathog Dis. 2013;10:31622. http://dx.doi.org/10.1089/
interventions to reduce the risk of illness caused by con- fpd.2012.1328
sumption of raw produce. Ways must be found to encour- 10. Swaminathan B, Barrett TJ, Hunter SB, Tauxe RV. CDC PulseNet
Task Force. PulseNet: the molecular subtyping network for
age implementation of interventions such as decreasing foodborne bacterial disease surveillance, United States.
contamination in the production of produce, vaccinating Emerg Infect Dis. 2001;7:3829. http://dx.doi.org/10.3201/
cattle, and irradiating foods (40). Consideration of the eid0703.017303
proximity of agriculture fields to cattle operations, as 11. Erickson MC, Doyle MP. Food as a vehicle for transmission of Shiga
toxin-producing Escherichia coli. J Food Prot. 2007;70:242649.
well as adherence to Good Agricultural Practices speci- 12. Hussein HS. Prevalence and pathogenicity of Shiga toxin
fied by the US Department of Agriculture (http://www. producing Escherichia coli in beef cattle and their products.
ams.usda.gov/AMSv1.0/gapghp) during production, will J Anim Sci. 2007;85(Suppl):E6372. http://dx.doi.org/10.2527/
also help reduce risk. Continued surveillance is needed to jas.2006-421
13. Luchansky JB, Phebus RK, Thippareddi H, Call JE. Translocation
monitor changes in food vehicles of contamination and to of surface-inoculated Escherichia coli O157:H7 into beef
improve our understanding of the geographic distribution subprimals following blade tenderization. J Food Prot. 2008;
of outbreaks. 71:21907.

1300 Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015
 E. coli O157, United States, 20032012

14. Laine ES, Scheftel JM, Boxrud DJ, Vought KJ, Danila RN, increased risk for most but not all serogroups. Vector Borne
Elfering KM, et al. Outbreak of Escherichia coli O157:H7 Zoonotic Dis. 2008;8:63544. http://dx.doi.org/10.1089/
infections associated with nonintact blade-tenderized frozen steaks vbz.2007.0237
sold by door-to-door vendors. J Food Prot. 2005;68:1198202. 29. United States Department of Agriculture. 2007 Census of
15. Stopforth JD, Lopes M, Shultz JE, Miksch RR, Samadpour M. Agriculture, United States, online census highlights, Ag Atlas
Microbiological status of fresh beef cuts. J Food Prot. 2006; Maps: USDA. 2009 [cited 2015 May 6]. http://agcensus.usda.gov/
69:14569. Publications/2007/Online_Highlights/Ag_Atlas_Maps/
16. Elder RO, Keen JE, Siragusa GR, Barkocy-Gallagher GA, 30. Edrington TS, Callaway TR, Ives SE, Engler MJ, Looper ML,
Koohmaraie M, Laegreid WW. Correlation of enterohemorrhagic Anderson RC, et al. Seasonal shedding of Escherichia coli O157:
Escherichia coli O157 prevalence in feces, hides, and carcasses H7 in ruminants: a new hypothesis. Foodborne Pathog Dis.
of beef cattle during processing. Proc Natl Acad Sci U S A. 2006;3:41321. http://dx.doi.org/10.1089/fpd.2006.3.413
2000;97:29993003. http://dx.doi.org/10.1073/pnas.97.7.2999 31. Hancock DD, Rice DH, Thomas LA, Dargatz DA, Besser TE.
17. Bell RG. Distribution and sources of microbial contamination Epidemiology of Escherichia coli O157 in feedlot cattle.
on beef carcasses. J Appl Microbiol. 1997;82:292300. J Food Prot. 1997;60:4625. http://www.ingentaconnect.com/
http://dx.doi.org/10.1046/j.1365-2672.1997.00356.x content/iafp/jfp/1997/00000060/00000005/art00001
18. Gould LH, Demma L, Jones TF, Hurd S, Vugia DJ, Smith K, et al. 32. Garber L, Wells S, Schroeder-Tucker L, Ferris K. Factors
Hemolytic uremic syndrome and death in persons with Escherichia associated with fecal shedding of verotoxin-producing Escherichia
coli O157: H7 infection, foodborne diseases active surveillance coli O157 on dairy farms. J Food Prot. 1999;62:30712.
network sites, 20002006. Clin Infect Dis. 2009;49:14805. 33. Chase-Topping M, Gally D, Low C, Matthews L, Woolhouse M.
http://dx.doi.org/10.1086/644621 Super-shedding and the link between human infection and livestock
19. Cimolai N, Carter JE, Morrison BJ, Anderson JD. Risk factors for carriage of Escherichia coli O157. Nat Rev Microbiol. 2008;6:904
the progression of Escherichia coli O157:H7 enteritis to 12. http://dx.doi.org/10.1038/nrmicro2029
hemolytic-uremic syndrome. J Pediatr. 1990;116:58992. 34. Cooley M, Carychao D, Crawford-Miksza L, Jay MT, Myers C,
http://dx.doi.org/10.1016/S0022-3476(05)81609-9 Rose C, et al. Incidence and tracking of Escherichia coli O157:H7
20. Chang HG, Tserenpuntsag B, Kacica M, Smith PF, Morse DL. in a major produce production region in California. PLoS ONE.
Hemolytic uremic syndrome incidence in New York. 2007;2:e1159. http://dx.doi.org/10.1371/journal.pone.0001159
Emerg Infect Dis. 2004;10:92831. http://dx.doi.org/10.3201/ 35. Davis RK, Hamilton S, Brahana JV. Escherichia coli survival in
eid1005.030456 mantled Karst springs and streams, Northwest Arkansas Ozarks,
21. Tserenpuntsag B, Chang HG, Smith PF, Morse DL. Hemolytic USA. J Am Water Resour Assoc. 2005;41:127987.
uremic syndrome risk and Escherichia coli O157:H7. Emerg Infect http://dx.doi.org/10.1111/j.1752-1688.2005.tb03800.x
Dis. 2005;11:19557. http://dx.doi.org/10.3201/eid1112.050607 36. Scallan E, Jones TF, Cronquist A, Thomas S, Frenzen P,
22. Rowe PC, Orrbine E, Lior H, Wells GA, Yetisir E, Clulow M, Hoefer D, et al. Factors associated with seeking medical care and
et al. Risk of hemolytic uremic syndrome after sporadic submitting a stool sample in estimating the burden of foodborne
Escherichia coli O157:H7 infection: results of a Canadian illness. Foodborne Pathog Dis. 2006;3:4328. http://dx.doi.org/
collaborative study. J Pediatr. 1998;132:77782. http://dx.doi.org/ 10.1089/fpd.2006.3.432
10.1016/S0022-3476(98)70303-8 37. National Association of County and City Health Officials.
23. Hancock DD, Besser TE, Rice DH, Herriott DE, Tarr PI. A lon- Local health department job losses and program cuts: findings
gitudinal study of Escherichia coli O157 in fourteen cattle herds. from January/February 2010 survey. Washington:
Epidemiol Infect. 1997;118:1935. http://dx.doi.org/10.1017/ The Association; 2010.
S0950268896007212 38. Voetsch AC, Kennedy MH, Keene WE, Smith KE, Rabatsky-Ehr T,
24. Dunn JR, Keen JE, Thompson RA. Prevalence of Shiga-toxigenic Zansky S, et al. Risk factors for sporadic Shiga toxinproducing
Escherichia coli O157:H7 in adult dairy cattle. J Am Vet Med Escherichia coli O157 infections in FoodNet sites, 19992000.
Assoc. 2004;224:11518. http://dx.doi.org/10.2460/ Epidemiol Infect. 2007;135:9931000. http://dx.doi.org/10.1017/
javma.2004.224.1151 S0950268806007564
25. Barkocy-Gallagher GA, Arthur TM, Rivera-Betancourt M, Nou X, 39. Ong KL, Apostal M, Comstock N, Hurd S, Webb TH,
Shackelford SD, Wheeler TL, et al. Seasonal prevalence of Shiga Mickelson S, et al. Strategies for surveillance of pediatric
toxin-producing Escherichia coli, including O157:H7 and hemolytic uremic syndrome: Foodborne Diseases Active
non-O157 serotypes, and Salmonella in commercial beef Surveillance Network (FoodNet), 20002007. Clin Infect Dis.
processing plants. J Food Prot. 2003;66:197886. 2012;54(suppl 5):S42431. http://dx.doi.org/10.1093/cid/cis208
26. Agricultural Marketing Resource Center. Lettuce Profile. 40. Doyle MP, Erickson MC. Opportunities for mitigating
2005 4/2012 [cited 2013 May 29]; http://www.agmrc.org/ pathogen contamination during on-farm food production.
commodities__products/vegetables/lettuce/ Int J Food Microbiol. 2012;152:5474. http://dx.doi.org/10.1016/
27. Sodha SV, Heiman K, Gould LH, Bishop R, Iwamoto M, j.ijfoodmicro.2011.02.037
Swerdlow DL, et al. National patterns of Escherichia coli O157
infections, USA, 19962011. Epidemiol Infect. 2014;143:26773.
Address for correspondence: Katherine E. Heiman, Centers for Disease
http://dx.doi.org/10.1017/S0950268814000880
28. Frank C, Kapfhammer S, Werber D, Stark K, Held L. Cattle density Control and Prevention, 1600 Clifton Rd NE, Mailstop A38, Atlanta, GA
and Shiga toxinproducing Escherichia coli infection in Germany: 303294027, USA; email: [email protected]

Emerging Infectious Diseases www.cdc.gov/eid Vol. 21, No. 8, August 2015 1301