Pork implicated in a Shiga toxin-producing Escherichia coli O157:H7 outbreak in Ontario, Canada.
Trotz-Williams, Lise A. ; Mercer, Nicola J. ; Walters, Janice M. 等
Shiga toxin-producing Escherichia coli O157:H7/NM (STEC O157) are
zoonotic pathogens associated with many food-borne and water-borne
outbreaks in North America and elsewhere. (1-3) Most STEC O157
infections have been linked to the consumption of beef, produce or water
contaminated directly or indirectly by cattle manure. (1,4-7) However,
STEC O157 carriage has been reported not only in cattle but also in
other animal species, including other ruminants, swine and poultry.
(2,8-11)
In October 2011, a local health department in southwestern Ontario
was notified of several cases of bloody diarrhoea in persons who had
attended a four-day gathering that had ended four days earlier. The 59
attendees had shared several meals prepared by attendees and caterers,
including pork from a pig roasted whole by a caterer at a pig roast,
served fresh on the second day of the event along with a meal prepared
by another caterer, and as cold and reheated leftovers the following
day. This report describes the investigation of this outbreak, in which
evidence implicated the pork as the source of illness.
METHODS
Case definition
A confirmed case was an attendee reporting enteric symptoms
(nausea, abdominal cramps, vomiting, and/or diarrhoea) beginning anytime
from day 2 of the event to 10 days after the last day of the event, with
laboratory confirmation of STEC O157:H7 infection. A probable case was
an attendee reporting enteric symptoms (nausea, abdominal cramps,
vomiting, and/or diarrhoea) during the same period without laboratory
confirmation of STEC O157:H7 infection.
Case finding and administration of outbreak questionnaire
Menu lists of all meals served were used to produce a questionnaire
on demographics, symptoms, onset and recovery dates, and food exposures
(consumption of individual menu items) at shared meals. A list of
attendees was provided by the host. Those who could be contacted were
asked to complete the questionnaire, either by telephone or in person.
Statistical analysis of outbreak questionnaire data
Questionnaire data were entered in a Microsoft Access database and
analyzed in Stata 11.0 (Statacorp, College Station, Texas) and Microsoft
Excel. Following descriptive analysis, exposure variables (consumption
of meals and individual menu items) were investigated for association
with illness by a retrospective cohort approach, using the <cs>
command with the <exact> option in Stata 11.0 for computation of
Fisher's exact p values, risks (attack rates), risk differences
(attack rate differences) and risk ratios. Variables with two-sided p
values >0.05 were considered statistically significant. Exposures
significantly associated with illness were included in a logistic
regression model with illness as the outcome variable, and a final model
was built by stepwise backward elimination.
Informal interviews
Public health inspectors informally interviewed the host and the
caterers to review how foods had been produced, prepared, transported,
stored and served. As much information as possible was gathered on who
had prepared each menu item, the ingredients the meals contained
(including brands), and where these had been purchased.
Inspection of food preparation premises and kitchen
Public health staff inspected the premises and kitchen where the
catered meal had been prepared, to review menus, refrigerator
temperatures, cleaning records, supply records and staff absenteeism
records, and to audit the hazard analysis and critical control points
(HACCP) systems for all menu items.
Traceback investigations
A whole, dressed, 42-kg pig had been cooked at the pig roast.
Investigations included identification of the plant of origin of the
pig, informal interviews with the plant managers, and identification and
inspection of the supermarket where the pig had been purchased and the
refrigeration unit where it had been stored before being cooked.
Traceback investigations were also conducted for meats, salads and other
foods served at the event that were considered common sources of
food-borne illness.
Collection of samples for laboratory analysis
Stool samples were requested from individuals still ill at the time
of their interview. Duplicate samples were submitted in Cary-Blair
medium for bacterial testing, (12) and in sodium acetate-acetic
acid-formalin (SAF) fixative for parasitic testing. (13)
[FIGURE 1 OMITTED]
Samples of leftover foods from the event (cooked potato, turkey,
carrots, beets, and pork) and water samples from the event premises were
also collected for testing for enteric pathogens.
Three batches of home-made ice cream served on the second and third
days of the gathering had been prepared using eggs from a local hobby
farm where goats and poultry shared a common housing area. Therefore,
water, environmental (soiled animal bedding) and fecal samples from
goats and poultry on this farm were collected for STEC O157 testing.
Laboratory analysis of samples
All samples were tested at the Public Health Laboratory in Toronto,
Ontario by routine enteric diagnostic methods, including testing of food
and water samples by immunomagnetic separation (IMS) of STEC O157 and
selective culture methods based on FDA and Health Canada methods.
(14,15) Isolates were confirmed biochemically as STEC O157 and
Clostridium perfringens, subtyped by pulse-field gel electrophoresis
(PFGE), and serotyped (STEC O157). The hobby-farm fecal and
environmental samples were also tested at the Public Health Agency of
Canada Laboratory for Foodborne Zoonoses, Guelph, Ontario, for any STEC
by screening enrichment broths for Shiga toxin by ELISA (16) and the
Verocell cytotoxicity assay, (17) and for STEC O157 by IMS. (15)
RESULTS
Summary of cases and results of analysis of outbreak questionnaire
data
Descriptive Epidemiological Analysis
Although some ill individuals may have been secondary cases, all
had attended the event and had onset dates within the period specified
in the case definition. Therefore, in analyzing the questionnaire data,
no distinction was made between primary and secondary cases.
The descriptive analyses included data for 52 of the 59 attendees.
Of the 52, 48 completed questionnaires, all within 5-11 days after the
end of the event. Information on demographics and symptoms was available
for 4 additional known cases who did not complete questionnaires.
Age and sex were known for 50 of the 52 persons: 23 (46%) were male
and 27 (54%) were female, ranging in age from 1-84 (median 31, mean
33.5) years. Twenty-nine (55.8%) were identified as cases according to
the case definition, of whom 11 were confirmed cases of STEC O157:H7
infection and 18 were probable cases. There was no significant
difference in either age or sex between those who reported illness and
those who did not (p > 0.1).
Illness and Symptoms
Onset dates of illness ranged from 0-5 days following the last day
of the event (early on Day 4 of the gathering, to the afternoon of Day
9) (Figure 1).
Table 1 summarizes symptoms reported by 27 of the 29 cases;
information on symptoms was unavailable for 2 cases. Recovery dates were
available for only 5 cases, for whom the durations of illness were: less
than one day (n = 1), three days (n = 1) and five days (n = 3).
Seven (24.1%) of the 29 cases were hospitalized: two adults <60
years, one adult >60 years, and four children aged 1-15 years. Verbal
updates from health care professionals and family members indicated that
of these, two children and one adult had signs of haemolytic uraemic
syndrome; one child and the adult required dialysis and blood
transfusions as a result. Another adult who initially reported
resolution of enteric symptoms was subsequently hospitalized with
pancreatitis suspected to be a complication of STEC O157:H7 infection.
No deaths resulting from the outbreak were reported.
Analysis of Exposure Variables
During the four-day event, five communal meals (including 50
individual menu items) were served before the onset date of the earliest
case (Day 4). Analysis of questionnaire data from 48 attendees (25 cases
and 23 non-cases) for exposures significantly associated with an
increased risk of illness (p < 0.05) revealed that of the 50 menu
items investigated, the leftover pork served on Day 3 (Sunday) produced
the highest level of statistical significance (p = 0.0003) and the
highest attack rate (risk) difference (59.6%; Table 2).
Many of the exposure variables were highly correlated. All
individuals included in the analyses had attended the pig roast (lunch)
on Day 2, and all except two had consumed the pork at that meal.
Multivariable logistic regression, using backward elimination to
produce a final model from an initial model that included exposures to
all five menu items listed in Table 2 as independent variables, resulted
in only leftover pork being retained in the model. In this final model,
the odds ratio associated with consumption of leftover pork was 9.0 (95%
CI: 1.8-45.3; p = 0.008).
Similar results were obtained when the exposure analyses were
repeated without five possibly secondary cases with onset dates later
than Day 7.
Informal interviews with food handlers, inspection of food
preparation premises, and traceback investigations
No concerns were identified on inspection and HACCP audit of the
meal caterer's kitchen or of food storage or preparation, and there
were no reports of illness or unexplained absenteeism among food
handlers.
Served foods initially considered high risk for causing food-borne
enteric illness included salads, beef lasagne, turkey, pork from the pig
roast, and the home-made ice cream. However, except for the pork, no
obvious concerns about preparation of those foods arose from
investigations.
According to the pig roast caterer, the pig had been turned
continuously for 12h on a spit positioned 1.5 to 2 feet (approximately
0.5 m) above the coals. Temperature probe readings taken at several
sites in the meat during cooking were reportedly 182[degrees]C, but
probe type, placement and readings had not been documented. After
cooking, the pork had been sliced into hot chafing dishes for immediate
serving. Leftover pork was refrigerated in large aluminium baking pans
of unknown dimensions. A potential concern was that the 42kg carcass had
been roasted in cool air temperatures (minimum 4[degrees]C overnight)
above the heat source and without a shelter, potentially resulting in
inadequate cooking of internal parts of the meat. Further, leftover pork
was stored in a refrigerator full of other leftovers, which may not have
allowed for quick cooling of the meat to optimal storage temperatures.
The pig had originated from a government-regulated and -inspected
plant that processed only pigs. No concerns arose from interviews with
the plant managers or provincial inspection authorities, or from
inspection of the supermarket where the pig had been purchased,
including the supermarket holding units. During inspection, the
temperature of pig carcasses in the store cooler was 3[degrees]C. Pig
carcasses were delivered through a separate entrance into a walk-in
cooler for only pigs and ducks, suggesting minimal potential for
cross-contamination between pig carcasses and other meats at the store.
Laboratory analysis
STECO157:H7 was isolated from 11 of 25 stool samples from
symptomatic attendees, and from the leftover pork--the only tested food
item positive for enteric pathogens. Isolates from eight cases and the
pork had the same PFGE pattern: (ECXAI.0221/ECBNI.0012). The PFGE
pattern of isolates from the three other cases (ECXAI.2684/ECBNI.0012)
differed from that of the other isolates by only one band, indicating
that they were closely related. (18) Clostridium perfringens was
isolated (but not quantified) from two STEC-O157:H7-positive cases, and
was also found at high levels (1.7 x [10.sup.6] organisms per gram) in
the pork. By PFGE, the three C. perfringens isolates were all unrelated
to each other.
The remaining 14 stool samples, other foods, water and hobby-farm
samples submitted for testing were negative for STEC or enteric
pathogens.
DISCUSSION AND CONCLUSION
Many ingredients of the meals for this event had been purchased
from large retailers, and/or were popular, locally-available brands.
However, review of surveillance data from the local area during the
months spanning the outbreak period indicated no concurrent increase in
reported STEC O157:H7 infections in the general population, nor
isolation of STEC O157:H7 with the PFGE patterns identified in this
outbreak. Therefore the source of STEC O157:H7 that caused this outbreak
appeared to have been limited to this event.
Several possible sources of STEC infection at the event were
considered. High-risk food items were identified from assessment--based
on information from interviews and inspections--of the potential for
contamination of food items on the menu lists, and from statistical
analysis of food exposures. In addition, as small ruminants and poultry
may carry STEC O157:H7, (8,9,11) and as at least one STEC outbreak has
been linked to consumption of home-made ice cream, (19) samples from the
hobby farm that provided ingredients for the ice cream were tested for
STEC. However, of all of the potential sources investigated, STEC O157
was isolated only from the pork.
The isolation of identical or closely-related strains of STEC
O157:H7 from 11 confirmed cases and the pork, as well as the negative
results for other possible sources tested, suggests that the pork was
the source of infection. Clostridium perfringens was also identified at
clinically significant levels in the pork, and was isolated from two
STEC O157:H7-positive cases. Though the three isolates were unrelated by
PFGE, (20) infection or co-infection by C. perfringens could have caused
symptoms experienced by some of the cases--probable or confirmed.
Given the high overall attack rate in this outbreak and the lack of
evidence for cross-contamination from other foods during slicing and
serving of the pork, it is likely that contamination of the pork existed
at the time of roasting on Day 2, possibly originating at the time of
slaughter from STEC O157:H7 infection in the living pig or from another
source. Improper cooking, cooling and storing, followed by inadequate
reheating before leftovers were served on Day 3, may have allowed C.
perfringens and STEC O157:H7 to survive and possibly proliferate. A few
attendees may have been infected by consuming the pork on Day 2;
however, freshly-cooked pork consumed on Day 2 was likely carved mainly
from the surface of the carcass, which would have been cooked more
thoroughly than the inner parts of the meat if a significant temperature
gradient from the outside to the inside of the carcass existed during
cooking. However, if cooling of the leftover meat occurred too slowly
following refrigeration, significant bacterial growth could have
occurred, causing the inner portions of the pork served as leftovers,
some cold, on Day 3, to contain significant levels of pathogens. This
hypothesis is supported by the highly significant association between
illness and the consumption of leftover pork on Day 3, with no such
association being found for the freshly roasted pork served on Day 2
when only a few attendees may have been exposed to the pathogen.
Additionally, the incubation period in this outbreak, assuming exposure
on Day 2 or Day 3, was 1-7 days with a median of 3-4 days: within the
reported ranges for STEC O157:H7 of 1 to >7 days, with a median of
3-4 days. (21-24)
Although most commonly found in ruminants, pigs may also carry STEC
O157. (10,11,25-27) However, there are few reports of STEC O157
outbreaks implicating pork as the likely source. (28) The results of
this investigation, including the fact that the plant that had processed
the pig did not process cattle, emphasize the importance of considering
other meat sources besides beef when investigating outbreaks or clusters
of STEC O157 infection.
This investigation also highlights the need for enhancement of
local and provincial educational resources for the public and for food
handlers on proper food handling and storage, to reduce the risk of
food-borne illness at events such as the one described here.
Specifically, there is need for clear provincial guidelines pertaining
to the safe handling and cooking of whole carcasses and large cuts of
meat outdoors on open spits at events such as a pig roasts, a cultural
norm in many rural communities in Ontario.
Acknowledgements: The authors thank the Enteric, Environmental and
Molecular Surveillance departments of the Public Health Laboratory,
Public Health Ontario, as well as the Public Health Agency of Canada
(PHAC) Laboratory for Foodborne Zoonoses, for the testing of outbreak
samples. We also thank all of the public health inspectors, public
health nurses and other staff of the health department who assisted with
the investigation. Thanks also to Dr. Scott McEwen, Dr. David Alves and
others of the Ontario Veterinary College and the Ontario Ministry of
Agriculture, Food and Rural Affairs, who provided information, advice
and other support as we investigated this outbreak.
Sources of support: PHAC; Ontario Ministry of Agriculture, Food
& Rural Affairs; Public Health Ontario; Ontario Ministry of Health
and Long-Term Care; Centers for Disease Control.
Conflict of Interest: None to declare.
REFERENCES
(1.) Bell BP, Goldoft M, Griffin PM, Davis MA, Gordon DC, Tarr PI,
et al. A multistate outbreak of Escherichia coli O157:H7-associated
bloody diarrhea and hemolytic uremic syndrome from hamburgers. The
Washington experience. JAMA 1994;272(17):1349-53.
(2.) Cody SH, Glynn MK, Farrar JA, Cairns KL, Griffin PM, Kobayashi
J, et al. An outbreak of Escherichia coli O157:H7 infection from
unpasteurized commercial apple juice. Ann Intern Med 1999;130(3):202-9.
(3.) Centers for Disease Control (CDC). Two multistate outbreaks of
Shiga toxin-producing Escherichia coli infections linked to beef from a
single slaughter facility--United States, 2008. Morb Mortal Wkly Rep
2010;59(18):557-60.
(4.) Waters JR, Sharp JC, Dev VJ. Infection caused by Escherichia
coli O157:H7 in Alberta, Canada, and in Scotland: A five-year review,
1987-1991. Clin Infect Dis 1994;19(5):834-43.
(5.) Tuttle J, Gomez T, Doyle MP, Wells JG, Zhao T, Tauxe RV, et
al. Lessons from a large outbreak of Escherichia coli O157:H7
infections: Insights into the infectious dose and method of widespread
contamination of hamburger patties. Epidemiol Infect 1999;122(2):185-92.
(6.) Jay MT, Garrett V, Mohle-Boetani JC, Barros M, Farrar JA, Rios
R, et al. A multistate outbreak of Escherichia coli O157:H7 infection
linked to consumption of beef tacos at a fast-food restaurant chain.
Clin Infect Dis 2004;39(1):1-7.
(7.) Talley JL, Wayadande AC, Wasala LP, Gerry AC, Fletcher J,
DeSilva U, et al. Association of Escherichia coli O157:H7 with filth
flies (Muscidae and Calliphoridae) captured in leafy greens fields and
experimental transmission of E. coli O157:H7 to spinach leaves by house
flies (Diptera: Muscidae). J Food Prot 2009;72(7):1547-52.
(8.) Warshawsky B, Gutmanis I, Henry B, Dow J, Reffle J, Pollet G,
et al. Outbreak of Escherichia coli O157:H7 related to animal contact at
a petting zoo. Can J Infect Dis 2002;13(3):175-81.
(9.) Wasteson Y, Johannessen GS, Bruheim T, Urdahl AM,
O'Sullivan K, R0rvik LM. Fluctuations in the occurrence of
Escherichia coli O157:H7 on a Norwegian farm. Lett Appl Microbiol
2005;40(5):373-77.
(10.) Jay MT, Cooley M, Carychao D, Wiscomb GW, Sweitzer RA,
Crawford-Miksza L, et al. Escherichia coli O157:H7 in feral swine near
spinach fields and cattle, central California coast. Emerg Infect Dis
2007;13(12):1908-11.
(11.) Doane CA, Pangloli P, Richards HA, Mount Jr, Golden DA,
Draughon FA. Occurrence of Escherichia coli O157:H7 in diverse farm
environments. J Food Prot 2007;70(1):6-10.
(12.) Dan M, Richardson J, Miliotis MD, Koornhof HJ. Comparison of
preservation media and freezing conditions for storage of specimens of
faeces. J Med Microbiol 1989;28(2):151-54.
(13.) Mank TG, Zaat JO, Blotkamp J, Polderman AM. Comparison of
fresh versus sodium acetate acetic acid formalin preserved stool
specimens for diagnosis of intestinal protozoal infections. Eur J Clin
Microbiol Infect Dis 1995;14(12):1076-81.
(14.) Feng P, Weagant SD, Jinneman K. Chapter 4A, Diarrheagenic
Escherichia coli, Bacteriological Analytical Manual Online, 8th ed,
Revised. Silver Spring, MD: U.S. Food and Drug Administration, 2011.
(15.) Health Protection Branch. Identification of E. coli O157 by
Dynabeads[R] AntiE. coli O157. Bureau of Microbial Hazards. MFLP-90.
Health Canada, 2006.
(16.) Atalla HN, Johnson R, McEwen S, Usborne RW, Gyles CL. Use of
a Shiga toxin (Stx)-enzyme-linked immunosorbent assay and immunoblot for
detection and isolation of Stx-producing Escherichia coli from naturally
contaminated beef. J Food Prot 2000;63(9):1167-72.
(17.) Rahn K, Wilson JB, McFadden KA, Read SC, Ellis AG, Renwick
SA, et al. Comparison of Vero cell assay and PCR as indicators of the
presence of vero-cytotoxigenic Escherichia coli in bovine and human
fecal samples. Appl Environ Microbiol 1996;62(12):4314-17.
(18.) Barrett TJ, Gerner-Smidt P, Swaminathan B. Interpretation of
pulsed-field gel electrophoresis patterns in foodborne disease
investigations and surveillance. Foodborne Pathog Dis 2006;3(1):20-31.
(19.) De Schrijver K, Buvens G, Posse B, Van den Branden D,
Oosterlynck O, De Zutter L, et al. Outbreak of verocytotoxin-producing
E. coli O145 and O26 infections associated with the consumption of ice
cream produced at a farm, Belgium, 2007. Euro Surveill 2008;13(7):pii
8041.
(20.) Johansson A, Aspan A, Bagge E, Baverud V, Engstrom BE,
Johansson KE. Genetic diversity of Clostridium perfringens type A
isolates from animals, food poisoning outbreaks and sludge. BMC
Microbiol 2006;6:47.
(21.) Salmon RL, Farrell ID, Hutchison JG, Coleman DJ, Gross RJ,
Fry NK, et al. A christening party outbreak of haemorrhagic colitis and
haemolytic uraemic syndrome associated with Escherichia coli O157.H7.
Epidemiol Infect 1989;103(2):249-54.
(22.) Bruce MG, Curtis MB, Payne MM, Gautom RK, Thompson EC,
Bennett AL, et al. Lake-associated outbreak of Escherichia coli O157:H7
in Clark County, Washington, 1999. Arch Pediatr Adolesc Med
2003;157(10):1016-21.
(23.) World Health Organization. Foodborne Disease Outbreaks:
Guidelines for Investigation and Control. Geneva, Switzerland: WHO,
2008;57.
(24.) Robinson AL, McKillip JL. Biology of Escherichia coli O157:H7
in human health and food safety with emphasis on sublethal injury and
detection. In: Mendez-Vilas A (Ed.), Current Research, Technology and
Education Topics in Applied Microbiology and Microbial Biotechnology.
Badajoz, Spain: Formatex Research Centre, 2010.
(25.) Feder I, Wallace FM, Gray JT, Fratamico P, Fedorka-Cray PJ,
Pearce RA, et al. Isolation of Escherichia coli O157:H7 from intact
colon fecal samples of swine. Emerg Infect Dis 2003;9(3):380-83.
(26.) Keen JE, Wittum TE, Dunn JR, Bono JL, Durso LM.
Shiga-toxigenic Escherichia coli O157 in agricultural fair livestock,
United States. Emerg Infect Dis 2006;12(5):780-86.
(27.) Farzan A, Friendship RM, Cook A, Pollari F. Occurrence of
Salmonella, Campylobacter, Yersinia enterocolitica, Escherichia coli
O157 and Listeria monocytogenes in swine. Zoonoses Public Health
2010;57(6):388-96.
(28.) Conedera G, Mattiazzi E, Russo F, Chiesa E, Scorzato I,
Grandesso S, et al. A family outbreak of Escherichia coli O157
haemorrhagic colitis caused by pork meat salami. Epidemiol Infect
2007;135(2):311-14.
Received: March 2, 2012
Accepted: July 18, 2012
Lise A. Trotz-Williams, BVetMed, MSc, PhD, [1] Nicola J. Mercer,
MD, MPH, FRCPC, [1] Janice M. Walters, BScN, MEd, MPH, [1] Anne M. Maki,
MLT, [2] Roger P. Johnson, MVSc, PhD [3]
Author Affiliations
[1.] Wellington-Dufferin-Guelph Public Health, Guelph, ON
[2.] Public Health Laboratories--Toronto, Public Health Ontario,
Toronto, ON
[3.] Laboratory for Foodborne Zoonoses, Public Health Agency of
Canada, Guelph, ON
Correspondence: Lise A. Trotz-Williams, Wellington-Dufferin-Guelph
Public Health, 600 Southgate Drive, Guelph, ON N1G 4P6, Tel:
519-846-2715, ext. 4666, Fax: 519-836-7215, E-mail:
Lise.Trotz-Williams@wdgpublichealth.ca
Table 1. Symptoms Reported by Confirmed and Probable
Cases, STEC O157:H7 Outbreak, Southwestern Ontario,
October 2011 (n=27 *)
Symptom Number of Cases Percentage
Reporting Symptom
([dagger])
Diarrhoea 24 89%
Abdominal cramps 20 74%
Nausea 7 26%
Bloody diarrhoea 6 22%
Fever 6 22%
Muscle aches 6 22%
Vomiting 5 19%
Blood in urine 1 4%
* Data on symptoms not available for two additional
cases.
([dagger]) Cases include both confirmed and probable
cases.
Table 2. Risk Differences and Ratios of Food Exposure Variables
Significantly Associated With an Increased Risk of Illness, STEC
O157:H7 Outbreak, Southwestern Ontario, October 2011 (n=48)
Exposure/Food or Persons Who Ate
Beverage (Exposed)
Risk of
Illness
[Attack
Rate]
Ill Not Total (% Ill)
Ill
Fri--Event Day 1 (dinner):
Ate at Fri dinner * 25 17 42 59.5
Garlic bread 23 12 35 65.7
Sat--Event Day 2 (lunch)
([dagger]):
Ate at Sat lunch
([dagger]) 25 23 48 52.1
Sat--Event Day 2 (dinner):
Ate at Sat dinner * 25 17 42 59.5
Pizza (various) * 25 17 42 59.5
Sun--Event Day 3 (lunch):
Ate at Sun lunch * 25 17 42 59.5
Roasted turkey 22 12 34 64.7
Sun--Event Day 3 (dinner):
Ate at Sun dinner 22 13 35 62.9
Leftover pork 18 4 22 81.8
Buns 18 5 23 78.3
Exposure/Food or Persons Who Did Not Eat
Beverage (Unexposed)
Risk of
Illness
[Attack
Rate]
Ill Not Total (% ill)
Ill
Fri--Event Day 1 (dinner):
Ate at Fri dinner * 0 * 6 6 14.3 *
Garlic bread 2 11 13 15.4
Sat--Event Day 2 (lunch)
([dagger]):
Ate at Sat lunch
([dagger]) 0 0 0 -
Sat--Event Day 2 (dinner):
Ate at Sat dinner * 0 * 6 6 14.3*
Pizza (various) * 0 * 6 6 14.3*
Sun--Event Day 3 (lunch):
Ate at Sun lunch * 0 * 6 6 14.3*
Roasted turkey 3 10 13 23.1
Sun--Event Day 3 (dinner):
Ate at Sun dinner 2 10 12 16.7
Leftover pork 4 14 18 22.2
Buns 5 11 16 31.3
Exposure/Food or Attack Risk
Beverage Rate Ratio
[Risk]
Difference
(%)
Fri--Event Day 1 (dinner):
Ate at Fri dinner * 45.2 * 4.2 *
Garlic bread 50.3 4.3
Sat--Event Day 2 (lunch)
([dagger]):
Ate at Sat lunch
([dagger]) - -
Sat--Event Day 2 (dinner):
Ate at Sat dinner * 45.2* 4.2 *
Pizza (various) * 45.2* 4.2 *
Sun--Event Day 3 (lunch):
Ate at Sun lunch * 45.2* 4.2 *
Roasted turkey 41.6 2.8
Sun--Event Day 3 (dinner):
Ate at Sun dinner 46.2 3.8
Leftover pork 59.6 3.7
Buns 47.0 2.5
Exposure/Food or p
Beverage Confidence (Fisher's
Interval exact,
of Risk two-
Ratio * sided)
Fri--Event Day 1 (dinner):
Ate at Fri dinner * 0.7-26.0 * 0.0082
Garlic bread 1.2-15.6 0.0029
Sat--Event Day 2 (lunch)
([dagger]):
Ate at Sat lunch
([dagger]) - -
Sat--Event Day 2 (dinner):
Ate at Sat dinner * 0.7-26.0* 0.0082
Pizza (various) * 0.7-26.0* 0.0082
Sun--Event Day 3 (lunch):
Ate at Sun lunch * 0.7-26.0 * 0.0082
Roasted turkey 1.01-7.8 0.0204
Sun--Event Day 3 (dinner):
Ate at Sun dinner 1.03-13.7 0.0078
Leftover pork 1.5-8.9 0.0003
Buns 1.2-5.3 0.0072
* For exposures with counts of zero in the "unexposed and ill"
cell, estimates of risks, risk differences and risk ratios were
computed by substituting a cell count of one in that cell.
([dagger]) All attendees included in these analyses had eaten at
the Saturday lunch, therefore no Fisher's exact p-value could be
computed for the exposure "Ate at Sat lunch". In addition, no menu
items served at the Saturday lunch were statistically associatedwith an increased risk of illness (p>0.05).
