. 1%. x

7.2.. a... r... ..

WI 1.1:... 3309::
. . n... .52.:

.auwﬁpuhﬁﬁwuﬁﬁ‘ﬁ
3 gm“?
Ryan;
2.
5:”! . 1::
.

I).
t
3
:

Haw C

V .IA: :0 ;
V aampurua):l.l. .ta
(“’31 n7.-

.ruu ,v a.
at. . i.

a» . Hz. b. u

but: .I zihn 23.!!!-
LJ '3... n 1 .

BJL ,.

.. . r...
:12.
2: 3m.-
..2 .25..
‘ . . .137. h.
. ‘5 l. I:
‘ . V H35... 3
.11.".
G (e Bail
.aqthulﬁxﬂtzd 1. . , ,
1.3.1:... : s
.5...

u an... .. er
..: huts... V :hmhw
: . c «"550:
Jr .4 3.1 u. .s .

‘3. A e...
z 1 .
1.71....

c

«an,
3...
:3. 1:2.
ID 1%”: I
2.1.:-

r

a...
.5 ....
any..."

A . .
yeah,

9! . £1: if)...
3.: a I I...) (r

3 9 6521
2.13:2: .i.
5.0.0.5115?

7
13.35 «a. ’ a.
(a? ‘y... .3}:
r. ..a~az..t»l
«33. 3. .3 iii.)
.1... u 5.19L! .7... 1
.33 .531. 352$-
. Ila-31d. $4
a

 

I“! . ..
2:20-43).- l.¢! .|.
aw‘u’ku.9};. ffﬂac? 4:353;

7.....5. i: . 7 ‘0’»...531 I A: 2‘. I.

.. o‘ .. al‘ \‘qr1sxa

.\ 9!)“. as... fart...
9‘31...

‘ ‘ A : n ‘ . E . ‘ v, . u. {$1
Eggﬂw _ 3% . mi 3 ,

.
{(21.53. t.
1. I .

 

.
.. a. 7.,

‘ . v 3.," ,... w 3...“..113 . uu‘
Eiﬁﬁﬁ? . . . ﬁclxéfggﬁs

 

DO 3.5
2‘:

THFmg

I
900 I

 

LIBRARY
michlgan State
University

'w

 

 

This is to certify that the

thesis entitled
Food-borne Illness: A Review from
National, State and Local Perspectives
with an Outbreak Investigational Report

presented by

Susan McIntosh

has been accepted towards fulﬁllment
of the requirements for

Wane in $315131:inng

WWW. gag;

Major professor

Date JEW—

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

N95! 11 9 ram

 

 

 

 

 

 

 

 

 

 

 

 

 

6’01 cJCIRC/DateoprS-DJS

 

 

FOODBORNE ILLNESS: A REVIEW FROM NATIONAL, STATE AND LOCAL
PERSPECTIVES WITH AN OUTBREAK INVESTIGATIONAL REPORT

By

Susan Elaine McIntosh

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Epidemiology

2001

ABSTRACT

FOODBORNE ILLNESS: A REVIEW FROM NATIONAL, STATE AND LOCAL
PERSPECTIVES WITH AN OUTBREAK INVESTIGATIONAL REPORT

By

Susan Elaine McIntosh

F oodbome illness has become a major public health concern in recent years.
Surveillance of foodborne illness at the national, state and local levels has been
instrumental in obtaining information necessary to track patterns and determine causes of
illness. Regulatory agencies use this information to form public health policy. The new
public health interventions included in the FDA 1999 Food Code are a result of
knowledge gained through surveillance of foodborne illness. Local health departments
must continue to irrrprove their investigational skills in order to effectively contribute to
this knowledge base. In 1998 the Ingham County Health Department developed a written
protocol to investigate foodborne illness outbreaks. The collaborative writing of this

protocol is discussed, and an example of an outbreak investigation is described.

ACKNOWLEDGMENTS

The development of the Ingham County Foodborne Illness Protocol was a team
eﬂ‘ort, including two registered nurses in the Communicable Disease Division: Barbara
VanDyke, RN. and Sharon Walker, RN.; and two other registered sanitarians ﬁom the
Environmental Health Division: Terry Anderson, RS. and Diane Gorch RS.

Other members of the Kellogg Center outbreak investigation team were: Betty
Wemette—Babian, RS. and Dick Ives, RS. from Michigan State University Department
of Public Safety; Dean Sienko, M.D., Judy Williams, RN., Sharon Walker, RN., Sharon
Allair, RN., F. Robert Godbold, R.S., Terry Anderson,RS., Diane Gorch, RS. and

1

Anthony Btmnell, RS. ﬁ'om Ingham County Health Department.

Thank-you all very much for your contributions to each of these tasks.

Members of this thesis committee: Michael Collins, M.D., M.S., Claudia

Holzman, D.V. M., M.S., Ph.D. and James Pestka, Ph.D. Thank-you for your Cﬁl’Cﬁll

guidanceinthewritingofthispaper.

iii

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... vi

LIST OF FIGURES ....................................................................................................... vii

INTRODUCTION ........................................................................................................... 1

CHAPTER 1

SURVEILLANCE AND INVESTIGATION OF FOODBORNE ILLNESS

OUTBREAKS IN THE UNITED STATES ..................................................................... 3
History of Surveillance of Foodborne Illness ....................................................... 3
Purpose of Foodborne Illness Surveillance ........................................................... 3
Reporting of Outbreaks ........................................................................................ 4
FoodNet ............................................................................................................... 6
Surveillance for Foodborne Disease Outbreaks 1993-1997 in the US .................. 7
Purpose of Outbreak Investigation ..................................................................... 12
Michigan Law .................................................................................................... 12
Surveillance for Foodborne Outbreaks in Michigan 1990-1999 .......................... 16
Surveillance for Foodborne Outbreaks in Ingham County 1996-2001 ................. 19

CHAPTER 2

MICHIGAN FOOD LAW 2000 .................................................................................... 21
Time and Temperature Parameters for Controlling Foodborne Pathogens .......... 23
Demonstration of Knowledge ............................................................................ 29
Employee Health Controls ................................................................................. 33
Controlling Hands as a Vehicle of Contamination .............................................. 36
Consumer Advisories ......................................................................................... 37
Conclusion ......................................................................................................... 39

CHAPTER 3

DEVELOPMENT OF THE ICHD FOODBORNE ILLNESS INVESTIGATION

PROTOCOL ................................................................................................................. 42
The Role of ICHD in Foodborne Illness Investigation ........................................ 42
Collaboration on the Writing of 3 Protocol ......................................................... 44
Purpose .............................................................................................................. 45
Policy Statement ................................................................................................ 45
Standard ............................................................................................................ 45
Guidelines for Outbreak Investigation ................................................................ 46

iv

CHAPTER 4

AN OUTBREAK OF FOODBORNE ILLNESS ............................................................ 58
Introduction ....................................................................................................... 58
Methods ............................................................................................................. 59

Environmental Investigation Methods .................................................... 59
Epidemiological Methods ....................................................................... 60
Laboratory Investigation Methods .......................................................... 61
Results ............................................................................................................... 63
Environmental Investigation ................................................................... 63
Epidemiological Results ......................................................................... 65
Laboratory Results ................................................................................. 66
Discussion ......................................................................................................... 67
Limitations ............................................................................................. 67
Conclusion and Discussion ..................................................................... 68
Recommendations to the Foodservice Establishment .............................. 71

CONCLUSION ............................................................................................................. 76

APPENDIX

BACKGROUND INFORMATION ON CLOST RID] UM PERF RINGENS AND

BA CILL US CERE US ..................................................................................................... 77

REFERENCES .............................................................................................................. 86

LIST OF TABLES

Table 1. Reported and estimated illnesses, ﬁ'equency of foodborne transmission and
estimated foodborne illness in the US ............................................................................. 9

Table 2. Incidence rate (per 100,000 population) of diagnosed infections for

pathogens at the ﬁve original FoodNet sites, 1996-2000, for all eight sites for 2000*,

by year and pathogen and the Healthy People 2010 goal” (when applicable)

(2000 rates are preliminary) ........................................................................................... 11

Table 3. State of Michigan Reported Outbreaks and Final Reports with Conﬁrmed
Etiology, 1998 and 1999 ................................................................................................ 18

Table 4. Summary of Food Speciﬁc Attack Rates, Odds Ratios, X2 and p-values
Kellogg Center Outbreak — May 1998 ........................................................................... 66

LIST OF FIGURES

Figure 1. Learning model for Public Health issues .......................................................... 2
Figure 2. Passive Surveillance Pyramid .......................................................................... 5
Figure 3. Reported Foodborne Disease Outbreaks in the US. 1991-1997 ........................ 8
Figure 4. Estimated frequency of foodborne illness in the US ...................................... 10
Figure 5. Foodborne illness complaints reported to MDA, 1990-1999 .......................... 17
Figure 6. Foodborne illness complaints received by ICHD, 1996-2001 .......................... 20
Figure 7. Epidemic Curve ............................................................................................. 65
Figure 8. Recipes for chicken augratin entree served at banquet, May, 1998 ................. 73
Figure 9. HACCP-based recipe for Chicken Entree served May, 1998

Indicating Critical Control Points (CCP) ........................................................................ 74
Figure 10. Outbreaks of C. perfringens in the us ........................................................ 78
Figure 11. Cases of C. perfringens in the US ................................................................ 79

Figure 12. Number of Outbreaks of Selected Bacterial Pathogens, 1991-1997 (Lower
graph is of smaller scale, excludes Salmonella to magnify the remaining pathogens.) 80

Figure 13. Number of Reported Cases of Selected Bacterial Pathogens, 1991-1997.
(Lower graph is smaller scale, Salmonella deleted, to magnify remaining pathogens.) 81

Figure 14. Reported B. cereus Outbreaks in the US ...................................................... 84

Figure 15. Reported B. cereus Cases in the US ............................................................ 84

vii

INTRODUCTION

Public health surveillance can be deﬁned as the routine collection, analysis and
dissemination of all data that may be relevant for the prevention and control of a public
health problem.” Epidemiology is deﬁned as the study ofthe distribution and
determinants of health-related states or events in speciﬁed populations and the
application of this study to control of health problems.31 The control of communicable
disease requires a clear understanding of the epidemiology of the disease in addition to a
reliable surveillance system. Reporting of communicable disease is an essential part of
any public health surveillance program.

Foodborne ilhless continues to be a serious public health problem. It is estirmted
that foodborne diseases cause 76 million illnesses, 325,000 hospitalizations and 5,000
deaths in the United States each year. The epidemiology of foodborne illness is
constantly changing, which makes it necessary to learn from and adapt to those changes
in order to effectively institute prevention measures. A typical learning model can be
used to illustrate how this is accomplished (Figure 1).

Surveillance systems are necessary to track patterns of foodborne illness in the
US. The epidemiological investigations of foodborne outbreaks are a necessary part of
the process. Surveillance at the local or state level will indicate if a foodborne illness
outbreak is occurring. Results ﬁ'om these outbreak investigations are compiled at a
national level of surveillance. At this level much can be learned about causes of
foodborne illness. Along with scientiﬁc research stimulated by these results, prevention

measures can be put into place.

 

 

Surveillance
Prevention Epidemiological
Measures 6 Investigations
Applied Research

 

 

 

Figure 1. Learning Model for Public Health Issues

This paper will discuss the current state of surveillance in the United States, the State
of Michigan and Ingham County. The FDA 1999 Food Code has recently been adopted into
the Michigan Food Law 2000. The major changes in the code are a result of knowledge
gained from the analysis of data collected by various surveillance methods at the national
level. These changes will be critiqued with regard to how effectively they will address
current problems in food service sanitation.

Finally, the development of a local health department protocol for investigating
foodborne illness outbreaks will be described, and an example of a local outbreak

investigation using the protocol will be outlined.

CHAPTER 1
SURVEILLANCE AND INVESTIGATION OF FOODBORNE ILLNESS

OUTBREAKS IN THE UNITED STATES

History of Surveillance of Foodborne Illness

Early in the 20‘“ century, state and territorial health ofﬁces in the United States
became concerned about the high levels of morbidity and mortality caused by infantile
diarrhea and typhoid fever. Investigation and reporting of “enteric fever” was
recommended in order to better understand the role of milk, food and water in outbreaks
of intestinal illness. This knowledge would become the basis for public health action.
The Public Health Service began publishing summaries of outbreaks of gastrointestinal
illness attributed to milk in 1925. By 1938 all foods were included in the summaries.8
The surveillance system in the US. continued to develop into the multi-faceted program
it is today. National surveillance systems for foodborne diseases now in place in the US.
include outbreak surveillance, laboratory based passive surveillance of individual cases,
physician-based surveillance and active surveillance. The Centers for Disease Control
(CDC) now assumes the responsibility for publishing reports of foodborne illness on a

national level.

Purpose of Foodborne Illness Surveillance
Foodborne disease outbreak surveillance has served three main purposes: 8’48
1. Disease prevention and control: Local and state public health ofﬁcials are

responsible for correction of faulty food preparation practices in food service

establishments and prompt identiﬁcation and removal of contaminated products
ﬁ'om the commercial market.

2. Knowledge of disease causation: Outbreak investigations have provided
invaluable information regarding new and emerging pathogens as well as the
long-standing pathogens. Prompt and thorough investigations of foodborne
outbreaks aid in the timely identiﬁcation of etiologic agents of the disease.

3. Administrative guidance: Trends in the prevalence of outbreaks caused by
speciﬁc organisms, food vehicles and food handling can be analyzed. Changes
can then be made in the regulatory and educational areas in order to more

effectively implement preventive actions.

Reporting of Outbreaks

The reporting of outbreaks to CDC F oodbome Disease Outbreak Surveillance
System is considered passive surveillance. A foodborne Disease Outbreak is deﬁned by
CDC as “the occurrence of two or more cases of a similar illness resulting from the
ingestion of a common food. Only a small percentage of foodborne illness is identiﬁed
through this passive system. In order for a foodborne illness to be reported all the way to
the CDC it must ﬁrst be reported to the local health department, recognized as an
“outbreak”, investigated, reported to the state health department and then passed on to the
CDC. Alternatively, a case of foodborne illness is reported when a person becomes ill,
seeks medical care, is culture-conﬁrmed with a particular pathogen and this case is
reported to the state health department and then to the CDC. It is estimated that only 5%

of bacterial foodborne illness is reported to the CDC by this system.3

Reported to Health Dept/CDC

 

Cultur conﬁrme case

 

 

 

Person becomes ill

Population \

Figure 2. Passive Surveillance Pyramid 53

 

There are several reasons why foodborne disease is underreported. Oﬁen the ill
person does not seek medical care. In many cases the illness is not severe and is of short
duration or is mistaken for the “stomach ﬂu” and not recognized as being transmitted by
food. Often the health care provider fails to obtain a specimen for diagnosis. The
laboratory may not have performed the necessary diagnostic test or the test results may
not be reported to the local health department. Michigan’s cormnunicable disease rules
(section 5111 of Act No. 368 of the Public Acts of 1978, as amended, being 333.5111 of
the Michigan Compiled Laws)45 require certain human pathogens to be reported to the
local health department. The foodborne pathogens that are required to be reported by
laboratories include: Campylobacterjejuni, Clostridium botulinum, Cryptosporidium
species, Cyclospora species, Giardia Iamblia, Hepatitis A, Listeria monocytogenes,

Salmonella species, Shiga toxin producing E. coli, Shigella species, Trichinella spiralis

and Yersinia enterocolitica.18

The illnesses associated with these organisms are required
to be reported by physicians to the local health department. Sporadic illness caused by
some pathogens such as Clostridium perﬁ'ingens or Bacillus cereus is not reportable.
Therefore the only cases reported for these two organisms and others that are not
reportable are those associated with outbreaks. In addition, local health departments may

not be equipped with the resources necessary to thoroughly investigate an outbreak.

FoodNet

In 1995 the Foodborne Diseases Active Surveillance Network (FoodNet) was
established as the foodborne disease component of the Emerging Infections Program
(EIP) of the CDC. FoodNet is a population-based active surveillance program that
provides a network for responding to emerging foodborne disease and for determining
more accurately the ﬁequency and severity of foodborne illnesses that occur in the
United States. By 1998 there were seven CDC EIP sites (Minnesota, Oregon and
selected counties in California, Connecticut, Georgia, Maryland and New York)
collaborating with the US. Department of Agriculture (USDA) and the US. Food and
Drug Administration (FDA). The total population of these sites was 20.5 million people
or 7.7% of the US. population.” In 1999 the remaining counties in Georgia and eight
counties in the metropolitan Albany, New York, area were added. In 2000, 11 counties
in Tennessee and Contra Costa County in California were added, bringing the FoodNet
surveillance population to 29.5 million persons (10.8% of the 1999 US. population). In

2001 selected counties in Colorado and Maryland will be added to the FoodNet area,

bringing the FoodNet population to approximately 33.1 million persons (12% of the 1999
population).

Data on nine foodborne pathogens (Campylobacter, Cryplosporidium,
Cyclospora, Esherichia coli 0157, Listeria, Salmonella, Vibrio and Yersinia) is
collected. More than 300 clinical microbiology laboratories in the catchment areas are
contacted regularly to obtain information on all laboratory-conﬁrmed cases of diarrhea]
ilhiess. Physician stool-culturing practices vary greatly. F oodNet surveyed more than
5000 physicians to determine how often and under what conditions they order these tests.
A population survey was conducted (approximately 9000 randomly selected people per
year) in order to estimate the percentage of the population that seeks medical care when
aﬂlicted with diarrheal illness. Case control studies of E. coli 0157, Salmonella
sero groups and Campylobacter have also been conducted to provide information on food
items and other risk factors associated with diarrheal illness. Information ﬁom all of
these investigations lms been combined and analyzed each year in order to more precisely

monitor the burden of foodborne illness and to compare these ﬁndings with time.

Surveillance for Foodborne Disease Outbreaks from 1993-1997 in the U.S,

During the years 1993-1997 2,751 outbreaks of foodborne illness were reported to
the CDC ( 489 in 1993, 653 in 1994, 628 in 1995, 477 in 1996 and 504 in 1997).8 This
study was based on the Foodborne Disease Outbreak Surveillance System. These

reported outbreaks caused 86,058 persons to become ill over the ﬁve year period.

 

Number of Reported Foodborne Disease Outbreaks in the
U.S.

 

 

 

 

 

 

 

1991 1902 19” 1994 1995 1996 1997
Y.

 

 

 

Figure 3. Reported Foodborne Disease Outbreaks in the US. 1991-1997 3'“

In 1999 the CDC published results of a comprehensive study designed to more
accurately estimate the burden of foodborne illness in the United States. Sources of data
for this project include the Foodborne Diseases Active Surveillance Network (FoodNet),
the National Notiﬁable Disease Surveillance System, the Public Health Laboratory
Information System, the Foodborne Disease Outbreak Surveillance System, the National
Ambulatory Medical Care Survey, the National Hospital Ambulatory Medical Care
Survey, the National Hospital Discharge Survey, and the National Vital Statistics System
as well as selected published studies.46 Selected results of this study are shown in Table
1. Known pathogens cause approximately 38.6 million illnesses per year, which include
5.2 million (13%) due to bacteria, 30.9 million (80%) due to viruses and 2.5 million (7%)
due to parasites. Of these 38.6 million, 13.8 million are believed to be of foodborne
transmission. For known foodborne transmitted illnesses, approximately 30% are caused

by bacteria, 67% by viruses and 3% by parasites (see Table 1).

Table 1. Reported and estimated illnesses, frequency of foodborne transmission and
estimated foodborne illness in the U.S 33

 

 

 

 

 

Disease or Organism Estimated Reported % Estimated
Bacterial Total Cases Cases Foodborne Foodborne
Bacillus cereus 27,360 792 100 27,3 60
Botulism, f.b. 58 29 100 58
Brucella spp. 1554 111 50 777
Campylobacter 2,453,926 102,219 80 1,963,141
C.perﬁingens 248,520 7,194 10 248,520
E.coli 0157:H7 73,80 6,899 85 62,458
E. coli, other 158,840 4,389 30-70 79,420
L. monocytogenes 2,5 1 8 1 ,632 99 2,493
Salmonella whi 824 412 80 659
Salmonella, other 1,412,498 78,653 95 1,341,873
Shigella 448,240 41,212 20 89,648
Staph, f.b. 185,060 5,357 100 185,060
Strep, f.b. 50,920 1,474 100 50,920
V. cholerae, 54 27 90 49
(toxigenic)

Vvulniﬁcus 94 47 50 47
Vibrio,other 7,880 505 65 5,122
Xenterocolitica 96,368 2,536 90 86,731
Subtotal 5,204,934 4,175,565
Parasitic

C.parvum 300,000 9,418 10 30,000
C. cayetanensis 16,264 526 90 14,638
Giardia lamblia 2,000,000 129,907 10 200,000
T oxoplasma gondii 225,000 15,000 50 1 12,500
T richinella spiralis 52 26 100 52
Subtotal 2,541,316 357,190
Viral

Norwalk-like virus 23,000,000 40 9,200,000
Rotavirus 3,900,000 1 39,000
Astovirus 3,900,000 1 39,000
Hepatitis A 83,391 27,797 5 4,170
Subtotal 30,883,391 9,282,170
Grand Total 38,629,641 13,814,924

 

Symptom-based data was used to estimate the total number of acute
gastrointestinal illnesses. Known cases were subtracted from this total to estimate illness
due to acute gastroenteritis of unknown agents. In Figure 4 the estimates for both the
known and unknown pathogens are added together to give a grand total of 76 million

foodborne illnesses per year in the United States.

 

Pathogen-based Data Symptom-based data

Known Pathogens Acute Gastroenteritis

(39,000,000 illnesses) (211,000,000 illnesses)
Known Pathogens- Known Pathogens- Unknown Agents-
Non-gastroenteritis Acute gastroenteritis Acute gastroenteritis
(300,000 illness) (38,000,000 illnesses) (173,000,000 illnesses)

l l l

Foodborn_e Transmission

120,000 illnesses + 14,000,000 illnesses + 62,000,000 illnesses =

 

 

76,000,000 Estimated Total Burden of Foodborne Illness

 

 

Figure 4. Estimated frequency of foodborne illness in the U.S.33

Preliminary results ﬁom FoodNet data on the incidence of foodborne illnesses in
2000 have recently been published. '5 The number of sites and the population under
surveillance has nearly doubled since 1996. Therefore, to be consistent, temporal trends
in rates of foodborne illness were compared using data from only the original ﬁve sites

(Table 2). Included in the table is the incidence rate for all sites for 2000.

10

Table 2. Incidence rate (per 100,000 population) of diagnosed infections for
pathogens at the ﬁve original FoodNet sites, 1996-2000, for all eight sites for 2000*,
by year and pathogen and the Healthy People 2010 goal" (when applicable)(2000
rates are preliminary) 8"

 

Pathogen 1996 1997 1998 1999 2000 2000* HP2010"

 

Campylobacter 23.5 25.2 21.4 17.5 20.1 15.7 12.3
Coptespoﬁdium NR 3.7 2.9 1.8 2.4 1.5

cyclospora NR 0.4 0.1 0.1 0.1 0.1
E. coli 0157:H7 2.7 2.3 2.8 2.1 2.9 2.1 1.0
Listeria 0.5 0.5 0.6 0.5 0.4 0.3 0.25
Salmonella 14.5 13.6 12.3 13.6 12.0 14.4 6.8
Shigella 8.9 7.5 8.5 5.0 11.6 7.9
Vibrio 0.2 0.3 0.3 0.2 0.3 0.2
Yersinia 1.0 0.9 1.0 0.8 0.5 0.4

 

For all ﬁve years Campylobacter was the most frequently diagnosed pathogen,
followed by Salmonella, Shigella and E. coli 0157:H7. Although the relative order of
incidence rates are the same between the 2000 ﬁve-site and 2000 eight-site statistics, the
differences show regional differences in pathogen isolation rates. The preliminary 2000
rates were based on the 1999 census population. Once the ﬁnal 2000 census is
determined the adjusted rates will be lower. These FoodNet foodborne disease rates will
continue to be a more accurate measure of the progress toward the Healthy People 2010
national health objectives as the percentage of total U.S. population encompassed by

FoodNet increases.

11

Purpose of Outbreak Investigation

Although the number of outbreaks investigated by local health departments

reported to the CDC represent only the tip of the iceberg, there are several speciﬁc

reasons why these investigations must be done: 27

1.

To identify illnesses associated with an incident or outbreak and to verify that the

causative agent is foodborne

2. To detect all of the cases, the causative agent, the implicated food and the location
where the food was mishandled

3. To determine the source and the mode of contamination or practices by which the
proliferation and/or survival of the etiologic agent occurred

4. To stop the outbreak or prevent further exposures

5. To gather information for the surveillance purposes on the epidemiology of
foodborne diseases and the etiology of the causative agents that can be used for
education, training and program planning which can make an impact on
preventing foodborne illness.

Michigan Law

By law, local health departments in Michigan are required to investigate

foodborne illness outbreaks. Michigan’s Public Health Code Sec. 2433 (2) (c) states that

“A local health department shall make investigations and inquiries to (i) the causes of

disease and especially of epidemics (ii) the causes of morbidity and mortality (iii) the

causes, prevention, and control of environmental health hazards, nuisances and sources of

illness.” 45 Under the authority of Michigan’s Public Health Code, Section 12909, Food

12

Service Sanitation Administrative Rule 5402 (2) states “Procedures for investigating
suspected foodborne illness outbreaks shall be implemented in accordance with the
publication entitled “Procedures to Investigate Foodborne Illness” prepared and
published by the International Association of Milk, Food and Environmental Sanitarians,
Incorporated, or an alternative local plan submitted to and approved by the
department. . .A complete summary report shall be prepared and submitted to the
department.” Section 12912 of the Michigan Public Health Code states that “(1) A local
health department shall investigate foodborne diseases and poisonings or suspected
foodborne diseases and poisoning connected with a food service establishment,
temporary food service establishment, or vending machine located within its jurisdiction
and shall promptly make a report of its ﬁndings to the department.” ’5

Until 1996, local health departments were conducting food service inspections
under the direction of the Michigan Department of Public Health. Retail grocery
operations were inspected by the Michigan Department of Agriculture using a different
food safety standard. Grocery stores have ventured into more food service activities over
the past decade. The result was tint many grocery stores were being regulated by both
agencies with two different standards. The need for a “uniﬁed code” became clear. In
1996 the food service program was moved to the MDA and the development of a uniform
code was initiated. Michigan Department of Agriculture convened a group of industry,
regulatory and academic experts to evaluate the current FDA Model Food Code for
possible adoption in Michigan. Michigan adopted the Michigan Food Law 2000 in
November of 2000, House Bill 5196.44 In this Law the FDA 1999 Food Code was

adopted by reference.

13

Section 3129 of the Food Law 2000 states “(1) A local health department shall
investigate an allegation of foodborne diseases and poisonings or suspected foodborne
diseases and poisonings connected with food service establishments within its jurisdiction
and delegated authority and shall promptly make a report of its ﬁndings to the
department. (2) If an investigation indicates that a source of a foodborne disease or
poisoning was from a food processing, food storage, or similar type of food establishment
over which the department has legal jurisdiction or responsibility, the local health
department shall immediately notify the director while the local health department is
completing the investigation.” Section 3131 states that “ (1) A local health department
shall develop and implement a communications system with other applicable
governmental agencies, individuals, and organizations including, but not limited to,
hospital emergency rooms and state employees and basic informtion necessary to
initiate a foodborne illness outbreak investigation. The information provided in the
cormnunications system shall be updated annually. (2) Procedures for investigating
suspected foodborne illness outbreaks shall be implemented, consistent with procedures
contained in the publication entitled “Procedures to Investigate F oodbome Illness, 5th
Ed.,” prepared and published by the International Association of Food Protection and
incorporated by reference or an equivalent plan submitted to and approved by the
department. (3) All information gathered during the investigation which is not exempted
from disclosure under section 13 of the Freedom of Information Act, 1976 PA 442, MCL
15.243, and shall be made available to the owner, operator, or his or her employees to
minimize the possibility of reoccurrence of the foodborne illness and to assure

compliance with the code and this act.” 4"

14

Generally, foodborne disease outbreaks investigated by the local health
department such as Ingham County are then reported to the state health department who
then report to the CDC through the F oodbome Outbreak Surveillance system. In
Michigan, the local health departments report directly to the Michigan Department of
Agriculture (MDA) Food and Dairy Division. This division provides support to local
health departments in the areas of consultation and training. The Science and
Technology Section of this division also maintains a F oodbome Illness Surveillance
Database. The information for this database is taken directly from complaints and reports
ﬁled by local health departments and MDA food inspectors. WA then forwards this
information to the MDCH and the CDC.

Complaints received by the MDA are classiﬁed as follows: 35

Famrjy: Two or more related cases with symptoms compatible with foodborne
illness etiology, i.e., gastroenteritis.

Al_ert_: Generally, a single reported case of foodborne illness

FBI/F BIO: (foodborne illness/foodbome outbreak): an incident involving two or
more unrelated cases having similar clinical features or having the same pathogen.
There is also a time, place or person association among the cases. Single cases of
botulisrn, mushroom poisoning, ciguatera or paralytic shellﬁsh poisoning or other rare
ilhress, or a case that can be deﬁnitely related to food also are classiﬁed as FBIs.

E1 (Isolated Consumer complaint): Incidents of adulterations of food or
complaint of food quality (ex: hair found in potato salad)

The Michigan local health departments rely on a cost-sharing system with the

State of Michigan to operate its mandated programs. Therefore they must conform with

15

certain requirements, “minimum program requirements” (MPRs) in each of their
mandated programs in order to continue the cost-sharing relationship. The MPRs became
effective October 1,2000 and contain detailed requirements for fulﬁlling the Michigan
Public Health Code (now Michigan Food Law 2000). The local health department is
reviewed yearly for compliance with these MPRs. Ingham County continues to improve
their system for reporting and surveillance of foodborne illness and the investigation of

foodborne disease outbreaks.

Surveillance for Foodborne Outbreaks in Michigan from 1990-1999

Although the WA has been collecting local health department data on suspected
foodborne illness and outbreaks for several years, the department began sumrmrizing
their yearly surveillance of foodborne illness beginning with the year 1998 in order to
provide oversight for the MPRs.

Figure 5 shows the number of foodborne illness complaints received by the MDA
from 1990-1999. In 1997 MDA began using the IAMFES subcategories described
earlier. In 1998 alerts and family related illnesses have been separated.

The number of people reportedly ill due to foodborne illnesses increased in 1997 but
leveled off in 1998 and 1999. In 1998 only 102 incidents of the total 1250 events met the
legal deﬁnition of foodborne illness outbreak. These 102 incidents included 986 ill
people. In 1999, 141 of the 1143 reported incidents met the legal deﬁnition, including

1236 ill people.36 In 1998 ﬁnal investigative reports were not received from local health

16

e2... 3363— £52 3 63.89.. EEaEEea n82: oEeneeeh .m charm

 

an;
«a: 03.. 52. ca: 32. a: «a: «a: 32. 82

 

 

5.1.3.3
iawam
tagged-.9
0034:-
33.50

.3. .88 I

355m .88-

 

 

 

II! reqwnN

 

525.2 5 32.82 "3526.58 sees: essence".

 

 

17

departments for 57 (56%) of the total 104 outbreaks. The MDA yearly report for 1999

pointed out that ﬁnal investigative reports were not received from local health

departments for 28 (19%) of the 141 outbreaks. Table 3 shows that the etiology of

foodborne outbreaks was identiﬁed in only 4 (9%) of the 45 outbreak reports in 1998 and

in 45 (40%) of the 113 reports received in 1999.

Table 3. State of Michigan Reported Outbreaks and Final Reports with Conﬁrmed

 

 

 

 

 

Etiology, 1998 and 1999 ’5"
1998 1999

_1_4_gent # events #111 #events #ill
B. cereus 2 44 l 2
C. perfringens 2 23
Campylobacter 5 5
E. coli 0157:H7 5 5
E. coli, other 1 2
Salmonella 1 1 25
Shigella 1 1
S. aureus 2 4
Calcivirus 1 62
Hepatitis A 1 3 11 51
Norwalk virus 1 34 3 75
Total (etiology 4 81 45 255
conﬁrmed)
Total reports 45 1 13
Total outbreaks 102 986 141 1236

 

Improvement is shown between 1998 and 1999. The percentage of ﬁnal

investigative reports submitted to MDA increased from 44% to 72% of total outbreaks.

With the enforcement of MPRs at the local level, this percentage will continue to

18

increase. The percentage of outbreaks where the etiology was conﬁrmed in the reports
also increased, from 9% to 40%. This may be due to several factors:

1. More testing of stool and food specimens is being requested as the local health
departments improve their investigative epidemiological skills and apply the
required guidelines according to MPR’s.

2. More testing requests are being accepted by the MDCH laboratories.

3. Testing for Norwalk-like viruses is becoming more available and local health
department personnel are becoming more educated on when to request testing for
these viruses.

The publication of the Year-End Summaries for Foodbore Illness Surveillance by the
MDA as well as increased opportunities for consultation with MDA staff and laboratory
support services will continue to improve participation by local health departments and

improve the surveillance system for foodborne illness in Michigan.

Surveillance for Foodborne Outbreaks in Ingham County, 1996-2001

Ingham County keeps a complete log of foodborne illness complaints received.
Figure 6 indicates a slight increase in the number of complaints received yearly. There
has also been a slight increase in the number of illnesses determined to be outbreak
related. The preliminary data for 2001 indicate a sharper increase for the year 2001. This
increase may be attributed to improved reporting by the public as well as improved

response by the ICHD in investigating possible outbreaks.

l9

 

 

Foodborne Illness Complaints: 1996-2001 in Ingham
County (2001 ﬁrst quarter date only)

I Total Events
ITotal III
as ill-FBIO

   

Year

 

Figure 6. Foodborne illness complaints received by ICHD, 1996-2001.

20

 

CHAPTER 2

MICHIGAN FOOD LAW 2000

The purpose of surveillance has been discussed earlier. When the etiology and
epidemiology of individual cases and outbreaks is documented and reported by the local
health agencies to the state level, this information is compiled at a national level. The
CDC, part of the U.S. Public Health Service, is not a regulatory agency. Its mission is to
monitor, investigate, control and prevent public health problems. The CDC works
closely with the state health departments to monitor the ﬁequency of speciﬁc diseases
and conducts national surveillance for them as well as providing expert epidemiologic
and microbiologic consultation to health departments. Trends in the occurrence of
outbreaks caused by speciﬁc organisms, food vehicles and food handling practices can be
analyzed. This information is then used by regulatory agencies to direct public health
policy.

The CDC has published summaries of the surveillance of foodborne disease
outbreaks. There have been three comprehensive studies surmnarizing data compiled
ﬁ'om the years 1983-1997 (1983-1987, 1988-1992 and 1993-1997) “"6” Individual
epidemiological studies performed by local or state health departments that were reported
to the CDC are included in these nationally comprehensive studies. Vehicles of
transmission and contributing factors were tabulated for these outbreaks. From this data

the CDC has concluded that there are ﬁve leading causes of foodborne illness outbreaks:

21

11,17

Five Leadmg' Causes of F oodbome Illness Outbrea_ks:
1. Improper holding temperatures
2. Poor personal hygiene of food workers
3. Inadequate cooking of foods
4. Contaminated equipment
5. Obtaining food ﬁ'om unsafe sources (example: oysters harvested from
contaminated oyster beds)

The FDA publishes the Food Code,5 I a reference that guides retail food outlets
such as restaurants and grocery stores on how to prevent foodborne illness. It is neither
federal law nor federal regulation and is not preemptive. It is commonly adopted and
used by agencies at all levels of government that have regulatory responsibilities for food
safety. The food code is updated every two years. This process is a collaborative effort
by FDA, CDC, Food Safety and Inspection Service (F SIS) and the Conference for Food
Protection. The conference is comprised of representatives ﬁom regulatory agencies at
all levels of government, the food industry, academia and consumer organizations that
work to improve food safety at the retail level. . The most recent edition is the 1999 Food
Code. In November 2000 The Michigan Food Law, Public Act 92 was enacted which
adopts the most recent edition of the Food Code. Prior to that the 1976 Edition was used
in Michigan.

The provisions of the Food Code establish a system of prevention and safeguards
designed to minimize foodborne illness, ensure employee health, industry manager

knowledge, safe food, nontoxic and cleanable equipment and acceptable levels of

22

sanitation of food establishment premises. The Food Code has addressed controls for risk
factors identiﬁed by the CDC as contributors to foodborne outbreaks (those listed above).
Five important public health interventions are established in this edition of the Food
Code designed to address these contributing factors:
1. Time and temperature parameters for controlling pathogens
2. Demonstration of knowledge by the foodservice manager
3. Employee health controls
4. Controlling lands as a vehicle of contamination
5. Consumer advisories of risks of eating raw or undercooked foods of animal
origin.
These additions and changes have been made based on the scientiﬁc evidence
gained through surveillance and supportive research, that foodborne illness is caused
primarily when these controls are not in place. Each of these new code items will be

discussed with regard to how successful they should be in limiting foodborne illness.

Time and Temperature Parameters for Controlling Foodborne Pathogens

The most frequently cited food handling practice associated with foodborne
illness outbreaks is improper food holding temperatures, both hot-holding and cold-
holding. New scientiﬁc research has resulted in the following changes.

One of the changes in the code is the lowering of the maximum cold-holding
temperature of potentially hazardous food from 45° F. to 41° F. In conjunction with this
requirement is the time limit for reﬁ'igerated storage of ready-to-eat potentially hazardous

foods. Until recently it was believed that pathogens would not grow at temperatures of

23

45° F. or below. However, in the last several years two emerging foodborne pathogens,
Listeria monocytogenes and Yersinia enterocolitica, have been found to grow in
refrigeration temperatures.

L. monocytogenes, recently found to be predominantly foodborne, is found in a
variety of ready-to-eat foods such as soft cheeses and meats. It causes an estimated 2500
illnesses and 500 deaths in the United States each yearn” Although Listeriosis is rare,
septicemia and meningitis may result from infection and the case fatality rate is estimated
at 20 percent.12 Listeriosis in pregnant women may lead to miscarriage, stillbirth, or
meningitis in newborns. The disease is more severe in individuals with weakened
irmnune systems.

Kenterocolitica has been recently found to be associated with undercooked
pork." It causes an estimated 96,000 illnesses per year with a very low case fatality
ratef“3 As a result of increased surveillance for these organisms and the discovery that
they grow at higher reﬁigeration temperatures, the FDA revised the code to 41° F. as a
maximum for reﬁ‘igeration.2|

The Michigan Food Law 2000 has given industry ample time to comply with this
change. Commercial refrigeration units built in the last few years have already been
designed to be capable of holding temperatures at 41° F. or below. Food service
establishments will have until May of 2006 to comply with this rule. All of their
reﬁ'igeration units must be capable of holding at 41° F. or below by that date. Local
health department sanitarians have begun to document and track the temperatures of
every refrigeration unit in every food service facility. Until May of 2006, ready-to-eat

potentially hazardous food must be consumed or disposed of within 7 days if held at 41°

24

F. or below. If the refrigeration units are holding temperatures between 41° F. and 45° F.,
the time limit is 4 days storage. These foods must be labeled with the date by which the
foods must be discarded if not sold or served. Since food service operators have been
given over ﬁve years to comply with the new temperature requirement and they will be
reminded of this at each inspection until then, it seems very reasonable that they will
comply. The practice of date marking is new to most food service facilities except some
national chains. New to almost all of them is the requirement for marking the end date.
It will take some time for management to train staff to date mark the appropriate food
items with the correct dates. Although these regulations will help prevent the growth of
these psyphrophilic pathogens, perhaps the greater responsibility for control rests on the
food processor where the foods are initially contaminated. The issue of control of

L. monocytogenes is currently being addressed by a multi-faceted action plan designed
collaboratively by the Health and Human Services (HHS) and the USDA to speciﬁcally
reduce the risk of illness and death caused by L .monocytogenes in ready-to-eat foods.
Lowering refrigeration temperatures is only a small part of the plan. The incidence of
illness caused by L. monogdogenes has decreased 29% ﬁ'om 1990 to 1997."3 This
reduction may also be a reﬂection of an intensive public education campaign directed at
those people most at risk, such as pregmmt women.

Another clmnge in the code allows for more time allowed in the cooling of
potentially hazardous foods to refrigeration temperatures. The former code allowed only
four hours to move the food item through the “ er zone” of 140° F. to 45° F. The new
code, based on scientiﬁc research requires that the food item cool ﬁom 140° F. to 70° F.

within 2 hours and then from 70° F. to 41°F. or below within an additional 4 hours. This

25

allows a total of 6 hours of cooling. The important feature, however, is that potentially
hamdous food must move more quickly through the higher end of the “ er zone”
than the lower end. Many outbreaks have been caused by inadequate cooling of foods.
The outbreak (to be described in Chapter 4) that occurred in 1998 in Ingham County was
most likely caused by inadequate cooling of a sauce for a chicken entree (too slowly
cooled). An earlier outbreak of C. perﬁ'ingens in Ingham County in 1990 was caused by
improper cooling (too slowly cooled) of minestrone soup.4|

This new regulation must be put into practice immediately. Proper cooling
practices have been required in former codes. The emphasis on the new code is quick
chilling. The most efﬁcient way to cool bulk foods such as soups and casseroles is in a
blast chiller. These are very expensive and therefore only available to a small number of
food service facilities. Ice baths, ice ﬁlled stirring paddles and cooling in shallow
containers in walk-in coolers are all affordable. These methodslof cooling will be
expected and monitoring of temperatures will also be expected. The person in charge
will need to know correct cooling procedures and implement them. During the plan
review process for new food service operations the need for cooling will be examined and
the necessary equipment and policies will be required to comply with the code. Training
ofthe person in charge and food handlers is necessary in order to convey the importance
of this rule.

Time alone is now an approved method of public health control (as opposed to
time in conjunction with temperature) when the process is done according to code. Four
hours is allowed ﬁ'om the point in time when food is removed from temperature control,

provided a written procedure is available describing how the food is marked and that the

26

food is discarded after 4 hours. For example, cooked rice may be kept at room
temperature for 4 hours if documented. Oriental restaurants will often prefer to keep
cooked rice at room temperature for stir-ﬁ'ying or for sushi. This has been considered a
critical violation in the past. Rice has been implicated in foodborne illness outbreaks
caused by Bacillus cereus due to this common practicez’9 It may appear that this new
code allowance is a step backward in food sanitation. However, it also appears that this
food handling practice has continued even after citation of code violation, due to
language barrier and other cultural differences. If the regulator can educate the operator
in the correct way to use time as a control and insist on the required documentation, it
may result in a compromise that is safe and acceptable to both the operator and the
regulatory agency. This has been done recently in Ingham County with oriental and
middle eastern restaurants. Falafel is kept at room temperature for a two hour period at
hmch time at a local middle eastern restaurant and is discarded at the end of that period.
These arrangements were documented very recently. It will take time, 6-12 months, to
determine if the practice is actually followed, by return visits and inspections of these
establishments.

Required cooking temperatures of some animal foods has changed. Raw shell
eggs that are broken and prepared to a customer’s order must be cooked to 145° F. for 15
seconds. Other eggs such as batched scrambled eggs must be cooked to 155° F. for 15
seconds. During 1993-1997 Salmonella caused 357 (55%) of the 655 reported bacterial
foodborne disease outbreaks with a known etiology (32,000 cases). Approxirmtely half
of those were caused by S. enteritidis, associated with shell eggs. S. enteritidis was the

most ﬁequently reported cause of outbreaks and resulted in more deaths than any other

27

pathogen. Of the 10 persons who died as a result of S. enteritidis, four were residents of
nursing homes. ‘7 An outbreak caused by S. enteritidis in Ingham County in 1997
associated with undercooked eggs resulted in 21 people becoming ill after eating at a
popular family restaurant.“ Illness with s. enteriditis is considered highly underreported,
and is estimated to cause over 1.4 million cases of foodborne ilhless annually during this
time period (1993-1997). Once these new requirements are in place, the number of
illnesses due to Salmonella should be greatly reduced. In fact, a “progress report” ﬁ'om
the CDC on the Healthy People 2000 states that the number of outbreaks of Salmonella
enteritidis has decreased 22 percent from 77 outbreaks in 1989 to 45 outbreaks in 1998.43
The HP2000 goal was 25 outbreaks. At that time 16 states had adopted the new 1999
Food Code and 23 were in the review process prior to adoption. It appears that the
number of outbreaks due to Salmonella will continue to decline as more states adopt the
new code.

Comminuted meats such as ground beef must now be cooked to 155°F. for 15
seconds. E. coli 0157:H7 was ﬁrst identiﬁed as a pathogen inl982 in an outbreak of
bloody diarrhea traced to hamburgers for a fast-food chain. It was subsequently shown to
have a reservoir in healthy cattle. During the years 1993-1997 E. coli spp. caused a
reported 84 outbreaks including 3,260 ill people.17 The estimated number oftotal cases of
E.coli 0157:H7 alone for this time period was 73,480 cases per year. 33 Some of the
E. coli 0157:H7 cases reported from 1993-1997 were caused by unusual foods such as
apple cider in 1993, apple juice in 1996, leaf lettuce in 1995 and 1996, alfalfa sprouts in
1997. 46 Most illness, however, has been associated with eating undercooked,

contaminated ground beef. The CDC Progress Report for Healthy People 2000 found

28

foodborne infections by E. coli 0157:H7 are down 75 percent from 1987 to 1997."3 This
exceeds the HP 2000 goal of a 50% reduction. This reduction can be attributed to the
massive public health education campaign resulting from the earlier outbreaks causing
death in young children. Consumer vigilance and food service retailers conforming to the
advisories recommended by the FDA to thoroughly cook burgers has continued to lower
the incidence. Solidifying this requirement in the code will provide a legal basis to

require food service establishments to recognize this hazard.

Demonstration of Knowledge

The code now requires a “person in charge” to be present in a food service at all
times during operation. This person must be able to demonstrate knowledge of
foodborne disease prevention, application of the Hazard Analysis Critical Control Point
(HACCP) principles and the requirements of the Code.

Consumers may be surprised to know that prior to the adoption of this code a food
service operator was legally allowed to open a business without any knowledge of food
safety. A plan review was conducted in accordance with the current food code, including
physical and structural aspects of the facility such as plumbing, mechanical, ﬁnishes of
ﬂoors, walls and ceilings and equipment. The proposed menu was reviewed to determine
suitability of equipment and storage space requirements for dry goods and refrigerated
foods. Once these plans were approved and the construction was approved at the ﬁnal
inspection, the facility was approved to operate. The burden of educating the food
service operator in food safety issues rested on the sanitarian at the time of the ﬁnal pre-

opening inspection and subsequent inspections when violations were cited.

29

Under the new code the person in charge is required to demonstrate knowledge in
any one of three different ways:
1. Compliance with the 1999 Food Code;
2. By responding correctly to the inspector’s questions as they relate to the speciﬁc
food operation;
3. Becoming certiﬁed as a food protection manager who has shown proﬁciency of
required information through passing a test that is part of an accredited pro gram.

The implied goal for this requirement is, of course, that this knowledge will be
passed down to the food handling employees in order to ensure safe food handling
practices. None of the above measures will guarantee the level of knowledge necessary
to prevent foodborne illness.

Compliance with the code simply means that during a routine unannounced
inspection at some time after the food service has been open and operating, no “critical”
violations are cited. These are the most serious violations most commonly associated
with causing illness. Most routine inspections are conducted during regular hours, but in
between busy meal times, out of courtesy to the food service manager. However, critical
violations are more likely to be seen during lunch or dinner “rushes” when there are
greater opportunities for failure in practicing safe food handling techniques. An example
is the grill cook who is responsrble for placing a raw burger on the grill, cooking it and
then assembling the burger on a bun with condiments. The opportunities for cross-
contamination of raw burger juices onto ready-to-eat food items, undercooking, as well as
poor employee hygiene are greatest when the cook is solely responsible for this process

during busy periods. In addition, if the employees are not adequately trained or

30

supervised, critical violations will continue to occur even if the manager or “person in
charge” is determined to be ﬁllly knowledgeable of the food code. Therefore, the
reliance on an inspection without critical violations is not a good measure of the level of
nmnager food safety knowledge. Further, having a well-trained manager does not ﬁllﬁll
the intent of this code requirement if the employees are not well trained or motivated to
follow code requirements. If the inspections were more comprehensive, including
evaluation of entire food preparation processes before and during meal times, then the
lack of critical violations would be a good indicator that the food safety knowledge of the
manager is being passed on to the employees and incorporated into practice

Similarly, answering questions pertaining to food safety relevant to the operation
may not cover all aspects of the particular food service. There is no standard set of
questions to use when interviewing the person in charge. This process will be largely
subjective, non-uniform and incomplete. The procedures for inspection of food service
establishments are evolving. Sanitarians are encouraged to take a more HACCP type of
approach in their evaluation of food preparation from the time it is delivered to the
facility until it is served. HACCP plans are not required for menu items of restaurants, so
this will be a difﬁcult task. Most local health departments lack the resources to
accomplish this task in every food service establishment to the level that will truly
ascertain the knowledge level of the person in charge.

Certiﬁcation of food service managers through an accredited program is the most
reliable measure of the level of knowledge of the code. The typical accredited program
includes 12-16 hours of training on all aspects of food safety and a standardized test to

pass. Public expectation will be that there is a “certiﬁed manager” on site at all times of

31

operation. Many of the natiorml restaurant chains require certiﬁcation of their
managers. Currently 21 states have laws requiring manager certiﬁcation of some sort,
and more are pending. Although certiﬁcation of managers is a more reliable measure of
manager knowledge, it still does not guarantee that safe food handling practices will be
passed along to employees.

The FDA, USDA and F SIS, co-lead agencies responsible for the Healthy People
2010 goals, recognize the need for thorough training of retail employees in order to
reduce the number of foodborne illness outbreaks. '2 The food service industry has a
large employee population with high rates of turnover. Language and literacy barriers as
well as a general lack of training pose additional challenges. It may be more of a hardship
on the smaller businesses that ﬁnd it difﬁcult to afford the cost of the program and the
time away from the business. It is becoming increasingly important for local health
departments to offer training programs that are affordable to food service operators that
cannot afford training from an independent consultant. In addition to manager
certiﬁcation training, it is also important for local health departments to offer shorter,
simpler comses for food handlers. In some cases interpreters are used as well as written
materials in different languages. These classes can be used as a tool by the food service
managers to incorporate as part of their general training, especially if there is no
corporate training program already in place.

It is unreasonable to expect compliance with the food code or to expect correct
answers to questions about proper food safety procedures if there has not been adequate
training of food service mamgers or employees. Michigan law still allows for the

requirement of food service sanitation certiﬁcation to be used as an enforcement tool in

32

cases of repeated non-compliance with the code. This is not cormnon and many months
of non-compliance with critical code requirements occur before this intervention is put
into place. A combination of all three methods of ascertaining rmnager knowledge
would adequately measure the level of knowledge required. By using all three methods
the intent of the code would more likely be fulﬁlled. Manager knowledge will be tested
in a tangible way through manager certiﬁcation. The other two methods, actual code
compliance and interviewing the manager on speciﬁc practices, will determine if that
knowledge is passed on in practice. Many states have required certiﬁcation and the
general public will expect it as awareness of food safety increases through the public
education eﬂ‘orts catalyzed by the Food Safety Initiative.

Michigan could have required manager certiﬁcation, but fell short of that in the
adoption of Michigan Food Law 2000. The Presidents Food Safety Initiative of 1997
should have provided enough political backing for this requirement. Funds could have
been appropriated to the state and local health departments to provide for this service
over a reasonable period of time. Until required at the state or federal level, this
requirement could be picked up by local governmental bodies. In fact, three counties in
Michigan now require manager certiﬁcation. Hopefully, the remaining counties will

follow suit.

Employee Health controls
The 1999 Food Code requires the person in charge of a food service facility to
require employees to report illness. Speciﬁcally, the employee must report to the

employer any past or present diagnosis of illness with one of the “Big Four” foodborne

33

pathogens: Salmonella T yphi, Shigella spp., Escherichia coli 0157:H7 or Hepatitis A
virus. The person in charge must, in turn, report these to the local health department.
These illnesses are separated out because they are especially virulent and infectious. The
“big four” illnesses have historically been linked to transmission by an ill employee
contaminating food. “Typhoid Mary” was a well known example. In the late 1800’s she
worked as a domestic cook for eight families in New York state. Mary became an
asymptomatic carrier of Salmonella typhi. Twenty-two people in seven of those families
became ill with typhoid fever and one died. Later, she changed her name and became
employed as a cook at a maternity hospital where twenty-ﬁve people became ill and two
died. If she had worked in a modern day ﬁrst-food restaurant the results would have been
even more disastrous.

There are close to 300 organisms associated with foodborne illness. Some of the
more common symptoms of illnesses that can be easily spread by food include: diarrhea,
vomiting, fever, jalmdice, sore throat with fever, discharges from the eyes, nose, and
mouth or infected wounds or boils. The symptoms listed are common to foodborne
illnesses transmitted by the hands or respiratory tract of an infected food handler. The
food service manager must recognize these symptoms and must exclude the ill employee
ﬁ'om working or restrict the employee to non-hazardous duties, depending on the type of
illness and type of population served ( high-risk or general).

The “big four” are already reportable diseases. There is a system already in place
in the more well-equipped and well-staffed local health departments to help exclude
employees with these diseases from work in a food service. In the Ingham County Health

Department, when a positive laboratory result is received by the Communicable Disease

34

Division for pathogens such as Shigella spp., Salmonella, E. coli 0157:H7 or Hepatitis A
virus, the patient is interviewed by a public health nurse. If it is found that the patient
works in a food service establishment, the nurse works with a sanitarian in Environmental
Health to inform the patient and the food service manager that the employee must be
excluded ﬁ'om work until reinstatement is approved by the health department and
physician documentation. Although this approach is effective in some health
departments, there is a possibility of delay that may result in the ill employee exposing
the general public to disease. Some health departments lack the infrastructure necessary
for this cooperative effort. In any case, it makes good sense to educate food service
nnnagers and employees of the importance of reporting these illnesses and the reasons
for excluding the ill employees so the food service manager can more quickly and
effectively prevent exposure of the food service environment to the ill employee. The
manager is the front line of protection and it is their responsibility to know and
understand the causes and prevention of foodborne illness.

This new code requirement will help to decrease the number of outbreaks caused
by the “big four” by catching them earlier than they have been in the past. Shigella is
estimated to be the most common of the four, followed by E. coli 0157:H7, Hepatitis A
and S. Ophi which is very rare. We should see a decline in foodborne illness outbreaks
caused by this pathogen after this code is in full effect. In addition, we should see a
decline in other types of outbreaks primarily caused by contamination of food by ill
employees such as those caused by Norwalk-like viruses, Salmonella, other E. coli species

and Staphylococcus aureus from infected wounds.

35

Controlling Hands as a Vehicle of Contamination

The requirement for hand washing by food handlers is not new. However, there
are more stringent guidelines on the proper procedure in the new code, including a
minimum of 20 seconds of lathered scrubbing prior to rinsing.

One of the most important and controversial additions to the food code is the

prohibition of bare-hand contact with ready-to-eat foods except when washing ﬁ'uits and
vegetables. Poor personal hygiene is the second most ﬁ'equently cited food handling
practice resulting in foodborne illness outbreaks.l "'6’” Pathogens on food handlers’
hands are passed on to the foods they are working with. If this food is ready-to-eat, such
as a sandwich or salad, the food will not be undergoing a subsequent cooking process that
would kill organisms left on the food. Therefore, preventing contact with ready-to-eat
foods should reduce greatly the number of ilhiesses caused by direct contamination of
food. This will be accomplished by providing a barrier between the food and the bare
hands. Examples include the use of utensils, plastic single-use gloves or deli paper.
These are not substitutes to proper hand washing. To be effective, these barriers must be
used correctly. Employees must be trained to wash hands before and after using gloves.
Care must be taken not to contaminate the gloves while in use and they must never be re-
used.

It is clear that consumers have expected this standard of “no bare-hand contact”
for quite some time considering the number of complaints received by local health
departments. Until now, bare-hand contact was allowed in Michigan. This has been the
most controversial of changes to the code in Michigan. Some of the food service industry

has had (and is having) a diﬂicult time conforming to this requirement. Use of utensils

36

and gloves is more awkward than bare-hand use and it will take some time to become
comfortable with these practices. The Michigan Law spells out guidelines that may be
used to develop a written alternative practice and procedure that allows bare-hand under
controlled circumstances. This is a very rigorous set of guidelines requiring the food
service establishment to be in consistent compliance with the critical requirements of the
food code and to conform with approved written HACCP-based operating procedures.
These guidelines were designed to ensure that consumer safety will not be compromised
when bare-hand contact with ready-to-eat foods occm's. Bare-hand contact with ready-to-
eat foods should be the rare exception to the rule.

Shigella spp., Norwalk-like viruses, hepatitis A and Staphylococcus aureus are
some of the organisms that are more commonly transferred to foods by direct handling of
foods. Poor food handler hygiene is the second most common documented cause of
outbreaks. The added protection of gloves or utensils (when used correctly) as barriers
between the contaminated hands of food service employees (with less than perfect
hygienic practices) and foods should result in a decrease in incidence of foodborne

illness.

Consumer Advisories

The third most ﬁ'equently reported cause of foodborne ilhless is inadequate
cooking of foods. The ﬁfth most ﬁ'equent cause is obtaining food from unsafe sources.
The general public has only recently been informed of some of the risks involved in
eating these foods. In 1995 the CDC, the FDA and several state health departments

collaboratively developed a multi-state surveillance study of food-handling, preparation

37

and consumption behaviors associated with foodborne diseases. Data were collected
from questionnaires administered with the 1995 and 1996 Behavioral Risk Factor
Surveillance Systems (BRFSS) in these states. The report indicates that behaviors
associated with foodborne diseases were common. Fifty percent of respondents reported
consumption of undercooked eggs. Twenty percent of respondents reported eating pink
hamburgers and eight percent reported eating raw oysters. 54

Another interesting result of this study was the observation that there was a direct
positive relationship between the education level and the ﬁequency of eating pink
hamburgers or raw oysters. The prevalence of consumption of pink hamburgers
increased with education (less than grade 12: 12.0%, high school graduate: 16.5% and
any college education: 24%) and also increased with yearly salary (less than $15,000:
11.8%, 315,000-834,999: 17.6%, $35,000-$49,999: 22% and greater tlmn or equal to
$50,000: 28.6%). This contrasts with other ﬁndings that persons with higher education
levels are more likely to pursue other health-promoting behaviors such as seat belt use
and abstinence ﬁ'om smoking that persons with lower education levels. This difference
suggests that highly educated persons are still not being educated on the associated risks
or they are choosing to ignore the hazards associated with these behaviors. In any case, it
is important that public education regarding these hazards continue to be emphasized.

The new food code now requires food service establishments to include a
“consumer advisory” in the menu (or other easily visible place) for ready-to-eat animal
foods that they choose to serve raw, undercooked or not otherwise processed to eliminate
disease causing organisms and poses a signiﬁcant risk. It is a customer’s right to know if

foods served at a restaurant may not be safe to eat. A food-safety survey of 2,197

38

homemakers found a high reliance on government inspection for the prevention of
bacterial contamination of raw meat and poultry, underrating their responsrhility for safe
food-handling and preparation practices.l Similarly, restaurant patrons believe that since
restaurants are inspected, the food served may be considered safe to eat. The food code
stops short of prohibiting raw or undercooked animal foods from being served, except for
those establishments serving high-risk populations. The general population is not ready
to give up some of these undercooked food items. Therefore, in order to leave the choice
up to the consumer and at the same time provide them with accurate information, the
requirement for an advisory makes sense.

An example of an advisory is the placement of an asterisk (*) next to the menu
item “hamburger” with a footnote reading “Can be cooked rare or to order. Consuming
raw or undercooked beef may increase your risk of foodborne illness, especially if you
have certain medical conditions.” The more the public sees this reminder in the menus of
their favorite restaurants, the more educated they will become. The result will be a
continued decrease in the incidence of foodborne illness caused by undercooked animal

foods.

Conclusion

The science-based knowledge gained by the interpretation of data ﬁ'om years
worth of surveillance for foodborne illness outbreaks is already proving to affect a
decline in certain types of foodborne illnesses. This cycle of surveillance, epidemiologic
investigation and the institution of prevention measures must continue. The continually

improving surveillance systems described earlier will more accurately measure the

39

burden of foodborne illness in the United States. In addition, the epidemiology of
foodborne illness outbreaks is changing. New pathogens are emerging along with new
vehicles of transmission. Surveillance has never been more important in determining
how to control foodborne illness. The evolution of the FDA Food Code is the result of
diligent analysis made possible by the cooperation and collaboration of local health
departments’ investigations and reporting of outbreaks.

The adoption of this code is an outstanding improvement item the former code
used by the State of Michigan. The ﬁve public health interventions introduced in the new
food code clearly address the most reported causes of foodborne illness outbreaks.
Excluding ill employees from food service operations and limiting bare-lmnd contact
with foods will prevent spread of foodborne illness organisms to foods eaten by the
public. Applying the new temperature requirements will kill pathogenic organisms in
raw animal foods and will prevent the growth of incidental contaminants that would
result in illness. Consumer advisories will educate the general public to make better
choices when considering eating raw or undercooked animal foods.

None of the above objectives will be realized without adequate training of food
service managers. The requirement in the new food code for demonstration of
knowledge is an important addition to the code. However, the code does not go far
enough to meet this objective. As discussed earlier, the method for assessing knowledge
of a “person in charge” is too subjective, non-uniform and incomplete. The ﬁrst step in
ensuring food service sanitation practices is to be sure the manager is well educated on
these matters. If the manager is not knowledgeable of the code requirements, then there

cannot be an expectation that the employees will follow through with the practices

40

required. In order to protect the public health we should not be relying on voluntary
enrollment in an accredited program run by private industry. The number of food safety
manager certiﬁcation training programs is rapidly growing. Many food service chains
and individual restaurants are acquiring training ﬁ'om these resources. More and more
local health departments are offering the same training at a lower cost. In order to offer
this service, local health departments must have adequate funding to provide and train
staﬂ‘ to fulﬁll this service. The food service support industries are also responding to the
need. Food service equipment and food suppliers are now offering certiﬁed training
programs to their clients. Some of the larger restaurant corporations hire their own food
service safety specialists to train their own managers. The National Restaurant
Association and afﬁliated state chapters support and publish the “Serv Safe” Training
Program which has been used by all the above industries.

The Michigan Food Law has fallen short of requiring mandatory food safety
training. However, several states are in similar situations. They have adopted the new
FDA code but do not require above that, mandatory training. Instead, localities, either
cities or counties have adopted ordinances requiring manager certiﬁcation. This is still a
possibility in Michigan. If it appears that food service establishments are not voluntarily
becoming “certiﬁed” and there is still a lack of conformity with the code, Ingham County

may attempt to pass an ordinance requiring certiﬁcation of managers.

41

CHAPTER 3

DEVELOPMENT OF THE ICHD F OODBORNE ILLNESS INVESTIGATION
PROTOCOL

The Role of ICHD in FoodBorne Illness Investigation

Foodborne illness outbreaks have been investigated by Ingham County Health
Department (ICHD) for many years in the compliance of the Michigan Public Health
Code, Act 368, PA. of 1978. Two divisions have been involved in these investigations.
The Communicable Disease Division (CD) of Ingham County Health Department is
responsible for the surveillance of Infectious Disease within Ingham County. Whenever a
food service establishment is implicated in any of their cases interviewed, the
Environmental Health Division (El-I) is consulted. The food service establishment may
be either a place of employment for the case or may be a suspected source of the
infection. In addition, if an illness or outbreak is suspected to be caused ﬁom food and
no food service establishment is implicated, the Environmental Health Division will also
be consulted.

For many years these two divisions of Ingham County Health Department have
worked together to investigate foodborne illness outbreaks. The expertise possessed by
each division has been invaluable in the professional teamwork approach used to
investigate outbreaks. In recent years the knowledge base in these two divisions has
overlapped somewhat due to the cooperative cross-training between the two divisions.
This has provided a greater assurance that foodborne outbreaks are investigated as

thoroughly and professionally as possible.

42

The Public Health Nurses in CD routinely monitor infectious disease. They have a
working knowledge of common communicable disease, including diseases spread by
foods, the clinical presentation and the incubation periods involved. They are very
knowledgeable and experienced in the task of interviewing clients. Training is necessary
in order to interview clients with or without a questionnaire to provide uniform and
standard results. Requisition forms for laboratory testing of patient specimens and
foods are completed by CD prior to collection of specimens and subsequent delivery to
the Michigan Department of Community Health Laboratory.

The Registered Sanitarians in EH have received specialized training in the area of
food service sanitation. Knowledge of the microbiology of foodborne pathogens is
necessary in educating food service managers in area of food safety. There are nearly
300 known foodborne pathogens and new pathogens are emerging constantly. The
science involved in the control of these pathogens is also changing. The Sanitarian must
be educated constantly to keep up. The Sanitarians spend many hours each week
inspecting food services. They are very knowledgeable of situations and conditions that
would cause foodborne illness, such as poor employee hygiene, improper cooling and
heating of potentially hazardous foods, cold and hot-holding practices of potentially
hazardous foods, cross-contamination, etc. This knowledge is necessary in the
investigation of outbreaks in order to determine the vehicle and improper food handling
techniques responsible for the outbreak. The rapport developed between the Sanitarian
and the foodservice manager is also important in order to effectively educate the
nmnager in ﬁrtm‘e prevention of foodborne illness. Follow-up by the Sanitarian is

necessary in order to help ensure that changes are made in faulty food handling.

43

Collaboration on the Writing of a Protocol

Early in 1998 the Divisions of Communicable Disease and Environmental Health
were charged by the Health Ofﬁcer to develop a written protocol for investigation of
foodborne outbreaks. Prior to that date the association of the two divisions was more
informal causing some confusion at times over which division would do certain tasks.

In an effort to solve this problem and to work more eﬂiciently during the ﬂurry of an
outbreak investigation, several members of each division met on 1-28-1998 including the
supervisors of each division to discuss the development of a protocol for foodborne
illness investigation. Two Public Health Nurses and two Registered Sanitarians
(including the author of this paper) were chosen to work on the committee that would
write the protocol. The two Sanitarians were (not coincidentally) working on their
Masters Degrees in Epidemiology at Michigan State University at the time. Their
experience and knowledge gained in the area of epidemiological analysis were very
helpﬁrl in developing a protocol.

Over the next several months this committee met to discuss all aspects of the
protocol after dividing up the responsibilities for each section. Periodically, the
document was circulated among the two staffs for cormnents. Although this committee’s
ﬁnal draft was completed late in 1998, the protocol remains a “ 'ving” document that has
already been modiﬁed since that date.” Revised MPRs (State of Michigan minimum
program requirements)37 will require some modiﬁcation and it will continue to be
updated as the science involved is also changing.

The format used for the document was the standard format used for many of the

protocols within the ICHD consisting of the “ se”, “Policy Statement” and

“Standard” written prior to the body of the document. The body of the document is an
outline of the investigation guidelines. An appendix includes a telephone list, forms

used, supporting documents and a glossary.

Purpose

In this section background is provided on the legal basis for intervention by the
ICHD as well as the objectives in the investigation of foodborne illness outbreaks. The
current Public Health Code at that time was the Michigan Public Health Code, Act 368,
PA. of 1978, as amended, Sections 12912 and 2422 which states that the local health
department is responsible for investigating all suspected and conﬁrmed foodborne related
outbreaks involving Ingham County residents or facilities. The “Purpose” also stresses
the importance of initiating the investigation as soon as possible in order to identify the
source of a potentially life-threatening situation. Prompt action will allow appropriate

interventions to be put into place in order to protect the public health.

Policy Statement
This section is standard in all ICHD protocol documents. It simply states that all
Ingham County Health Department employees will follow the guidelines for suspected

and conﬁrmed foodborne outbreaks.

Standard
This statement is also standard in all ICHD protocol documents. It is a
commitment to “achieve and maintain a performance level of employees at ICHD

through a systematic process which protects the health and well being of the residents of

45

Ingham County”. This document will provide a means to investigate outbreaks in a

consistent and professional manner.

Guidelines for Outbreak Investigation
53%

Section 1 covers the initial surveillance stage of foodborne illness or outbreak
situations. This is a very important stage. In the past, some outbreaks may have been
undiscovered due to a lack of attention at this point. When a call is received ﬁom the
public reporting an illness suspected of being a foodborne illness it could be directed to
either the oﬂice of Environmental Health or the ofﬁce of Communicable Disease. There
must be a system in place that assures that EH is notiﬁed immediately if a commercial
food service is implicated. The clerical staff of both divisions must have clear guidelines
on referring these calls to the appropriate staff members in order for to effectively
determine if there is an outbreak. It is also important that the pertinent information be
recorded on the correct form. Section I is divided into two parts, clearly deﬁning the
responsibilities of each division when a report is called in.

In Environmental Health , these calls are usually from a member of the public
who is reporting their own illness suspected to have been caused by eating food at a
particular food service establishment. When such a call is received by a clerk, it is
directed to the Sanitarian who is responsible for the geographic area the food service is
in. If the Sanitarian is unavailable and the call is forwarded to an answering machine, it
maynotbehandledinatirnelyfashion. InitiallyitwaswrittenthattheareaSanitarian

would be paged, but after input ﬁ'om the Supervisor it was decided that the clerk would

collect essential information on a short form, “Clerical Ilhress Reporting Form” and give
this form to the Sanitarian as well as a copy to the Supervisor. The Supervisor would
then be able to re-delegate or take other appropriate action if necessary in the absence of
the area Sanitarian.

The next step is collecting the information necessary ﬁom the complainant in
order to determine if an outbreak is occurring. On the day the complaint is received, the
“Foodborne Illness Form” is completed by the Sanitarian. In addition to demographic
information the complaint is summarized. The complainant is asked several questions
pertaining to the illness, such as: how many were ill, if they were seen by a physician,
who the physician was, other group meals eaten within 72 hours, onset date and time,
date and time the meal was eaten, duration of illness, symptoms, other people who ate
with them and then a complete three day food history ﬁ'om two days before onset through
the day of onset of illness.

The Sanitarian would then call the food service establishment implicated to
discuss the complaint and ask if they had received any other complaints. The manager
would be asked to contact us immediately if they received any similar complaints.

The Sanitarian would then enter information about the complaint into an “Alert
Log” posted outside the Supervisor’s ofﬁce. This is the master compilation of all
suspected foodborne illness complaints received in Ingham County and serves as the
primary record of foodborne illness surveillance at ICHD. It is a line listing of
abbreviated information including Sanitarian’s initials, complaint number, date received,
onset date, number ill, predominant symptoms, alleged suspect food and eating places

other than home in the three days prior to illness. When a complaint is recorded,

47

previous entries are reviewed for relevance to the current case. In addition, a short
summary is written on a bulletin board seen daily by all Sanitarians that will alert all area
Sanitarians of the report.

The ﬁle for the establishment implicated may be reviewed for more information
on compliance, past complaints, manager’s name and phone number, etc.
The original Foodborne Illness Form is placed in the food service establishment ﬁle and a
copy is sent to the Food Sanitation Section of Michigan Department of Agriculture.

Occasionally a complaint about a possible foodborne illness ﬁom a commercial
food service will be received by the CD oﬂice. The clerk will transfer the call to EH
Division. Occasionally a possrhle foodborne illness will be reported directly to a CD
nurse. The nurse will ﬁll out the Foodborne Illness Form and provide a copy to the area
Sanitarian in EH. The nurse will then enter the appropriate information into the “Alert
Log”. Sometimes foodborne illness is recognized through the passive surveillance
system. During the course of a routine enteric illness case investigation by a CD nurse,
the patronage of a commercial food service is mentioned during the typical incubation
period for that disease. In this case also, the nurse will ﬁll out a Foodborne Illness Form
and provide a copy to the area Sanitarian and enter information into the “Alert Log”.

If all the above steps are followed, it would be highly unusual for a foodborne
illness to be missed or for an outbreak to be unrecognized due to lack of proper
notiﬁcation of the Sanitarian. Proper recording and notiﬁcation of these calls will result

in timely decisions on whether an outbreak is occurring or has occurred.

48

Section H

Section H covers the deﬁnition of an outbreak. Further investigation is not
initiated unless the illness reported is determined to be an outbreak. An outbreak is
oﬁcially determined when two or more unrelated persons have the same condition, or
similar symptoms with a time, place and person association between them. The deﬁnition
of the term “unrelated” was debated. After consultation with MDA it was concluded that
persons were considered related if they lived in the same household. For example, if two
co-workers both became ill after eating at the same restaurant together, they would be
considered unrelated. Likewise, if two fratemity brothers living in the same ﬁaternity
house both become ill after sharing a meal at a restaurant, they would be considered
related. Along with the deﬁnition given in IAMFES Procedures to Investigatg
Foodborne Illgss exceptions include single cases of suspected botulisrn, mushroom
poisoning, ciguatera, paralytic shellﬁsh poisoning, typhoid fever, other rare diseases, or a
case of a disease that can be deﬁnitely related to ingestion of a food. These exceptions
are considered incidents of foodborne illness that warrant immediate further
investigation.” Once it is determined that a foodborne outbreak has occurred (or an

incident of one of the exceptions listed), outbreak investigation activities are initiated.

Section HI

Section IH delineates the steps involved in initiating investigations. First, the
supervisors in EH and CD must be informed that an outbreak has been identiﬁed. If one
of them are not available, the next in charge should be notiﬁed. These supervisors and

Medical Director (if available) will confer about the advisability of initiating a full

49

investigation. Once this is approved the two divisions will work together to conduct the
investigation.

If there is a suspected food item it should be set aside for collection as a specimen
at a later time. The manager of the food service establishment will be directed to mark
the food item and keep it refrigerated until it is picked up for delivery to the laboratory.
If there are any other control measures that could be taken to halt an ongoing outbreak
they should also be done. For instance, if complaints are continuing to be reported of
illness after eating at a particular food service, the food service may need to be closed
until the cause has been identiﬁed.

Other agencies must be notiﬁed at this time. Michigan Department of
Community Health and Michigan Department of Agriculture Food Sanitation Division
work together in state-wide surveillance of foodborne illness and may be able to track
clusters of outbreaks not seen on a local level. If outbreak related specimens, either stool
samples or food samples are available, MDCH Division of Laboratory Services must be
contacted for approval prior to testing. A phone list was compiled and has been attached

as an appendix for easy reference.

Section IV

Section IV outlines the investigation process. Once an outbreak has been
identiﬁed, a meeting is held including administrative staff, EH and CD personnel. It is at
this meting that the following assignments are made:

Staff members from each division will be designated to work on the investigation

to work as a team. Lead investigators will be assigned from each division. These lead

50

persons will be responsible for assigning tasks to other team members throughout the
investigation as they become necessary in consultation with the supervisors. They will
also jointly complete the summary report at the end of the investigation.

A preliminary case deﬁnition will be determined at the initial meeting. The most
pronounced symptom (usually diarrhea or vomiting) should be included along with a
reasormble incubation period from the time of common exposure.27 For example, a case
of gastroenteritis may be deﬁned as a person who was at risk or exposed to the implicated
food or meal and developed diarrhea within a speciﬁed period of time. The case
deﬁnition will become more well-deﬁned as the investigation proceeds.

An initial hypothesis is formed based on preliminary information. Referring to a
manual such as IAMFES or Benenson’s Control of Communicable Diseases, onset times
of the predominant symptoms as well as the symptoms themselves will provide clues to
narrow down the possibilities of organisms responsible for the outbreak. As much as
possible the hypothesis should explain the most likely type of illness, the most likely food
involved, where and by what manner the food may have become contaminated. The
hypothesis will become more well-deﬁned as the investigation continues. Before the
initial meeting is over, assignments must be made for both EH and CD personnel. The
clerical staff must be informed of the outbreak and directed to forward all calls that come
in regarding the outbreak to a designated staff member. These calls may be additional
cases reporting illness or persons concerned about potential exposure or they may be calls
ﬁom the media. Plans must be made to handle the different type of calls that may come

in.

5]

The CD oﬂice will be responsible for the preparation of stool specimen
containers so staff ﬁ'om CD must be assigned to this task. Procedures for assembling
these client mailable specimen kits can be found in the appendix. Included in this kit will
be instructions for the client on how to collect the specimen and mail it in. Staff
members (usually including clerical staﬂ) will be assigned to assemble these kits and
various other staff members (ﬁ'om either division) will be assigned to deliver them to
clients. In ICHD the EH division employs more ﬁeld staff than CD so EH usually
delivers specimen containers. MDCH laboratory must approve testing of the specimens
prior to delivery to the lab. The EH oﬂice will be responsible for obtaining and
submitting food specimens so this task must be delegated. Directions for these
procedures are also included in the appendix. The MDCH laboratory will not accept food
samples unless prior arrangements have been made.

A menu or list of all food and beverage items served at the event must be
obtained. This list must be as complete as possrhle. It is very difﬁcult to add items to a
questionnaire at a later time. This task could be done during a site visit. However, it is
often more efﬁcient to get a list started by phone or fax by staﬂ‘ members working on the
questionnaire. The ﬁeld Sanitarian or Nurse could then conﬁrm the menu when they are
on site at the food service establishment.

Astaﬂ'membermustbeassignedto obtainalistofallpersons exposedto the
suspect meal must be obtained. Attendance lists, reservation lists, credit card receipts or
a roster of usual clientele may be used to help make the list as complete as possible.

Staff members must be assigned to develop a questionnaire, conduct interviews

and enter data onto Epi-Info statistical program for analysis. '0 A sample of the

52

questionnaire is in the appendix of the protocol. It should include demographic
information, the health of the individual, foods eaten, date and time of meal and their
willingness to submit a stool specimen if asked. The questionnaire may need to be
changed and updated as new information is obtained or if additional analysis is necessary.
Both ill and well clients will be interviewed using this questionnaire. Interviewers will

be directed by a lead person on how to conduct the interview in a consistent and non-
biased manner. Clients may be re—interviewed as warranted by progress and resolution of
the outbreak.

As information on individual incubation periods become available, an epidemic
curve must be plotted. One of the team members working with this information should
be assigned the “epi curve”. This graph will have as its X-axis the time of onset in hours
or days and will have the number of ill people as its Y-axis.

F ollow-up meetings will be scheduled for team members in order to update the
whole team on the progress made on the investigation. The preliminary case deﬁnition
rmy be revised to more accurately reﬂect the situation. The study design will be
determined based on the information available and the results of case-ﬁnding activities.
Control measures may be re-evaluated and changed if necessary. The team will review
the information received from the environmental inspection (outlined below) and results

ﬁom the Epi-Info analysis and make any changes necessary to the hypothesis.

Section V

Section V describes the environmental inspection or “site visit”. This was

included as a separate section because it is usually done at the same time section IV is

53

taking place at the ofﬁce. It is considered an important part of the overall investigation
because this physical investigation will usually provide clues to the mystery of how the
foodborne outbreak began. It is preferable that the area Sanitarian perform this
inspection along with a nurse from CD. The Sanitarian has a pre-established rapport with
the manager which is helpful in this stressful situation. At Ingham County our policy has
always been to have more of an educational relationship with food service managers than
a strictly regulatory one. In addition to having a second set of “eyes”, the nurse’s
presence is helpful in obtaining inforrmtion on sick employees and answering questions
they are more experienced with answering.

Guidelines for the Environmental Investigation can be found in the appendix of
the protocol. The manager should already have been called to alert him or her of the
outbreak. If the outbreak appears to be limited and no more cases are calling in, it is
actually preferable for the food service to stay open. Routine operations can then be
observed and more clues may be found regarding the source of the outbreak.

The manager should be asked again if any reports of illness were received by the
food service establishment. If so, names and phone numbers must be obtained. They
should also be asked if any customers were ill in the restaurant around the time of the
alleged meal or if any employees were ill during that time. Employee work attendance
records should be checked by the manager and any employees reporting ill during the
week prior to and the week of the suspect meal should be reported to the Sanitarian.
These employees will be interviewed to determine if their illness meets the case

deﬁnition or if they may be a possrhle source.

54

Speciﬁc information about the meal should be obtained, including number of

meals served of the suspected food. If any of the food is left from the suspect meal, the

food items will be tagged or labeled to identify it as a specimen and must remain

reﬁigerated. The manager must be notiﬁed that this is a hold order for possrhle delivery

to the MDCH laboratory. A written notice must be left with the owner or manager.

The inspection of the facility must be tailored to the type of food service and the

foods initially implicated in the outbreak. There are some elements of investigation that

should always be included:

1.

Check personal hygiene of food handlers. Observe hand washing practices, if
they are wearing bandages or have any open sores, etc.
Check the producer and supplier of the suspect food product. Record any

information that may be important if a trace back is required.

. Check food handling and preparation practices of the suspect food product from

delivery to actual service to the customer.

Check time/temperature relationships of the following: thawing, preparation,
cooking, display, storage, cooling, reheating. Pay close attention to foods
implicated in outbreak.

Check if there were any problems with reﬁigeration or power outages at the time
of the complaint.

Check for food additives used.

Check for contamination by vermin such as mice or cockroaches.

55

8. If the facility has an on-site sewage disposal system or well, these must be
evaluated. A water sample must be taken for bacterial and chemical analysis if
there is a well
After the inspection is completed, the manager must be provided with a complete

report documenting the investigation. If necessary, the ﬁeld Sanitarian or Nurse may call
a lead person back at the ofﬁce to report any pertinent ﬁndings or information that would

be useful immediately instead of waiting to return to the oﬂice.

W
Section VI states that active surveillance including area ambulatory health care
centers and emergency rooms may be necessary depending on the scope of the outbreak

at the discretion of the Medical Director.

Section VII

Section VII describes the activities involved with the sunnnary report of the
investigation. The format to follow for the ﬁnal report is included in the appendix of the
protocol. The lead persons of the investigation from both CD and EH will collaborate in
the writing of this report. The scientiﬁc format includes four sections: Introduction,
Methods, Results and Discussion. The Department of Agriculture has requested this '
speciﬁc format and would like it to be approximately 2-3 pages of narrative and 2-3
tables or graphs.

The Introduction will give a brief description of the outbreak. It will explain why

the investigation was done and who was involved in the investigation.

56

The Methods section will be of more detail and will include epidemiological
methods and ﬁeld investigation methods. In this version of the protocol the laboratory
methods are included in the ﬁeld methods. Under Epidemiologic Methods the following
will be summarized: Case deﬁnition, Study Design, Case and Control selection
techniques, Questionnaire procedures, Interview technique and analysis methods. Under
Field Investigation Methods the following will be summarized: Food sampling
technique, Client specimen collection technique, Laboratory methods used, Scope of food
service inspection and employee interview technique.

The Results section will include only the following pertinent data: Graphs and
tables including the epidemic curve graph, an attack rate table, a food-speciﬁc attack rate
table and/or any other important visual aid. It should also include a description of
ﬁndings from interviews, the ﬁeld investigation, laboratory results and the data analysis.

The Discussion will explain the data, draw conclusions and make
recommendations for prevention of foodborne illness. Rarely do all the lab results and
ﬁeld investigation results positively conﬁrm a causative link between a food item and
illness. Conclusions must be made with the inforrmtion available.

After the draft copy is completed, corrections are made and the draft is circulated
to the other team members for review. The comments are collect and ﬁnal clmnges
made. The Final report is than ﬁled in the food service establishment’s ofﬁcial ﬁle in the
EH Division. All conﬁdential interview information and laboratory reports will be ﬁled
in the CD Division. Copies are also distributed to the following people: Medical
Director of ICHD, Health Ofﬁcer of ICHD, EH and CD supervisors, Michigan

Department of Agriculture and the Michigan Department of Community Health.

57

CHAPTER 4

AN OUTBREAK OF F OODBORNE ILLNESS

Introduction

The Ingham County Health Department (ICHD) was notiﬁed on May 5, 1998 by
Michigan State University (MSU) Department of Public Safety (DPS), Safety and Public
Health Division of a potential foodborne illness outbreak. A woman who worked at an
MSU dormitory cafeteria reported directly to the MSU Registered Sanitarian at DPS that
her boyfriend and her brother became ill after eating a meal for a banquet at the Kellogg
Center on May 3, 1998. Supervisory personnel for the MSU dormitory cafeterias have
beneﬁted greatly from the good rapport with the Department of Public Safety Sanitarians.
In addition to providing a regulatory oversight service the DPS sanitarian regularly offers
training sessions for food service personnel. The cafeteria employee who reported the
illnesses therefore knew to report the illnesses to DPS. Kellogg Center was contacted by
the sanitarian to request booking information for banquets held that day. A banquet meal
for 190 of MSU Minority Aﬂ'airs staff and students was served at Kellogg Center on
May 3, 1998 at 4:30 pm. When the Oﬂice of Minority Affairs was contacted to obtain
the guest list, the sanitarian was informed of an employee of the department had also
become ill after attending the banquet. The initial report to Ingham County Health
Department included these three unrelated persons ill with enteric symptoms who ate the
banquet meal. The deﬁnition of an outbreak was met (Two or more unrelated persons

having the same clinical features as well as a time, place or person association ).

58

Supervisors of the Environmental Health Division and the Communicable Disease
Division were notiﬁed. The investigation process was initiated. Several meetings were
held including members of all three agencies in a teamwork approach to accomplish the

tasks described below using professional, efﬁcient and well-organized methods.

Methods
Environmental Investiggtion Methods

A joint environmental inspection was performed by staff ﬁ'om MSU Public Safety
and ICHD Environmental Health Division in the afternoon of May 5, 1998. The executive
chef of Kellogg Center was consulted. The menu and recipes for the banquet meal were
provided. Chicken was the main entrée for the banquet. The chef indicated that two
servings of the chicken AuGratin were leftover and available. These were marked and held
under reﬁigeration until they could be picked up for delivery to the Michigan Department of
Community Health (MDCH) laboratory for testing.

The environmental investigation was performed by two Registered Sanitarians.
Accompanying them were the executive chef and sous chef of Kellogg Center. General
operational procedures were reviewed with the chef, including hot and cold holding
temperatures, food handling and storing practices. Employee hygiene practices were
reviewed including proper hand washing. The speciﬁc food preparation procedures used for
this pmicular banquet meal were studied. In addition, a list of employees who were absent
from work due to illness within the previous two weeks was obtained from the executive

chef.

59

Epidemiological Methods
Case Deﬁnition

During the ﬁrst outbreak meeting on May 6, a “working” case deﬁnition was
formulated. A case was any person who consumed the meal served at the Kellogg Center
Banquet for MSU Minority Affairs on May 3, 1998 who had onset of abdominal
cramping and diarrheal illness within 36 hours of the suspect meal or laboratory
conﬁrmation of an enteric condition within 36 hours of the meal.

I_r_1_terview Process

The MSU Minority Affairs Ofﬁce was contacted by MSU DPS and a guest list
was provided for the May 3 banquet and faxed over to ICHD Communicable Disease
Division. A questionnaire was developed by members of both ICHD divisions based on
the menu provided by Kellogg Center. It was determined that the team would attempt to
interview all of those people attending the banquet in order to ﬁnd both ill and well
people (cases and controls). Telephone numbers for the attendees were provided by the
MSU DPS and included local and home town (when listed) numbers. The interview
process was delegated to four public health nurses from the Communicable Disease
Division and one registered Sanitarian who worked as a team to contact and interview all
attendees of the banquet. The interviewers were provided with a standard and uniform
methodology for conducting the interviews. The data obtained from the interviews was
designed to be entered into the Epi-Info (version 6) statistical analysis program from

Centers for Disease Control.

60

Employees who had been ill were interviewed in order to ascertain whether or not
they had any enteric illness during the two weeks prior to the outbreak. All persons
interviewed were asked if they would be willing to submit stool specimens.

By the fourth day of the investigation a total of 69 of the 190 attendees of the
banquet were interviewed. Local campus telephone numbers as well as home town
numbers were used. Attempts to reach the remainder of the students failed. No further
attempts were made once the team determined that we had enough ill and well people to
continue with the investigation.

A_n_a_lysis Methods

Epi-Info (6) software was utilized to calculate ﬁequencies, attack rate, chi square
and statistical signiﬁcance of results for a case-control study. Generally, for foodborne
illness outbreaks a case-control approach is used when all of those at risk cannot be
identiﬁed or only a pr0portion of ill persons and well persons can be interviewed about
their exposures. A cohort study would be done if the entire group at the event is known
and interviewed about illness and exposure.27 Only 69 out of the 190 attendees could be
reached for an interview, so the case-control study was chosen. All attendees interviewed
that were well became the controls.

Laboratory Investigation Methods

Stool Specimen Collection and Testm' g

After studying the onset times as well as the symptoms of the cases, it was
decided that testing for Clostridium perfringens and Bacillus cereus be conducted along
with the standard enteric pathogen testing so the necessary forms were completed for this

special request. Patient mailable specimen kits were utilized. Instructions for collection

6]

were provided in the kits along with the laboratory request forms. Ten cases had agreed
to submitting stool specimens by May 6. MSU DPS staff delivered specimen kits to
these people during the afternoon of May 6, 1998. They would then mail them to MDCH
laboratory for analysis. None of the ill people had required medical attention and no
specimens were obtained from them prior to this point. Standard plate counts were
conducted for C. perfringens, B. cereus and the standard enteric panel.”39

Food SpECimen Collection a_nd Testing

The food samples were also delivered by MSU staff to MDCH laboratory for
analysis on May 6. Testing for B. cereus and C. perﬁingens was requested in addition to
the standard enteric testing. For C. perﬁ'ingens a standard plate count on tryptose-
sulﬁte-cycloserine (TSC) agar plates is conducted.39 An isolation of 105/gm is required
for conﬁrmation.” In addition, a test to detect C. pery‘ringens enterotoxin is perforrned.’0
This is a reversed passive latex agglutination test. This technique enables soluble antigen
such as the diarrhea] enterotoxin to be detected in an agglutination assay. In a standard
agglutination assay, soluble antibody reacts with particulate antigen such as bacterial
cells.

For B. cereus a standard plate count on mannitol—egg yolk-polymyxin (MYP) agar
plates is conducted.38 The same number of cells (105/gm) are required for conﬁrmation
of B. cereus.” In addition, a test to detect the B. cereus enterotoxin is done. This is also

a reversed passive latex agglutination test.49

62

Results
Environmental Investigation

General Practices

The ﬁeld inspection on May 5 performed by two Sanitarians (one from ICHD,
one from MSU) took approximately 1‘/2 hours. Both the Executive Chef and the Sous
Chef attended.

Temperatures of reﬁigeration units appeared to be acceptable with temperatures
reading 38-44° F. The baine-marie for hot food holding was set at 190° F. and the foods
held in the unit were above 140° F.

Storage practices of cold foods and temperatures were monitored. Individual
portions of cooked food items placed in single layers on sheet trays in walk-in coolers
was noted. This is an acceptable cooling practice.

The potential for cross contamination was noted with raw beef stored below raw
chicken and raw sausage patties stored on top of cooked potatoes. Open containers of
sheet trays of chicken breast were noted with parchment paper or toweling on top of
them.

Inadequate cooling of food products prior to cold storage in a large container was
observed A ﬁve gallon container of chicken minestrone soup was marked 5-5-98 as a
preparation date with no time given. The temperature of the soup was ﬁom 50-60° F.
The chef indicated that ice baths are commonly used to cool soups but no cooling wands
were used (ice ﬁlled plastic stirring wands). He estimated that the soup was prepared

late that morning.

63

Proper hand washing was found to be an issue. At least two sinks were blocked
by carts, empty boxes or clean dish racks. Hand wash sinks were supplied with paper
towel and liquid hand soap. The use of the hand wash sinks was questionable. No food
preparation employee was observed washing their hands during the inspection. One of
the sanitarians observed employees rinsing their hands at food preparation sinks several
times. He also observed am employee lick his ﬁngers and use a common towel to wipe
his face, then lay the towel back down on the preparation area at the grill station. The
employee failed to wash his hands prior to continuing food preparation. Plastic gloves
are used by food preparation employees regularly. However, some bare-hand contact of
food was noted. Strict personal hygiene was discussed with the chefs, including the
importance of thorough hand washing.

Practices Smelﬁc to the Outbreak

There was a set menu for the banquet. The choices included either a Chicken
AuGratin entree or a Vegetarian Strudel entree. The recipes for the meal and speciﬁc
procedures for the Chicken AuGratin were scrutinized. Chicken breasts were seasoned
and sauteed lightly on both sides and then placed on sheet trays to cool in the walk-in
cooler the day before serving. The vegetable mixture of mushrooms, leeks, red peppers
and green peppers were cut up the day prior to serving. The parmesan cream sauce
consisting of whipping cream, white wine, parmesan cheese, bay leaf, salt, pepper, roux,
cayenne pepper and chopped ﬁesh basil was prepared the day prior to service as well.
The four gallon batch was then cooled in the walk-in cooler. The chef indicated that it
should have been quickly cooled in a water bath, but no temperature record was kept. He

could not be sure that the sauce was properly cooled. On the day of service the chicken

breasts were taken out of the cooler and baked for approximately 25 minutes at 325° F.
until the internal temperature reached 165° F. The vegetables were sauteed and placed on

the cooked chicken, topped with warmed parmesan cream sauce, parmesan cheese/bread

mix and brieﬂy broiled just before serving.

Epidemiological Results
Of the 69 attendees that were contacted, 35 met the case deﬁnition. They had all

attended the banquet on May 3, 1998 and had onset of cramping and/or diarrhea within
36 hours of the meal. The remaining attendees interviewed who were not ill became the
controls for the study. The breakdown of symptoms was as follows: Diarrhea: 91.7%,
Crarnping: 63.9%, Vomiting: 5.6%, Fever: 2.8%. The median onset time was 10 hours

with a range of 2-24 hours. (See Figure 6)

 

Kellogg Center Outbreak - May 1998
Epidemic Curve

see
GNhOQ

 

0 2 4 6 8 101214161820222426
Onset Time (hours after meal)

 

 

 

Figure 7. Epidemic Curve

65

Table 4. Summary of Food Specific Attack Rates, Odds Ratios, x2 and p-values
KelloggCenter Outbreak — May 1998

 

 

Ate Did Not Eat
Food ILL Total AR ILL Total AR OR 95%CL x2 p
% 94
Chicken 35 64 55 0 5 0 5.6 .025
Veg. Strud 3 7 43 32 62 52 0.7 .11,4.2 0.19 .483
Potatoes 33 63 52 2 6 33 2.2 31,19 0.80 .323
Gr. Beans 27 50 54 8 19 42 1.6 49.5.4 0.78 .377
Rolls 28 57 49 7 12 58 0.69 .1629 0.34 .562
Red Pepper 3 9 33 31 59 52 0.45 .0824 1.15 .238
Salad 30 58 52 s 11 45 1.3 305.7 0.15 .703

Ranch Dr 22 43 5] 13 26 50 1.05 .35,3.] 0.0] .925
Italian Dr 8 13 61 27 56 48 1.72 .43,7.l 0.75 .387

Cheesecake 3 l 6] 51 4 8 50 1.03 . 19,5.6 0.0 .63
Water 34 63 54 l 6 17 5.86 59,143 3.0 .09
Soft Drink 19 40 47 16 29 55 0.74 25,] .9 0.40 .529
Tea 3 4 75 32 65 49 3.09 26,83 1.00 .318
Coﬁee 3 l4 2] 32 55 58 0.2 .04,.90 6.0 .014
Milk 0 l 0 35 68 5 l l .0 .493

 

 

 

The overall ilhress attack rate was 50.7% (35/69). See Table 4 for food-speciﬁc
attack rates. Notable was the attack rate for those who ate the chicken entree which was
55% (35/64). The attack rate for those who did not eat the chicken entree was 0%.

Using Epi-Info software, the only food item with a positive association with
illness was the chicken entree with a p-value of .025 ( Fisher’s Exact Test). The odds
ratio could not be calculated due to the fact that one cell in the table was 0 (0 people were
ill who did not eat the chicken). The chi-square value is also not accurate due to low cell

counts. Therefore the Fisher’s Exact probability test was recormnended.

Laboratory Results

Stool Spgimen Rm
Five of the cases rettu'ned a stool sample to MDCH laboratory. All of them tested

negative for the foodborne pathogens tested for. The specimens were all collected 3-4

daysafterthemealwasconsumed.

Food Smec'angn Results

Two complete chicken entrees were tested. Bacillus cereus was found at a level of
103/gm. There was no way to separately test the sauce and chicken so this is the result of
the homogenation of the two. In addition, Clostridium perﬁingens was found, but only

after enrichment. The test for Bacillus cereus enterotoxin was negative.

Disscussion
Limitations

The duration of the illness was short, about 24 hours, and the symptoms were not
severe. None of the cases required medical attention. No initial stool specimens were
collected. The college student population dining this week of ﬁnal exams was transient.
These factors contributed to the low response rate of stool specimen collection and the
inability to interview a greater percentage of the attendees of the banquet. However, 69
of the 190 attendees were interviewed (3 6%), including 5 of the 20 vegetarian attendees,
a large enough number to produce statistically signiﬁcant results.

All the stool specimen test results were negative for all organisms tested. The
specimens were collected after acute illness was over. Some pathogens such as C.
perﬁingens remain in the intestinal tract only for a few days after onset of the illness.27
It would have been preferable to collect the specimens during acute illness but this was
not practical considering the short duration of symptoms. The collection time was most
likely past the prime “shedding time” when an adequate number of organisms or level of

toxins in the stool can be detected by culturing or toxin testing.

67

Conclusion and Discussion

The results of the statistical analysis show that the most likely source was the
chicken entree. Considering attack rates, none of the attendees who ate only the
vegetarian entree ( which were those who did not eat any chicken entree) were ill
compared to 55% of those eating the chicken entree were ill. The Fisher’s exact p-value
was <05 which tells us that this difference is signiﬁcant. No other food item stood out
with any signiﬁcant results.

Bacillus cereus was found in the chicken entree specimen submitted (103/gm) but
not at a deﬁnitive level. IAMFES Procedures to Investigate F oodborne Illn_e_s_s states that
to be conﬁrmation of the cause it must be found at a level of lOs/gm in the
epidemiologically implicated food. C. perfringens was found only after emichment and
was therefore not deﬁnitive. To be considered conﬁrmation it would also have to be
found at a level of at least 103/gm of epidemiologically implicated food. No other enteric
pathogens were found. The sample included chicken with sauce and the laboratory could
not test the two separately. It is unusual to ﬁnd two different pathogenic organisms in a
food sample, even at these lower levels. C. perfringens would most likely have
originated ﬁom the ﬂesh chicken and the B. cereus ﬁ'om the sauce ingredients, but there
is no way of determining this without extensive testing of the food items in the same lots
as the original meal ingredients.

Inadequate cooling procedures in the sauce preparation would allow for
germination of surviving bacterial spores and subsequent growth. Inadequate cooling
procedm'es were found in this kitchen for an unrelated food item during the

environmental inspection, making this scenario more plausrhle. The partial cooking of

68

the chicken breasts the day prior to serving may also have allowed survival of pathogenic
organisms that would grow if not quickly cooled. The environmental inspection report
highlights the need for improved employee hygienic procedures and possible cro ss-
contamination in the walk-in cooler.

Although positive stool cultures would have been desirable, positive food
specimen results (although not high enough to be conclusive), the clinical features of the
illness, including incubation period and the statistical Signiﬁcance of the attack rates all
support the following conclusion: The outbreak was caused by ingestion of the chicken
entree that had been naturally contaminated with the pathogenic organisms. The
ingredients were heated and then cooled slowly, aﬂong germination of spores of one or
both of these two organisms and subsequent toxin formation in the intestines which in
turn caused the illness.

A similar outbreak occurred in Napa, California in 1989.42 A wedding reception
including 140 guests was catered by a local restaurant. The outdoor buffet included
Cornish game hens, Kentucky Fried Chicken and various side dishes. The guest list was
obtained and 56% of the group (79/140) were interviewed. Of those interviewed, 70%
(55/79) met the case deﬁnition. The overall attack rate was at least 39%(55/ 140). The
median incubation period was 12 hours and the median duration was 19 hours.

The statistical analysis showed that consumption of the Cornish game hen was
highly associated with illness (OR=29.33, p=.0001) Fecal samples were only collected
ﬁ'om two guests. These were only tested for the common enteric bacteria and were not
tested for B. cereus or C. perfringens. However, both B. cereus and C. perfringens were

found in the Cornish game hen sample. B. cereus was found at 14 million per gram of

69

food. C. perfringens was present at 99,000 per gram. The plate count for B. cereus was
suﬂicient as conﬁrmation of causative organism. The C. perfringens was not at a high
enough level for conﬁrmation, although the investigators left open the possibility that
some of the illness may have been caused by C. perﬁingens. The actual dosage could
have been high enough in the Cornish hens eaten to cause illness.

In this outbreak bacterial multiplication was facilitated at several points during the
preparation and transportation of the food:

1. The main reﬁigeration unit used in the catering operation was only holding
temperatures of 48° F.-50°F.;

2. The birds were probably thawed incompletely, resulting in inadequate cooking at
the center;

3. Cross-contamination occurred when the same brush was used for basting both
before and after cooking without being washed; and

4. On the day of the event, when the outdoor temperature reached 90°F ., the hens
were held for approximately 4 1/2 hours in an unrefrigerated van while a delivery
was made elsewhere in the county.

Like the MSU outbreak, this outbreak resulted ﬁom improper temperature
control that was conducive to the growth of both B. cereus and C. petﬁingens and
resulting toxin formation. In addition to the onset times and symptomology being
similar, the fact that these two organisms grow under the same conditions should lead

investigators to test for both of these organisms and toxins when one of them is

suspected.

70

Recommendations to the Foodservice Establishment

Good employee hygiene is always stressed in the prevention of foodborne illness.
It is recommended that employee training be increased in this area and reviewed
ﬁ'equently. This food service estabishment is one of the more well-trained facilities in
Ingham County. The new Food Code requires that employers have an employee hygiene
training policy in place. In ﬁct, the MSU DPS Registered Sanitarian offers regular
training for MSU employees. All ﬁll] time employees working for the food services at
MSU are required to attend a %-day food service sanitation class that includes basic
HACCP principles. The chefs and other managers are required to be certiﬁed with Serv-
Safe or an equivalent course, including the chef at Kellogg Center at the time of the
outbreak.

The Hazard Analysis Critical Control Point (HACCP) system is used as a method
of ensuring safe food processing “ﬁom the ﬁrm to the table”. Developed in the 1970’s
for the National Aeronautics and Space Administration (NASA) for the astronauts’ foods,
it is now applied at all levels of the food industry. The HACCP system emphasizes the
quality of all ingredients and all processing steps on the premise that the resulting product
will be safe if these steps are properly controlled. The system addresses all of the leading
causes of foodborne illness listed earlier. A “critical control point” (CCP) is any point in
a speciﬁc food system where control can be exercised and a hazard can be minimized or
prevented.” The essential elements of HACCP are:

1. Assess hazards and risks of the food process.

2. Determine the critical control points to ensure control of or minimize a hazard

71

3. Establish critical limits (prescribed tolerances that must be met to ensure that a
CCP eﬂ‘ectively controls a microbiological hazard. An example is to cook a
hamburger to at least 155°F. for 15 seconds.)

4. Establish procedures to monitor CCP’s, such as charting temperatures of
reﬁigeration units.

5. Establish corrective actions to be taken when deviations occur in CCP monitoring.

6. Establish effective record-keeping to document the HACCP plan.

7. Establish procedures for veriﬁcation that the HACCP system is working correctly.
The recipes used at Kellogg Center are not HACCP recipes. The principles are

usually incorporated into the recipes, but there is no requirement for the recipes to be in
HACCP style. However, the dormitory food service cafeterias are in the process of
changing their recipes to reﬂect good HACCP procedures. Figure 7 shows the actual
recipe used for the banquet and Figure 8 shows the recipe converted to a HACCP style
recipe. The inspection process for the food services has also evolved into a HACCP

inspection, from the point of delivery to the point of service at the cafeteria line.

72

 

Breast of Chicken AuGratin- Recipe used in May, 1998

6 oz. chicken breast
V2 oz. mushrooms
'/2 oz. leeks, sliced

'/2 oz. red peppers
2 oz. cream sauce

'/2 oz. grated parmesan
'/2 oz. bread crumbs

6 oz. Chicken breast sauteed lightly on both sides, season with salt & pepper.
Place chicken breast on sheet tray.

Cook for 25 minutes at 325° , until internal temp reaches 165°.

Saute mushrooms, leeks, green and re peppers.

After chicken is cooked, top each breast with 2 oz. of vegetable mixture.

Sprinkle 1 oz. of Parmesan bread mixture on top of chicken, vegetables and cream
sauce.

0 Place chicken tmder broiler for 5-10 seconds to brown off Parmesan mixture.

Parmesan Cream Sauce- Recipe used in May, 1998

4 qt. Whipping cream
'/2 qt. White wine
4 bay leaves

salt and pepper
16 oz. roux (mixture of butter and ﬂour)

1 t. cayenne pepper
1 T. dry basil, chopped

Heat together wine and bay leaves. Reduce to '/2.

Add cream and simmer for about '/2 hour on low-med. Heat.
Thicken with roux to a smooth consistency.

Add seasoning and hold in bay.

 

 

 

Figure 8. Recipes for chicken augratin entrée served at banquet, May, 1998

73

 

Breast of Chicken AuGratin- Sagested HACCP-based Recipe

Cream Sauce

4 qt. Whipping cream- held at 41° F. or below 16 oz. roux (butter and ﬂour)
'/2 qt. White wine 1 t. cayenne pepper

4 bay leaves 1 T. dry basil, chopped

salt and pepper

In a pot heat together wine and bay leaves. Reduce to '/2.
Add cream and simmer for about '/2 hour on low-med heat.
Thicken with roux to a smooth consistency.
Add seasoning.
If ﬂing immediately:
0 CCP: immediately transfer to hot-holding unit and hold at 140°F. or higher until chicken
entrees are assembled.
If using the next day:
- CCP: Cool quickly down to 70° F. or lower within 2 hours using an ice bath and ice paddles
for stirring. Continue cooling to 41°F. or lower within another 4 hours using the same
procedure or transfer sauce to shallow containers and store in walk-in cooler. Monitor
temperature.

Chicken Entree

6 oz. chicken breast held ﬁ'ozen in ﬁeezer, then thawed in walk-in cooler in drip-proof
container.

'/2 oz. mushrooms

'/2 oz. leeks, sliced ‘/2 oz. grated parmesan

'/2 oz. red peppers '/2 oz. bread crumbs

2 oz. cream sauce prepared as described above

0 Saute chicken breast lightly on both sides. Season with salt and pepper.
Place chicken breast on sheet tray.
CCP: Cook for about 25 minutes at 325° F. oven temperature until internal temperature of
chicken breast reaches 165° F.
o CCP: Hold at 140°F. or higher until entrees are assembled.
o CCP: lfassembling and serving the next day, cool quickly to 70° F. or lower within 2
hours and to 41° F. or lower within another 4 horns by chilling in coldest part of walk-in
cooler in single layers on trays. Monitor temperature.
Ifserving immediately:
Saute mushrooms, leeks, green and red peppers.
Top each breast with 2 oz. of vegetable mixture.
Pour 2 oz. of hot cream sauce on top of chicken and vegetable mixture.
Sprinkle 1 oz. of parmesan bread mixture on top of chicken, vegetables and cream sauce.
Place chicken under broiler for 5-10 seconds to brown off parmesan mixture.
CCP: Hold entrees at 140° F. or higher until served.
If assembling and serving the next day:
0 CCP: Heat cream sauce and chicken portions to 165°F. or higher prior to assembling with
vegetables as described above.
0 CCP: After assemblina. hold at 140°F. or higher until served.

 

 

 

Figure 9. HACCP-based recipe for Chicken Entrée served May, 1998
Indicating Critical Control Points (CCP)

74

The practice of cooling cooked foods was reviewed. In order to prevent
sporulation and subsequent growth of organisms surviving the cooking stage, the foods
must be cooled quickly, within four hours to 45°F or below based on the 1979 FDA
Code. The new 1999 FDA code requires cooling to a temperature below 70°F. within 2
hours and then to 41°F . or below within another 4 hours.

Sauces prepared ahead of time should be cooled in a water bath and stirred with
an ice-ﬁlled stirring paddle or transferred to shallow pans to bring the hot sauce down to
proper temperatures within the safe time requirements. Water bath use with ice paddles
is the standard operating procedure for cooling soups and sauces at this food service
establishment. However, it appears that during this unusually busy period this practice
was not completely followed or documented.

The cooperative effort of staff members in the bureaus of Communicable Disease
and Environmental Health and Michigan State University was instrumental in the quick
response and thorough investigation of this outbreak. Notable also was the highly
professional attitude and total cooperation of the management of this highly respected
food service. This allowed for the most complete investigation possible and resulted in a

valuable learning experience for all parties involved.

75

CONCLUSION

The 1998 foodborne illness outbreak described is an example of many
investigations that have been done in the last few years at ICHD. The protocol has
helped to make the process more efﬁcient. Every outbreak is unique and it is helpﬁll to
have a guide to follow in order to be sure steps are not missed. More outbreaks are
discovered and investigated each year in Ingham County. The information gained by
thorough investigation is valuable not only to the particular food service establishment,
but at the state and federal level, as it is tabulated into the appropriate surveillance
systems. These surveillance systems, both active and passive, are improving in their
capabilities to more accurately measure the burden of foodborne disease and its causes.
The FDA has used this information to update the FDA Food Code, the 1999 version
being the most recent. This updated version is an outstanding improvement, having
incorporated public health interventions designed to address the most ﬁ‘equent causes of
foodborne illness. As the epidemiology of foodborne illness continues to change, public
health policy will also change to more effectively prevent foodborne illness. To
accomplish this goal, the importance of foodborne illness surveillance and outbreak

investigation must become a priority at the national, state and local levels.

76

APPENDIX

BACKGROUND INFORMATION ON CLOSTRIDIUMPERFRLNGENS
AND BACILL US CERE US

Background Information on Clostridium pedringens

Clostridium petfringens is an anaerobic, gram-positive, spore-forming rod. It is
widely distributed in the environment and is commonly found to inhabit the
gastrointestinal (GI) tract of humans, domestic livestock and other warm-blooded
animals. Spores of the organism persist in soil, sediments, and areas subject to human or
animal fecal contamination. Meats become contaminated during the slaughtering process
and poultry during line processing. The organism can then grow on the food surfaces.
Cooking usually kills the vegetative cells but heat-stable spores survive. Then if the
food is held at improper temperatures the spores can germinate and proliferate in the
food. The most ﬁ'equently implicated foods include roast beef and gravy, turkey,
dressing and gravy, casseroles containing meat/poultry, stews, Mexican foods, sauces,
salads, and pork.4°’“ The illness is a relatively mild, self limiting disease characterized
by diarrhea and abdominal cramps 6-24 (usually 10-12) hours after ingestion of the
contaminated food. The illness usually resolves within 12-24 hours.

C. perﬁingens has been recognized as a cause of foodborne illness for more than
50 years.34 C. perﬁingens can cause other diseases such as gas gangrene and at least 13
different toxins have been identiﬁed ﬁ‘om the organism. Any individual strain will
produce only a few of the toxins. The production of four certain major toxins (a,B,c,t) is
the basis for subtyping into ﬁve toxigenic types (A,B,C,D and E).34 The toxin
associated with foodborne illness is termed the C. perﬁingens enterotoxin (CPE) and can

be present in any of these. Foodborne illnesses due to C. perfringens are almost always

77

caused by serotype A. Ingestion of a large number of vegetative organisms (>=105) is
followed by sporulation of the bacterial cells in the small intestine with the production of
the enterotoxin. CPE synthesis is increased by sporulation. Sporulation is also required
for CPE release ﬁom the cell. This second sporulation may occur in the food during
reheating, but most likely occurs when the bacterial cell reaches the low pH of the
stomach. The pathogenesis of CPE involves binding to the tips of the villi of the
epithelial cells of the ileum, insertion into the cell membrane and a formation of a
complex which causes pore formation, alteration in permeability and ion ﬂuxes,
inhibition of cell metabolism, cytoskeleta] disintegration and ultimately cell lysis.34 The

result of this process is cramping and diarrhea.

 

Reported C.parfrlngens Outbreaks

Number of Outbreaks

 

Year

 

 

 

Figure 10. Outbreaks of C. perfringens in the U.S."n’l7

78

 

Reported C.perlrtngens Cases

.M\

V T

1991 1992 1993 1994 1995 1996 1997
Year

dd

 

 

 

 

Number of Cases

 

 

 

 

 

 

Figure 11. Cases of C. perfringens in the U.S."“’”

During 1991, 528 outbreaks involving 14,876 cases of foodborne disease were
reported to CDC. Only 214 (41%) were of conﬁrmed etiology. Of those 214 outbreaks,
173 were ﬁ'om bacterial pathogens. Of the bacterial pathogens Figure 12 shows
C. perﬁ'ingens was third in number of outbreaks (10 outbreaks, 1,213 cases). Salmonella
caused most of the conﬁrmed outbreaks (122 outbreaks, 4,146 cases) and second was
C. botulinum (11 outbreaks, 25 cases).4 In 1992 C. perﬁ'ingens jumped to second place
(12 outbreaks, 912 cases) following Salmonella (117 outbreaks, 4,156 cases) of
etiologically conﬁrmed outbreaks reported. For both years C. perfringens claimed
more cases than any other bacterial pathogen except for Salmonella (Figure 13)

The number of reported illnesses due to C. perfringens has dropped in recent
years. It still remains the third in number of outbreaks caused by bacterial pathogens
from years 1993, 1994, 1995,1996 and ﬁfth in 1997. The number of reported cases due

to outbreaks has dropped also.17

79

 

 

 

 

 

 

 

 

 

    

 

 

Number of Outbreaks Reported for Selected
Bacterial Pathogens, 1991 -1997

8
t 140
8, 120 +Salmonella
2 100
a 80 -l- C.botulinum
.1:
3 so C.perfrlngens
3 40 +E.coli(all)
'5 20 +S.aureus
O 0 +Campylobacter

1991 1992 1993 1994 1995 1996 1997 +33”;

Year
30
E 25 +C.botulinum
§ 20 c.perfringens
1: 15 +E.coll (all)
% +s.aureus
3 1° +Campylohacter
3 5 —l—B.cereus
o ‘A, ,. r z ,
1991 1992 1993 1994 1995 1996 1997
Year

 

 

Figure 12. Number of Outbreaks of Selected Bacterial Pathogens, 1991-1997
(Lower graph is of smaller scale, excludes Salmonella to magnify the remaining
pathogens.)"’"

80

 

 

Number of Reported Outbreak Cases for Selected
Bacterial Pathogens, 1991 -1997

 

+ Salmonella

—I— C.botulinum
C.perfrlngens

—x-— E.coli (all)

+ Saunas

+ Campylobacter

—+—— B.cereus

Cases reported

‘-h-—__.

1991 1992 1993 1994 1995 1996 1997

 

 

 

Year

 

 

 

1000
1400
1200
1000

 

+ C.botulinum
C.perfringens

+ E. coli

+ S.aureus

+ Campylobacter

—+— B. cereus

 

 

 

 

Outbreak Cases Reported
c. § § § §

1991 1992 1993 1994 1995 1999 1997
Year

 

 

 

Figure 13. Number of Reported Cases of Selected Bacterial Pathogens, 1991-1997.
(Lower graph is smaller scale, Salmonella deleted, to magnify remaining
pathogens.)1“”

Standard bacteriological culturing procedures are used to detect the organism in
implicated foods and in fecal samples from patients. Serological assays are used for

detecting enterotoxin in the feces of patients and for testing the ability of strains to

produce toxin. The criteria used for conﬁrmation of a case or outbreak include?"

8]

a. Isolation of the same serotype of C. perfringens from specimens from most ill

persons but not from controls; or

b. Isolation of the same serotype of C. perﬁingens from ill persons and

epidemiologically implicated food; or

c. Fecal spore count of >106/g C. perﬁingens in most ill persons examined

within a few days of illness; or

(1. Isolation of >=105 C. perﬁingens in epidemiologically implicated food; or

e. Demonstration of toxin in feces by reverse passive hemagglutination,

ﬂourescent antibody or other techniques.
(Methods c, d, and e are the most frequently used methods.)

A conﬁrmed outbreak of C. perﬁ'ingens occurred in Ingham County in November
1990.41 Thirty-two of forty two (76%) of attendees of a cake decorator’s conference
became ill with gastroenteritis and a median incubation period of 6.5 hours (range 1.5-15
hours). 11] persons were ﬁve times more likely to have consumed minestrone soup than
were those who were not ill (relative risk=4.92, 95% conﬁdence interval 1.23-00). C.
perfringens (>1050rganisms/g) was isolated from 11 of 12 stool samples collected. C.
perfringens (lO7organisms/ g) was isolated from the minestrone soup. The environmental
investigation found that the soup was cooked two days prior to serving and was slowly
cooled before refrigeration. On the day it was served it was brieﬂy reheated. The lack of
time and temperature controls during the cooling and reheating of the soup most likely
allowed for the proliferation of the C. perfringens in the soup and subsequent toxin
formation. The relatively short incubation period (median 6.5 hours, range 1.5-15 hours)

suggests that sporulating cells or preformed toxin may have been ingested.

82

Background Information on Bacillus Cereus

Bacillus cereus is an aerobic or facultatively anaerobic, gram-positive spore-
forming rod. It is commonly found in soil, air, dust, water and in many raw and
processed foods. B. cereus may be isolated ﬁom rice, dairy products, meat, spices and
eggs. The bacterium is capable of surviving different environmental stresses, making it
very diﬂicult for the food industry to control.

B. cereus causes two diﬁ‘erent types of foodborne illness: diarrheal and emetic.
The diarrheal type of illness is caused by complex, heat labile enterotoxins produced
during vegetative growth of B. cereus in the small intestine. This type of illness is
considered an infection. For the diarrheal type the incubation period is ﬁ'om 8 to 16
hours. The emetic toxin causes vomiting and is produced in the food by growing cells.
The incubation period for the emetic type of illness is from 1/2 hour to 6 hours. The toxin
is injested while in the food and the resulting illness is considered an intoxication. This
toxin is resistant to heat and is therefore not inactivated by reheating of foods.20 For
both types of illness the duration is usually less than 24 hours.

In both types the food involved has usually been heat-treated. The surviving
spores then produce either the enterotoxins or the emetic toxin. B. cereus grows well after
cooking and cooling. The heat treatment (cooking) causes spore germination and
subsequent growth. Different strains of B. cereus produce different amounts of
enterotoxin. The infectious dose appears to range ﬁom 10 5 — 108 viable cells. Factors
contributing to foodborne outbreaks include: Storing cooked foods at room temperature;

storing cooked foods in large containers in reﬁigerator; preparing foods several hours

83

before serving; inadequate reheating of leftovers. The emetic type of illness has most
often been associated with boiled or ﬁ'ied rice, cooked cornmeal dishes, cooked cereals
and pasta. The diarrheal type has been associated with cereal product, soups, custards
and sauces, meatloaf, sausage, cooked vegetables, reconstituted dried potatoes and reﬁied
beans?”

 

Reported Bacllus cereus outbreaks in the
U.S.

    

ON‘O

v «a, _ , J. .r i,» n7 if, BJOSN’A‘YJJﬁ-sy-.v'EZRwAﬁ-I71-1.. ,
1991 1992 1993 1994 1995 1996 1997

Year

number of outbreaks

 

 

 

Figure 14. Reported B. cereus Outbreaks in the U.S."l7

 

Reported Bacillus cereus cases in the
U.S.

 

numberofcases
.. '2‘ § § § §

1991 1992 1993 1994 1995 1990 1997

Year

 

 

 

Figure 15. Reported B. cereus Cases in the U.S."l7

Cases of B. cereus foodborne illness are not routinely reported (Table 1). No more
than 5 outbreaks per year have been reported to the CDC ﬁom 1991-1997 (Figure14).
Those outbreaks accounted for as few as 22 cases in 1996 to a high of 438 in 1997
(Figure 15). Because it is a mild illness, it is considered to be highly underreported and
has been estimated by the CDC to actually be 27,360 cases per year in the U.S.(Table 1).

Standard bacteriological culturing procedures are used to detect the organism in
implicated foods and in fecal samples from patients. Serological assays are used for
detecting enterotoxin in the feces of patients and for testing the ability of strains to
produce toxin. The criteria used for conﬁrmation of a case or outbreak include:

a. Isolation of the same serotype of B. cereus from stool specimen from ill

person(s) but not from controls; or

b. Isolation of >=1 05 B. cereus per gram ﬁom epidemiologically implicated

food;or

0. Detection of enterotoxin by ligated rabbit ileal loop, vascular permeability

reaction, immunogel-diﬂ‘usion, aggregate-hemagglutination, reverse passive
latex agglutination; or

(1. Detection of emetic enterotoxin by monkey feeding, Hep-2 cell, or other test.

85

10.

ll

12.

13.

REFERENCES

. Albrecht J. Food safety knowledge and practices of consumers in the U.S.A.

J Consumer Studies and Home Economics 1995;19:119-34.

Andersson A, Ronner U, Granum P. What problems does the food industry have with
the spore-forming pathogens Bacillus cereus and CIostridium perfringens? Internat J
Food Microb 1995;28(1995):145-155.

Angulo F, Voetsch A, Vugia D, et a1. Determining the Burden of Human Illness from
foodborne diseases: CDC’s Emerging Infectious Disease Program Foodborne Disease
Active Surveillance Network (FoodNet).[Online] Available
http://www.cdc.gov/ncidod/dbmd/foodnet/pub, 1998.

Bean N, Goulding J, Lao C, et a1. Surveillance for Foodborne-Disease Outbreaks——
United States, 1988-1992. WWR 1996;45(SS-5):1-55.

Bryan F. Hazard Analysis Critical Control Point (HACCP) Systems for Retail Food
and Restaurant Operations. J Food Protection 1990;53(11): 978-983.

Bryan F. Risks Associated with Vehicles of Foodborne Pathogens and Toxins. J Food
Protection l988;51(6):498-508.

Bryan F. Risks of Practices, Procedures and Processes that Lead to Outbreaks of
Foodborne Diseases. J Food Protection 1988;51(8):663-673.

Centers for Disease Control (CDC). Appendix B Guidelines for Conﬁrmation of
Foodbome-Disease Outbreaks. MMWR 2000;49(SSOI):54-62.

CDC. Bacillus cereus Food Poisoning Associated with Fried Rice at Two Child Day
Care Centers-Virginia, 1993. MMWR 1994;43(10):177-178.

CDC. Epi-Info (Version 6) Statistical Software package.

. CDC. Foodborne Disease Outbreaks, 5-Year Summary, 1983-1987. MMWR

1990;39(SSOl):15-23.

CDC. Healthy People 2010 Chapter 10 Food Safety. [Online] Available
http://www. health. gov/healthypeople/Document/HTML/Volume1/1 OFood.htm.
2001.

CDC. Incidence of Foodborne Illnesses: Preliminary Data ﬁ‘om the F oodbome

Diseases Active Surveillance Network (F oodNet)——U.S., 1998. MMWR
1999;48(9):189-194.

86

14. CDC. Preliminary FoodNet Data on the Incidence of Foodborne Illnesses—Selected
Sites, United States, 1999. MMWR 2000;49(10):201-5.

15. CDC. Preliminary FoodNet Data on the Incidence of Foodborne Illnesses—Selected
Sites, United States, 2000. MMWR 2001;50(13):241-6.

16. CDC. Surveillance for Foodbome-Disease Outbreaks—United States, 1988-1992.
MMWR 1996;45(SS-5):1-55.

17. CDC. Surveillance for Foodborne Disease Outbreaks—United States, 1993-1997.
MMWR 2000;49(SSOl):1-51.

l8. Chin J. Control of Communicable Diseases Manual, 17th Edition. Washington, DC,
American Public Health Association, 2000.

19. Crane J. Preformed Bacterial Toxins. Clinics in Laboratory Medicine
l999;l9(3):583-599.

20. Dufrenne J, Soentoro P, Tatini S, et a1. Characteristics of Bacillus cereus related to
safe food production. Int J Food Micro 1994;23:99-109.

21. F SIS. Reducing the Risk of Listeria monocytogenes Joint Response to the President,
Executive Summary. [Online] Available
http://www.foodsafetlgovhdmg/lmriplan.html. 2001:1-10.

22. Frewer L, Shepherd R, Sparks P. The interrelationship between perceived knowledge,
control and risk associated with a range of food-related hazards targeted at the
individual, other people and society. J Food Safety 1994;14:19-40.

23. Granum PE, Lund T. Bacillus cereus and its food poisoning toxins. FEMS
Microbiology Letters 1997;157:223-228.

24. Gregg M. Field Epidemiology. New York: Oxford University Press, 1996:92-131.
25. Ingham County Health Dept. Foodborne Illness Investigation Protocol 1998.

26. Ingham County Health Dept. Press Release: An Outbreak of Salmonella enteriditis in
Ingham County, Aug. 21,1997.

27. International Association of Milk, Food and Environmental Sanitarians, Inc.
(IAMFES). Procedures to Investigate Foodborne Illness Fifth Edition. Des Moines,
Iowa: IAMFES 1999.

28. Jackson L, Wenger J. Listeriosis: a foodborne disease. Infections in Medicine
1993;10:61-66.

87

29. Jay J. Modern Food Microbiology. New York, N.Y.: International Thomson
Publishing 1996: 408-417.

30. Knabel S. Role of Home Food Handling Practices. Food Technology 1995; 49(4):
1 19- 1 3 1 .

31. Last J. A Dictionary of Epidemiology Third Edition. New York: Oxford University
Press, 1995.

32. Lindsay J. Clostridium perﬂingens Type A Enterotoxin (CPE): More Than Just
Explosive Diarrhea. Crit Rev Microb l996;22(4):257-277.

33. Mead P, Slutsker L, Dietz V, et a1. Food-Related Illness and Death in the United
States. Emerg InfDis 1999;5(5):607-625.

34. Meer R, Songer G, Park D. Human Disease Associated with Clostridium perfringens
Enterotoxin. Rev Environ Contam Toxicol 1997;150:75-94.

35. Michigan Department of Agriculture, Food and Dairy Division. 1998 Year-End
Surmnary Foodborne Illness Surveillance. Oct 1995.

36. Michigan Department of Agriculture, Food and Dairy Division. 1999 Year-end
Summary Foodborne Illness Surveillance. Sept 2000.

37. Michigan Department of Agriculture, Food and Dairy Division, Food Service
Sanitation Section. Minimum Program Requirements- A manual of food Service
Sanitation Regulatory Minimum Program Requirements. August 2000.

38. Michigan Department of Community Health Bureau of Laboratories, Division of
Infectious Diseases. Foodborne Illness and Poisoning: Bacillus cereus Colony Count
(Procedure) 1998.

39. Michigan Department of Community Health Bureau of Laboratories, Division of
Infectious Diseases. Foodborne Illness and Poisoning: Clostridium perjﬁ'ingens
Colony Count (Procedure) 1998.

40. Reed G. F oodborne Illness O’art 3) Clostridium perﬁingens Gastroenteritis. Dairy,
Food and Enviromnental Sanitation 1994;l4(l):l6-17.

41. Roach R, Sienko D. Clostridium perfringens Outbreak Associated with Minestrone
Soup. Am J Epidemiol 1992;136(10):1288-1291.

42. Slaten D, Oropeza R, Werner S. An Outbreak of Bacillus cereus Food Poisoning—
Are Caterers Supervised Suﬁiciently? Pub Health Reports l992;107(4):477-480.

88

43. Sondik E. Healthy People 2000 Progress Review, Food Safety (Part 2) [Online].
Available http://www.cdgov/nchs/aboutotheracUhpZOOO/foodsafety/foodcharts.htm.
1999.

44. State of Michigan, Michigan Compiled Laws, Act 92 of 2000 sections 289.3103-
289.3139.

45. State of Michigan, Michigan Compiled Laws, Act 368 of 1978, Mich Public Health
Code PA. 1978.

46. Taormina P, Beuchat L, Slutsker L. Infections Associated with Eating Seed Sprouts:
An International Concern. Emerg InfDis l999;5(5):626-634.

47. Tauxe R,Vandepitte J, Wauters G. Yersinia enterocolitica infections and pork: the
missing link. Lancet 1987;]:1129-32.

48. Tauxe R. Emerging Foodborne Diseases: An Evolving Public Health Challenge.
Emerg Inf Dis l997;3(4):425-434.

49. Unipath Limited. Oxoid Toxin Detection Kits, PET-RPLA CIostridium Perfringens
Enterotoxin Test Kit. 1990.

50. Unipath Limited. Oxoid Toxin Detection Kits, BCET-RPLA Bacillus Cereus
Enterotoxin (Diarrhoeal type) Test Kit. 1990.

51. U.S. Department of Health and Human Services, Public Health Service, Food and
Drug Administration. Food Code-1999 Recommendations of the U.S. Public Health
Service Food and Drug Administration. 1999.

52. United States Food and Drug Administration, Center for Food Safety & Applied
Nutrition. Foodborne Pathogenic Microorganisms and Natural Toxins Handbook.

53. Yang S. FoodNet and Enter-net: Emerging Surveillance Programs for Foodborne
Diseases. Emerg InfDis 1998;4(3):457-8.

54. Yang S, Leﬁ‘ M, McTague D. Multistate Surveillance for Food-Handling, Preparation

and Consumption Behaviors Associated with Foodborne Diseases: 1995 and 1996
BRF SS Food-Safety Questions. MMWR l998;47(SS-4):33-54.

89