u.
3

u in
a

a:
. .weafia

.

 

'

12% “V."
“mfimmnn
anti.

.1

 

 

.13" .99‘ ‘ Vol 5
. LALCZ u v 41‘! ‘3 w * a
.‘J‘. ‘ . . “2.!th . J...) 1
. .3... i. nit ‘~ $4.23. 1,.er V . :3?
‘ . . . . . . .. Jan: ! .4
A . ‘ , u i. .
‘ ins". flat. ' x .

2
2 00 4
59590753 LIBRARY
Michigan State
University

This is to certify that the
thesis entitled

Microbiological Assessment of Soft Serve Desserts as a
Means of Determining the Effectiveness of the Current Frozen
Dessert Standard

presented by

Ijeoma O. Okpala

has been accepted towards fulfillment
of the requirements for the

Master of degree in Food Science
Science

 

(Sada £4.“

Major Professor’s/Signature

11/42/53

Date

MSU is an Afl'innative 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/0_1 c:/CIRCIDateDue.p65-p.15

 

MICROBIOLOGICAL ASSESSMENT OF SOFT SERVE DESSERTS AS A MEANS
OF DETERMINING THE EFFECTIVENESS OF THE CURRENT FROZEN
DESSERT STANDARD
BY

Ijeoma 0.0kpala

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Food Science and Human Nutrition

2003

ABSTRACT
MICROBIOLOGICAL ASSESSMENT OF SOFT SERVE DESSERTS AS A MEANS
OF DETERMINING THE EFFECTIVENESS OF THE CURRENT FROZEN
DESSERT STANDARD.
By
Ijeoma O. Okpala

The Frozen Desserts Act of 1968, PA. 298, was implemented to protect public
health, and to prevent fiaud and deception in the manufacture and sale of adulterated or
deleterious ice cream and ice cream mix. In cooperation with Michigan State University
(MSU) and local health departments, the Michigan Department of Agriculture
commissioned a soft serve ice cream sampling project to assess the effectiveness of
practices used in the manufacture of soft serve ice cream and to determine the level of
compliance with the Act. Over a 7-month period, a total of 227 samples of frozen soft
serve desserts (soft serve ice cream) were collected, 133 of those samples being vanilla
and 94 being chocolate. The samples were then tested for numbers of coliform and
mesophilic aerobic bacteria (SPC), and for the presence of Listeria spp.

Overall, 49.8% (113/227) ofthe samples exceeded the allowable number of
aerobic mesophilic bacteria as determined by standard plate count (SPC), with 39.6%
(53/134) and 65.6% (61/93) of the vanilla and chocolate samples, respectively, being
non-compliant. For coliform counts, 30.6% (41/134) of the vanilla samples and 29%

(27/93) of the chocolate samples were non-compliant. Listeria spp. were not found in any

of the samples.

DEDICATION

To my family
who have supported and encouraged me throughout my study and my life.

ACKNOWLEDGEMENTS

I would like to express my heart-felt gratitude to my advisors, Dr. Thomas Wilson
and Mr. Gerald Wojtala (of the Michigan Department of Agriculture) for their guidance,
encouragement and enthusiasm throughout the course of my study. I would like to extend
my appreciation to my guidance committee, Dr. John Kaneene, Dr. Vincent Hegarty, and
Dr. Elliot Ryser for all their efforts and patience.

I would like to thank Dr. John Tilden and J. Douglas Park fiom the Michigan
Department of Agriculture for their mentorship and friendship. I am thankful to Siobhan
Kent for her friendship and enthusiasm throughout my study. I am equally thankful to all
the lab technicians and inspectors who helped me with specific phases of this project. I
am grateful to Roseann Miller at the Population Medicine Center for all her help in the
analysis of the data. Furthermore, I would like to thank MaryAnne Verleger at the
Institute for Food Laws and Regulations, for all her help, both administratively and
personally.

Finally, I would like to thank my parents, Dr. Donatus Okpala and Mrs. Cecilia
Okpala for all their support and for always encouraging me and making me believe that I
can achieve whatever I set out to do. Last, but certainly not least, I would like to thank
my brothers, Dozie and Emenem and my sisters, Ogo, Ify and Aku for all their love and

support.

TABLE OF CONTENTS

List of Figures

List of Tables

INTRODUCTION

3.

Objectives

. LITERATURE REVIEW

1.1. Food and Dairy Division
1.2. Standards
1.3. Sofi Serve Ice Cream
1.3.1. Cleaning and Sanitizing
1.3.2. Storage of Mix
1.3.3. Sanitizers and Cleaners
1.4. Soft Serve Freezers
1.5. Microbiological Problems
1.5.1. Coliforrn Bacteria
1.5.2. Listeria monocytogenes
1.6. Ice Cream Surveys
1.7. Ice Cream Recalls
1.7.1. Complaints in Michigan

Microbiological Assessment of Soft Serve Desserts
2. 1 . Introduction
2.2. Materials and Methods
2.3. Results and Discussion
2.4. Conclusions and Recommendations
2.4.1. Summary and Conclusions
2.4.2. Recommendations for Future Research

APPENDIX

3.1. Data Tables

3.2. Definitions

3.3. Regulations

3.4. Map of Regional Offices
3.5. Survey of Sanitary Practices

BIBLIOGRAPHY

vi

vii

Ar—I

\IQMUI

11

15
15
20
23
25
28
31
32

35
35
36
40
60
60
62

64
65
77
81
86
87

91

List of Figures
Figure

1.1. Typical composition of soft serve ice cream mix
1.2. Process of ice - cream production

1.3. Model PH71

1.4. Model PH84

1.5. Model PH85

2.1. Summary of Standard Plate Counts

2.2. Summary of Total Coliforrn Counts

vi

Page

12
17
17
18
41

42

List of Tables
Table

1.1 Hazards and typical control in the production of ice cream

1.2 Coliforrn counts obtained from 93 milkshake samples

1.3 Standard plate counts obtained from 93 milkshake samples

2.1 Acceptable bacterial levels in vanilla soft serve ice cream

2.2 Analysis of "Plant A" soft serve mix

2.3 Analysis of "Plant B" soft serve mix

2.4 Bacterial levels in soft serve ice cream, by establishment

2.5 Bacterial levels in soft serve ice cream, by volume processed/wk

2.6 Acceptable bacterial levels in soft serve ice cream, by number of times the
machine is completely broken down and cleaned

2.7 Acceptable bacterial levels in soft serve ice cream, by type of establishment

2.8 Acceptable bacterial levels in sofi serve ice cream, by observed employee
practices

2.9 Acceptable bacterial levels from "Establishment S"

2.10 Final multivariable logistic regression model for acceptable standard
plate counts, adjusting for enterprises using "Mix S"

2.11 Final multivariable logistic regression model for acceptable coliform
counts, adjusting for enterprises using "Mix S"

2.12 Summary of standard plate count test results on routine frozen dessert

samples submitted to the Michigan Department of Agriculture
Laboratory Division, 1998 through June 2000.

vii

Page
24
28
29

43

44
45
46

47

50

51

52

53

54

Table Page

2.13 Summary of total coliform count test results on routine frozen dessert 55
samples submitted to the Michigan Department of Agriculture
Laboratory Division, 1998 through June 2000

2.14 Acceptable bacterial levels in soft serve ice cream by County 56

2.15 Acceptable bacterial levels in soft serve ice cream by MDA region 57

2.16 Acceptable bacterial levels in sofi serve ice cream by Jurisdiction 57

2.17 Acceptable bacterial levels in soft serve ice cream, by product temperature 57

2.18 Acceptable bacterial levels in soft serve ice cream, by whether unused mix 58
is rerun.

2.19 Acceptable bacterial levels in sofi serve ice cream, by time mix spends in 58
machine

2.20 Acceptable bacterial levels in soft serve ice cream, by number of days 59

between last cleaning and sample collection

viii

INTRODUCTION

Ice cream is a milk product which contains a variety of ingredients in addition to
milk, cream and sugar. It is legally defined as “any frozen, sweetened milk product
containing no less than 10% milk fat, which is stirred during the process of freezing and
includes every such frozen milk product which contains milk fat or milk solids not fat
and which in any manner simulates the texture or characteristics of ice cream no matter
what coined or trade name it may be sold “ (21 CFR 135.110). It is a very popular
product and as a result of this, its production and consumption is increasing rapidly. The
International Ice Cream Association reported that in 1859, the production of ice cream
was 400 gallons (Andreasen et a1. 1998). In 1955, production increased to 59 million
gallons with 240 million gallons produced in 1981 (Tobias et al., 1981). Today, annual
production in the United States has reached more than 2 billion gallons for ice cream and
related products and 34 million gallons for soft serve ice cream (USDA, 2002). Soft serve
ice cream has been and continues to be a great addition to any food service operation and
is also profitable as an individual operation. The soft serve fi'eezer has allowed ice cream
to be available in a wide variety of operations such as fast food restaurants and retail
outlets.

In the mid 1980’s, several ice cream products and other frozen dairy desserts were
recalled in the United States because of contamination with Listeria monocytogenes and
the recalls incurred expenses that exceeded several million dollars (Anon, 1986).
Increased awareness and improvements in cleaning and sanitation practices have helped
to minimize the presence of L. monocytogenes, as well as other bacteria in dairy plant

environments. However, continuous recalls of ice cream products indicate that problems

still exist in current retail ice cream manufacturing practices.

Typically, soft serve frozen dessert mixes are manufactured (and pasteurized) at a
processing plant and then delivered to the retail establishment in the form of a mix. The
mix is then held at refiigeration temperature in the retail store until it is placed in the soft-
serve fi'eezer where it is made ready for consumption. This type of setting provides
opportunities for grth of contaminating organisms. Storage space for the mixes is used
for other food products as well. This means that the doors to the refrigerated storage area
are opened and closed frequently, which leads to temperature abuse of the mix.

The US Centers for Disease Control and Prevention (CDC) estimates that there
are approximately 76 million cases of food-related illnesses per year, 325,000 of which
result in hospitalizations and 5,000 deaths (Mead et a1. 1999). Infectious diseases are the
third leading cause of death in the United States. According to the CDC, the five leading
causes of foodbome illness outbreaks are inadequate cooling of foods, improper holding
temperatures, poor personal hygienic practices, contaminated equipment, and obtaining
food from unsafe sources.

According the to the Michigan Department of Agriculture, basic food safety
practices that will help reduce the risk of foodbome illness include preventing sick
employees from working with foods, demonstrating food facility manager knowledge,
ensuring proper time and temperature controls for food, preventing contamination of
foods by employee hands, and advising consumers of the risks of eating raw or
undercooked foods of animal origin.

The public’s awareness of the safety of frozen desserts is heightened during the

summer months with the news media often running stories on bacteria levels in ice cream

as was the case in both New York and Michigan. On August 2“, 1998, WDIV TV-4 in
Detroit aired a segment on their 11pm news entitled “Dirty Desserts II” (Dietz, K, 1998).
Kevin Dietz, a correspondent for the news station collected about 48 samples of soft
serve ice cream (three samples collected from each local establishment every two weeks
and tested by AB labs (Detroit, MI) a state certified lab) over a period of two months.
They referred to their findings as “shocking”. The results indicated that 19 of 48 (40%)
samples exceeded the standard for vanilla and flavored soft serve ice cream. The counts
ranged from 690 — 81,000 coliforms/g (69- 8100 times the state limit). At some
establishments, all three samples exceeded the limit. They concluded that if these soft
serve machines are not kept clean, bacteria could build up and cause illnesses.

The Michigan Department of Agriculture, which regulates retail soft serve ice
cream facilities, has encouraged establishments that serve frozen desserts to focus on
important issues such as equipment maintenance, cleaning and sanitizing, product
storage, and employee practices and hygiene, in order to reduce the risk of illnesses
associated with soft serve ice cream. The department also recommends that owners
establish a routine sampling and inspection schedule. This will help to ensure that the
cleaning and sanitizing methods are adequate to maintain a safe product. Emphasis
should also be paid to routine maintenance of machines and employee knowledge of
hygiene and cleaning practices. The department also recommends that establishments
maintain and keep records of sampling results and cleaning schedules that can help
document due diligence in managing their operations. In an effort to assess the practices

used in soft serve operations, the objectives of this project were to:

determine the baseline levels of mesophilic aerobic bacteria, coliform bacteria
and Listeria spp. found in soft serve desserts

assess the effectiveness of handling practices currently used in soft serve
production and to determine if practices have an impact on the bacterial levels,
and

assess the effectiveness of the current frozen dessert standard.

CHAPTER 1
LITERATURE REVIEW

1.1. Food and Dairy Division

The mission of the Food and Dairy Division (FDD) of the Michigan Department
of Agriculture (MDA) is to “protect public health by ensuring a safe and wholesome food
supply, while working to maintain a viable food and dairy industry”. The FDD protects
the public well - being through regulatory enforcement, problem - solving, leadership and
expertise in food safety issues. It inspects more than 20,000 licensed food establishments
annually. The primary goal of a food establishment inspection is to prevent foodbome
disease. Inspections are the primary tool used by a regulatory agency for identifying
procedures and practices that may be hazardous and taking actions to correct deficiencies.
These inspections address sanitary conditions, cleanliness and safety of infrastructure,
freshness and wholesomeness of foods, and truth in labeling. The department has close
working relationships with local health departments, the Michigan Department of
Community Health and the US. Food and Drug Administration. These agencies work
together to identify and respond to outbreaks of foodbome illness. MDA is responsible
for inspecting facilities such as grocery stores, convenience stores, warehouses and food
processors, while local health departments are responsible for inspecting restaurants
(including carryout food), catering trucks and commissaries, food carts and coffee
houses.

In May 2000, Governor John Engler signed Michigan Food Law 2000,
effectively rewriting Michigan Food regulations for the first time in nearly three decades

(Michigan Food Law, 2000). The new law helped food industry managers and regulators

focus on food safety issues and prevent foodbome illnesses. Michigan Food Law 2000 or
Public Act 92, establishes regulatory standards for all licensed food establishments in

Michigan.

1.2. Standards

The Frozen Desserts Act of 1968, PA. 298. (herein after referred to as the Act) was
designed to protect public health, prevent fraud and deception in the manufacture and
sale of adulterated, or deleterious ice cream and ice cream mix and also apply standards
for ice cream.

The Act, however, does not apply to establishments that are under jurisdiction of
the local health department. The Michigan Milk Manufacturing Law of 2000, See 2, part
1 states that “ Frozen desserts manufactured from pasteurized mix in the soft form at food
service or retail food establishments licensed pursuant to the Food Law of 2000, PA 92 of
2000, MCL 289.1101 to MCL 28918111 are exempt from the licensing requirements of
this law and rules promulgated pursuant to this law (Act No. 298)”. This means that
restaurants or other such establishments that offer soft serve ice cream for sale are exempt
from the rules stated in the Act. The Michigan Food Law regulates retailers and the
practices in retail establishments, whereas the Milk Manufacturing Law, which covers the
frozen desserts standard, sets standard for manufacturing.

Under the original law, frozen desserts should not have a standard plate count
(SPC) of more than 50,000 cfu/ml and should not have a coliform count of more than 10
cfu/ml for vanilla products and 20 cfu/ml for chocolate and other products. Under the

newly amended law, the SPC standard was reduced to 30,000 cfir/ml. This is because

under the new law, the standard for condensed milk was 30,000 cfu/ml and hence the
standard for soft serve was reduced to achieve consistency in the law. This was done, as it
would not have been acceptable to increase the standard for condensed milk to match

soft serve mix.

Initially, if a retail establishment wanted a soft serve freezer, they would have
needed a separate “soft serve license” in order to sell the product fiom their store. Under
the new law, establishments licensed under MDA do not need a separate license to sell
soft serve ice cream in their establishment. However, the manufacturer of the nrix, is
required to have a license to produce, distribute and sell the mix. The department expects
that all soft serve samples comply with the given standard, but will not take any action, if
a single sample from an establishment violates the standard. However, if 2 of 4 and 3 of 5
consecutive samples violate the standard, then action will be taken against the

establishment.

1.3. Soft Serve Ice Cream

There are two main categories of ice cream, according to the methods by which it
is made and stored (Hyde, 1973). Ice cream is a frozen dairy dessert which can be
manufactured in the soft or hard form. The term soft serve is used when the product is
sold in a soft form and usually sold on the same day that it is made or as soon after as
possible. Soft serve is distinctly different from hard ice cream for many reasons.
Different procedures and equipment are used in their preparation, they are served at a
different temperatures (18-20°F for soft serve) and they are usually frozen on the

premises of the retailer (Arbuckle, 1986). With soft serve freezers, the freezer maintains

the ice cream within the freezer barrel at the correct serving temperature, and therefore
with intermittent re-fi'eezing and whipping, there is little problem of lactose crystallizing
out and giving a “rough” texture (Crowhurst, 1974). There are three main benefits of soft
serve. Firstly, it is a nutritious product, as it is low in butterfat and contains an average of
135 calories/4oz. serving (Cummings, 1985). Secondly, it is manufactured and dispensed
directly to the patron with no need for further handling or preparation. Lastly, it is highly
profitable, with a 400-500% mark-up (Casper, 1987).

Soft serve mix is commercially produced at a plant. The mix is then sold to a
retailer who then places it in a special freezer and also serves it at that location. Soft serve
mix goes through the steps of homogenization, pasteurization, and aging. After freezing,
the semisolid ice cream can be sold at retail as “soft serve” ice cream. Soft serve ice
cream mix can be processed with an ultra high temperature (UHT) processing system,
which aseptically processes the mix to avoid the cooked taste and gold color sometimes
associated with UHT milk and its products (Catalana et al. 1986).

Ice cream, whether it be in the soft or hard form, is a dairy product produced by
freezing a pasteurized mix of milk, cream, nonfat milk solids, sugars, emulsifiers and
stabilizers. The main ingredients of ice cream include whole milk, skimmed milk, cream,
condensed milk products and/or milk solid. Other ingredients such as fruits, nuts, candies
and syrups are added for flavor enrichment and for variation. Figure 1.1 illustrates the
composition of ice cream mix. Water is the major component with milk and cream being
the source of the water. The composition of soft serve ice cream mix varies between
manufacturers. A consumer panel composed of 259 males. and 250 females found that

soft serve ice cream products were widely accepted by consumers even when drastic

Stabilizers & Emulsiflers
1 %
, Butterfat
" 4%

i
l
l
l
l

  

Milk Solids

Water 18%

61%

Fig 1.1. Typical composition of soft serve ice cream mix
(MDA, soft serve facts, 2001).

variations (substituting vegetable fat for milk fat) were made (Lowenstein et a1, 1972).

A disadvantage of soft serve is that one cannot blend in any extra ingredients
during manufacture and therefore toppings must be added later to provide variety
(Cummings, 1984).

One of the sugars most widely used in mixes is high fructose corn syrup. Mixes
that utilize a high percentage of this sugar will have a lower freezing point than mixes
prepared with cane sugar. This partly explains why soft serve temperatures of 14°F and
16°F are more common now than previously.

Milk solids form the bulk of the solids found in soft serve mixes. These are
obtained from condensed skim milk or powdered skim milk. Government ruling has
allowed the use of some whey in place of skim milk solid. Whey is primarily lactose
sugar and does not add any sweetness to the mix nor does it add any body to the soft
serve product. Milk solids are primarily proteins that contribute greatly to the body and
smoothness of the soft serve product. Some lowering of the freezing point is noticeable
with high percentages of whey usage. Stabilizers are added to increase the firmness of
the frozen product. The ability of the mix to hold air is also increased by the addition of
stabilizers. Stabilizers are primarily carbohydrates derived fiom plants and include
carrageenan (Irish moss), sodium carboxymethylcellulose, locust bean gum, guar gum
and pectin. Each has different advantages and most often combinations of the above are
used. Excessive use of stabilizers will mask off-taste of the product.

By definition, soft serve has a butterfat content of less than 10 percent
(Cummings, 1982). Anything more than that is classified as “ice cream”.

Emulsifiers are added to increase the dryness and firmness of the soft serve

10

product. Emulsifiers act to combine fats with water. Commonly used emulsifiers include
mono-and diglycerides and polysorbates with polysorbate 80 being the most popular.
However recent shortages have almost eliminated its use. Polysorbate 80 was extremely
effective and replacement with other emulsifiers does not give the product equivalent
qualities. Formulations are constantly changing and freezers must be adapted for the mix
available. Temperatures of 20°F with some mixes will give the same consistency as 14°F
with other mixes. Overrun (% of volume increase to a product by the addition of air) may
also vary as well as the rapidity of product breakdown.

Ice cream production is an 8-step process which consists of pasteurization,
homogenization, aging, freezing, packaging, hardening, storage and transportation (Fig

1.2).

1.3.1. Cleaning and Sanitizing

Sanitation is the most significant concern for any operation involving frozen
desserts, especially those with dairy ingredients. Most health departments require systems
to be cleaned and sanitized daily (except for heat treatment freezers). This is necessary
for the health and safety of consumers since dairy products are susceptible to bacterial
contaminations (Bendall, 1997). Cleaning and sanitizing a machine is very time
consuming and sometimes tends to be overlooked by employees. Equipment cleaning and
set up takes an hour or more of an employee’s time each day. Consequently, automated
heat treatment systems were introduced into soft serve operations. The machine still
requires daily cleaning and sanitizing, but rather than manually disassembling the

equipment and discarding leftover mix, the heat treatment systems automatically heats

11

Fig 1.2.Process of ice cream Production (Andreasen et al., 1998)

Mixing

t

Ice Cream Mix Making

 

 

 

 

 

Pasteurization ‘( 70°C, 30min)

 

 

 

I

 

 

 

 

 

 

Homogenization ——p

 

May be distributed as
ice-cream mix to

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fiozen confections
V retailers
T v
. Aging
Soft serve ice-cream making i
. 0 Adding other ingredients
l Freezrng ('5 C) (fiuits, nuts, etc.)
J
Packaging
Hard ice-cream Hardening (-34°C)
Production

 

 

 

1

Storage (<-15°C)

l

 

 

 

 

 

 

 

Transportation (<-15°C)

 

 

12

the mix to kill the bacteria and then re-chills it (Bendall, 1997). All soft serve freezers
should be disassembled, cleaned, and sanitized periodically. Ice cream freezer
manufacturers like TaylorTM (Rockton, IL) have manuals that outline procedures that
should satisfy local laws. Deviations from the manuals should occur only when manual
procedures contradict local laws or when an insert accompanies the manual. To keep
bacterial growth to a minimum, levels of coliforms and other bacteria must be below
legal limits and sanitary conditions must be used when handling mix. Also, all
components of the freezer must be lubricated with acceptable food grade lubricants and
the parts must be sanitized with an approved sanitizing solution. Sanitary conditions must
be used when removing product from the freezer to be used as rerun.

Lapses in good practices when handling the mix could lead to problems. For
example, inadequate brush scrubbing will allow the build up of milkstone. Milkstone
harbors bacteria and contributes to counts above 50,000 cfu/ml. Improper mix handling

greatly accelerates bacterial growth.

1.3.2. Storage of Mix

The mix is received in a refi'igerated truck about twice a week (varies by
establishment). The temperature at which the mix is received may or may not be checked.
The mix is then placed in a holding refrigerator or a walk-in cooler for 1-5 days at 32-
42°F. The mixes are rotated based on dates of delivery (which is also based on dates of
production). The mix is then moved to a second holding refi'igerator (34°F). Mix is
poured into a hopper in the lower refrigerated unit of the machine with the hOpper refilled

as needed. During summer, the hopper is filled more frequently than in spring and the

13

fall. The hopper is at about 36°F. At some locations, mix is poured into metal or plastic
storage buckets (usually covered) and conveyed through tubing, to the soft serve
machine. The buckets are emptied and sanitized about twice a week, while the tubing is
cleaned and sanitized daily.

At some locations, the product is drained nightly from the machine and the
machine is washed, rinsed and sanitized each morning before use. For others with heat
treatment freezers, the external and removable parts are cleaned and sanitized each night,
while the mix remains in the hopper. Mix is kept and used as re-run at every other
complete breakdown. Complete breakdown varies depending on what type of machine
the establishment is using.

Most frozen dessert mixes are not sterile, but generally have a low bacteria count
when fresh (Cornell University, Dairy Science F acts). Storing soft serve mixes under
proper refrigeration until used is very important in maintaining quality. Body and texture
are particularly important attributes to control in frozen dairy desserts (Chandan, 197 7)
and proper temperature maintenance will do so. Soft serve freezer operators can avoid
problems with high bacteria counts and flavor defects developing in their mix from these
high counts if the mix is handled properly.

Other types of bacteria also may be present in the mix if the mix bags and other
processing equipment in the dairy plant that the mix contacts are not properly cleaned and
sanitized. Milk cans are hard to clean properly and as a result many mix manufacturers
are marketing mix in single service plastic lined corrugated boxes. Use of these boxes
greatly reduces the chance of contaminating the pasteurized mix with spoilage bacteria.

Storage of the mix at a low temperature (below 40°F) is the only practical way of

14

controlling the multiplication of spoilage bacteria. Many of these bacteria can grow in

cold mix but they grow much slower than at their usual 70° - 80°F optimum temperature.

1.3.3. Sanitizers and Cleaners

Commercial sanitizers and cleaners are used to clean and sanitize the internal and
external surfaces of various food preparation utensils and equipment. Heat is a reliable
sanitizing agent, but chemical sanitizing agents such as hypochlorites, chloramines,
iodophors, and quaternary ammonium compounds (QUATS) can be used and are very
effective under the right set of conditions (Arbuckle, 1986). Cleaners and detergents are
chemicals that are used to remove debris from food surfaces, while sanitizers are intended
to sanitize surfaces by deposition of chemical onto surfaces. A detergent is typically
combined with a chlorine compound, to produce a dual-purpose, dry chemical. When
mixed with water, in the proper ratio, a suitable 100 ppm (free chlorine) solution can be
produced. This will eliminate the need for separate cleaning and sanitizing. A proper
balance of all ingredients must be maintained to ensure easy removal of butterfat and
other fats deposited on surfaces. A good sanitizer will also prevent damage to materials in

contact, injury to the user, and dissolve easily and completely in water.

1.4. Soft Serve Freezers

A soft serve machine is a horizontal cylinder into which ice cream mix is poured
or pumped (Cummings, 1982). The first soft-serve freezers were in operation in the US
in 1930 (Hyde, 1973). Many individual brands of these freezers exist, all of which vary

by design. However, they all contain a small mix reservoir, a freezing cylinder and a

15

refrigeration unit. The detailed designs differ, but generally soft serve freezers are small-
scale horizontal continuous freezers (Rothwell, 1982). These machines are relatively
simple to operate with temperature control being an integral part of maintenance. The
product must be dispensed at 18 - 20°F (Cummings, 1984). Ifthe temperature is too
high, the product will appear soft and wet. If it is too low, the product will add pressure to
the moving parts which could result in equipment failure.

Although these freezers are very convenient and serve as the point of sale for the
soft serve product, this machine can serve as a significant source of contamination if the
machine is not adequately cleaned and maintained. As with the State of Michigan, the
Pennsylvania Department of Agriculture (PDA), and other regulatory agencies also have
been concerned about the number of licensed food service operations that make soft serve
products that do not meet quality standards.

TaylorTM Company (Rockton, IL) manufactures and distributes an extensive line
of soft serve and yogurt dispensing freezers to meet the space and volume needs of food
service operations. These freezer units come in a variety of models. They range from
Single Flavor Units to Twin Twist Units and Heat Treatment Units (Fig 1.3-1.5).

The single flavor units yield single cones, while the twin twist units allow
consumers a choice of two individual flavors or an equal combination of both with a twist
effect. The heat treatment process reduces labor and daily maintenance costs because the
freezer is able to run for two weeks without dismantling and cleaning. The machine re-
pasteurizes the mix on a daily basis. The contents of the mix hopper and the freezing

cylinder are raised to 150°F within 90 minutes during heating (Durocher, 1992).

16

 

Fig 1.3. Model PH71. Single flavor freezer. Used for low or non-fat soft serve ice
cream, yogurt, and sorbet. Has one 3.4quart freezing cylinder and one 20-quart hopper.
Features an Air/Mix Pump. Heat Treatment System. Similar to Model 8751. (Courtesy of
Taylor” Freezer, Rockton, IL)

 

 

Fig 1.4. Model PH84. Twin Twist Freezer. Used for two flavors and twist: low or non-
fat soft serve ice cream, yogurt, sorbet, and sundaes. Has two 3.4 quart freezing cylinders,
two 20-quart mix hoppers and features an Air/Mix Pump. Heat Treatment System.
Similar to Model 8754 (Courtesy of TaylorTM Freezer, Rockton, IL).

 

L|
I

Fig 1.5. Model PH85. Twin Twist Freezer. Used for two flavors and twist, low or non-fat
soft serve ice cream, yogurt, sorbet, and sundaes. Has two 3.4-quart freezing cylinders,
two 20-quart mix hoppers and features an Air/Mix Pump with a syrup rail. Heat
Treatment System (Courtesy of TaylorTM Freezer, Rockton, IL).

18

The survival of naturally occurring microorganisms in soft-serve ice cream mix
stored in a dispensing freezer equipped with a heat treatment system was examined by
Ayoung and Doores, 1995. L. monocytogenes was inoculated into the mix at 105-106
cfu/ml and samples were collected from the hopper and the spigot before and after the
daily heat treatment. The heat treatment system effectively inactivated up to 106 cfir/ml
of L. monocytogenes because the organism was not recovered in any of the hopper or
spigot samples. Results also showed that there were no apparent changes in the microbial
quality of the un-inoculated commercial mix and hence there was no increased hazard
associated with keeping the mix for two weeks in the fi'eezer unit equipped with a heat
treatment cycle. Although ice crystal diameter was affected after 9 days due to
continuous heating and agitation, texture, firmness and meltdown were not affected.

A “Heat treatment cycle” is a cycle in which the heat treatment dispensing freezer
elevates the product temperature during the heating phase to at least 150°F (655°C)
within no more than 90 minutes, maintains the product at that temperature during the
holding phase for at least 30 minutes, then cools it to of 41°F (5°C) or below within no
more than 2 h. The heat treatment freezer completes a heat treatment cycle at least once
every 24 h. The heat treatment dispensing freezer is equipped with a monitoring device
which indicates the length of time since the last heat treatment cycle, the length of time
that the most recent heat treatment cycle was at 150°F (655°C) or above, and the length
of the heating, holding and cooling phases. The freezers have a temperature probe,
accurate to plus or minus 2°F (-1°C), showing the product temperature in the hopper. The
freezer is equipped with an internal lockout device that cannot be reset without complete

disassembly of the machine. The internal lockout device mechanically shuts down the

19

heat treatment dispensing fi'eezer so that the unit is unable to dispense frozen product if
the heat treatment cycle is not properly completed, the cycle has not been completed at
least once in the preceding 24 h, or the freezer has not been disassembled for cleaning
and sanitizing within the preceding 14 days.

The freezer needs to be disassembled, cleaned and sanitized at least every 14
days, except for those parts specified by the manufacturer such as hopper covers, design
caps, door spouts, and bottoms of draw valves that need to be cleaned and sanitized daily.
The product in the hopper must be maintained at 41°F (5°C) or below during a heat
treatment cycle. A daily log should be maintained to record the cycles. All remaining
products in the fi'eezer should be discarded whenever the heat treatment dispensing
freezer is disassembled for cleaning.

The effect of the sanitation schedule on microbiological quality of soft-serve ice
cream was studied at the Louisiana Agricultural Experiment Station (White et al., 1985).
The study compared bacterial quality of soft serve mixes taken from two different types
of dispensers that recommended either daily or weekly cleaning and sanitizing. Bacterial
counts (coliform and standard plate count) from the machine that required bi-weekly
cleaning and sanitation were equal to or below those of the machine that required daily

cleaning and sanitizing.

1.5. Microbiological Problems
Soft serve mix must arrive at a temperature of 40°F (44°C) or below. The product
should also have an SPC count of less than 250 CFU per ml and less than 10

coliforms/ml. An increase in the standard plate count and coliforms can be due to a

20

number of factors, including time (shelf life), temperature, mix handling and the
condition of the storage containers.

Local codes require that machines be cleaned and sanitized at the end of each day
(except for heat treatment machines) and that the parts that come in contact with the mix
be washed, left out overnight and then rinsed with a sanitizing solution before being
reassembled (Cummings, 1984). Soft serve ice cream is rich in nutrients making it a good
medium for microbial growth. It has a pH value of approximately 6.6. Pasteurization will
eliminate these microbiological hazards. Pasteurization, which is the most commonly
applied heat treatment in the dairy industry, can destroy all pathogenic bacteria in milk.
Freezing can also inhibit the grth of any flora that was not killed during pasteurization.

There have been recent attempts to implicate frozen dairy products as vehicles of
infection in listeriosis cases. In 1986, a 21-year old mother gave birth to a premature
infant who died 5 days later. The mother had consumed ice cream sandwiches a few days
before her delivery, and although it was determined that ice cream was not the source of
infection (L. monocytogenes was isolated from amniotic fluid following the premature
birth of the baby), all suspect product was withdrawn from store shelves (Ryser et al.,
1999). Although well over 4 million gallons of frozen dairy products thought to be
contaminated with Listeria have thus far been recalled from the market at a cost in excess
of $88 million, not one case of listeriosis has been directly linked to the consumption of
frozen dairy products in the United States. The only proven case comes from Belgium
involving a 62-year old immune-compromised man (Ryser et al. 1999).

Studies have shown that the components of ice cream (fat, casein and lactose) are

protective against freeze damage to L. monocytogenes. (El-kest and Marth, 1991).

21

Palumbo and Williams (1991) also noted that the presence of food components might
partially explain increased resistance to fi'eezing. Peptides, carbohydrates, milk and
viscous products have been reported to protect bacterial cells against freezing damage
(Speck et al., 1977).

Typically, soft serve mixes are stored in refrigeration units until they are placed in
the freezer for consumption. In the meantime, the refiigerator door is constantly opened
to obtain other stored products. Martin et al. (1971) studied the effect of pasteurization
conditions, type of bacteria and storage temperature on the keeping quality of UHT-
processed soft —serve-frozen dessert mixes. Some processing plants have changed to
UHT pasteurization in an attempt to improve the shelf - life of these products (Martin et
al., 1971). Reports from industry reveal that UHT - treated products may actually have a
shorter shelf - life because it was not established that the spoilage organisms did not
survive UHT treatment (Martin et al., 1971). The results of the study showed that when
approximately one million spores/ml were present, only 12.4% were destroyed by heating
for 3 sec at 104.5°C, and 99.5% at 137.7°C. During storage at 44°C, even with more than
1 million spores/ml surviving 104.5°C, the products had a shelf life of more than 8 weeks
and the number of viable B. cereus organisms actually decreased by more than one-third
during this time. At a storage temp of 10°C, the product pasteurized at 104.5°C was
spoiled after 4 weeks, and that pasteurized at 877°C was spoiled after 5 weeks with B.
cereus numbering in the millions. When stored at 15°C, the numbers of B. cereus were in
the millions and the product was spoiled after 2 weeks regardless of the UHT treatment.

There are some steps in the production of ice cream that can promote microbial

contamination of the product. However, the potential microbiological hazards found in
\

22

the final product can still be introduced by adding contaminated ingredients to the
product and improper handling procedures (Bell et al., 1998). This is especially important
in the preparation of soft-serve ice cream as its final stage of production is carried out at
the point of sale, i.e. freezing. F lavorings are not added into the pasteurized mix, but
rather after freezing. Addition of these ingredients has the potential to contaminate the
product if they are themselves contaminated. Some pathogens that can survive in food
even at low temperature include Salmonella spp., Listeria monocytogenes, and Y ersinia
spp. (Marshall, 1998). Table 1.1 summarizes the possible hazards associated with

different stages of ice-cream production (ICMSF, 1998).

1.5.1 Coliform Bacteria

Coliform and fecal coliform are commonly used as a measure of sanitation.
These microorganisms thrive in the human intestinal tract and include some strains that
are pathogenic for infants and adults. They are facultatively anaerobic, gram-negative,
non-spore forming, rod - shaped bacteria that produce acid and gas from lactose within
48 hours at 35°C. Coliform bacteria include E. coli, Citrobacter spp., Enterobacter
aerogenes, and Klebsiella pneumoniae. Not all of these organisms are of intestinal origin,
with some species living free in the environment. They have "sanitary significance," in
that their presence at detectable levels in finished products or at higher than minimal
levels in pasteurized milk indicates unsanitary manufacturing practices. Most are not

capable of causing disease in humans.

23

Table 1.1. Hazards and typical control in the production of ice cream

 

Process

Potential Hazard

J

Control Measure

 

Raw Materials

Presence of pathogens

I

Test materials before
intake

 

Pasteurization

Survival of pathogens

I

Ensure correct
time/temp control
Equipment maintenance

 

Aging

Recontanrination

Growth of microorganisms

\\

Hygienic design,
cleaning and dis-
infection of equipment
and utensils

The temp should be
<5°C

 

Freezing

Survival of microorganisms

The product should be
at correct temp

 

Addition of ready-to-eat
ingredients

Recontamination

Purchase materials from
a reputable source
Environmental hygiene
of storage area,
equipment and utensils
Hygiene during addition
step

 

Filling in packing step

Recontamination

Hygienic design and
environmental hygiene
of equipment and
utensils

 

 

Storage & Transportation

 

Survival of microorganisms

 

Keep temp at <-18°C

Discard the defiosted
products.

 

Source: ICMSF. Microorganisms in Foods 6-Microbial Ecology of Food Commodities. p 563

24

 

Fecal coliform bacteria are a group of bacteria that are passed through the fecal
excrement of humans, livestock and wildlife. Fecal coliforms ferment lactose to acid and
gas within 48 h at 44.5 to 455°C. E. coli is one specie of fecal coliform bacteria. The
presence of fecal coliform bacteria in food indicates that it has been contaminated with
the fecal material from man or other animals. They will grow at 445°C (above body
temperature). E. coli and K. pneumoniae belong to this group, although the latter is not
necessarily of fecal origin.

The presence of E. coli in soft serve ice cream or any food product is an
indication that the food has been contaminated due to manufacturing practices conducted
under unsanitary conditions. Fecal coliforms are destroyed by pasteurization and the
presence of these microorganisms in a pasteurized product indicates that there has been
post-process contamination. Contamination with coliforms is a reliable indicator of post-

process contamination of the product, even if the source of contamination is not fecal.

1.5.2. Listeria monocytogenes

Listeria monocytogenes is a ubiquitous bacterium commonly found in the
environment and is a foodbome pathogen for humans. This pathogen is able to survive
and even grow in the food-processing environment (Eklund et al., 1995). In one study in
which 2545 samples were collected over a 7-year period from various locations
(environment, equipment, raw materials, unfinished product, and ice cream), 71 of 2545
(2.8%) samples contained L. monocytogenes (Miettinen et al., 1999). In 1994, when most

of the samples were taken, and when sampling focused on previously known L.

25

monocytogenes positive sites, the prevalence of L. monocytogenes was 3.2%. In the
equipment and environmental samples, the prevalence of L. monocytogenes was 5.1%,
and was 0.6% in raw material and ice cream samples. The 71 L. monocytogenes positive
samples obtained during 1990-1996 yielded 41 isolates for serotyping and strain-specific
typing by pulse-field gel electrophoresis (PFGE) (Miettinnen et al., 1999). The isolates
characterized were obtained from the ice cream processing environment and equipment
with six strains isolated from the finished product. The findings indicated that isolates of
L. monocytogenes PF GE type H had survived in an ice cream plant for at least 7 years.

Most strains of L. monocytogenes are pathogenic and have the ability to grow at
temperatures as low as -2°C, permitting multiplication in refrigerated foods. This
organism is also very resistant to freezing, drying and heat. Although, L. monocytogenes
is killed by pasteurization, post-process contamination can occur within the plant.
Foodstuffs associated with listeriosis epidemics have included among other foods, both
milk and dairy products with soft cheese most commonly associated with listeriosis
outbreaks.

L. monocytogenes is widespread in the environment and has been isolated from
water, soil, dust, vegetation, animal feed, feces and sewage and has been associated with
mammals and birds. Many animals, including dairy cows, can carry the bacterium in their
intestinal tract without becoming sick. Foods from which L. monocytogenes has been
isolated include unpasteurized milk, ice cream, cheeses, red meat, poultry, seafood,
vegetables and fruits. It is also a common contaminant in the dairy industry, both on the
farm and in the processing plant. On the farm, sources include improperly fermented

silage and manure.

26

Dairy processors can take steps to prevent Listeria contamination in the
processing environment by separating raw and finished product areas. Access to
processing areas should be restricted, especially to individuals such as truck drivers and
raw milk handlers. Employees should practice good sanitary practices and adequate
training and guidance should be provided for all workers. Cleaning and sanitizing
procedures are important activities that should not be delegated to any employee without
proper training.

Since 1985, enhanced surveillance of the dairy industry by the FDA has prompted
over 50 recalls of Listeria —contaminated frozen dairy products, including ice cream and
ice cream novelties (Ryser et al., 1999). Data from several environmental surveys of
dairy plants show that Listeria can be found throughout the processing facility (Nelson,
1990). In one study, nisin effectively reduced the numbers of viable Listeria in ice cream
with L. monocytogenes populations in ice cream prepared without nisin remaining
constant during 3 months (Dean et al., 1996). However, other studies contradict the
findings of Dean et. al, 1996. El-Kest and Marth, 1991, found that although fat, casein
and lactose were protective against freeze damage, a substantial reduction in the viable L.
monocytogenes population was seen throughout frozen storage.

In the United States, an estimated 2500 persons become seriously ill with
listeriosis each year, with a fatality rate of approximately 20-30% (CDC). This high
mortality rate makes listeriosis an important concern for the food industry. Since the
early 1980’s, L. monocytogenes has been implicated in several major foodbome
outbreaks in North America. In 1981, coleslaw was implicated in a Canadian outbreak

involving 41 cases and 11 deaths. In 1983, pasteurized milk was the probable vehicle for

27

an outbreak in Massachusetts, with 49 confirmed cases and 14 deaths. A third major
outbreak associated with a Mexican-style cheese occurred in 1985 in California, with 142
cases and 48 deaths (Weinstein et al., 2001). In response to the later outbreak, the USDA
and FDA established a “zero” tolerance policy for L. monocytogenes in ready-to-eat
products. However, in 1999, an outbreak was linked to hot dogs and deli meats (100
cases with 21 deaths). The recall cost the Sara Lee Corporation $76 million) and meat

sales dropped about $200 million in the six month period after the recall (USDA).

1.6. Ice cream surveys
A) Milkshakes purchased at fast food outlets (Pennsylvania Department of Agriculture)
In order to determine whether the fast food outlets were complying with set
standards, milkshakes were purchased from fast food outlets and tested for mesophilic
aerobic bacteria and coliforms (Barnard et al., 1991). Ninety-three samples were
purchased from January through June 1998, just as consumers would have purchased
them (in a container with a cover). Standard plate and coliform counts were obtained
fi'om samples within 24 hours of purchase. Between 33% and 41% of the sample did not
meet the maximum bacterial standards (<50,000 cfu/ml for SPC and <10 cfu/ml for CC)
of the Pennsylvania Department of Agriculture, as shown in Table 1.2 and 1.3.

Table 1.2. Coliform counts obtained from 93 milkshake samples.

 

 

Range Percentage
Lessthan 1 perml 38.7
1 to 10 per ml 20.4
More than 10 per ml 40.9

 

28

Table 1.3. Standard Plate counts tests done on 93 milkshake samples.

 

 

Range Percentage
Less than 1000 per ml 41.3
1000 to 10000 perml 13.8
10000 to 50000 per ml 12.1
More than 50000 per ml 32.8

 

For coliform counts, 59.1% of the samples complied with the standards while 40.9%
were in violation of the standard (Table 1.2). For the standard plate count, 32.8% of the
samples were in violation of the standard while 67.2% of the samples complied with the

standard (Table 1.3).

B) Soft Serve Ice cream

Another study was conducted by the New York State Department of Agriculture
to evaluate retail soft serve ice cream and frozen yogurt for microbial quality (Brown et
al. 1991). New York State has no regulatory requirement for routine testing of retail soft
serve ice cream and frozen yogurt. During the summer of 1990, samples were collected
from the point of sale, i.e. dispensed from the freezer, from retail operations located in
rural and major metropolitan markets of New York. The samples were evaluated for total
numbers of mesophilic aerobic bacteria and coliforms. The results indicated that there
was concern over the microbial quality of soft frozen desserts as offered for sale. Overall,
40% of the 58 ice cream samples had an SPC that exceeded the 100,000 CFU/g standard
and 52% were above the 20 CFU/g coliform standard. In addition, 24% of the 76 samples

of frozen yogurt also exceeded the coliform standard.

29

C) Soft Serve Ice-cream

A study on the bacteriological quality of soft-serve mixes and products in
Louisiana was conducted by the Louisiana Agricultural Experiment Station (Ryan et a1,
1982) over a 21-month period. The study focused on standard plate counts and coliform
counts of soft-serve mixes and products from retail outlets. A total of 252 mix and 817
frozen dessert samples were collected. Of the mixes, 89.3% contained less than 50,000
cfu/g (10.7% contained >50,000 cfir/g), 7.1% contained between 50,000 and 300,000
cfu/g, and 3.6% contained over 300,000 cfu/g. In addition, 7.5% of the mix samples had
coliform counts that exceeded 10 cfu/g. Overall, 61.4% of the soft serve product samples
contained less than 50,000 cfu/g (38.51% contained >50,000 cfir/g), 22.3% contained
between 50,000 and 300,000 cfir/g, and 16.3% contained over 300,000 cfu/g. In addition,
51.2% of the samples had a coliform count greater than 10 cfu/g (Ryan et. al, 1982).
These results indicate that the mixes had relatively low bacterial counts, whereas the
actual product, as dispensed from the machine was far more heavily contaminated.
Hence, contamination of the mix likely occurred before or during freezing (i.e.
contaminated by the freezer).

In other studies, Van der Zant and Moore (1954) examined 50 soft ice milk
samples obtained from vendors in 19 Texas cities. The results showed that 50% of the
samples contained >50,000 SPC/ml and 36% contained >10 coliforms/ml. In a similar
study, 44% of the soft —serve ice milk samples contained >50,000 SPC/ml.

In other work, F oltz and Mickelsen (1964) examined 100 vanilla milkshakes
purchased in Manhattan, Kansas and found that 61% of the samples contained >10

coliforms/ml.

30

Finally, Martin et al (1968) examined soft serve mixes and products from several
Georgia soft-serve retail outlets which were collected over three one-year periods. An
average of 21 .5% of the mix samples had an SPC >50,000cfu/ml, while 47.7% of the
soft-serve product had an SPC >50,000 cfir/ml. Overall, 27.3% and 50% of the mixes and

products contained >10cfu/ml respectively.

1.7. Ice cream recalls

The 1980’s saw a large number of recalls that were associated with ice cream. In
July of 1986, ice cream bars were recalled because of possible contamination with
Listeria spp. In August of the same year, a nationwide recall of ice cream was conducted
because of suspected contamination with L. monocytogenes. By the end of 1987, over
500 million Listeria- contaminated products were recalled, with a cost of $70 million to
the industry (MMWR 1990)

In the summer of 1994, consumption of Schwans ice cream (MMWR 1994.)
contaminated with Salmonella enteritidis sickened 224,000 people in the United States.
This was the largest outbreak of salmonellosis ever reported and was caused by
transporting the ice cream in tainted tanker trailers. This was an unusual outbreak as most
other ice cream - related food poisoning cases had been caused by homemade or
unpasteurized commercial ice cream. However Schwan's ingredients had undergone
pasteurization to kill any disease-causing microbes. It was later determined that the
delivery truck had also carried unpasteurized eggs from Minnesota to the Midwest with
this company also contracted for transporting ice cream mix to Schwan's factory. Health

regulations require that tanks be cleaned and washed after hauling liquid eggs. However,

31

the trucks were not clean and the trailers were not inspected regularly. After inspection,

the trailer was found to have cracks in its lining that could harbor bacteria.

1.7.1. Complaints in Michigan

Although none of the following cases were confirmed, during the course of
sampling, several local Michigan health departments received complaints from
consumers that were potentially related to the consumption of soft serve ice cream.

Mint #1 (District Health Deggtment #2; Five members of a family ate soft
serve ice cream at a restaurant in Hubbard Lake City, Caledonia Township in Alcona
County and experienced nausea, bloating, lightheadedness, weakness, difficulty breathing
and indigestion approximately 15 minutes after ingesting the ice cream. Some members
of the group had strawberry sundaes, while others had Oreo blizzards with vanilla soft
serve as the base in both. On another occasion a consumer ate ice cream at the same
establishment which reportedly had a “sour” taste with symptoms disappearing within
24 h. Ice cream samples were collected from the restaurant, however they were not of the
same batch that the family members consumed. No testing was done on these samples
because the incident was not considered to be a foodbome illness. The extremely short
onset time of 15 minutes suggests that the problem may have been due to chemical
poisoning, perhaps from a sanitizer. However none of the clients reported a burning
sensation within their mouths or throats. For the chemical used to sanitize the machine,
the MSDS listed burning of the mouth, throat and stomach. Also, family members spent
time together that day and might have shared other common meals. No other complaints

were noted at the hospitals or at the facility.

32

Complaint #2 (Genesee County): A consumer, her daughter and her sister ate soft
serve ice cream and approximately five hours later, began to experience symptoms of
abdominal pain and diarrhea. No follow up was done on this case, because it was not

believed to be a foodbome disease.

Complaint #3 (Livingston County Health Department): A woman and her family

had eaten soft serve ice cream at a local establishment and had later experienced some

 

symptoms of foodbome illness. The Livingston County Environmental Health Food
Service Sanitarians conducted an onsite investigation and found the establishment to be
clean and orderly. However, the chocolate soft serve supply line from the cooler to the
soft serve machine contained product at 55°F. The chocolate product was not used as
frequently as the vanilla product, thereby remaining in the line longer than vanilla. The
department recommended drawing product from this line every hour to keep the product
below 41°F. No food or stool samples were made available and all of the complainants
refused to see a physician as requested, therefore no determination could be made as to
the cause of illness pertaining to this complaint.

Cgmflaint #4 (Macomb County Health Depaitmentk A family of five (3 children
and 2 adults) ate chocolate and twist cones at a local establishment and complained of
illness approximately 2.5 hours later. The family had consumed previous meals together
and upon inspection of the establishment, no critical violations were found. There were
no power outages, no sewage back ups, and no ill employees. No leftover samples were
available fi’om either the consumer or the establishment, therefore the product was
sampled on a different day. On the day of inspection, both vanilla and chocolate ice

cream were at a temperature of 20°F and the walk- in cooler was at a 40°F. The person in

33

charge also indicated that the machines were sanitized twice a day and all disassembled
parts were washed, rinsed and sanitized in a three-compartment sink. No other complaints

were received that implicated this particular establishment.

CHAPTER 2
Microbiological assessment of soft serve desserts as a means of determining the
effectiveness of the current frozen dessert standard.
2.1. Introduction
The microbiological safety of soft serve desserts has received increased

awareness during summer and with the Michigan Department of Agriculture being
concerned about the number of licensed food service establishments that produce soft
serve ice cream from freezers that may not meet quality standards. Therefore, a study
was undertaken to determine the microbial quality of soft serve ice cream. In cooperation
with Michigan State University (MSU), the Michigan Department of Agriculture OVIDA)
and local health departments, soft serve ice cream commercially available in the State of
Michigan was examined for baseline levels of mesophilic aerobic bacteria and coliform
count as well as presence of Listeria spp. In an effort to supplement information on soft
serve ice cream as well as to regulate the sale of soft serve, a study was undertaken in
which the objectives were to (1) determine the baseline levels of mesophilic aerobic
bacteria, coliform bacteria and Listeria spp. found in soft serve desserts, (2) assess the
effectiveness of current handling practice in minimizing microbial contamination and (3)

assess the effectiveness of the current frozen dessert.

35

2.2. Materials and Methods
Images in this thesis are presented in color.

A total of 250 samples of Soft serve ice cream samples were collected once a
week over a 7 - month period, from May 13 to November 26, 2001. With the help of food
service inspectors from local MDA office, samples were dispensed directly from the soft
serve freezer into 300ml vile.

Factors such as leaking samples, improper storage temperatures and incomplete
surveys reduced the useable sample size from 250 to 227. Samples covered the range of
available soft serve freezers from retail stores and food service establishments. All of the
samples were assessed for Standard Plate Count (SPC) and Coliform Count (CC) and for
the presence of Listeria spp. Bacteriological testing procedures were done in accordance
to FDA/BAM methods.

Establishments were selected in alphabetical order from a pre-established list of
food service and retail establishments. The list was then divided into Regions (see
appendix) and sampled weekly, with Regions 1& 2 being sampled on the same day. If
the chosen establishment had a soft serve freezer, a sample was obtained from that
establishment. Ifthe establishment did not have a soft serve machine, then the next
establishment on the list that did have a machine was sampled. The number of samples
collected from each region varied with the size of the region and the availability of
establishments with that particular region.

Control tests done on mixes from dairy plants were used as a comparison to
ensure that the samples complied with the standard for mixes when delivered to the

establishment. “Plant A” (Detroit, MI) and “Plant B” (Port Huron, MI) supplied mixes to

36

most of the establishments.

Using gloves, approximately, 240 ml of product was aseptically dispensed from
the freezer into sterile 300-ml viles with twist caps. The vials were then sealed, labeled
and placed in a plastic zip lock bag. This zip-lock bag was placed in a cooler that was
filled with ice, and transported to the lab for testing. At each sampling time, a survey
questionnaire (see appendix) was completed, either with the manager or the person in
charge.

According to “Standard Methods for the Examination of Dairy Products”, soft
serve products should be maintained at 0-4.4°C and must arrive at the laboratory so that
microbiological analysis can begin within 36 hours. Samples were transported to Geagley
Laboratory (East Lansing, MI), within 24 hrs of collection and logged in by a licensed lab
technician. The time, date and temperature at which the samples were received were
recorded. To determine the temperature, a pre-cooled thermometer was inserted into the
temperature control (an extra vial of sample was collected during the day to be used as a
temperature control). The temperature control was at least half the size of the largest test
container. Samples were rejected if there was no temperature control. However, one test
sample could be used as a temperature control when necessary. Samples were also
rejected if the containers were leaking or if the temperature was deemed inadequate, as
two sets were.

Testing included standard plate counts, coliform counts and an assessment for the
presence of Listeria spp. Testing was done in accordance with FDA/BAM methods. The

test kit for Listeria spp. was supplied by BioControl Systems, Inc., Bellevue, WA).

37

a) Standard Plate Count
Soft serve mix (11g) was pipetted into 99ml of buffer solution. Thereafter, 1 ml
and 0.1 ml of the mix was pour-plated using standard methods agar. These plates were

counted after 48 hrs :3 hours of incubation at 32 i2-3°C.

b) Coliform Count
Soft serve mix (11g) was pipetted into 99ml of buffer solution. Thereafter, 3.3 ml
was pour -plated in triplicate and later overlayed using Violet Red Bile Agar (VRBA).

All plates were counted after 24 h :l: 2 h of incubation at 32 i2-3°C.

c) Listeria spp.

Soft serve ice cream (25g) was pipetted into 225 ml of modified Fraser Broth with
lithium chloride. The broth was then incubated at 28i2 hours at 30°C. Thereafter, lml of
modified broth was transferred into 9 ml of Buffered Listeria Enrichment Broth (BLEB).
The tube was then incubated at 24i2 h at 30°C. Following incubation, lml of this
enrichment was transferred into a test tube immersed in boiling water for 15 minutes to
inactivate the organism. After cooling, samples were tested along with 2 positive controls
and one blank. In each case, 100 microliter (11L) of sample and positive control were
pipetted into their respective wells (microwells were fitted into a holder). The wells were
incubated for 30 min at 35-37°C. Following incubation, the wells were washed, after
which the antibody, conjugate and substrate were added. The microwells were read at

410nm.

38

d) Data Analysis

Results from the standard plate and coliform counts were analyzed using
multivariable logistic regression with the final result deemed either acceptable or
unacceptable. Because the outcome of interest had a binomial distribution (yes or no),
logistic regression was used. In order to evaluate the effect of different risk factors
simultaneously, a multivariable logistic model was used. This program essentially
determined how each variable contributed to the results that were obtained. A chi-square
(x2) analysis yielded results based on each individual factor, whereas this method took all
factors into consideration.

Multivariable Models: Variables were included for analysis in the multivariable
logistic regression model if: (1) they demonstrated some level of association with
bacterial counts (p $0.1), and 2) had low levels of missing data (excluding variables for
time from machine cleaning to sample collection and observed employee practices). The
remaining variables were combined into a firll model. Interaction terms were generated
where interaction between two model variables were expected.

A modified backwards model building approach was used. A variable was
removed from the model based on its Wald x2 probability value (checking for p >005),
with the odds ratios of the remaining variables also checked. When removal of the
variable resulted in significant changes in odds ratios of the remaining variables (a 210%
change in the odds ratio), the presence of a confounding variable was suspected with the
original variable then remaining. This process of removal and checking was repeated

until the “final” model was obtained.

39

 

2.3. Results and Discussion

According to the Frozen Desserts Act, frozen desserts should have an SPC of
S 30,000 cfu/ml and a coliform count of S 10 cfu/ml for vanilla and 20 cfu/ml for
chocolate. Overall, 50.2% (114/227) ofthe samples had an SPC no more than 30,000
cfu/ml and therefore were in compliance with the current standard requirement. Of the
134 vanilla samples, 81 (60.4%) complied with the standard for SPC with 32 of 93
(34.4%) chocolate samples also in compliance (Figure 2.1).

Up to 70% of the soft serve samples met the current standard for Coliform

Counts, whereas 67 of 277 (29.5%) samples did not comply with the standard. (Appendix
Table 3.4). While 69.4% (93/134) of the vanilla samples and 71% (66/93) of the
chocolate samples complied with the standard, 30.6% and 29% of the vanilla and
chocolate samples, respectively were out of compliance (Figure 2.2). This means that for
vanilla, 93 of the 134 samples had a coliform count _<_ 10 cfu/ml, while 66 of the 93
chocolate samples had coliform counts S 20 cfu/ml. Listeria monocytogenes was not
isolated from any of the samples.

Overall, more vanilla samples had acceptable standard plate counts, while there
was no difference in the percentage of acceptable coliform counts between vanilla

samples and chocolate samples. (Table 2.1).

40

% of samples

Fig 2.1. Summary of Standard Plate Count (SPC) results on 227 soft
serve ice cream samples

 

70

EPE Fomfiiainf
I SPC Non-Compliant

 

 

 

Total . Vanilla Chocolate
Flavor

41

96 of samples
A
0

Figure 2.2 Summary of Coliform Count (CC) results on 227 soft serve
ice cream samples

 

 

EEEEFmBEi?’ ’
I CC Non-Compliant

 

 

 

 

Total Vanilla Chocolate
Flavor

42

Table 2.1. Acceptable bacterial levels in soft serve ice cream

 

 

 

Type Overall Accgrtable SPC (%) Acceptable CC (%)
Vanilla 134 81 (60.44) , 93 (69.40)
Chocolate 93 32 (34.41) 66 (70.96)
Fisher’s Exact p = .0012 Fisher’s Exact p = .5588
Mantel-Haenszel X2 I l 1.08* * Mantel-Haenszel X2 .. ' 40
OR. = 2.5 (1.45 - 4.31) OR. = .82 (.46 - 1.48)

Ryan et al., (1981) examined the bacteriological quality of soft serve mixes and products
in Louisiana and found that 61.4% of the samples contained < 50,000 cfu/ml (the
standard was higher at the time) and were therefore in compliance with the standard for
SPC. The slightly higher level of compliance in their study (61.4%) is likely due to the
higher standard in 1981 (50,000 cfu/ml) as compared to the present Michigan standard
(30,000 cfu/ml). Overall, 50.2 % of the product data set exceeded the standard for
coliform count (10 cfu/ml), whereas in the current study, 29.5% of the product data set
were in violation of the standard. In a similar study by Brown et al.(1991), 40% of the
samples had SPC greater than the 100,000 cfu/ml standard and 52% were above the 20

cfu/ml coliform standard.

 

*= significant at ps 0.1
** significant at ps0.05

43

Soft serve mix from two different distribution plants was tested for SPC and CC.
Soft serve mixes had low bacterial counts when they left the plant but had higher
microbial counts by the time the mix was been served as a soft serve ice cream to the

consumer (Table 2.2 and 2.3).

Table 2.2. Analysis of “Plant A” Soft Serve Mix

 

 

Date Sample SPC(cfu/ml) CC (cfu/ml)
10/24/01 Premium Vanilla Soft Serve Mix <250 <1
9/25/01 Premium Vanilla Soft Serve Mix <250 <1
8/28/01 Premium Vanilla Soft Serve Mix <250 <1

Premium Chocolate Soft Serve Mix 290 <1
7/30/01 Premium Vanilla Soft Serve Mix <250 <1
Premium Chocolate Soft Serve Mix <250 <1
6/27/01 Premium Vanilla Soft Serve Mix <250 <1
Premium Chocolate Soft Serve Mix <250 <1
5/29/01 Premium Vanilla Soft Serve Mix <250 <1
Premium Chocolate Soft Serve Mix 620 <1

 

Table 2.3. Analysis of “Plant B” Soft Serve Mix

 

 

Date Sample SPC (cfu/ml) CC (cfir/ml)
9/4/01 Vanilla Soft Serve Mix <250 <1
' Chocolate Soft Serve Mix 950 3
8/7/01 Vanilla Soft Serve Mix <250 11
7/ 17/01 Vanilla Soft Serve Mix <250 <1
Chocolate Soft Serve Mix <250 <1
5/15/01 Vanilla Soft Serve Mix <250 <1
Chocolate Soft Serve Mix <250 <1

 

The data further proves that the “brand of mix” was not a determining factor in the
microbial quality of the soft serve product (Table 2.4) since the mixes were delivered to a

wide range of establishments. Contamination likely occurred after the product was

opened and handled in the food service establishments.

Table 2.4. Bacterial levels in soft serve ice cream, by establishment

 

 

Establishment # of samples % Exceeding SPC % Exceeding CC
standard standard

A 14 71 (10/14) 43 (6/14)

B 6 83 (5/6) 50 (3/6)

C 10 50 (5/10) 40 (4/10)

D 3 33 (1/3) 0

E 4 100 (4/4) 0

F 4 100 (4/4) 50 (2/4)

G 2 100 50 (1/2)

H 34 74 (25/34) 44 (15/34)

I 10 90 (9/10) 20 (2/ 10)

J 7 29 (2/7) 14 (1/7)

K 1 100 100

L 1 0 O

M 20 60 (12/20) 35 (7/20)

N 3 67 (2/3) 0

O 2 100 50 (1/2)

P 2 0 50 (1/2)

Q 2 0 0

R 14 29 (4/ 14) 43 (6/ 14)

S 50 6 (3/50) 8 (4/50)

T 5 60 (3/5) 0

U 1 100 0

V 12 92 (11/12) 50 (6/12)

w 3 33 (1/3) 0

X 2 0 0

Y 2 100 50

Z 2 100 50

A1 4 25 (1/4) 50 (2/4)

A2 1 0 0

A3 4 75 (3/4) 100

A4 1 100 100

A5 1 0 0

 

There were significant differences in SPC based on volume processed per week, with

higher volumes associated with acceptable standard plate counts. Table 2.5 indicates that

contamination increased as the product was held in the machine longer. Residence time in

45

the machine can be determined by the volume used per week. The lower the amount used
per week, the longer the product must have been in the soft serve machine.

Table 2.5. Bacterial levels in soft serve ice cream, by volume processed/wk.

 

 

Vol/Week # of samples % Exceeding SPC % Exceeding CC

(gal) standard standard

0-10 27 81 (22/27) 44 (12/27)

11-20 34 65 (22/34) 15 (5/34)

21-30 26 46 (12/26) 38 (IO/26)

31-40 23 48 (11/23) 26 (6/23)

41-50 21 43 (9/21) 33 (7/21)

51-60 13 38 (5/13) 23 (3/13)

61-70 15 33 (5/15) 20 (3/15)

71-80 17 59 (10/17) 35 (6/17)

81-90 3 67 (2/3) 0

91-100 9 11(1/9) 0

101-110 1 0 0

111-120 2 50 (1/2) 50 (1/2)

121-130 2 0 0

131-140 2 0 0

141-150 8 63 (5/8) 63 (5/8)

191-200 6 50 (3/6) 33 (2/6)

201-210 2 50 (1/2) 100 (2/2)

221-230 1 0 0

241-250 1 0 0

271-280 1 0 0

291-300 8 50 (4/8) 13 (1/8)

321-330 1 0 0

391-400 1 O 0

591-600 3 67 (2/3) 100 (3/3)

 

This study also examined the association between the cleaning schedule and the
levels of bacteria found in the mix. Acceptable standard plate and coliform counts were
associated with decreasing numbers of times the machine was cleaned during one month

(Table 2.6).

46

Table 2.6. Acceptable bacterial levels in soft serve ice cream, by number of times
the machine is completely broken down and cleaned

 

 

 

 

Times per month Overall Acceptable SPC (%) Acceptable C Cl %)

Less than weekly 58 53 (91.38) 54 (93.10)

Weekly 43 13 (30.23) 23 (53.49)

8 - l4 times/month 70 18 (25.71) 47 (67.14)

Daily 44 21 (47.73) 28 (63.64)

Total 215 105 (48.83) 152 (70.70)
Fisher’s Exact p 30.0005 Fisher’s Exact p _<_0.0005

Mantel-Haenszel x2 = 27.08" Mantel-Haenszel x2 -—- 8.78"

These findings are in support of White et al, (1985) who examined the effect of
sanitation schedule on microbiological quality of soft-serve ice cream. Dispensers
cleaned and sanitized only on day zero (i.e. those cleaned less frequently) maintained
bacterial counts equal to, or below those of the dispenser cleaned and sanitized daily, for
an average of seven days. The more a machine is broken down and “cleaned”, the more
it’s interior components are coming in contact with hands that may not necessarily be
clean. Soft serve freezers with heat treatment capabilities require bi-weekly cleaning
because during heat treatment, the machine pasteurizes the mix on a daily basis. In this
situation, the interior of the machines has less contact with the environment. One would
expect that the more a machine is dismantled and cleaned, the cleaner it will remain.
However, in reality, one cannot ensure that cleaning is done thoroughly. The heat
treatment machine, however, reduces the amount of human contact with the inner parts of

the machine.

47

The levels of bacteria found in the soft serve samples were also analyzed, by type
of establishment. Results indicated that soft serve ice cream taken from certain
establishments had higher microbial counts than those taken from other establishments.
There were significant differences in standard plate counts by type of establishment, but
not in coliform counts. Fast food establishments had higher levels of acceptable standard
plate counts, while gas station/convenience stores had no acceptable standard plate counts
(Table 2.7).

Table 2.7. Acceptable bacterial levels in soft serve ice cream, by type of
establishment

 

 

 

 

Establishment Overall Acceptable SPC (%) Acceptable CC (%)

Convenience Store 18 3 (16.67) 10 (55.56)

Fast Food 58 50 (86.21) 51 (87.93)

Gas/Convenience 7 0 4 (57.14)

Ice Cream Shop 39 18 (46.15) 21 (53.85)

Restaurant 72 30 (41.67) 53 (73.61)

Retail Store 33 13 (393$ 20 (60.61)

Total 227 114 (50.2) 159 (70.04)
Fisher’s Exact p 5 0.0005 Fisher’s Exact p 5 0.0005

Mantel-Haenszel x2 =7.36** Mantel-Haenszel x2 = 234*

All 7 samples obtained from “Gas /Convenience” were in violation of the SPC
standard, while 57.14% of the samples violated the standard for coliforms. Overall, 83%
of the samples obtained from “Convenience Stores” (convenience stores with no gas
stations) violated the SPC standard, while 50% of the samples were in violation of the
coliform standard. Overall, fast food restaurants had the lowest standard plate and
coliform counts. This is an expected result, as it was noted that most fast food restaurants

sampled had soft serve ice cream machines with “heat treatment”. As stated earlier, these

48

machines must be completely broken down and cleaned every 14 days, whereas older
model machines that were found in, for example, gas stations, needed to be dismantled
and cleaned daily. Heat treatment machines pasteurize the mix in the hopper every night,
while employees of establishments with older machines had to break the machines apart
and either dispense the mix, or use it as re-run on the next business day. Results from the
study showed that the more the machine was broken down and cleaned, the higher the
microbial counts (Table 2.6).

Employee practices were also observed during sampling. Hand washing, use of
gloves, use of haimets and use and changing of gloves were observed. Significant
associations were seen between acceptable standard plate counts and hand washing and
changing gloves, and coliform counts and changing gloves (Table 2.8).

“Establishment S” which was sampled more frequently than the others had
significantly higher levels of acceptable standard plate counts and coliform counts (Table
2.9). In addition, “Establishment 8” used soft serve machines that had heat treatment
cycles, firrther supporting the fact that these types of machines are better at producing

product with acceptable standard plate and coliform counts.

49

Table 2.8. Acceptable bacterial levels in soft serve ice cream, by observed employee

 

 

practices
Practice N Acceptable SPC (%) Acceptable CC (%)
Washed hands 107 63 (58.88) 72 (67.29)
Did not wash 65 22 (33.85) 47 (72.31)
Fisher ’5 Exact p = .001 7 Fisher ’s Exact p = .6097
Mantel-HaenSel X 2 =10. 08 * * Mantel-HaenSel X 2 = .47
O. R. =2.80 (1.47- 5.32) O. R. =.78 (.40-1.55)
Used gloves 8O 39 (48.75) 60 (75.0)
Did not use gloves 39 20 (51.28) 28 (71.79)
Fisher ’s Exact p = .8466 Fisher ’s Exact p = .8243
Mantel-Haensel X 2 = .07 Mantel-HaenSel X 2 = .14
OR. =.90 (.42-1.94) O. R. =1.18(.50-2. 79)
Used hair covers 83 45 (54.22) 59 (71.08)
Did not cover hair 22 7 (31.82) 17 (77.27)
Fisher ’5 Exact p = .0922 Fisher’s Exact p = . 7890
Mantel-Haemel X2 = 3. 46* Mantel-HaenSel X2 = .33
O.R. =2.54(.94- 6.80 O. R. =.72 (.24-2.18)
Changed gloves 77 45 (58.44) 60 (77.92)
Did not change 65 21 (32.31) 38 (58.46)

Fisher ’5 Exact p = .0024

Mantel-Haensel X 2 = 9. 61 * *

0.12 = 2.95 (1.48- 5.8a

Fisher ’s Exact p = .01 76

Mantel-HaenSel X2 = 6. 20"

o. R. = 2.51 (1.21 - 5.21)

 

5O

Table 2.9. Acceptable bacterial levels in mix from “Establishment S” .

 

 

 

Mix S Overall Acceptable SPC (%) Acceptable CC (%)

Yes 38 36 (94.74) 35 (92.11)

No 189 76 (40.21) 124 (65.61)
Fisher’s Exact p 5 .0001 Fisher ’5 Exact p g. 001

Mantel-Haenszel X2 = 37.4 7 ** Mantel Haenszel X2 = 10. 54"
O.R. = 26. 76 (6.26-114.4fi O.R. = 6.12 (1.81 - 20.65)

Significant associations were seen between increasing levels of acceptable
standard plate counts and “Establishment S” and cleaning soft serve machines less than
once a week (Table 2.10). This means being supplied by “Establishment S” and having
your machines cleaned less than once a week would result in acceptable standard plate
counts. However, convenience stores were associated withdecreasing levels of

acceptable standard plate counts.

51

Table 2.10. Final multivariable logistic regression model for acceptable standard
plate counts, adjusting for enterprises using “Mix S” (n = 214, 105 acceptable

samples, 109 unacceptable samples)

 

 

Risk Factor Wald X’ p Odds Ratio
Point Est. 95 ”a CI.

“Mix S” used .0274 6.96 1.24 - 39.02
Ice cream shop .7230 1.22 .40 - 3.72
Restaurant .6751 1.22 .48 - 3.12
Convenience store .0467 .22 .05 - .98
Gas station/convenience storeT - - -
Vanilla sample .1065 1.75 .89 - 3.47
Machine Less than weekly < .0001 10.08 2.68 - 37.88
“mplaely Weekly .2080 1.03 .37 - 2.88
cleaned

8 - l4 times/month .0003 .57 .23 - 1.41

Daily - - -
Volume served (in units of 10) .2147 1.02 .99 - 1.06

 

T - baseline of comparison for type of establishment
Model log likelihood = 207.62; AIC score = 22 7.62

Estimated modele = 34.30 %,- LikelihoodRatio X2 = 90.32, 9 df, p < .0001

Significant associations were seen between increasing levels of acceptable aerobic

plate counts and using Mix “S” and cleaning soft serve machines less than once a week

(Table 2.11). Convenience stores were significantly associated with decreasing levels of

acceptable standard plate counts.

52

Table 2.11. Final multivariable logistic regression model for acceptable coliform
counts, adjusting for enterprises using “Mix S” (n = 215, 152 acceptable samples, 63
unacceptable samples)

 

 

Risk Factor Wald x2 p Odds Ratio
Point Est. 95 ”a CI.
Mix “S” .2362 2.44 .56 - 10.68
Convenience Store .5440 1.40 .47 - 4.15
Restaurant .0215 2.43 1.14 - 5.19
Gas station/convenience store'r - - -
Vanilla sample .0957 .57 .29 - 1.11
Mix rerun .6638 1.18 .55 - 2.54
Machine Less than weekly .0005 8.08 2.10 - 31.15
323136” Weekly .0048 .58 .21 - 1.64
8 - 14 times/month .3000 1.12 .46 - 2.73
Daily - - -
Volume served (in units of 10) .1740 .98 .94 - 1.01

 

* - baseline of comparison for type of establishment

Model log likelihood = 222.22; AIC score = 242.22
Estimated modele = 16.15 %,- LikelihoodRatio X2 = 37.86, 9 df, p < .0001

53

When data collected from previous years were compared to those collected in this
study, average SPC and CC decreased markedly from 1998 —2000 (Table 2.12 and 2.13).
Table 2.12. Summary of Standard Plate Count (SPC) test results on routine frozen

dessert samples submitted to the Michigan Department of Agriculture Laboratory
Division, 1998 through June 2000.

 

 

Year n # exceeding SPC standard (%)
1998 343 39 (11.37)

1999 345 26 (7.54)

2000 129 1 (0.78)

1998-2000 817 66 (8.08)

 

However, the number of samples collected in 2000 was significantly less than that
collected in 1998, which could reduce the chance of finding a violative sample. An
accurate comparison cannot be made with the 2001 data, as one would be comparing data
against two different standards. In 2001, 49.8% of the samples exceeded the standard.
However, the data collected in 2001 was analyzed against a lower standard than that of
2000. The data collected in 2001 were to be analyzed against the old standard of
50,000/ml, the percentage of samples exceeding the standard in 2001 would be 46.26%.

This still represents a drastic increase from the numbers seen during 1998-2000.

Table 2.13. Summary of Total Coliform Count (TCC) test results on routine frozen
dessert samples submitted to the Michigan Department of Agriculture Laboratory
Division, 1998 through J unc 2000.

 

 

YEAR 11 # exceeding TCC standard (%)
1998 415 87 (20.96)

1999 375 45 (12.00)

2000 138 13 (9.42)

1998-2000 928 145 (15.63)

 

Coliform counts decreased markedly from 1998-2000. However, the number of
samples collected in 2000 was significantly less than that collected in 1998, which could
reduce the chance of finding a violative sample. The standard for coliform counts has not
changed, therefore, an accurate comparison can be made with the data collected in
2001.The percentage of samples exceeding the standard in 2001 was 29.5%, which still
represents an increase from the numbers seen during 1998-2000.

County — wide sampling was unequal and often insufficient to ascertain microbial
differences between counties. The counties did not have the same types of establishments
and the same types of soft serve fi'eezers, therefore, an accurate conclusion cannot be
made based on county. For example, only one sample was obtained fi'om Alcona County
(Table 2.14). If the establishment happened to be taken from a fast food establishment,
based on prior information from this study, one can expect that it will meet SPC and
coliform standard. The types of establishments influenced the results seen by county. The
same reasoning would apply to counties like Cheboygan which had 100% compliance.
This does not necessarily mean that the entire county is in compliance, but rather the
single sample taken happened to be one from an establishment that complied with the

standard.

55

Table 2.14. Acceptable bacterial levels in soft serve ice cream, by County

 

 

County # of samples % exceeding SPC % exceeding CC
standard standard

Alcona l 100 100
Allegan 2 100 0
Alpena 1 100 0
Antrim 1 100 100
Barry 2 100 50
Calhoun 2 100 50
Cheboygan 1 0 0
Clinton 8 38 13
Delta 4 75 75
Dickinson 2 50 0
Eaton l 0 100
Genesee 15 40 27
Grand Traverse 6 67 33
Holland 1 0 0
Inng 28 46 21
Iron 1 0 0
Kalamazoo 17 65 35
Kalkaska 2 100 0
Kent 1 100 100
Mackinac 1 0 0
Macomb 5 40 60
Menominee 1 0 100
Oakland 41 49 32
Otsego 9 67 0
Ottawa 17 65 47
Saginaw 10 50 20
St. Clair 5 4O 20
Van Buren 1 100 100
Wayne 41 34 27

 

The same would apply to any comparison made by Region and Jurisdiction (Table

2.15 and 2.16). The different regions and jurisdictions did not have the same types of

establishments withthe same types of soft serve machines, therefore, an accurate

conclusion cannot be made on the effect the region had on levels of bacteria in soft serve

ice cream (Table 2.15).

56

Table 2.15. Acceptable bacterial levels in soft serve ice cream, by MDA Region

 

Region # of samples % exceeding SPC standard % exceeding CC standard

 

l 9 44 44
2 18 78 28
3 24 67 50
4 30 47 23
5 19 74 32
6 32 38 22
7 95 42 28

 

Table 2.16 Acceptable bacterial levels in soft serve ice cream, by Jurisdiction

 

 

Jurisdiction # of samples % exceeding SPC % exceeding CC
standard standard
LHD: Local Health 173 45 28
Department
MDA: Michigan 54 69 35
Department of
Agriculture

 

There were no visible trends associated with temperature and the levels of

bacteria found in soft serve desserts (Table 2.17).

Table 2.17. Acceptable bacterial levels in soft serve ice cream, by product
temperature

 

 

Product Temp (°F) 11 % exceeding SPC % exceeding CC
standard standard
14-16 9 44 44
17 14 57 21
18 31 42 16
19 40 53 33
20 55 38 35
21 3O 77 40
22 28 64 29
23 3 33 0
24 8 75 1 3
25 3 0 67
26-40 6 33 17

 

57

The use of rerun was also examined to see for potential impact on microbial levels
in soft serve ice cream. Samples having the highest SPC and coliform counts were
associated with those establishments that used re-run (Table 2.18). This may be due to
improper storage and handling of the re-run product.

Table 2.18. Acceptable bacterial levels in soft serve ice cream, by whether unused
mix is rerun.

 

 

Re-run used 11 % exceeding SPC % exceeding CC
standard standard
No 177 49 29
Yes 50 62 32

 

Based on the data, samples remaining in the machine for 24-48 h yielded higher

percentages of violative coliform standard plate counts (Table 2.19).

Table 2.19. Acceptable bacterial levels in soft serve ice cream, by time mix spends in
machine.

 

 

Time in machine 11 % exceeding SPC % exceeding CC
standard standard
<24I-IRS 161 47 29
24-48HRS 33 64 36
>481-IRS 33 55 30

 

Significant associations were seen between bacterial counts and the number of
days between the machine’s last cleaning and when samples were collected (Table 2.20).
Acceptable levels of standard plate counts and coliform counts were associated with

increasing numbers of days between cleaning and sample collection.

58

Table 2.20. Acceptable bacterial levels in soft serve ice cream, by number of days
between last cleaning and sample collection

 

 

 

Days 11 Acceptable SPC (%) Acceptable CC (%)
0 41 25 (60.98) 28 (68.29)

1 56 25 (44.64) 39 (69.64)

2 - 3 40 10 (25.0) 23 (57.50)

4 24 11 (45.83) 19 (79.17)

5+ 31 25 (80.65) 26 (83.87)
Total 192 96 (50.0) 135 (70.31)

 

Fisher 's Exact p 5 0. 0005 Fisher ’5 Exact p = .1491
Mantel-HaenSzel X2 = 1.62 Mantel-HaenSzel X2 = 2.13

59

2.4. Conclusions and Recommendations

2.4.1. Summary and Conclusions

The microbial quality of 227 samples of soft serve ice cream was assessed over a
7-month period (133 of those samples being vanilla, and 94 being chocolate). Overall,
50.2% of the samples complied with the standard for standard plate count. Of these,
60.4% of the vanilla samples and 34.4% of the chocolate samples complied with the
standard. Furthermore, 69.4% of the vanilla samples and 71% of the chocolate samples
complied with the standard for coliform count. Listeria spp. were not detected in any of
the samples.

Significant increases in standard plate and coliform counts of soft serve ice cream
mixes were noted after mixes originally containing low bacterial counts were dispensed
from retail outlets. A greater number of acceptable standard plate and coliform counts
were seen with heat treatment machines. Samples taken fi'om establishments with heat
treatment fi'eezers had lower standard plate and coliform counts than their counterparts.
Increased dismantling and cleaning of the machine increased microbial contamination.
This also explains why samples that had the highest number of unacceptable counts were
associated with those establishments that use re-run.

Although the findings indicate that the type of soft serve freezer used is a
determining factor in acceptability of the product, it is necessary and recommended that
regulators, manufacturers and industry focus on preventative factors as well as conduct
regular sampling to ensure that these preventative measures are effective. These
preventative factors could include, but are not limited to the training of employees in

cleaning and sanitizing equipment, as well as providing manuals for re-training and for

60,

reference. It is logical to assume that if done properly, cleaning and sanitizing will
decrease microbial levels in product. Preventative factors are especially important, as it is
not feasible for all establishments to be equipped with a soft serve machine that has a heat
treatment cycle. While the standard is set to serve as a guideline for manufacturers and
retail outlets, without regular sampling, there is no way of determining and ensuring that
the products are meeting these standards. Therefore, continuous sampling is essential in
assuring that the manufacturers and workers at retail outlets are not only self-regulating,

but are supplying products that are low in bacterial counts.

61

2.5.2 Recommendations for Future Research

1) Due to a lack of current information on soft serve products and in order to supplement
and update already existing data, State Departments of Agriculture across the country
should implement programs or add to their already existing programs, an inspection
schedule that includes soft serve products. Local governments should make food safety a
priority and take firnding of such projects into consideration when making budget

increases or cutbacks.

2) Further, more in depth sampling projects should be done that will include

3. Taking swabs from different areas of the machine, such as in the hopper and in
the spigot in order to identify the location where contamination would most likely
occur.

b. The role that economics plays in the results obtained from different counties and
regions. This is important, as business owners living in affluent counties have
access to better resources and may be able to equip their establishments with
modern soft serve freezers, such as those with heat treatment capabilities. In
addition, local governments in more affluent counties may be able to provide
better training and better inspection services to establishments in their counties.

c. An assessment of county populations and the relationship between the number of
samples collected i.e. the risk of population exposure to violative samples.

(1. Proper sampling methods that would allow adequate sampling of all counties.

62

3) Investigate the age and type of freezers and its effects on the microbial levels found in
soft serve products.
a. Set up an experimental design simply based on heat treatment freezers vs. non

heat treatment freezers.

4) A study that focuses on training and preventative measures for establishments that are

not equipped with heat treatment machines, e. g. cleaning and sanitizing strategies.

63

APPENDIX

64

Appendix 3.1. Data Tables

Table 3.1. Summary of Standard Plate Counts (Total)

 

 

Standard Plate Count (/ml) Number of Samples %
<250 53 23.3
250-30,000 61 26.9
30,001-49,999 8 3.5
50,000-99,999 22 9.69
100,000-199,999 14 6.18
200,000-299,999 5 2.20
300,000-3 99,999 1 0.44
400,000-499,999 O 0.00
500,000-600,000 4 1 .76
>600,000 59 26.00
TOTAL 227 100.00

 

Summary: 50.2% (1 14/227) of the samples complied with the standards, while 49.8%
(113/227) did not comply with the standard.

Table 3.2. Summary of Standard Plate Counts (Vanilla Samples)

 

 

Standard Plate Count (/ml) Number of Vanilla Samples %
<250 45 ' 33.58
250-30,000 36 26.87
30,001-49,999 5 3.73
50,000-99,999 12 8.96
100,000-199,999 5 3.73
200,000-299,999 2 l .49
300,000-399,999 1 0.75
400,000-499,999 O 0.00
500,000-600,000 3 2.24
>600,000 25 18.66
TOTAL 134 100.00

 

Summary: 60.4% (81/134) of the vanilla sample complied with the standard, while 39.6%
(53/134) were in violation of the standard.

65

Table 3.3. Summary of Standard Plate Counts (Chocolate Samples)

 

 

Standard Plate Count (/ml) Number of Chocolate Samples %
<250 8 8.6
250-30,000 24 25.8
30,001-49,999 3 3.23
50,000-99,999 l l 1 1.83
100,000-199,999 9 9.68
200,000-299,999 3 3.23
300,000-399,999 0 0.00
400,000-499,999 . O 0.00
500,000-600,000 0 0.00
>600,000 35 37.63
TOTAL 93 100.00

 

Summary: 34.4% (32/93) of the chocolate samples complied with the standard, while
65.6% (61/93) of the chocolate samples were in violation of the standard.

Table 3.4. Summary of Coliform Counts

 

 

Coliform Counts (lml) Number of Samples %
<1 130 57.27
1--10 19 8.37
ll--20 11 4.85
21--49 13 5.73
50--99 9 3.96
100--150 10 4.41
>150 35 15.42
TOTAL 227 100.00

 

Summary: 29.5% (67/277) of the samples were in violation of the standard, whereas 66-
70% complied with the standard (depending on whether it was chocolate or vanilla).

66

Table 3.5. Summary of Coliform Counts (Vanilla Samples)

 

 

Coliform Counts (lml) Number of Vanilla Samples %
<1 83 61.94
1--10 10 7.46
11-20 4 2.99
21--49 8 5.97
50--99 3 2.24
100--150 6 4.48
>150 20 14.93
TOTAL 134 100.00

 

Summary: 69.4% (93/134) of the vanilla samples complied with the standard while
30.6% (41/134) of the vanilla samples were in violation of the standard.

Table 3.6. Summary of Coliform Counts (Chocolate Samples)

 

 

Coliform Counts (lml) Number of Chocolate Samples %
<1 50 53.76
1--10 9 9.68
11--20 7 7.53
21--49 7 7.53
50--99 4 4.30
100--150 4 4.30
>150 12 12.90
TOTAL 93 100.00

 

Summary: 71% (66/93) of the chocolate samples complied with the standard while 29%
(27/93) of the chocolate samples were in violation of the standard.

67

Table 3.7. Acceptable bacterial levels in soft serve ice cream, by MDA regions

 

 

 

 

Region Overall Acceptable SPC (%) Acceptable CC (%)
1 9 5 (55.56) 5 (55.56)

2 18 4 (22.22) 13 (72.22)
3 24 8 (33.33) 12 (50.0)

4 30 15 (50.0) 23 (76.67)
5 19 5 (26.32) 13 (68.42)
6 32 20 (62.50) 25 (78.13)
7 95 55 (57.89) 68 (71.58)
Total 227 112 (49.34) 159 (70.04)
Fisher’s Exact p _<_ 0.0005 Fisher’s Exact p<0.0005
Mantel-Haenszel X2 = 1.68 Mantel-Haenszel X2 = 7.81"

There were no significant differences in standard plate counts by MDA region, but there
were differences in coliform counts: lower levels of acceptable coliform counts were seen

in regions 3 and 1.

*significant at p501
** significant at p.<.0.05

68

Table 3.8. Acceptable bacterial levels in soft serve ice cream, by product
temperature

 

 

Kruslrall-

11 Mean Std. Dev. Quartiles F Wallis X:
Overall 227 20.08 2.58 14, 19, 20, 21, 38
Aerobic Plate Counts
Acceptable 112 20.01 2.89 14, 18, 20, 20.5, 38 13 1 74
Unacceptable 115 20.14 2.25 15, 19,20, 21, 29 ° '
Coliforms
Acceptable 159 20.09 2.73 14, 18, 20, 21, 38 02 21
Unacceptable 68 20.04 2.19 15, 19,20, 21, 29 '

 

69

Table 3.9. Acceptable bacterial levels in soft serve ice cream, by brand of mix

 

 

 

 

Brand Overall Acceptable SPC (%) Acceptable
CC (%)
A 14 4 (28.57) 8 (57.14)
B 10 5 (50.0) 6 (60.0)
C 33 8 (24.24) 18 (54.55)
D 20 8 (40.0) 13 965.0)
E 38 36 (94.74) 35 (92.11)
F 12 1 (8.33) 7 (58.33)
G 100 50 (50.0) 72 (72.0)
Total 227 112 (49.34) 159 (70.04)
Fisher’s Exact p: 0.0005 Fisher’s Exact p 3 0.0005
Mantel-Haenszel X2 = 3.15* Mantel-Haenszel X2 = 3.15*

*significant at p301
** significant at p30.05

While not significant at pS0.05, there were differences in levels of acceptable bacterial
counts and brand of soft serve mix used. In particular, levels of acceptable bacterial
counts were very high for establishments using “Mix S”(did not include subsets of

Mix S).

70

Table 3.10. Acceptable bacterial levels in soft serve ice cream, by volume processed

 

 

per week
11 Mean Std. Dev Quartiles F Kruskall-
Wallis xz
Overall 227 74.98 95.12 2, 20, 45, 80, 600
Aerobic Plate Counts
Acceptable 112 85.53 92.65 6, 30, 58, 100, 600
274* 13.43"
Unacceptable 115 64.7 96.76 2, 15, 35, 75, 600
Coliforms
Acceptable 159 67.72 72.30 2, 20, 50, 80, 325 3.1" .01
Unacceptable 68 91.94 133.31 2, 22.5, 43.5, 80, 600

 

There were significant differences in volume processed per week by standard plate count,

with higher volumes associated with acceptable standard plate count.

Table 3.11. Acceptable bacterial levels in soft serve ice cream, by time mix spends in

 

 

 

machine
Time in machine Overall Acceptable SPC (%) Acceptable CC (%)
< 24 hours 161 85 (52.8) 115 (71.43)
24 - 48 hours 33 12 (36.36) 21 (63.64)
> 48 hours 33 15 (45.45) 23 (69.70)
Total 227 112 (49.34) 159 (70.04)

 

Fisher’s Exact p = .2133

Mantel-Haenszel X2 = 1.53

71

Fisher’s Exact p = .6329
Mantel-Haenszel X2 = .21

Table 3.12. Acceptable bacterial levels in soft serve ice cream, by whether unused
mix is rerun

 

 

Mix rerun Overall Acceptable SPC (%) Acceptable CC (%)
Yes 50 19 (38.0) 34 (68.0)
No 177 93 (52.54) 125 (70.62)

 

Fisher’s Exact p = .0791 Fisher’s Exact p = .7290
Mantel-Haenszel x2 = 3.28 Mantel-Haenszel x2 = .13
O.R. = .55 (.29 - 1.05) O.R. = .82 (.46 - 1.48)

Table 3.13. Acceptable bacterial levels in soft serve ice cream, by number of times
the machine is completely broken down and cleaned

 

 

 

Times per month Overall Acceptable SPC (%) Acceptable CC (%)

Less than weekly 58 53 (91.38) 54 (93.10)

Weekly 43 13 (30.23) 23 (53.49)

8 - l4 times/month 70 18 (25.71) 47 (67.14)

Daily 44 21 (47.73) 28 (63.64)

Total 215 112 (49.34) 152 (70.70)
Fisher’s Exact p _<_ 0.0005 Fisher’s Exact p 5 0.0005

Mantel-Haenszel x2 = 27.08“ Mantel-Haenszel x2 = 8.78"

There were significant associations between bacterial counts and the number of times the
soft serve machine was completely broken down and cleaned in a one-month period.
Acceptable levels of standard plate counts and coliform counts were associated with

decreasing numbers of times the machine was cleaned during one month.

72

Table 3.14. Acceptable bacterial levels in soft serve ice cream, by type of
sanitizer/disinfectant used

 

 

Type of Overall Acceptable SPC (%) Acceptable CC (%)
disinfectant
SDSTD 157 84 (53.50) 112 (71.34)
Other 70 28 (40.0) 47 (67.14)
Fisher’s Exact p = .0636 Fisher’s Exact p = .5337
Mantel-Haenszel X2 = 352* Mantel-Haenszel X2 = .40
O.R. = 1.73 (.97 - 3.06) O.R. = 1.22 (.66 - 2.23)

Table 3.15. Acceptable bacterial levels in soft serve ice cream, by number of days
between last cleaning and sample collection

 

 

Days Overall Acceptable SPC (%) Acceptable CC (%)
0 41 25 (60.98) 28 (68.29)

1 56 25 (44.64) 39 (69.64)

2 - 3 40 10 (25.0) 23 (57.50)

4 24 11 (45.83) 19 (79.17)

5+ 31 25 (80.65) 26 (83.87)

Total 192 96 (50) 135 (70.31)

 

Fisher’s Exact p 5 0.0005 Fisher’s Exact p = .1491
Mantel-Haenszel X2 = 1.62 Mantel-Haenszel X2 = 2.13

There were significant associations between bacterial counts and the number of days
between the machine’s last cleaning and when samples were collected. Acceptable levels
of standard plate counts and coliform counts were associated with increasing numbers of

days between cleaning and sample collection.

73

Table 3.16. Acceptable bacterial levels in soft serve ice cream, by cleaning water
temperature (Fahrenheit)

 

 

. Kruskall-

n Mean Std. Dev. Quartrles F Wallis x1
Overall 201 122.34 33.27 40, 100, 120, 150, 190
Aerobic Plate Counts
Acceptable 95 122.13 35.14 40, 100, 120, 165, 190 01 07
Unacceptable 106 122.54 31.66 75, 100, 120, 150,190 ' '
Coliforms
Acceptable 138 122.77 34.97 40, 100, 120, 155,190 07 02
Unacceptable 63 121.41 29.44 75, 100, 120, 150, 170 ' °

 

74

For both bacterial counts, two different multivariable logistic regression models were
developed: one with and one without adjustment for whether an establishment was an
“Establishment S” restaurant.

Table 3.17. Final Multivariable Logistic Regression Model for Acceptable Aerobic
Plate Counts (n = 214, 105 acceptable samples, 109 unacceptable samples)

 

 

Risk Factor Wald x2 P Odds Ratio 95% CI.
Point Est.

Fast food establishment .2408 2.11 .61 - 7.36

Restaurant .3201 1.58 .64 - 3.89

Ice cream shop .4527 1.54 .40 - 4.72

Gas station/convenience storeT - - -

Vanilla sample .0587 1.90 .98 - 3.71
Less than weekly < .0001 9.70 2.57 - 36.67

Machine Weekly .0584 .71 .25 - 2.00

completely .

cleaned 8 - l4 times/month .0003 .49 .20 - 1.20
Daily - - -

Volume served (in units of 10) .2421 1.02 .99 - 1.06

Use of sanitizer with STS’ID .5921 1.22 .59 - 2.54

 

Model log likelihood = 218.93; AIC score = 238.93
Estimated model R2 = 30.43 %; Likelihood Ratio x2 = 77.66, 9 d.f., p < .0001

l - baseline of comparison for type of establishment

Based on the model developed without adjusting for “Establishment 8’, significant
associations were seen between acceptable levels of aerobic plate counts and

establishments cleaning soft serve machines less than once a week.

75

Table 3.18. Final Multivariable Logistic Regression Model for Acceptable Coliform
Counts (n = 215, 152 acceptable samples, 63 unacceptable samples)

 

Odds Ratio
Risk Factor Wald X2 p Point Est. 95% CI.
Restaurant .0303 2.19 1.08 - 4.43
Less than weekly < .0001 8.37 2.41 - 29.13
Machine Weekly .0005 .51 .20 - 1.31
completely
cleaned 8 - 14 times/month .1730 1.01 .43 - 2.35
Daily - - -
Volume served (in units of 10) .0771 .97 .94 - 1.00

 

Model log likelihood = 226.40; AIC score = 238.40
Estimated model R2 = 14.5 %; Likelihood Ratio x2 = 33.68, 5 d.f., p < .0001

Based on the model developed without adjusting for “Establishment S”, significant
associations were seen between increasing acceptable levels of aerobic plate counts with

restaurants and establishments cleaning soft serve machines less than once a week.

76

Appendix 3.2. Definitions
Ice cream: “any frozen, sweetened milk product which is stirred during the process of
freezing and includes every such frozen milk product which contains milk fat or milk
solids not fat and which in any manner simulates the texture or characteristics of ice

cream no matter under what coined or trade name it may be sold.”

Ice cream mix: an “unfrozen food product, made from wholesome ingredients as
provided for use in the manufacturing or processing of ice cream. . . .respectively, and

shall meet the same requirements as the foods made from those mixes.”

Frozen desserts: means ice cream. . .and other new related products which are defined in

the future by rule.

Ice cream plant: any place, premises, or establishment where ice cream. . . .is

manufactured, prepared, processed, or frozen for distribution or sale.

Soft Serve: a generic term used to describe a complete category of ice cream and frozen
dessert products. Soft Serve products are fi'ozen in and dispensed directly from the
freezer to the customer. The mix used determines the type of product served. Reduced
Fat, Low Fat, Non-Fat, Fat Free Ice Creams, Ice Milk, Frozen Custard, Sherbet, and
Sorbet are some of the products consumers know and order. Soft Serve is a profitable
menu item because the food cost associated with Soft Serve is low, resulting in a 70-80%

profit margin.

77

Bacteria Count: The amount of bacteria in the soft-serve mix or the finished product. If
the bacteria count in test samples taken by the health inspector exceeds the allowable

limit, penalties can result, including suspension or total shutdown of business.

Coliform: Gram negative, non-spore forming, facultatively anaerobic rods that ferment
lactose to produce acid and gas at 35°C within 48hrs. Coliform bacteria include E. coli,

C itrobacter spp., Enterobacter aerogenes, and Klebsiella pneumoniae.

Re-Run: Mix that has been frozen and allowed to melt down for reuse. Re-run is most
commonly acquired while cleaning the freezer. Re-run should never be used during the
initial start-up. Where health codes allow, re-run should be added to the freezer during
high volume periods during the business day. There is danger that, if improperly handled,
re-run can contaminate fi'esh mix and lead to poor product quality and a high coliform

count .

Overrun: As mix is frozen, air is introduced into it to increase its volume. Overrun is the
percentage of volume that a gallon is increased when air is added. Ice cream mix has a
composition of : 63% moisture, 37% dry material (10% fat, 11.5% MSNF, 15% Sugar,

0.5% Emulsifiers and Stabilizers).

Pressurized Freezer: a freezer equipped with mix pumps that transfer mix from a
container or hopper to the freezing cylinder. As mix is pumped into these freezers, a

controlled amount of air is added to the product. These freezers use the pressure item the

78

pump to force product fi'om the cylinder when dispensing.

Twist Freezer: any soft serve freezer that has two freezing cylinders and one dispensing
head designed to dispense two flavors individually or combine the two flavors into a

single serving.

Butterfat/Milkfat Content: The percentage of dairy fat contained in the mix. A 6%
butterfat mix, for example, contains 6 pounds of nrilkfat per 100 pounds. Product richness

will normally increase as the butterfat content increases.

Hopper Models: Freezers that have mix reservoirs on the top of the machine.

Mix: A general term that refers to any unfrozen liquid mixture designed for use in the
manufacture of frozen dairy products such as soft serve. Mixes can be both dairy and no-
dairy products in either liquid or powdered form. Mixes come packaged in cartons, bags,

cans, etc.

Mix Level Indicator: A flashing light on the machine automatically tells the operator

that mix needs to be added to the mix container or hopper.

C.I.P: Clean in Place, the procedure by which sanitary pipelines or pieces of dairy

equipment are mechanically cleaned in place by circulation

79

Frozen Dairy Dessert: any frozen or partially fiozen combination of two or more

products such as milk, milk products, eggs, etc.

Dairy Plant: a place, premise, or establishment where milk or dairy products are

received or handled for the processing or manufacturing of mix and fi'ozen dairy desserts.

Heat Treatment Dispensing Freezers: This is a freezer that has a product reservoir that
processes previously pasteurized products, fi'eezes the product, dispenses frozen dairy
products, and maintains microbiological quality by elevating the temperature of the

product using heating methods that are an integral part of the dispensing fieezer.

80

Appendix 3.3. Regulations

Regulation No. 405: Frozen Desserts

Reg. 285.4051. Definitions A to F
Rule 1. (1) The definitions of terms in the act are applicable to these rules.
(2) “Act” means Act No. 298 of the Public Acts of 1968, being sections 288.321 to
288.334 of the Compiled Laws of 1948.
(3) “Department” means the State Department of Agriculture
(4) “Freezer” means mechanical equipment used to lower the temperature of a mix, while
at the same time incorporating air into the mix.

Reg.285405.2. Definitions M to T
Rule 2. (2) “Mix” means ice cream mix, ice milk mix, sherbet mix, and any other
unfrozen pasteurized liquid mixture which is to be manufactured into a frozen dessert,
including a liquid mixture intended for processing into fi'ozen confections.
(3) “Rerun” means a frozen dessert that is not placed in its final container immediately
after passing through the freezing process and is intended to be melted and reprocessed or
refrozen.
(4) “Sanitation” means the application of an effective method or substance to a clean
surface for the destruction of organisms as fast as is practicable, which treatment does not
adversely affect the equipment, the fi'ozen dessert or the health of consumers, and is

acceptable to the department.

81

285.405.14. Cleaning of equipment.

(1) Immediately before use, equipment in an ice cream plant coming in contact
with milk, dairy products, mix or fiozen desserts shall have an effective sanitizing
treatment. (2) After each use, equipment not designed for cleaning in place, shall be
disassembled and thoroughly cleaned. Dairy cleaners, wetting agents, detergents,
sanitizing agents, or other similar material may be used that will not contaminate or
adversely affect dairy products. Steel wool or metal sponges shall not be used in
cleaning dairy equipment or utensils. (3) Circulating-in-place cleaning shall be used
only on equipment and pipeline systems designed and engineered for that purpose,
installation and cleaning procedures shall be in accordance with 3-A accepted
practices for permanently installed sanitary product pipelines and cleaning systems,
and be approved by the department. An outline of the cleaning procedures to be

followed should be posted near the CIP equipment.

Reg.285.40515. Storage of equipment and supplies.
After cleaning, multi-use utensils, containers and equipment in an ice cream plant shall
be stored to drain dry, and in such a manner as not to be contaminated before usage.
Equipment stored in sanitizing solution is permissible. Caps, parchment paper, wrappers,
liners, gaskets, and single service sticks, spoons, covers and containers for frozen
desserts, mix or their ingredients, shall be purchased and stored only in sanitary tubes,
wrappings or cartons; shall be kept therein in a clean, dry place until used; and shall be

handled in a sanitary manner.

82

Reg. 285,405.16. Handling of ingredients, mix and frozen desserts.
(1) Milk, cream and milk products in fluid form received at an ice cream plant for use in
mixes, shall immediately be cooled at a temperature of 50°F or less and maintained at
that temperature until pasteurized. Mixes shall be assembled and pasteurized in an ice
cream plant. Mix which is not frozen at the plant in which it was pasteurized shall be
transported to the place of freezing in properly covered and protected containers. (2)
Spilled fi'ozen desserts and ingredients shall be discarded. Rerun shall be handled in
sanitary containers properly covered and stored, or shall be piped directly back to vats.
Rerun, which has been strained to remove nuts, fruits or other ingredients, should be re-
pasteurized. Partially full containers, if not damaged, broken or contaminated, may be
returned to the manufacturer for re-pasteurization and processing. After opening, it is
recommended that all syrups used for flavoring or topping be refiigerated at all times.
Flavoring ingredients may be added to mix after pasteurization. (3) Frozen desserts and
mix shall be packaged in commercially acceptable containers and packaging material that
will protect the quality of the contents in regular channels of trade. The packaging,
cutting, molding, dispensing, and other handling or preparation of mix or frozen desserts
and their ingredients shall be done in a sanitary manner. Plastic or rubber gloves should
be worn when handling frozen desserts for molding, cutting, or similar hand contact
work. (4) When an ice cream plant is identified by code number on containers of fi'ozen
desserts, the distributor’s name and address or the retailer’s name and address shall be

shown on the container label.

83

Reg. 285.405.17. Standards for mix, ingredients and frozen desserts.
(1) Pasteurized mix, pasteurized dairy ingredients and frozen desserts shall
comply with the following standards:

Table 2.1. Bacterial standards for frozen desserts and frozen desserts mix

 

 

 

 

 

 

 

Standard Plate Coliform Count Storage
Count Temperature
Mix $30,000/ml SIG/ml 3 45°F
Frozen desserts $30,000/ml SIO/ml" 3 32°F

 

* $20/ml for chocolate, fruit, nuts or other bulky flavored frozen desserts.

(2) When 2 of the last 4 consecutive Standard Plate Counts, Coliform Counts, or
temperatures taken on separate days exceed the limit of the standard established in the
paragraph (1), the department shall send a written notice thereof to the person concerned.
This notice shall be in effect so long as 2 of the last 4 consecutive samples exceed the
limit of the standard and shall inform the person that an additional sample will be taken
within 14 days of the receipt of such notice, but not before the lapse of three days. It shall
further state that when the standard is in violated by 3 of the last 5 bacteria counts,
coliform determinations, or cooling temperatures, the plant shall discontinue
manufacturing the specific product involved until the cause of the violation is corrected.
Court action may also be instituted. An inspector may send the establishment a “notice of

intent to suspend permit”, a “notice of seizure” or a “frozen desserts warning notice”.

Reg. 285.405.21. Personal Health.
The plant management shall take all reasonable measures and precautions to assure

the following:

 

(a) No person affected with a communicable disease or while a carrier of such
disease shall work at an ice cream plant in a capacity which brings them in
contact with the processing, handling or storage of frozen desserts, or the
containers, equipment and utensils.

(b) No person shall be employed in an ice cream plant that is suspected of having any
diseases in a communicable form or of being a carrier of such disease.

Reg. 285.405.23. Vehicles.

A vehicle ....used for the transportation of mix, frozen desserts and their
ingredients, shall be constructed and operated so as to protect its contents from heat, sun
and contamination. The vehicle shall be kept clean, and no substance capable of
contaminating mix, frozen desserts and their ingredients, shall be transported therein.
Where applicable, an ice cream plant shall provide an area for unloading vehicles that can
be maintained in a sanitary condition. This area should be surfaced with concrete or

blacktop.

85

Appendix 3.4. Map of Regional Offices

 

 

[REGION 6__-;SOUT_H_ CENTRAL]

 

Appendix 3.5. Soft Serve Machines

Survey of Sanltary Practlces

Type of Establishment: Est. ID#:

 

 

Jurisdiction: MDA 1:] LHD [:1 County: Date:

 

 

1. Brand of Mix:

 

2. What is the volume of mix used per week in the summer (peak
season) Gallons:

3. On the average, how long does one “batch” of soft serve mix
remain in the machine?

Cl Less than 24 hours
Cl Between 24-48 hours
Cl More than 48 hours

4. Is the re-run used?

 

5. Does the firm have a HACCP program in place for maintaining the
safety and quality of the soft serve mixes?

 

6. If yes, briefly describe the plan.

 

I!
E
o
E
o
in
t
o
in

 

 

 

7. Using a simple flowchart, describe the protocol for storage and
handling of the soft serve mixes including holding times and
temperatures. Please place flowchart on back.

8. Briefly describe the protocol for storage and handling of cones,
cups, spoons, etc.

 

 

 

 

 

 

 

 

 

 

87

 

1. Describe the protocol for the cleaning/sanitizing of the soft serve
machines.

Check all boxes that apply: How Often ?
DEr Once/wk Twice /wk Other
Cl Clean in Place

 

0 Complete Breakdown D E] D ——
CI Partial Breakdown E] E El

2. Are the above methods in compliance with the manufacturer's
recommendations?

3. Name and type of cleaning product used

 

4. Type of Sanitizer
Active Ingredient
Concentration of Sanitizer

 

 

 

5. Temperature of water used during cleaning

 

m
.E
1:
e
2
U

6. Are cleaning procedures readily accessible? Yes No

7. Type of Utensil Sink: Two-compartrnent D Three-compartment El

8. How/where are the cleaning materials (brushes, buckets) stored?
Do you see a problem with where the cleaning materials are
stored? Are they adequately maintained?

 

 

9. Last day the machine was cleaned prior to sampling?

10. Last day the machine was sanitized prior to sampling?

11. Water Source: Well:

 

Municipal:

 

12. Is the water source certified?

 

 

 

 

 

88

Employee Hygiene

 

Recordkeeping

 

1. What measures are taken to ensure adequate employee hygiene? Check if
observed.

Yes No Not Observed
Handwashing 1:1 1:3 1:3
Nearest Hand sink is __ feet away
Gloves 1:] 1:] Cl
Hairnets or Covers CI 1:] D
Clean Garments [:1 1:] :1

Are gloves properly changed?

 

2. How often are employees encouraged to wash their hands? Describe employee-
hand-washing procedure and indicate frequency.

After every operation Once a day _Twice a day—Other _

 

 

 

“
C
O
E
.s
3
8
III
h
0
2
.2
H
1‘
h
o .
n.
O

 

1. Make & Model of Soft Serve Machine: Record temp. Indicated on machine:

 

 

2. How many days a week does the machine run?

 

3. In what area of the establishment is the machine located?

Cl Employee service area
Cl Customer self serve

4. Number of employees operating the machine during a work shift?

 

5. Do the employees have dedicated job duties for the soft serve service?

6. Is there any other food/non-food operation near the soft serve machine that could
be a possible source of cross-contamination? Explain.

 

 

 

 

 

 

1. Are there accurate record keeping procedures in place?

 

2. What types of activities are logged? (temperature, cleaning, etc.)

 

 

 

89

 

 

 

1. How do you rate the following about this establishment’s soft serve practices?

P M A G N‘
a. Cleaning & Sanitation :1 1:1 I: 1:] Cl
C
'2 b. Employee Hygiene 1:] :1 II] :1 [:1
H
g c. Equipment use & maint. 1:1 [:1 II] CI] [:1
g (1. Compliance w/GMP’s [:1 :3 1:1 1:1 1:1
I: e. Soft Serve Mix Handling [:1 :1 1:1 Cl 1:]
'3 f. Record-keeping 1: 1:1 1:1 I: :1
fl
3 3. Knowledge of PIC 13 I: [:1 [:1 El
9; * P = Poor
5 M = Marginal
A = Adequate
G = Good

N = No observation

 

 

 

 

9O

BIBLIOGRAPHY

91

Anon. 1986. Food Chem. News. 28 (19): 31

Arbuckle, w.s. 1986. Ice cream. 4th ed. Early, R (ed.). AVI Publishing Co., Inc.
Westport, CT.

Andreasen, T.G., and Nielsen, H. 1998. Ice Cream and aerated desserts. In: The
Technology of Dairy Products, 2”d edition. Early, R (ed.). Blackie Academic &
Professional, New York.

Ayoung, C.W., and Doores, S. 1995. Survival of L. monocytogenes and other naturally
pathogenic microorganisms in a soft-serve freezer equipped with a heat treatment
system. Inst. Food. Technol. Ann Mtg. Chicago, IL. (Abstr #12B-11)

Barnard, S. E., Smeltz, R.A., Anthony, BA. and Full, NA. 1991. Comparison of
PetrifilmTM Test Kit-L and standard cultural methods for determining general
sanitation condition of fast food freezers by sampling milk shakes. Dairy Food
Environ. San. 11:138-139.

Bell, C., and Kyriakides, A. 1998. Listeria: A practical approach to the organism and its
control in foods. Blackie Academic and Professional. London.

 

Bendall, D. 1997. Soft-serve cool. Food Management. 32: 121-129.

Brown, D.P., Bandler, UK. and Wolff, RS. 1991. Evaluation of retail soft serve ice
cream and frozen yogurt for microbial quality. J. Dairy Sci. 74:141. (Abstr.)

Casper, C. 1987. Soft-serve systems. Restaurant Business. 86: 190-192.

Catalana, C., Contino, R., and LaBell, F. 1986. UHT systems produces extended shelf
life dairy products. Food Process. 47: 66-67.

Chandan, R. C. 1977. Considerations in the manufacture of frozen and soft serve yogurt.
Food Prod. Develop. 1 1 :1 18-121.

CDC. 2001. Diagnosis and management of foodbome illnesses. MMWR. 5021-69.

CDC. 1994. Emerging infectious diseases outbreak of salmonella enteriditis associated
with nationally distributed ice cream products- Minnesota, South Dakota,
Wisconsin, 1994. MMWR. 43:740-741.

CDC. 1990. Foodbome disease outbreaks, five-year summary, 1983-1987, 1990. Morb.
Mortal.Wkly.Rep. 39(SSOl): 15-23.

Crowhurst, B. 1974. Making soft-serve ice cream. Ice Cream Frozen Confect. 26: 814,
816.

92

Cummings, G. 1982. Soft serve machines. Restaurant Bus. 81 : 126

Cummings, G. 1982. Swirling profits from soft serve. Restaurant Bus. 81 :68-72.
Cummings, G. 1984. The Software scenario. Restaurant Bus. 83:156-162.
Cummings, G. 1985. Soft serve made simple. Restaurant Bus. 842158 -160.

Dean, J .P., and Zottola, EA. 1996. Use of nisin in ice cream and effect on the survival of
Listeria monocytogenes. J. Food Prot. 59:476-480.

Dietz, K. 1998. NBC News. New York: NBC-TV. 2 Aug. Detroit, MI.

Durocher, J. 1992. Hard facts on soft serve. Restaurant Bus. 91 : 192-197.

Eklund, M.W., Poysky, F.T., Paranjpye, R.W., Lashbrook, L.C., Peterson, ME, and
Pelroy, GA. 1995. Incidence and source of L. monocytogenes in cold smoked

fishery products and processing plants. J. Food Prot. 58:502-508.

El-Kest, S., and Marth, EH. 1991. Injury and death of fiozen L. monocytogenes as
affected by glycerol and milk components. J. Dairy Sci. 74: 1201-1208.

FDA/CFSAN Bad Bug Book Listeria monocytogenes
(http://vm.cfsaggov/~mow/chap6.html)

Foltz, V. D., and Mickelsen, R. 1964. Sanitary studies on malted milk shakes. J. Milk
Food Technol. 27:139-141.

Frozen Desserts. 1994. Regulation No. 405. Michigan Dept. of Agriculture.
Frozen Desserts Act, Act No. 298, Public Acts of 1968, amended 1994. MDA.

Grant, HE. 1977. 25th annual survey of frozen dessert trends. Dairy Ice Cream Field.
160:32-38.

Hansen, AP. 1982. The effect of ingredients, process temperature, and storage on the
shelf life of UHT soft serve mixes. J. Dairy Sci. 65: 64.

Hyde, K.A., and Rothwell, J. 1973. Ice cream. Edinburgh, Churchill Livingstone. UK. pp.
195-209.

ICMSF. 1998. Microorganisms in Foods 6- Microbial Ecology of Food Commodities.
Blackie Academic & Professional. London. pp.563.

93

Lowenstein, M., and Dell, W.J. 1972. Development and consumer evaluation of soft-
serve fi'ozen desserts. Food Product Dev. 5:32 - 36.

Marshall, R.T.1998. Ice Cream and Frozen Yogurt. In: Applied Dairy Microbiology.
Marth, E..,H and Steele, J .L.(eds.). Marcel Dekker, New York.

Marshall, RT. 1993. Standard Methods for the Examination of Dairy Products.
Amer.Pub.Health Assoc, New York. 16:76.

Martin, J .H., Roberts, RE, and Sheuring, J .J . 1968. Bacteriological quality of soft serve
frozen desserts. J. Milk Technol. 31:31-33.

Martin, J .H. and Blackwood, PW. 1971. Effect of pasteurization conditions, type of

bacteria, and storage temperature on the keeping quality of UHT-processed soft-
serve frozen dessert mixes. J. Milk Food Technol. 34: 256-259.

Mead, P.S., Slusker, L., and Dietz, V. 1999. Food-related illness and death in the United
States. Emerg Infect Dis. 5: 607.

Michigan Department of Agriculture, Soft Serve Facts, 2001.

Michigan Food Law of 2000 (Act No. 92 of 2000).

Miettinen, M.K., Bjorkroth, K.J., and Korkeala, H.J. 1999. Characterization of Listeria
monocytogenes from an ice cream plant by serotyping and pulsed-field get
electrophoresis. Int. J. Food Microbiol. 46:187-192.

“Milk Manufacturing law of 2001”. State of Michigan. Enrolled House Bill No. 4829.
Act No. 267, Public Acts of 2001.

National Agricultural Statistics Service, USDA, Ag Stats Board, July 3, 2002.

Nelson, J. 1990. Where are Listeria monocytogenes likely to be found in dairy plants.
Dairy Food Environ. San. 10:344-345.

Nickerson, J .T., Sinskey, A.J. 1992. Microbiology of Foods and Food Processing.
London: Elsevier. Chap 1,2,10.

Official Methods of Analysis, 14th edition, 1984, Association of Official Analytical
Chemists, Arlington, VA.

Palumbo, SA, and Williams, AC. 1991. Resistance of Listeria monocytogenes to
freezing in foods. Food Microbiol. 8:63-68.

Rothwell, J. 1982. Ice cream freezers, freezing techniques, and equipment performance.
Ice Cream Frozen Conf. 43: 531-533, 619-621, 623.

94

Ryan, J. J ., and Gough, RH. 1982. Bacteriological quality of soft serve mixes and frozen
products (ice cream, milkshakes). J. Food Prot. 452279-280

Ryser, ET, and Marth, EH, 1999. Listeria, Listeriosis, and Food Safety. 2"‘1 edition.
Marcel Dekker, Inc. New York.

Richardson, G.H. 1985. Standards Methods for the Examination of Dairy Products, 15th
edition. Amer. Public Health Assoc, Washington, DC, pp. 106-109, 214-216.

Speck, ML, and Ray, B. 1977 . Effects of freezing and storage on microorganisms in
frozen foods: A review. J. Food Prot. 40:333.

Taylor TM Corp. 2000. Manual of Operations (705 N. Blackhawk Blvd, Rockton, IL).

Tobias, J ., and Muck, G. A. 1981. Ice cream and frozen desserts. J. Dairy Sci. 64: 1077-
1086.

USDA-F SIS. 1997. “Bacteria That Cause Foodbome Illness.”
USDA. 2001. Dairy products annual summary. 1995-2002. USDA, Washington, DC.

Van der Zant, W.C., and Moore, A.V. 1955. Some factors affecting bacterial counts of
soft ice milk. J. Food Technol. 9:285-289.

Van der Zant, W.C., and Moore, WC. 1954. Organoleptic, chemical and bacteriological
examination of soft serve ice milk. Ice Cream Rev. 38:48.

Weckel, KG. 1966. A suggested sanitary standard for freezer —dispensed ice milk,
malted milk and shakes. J. Milk Food Technol. 29: 180-181.

Weinstein, KB, and Ortiz, J. 2001. Listeria monocytogenes. J. Med. 2: 2-10.

White, CH, and Bishop, J .R. 1985. The effect of sanitation schedule on microbiological
quality of soft serve ice cream. 68:261. (Abstr.)

95

     

IIIIIII

1(1111111111111111111

504 8442