University of Memphis University of Memphis University of Memphis Digital Commons University of Memphis Digital Commons Electronic Theses and Dissertations 4-28-2016 Microbiological Safety of Retail Foods Available in Low and High Microbiological Safety of Retail Foods Available in Low and High Socioeconomic Neighborhoods in Memphis Metropolitan Area Socioeconomic Neighborhoods in Memphis Metropolitan Area Daleniece Higgins Follow this and additional works at: https://digitalcommons.memphis.edu/etd Recommended Citation Recommended Citation Higgins, Daleniece, "Microbiological Safety of Retail Foods Available in Low and High Socioeconomic Neighborhoods in Memphis Metropolitan Area" (2016). Electronic Theses and Dissertations. 1334. https://digitalcommons.memphis.edu/etd/1334 This Thesis is brought to you for free and open access by University of Memphis Digital Commons. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of University of Memphis Digital Commons. For more information, please contact [email protected].
80
Embed
Microbiological Safety of Retail Foods Available in Low ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
University of Memphis University of Memphis
University of Memphis Digital Commons University of Memphis Digital Commons
Electronic Theses and Dissertations
4-28-2016
Microbiological Safety of Retail Foods Available in Low and High Microbiological Safety of Retail Foods Available in Low and High
Socioeconomic Neighborhoods in Memphis Metropolitan Area Socioeconomic Neighborhoods in Memphis Metropolitan Area
Daleniece Higgins
Follow this and additional works at: https://digitalcommons.memphis.edu/etd
Recommended Citation Recommended Citation Higgins, Daleniece, "Microbiological Safety of Retail Foods Available in Low and High Socioeconomic Neighborhoods in Memphis Metropolitan Area" (2016). Electronic Theses and Dissertations. 1334. https://digitalcommons.memphis.edu/etd/1334
This Thesis is brought to you for free and open access by University of Memphis Digital Commons. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of University of Memphis Digital Commons. For more information, please contact [email protected].
MICROBIOLOGICAL SAFETY OF RETAIL FOODS AVAILABLE IN LOW AND HIGH SOCIOECONOMIC
NEIGHBORHOODS IN MEMPHIS METROPOLITAN AREA
by
Daleniece Higgins
A Thesis
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Master of Public Health
Major: Public Health
The University of Memphis
May 2016
!!
ii!!
Copyright @ 2016 Daleniece Higgins All rights reserved
!!
iii!!
ACKNOWLEDGEMENTS
I would like to thank Dr. Pratik Banerjee, whom I am incredibly grateful to, for
the abundant amount of help he has given to help me shape the research for this project.
Dr. Banerjee has guided me through each step of the thesis work, allowing me to choose
a unique project that brings attention to food safety in Memphis, TN. I would also like to
thank Dr. Tyler Zerwekh and Dr. Chunrong Jia, whom are also on my thesis committee,
for their guidance and support throughout this project. Their thoughts and wisdom in
environmental health has helped me complete research that is both meaningful and useful
to society. I would also like to give a special thanks to Nabanita Mukherjee and Bhavin
Chauhan for all the assistance given throughout the intense lab portion of this research
project.
!!
iv!!
ABSTRACT
Retail foods available in areas with higher food insecurity and Low
Socioeconomic Status (SES) are known to be of inferior quality than High SES areas.
The purpose of this research was to assess the availability of different food choices and
evaluate the microbiological quality of foods available at retail outlets in Low SES and
High SES areas in Memphis metropolitan. Survey of Low and High SES stores, aerobic
plate count, selective plating, and multiplex polymerase chain reactions were conducted
to determine the differences in food availability, microbial load, and the microbial
composition of selected retail foods procured from Low and High SES areas. Foods from
Low SES areas were found to have higher bacterial loads and a differential microbial
composition (with an abundance of generic E. coli) as compared to food items obtained
from High SES areas. The results indicate the disparity in microbiological quality of
foods available to populations.
!!
v!!!
Table of Contents
LIST OF TABLES ............................................................................................................ vii LIST OF FIGURES ......................................................................................................... viii LIST OF ABBREVIATIONS ............................................................................................ ix CHAPTER 1-INTRODUCTION…………………………………………………………1
The Overall Objective…………………………………………………………….3 Justification of Research………………………………………………………….3
CHAPTER 2-LITERATURE REVIEW………………………………………………….5 Food Safety……………………………………………………………………….5 Food Security……………………………………………………………………..7 Economic burden of foodborne illnesses………………………………………...10 Policies and regulations on Food Safety and Security…………………………...11 Low Socioeconomic Status vs. High Socioeconomic Status…………………….13 Vulnerable Populations…………………………………………………………..15 Food Quality associated with Food Deserts……………………………………...17 Outbreaks Associated with Foodborne Illness…………………………………...20 Pathogens Associated with Foodborne Illness…………………………………...22 CHAPTER 3-MATERIALS AND METHODS…………………………………………26 Study Area and Sampling Plan…………………………………………………..26 Microbiological Analysis………………………………………………………...27 DNA Analysis by Polymerase Chain Reaction (PCR) ………………………….29 CHAPTER 4-RESULTS…………………………………………………………………33 Availability of Foods in Low-SES Areas………………………………………..33 Microbiological Quality of Food Commodities Tested………………………….35 CHAPTER 5-DISCUSSION…………………………………………………………….45 CHAPTER 6-CONCLUSIONS………………………………………………………….51 Limitations……………………………………………………………………….51 Recommendations………………………………………………………………..52 REFERENCES…………………………………………………………………………..53
!!
vi!!
APPENDIX………………………………………………………………………………70 Primer Sequence Tables………………………………………………………….70
!!
vii!!
List of Tables
Table Page 1. Recent Data of Foodborne Outbreak in the United States. ................................... 20 2. Frequency of different food commodity availability at stores in low SES areas .. 33 3. Store characteristics based on availability of different categories of foods in low
SES areas…. ......................................................................................................... 34 4. Distribution of Quantified APC in the Cabbage Samples .................................... 37 5. Distribution of Quantified APC in the Lettuce Samples ...................................... 38 6. Distribution of Quantified APC in the Ham Samples ........................................... 39 7. Distribution of Quantified APC in the Chicken Leg Samples .............................. 40 8. Prevalence of Selected Foodborne Bacteria in the Food Products Procured from
Low-SES Stores .................................................................................................... 41 9. Prevalence of Selected Foodborne Bacteria in the Food Products Procured from
High-SES Stores ................................................................................................... 41 1a. Primer sequences used in Multiplex PCR amplification of Salmonella and E. coli ................................................................................................................... .70 1b. Primer sequences used in Multiplex PCR amplification of Listeria ..................... 70
!!
viii!!
List of Figures
Figure Page 1. Map showing the sampling area and sampling points .......................................... 27 2. Aerobic Plate Count (APC) of different food commodities ................................. 36 3. Multiplex PCR Amplification profile of Salmonella and E. coli (sample group
1)… ………………………………………………………………………………42 4. Multiplex PCR Amplification profile of Salmonella and E. coli (sample group
2)… ............................................................................................................... ……42 5. Multiplex PCR Amplification profile of Salmonella and E. coli (sample group
are more readily adopted if they are beneficial to the food facility, like helping to increase
the amount of consumers or customers. Private standards have evolved in response to
regulatory developments and, more directly, consumer concerns, and as a means of
competitive positioning in markets for high-value agricultural and food products; as a
result, private standards are predominate (Henson and Reardon, 2005).
There are other regulations that use incentives to help control diseases in
vulnerable populations. Recently, policy makers are beginning to focus on populations
that are most food insecure. Regulating retail food establishments can be a powerful tool
for improving a community’s food environment, especially in low-income food deserts
(Diller and Graff, 2011). Another action policymakers are beginning to take on the
problem of limited and disparate healthy food availability – most notably in the U.S.
through a federal initiative that will bring grocery stores and other healthy food retailers
to underserved communities (Lee, 2012).
Low Socioeconomic Status vs. High Socioeconomic Status
Socioeconomic status varies based on income, occupation, and education.
Economic status is measured by income, social status is measured by education, and
work status is measured by occupation; each status is considered as an indicator, related
but not overlapping (CDC, 2014). In comparison to High Socioeconomic Status (SES),
individuals that live in Low SES areas are characterized with lifestyles that have more
health risks with poor health outcomes, like chronic diseases. In the United States, several
!!
14!
studies have shown that low-SES and minority groups have a higher prevalence of
obesity (Wang and Zhang, 2006). Low SES is associated with a multiple negative
outcomes for children and adults. Studies in a variety of industrial countries have shown
that lower SES is generally associated with higher rates of smoking, poorer dietary
habits, lower levels of physical activity, and higher prevalence of psychosocial
orientations that are related to poor health outcomes (Lynch et al., 1997). In children,
there are higher rates of chronic illnesses, vision and hearing problems, injury, and acute
illnesses and in adults, greater rates of morbidity and mortality, including cardiovascular
disease, hypertension, osteoarthritis, asthma, and cancer (Hanson and Chen, 2007).
Compared to low SES, individuals that have a high SES have less health problems
and lead healthier lifestyles and lives. Individuals with lower SES report greater exposure
to stressful life events and greater impact of these events on their lives than do
individuals with a higher SES, and this relationship between SES and health begins at the
earliest stages of life (Lupien et al., 2000). People from lower SES groups may
experience more distress and poorer health outcomes because they lack the ability to
purchase goods or services that reduce stress, minimize sources of stress, or that can be
used to prevent or treat illness (Baum et al., 1999). As another comparison, low SES
individuals live in worse physical environments, disproportionately located near
highways, industrial areas, and toxic waste sites, since land there is cheaper and
resistance to polluting industries, less visible, and housing quality is poorer; this results in
six fold increases in rates of high blood lead levels (Adler and Newman, 2002). SES can
affect individuals from childhood to adulthood. Adult and childhood SES are correlated;
for example, those with college educated and relatively wealthy parents are more likely to
!!
15!
have access to educational opportunities and to higher status, well-paying careers (Cohen
et al., 2010).
Vulnerable Populations
A defining contradiction of the American food and agriculture system has been
the persistence of hunger despite the world’s most productive agriculture (Allen, 1999).
There are still many individuals in the United States that do not live in environments
where healthy food is accessible and affordable. One strategy used to decrease the
amount of hunger in society is through food stamps. The Supplemental Nutrition
Assistance Program (SNAP), formerly the Food Stamp Program (FSP), is the largest of
the 15 federal nutrition-assistance programs; it aims to alleviate hunger among poor by
providing benefits to purchase nutritious food items (Leung et al., 2012). Disparities exist
across different neighborhoods in terms of access to healthy or higher quality foods; these
disparities put certain communities at higher risk for illnesses (Lewis et al., 2011). When
taking into account the accessibility and availability to healthy food items of good quality
and nutritious values, one must consider what groups of people are the most affected. In
2011, 14.9 percent (17.9 million households) of minority American households
experience food insecurity at times during the year, meaning that their access to adequate
food is limited by a lack of money and other resources (Coleman-Jensen et al., 2014).
Many individuals that live in low-income areas are more susceptible to a lifestyle
resulting bad dietary habits and chronic disease, like obesity. Many children have high
rates of obesity due to the location of their households in areas of food insecurity. Obesity
among those aged 2–19 years increased steadily from 14% in 2000 to 17% in 2008.
Considering these facts, White House Childhood Obesity Task Force proposed to
!!
16!
increase the number of supermarkets in order to reduce childhood obesity (An and Sturm,
2012). This suggests that children living in poverty may retain the same life styles, bad
diets and no physical exercise, resulting drastic effects on their health in adulthood.
Although poverty in the United States rarely leads to clinical manifestations of
malnutrition, poverty was a significant predictor of hunger and food insecurity; adults
from low-income families were more likely to be overweight than other adults
(Townsend et al., 2001).
When compared with other populations in the United States, African Americans
tend to have diets of poorer quality such as being lower in fruits, vegetables, milk and
whole grain products (Sharma et al., 2009). Therefore, determining what group of people
are the most effected will help to set interventions to increase food quality throughout all
populations. Studies have shown that neighborhoods with a higher proportion of African
American residents have fewer supermarkets and fewer high-quality food options, as well
as a disproportionate number of fast food restaurants (Lewis et al., 2011). Residents of
African American and low-income neighborhoods tend to face more environmental
barriers to healthy eating than residents of other neighborhoods (Zenk et al., 2011). From
National Health and Nutrition Examination Survey (NHANES) 2003–2004 data, African
American adults had one of the highest prevalence rates of obesity (45.0% had a body
mass index > 30 kg/m2) and extreme obesity (10.5% had a body mass index > 40 kg/m2),
due to higher rates of obesity related chronic diseases (Suratkar et al., 2010). These bad
diets lead to higher rates of chronic diseases, not just in African Americans, but also in
other races as well. A large gap exists in the health status of American Indians compared
with Caucasians and other races, for example mortality from cardiovascular disease was
!!
17!
195.9 per 100 000 for American Indians/Alaska Natives compared with 159.1 and 166.1
for the Caucasian population and other races, respectively (Sharma et al., 2007).
Food Quality Associated with Food Deserts
“Food deserts” is a term that first originated in Scotland in the early 1990s and
was used to describe poor access to an affordable and healthy diet (Beaulac et al., 2009).
During the last two decades, several studies have reported increasing challenges in access
to quality food commodities for the populations living in inner-city areas in major
metropolitans (Walker et al., 2010; Morland et al., 2002; Dubowitz et al., 2015). During
this period, in the U.S., several major grocery stores had moved away from inner cities
resulting in expansion or creation of food deserts (Cummins and Macintyre, 2002).
Industrialization and globalization have dramatically changed the American food system
over the past century, and consolidation of the retail food industry has left some rural and
inner-city areas with inadequate food resources (Smith and Morton, 2009). Having access
to healthy foods (e.g., foods low in sugar, such as fresh produce) in these areas has
become a greater issue in public health. Without accessibility to healthy food items at
affordable prices, the food items bought in the stores may adversely affect dietary intake
and eventually lead to nutrition related negative health outcomes such as obesity,
diabetes, and cardiovascular diseases (Martin et al., 2014).
Many of the retail food stores in a food desert area are located in areas with high
concentrations of poor residents. Low-income and populations of color appear to be at
particular risk of living in poor food environments and bear much of the burden of
chronic disease (Gittelsohn and Sharma, 2009). Food desserts have very limited
nutritional resources, which makes it difficult for residents to sustain any effort to eat a
!!
18!
healthy diet (Lewis et al., 2011). Consumers in these areas are more likely to stock
energy dense foods that do not hold nutritious value because it is cheaper. These residents
are likely to stock foods that are of lesser quality, i.e., full of empty calories, high in
carbohydrates, sugar, fats, and sodium, but are more effective at filling up the family
(Hendrickson et al., 2006). Many individuals in these areas cannot afford to buy healthy
food items since the relative costs of fruit and vegetables have increased greatly
compared with the prices of refined grains and sugar (Gordon-Larsen, 2014).
There have been many initiatives to increase quality of the food in areas of food
deserts. The most promising efforts for the metropolitan areas have been to improve
healthy food access through corner and convenience stores (Larson, 2013). The majority
of corner-store shoppers report shopping every day and purchase significantly more
unhealthy food than supermarket shoppers. This makes store interventions important to
promote healthy food in corner stores and encourage corner-store owners to stock
healthier food items (D'Angelo et al., 2011). There are multiple reasons why store owners
do not sell healthy food items, this may include any of the factors ranging from the
refrigerators not being able to hold the correct temperatures or the owners not thinking
the population will buy the healthy foods because it is more expensive. Previous studies
have shown that by partnering with corner stores, primary sources of food in
neighborhoods lacking comprehensive supermarkets are able to greatly increase food
quality in food deserts (Langellier et al., 2013).
Intervening with corner stores could also help to better neighborhood community.
Corner stores are a predominant food source in low-income urban communities and are
frequently characterized by less availability of healthy foods, higher prices, and often
!!
19!
tense relationships with community residents (Song et al., 2011). It is also found that
more individuals eat the healthy foods sold in the area where they live (Sharkey et al.,
2010). Incentives are available to encourage a partnership with the government to help
communities that are located in food deserts. Examples of these types of incentives are
centered on financial help from the federal government, like taxes, training and technical
assistance in community development, grants, or low interest financing; each of these
incentives have goals to improve labor market opportunities, housing options, and
spurring development in low-income areas (Ver Ploeg, 2010).
There have been multiple strategies used in low-income areas to improve the
health of the populations through changing access to healthy eating. Because of the
importance of healthful nutrition to large populations, population-based interventions are
necessary (Glanz and Yaroch, 2004). The three main strategies used in food store
interventions are: creating supermarkets in areas where none currently exists, upgrading
the facilities of existing small stores to enable them to carry fresh produce and a wider
range of healthy foods, and increasing the availability of healthy food options at small
stores using existing facilities (Gittelsohn et al., 2010). Many public health officials work
together to create standards, programs, or policies to incorporate the interventions
appropriate for a specific environment. An example of one such programs used to create
interventions in deserving populations is the Fruit and Vegetable Environment, Policy,
and Pricing Workshop sponsored by the Centers for Disease Control and Prevention
(CDC) and the American Cancer Society (ACS). The workshop aimed to identify types
of interventions, specific programs that may be ready for national dissemination, and
!!
20!
research needs related to environmental, policy, and pricing strategies to promote greater
consumption of fruits and vegetables (Glanz and Yaroch, 2004).
Outbreaks Associated with Foodborne Illness
Foodborne diseases are a major cause of illness and death in the United States
(Scallan et al., 2011b). Estimating the burden of foodborne disease is complicated by the
fact that very few illnesses can be definitively linked to food (Flint et al., 2005).
Moreover, a recent CDC estimate reveals that majority of hospitalization and deaths due
to foodborne illness in the US occur to due to unspecified agents transmitted through
food (Table 1).
Table 1. Recent Data of Foodborne Outbreak in the United States. Estimated annual number of domestically acquired, foodborne illnesses, hospitalizations, and deaths due to 31 pathogens and unspecified agents transmitted through food, United States (CDC, 2011). Foodborne Agents
Estimated annual number of illnesses (90% credible interval)
%
Estimated annual number of hospitalizations (90% credible interval)
%
Estimated annual number of deaths (90% credible interval)
%
31 known pathogens !
9.4 million (6.6–12.7 million) !
20 55,961 (39,534–75,741)
44 1,351 (712–2,268)
44
Unspecified agents
38.4 million (19.8–61.2 million)
80 71,878 (9,924–157,340)
56 1,686 (369–3,338)
56
Total 47.8 million (28.7–71.1 million)
100 127,839 (62,529–215,562)
100
3,037 (1,492–4,983)
100
!!
21!
Many individuals consume fresh fruits and vegetables, which are important
components of a healthy and balanced diet, providing important vitamins, minerals, and
phytonutrients that lead to healthy living. However, foodborne illnesses and outbreaks in
the United States linked to fresh produce increased from 4% in the 1970s to 6% in the
1990s (Lynch et al., 2009). These millions of cases of foodborne illness that occur cause
an economic impact of $6.5 billion to $34.9 million annually (Finch and Daniel, 2005). In
a recent survey of outbreaks with an identified food source, produce outbreaks accounted
for 13% (713/5,416) of outbreaks and 21% (34,049/161,089) of associated illnesses from
1990 through 2005 (Dewaal and Bhuiya, 2007).
As a result of promotion by numerous companies, presently, more individuals are
eating healthier food items, like lettuce, cabbage, and deli meat. More individuals are
taking their lunches, making their own meals, and using fresh produce. As a result, the
per capita consumption of fresh produce has increased in the United States in recent
years. This has caused an increased risk for human illness associated with pathogenic
bacteria, mycotoxigenic molds, viruses, and parasites (Beuchat and Ryu, 1997). The
increased risk can be from a number of things, such as farmers using manure or untreated
water.
Estimates of the overall number of episodes of foodborne illness are helpful for
allocating resources and prioritizing interventions, but can be difficult to compute
because of the many different agents, food proportion that disease is transmitted through
differs by pathogen, and only a small proportion of illnesses are reported and confirmed
(Scallan et al., 2011a). Recently, in 2006, four separate outbreaks of foodborne illness
associated with the consumption of fresh produce occurred in the United States, with
!!
22!
lettuce being one of the main vehicles of the outbreak (Doyle and Erickson, 2008).
Outbreaks in food can occur in different manners, from an employee not following the
correct protocol of keeping their station clean to a refrigerator not holding food at the
correct temperatures. It is in these instances that pathogens grow. Pathogens can also
attach themselves on the surface of fruits and vegetables or through cuts or crevices on
the produce (Montgomery and Banerjee, 2015). Consumers trust that the food they buy is
free from contamination, and that what they are eating will not make them sick.
Pathogens Associated with Foodborne Illness
Outbreaks of foodborne illnesses often occur from pathogens. In the United States
alone over nine million foodborne illnesses occur from major pathogens each year
(Painter et al., 2013). In many cases pathogens are not identified because of delayed or
incomplete laboratory investigation, inadequate laboratory capacity, or inability to
recognize a pathogen as a cause of foodborne disease (Lynch et al., 2006). The pathogens
that are identified show significant results in foodborne illnesses and death. The
identified pathogens account for an estimated 14 million illnesses, 60, 000
hospitalizations, and 1,800 deaths (Mead et al., 1999). U.S. Food and Drug
Administration (FDA) researchers estimate that 1 to 3 percent of all foodborne illness
cases later develop secondary illnesses or complications that can occur in any part of the
body, including the nerves, joints, and heart, which can be chronic or cause premature
death (Buzby and Roberts, 1996). Since there is underreporting in foodborne illnesses,
these numbers are actually less than the amount of individuals that are affected by
contaminated food. Although there are many pathogens, there are only a few that
predominate in food. In a recent study by the CDC, the results show that the
!!
23!
predominating pathogens in foodborne outbreaks are Salmonella, Shiga toxin-producing
E. coli (STEC), and Listeria monocytogenes (Crowe et al., 2014). It is important to study
these pathogens to determine their prevalence in daily store bought items.
Salmonella. Salmonella is a pathogen that can cause diarrhea, abdominal cramps,
fever, and nausea. The serotypes of Salmonella that are most common in outbreaks are
Enteritidis, Typhimurium, Newport, and Javiana (CDC, 2015b). There have been
numerous outbreaks caused by Salmonella in recent years. In 2008, Salmonella was
diagnosed in 1407 persons in 43 states, the District of Columbia, and Canada. Ultimately,
282 patients were hospitalized, and 2 elderly patients died (Maki, 2009). Annually in the
United States, the CDC estimates that approximately 1.2 million illnesses and 450 deaths
occur due to non-typhoidal Salmonella (Crowe et al., 2014).
Many Salmonella infections are caused by undercooked shell eggs, which may be
contaminated by hens infected by Salmonella serotype Enteritidis, one of the most
common Salmonella strains (Frenzen et al., 1999). In most cases individuals eat
contaminated meat or poultry to contract food poisoning from pathogenic Salmonella.
Salmonella and other pathogens can be commonly found in wastes such as human
sewage, farm effluents, poultry litter, and other types of materials containing fecal matter
(Santos et al., 2005). In recent studies it has become a concern of the prevalence of the
contamination in retail meats (Zhao et al., 2001). In other studies Salmonella outbreaks
have been associated with consumption of celery, watercress, watermelon, lettuce,
cabbage, and raw salad vegetables (Wells and Butterfield, 1997). Salmonella being found
in RTE items, like fruits and vegetable, is a major concern. Since individuals are not
cooking these items, so there is no intervention measure. This also raises a concern for
!!
24!
rates of foodborne illnesses increasing in the summer months if the pathogen is being
found in fruits and vegetables.
Listeria monocytogenes. Listeria monocytogenes is a highly virulent pathogen that
is most commonly found in ready-to-eat (RTE) foods, deli meats, processing plants, and
dairy products, like soft cheeses (Lund, 2015). It can cause a serious infection called
Listeriosis, which results in headaches, stiff neck, confusion, loss of balance, and
convulsions, in addition to fever and muscle aches (CDC, 2016a). Listeriosis is rare when
compared to other food-borne infections, but it has the high mortality rate of 15 to 40%,
which causes great concern (Guenther et al., 2009). Pregnant women, the unborn,
newborns, the elderly and immunocompromised people are most commonly affected by
Listeriosis (Gillespie et al., 2010).
Listeria monocytogenes is mostly found in RTE items. When consumers buy
these items they are at risk for a harmful infection unless they undertake preventative
measures. Listeria monocytogenes is different from most known foodborne pathogens, in
that it is ubiquitous, resistant to diverse environmental conditions including low pH and
high NaCl concentrations, and can grow in refrigerators (Rocourt et al., 2003). There
should be more precautions put in place for foods where the pathogen is found to reduce
the incidence of foodborne infections caused by Listeria monocytogenes. Some measures
that have been put in place include post-packaging decontamination methods, such as in-
package thermal pasteurization and irradiation, and formulating meat products with
antimicrobial additives. These measures are common approaches in controlling the
incidence of the pathogen in RTE meat (Zhu et al., 2005). These measures are helpful,
but not infallible.
!!
25!
Escherichia coli O157:H7. E. coli O157:H7 was first discovered in 1982 in an
investigation by the Centers for Disease Control and Prevention (CDC) of two outbreaks
of severe bloody diarrhea. These outbreaks identified of a strain of Escherichia coli, one
that expressed O- antigen 157 and H-antigen 7, which had not previously been
recognized as a pathogen (Armstrong et al., 1996). It still was not broadly recognized
until a large multistate E. coli O157 outbreak linked to undercooked ground beef patties
sold from a fast-food restaurant chain happened in 1993 (Rangel et al., 2005). Since that
time, E. coli O157:H7 has caused an estimated 73,000 cases of infection and 61 deaths in
the United States each year. These cases often result in an infection often leads to bloody
diarrhea, and vomiting (CDC, 2016b).
E. coli O157:H7 is also known to be responsible for severe cases of hemorrhagic
colitis (HC) and hemolytic-uremic syndrome (HUS), kidney failure, around the world
(Rump et al., 2015). It is known to use ground beef as a vehicle, but has recently been
found in produce like RTE salads. In September 2006, a total of 183 persons were
infected with the outbreak strain of E. coli O157:H7 from fresh Spinach (Sep, 2006).
Another example of this pathogen causing an infection in many individuals happened in
November 2013. A multistate outbreak occurred where numerous ready-to-eat salads and
sandwich wrap products that may have been contaminated with E. coli O157:H7 were
recalled (CDC, 2013). Other foods that have been increasingly associated with this
pathogen include water, vegetables, cantaloupe, and apple cider (Ackers et al., 1998).
Since this pathogen affects a diverse array of foods, it can be hard to detect the
contamination. That contamination can lead to a wide distribution of individuals affected
by E. coli O157:H7.
!!
26!
CHAPTER 3
MATERIALS AND METHODS
Study Area and Sampling Plan
This study was performed in Memphis metropolitan area in Shelby County,
Tennessee, USA. Data was collected during a four-month period from July 2015 to
October 2015. Data from the Shelby County Health Department (SCHD) was used to
create an initial list of the stores in Shelby County, TN.
Retail Store Selection and Survey of availability of food commodities. The Food Access
Research Atlas of the Economic Research Service (ERS), U.S. Department of Agriculture
(USDA) was utilized for selection of the stores. To designate “Food deserts” the ERS
Food Access Research Atlas (www.ers.usda.gov/data-products/food-access-research-
atlas.aspx) maps census tracts that are both low income (li) and low access (la). By using
the ERS map, stores in Memphis metropolitan area was selected which fulfilled the
criteria of food access indicators for census tracts using ½-mile and 1-mile demarcations
to the nearest supermarket (for urban areas). The selected stores (Figure 1) were visually
surveyed to list the food items sold.
Sampling plan. Based on the initial store survey, twelve stores were chosen in the Low
SES area and ten stores were chosen in the High SES area for food commodity sampling.
The selection criteria for the store for sampling were based on a) the store must be
previously or currently evaluated by SCHD, and b) the stores must sell at least two of the
following food items (including at least one produce): deli meat, cabbage, lettuce, and
chicken legs. The food items were selected based on availability in both convenience
stores and supermarkets (stratified random sampling). A total of 200 samples were
!!
27!
collected. The location of each store was noted and the food samples brought back to the
lab for preparation within 24 to 48 hours.
Figure 1. Map showing the sampling area and sampling points. The stores in Low-SES areas is denoted by (■); while the High-SES area stores are marked by (■). The background colors are indicative of food security measures. Data Source: Food Access Research Atlas, USDA-ERS, (can be accessed at: http://www.ers.usda.gov/data-products/food-access-research-atlas/go-to-the-atlas.aspx (as of 3/13/2016). Microbiological Analysis
Sample Preparation. The procured samples were kept in a laboratory refrigerator (5°C)
prior to processing. All samples were processed within one day following the Food and
Drug Administration Bacteriological Analytical Manual (FDA BAM) methods. A 25g
portion of sample was placed into a stomacher bag containing 225 ml of appropriate
!!
28!
broth. Both non-selective and selective broths were used. The non-selective broth used
was Brain-Heart Infusion (BHI). The selective broths that were used are: Escherichia coli
(EC) broth for E. coli, Buffered Listeria Enrichment Broth (BLEB) for Listeria, and
Rappaport-Vassiliadis (RV) for Salmonella. Each food sample had a stomacher bag for
all four broths. The stomacher bag holding the broth was then placed into a stomacher
machine for one minute to thoroughly mix the broths and the food samples together. At
this point a 10 ml sample was withdrawn from stomacher bag containing the BHI broth
for the aerobic plate count analysis. The bags were then placed in an incubator overnight,
18 to 24 hours, at 35oC.
Aerobic Plate Count (APC). For APC, 10 ml samples from BHI broth was vortexed for
mixing. A 100 µl aliquot of the appropriate dilutions (in duplicate) of the solution was
plated on BHI agar (in duplicate for each dilution) for non-selective enumeration of APC.
The APC was calculated based on a formula (FDA-BAM):
where: N = Number of colonies per ml or g of product ∑ C = Sum of all colonies on all plates counted n1 = Number of plates in first dilution counted n2 = Number of plates in second dilution counted d = Dilution from which the first counts were obtained
!!
29!
Selective Enrichment. Overnight growth from the sample broths was used for microbial
analysis. A 10ml portion of broth from the overnight growth was placed into a 15ml
falcon tube. The rest of the broth was put into autoclaved 500ml jar. There was one
500ml jar for each sample to make a mixture of broths. After broths were taken from all
samples, the 500ml jar was set aside to use for DNA extraction. The broth filled falcon
tubes were used to make dilutions, plating, and streaking of plates. The selective isolation
was done in the following way: EC to McConkey’s agar, the BLEB to PALCAM agar,
and the RV to XLD agar. Both the streaked plates and the plates holding the dilutions
were placed in an incubator overnight, 18 to 24 hours, at 37 oC. After the overnight
growth occurred, the bacterial counts for the BHI plates were noted. It was noted
positive, if there is a colony and negative, if there are no colonies, for the selective plates.
DNA Analysis by Polymerase Chain Reaction (PCR)
DNA Extraction. A microbial DNA extraction kit, MO-Bio Ultra Clean Microbial DNA
Isolation Kit, was used to extract DNA from each food sample. First, a pellet for each
sample was formed, from the set aside broth in the 500ml jar, through multiple
centrifuging in a 50ml falcon tube. Then 300µl of microbead solution was placed into the
falcon tubes and vortexed until the pellets are dissolved. The mixtures were put into a
micobead tubes with 50µl of MD1 solution and vortexed for ten minutes. The tubes were
placed into a centrifuge for 30 seconds at 10,000 x g. The supernatants were taken out of
the tubes and placed into 2ml centrifuge tubes and 100µl of MD2 solution placed into the
same tubes. The mixtures were then sat in an ice bath for 5mins. Then they were
centrifuged for one minute at 10,000 x g. After centrifuging, the supernatants were
transferred to new 2ml centrifuge tubes and 900µl of MD3 solution were placed into the
!!
30!
tubes. The mixtures were vortexed for 5 seconds and 700µl of the mixtures were placed
into 2ml spin filter tubes. The spin filter tubes were placed into the centrifuge for 30
seconds and the filtrated poured out. The process was repeated until no mixture is left in
the 2ml centrifuge tubes. Then 300µl of MD4 was placed into the spin filter tubes and
centrifuged for 30 seconds at 10,000 x g. The filtrate was discarded and the tube
centrifuged again for one minute. Then the filters were placed into permanent centrifuge
tubes. Then 50µl of MD5 solution was placed into the spin filter and centrifuged for 30
seconds at 10,000 x g. The spin filters were then discarded and the DNA in the tube was
stored at 20oC
Multiplex Polymerase Chain Reaction (PCR). The bacterial strains used were positive
controls for detection of pathogens used for multiplex PCR were Salmonella Newport, E.
coli O157:H7 EDL 933, and Listeria monocytogenes 10403S. The strains were grown
overnight at 37oC with rotary shaking in 5ml of BHI broth. The DNA was extracted from
each strain using the process mentioned before. Multiplex PCR was used to detect and
verify the pathogens, pathogen analysis, in the food samples. Two different PCR
procedures were used to identify the pathogens, one for Salmonella and E. coli and
another for Listeria. The PCR mixture contained 25ul of solution. The 25ul mixture for
Salmonella and E. coli Multiplex PCR consisted of: 4ul of nuclease free water, 12.5ul of
Master mix, Sigma Ready Mix Taq PCR reaction with MgCl2, 3ul of forward primer, 3ul
of reverse primer, and 2.5ul of DNA in sample. The 25ul mixture for Listeria Multiplex
PCR consisted of: 4ul of nuclease free water, 12.5ul of Master mix, 8ul of primer (1ul
each), and 2.5ul of DNA in sample.
!!
31!
Primers were combined in different mixtures to amplify the different strains of the
pathogens simultaneously. Specific primers, 18-24bp in length (Henegariu et al., 1997).
will be chosen for each pathogen. The primers used to identify Salmonella Newport, E.
coli O157:H7 EDL 933 in Multiplex PCR were TS-11F: 5’-
GTCACGGAAGAAGAGAAATCCGTACG-3’ (Sal), TS-5R: 5’-
GGGAGTCCAGGTTGACGGAAAATTT-3’ (Sal), VS8F: 5’-
GGCGGATTAGACTTCG GCTA-3’ (Ec), and VS9R: 5’-
CGTTTTGGCACTATTTGCCC-3’ (Ec) (Kawasaki et al., 2005), respectively as seen in
Appendix 1a. The primers used to identify Listeria and Listeria monocytogenes in
Multiplex PCR were LIS-R: 5’-AAGCAGTTACTCTTATCCT-3’, LIS-F: 5’-
AGCTTGCTCTTCCAATGT-3’ (Somer and Kashi, 2003), respectively as seen in
Appendix 1b. The Salmonella and E. coli reaction was carried out in the PCR
thermocycler, Bio-Rad CFX96 Real-Time System C1000 Touch Thermocycler, under the
following conditions: 50oC for 2 min; 95oC for 10 min, 40 cycles of 95oC for 20 s, 60oC
for 30 s; 72oC for 30 s, and 72oC for 7 min. The Listeria was carried out in the PCR PCR
thermocycler under the following conditions: 95oC for 5 min; 10 cycles of 95oC for 45 s,
63oC for 45 s, 72oC for 45 s; 30 cycles of 95oC for 45 s, 58oC for 45 s, and 72oC for 45 s;
and 72oC for 7 min.
!!
32!
Gel Electrophoresis. PCR products were analyzed by agarose gel electrophoresis to
visualize and get the expected band sizes for Salmonella, L. monocytogenes and E. coli
O157:H7. The expected size of Salmonella, L. monocytogenes and E. coli O157:H7 were
375, 287, and 120bp. A 500ml glass flask was used to make the gel; 150ml x TAE
Buffer, Bio-Rad 50xTAE buffer, and 3g of agarose were mixed in the flask to make 2%
gel agarose. The mixture in the glass was then heated to boiling until the mixture was
clear. Then 4ul of ethidium bromide (EtBr) was added to the mixture and slowly swirled
until mixed. The mixture was then poured into the gel tray to sit until solidified,
approximately 25 min. The comb was taken out of the tray and the chamber was placed
into the gel chamber. The wells in the gel were then filled with a mixture of 2.5ul of PCR
product and 2ul of loading dye, Bio-Rad Nucleic Acid Sample Loading Buffer. The DNA
ladder was added to the first well in the gel; 3.5ul of ladder was used. The gel was then
ran and picture was taken of the gel using the, Bio-Rad Gel Doc EZ Imager.
Statistical Analysis. Statistical analysis was conducted using Microsoft Excel for Mac
2011. All experiments of bacterial counts were performed twice in duplicates. Results are
presented as means ± Standard Deviation (SD). t-tests were used to determine differences
among counts (APC) in samples from areas of High SES and Low SES (Koro et al.
2010). Significance was determined at p < 0.05.
!!
33!
CHAPTER 4
RESULTS
Availability of Foods in Low-SES Areas
The stores surveyed for this study were located in low SES areas. The total
number of stores surveyed is 15. Each store was surveyed to determine availability of
fruit and vegetables, animal products, seafood, cooked foods, juices, and dairy products.
In the fruits and vegetables category apples, avocados, cucumbers, nectarines, peaches, or
strawberries were available in 1 store, 6%, lemons in 2 stores, 13%, grapefruit or peppers
in 3 stores, 20%, lettuce in 4 stores, 27%, and bananas, cabbage, oranges, or tomatoes in
6 stores, 40%. In animal products ground beef was available in 2 stores, 13%, chicken
legs in 6 stores, 40%, eggs in 11 stores, 73%, and deli meat in 12 stores, 80%. Seafood
was available in 1 store, or 6%. Cooked food items were available in 8 stores, or 53%.
Juices were available in 7 stores, or 47%. In the dairy category, milk was available in 13
stores, 87%, and butter was available in 5 stores, or 33% (Table 2).
Surveying was also done from the store prospective to show the amount of option
the 15 stores have. In the fresh produce and fruits category 7 stores, 47%, had 0 to 1
options, 1 store, 14%, had 2 to 3 options, 6 stores, 40%, had 4 to 5 options, and 1 store,
14%, had more than 5 options. In the animal product category, 6 stores, 40%, had 0 to 1
options and 9 stores, 60%, had 2-3 options (Table 3).
Table 2. Frequency of different food commodity availability at stores in low SES areas Food Commodity Availability frequency Number (total count) Availability (% of all
stores surveyed) Fresh Produce and Fruits
Apples 1(15) 14 Avocados 1(15) 14
!!
34!
Table 2. Continued Food Commodity Availability frequency Number (total count) Availability (% of all
Microbiological Quality of Food Commodities Tested
Aerobic Plate Count (APC)
To determine the microbiological load of the food commodities procured from
Low- and High-SES areas APC was performed. The logarithmic microbial count showed
significant differences between low and high SES for cabbage and lettuce (p < 0.05),
whereas no statistically significant differences observed for deli meats (ham) and chicken
legs. Figure 2 shows the Log CFU/ml for all chosen items. The Log CFU/ml was
7.1±0.96 and 5.2 ±0.82 for low and high SES of cabbage, respectively. The Log CFU/ml
was 6.8±0.87 and 4.81±0.39 for low and high SES of lettuce, respectively. The Log
CFU/ml was 6.4±0.6 and 5.9±0.7 for low and high SES of ham, respectively. For chicken
legs, Log CFU/ml was 5.75±0.84 and 5.04±0.64 for low and high SES, respectively
(Figure 2).
!!
36!
Figure 2. Aerobic Plate Count (APC) of different food commodities. The food products were acquired from Low- and High-SES areas of Memphis Metropolitan, Shelby County, Tennessee. The results depict log(10) transformed counts of the bacterial loads. Values are presented as Mean ±SD of two experiments done in duplicates. Columns, mean; bars, SD. Columns with (∗) indicate significant differences (p < 0.05) among those food categories.
Commodity-wise Distribution of the Quantified APC Values
Cabbage. For low SES, there were no samples with APC values of CFU/ml range
of 10-1,000. There were 2 samples, 8.3%, with a CFU/ml range of 1,001-10,000. There
were 5 samples, 20.8%, with a range 10,001-100,000. There were 5 samples, 20.8%, with
a range 100,001-1,000,000. There were 12 samples, 50%, with a range of 1,000,001-
10,000,000. In high SES there were no samples with a range of 10-100 or 1,000,001-
10,000,000. There was 1 sample, 3.7%, with a range of 101-1,000. There were 7
samples, 25.9%, with a range of 1,001-10,000. There were 17 samples, 63%, with a range
!!
37!
of 10,001-100,000. There were 2 samples, 7.4%, with a range of 100,001-1,000,000. The
APC for cabbage showed 70% of stores with the range 100,001-10,000,000 CFU/ml in
low SES and only 7% of stores with the same range in high SES, as seen in Table 4. In
both low and high SES no percentage of samples had the range of 10-100, 2% had the
range 101-1,000, 18% had the range 1,001-10,000, 43%, had the range of 10,001-
100,000CFU/ml, 14% had the range 100,001-1,000,000, and 24% had the range of
1,000,001-10,000,000.
Table 4. Distribution of Quantified APC in the Cabbage Samples
Range, CFU/ml Number of Samples Percent of Total Cumulative
Multiplex PCR specific for Salmonella and E. coli was evaluated utilizing the
food items, chicken legs, lettuce, cabbage, and ham. Figures 3-5 show the findings from
Multiplex PCR of Salmonella and E. coli. The expected sizes for them are 375 and 120bp
respectively. The NTC and PC stand for no template control and positive control, each
lane identifies if Salmonella or E. coli is in that food sample. In figure 1, all samples were
positive for E. coli. Also in figure 1, lane 20, food sample 197, shows positive results for
both Salmonella and E. coli. In figure 2, all samples were positive for E. coli. In Figure 3,
!!
42!
all samples were positive for E. coli and food sample 57, lane 9, was positive for both E.
coli and Salmonella.
Figure 3. Multiplex PCR Amplification profile of Salmonella and E. coli (sample group 1). Lane 1, 100 bp ladder ; lane 2, blank; lane 3, no template control; lane 4, blank; lane 5, positive control; lanes 6-20, samples.
Figure 4. Multiplex PCR Amplification profile of Salmonella and E. coli (sample group 2). Lane 1, 100 bp ladder ; lane 2, blank; lanes 3-16, samples.
!!
43!
Figure 5. Multiplex PCR Amplification profile of Salmonella and E. coli (sample group 3). Lane 1, 100 bp ladder ; lane 2, blank; lanes 3-15, samples.
Figures 6 and 7 show results from Multiplex PCR Listeria spp. and Listeria
monocytogenes. The expected size of Listeria monocytogenes is at 400 and 287bp. The
positive controls for Listeria spp. were L. innocua and L. ivanovii. All other controls were
specific to L. monocytogenes, strains 10403S, 19113, 4244, and Scott A. Detection of
Listeria spp. was not successful with the multiplex PCR protocol used in this study. In
Figure 4, lanes 1-8 are Singleplex PCR. Positive results are found for only two of the
positive controls used, strains 10403S and Scott A. All other controls in figure present
negative results. In lanes 9-19 are multiplexed with different DNA concentrations; it
presents negative results for all controls. Figure 5 shows results from singleplex, lanes 1-
5, and Multiplex, lanes 6-8. Again only negative results are seen from all controls.
!!
44!
Figure 6. Multiplex PCR Amplification profile of Listeria (test optimization 1). Lane 1, 100 bp ladder ; lane 2, blank; lanes 3-19, different Listeria species.
Figure 7. Multiplex PCR Amplification profile of Listeria (test optimization 2). Lane 1, 100 bp ladder ; lane 2, blank; lanes 3-8, different Listeria species.
!!
45!
CHAPTER 5
DISCUSSION
This research highlights the disparities in access to nutritious foods among poor
urban areas compared to wealthy areas. Acquiring fresh produce, fruit, or meat in low
SES convenience stores in metropolitan Memphis is challenging, which is why stratified
random sampling was used to depict both low and high SES. Since all these stores are
located in food desert areas, individuals in the community cannot go to a supermarket and
get any nutritious foods to include in their diet. The choices that were available for
nutritious foods in the convenience stores were more costly and of lower microbiological
quality than in supermarkets in high SES areas. Transporters and suppliers also account
for the quality differences in food. Supermarkets have the means to rely on traditional
transportation methods, insulation and refrigerated containers, whereas convenience
stores do not have those means. Also, supermarkets are bigger businesses and have
suppliers that have more strict standards of quality than suppliers for small business
owners.
The majority of the stores in low SES areas only had one option for fresh produce
and fruit, like bananas or strawberries; or lettuce (used only in sandwiches but not being
sold). If chicken legs were sold, in many cases, they were kept frozen and not fresh. Deli
meat and cooked foods bring the convenience stores the most business, probably that is
the reason why they were found so abundantly throughout the low SES area. Other foods
like eggs and milk are also bought frequently, which accounts for their abundance. Since
eggs and deli meat are bought so frequently, the majority of the stores had 2-3 options for
animal products.
!!
46!
Lettuce was rarely found in the low SES convenience stores, contrasting to high
SES supermarkets. There were 16 samples of lettuce taken in low SES and 30 samples of
lettuce taken in high SES. All lettuce samples were taken from a head of lettuce, in both
supermarkets and convenience stores, or available lettuce used on sandwiches, in
convenience stores. In the low SES convenience stores lettuce was found to have
significantly higher microbial load than the high SES supermarket counterparts. In the
other selected vegetable for this study, cabbage was more readily found in low SES, 24
samples taken. In high SES 27 samples of cabbage were taken. In all areas cabbage
samples were taken from a head of cabbage. As expected, in lower SES stores cabbage
was found to have significantly higher microbial growth than high SES stores.
The deli meat used in this study is chopped ham. Chopped ham was widely found
in all the convenience stores, but uncommonly found in supermarkets. Finding chopped
ham was difficult in supermarkets; most supermarkets sold only premium and expensive
deli meats. In low SES 38 samples of chopped ham were taken. In high SES 12 samples
of chopped ham were taken. In the other selected animal product for this study, chicken
legs were of better quality and more readily found in high SES stores. In low SES
chicken legs were only found in a few stores, but were found in all the high SES stores.
In low SES 22 samples of chicken legs were taken. In high SES 31 samples of chicken
legs were taken.
Aerobic plate counts provided an overall understanding on the microbial loads of
foods obtained from retail outlets in the Low SES communities with respect to
microbiological quality than those in the High SES communities. It gives a review of the
quality of produce based on how it deteriorates. Produce deteriorating in low SES areas
!!
47!
faster than high SES areas could be a result of poorer standards of quality in low SES,
than high SES. This study showed that there were more aerobic counts in low SES for
each food item, in comparison to high SES. Our result is in agreement with a previous
study done by Drexel University of microbial quality of food available to populations of
different socioeconomic status, findings showed higher microbial loads on produce from
markets in low-SES areas (Koro et al., 2010). The results of this study as well as the
mentioned study show those individuals that live in areas of low SES have a greater
probability of lower quality food that can cause food safety issues in the produce and
animal products being sold. In another study done at Drexel University on retail food
safety risks for populations of different races, ethnicities, and income levels, the results
also show that the food samples taken from low SES area have higher APCs than high
SES areas. It also presented that ready-to-eat (RTE) fruits and greens were most likely to
be found in markets in high-SES census tracts, which is consistent with research on food
access for populations of different demographics (Signs et al., 2011). This finding is
synonymous with the findings in this study. After surveying RTE items were not found in
any of the convenience stores in the low SES areas, but were found in all the high SES
supermarkets. Many of the low SES stores have an older facility and use older
appliances when cooking, preparing, and storing food. The significant differences in
APCs in cabbage and lettuce, items with a fast deterioration rate, may not have been
stored at the correct temperatures, was from a supplier with low quality standards, or was
transported through low quality standards. Other studies have found other explanations to
justify why lettuce has higher microbial counts. In the National University of Singapore
in a study done on the microbiological quality of fresh vegetables and fruits sold in
!!
48!
Singapore, found that since lettuces are leafy vegetables with large surface areas and
folds, this makes them more susceptible to bacterial contaminations and adhesions (Seow
et al., 2012). Another study done in the University of Lleida on the microbial quality of
fresh fruit, vegetables, and sprouts from retail establishments found that the open leaves
of lettuce might also be in contact with soil and irrigation water, trapping dirt in the folds
(Abadias et al., 2008)
The potential pathogens found in the samples gave normal findings. The generic
E. coli was most found in more samples than both Listeria spp. and Salmonella. In both
low and high SES generic E. coli was the most prevalent bacterium in cabbage, which
may be caused by manure being commonly used in agriculture to fertilize soil. Listeria
spp. and Salmonella both prevailed most in the chicken leg samples in comparison to
other tested food items, which may be caused by various processing techniques from
suppliers to the shelf life in supermarkets and convenience stores. In many studies finding
pathogens prevail less than 5% is normal. For example, in a study done at North Carolina
State University on the microbiological quality of fresh produce no E. coli was found and
Salmonella prevailed in less than 1% of samples (Johnston et al., 2005).
Multiplex PCR was also used to confirm presumptive plates for the selected
pathogens. In most studies pathogens are found in small amount of samples. For example
in one study on evaluation of a multiplex PCR system for detection of multiple
pathogens, Salmonella, Listeria, and E. coli only one sample gave a positive result for E.
coli O157:H7 with the multiplex PCR method (Kawasaki et al., 2005), where as in this
study 189 samples gave positive results. This study shows that E. coli is found in both
high SES and low SES areas. There were not significant differences in which areas they
!!
49!
are found in. The primers used in the Salmonella and E. coli PCR were specific for
Newport and EDL 933. All positive results show the presence of pathogens in the food
item. Many trials were completed for multiplex PCR of Listeria spp., but they were not
successful in identifying pathogens in the food samples. In a study done in the Israel
Institute of Technology on a PCR method based on 16S RNA sequence for detection of
Listeria specific forward primers, L. ivanovii (IVA-F), L. grayi and L. murrayi (MG-F),
and Listeria genus (LIS-F), were used in the PCR mix for the identification of the
presence of one or more of the Listeria spp. (Somer and Kashi, 2003). This primer
control is identical to the primer control in this study, but the results vary widely. The
results in this study may be negative due to the difference in cycles used in amplification
of PCR. It can also be because a difference in the PCR machines used in both studies.
Overall, the multiplex PCR was helpful in determining the quality of food in both low
and high SES by identifying the type of bacteria and if it was pathogenic. Therefore, the
difference in findings compared to this study may be due to the primers used. This study
may amplify a gene that is commonly found in the food items, while the Kawasaki study
amplified a more obscure gene using a different primer set.
In another study done by the Environmental Surveillance Unit in London on the
microbial quality of open ready-to-eat salads vegetables, findings showed that there was
only one pathogen was found, Listeria monocytogenes, out of four, E. coli O157,
Campylobacter spp., and Salmonella being the other three, 3% of samples were
unsatisfactory, and less than 1% unacceptable for microbial quality (Sagoo et al., 2003).
Compared to the North Carolina and the Environmental Surveillance Unit study the
results in this study have higher percentages of generic E. coli, Salmonella, and Listeria
!!
50!
spp. and APCs. The differences in sample size can account for the variances in
percentages. The entire sample size in this study is 200, whereas for the Johnston study it
is 400. A food safety and access study done at Johns Hopkins shows 81% of samples
positive for generic E. coli (Silbergeld et al., 2013). This finding is more similar to the
ones in this study. This is because the sample size for the chicken is 32. When the
bacteria for each food item is taken into account higher percentages for prevalence of
general bacteria and pathogens can be perceived. Other studies have discussed the
reasons potential food safety issues in low SES areas compared to high SES areas. For
example, in previous report on the identification of unique food handling practices that
could represent food safety risks for minority consumers the findings presented common
food mishandling practices like cooking poultry without using a thermometer, occasional
thawing of frozen poultry at room temperature, and consuming eggs with runny yolks
(Henley et al., 2012). Since cooked foods are sold often in low SES areas, these food
safety issues are important to ensure consumers in these areas do not become ill.
Decreasing food safety through better quality standards can be expensive. Therefore, this
issue will persist if small business owners do not have an incentives or regulations to
adhere to more strict standards of food handling and storing. Small retail facilities that
serve populations in low-SES urban areas may lack the resources, time, or knowledge to
focus on sanitation and proper refrigeration (Koro et al., 2010). Many of the urban stores
only stock what they can make a profit from. Since, fruits and vegetables are more
expensive; consumers will not often purchase them.
!!
51!
CHAPTER 6
CONCLUSIONS
In metropolitan Memphis, many communities are located in food deserts and the
only nearby place to obtain food is the local convenience store, where healthy food
choices are either limited, or of inferior microbiological quality. The current study has
also underscored the different access to healthy and nutritious foods available through
retail outlets in both low and high SES communities. The differences in the microbial
quality of food in the retail outlets in low and high SES communities show different
microbial composition with respect to food safety risks. The current study presents
similar findings with studies conducted in other metropolitan areas, that there is a
disparity in microbiological quality of foods available to populations; the microbial
quality of food in high SES areas is better than low SES areas.
Limitations
This study employed a stratified random sampling method and was vulnerable to
bias. The food items were chosen based on the availability in low SES areas. In this type
of sampling, selection bias can influence the results based on the commodities or stores
chosen. However, the selected stores and food products for this study was based on a
random selection of food establishments, and the commodities available in those venues.
Consequently, similar/same food types were procured from the comparison group (high
SES stores). Therefore, the selection bias equally affected sampling from low and high
SES stores. Compared to other studies, (Signs et al., 2011; Koro et al., 2010; Johnston et
al., 2005; Sagoo et al., 2003), the sample sized used in this study is smaller and for a
!!
52!
shorter period of sampling. Also, most of the sampling was completed in the summer
months where there is a rise of food borne illnesses.
Recommendations
In efforts to decrease the amount of disparities in food safety, small business
owners should be educated on food handling procedures with more strict standards. This
education will allow the business owners to incorporate some of the techniques to
increase microbial quality of food. Also, the public needs to be made aware of the issue
and how it especially affects their lives. If the public is not made aware, then they will
not perceive food safety disparities as an issue that affects their physical bodies or
lifestyles. Future research should look into partnering with the local health department to
talk to storeowners to acquire an overview of why storeowners do not sell fresh or
nutritious items and creating an intervention to improve food quality standards in low
SES convenience stores. A collaborative effort between community partners, local health
departments, public health researchers and practitioners, and stakeholders (including
storeowners and customers) could help create this awareness of food safety in
impoverished neighborhoods to improve public health.
!!
53!
REFERENCES Abadias M, Usall J, Anguera M, Solsona C, Vinas I. 2008. Microbiological quality of
fresh, minimally-processed fruit and vegetables, and sprouts from retail
establishments. Int J Food Microbiol 2008 Mar 31;123(1-2):121-9 doi:
101016/jijfoodmicro200712013 Epub 2008 Jan 30.
Ackers ML, Mahon BE, Leahy E, Goode B, Damrow T, Hayes PS, et al. 1998. An
outbreak of Escherichia coli O157:H7 infections associated with leaf lettuce
rot of fresh fruits and vegetables in the marketplace. Plant Disease 81(8): 867-
872.
(WHO), World Health Organization. 2015. 10 facts on food safety. Available:
http://www.who.int/features/factfiles/food_safety/en/ [accessed February 12
2016].
(WHO), World Health Organization. 2016. Food safety. Available:
http://www.who.int/topics/food_safety/en/ [accessed February 12 2016].
!!
69!
Zenk SN, Odoms-Young AM, Dallas C, Hardy E, Watkins A, Hoskins-Wroten J, et al.
2011. " You Have to Hunt for the Fruits, the Vegetables": Environmental Barriers
and Adaptive Strategies to Acquire Food in a Low-Income African American
Neighborhood. Health Education & Behavior: 1090198110372877.
Zhao C, Ge B, De Villena J, Sudler R, Yeh E, Zhao S, et al. 2001. Prevalence of
Campylobacter spp., Escherichia coli, and Salmonella serovars in retail chicken,
turkey, pork, and beef from the Greater Washington, DC, area. Applied and
environmental microbiology 67(12): 5431-5436.
Zhu M, Du M, Cordray J, Ahn DU. 2005. Control of Listeria monocytogenes
contamination in ready�to�eat meat products. Comprehensive Reviews in
Food Science and Food Safety 4(2): 34-42.
!!
70!
APPENDIX
PRIMER SEQUENCE TABLES Table 1a: Primer sequences used in Multiplex PCR amplification of Salmonella and E. coli. Primer Specificity Sequence TS-11� Salmonella� 5’-GTCACGGAAGAAGAGAAATCCGTACG-
3’ TS-5� Salmonella� 5’-GGGAGTCCAGGTTGACGGAAAATTT-3’ VS8 E. coli O157: H7 5’-GGCGGATTAGACTTCG GCTA-3’ VS9 E. coli O157: H7 5’-CGTTTTGGCACTATTTGCCC-3’
Table 1b: Primer sequences used in Multiplex PCR amplification of Listeria.
Primer Specificity Sequence IVA-F� L. ivanovii� 5’-AGCTTGCTCTTCCAATGT-3’ MG-F� L. grayi, L.
murrayi� 5’-GCTTGCTCCTTTGGTCG-3’
LIS-F L. monocytogenes, L. innocua, L. seeligeri, L. welshimeri