Assessing vegetable producers' beliefs regarding food safety issues

Full text

Vegetable producers’perceptions of food safety hazards in the Midwestern USA
Melanie L. Lewis Ivey
a
, Jeffrey T. LeJeune
b
, Sally A. Miller
a
,
*
a
Department of Plant Pathology, The Ohio Agricultural Research and Development Center, The Ohio State University, 1680 Madison Avenue, Wooster, OH 44691, USA
b
Food Animal Health Research Program, The Ohio Agricultural Research and Development Center, The Ohio State University, 1680 Madison Avenue, Wooster, OH 44691, USA
article info
Article history:
Received 11 November 2011
Received in revised form
19 January 2012
Accepted 20 January 2012
Keywords:
Food safety
Pre-harvest contamination
Post-harvest contamination
Mental models
Producer perceptions
Science-based management decisions
Risk communication
abstract
The perspectives, practices and potential gaps in knowledge regarding fresh produce safety hazards
among Midwestern US vegetable producers were measured using a survey-based confirmatory assess-
ment. Although the majority of vegetable producers considered themselves familiar with national Good
Agricultural Practices (GAPs) and agreed that implementing GAPs could reduce the risk of produce
contamination, they were not consistently practicing GAPs. Irrigation and run-off water, worker hygiene,
raw and composted animal manure, wildlife droppings, field proximity to livestock or wildlife habitats,
plant diseases and insects were acknowledged as important potential sources of pre-harvest microbial
contamination of produce, but paradoxically, producers disagreed that contamination most frequently
originated on-farm. There was significant variation in producers’level of agreement with regard to the
importance and economic feasibility of various management practices for the prevention of on-farm food
contamination. In general, vegetable producers did not declare an immediate need for more information
on food safety, but did nevertheless, indicate that they would like more information on the sources of
produce contamination, how contamination occurs, and GAPs guidelines. Vegetable producers preferred
in-person modes of communication over mass media, fact sheets or electronic modes, with only 17%
having a preference for Internet or email based information. These findings aid in the development and
delivery of targeted, science-based, farm management guidelines and knowledge translation programs
aimed at enhancing the safety of produce on the farm.
Ó2012 Elsevier Ltd. All rights reserved.
1. Introduction
Recent years have been marked by an increase in foodborne
disease outbreaks caused by the consumption of fresh produce
contaminated with pathogenic bacteria and these outbreaks
currently account for more illnesses than outbreaks caused by beef,
poultry and seafood (DeWaal, Tian, & Plunkett, 2009). Much of the
current research emphasis is directed toward identifying the
sources and mechanisms of pre- and post-harvest contamination
(Lynch, Tauxe, & Herberg, 2009), understanding plant pathogen/
human pathogen interactions (Aruscavage, Lee, Miller, & LeJeune,
2006; Parish et al., 2003), and assessing the efficacy of post-
harvest decontamination practices (Teplitski, Barak, & Schneider,
2009) with the long-term goal of reducing the number of food-
borne illnesses and deaths related to these illnesses. Long-term
research outcomes of such studies are to develop improved
science-based guidelines for pre- and post-harvest protection of
fresh produce from microbial threats. However, in the short-term,
there is a need to transfer and translate generated knowledge to
the end-users and to provide producers with the tools they may
need to integrate vegetable food safety practices into daily
management of food production systems.
Historically, information transfer in agriculture is a top-down
process, from “expert”source (i.e. scientist or academic) to “inex-
pert”recipient (i.e. practitioner or layperson). Experts often provide
information that is influenced by their own perceptions, biases,
preferences, beliefs and knowledge, often without first identifying
these same attributes in the recipients of the information (Morgan,
Fischhoff, Bostrom, & Atman, 2002). As a result, the recipients often
miss the intended message and become confused, annoyed or
disinterested (Morgan et al., 2002) while the communicator
becomes frustrated that the message is not being accepted or
adopted (Francis, 2009).
Effective risk communication is grounded in an understanding
of the mental models of communicators and target audiences
(Morgan et al., 2002). The mental models methodology has been
previously applied in risk communications related to natural
disasters (Zaksek & Arvai, 2004), global climate change (Bostrom,
Fischhoff, & Morgan, 1994), public health concerns (Atman,
Bostrom, Fischhoff, & Granger Morgan, 1994; Cox et al., 2003;
*Corresponding author. Tel.: þ1 330 263 3678; fax: þ1 330 263 3841.
E-mail address: miller.769@osu.edu (S.A. Miller).
Contents lists available at SciVerse ScienceDirect
Food Control
journal homepage: www.elsevier.com/locate/foodcont
0956-7135/$ esee front matter Ó2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodcont.2012.01.065
Food Control 26 (2012) 453e465

Jungerman, Shutz, & Thuring, 1988; Marik & Fischhoff, 1992),
hazardous processes (Bostrom, Fischhoff, & Morgan, 1992) and
most recently in agriculture-related disciplines (Cattaneo, Wislon,
Doohan, & LeJeune, 2009; Wilson, Parker, Kovacs, Doohan, &
LeJeune, 2009; Wilson, Tucker, Hooker, LeJeune, & Doohan, 2008).
The mental models approach (Morgan et al., 2002)isafive-stage
formal analysis that can provide insight into ways of broadening
the deeply held beliefs of a particular audience through first
identifying individuals’beliefs, then addressing them with highly
specific and targeted risk communication messages (Atman et al.,
1994; Morgan et al., 2002). Classically, the first two stages of
a mental models approach are the elucidation and description of
thought patterns, or mental models, of the problem and associated
decision-making process held by 1) experts in the field and 2)
practitioners or routine decision-makers. The third stage is the
development and implementation of a survey-based confirmatory
and quantitative assessment of the first two models. The last two
stages involve the development and evaluation of risk communi-
cation messages.
In this and companion studies, expert-based mental models
depicting expected influences on farmers’decision-making
regarding the safety of fresh produce were developed (Parker,
Wilson, LeJeune, Rivers III & Doohan, 2012; Wilson et al., 2009).
The primary goal of this study was to determine the extent to which
the beliefs, attitudes, perceptions and practices identified in these
previously described mental models were shared among producers
in the Midwestern United States. The results of this study will be
used to prioritize the development and delivery of risk communi-
cation materials about pre-harvest vegetable food safety hazards so
that producers can make more informed, science-based manage-
ment decisions.
2. Methods
2.1. Research approach and survey design
In a companion study (Wilson et al., 2009), qualitative expert
and grower mental models depicting expert and grower
perspectives on decision-making regarding contamination
prevention and response for fresh and fresh-cut produce were
developed. In this study, major factors influencing these growers’
perspectives and practices, and potential gaps in knowledge were
identified from the qualitative companion models and a survey-
based confirmatory assessment was developed. The survey con-
sisted of four sections (Tab le 1) and the questions were either non-
weighted and discrete (yes or no, “select one response”or “select
all that apply”), ranks, or Likert Scales. The survey was pre-tested
by various representative stakeholders (n ¼10) including vege-
table producers, Extension specialists and researchers from Ohio.
Survey pre-testers were asked to note the following as they per-
formed their validation: 1) approximate time needed to complete
the survey; 2) questions that were unclear or difficult to interpret;
3) terminology that was not understood or needed to be clarified;
4) formatting concerns (i.e. font type and size, amount of content
per page, grammatical or spelling errors) and 5) any additional
comments about the survey. The reviewers’comments were
ascertained in a group via teleconference and the survey was
modified accordingly.
2.2. Participant recruitment
All survey and contact documents were reviewed by The Ohio
State University Office of Risk Protection and deemed exempt
from further review by the Institutional Review Board. Partici-
pants were selected from a mailing list from a commercial lay
publisher of farming magazines (Farm Journal Media Inc., Park
Ridge, IL). Individuals self-identified as vegetable growers were
randomly selected from 195, 416, 151, and 42 names available in
the states of Ohio (OH), Michigan (MI), Indiana (IN) and Kentucky
(KY), respectively. One hundred and fifty-three OH growers, 318
MI growers, 119 IN growers and 31 KY growers were mailed
surveys, representing 76% of the qualified names provided from
each state on the mailing list. A five-contact, Tailored Design
Sequence (TDS) (Dillman, 2007) was used, consisting of a pre-
notice letter, questionnaire, reminder postcard and a return
incentive postcard sent to selected participants. Participants who
returned the incentive postcard within 90 days received a thank
you letter and a $5 gift card as compensation. For surveys
returned as “undeliverable”, additional questionnaires were sent
to alternate, randomly selected participants from the same state.
The survey was conducted from December 2008 through March
2009.
2.3. Data analysis
The survey was manually coded by assigning a unique number
to each possible response and the response codes were entered into
a multivariate Excel spreadsheet (Microsoft Excel 2008, 2007
Microsoft Corp.). The PivotTable report function in Excel 2008 was
used to combine, summarize and identify trends in the participants’
responses to each question. Descriptive statistics were determined
for the demographic and vegetable production responses and
research questions and hypotheses were tested using either the
KruskaleWallis (Kruskal, 1952; Kruskal & Wallis, 1952) or Mood
Table 1
Survey sections, question types and content of the vegetable food safety survey.
Section No. Question type Content
1 Production and management practices Years producing vegetables, acres produced, production locations, types of vegetables produced,
irrigation and wash water sources and practices, Good Agricultural Practices (GAP)
implementation, auditing procedures, Integrated Pest Management (IPM) practices,
production challenges
2 Knowledge and beliefs regarding
contamination threats
Organizations responsible for ensuring food safety, organizations that influence management
practices, pathogens that contaminate fresh produce, sources of pre- and post-harvest
contamination, activities likely to cause contamination, management practices that prevent
contamination, economically feasible management practices to prevent contamination, stages
in the chain of custody where contamination occurs
3 Knowledge and beliefs regarding
contamination response
and communication
Opinions regarding producer’s preparedness for a contamination event, need for information
about food safety or GAP guidelines, preferred modes or sources of communication, types of
research that is needed to understand and respond to a contamination event and the impact
a contamination event would have on production
4 Demographics Age, gender, education, heritage, religious affiliations, production sales
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465454

one-way analysis of variance by ranks tests (Mood, 1954), or
Spearman’s rank-order correlation coefficient (Spearman, 1904)as
appropriate. Medians were separated using the Dunn test (Dunn,
1964) with a Minitab macro (Orlich, 2000) in combination with
the KruskaleWallis test. Data were considered to be statistically
significant at a 95% confidence level (
a
¼0.05) unless otherwise
noted. All statistical tests were conducted using Minitab 15.1.20.0
(2007 Minitab Inc.).
3. Results
3.1. Survey response rate and respondent demographics
Of the 621 surveys mailed, 210 were returned (33.8% response
rate) within 90 days. At least 50% of the questions were answered
for 164 (26.4%) of the surveys, and 46 (21.9%) were returned with no
responses. Fifty-two percent (52.4%) of the respondents grew
vegetables in MI, 33.5% in OH, 20.1% in IN and 8.5% in KY. One
percent (1.2%) of the respondents grew vegetables in Florida (FL)
and Illinois (IL) and less than one percent grew vegetables in
Pennsylvania (PA) and Georgia (GA). Respondents’demographic
information including age, level of education, gender, vegetable
sales (for 2007) and number of acres and years of vegetable
production is summarized in Table 2.
3.2. Production practices
The majority (95%) of the respondents produced vegetables
using conventional farming practices. One respondent from OH and
one from MI (1%) produced certified organic vegetables, 4% used
both certified organic and conventional management practices
(mixed farming practices) and 2% indicated transitional organic
production systems. Ten (6%) respondents grew vegetables under
high tunnels (hoop houses) with two of those respondents indi-
cating that more than 21% of their production was under high
tunnels.
Cucurbits, sweet corn and tomato were most often listed as the
top three vegetables produced in terms of total acres of production.
In OH, 31% of the respondents listed cucurbits and tomato as the
most cropped vegetables. In MI, cucurbits and asparagus were top
ranked and in IN and KY tomato and sweet corn were ranked first.
3.3. Growers’perspectives on management decisions regarding
produce food safety
The distribution of producer responses as to the types of
groups or organizations that influence their management decisions
varied significantly regarding produce food safety (
c
2
¼183.4,
DF ¼9, p0.001). Over 68% of producers either strongly or
somewhat agreed that consumers, crop consultants, retailers,
government regulators and university representatives influence
their management decisions regarding produce food safety (Fig. 1).
Only 32% of producers were in agreement that media influenced
their produce food safety management practices (Fig. 1). However,
55% agreed that news reports or stories specific to disease
outbreaks associated with fresh produce changed their manage-
ment practices. Producers strongly or somewhat disagreed (72%)
that farm management practices should be regulated by govern-
ment agencies. When asked to rank six groups or organizations
(consumer, producer, retailer, processor, government agencies and
universities) from most to least important with regard to which
should be responsible for ensuring food safety, producers ranked
themselves first followed by processors, retailers, and consumers
(
c
2
¼411.3, DF ¼5, p0.001). Government agencies and university
extension were equally ranked last.
3.4. Producer familiarity with and implementation of good
agricultural practices
Specific survey questions were included to determine
producers’self-perception of compliance with GAPs. Questions
were also asked about their specific management practices. These
questions permitted a comparison of how producers thought
they were complying with GAPs and actual management prac-
tices employed on the farm. Producers agreed (83%) that
following GAPs could reduce the risk of produce contamination,
however less than two thirds (60%) indicated that they practiced
GAPs all or most of the time. Approximately one quarter of
respondents (23%) were not at all familiar with GAPs and of those
respondents 89% indicated that they didn’t know if they were
practicing GAPs, 3% indicated they never practiced GAPs, and 8%
indicated that they practiced GAPs all of the time. There was
a positive correlation between producers’familiarity with GAPs
and their implementation of them, but only amongst those who
Table 2
Descriptive summary of Midwestern vegetable producer demographics.
Variable Variable
description
Distribution of responses (%)
Ohio Michigan Indiana Kentucky All
Age 35 12.2 5.1 3.4 0.0 6.3
36e65 75.5 79.7 86.2 90.0 80.5
66 12.2 15.2 10.3 10.0 13.2
Gender Male 93.9 98.8 89.7 70.0 94.4
Female 6.1 1.2 10.3 30.0 5.6
Education High school
or less
40.8 43.8 41.4 30.0 42.5
College
or higher
59.2 56.2 58.6 70.0 57.5
Years in
vegetable
production
1e20 28.0 22.5 36.7 45.5 28.8
21e30 14.0 23.8 23.3 9.0 20.9
>30 58.0 53.7 40.0 45.5 50.3
Acres in
vegetable
production
<10 18.0 3.7 16.7 63.6 14.6
10e50 12.0 13.6 10.0 27.3 14.0
51e100 16.0 17.3 6.7 0.0 14.0
>100 54.0 65.4 66.7 9.1 57.3
Vegetable sales
a
100 36.7 31.3 34.5 80.0 37.5
100 <500 30.6 37.5 27.5 10.0 33.8
500 32.7 31.3 37.9 10.0 28.8
a
Vegetable production sales (X 10
3
) in 2007.
Fig. 1. Response percentage of Midwestern vegetable producers who indicated that
they strongly or somewhat agree that various groups or organizations influence their
management decisions regarding produce food safety. The line on the graph represents
the top five groups or organizations that influenced producers’management decisions
regarding produce food safety.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465 455

were very familiar, somewhat familiar or indifferent to GAPs
(
r
¼0.347, p0.001; Fig. 2A). For those producers who ranked
(in terms of acreage) a fresh market vegetable as their top ranked
crop the correlation between producer familiarity and GAPs
implementation increased (
r
¼0.394, p¼0.006; Fig. 2B).
No associations between the frequency with which producers
cleaned or sanitized packaging equipment and the extent to which
they practiced GAPs were identified. Over half (55%) of the
producers who packaged produce on the farm cleaned and sani-
tized their packaging equipment (bins, totes etc.) at least once
a week (Fig. 3A). However, 29% never cleaned or sanitized pack-
aging equipment and of those producers, 18% indicated that they
practiced GAPs all or most of the time.
There was a positive correlation between the frequency of
testing water (
r
¼0.343, p0.001) or testing wash water for
residual disinfectants (
r
¼0.227, p¼0.013) and the extent to which
producers indicated that they practiced GAPs. The majority (70%) of
growers who washed produce on the farm tested their water at
least once per year (Fig. 3B). However, 20% never tested their water
(Fig. 3B) and 67% did not test for residual disinfectant in their wash
water (Fig. 3C). Only 16% of the producers who washed produce on
the farm considered themselves to be very familiar with GAPs and
indicated that they practiced them all of the time. An additional
30% indicated that they were somewhat familiar with GAPs and
practiced them most of the time. The level of practice of GAPs
indicated by producers was associated with on-farm packaging of
produce (
r
¼0.161, p¼0.045) but not with the washing of produce
on the farm.
Fig. 2. Box plots showing Midwestern vegetable producers’reported familiarity and
implementation of national Good Agricultural Practices (GAPs). A: All vegetable
producers who responded to both of the questions “Are you familiar with GAPs?”and
“To what extent does your farm practice national GAPs?”. B: Includes only those
producers that indicated a fresh market vegetable as their top ranked crop in terms of
acreage and responded to both of the questions listed in A. Gray boxes represent the
interquartile range where the top of the box is at third quartile, the bottom of the box
is the first quartile and the middle line is placed at the second quartile, or median.
Lines extending from the boxes represent the range of data from the first and third
quartiles to their respective extremes.
A
B
C
Fig. 3. Reported frequency of cleaning or sanitizing packaging equipment (A); testing
produce wash water (B) and testing residual disinfectant in wash water (C) by Mid-
western vegetable producers who wash produce on their farm.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465456

3.5. Producer beliefs regarding sources and prevention of the
contamination of fresh produce
Producers ranked the importance of some microorganisms
higher than others (
c
2
¼253.0, DF ¼9, p0.001; Fig. 4). Producers
considered Escherichia coli (E.coli; pathogenic or non-pathogenic)
and Salmonella to be significantly more important contaminants of
fresh produce than coliforms, Listeria, parasites, Clostridium,
viruses, Cyclosporia,Cryptosporidium or the mad cow disease
agent.
Producers had significant differences in their level of agreement
as to where in the food chain produce contamination is most likely
to occur (
c
2
¼95.8, DF ¼6, p0.001; Fig. 5). Producers believed
that the home (50%), processing (43%) and wholesale or retail
operations (38%) were the most common sites of produce
contamination and that contamination was least likely to occur on
the farm (19%).
The majority of producers agreed that certain vegetables are
more likely to become contaminated than others (89%;
c
2
¼227.7,
DF ¼2, p0.001) and that knowing the source of contamination is
important to making management decisions (94%;
c
2
¼269.0,
DF ¼2, p0.001). Only 11% of producers disagreed with the
statement “organic farms are more likely to produce contaminated
vegetables than conventional farms”. Forty-four percent agreed
with the statement and the remaining 45% were of neutral opinion.
The two organic producers who responded to the surveysomewhat
disagreed that organic farms are more likely to produce contami-
nated vegetables.
Producers’level of agreement regarding potential sources of
pre- and post-harvest contamination (
c
2
¼259.5, DF ¼18,
p0.001 and
c
2
¼83.4, DF ¼14, p0.001 respectively) of fresh
produce varied significantly. At least one third of the respondents
strongly agreed that animal and bird droppings, raw manure,
handling by workers and criminal acts or terrorism could
contribute to both pre- and post-harvest contamination (Fig. 6).
Producers also strongly agreed that wash water (42%) and raw
animal manure (50%) were potential sources of contamination.
Producers somewhat agreed (median response ¼2) that insects
(pre- and post-harvest) and pre-harvest plant pests could be
sources of contamination and they were indifferent (median
response ¼3) to whether or not post-harvest plant diseases and
weeds were sources of contamination (Fig. 6). Producers who
strongly agreed that pests could be sources of contamination
believed insect infestations (pre- and post harvest) were more
likely to be a source of contamination than plant diseases or weed
infestations (
c
2
¼17.5, p¼0.002, DF ¼4).
3.6. Producer beliefs regarding the importance and economic
feasibility of management practices for the prevention of on-farm
food contamination
Producers’level of agreement regarding the importance
(
c
2
¼765.3, DF ¼27, p0.0001) and economic feasibility
(
c
2
¼696.3, DF ¼26, p0.0001) of management practices for the
prevention of on-farm food contamination varied significantly.
Producers equally believed that the control of plant diseases and
insect infestations is somewhat important in the prevention of
produce contamination and they somewhat agreed that the control
of these pests is an economically feasible practice (Table 3).
Producers were indifferent as to whether or not minimizing contact
between irrigation water and edible plant parts would be an
important management practice to reduce food contamination and
somewhat disagreed that eliminating overhead irrigation would be
economically feasible (Table 3). More producers believed that the
treatment of wash water would be very important for the
prevention of food contamination than would treatment of surface
water or well water used for irrigation (Fig. 7). Producers equally
believed that the routine testing of wash water and irrigation water
could be somewhat important in the prevention of produce
contamination but only somewhat agreed that testing wash water
would be economically feasible and were indifferent as to the
extent to which testing irrigation water would be economically
feasible. Routine on-farm inspections by trained upper manage-
ment employees were thought to be more important and
economically feasible than inspections by government agents or
third party auditors. Over one third (33%) of producers thought that
banning raw manure as an amendment would be very important
for the prevention of food contamination and 24% strongly agreed
Fig. 4. Percentages of Midwestern vegetable growers who believed that various
foodborne microorganisms were “very important”or “somewhat important”
contaminants of fresh produce.
Fig. 5. Cumulative distribution of Midwestern vegetable producers’reported
responses as to where contamination of fresh produce most likely occurs.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465 457

that it would be economically feasible (Fig. 7). Producers also
believed that the appropriate timing of both raw and composted
animal manure would be a very important management practice
for the prevention of produce contamination (56 and 43% respec-
tively; Fig. 7). While producers believed that separating livestock
and poultry farms (38%) from vegetable producing areas and the
use of barriers to restrict animal movement in the field (35%) were
very important in the prevention of food contamination, only 17%
and 8% strongly agreed that these practices were economically
feasible, respectively (Fig. 7).
3.7. Food safety information needs of producers
The majority (82%) of producers declared that they had no
immediate need for food safety information, that they knew where
to find information regarding food safety (77%) and that they were
content with the quality of information provided by university
extension (70%). Producers most frequently indicated that they
wanted more information on the sources of contamination (41%)
and GAPs guidelines (40%; Fig. 8).
3.8. Producer perspectives on contamination response
There was significant variation in the distribution of responses
as to the type of impact that the occurrence of an on-farm food
safety contamination event might have on a producers’farm
(
c
2
¼218.3, DF ¼8, p0.001; Fig. 9). The perceived impacts
separated into three groups based on the frequency of responses.
Poor reputation, adoption of additional food safety practices and
loss of market share (Group 3) were most frequently selected, fol-
lowed by additional inspections, changes to the type of commodity
being produced, going out of business, lower product pricing and
worker layoffs (Group 2). Group 1 contained three producers (1.8%)
that selected the “other”category with two producers specifying
the type of impact (1: an on-farm contamination event would be
“highly unlikely”because the commodity that they produce is
processed and 2: a “loss of buyers”would also be an impact). Most
producers (80%) indicated that a contamination event on a neigh-
boring farm could impact their own farms operations (
c
2
¼56.0,
DF ¼1, p0.001) and that their farms would be prepared to deal
with a contamination event should one occur (71%;
c
2
¼27.9,
Fig. 6. Midwestern vegetable producers’reported level of agreement about potential sources of pre- and post-harvest microbial contaminants.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465458

DF ¼1, p0.001). Producers (83%) also believed that their
employees are very or somewhat knowledgeable about food safety
hazards (
c
2
¼68.4, DF ¼1, p0.001).
Vegetable producers most frequently selected university
extension (66%) as the group or organization that they would
contact in response to an on-farm contamination event (
c
2
¼284.3,
DF ¼8, p0.001; Fig. 10). Producers’second choice was crop
consultants (42%) followed by food safety consultants (29%),
commodity group representatives (27%), government agency
representatives (20%), retailer (16%), third party auditors (10%), no
group or organization (7%) and media (2%).
Producers ranked university extension significantly higher than
any other regarding the group or organization from which they
would seek help during an on-farm contamination event
(
c
2
¼114.3, DF ¼7, p0.001). Commodity group representatives,
crop consultants and food safety consultants ranked second, and
third party auditors ranked third. Producers ranked both retailers
and government agency employees fourth and media representa-
tives fifth.
3.9. Communication of food safety risks
Producer responses to the question “What modes of commu-
nication do you prefer to receive information regarding food safety
hazards?”were combined into four groups: 1) mass media that
included newspapers, television or radio and magazines; 2) in-
person communication that included training sessions and
grower meetings; 3) electronic communication that included
internet or email; and 4) fact sheets or pamphlets. Producers
preferred in-person modes of communication to all other
communication modes (
c
2
¼62.1, DF ¼3, p0.001). However, for
their second most preferred mode of communication, the distri-
bution of responses varied depending on age (
c
2
¼21.3, DF ¼6,
p¼0.002) and level of education (
c
2
¼7.5, DF ¼3, p¼0.057).
Producers 35 years of age preferred electronic modes of
communication while those between the ages of 36e65 preferred
fact sheets or pamphlets and producers 66 years old preferred
mass media modes of communication. Post secondary school
educated producers preferred fact sheets or pamphlets and those
with any level less than post secondary preferred mass media.
Producers most frequently selected training sessions held in
the off-season (74%) as the most likely way to encourage producer
attendance or participation in learning programs or training
sessions (
c
2
¼300.2, p0.001, DF ¼7; Fig. 11). Off-season, short
training sessions (1e2 h) held close to the farm were preferred
over off-season, all day, on-farm training sessions (
c
2
¼49.4,
p0.001, DF ¼3). Nearly half of the producers (47%) indicated
that training sessions encompassing multiple topics such as food
safety, disease management, or soil fertility would encourage
attendance and participation. Producers who selected “other”
indicated that “getting paid to be trained in issues regarding food
safety”,“training sessions leading to post test credits”,“being
trained by people who know what we need”,and“training
sessions required by the buyer”would encourage attendance and
participation.
4. Discussion
Based upon expert models and previous in-depth interviews
with vegetable producers we were able to determine the extent to
which beliefs and behaviors pertaining to pre- and post-harvest
food safety hazards were held among a wider population of vege-
table growers in the Midwest. The majority of surveyed vegetable
producers:
Table 3
Midwestern vegetable producers’median responses regarding the importance and
economic feasibility of management practices for the prevention of on-farm food
contamination.
Management practice Median response
Prevention
a
Economic
feasibility
b
Control of crop pests
Plant diseases 2 eei
c
2fei
Insect infestations 2 hek2hek
Weeds 2 def2eeh
Irrigation
Minimizing plant and water contact 3 bc e
Use of drip or in-furrow irrigation e
d
2de
Eliminating overhead irrigation e4ab
Amendments
Excluding or banning the use of raw manure 2 eeg3cd
Appropriate timing of raw manure applications 1 lm e
Appropriate timing of composted manure
applications
2kl e
Use of composted manure e3c
Use of chemical fertilizers e2de
Wildlife
Separation of fields from livestock and poultry
farms
2iek2cd
Use of physical barriers to restrict animal
movement in fields
2eei4ab
Routine testing of:
Wash water 2 gei2eei
Irrigation water 2 de 3 cd
Produce 2 eeg2def
Compost e2deg
Treatment of:
Wash water 2 iek2eeh
Irrigation surface water 2 cd 3 b
f
Irrigation well water 3 b
Routine on-farm inspections by:
Government agents 3 a 4 a
Trained upper management employees 2 eeh2def
Third party auditors 3 b 3 b
Employee training in:
Personal hygiene 1 m 2 jk
Hand washing techniques 1 m 2 jk
Sanitation
Machinery and equipment Providing convenient 2 jk 2 gek
Restrooms and hand washing stations 1 m 2 k
Proper disposal of sewage 1 n 1 l
Documentation and signage
Posting personal hygiene signs in multiple
languages
e
2kl 2jk
Food safety standard operating procedures (SOP) 2 fei
Access to food safety SOP 2 hek2eeg
Other
Neighboring farms participation in food safety
prevention practices
2eege
Controlling movement of on-farm visitors 2 de e
c
2
¼765.3,
DF ¼27,
p0.0001
c
2
¼696.3,
DF ¼26,
p0.0001
a
Median response to the question “To what extent do you consider the following
management practices to be important in the prevention of food contamination?”
where 1 ¼very important; 2 ¼somewhat important; 3 ¼indifferent; 4 ¼somewhat
unimportant and 5 ¼not important.
b
Median response to the questions “To what extent do you agree or
disagree that the following on-farm management practices are economically
feasible options to prevent food contamination?”where 1 ¼strongly agree;
2¼somewhat agree; 3 ¼neither agree nor disagree; 4 ¼somewhat disagree and
5¼disagree.
c
Differences were tested and medians separated using KruskaleWallis one-way
analysis test on ranks. Medians followed by the same letter within a column are not
statistically significant at a 95% confidence level (
a
¼0.05).
d
Indicates that the management practice was not a choice for the relevant
question on the survey.
e
Question on the survey actually reads as “posting of hand washing techniques in
all languages spoken by all workers”.
f
Question on the survey did not distinguish between surface and well water.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465 459

Believe that they are practicing national GAPs all or most of the
time and that GAPs can reduce the risk of produce contami-
nation. However, their responses indicated that many do not
implement GAPs at a level of consistency needed to effectively
minimize the risk of contamination.
Do not believe that contamination is most likely to occur on the
farm.
Believe that many preventative management practices are
economically feasible.
Believe (amongst conventional producers) that contaminated
vegetables are more likely to originate from organic farms than
non-organic farms.
Do not want GAPs or other management guidelines to be
government-regulated.
Did not declare an immediate need for food safety information
and thought that the majority of their employees were
“somewhat”or “very knowledgeable”about food safety
hazards. Nevertheless, if they were to obtain more information
Fig. 7. Percentage of Midwestern vegetable producers who strongly agreed that various management practices were important for the prevention of fresh produce contamination
or were economically feasible.
Fig. 8. Midwestern vegetable producers’reported preference for food safety-related topics for which they would like to receive more information.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465460

on food safety, they wanted to learn more about sources of
produce contamination.
Believe that bacteria are the most important contaminants of
fresh produce.
Overwhelmingly prefer to receive their information in-person
and from university extension.
4.1. GAPs are not being implemented consistently by Midwestern
vegetable producers
Producer-proclaimed familiarity with GAPs correlated poorly
with producers’usage of GAPs. Differences between knowledge
and behavior are not uncommon and research related to food safety
and other health risks indicates that there is often a discrepancy
between self-reported behavior and what is actually practiced
(Abbot, Byrd-Bredbenner, Schaffner, Bruhn, & Blalock, 2009; Patil,
Cates, & Morales, 2005; Rimal, Fletcher, McWatters, Misra, &
Deodhar, 2001). The discrepancy between knowledge and behavior
has been linked to a person’s feelings, emotions, or degree of
acceptance or rejection of some concept (or an affective dimension,
Galli, 1978). It is therefore important for communicators to be
aware of the recipients’attitudes toward a subject or concept. We
posit several possible reasons for the inconsistency between
reported GAPs familiarity and implementation: 1) a lack of under-
standing of the specific practices required to be in compliance with
GAPs guidelines or a lack of clarity of the current guidelines; 2)
a belief among producers that they are unable to comply (self-
efficacy) or that following GAPs will not have an effect on the safety
of fresh produce (outcome expectancy); or 3) a conscious and
rational choice of the producer based on personal beliefs with
respect to issues related to on-farm food safety, including but not
Fig. 9. Distribution of producers’reported perceptions of the type of impact a food safety contamination event would have on their farms.
Fig. 10. Outside group or organization most likely to be contacted by Midwestern
vegetable producers in response to an on-farm contamination event. Differences
among vegetable producers’responses were tested using the Mood one-way analysis
of variance by ranks test where
c
2
¼284.3, DF ¼8, and p0.001.
Fig. 11. Types of training sessions or learning programs geared toward on-farm food
safety that would encourage participation of Midwestern vegetable producers.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465 461

limited to the role producers believe they and others should play in
ensuring a safe food supply (fundamental attribution error).
Communicators of produce food safety information should
consider and attempt to integrate these possibilities in their
development of outreach materials. Ultimately, for a mechanism of
knowledge transfer to positively effect change, the GAPs guidelines
will need to be re-communicated in a way that encourages indi-
vidual empowerment and is sensitive to regional and geographical
differences in producer attitudes. Careful thought must also be
given to the content of the message. Wittwer, Nuckles, and Renkl
(2005) demonstrated that the negative effects of instructional
explanations could persist if the message is not adapted to accu-
rately reflect the learners’knowledge. Therefore we propose that
the addition of the most current science-based information per-
taining to on-farm produce safety, the identification of reliable
metrics, providing producers with clear standards and emphasizing
the immediate production impacts and economic benefits of safe
management practices, will have the potential to re-engage vege-
table producers and increase GAPs implementation. Although those
producers who thought themselves to be very familiar or some-
what familiar with GAPs have the potential to be impervious to
GAPs-related messages, nearly half of the respondents indicated no
familiarity at all with GAPs, suggesting that they may be receptive
to such messages.
4.2. Vegetable producers do not believe that contamination is most
likely occurring on the farm
Midwestern vegetable producers believed that contamination of
fresh produce most often occurred in the home and not on the farm.
This is not surprising since the majority of reported foodborne
disease outbreaks over the last decade have been associated with
food prepared or consumed in the home (Redmond & Griffith,
2003). Nonetheless, contamination can occur at any time during
growing, harvesting, packing, processing, distributing or handling
and outbreaks associated with on-farm contamination events have
been steadily increasing (Alliance for Food and Farming, 2010).
Many of the more recent outbreaks, especially those associated
with tomatoes contaminated with Salmonella, have been traced
back to the farm or packinghouses (CDC, 2007; Cummings et al.,
2001). For these reasons the successful prevention of foodborne
diseases associated with fresh produce must be a continuous chain
of prevention from the field to the home and have the cooperation
of all participants in the chain of custody.
4.3. Vegetable producers believe that preventative food safety
management practices are economically feasible
Although Midwestern vegetable producers who responded to
this survey did not implement GAPs consistently; they indicated
that they believe most preventative management practices listed in
the survey were economically feasible. This could suggest that
economic potential outcome is not the primary driving force
for GAPs implementation, as has been found for the adoption
of sustainable agricultural practices (SAP); (D’Souza, Cyphers,
& Phipps, 1993; Rodriguez, Molnar, Fazio, Sydnor, & Lowe, 2008;
Saltiel, Bauder, & Palakovich, 1994). However, the rationale
driving this conundrum is not known. Although some of the
management practices listed in the survey such as pest control and
the use of soil amendments are not new and are commonly used to
produce vegetables, many would need to be newly adopted and
may add substantial costs to producers’operations. The respon-
dents in this survey may have underestimated such costs.
4.4. Vegetable producers using conventional production practices
believe contaminated vegetables are more likely to originate from
organic farms than non-organic farms
Even though nearly 35% of the respondents using conventional
farming practices indicated that they used either manure or com-
posted manure to produce vegetables they believed organic farms
were more likely to produce contaminated vegetables than non-
organic farms. The risk of microbial contamination on organic
farms has been hotly debated both in the United States and United
Kingdom over the last decade. Proponents of organic farming
believe that organically produced foods are inherentlysafe (Powell,
Jacob, & Chapman, 2010; Soil Association, 2001), while critics assert
that organically produced foods are at higher risk of contamination
because of the use of animal manure (Avery, 1998; Stephenson,
1997). Although the microbial quality of organically grown
produce has been evaluated (Arthur, Jones, Fabri, & Odumeru,
2007; Bohaychuk et al., 2009; Oliveira et al., 2010), these studies
provide no evidence that organic production practices are linked to
higher microbial food safety risks than those practices used in
conventional production systems. This is not surprising since there
are numerous obstacles associated with conducting such studies,
especially at the pre-harvest level of production. For example,
because contamination rates are very low, large samples sizes are
needed in order to have adequate statistical power. In addition,
deliberate release of genetically modified (marked strains, for
example) or non-modified human pathogens in a field setting such
as needed for pre-harvest studies is problematic. A better under-
standing of the risks, if any, associated with organic production
practices and a more practical means to identify these risks is
needed before adequate information translation can occur.
However, even with sound, reproducible scientific evidence it can
be very difficult to change individuals’beliefs and behavior with
regard to sustainable agricultural practices (Francis, 2009; Prokopy,
Floress, Klotthor-Weinkauf, & Baumgart-Getz, 2008; Rodriguez
et al., 2008; Vanclay & Lawrence, 1994). Francis (2009) argues that
studies that compare organic farming practices to conventional
practices are sometimes biased toward the researchers’personal
beliefs resulting in “clouded”conclusions that will not likely
convince stakeholders to change their taut beliefs’about either
type of production system. Francis (2009) goes on to suggest that
a better research approach to such a complex issue is to focus
resources on improving the agro-ecosystems as a whole and not
focusing on comparative studies that consider just one or two
factors between the systems. Leifert, Volakakis, and Cooper (2008)
made this same point and used a Hazards Analysis Critical Control
Points (HACCP)-like system to identify six Risk Reduction Points
(referred to as Critical Control Points for HACCP) where risks from
enteric pathogens could be reduced in organic and ‘low-input’
ready-to-eat vegetable crops. They also identified approaches that
could be used to minimize the risk of pathogen transfer onto these
crops. In a study by Franz, Semenov, and van Bruggen (2008),
probabilistic modeling was used to estimate the probability of
lettuce contamination with E.coli O157:H7 from manure-amended
soil and the effect of intervention strategies was determined.
Interestingly, a model developed by Franz et al. (2008) demon-
strated that traditionally produced organic manure (30-day storage
and applied at least 60 days prior to planting) was more risk-
reducing than manure produced using intensive conventional
practices. Regardless of the production method, all foods in the
United States, including fresh produce are covered under U.S. food
safety laws and must meet the minimal requirements of the United
States Food and Drug Administration (US-FDA). Until additional
science-based information is available on the added risks, if any, in
using organic management practices to produce vegetables the
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465462

best approach may be to explain to conventional vegetable
producers how sustainable agriculture practices, such as those used
in organic farming, can and must be compatible with US-FDA food
safety guidelines.
4.5. Vegetable producers do not want GAPs or other management
guidelines to be government regulated
Public reaction to recent foodborne related outbreaks on fresh
produce has motivated the United States federal government to
increase its efforts to enhance food safety. In the past, the FDA has
issued non-mandatory food safety guidelines designed to prevent
microbial contamination. However there is increasing pressure
from consumers, buyers and retailers for the enactment of
mandatory state and federal regulations. While Midwestern vege-
table producers agreed that government agencies influence their
management decisions regarding food safety hazards, they strongly
disagreed that management practices should be regulated.
Producers’reasons for not wanting government-issued regulations
were not investigated in this study. Nonetheless, the process of
bringing federally mandated regulations for produce food safety
into law has already begun (see Food Safety Modernization Act,
Office of Information and Regulatory Affairs, Executive Office of the
President; www.reginfo.gov/public/). Palma, Ribera, Paggi, and
Knutson (2010) explored how government, producers, and others
in the private and public sectors could work together to share the
responsibility of assuring a safe food supply. While they provided
reasonable options, the analysis lacked a nuanced understanding of
farmers’reluctance to mandated government regulations. Ascer-
taining Midwestern vegetable producers’perceived burdens or
benefits of mandatory food safety regulations, especially those
related to enforcement procedures and cost sharing, could provide
valuable insight into how to best work with producers to positively
influence their adoption of unavoidable food safety regulations.
4.6. Vegetable producers want more information on the sources of
contamination
Academic and government researchers have not yet confirmed
the major sources of pre-harvest contamination nor have the
mechanisms of survival and persistence of human pathogens on
produce been deciphered. However, progress is being made in both
areas (Beuchat, 2006; Ng, Fleet, & Heard, 2005; Tyler & Triplett,
2008; Wachtel, Whitehand, & Mandrell, 2002). Identifying high-
risk sources of contamination will be essential to developing
effective preventative management strategies. National GAPs
guidelines focus on management practices pertaining to the “Four
W’s”ewater, waste, worker health and hygiene and wildlife. For
each, guiding principles for crop production have been developed
based on the highest perceived risk factor in each category. In
addition to the “Four W’s”, vegetable producers indicated that pre-
harvest pests, specifically plant pathogens and insects, were likely
sources of produce contamination.
Despite the fact that seed and transplants have long been known
sources of phytopathogens (Baker & Smith, 1966), most respon-
dents did not believe that they were potential sources of human
pathogen contamination. The role of plant pathogens (Aruscavage,
Lee, Phelan, & LeJeune, 2010; Aruscavage, Miller, Lewis Ivey, Lee, &
LeJeune, 2008; Barak & Liang, 2008; Brandl, 2008; Warriner &
Namvar, 2010) and insects (Sela, Nestel, Pinto, Nemny-Lavy, &
Bar-Joseph, 2005) as contaminants of fresh produce is currently
being investigated and there is great potential for this new
knowledge to be leveraged and added to the current GAP
guidelines.
4.7. Vegetable producers believe that bacteria are the most
important contaminants of fresh produce
Although viruses account for approximately 20 million cases of
foodborne illnesses in the United States (Scallan et al., 2011), more
than any other pathogenic group, the majority of vegetable
producers believed that the bacteria E. coli (pathogenic or generic,
non-pathogenic) and Salmonella spp. were the most important
pathogens of fresh produce. The origin of this belief was not
explored in this survey but could be explained by the mass media
campaigns ensuing recent foodborne disease outbreaks. While
recurrent and deadly multistate outbreaks caused by E. coli
O157:H7 and Salmonella continue to make national and interna-
tional news headlines, those caused by viruses such as Norovirus
(Norwalk-like virus) often do not. Since the public makes judg-
ments on an event based on how easily they can recall past
occurrences (Tversky & Kahneman, 1983), how memorable the
event or the likelihood of a fatal impact or catastrophe (Slovic,
1993; Slovic, Fischhoff, & Lichtenstein, 1981; Slovic, Finucane,
Peters, & MacGregor, 2004), it is reasonable that bacterial patho-
gens are in the forefront of producers minds. This belief may alsobe
explained by the fact that viruses, such as Calicivirus (Norwalk-like
virus) produce milder symptoms than other gastroenteric viruses
or bacteria, resulting in fewer hospital visits, less reporting to the
CDC and less media attention (Mead et al., 1999).
Most producers also believed that coliforms were important
pathogens of fresh produce. Communications by experts should
better explain that these bacteria are not considered to be human
pathogens, generally do not present a health risk to consumers, and
when detected in water or on fresh produce, only indicate
a possible presence of harmful, disease-causing bacteria.
4.8. Vegetable producers prefer to receive their information in-
person and from university extension
Recognizing individual perceptions and beliefs is key to
successful message development (Lundgren, 1994). To effectively
disseminate information that target audiences self-report as being
most needed, audience-appropriate channels and trusted sources
(Morgan et al., 2002) for dissemination must be identified.
However, the number of choices available for information transfer
is extensive and leveraging the best and most preferred way to
reach desired audiences is a difficult challenge. Consumer prefer-
ence for mass media and internet tools has been documented in
recent studies (FDA, 2000; Cody & Hogue, 2003; Freimuth,
Linnman, & Potter, 2000; International Food Information Council
Foundation, 2010; Redmond & Griffith, 2006). However, Midwest-
ern vegetable producers overwhelmingly preferred personal
channels to receive information about food safety. Similar to other
farmer-targeted surveys throughout the United States (Ngathou,
Bukenya, & Chembezi, 2006; Riessenberg & Obel Gor, 1989;
Schnitkey, Batte, Jones, & Botomogno, 1992; Suvedi, Lapinski, &
Campo, 2000), our survey indicated that electronic modes of
communication were the least preferred. Given the decreased
funding of US Cooperative Extension Services (CES), emphasis has
been placed on electronic means of information transfer. While
there are obvious benefits to electronic approaches, a substantial
portion of the target audience may be overlooked or unreached. For
example, the Amish, who are believed to be the fastest growing
population in the world (Ericksen, Ericksen, Hostetler, &
Huntington, 1979) and have an estimated population size of
249 500 in the United States (of which 48.7% are located in OH, MI,
KY and IN) (Young Center for Anabaptist and Pietist Studies, 2010)
voluntarily forego use of technologies such as television, radios and
personal computers (Kraybill & Nolt, 2004) and hence do not access
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465 463

their information using on-line resources. In addition, ethnic
minority farmers (i.e. Blacks or African Americans, American
Indians and Asians) have less access to the Internet, especially high-
speed Internet, than non-minority farmers (USDA-NASS, 2007).
Overall, vegetable producers in the Midwest prefer to receive
food safety-related information from university extension. In other
studies conducted in OH, Oregon (OR), and FL, agricultural
producers and other clients also indicated a high level of satisfac-
tion with university extension (Meadowbrook & Fletcher, 1988;
Schnitkey et al., 1992; Warnock, 1992). In contrast, in an Internet-
based survey (International Food Information Council Foundation,
2010), consumers cited government agency/officials and health
professionals as their most trusted sources for food safety infor-
mation and CES as their least trusted source. However, consumers
are less likely than farmers to be aware of CES (Warner,
Christenson, Dillman, & Salant, 1996). The identification of CES as
a preferred source of information by Midwestern vegetable
producers should not be generalized to all vegetable producers nor
should it be considered as an “either/or”choice. Producer prefer-
ences can be regionally biased and depend on other factors such as
accessibility to CES specialists and producer usage of CES (Lavis &
Blackburn, 1990). For example, vegetable CES Specialists in the
western United States have reported that for day-to-day concerns
growers mostly consult industry sources for their food safety
information but these sources typically obtain information from
university researchers or CES (S. Koike, personal communication).
In society the need for communication is heterogeneous (Renn,
2006), requires differentiation according to target groups (Yosie &
Herbst, 1998) and must therefore be accurately defined for each
group. It is critical that food safety educators in both the private and
public sectors recognize that there are clear differences between
consumers and producers with regard to how they prefer to receive
risk communications as they seek to develop the most effective
means of communication given current funding constraints.
5. Study limitations
As in similar survey-based research, the main limitations of
these types of studies pertain to survey data. These include,
coverage errors, non-response and measurement errors and
selection bias. The response rate in this study was low (26.4%) with
a high margin of error (6.6%). There was an unequal response rate
from each state with only 7.1% of KY producers responding to the
survey.
6. Conclusions
The dominant beliefs, perceptions and food safety practices of
vegetable producers in the Midwest were identified using a survey-
based confirmatory assessment. Having a better understanding of
the beliefs and management practices of vegetable producers, as
they pertain to produce safety, will allow us to develop risk
management and communication materials that specifically target
the knowledge needs and wants of our stakeholders. Through the
development of highly targeted informational materials we hope to
provide vegetable producers accurate information on pre-harvest
vegetable food safety hazards so that they can make science-
based management decisions that will allow them to produce
fresh vegetables that are safe, sound and secure.
Acknowledgments
This work was supported by the National Integrated Food Safety
Initiative (NIFSI)-United States Department of Agriculture (Grant
No. 2007-51110-03817). We thank Pamela Schlegel for her
administrative expertise and liaisons with The Ohio State Univer-
sity Office of Responsible Research Practices. We thank Drs. Pierce
Paul and Laurence Madden for their invaluable suggestions on data
analyses. We also thank Drs. Sanja Ilic and Sunjeong Park for critical
reviews of this manuscript.
References
Abbot, J. M., Byrd-Bredbenner, C., Schaffner, D., Bruhn, C. M., & Blalock, L. (2009).
Comparison of food safety cognitions and self-reported food-handling behav-
iors with observed food safety behaviors in young adults. European Journal of
Clinical Nutrition, 63, 572e579.
Alliance for Food and Farming. (2010). Analysis of produce related foodborne illness
outbreaks. Alliance for Food and Farming, Available at www.foodandfarming.
info.
Arthur, L., Jones, S., Fabri, M., & Odumeru, J. (2007). Microbial survey of selected
Ontario-grown fresh fruits and vegetables. Journal of Food Protection, 70,
2684e2867.
Aruscavage, D., Lee, K., Miller, S. A., & LeJeune, J. T. (2006). Interactions affecting the
proliferation and control of human pathogens on edible plants. Journal of Food
Science, 71,89e99.
Aruscavage, D., Lee, K., Phelan, P. L., & LeJeune, J. T. (2010). Impact of changes in
sugar exudate created by biological damage to tomato plants on the persistence
of Escherichia coli O157:H7. Journal of Food Science, 75,187e192.
Aruscavage, D., Miller, S. A., Lewis Ivey, M. L., Lee, K., & LeJeune, J. T. (2008). Survival
and dissemination of Escherichia coli O157: H7 on physically and biologically
damaged lettuce plants. Journal of Food Science, 71, 2384e2388.
Atman, C. J., Bostrom, A., Fischhoff, B., & Granger Morgan, M. (1994). Designing risk
communications: completing and correcting mental models of hazardous
processes, Part 1. Risk Analysis, 14, 779e788.
Avery, D. T. (1998). The hidden dangers in organic food: Products most people think are
purer than other foods are making people seriously ill. Washington, D.C.:
Hudson Institute. Available at http://www.hudson.org/index.cfm?fuseaction¼
publication_details&id¼119 6 .
Baker, K. F., & Smith, S. (1966). Dynamics of seed transmission of plant pathogens.
Annual Review of Phytopathology, 4,311e332.
Barak, J. D., & Liang, A. S. (2008). Role of soil, crop debris, and a plant pathogen in
Salmonella enterica contamination of tomato plants. PLoS ONE, 3,e1657.
doi:10.1371/journal.pone.0001657.
Beuchat, L. R. (2006). Vectors and conditions for preharvest contamination of fruits
and vegetables with pathogens capable of causing enteric diseases. British Food
Journal, 108,38e53.
Bohaychuk, V. M., Bradbury, R. W., Dimock, R., Fehr, M., Gensler, G. E., King, R. G.,
et al. (2009). A microbiological survey of selected Albert-grown fresh produce
from farmers’markets in Alberta, Canada. Journal of Food Protection, 72,
415e420.
Bostrom, A., Fischhoff, B., & Morgan, M. J. (1992). Characterizing mental models of
hazardous processes: a methodology and an application to radon. Journal of
Social Issues, 48,85e100 .
Bostrom, A., Fischhoff, B., & Morgan, M. J. (1994). What do people know about global
climate change? Risk Analysis, 14,959e970.
Brandl, M. (2008). Plant lesions promote the rapid multiplication of Escherichia coli
O157:H7 on postharvest lettuce. Applied and Environmental Microbiology, 74,
5285e5289.
Cattaneo, A., Wislon, R. S., Doohan, D., & LeJeune, J. T. (2009). Bovine veterinarians’
knowledge, beliefs, and practices regarding antibiotic resistance on Ohio dairy
farms. Journal of Dairy Science, 92,3494e3502.
Centers for Disease Control and Prevention. (2007). Multistate outbreaks of
Salmonella infections associated with raw tomatoes eaten in restaurants-United
States, 2005e2006. MMWR, 56, 909e911.
Cody, M. M., & Hogue, M. A. (2003). Results of the “home food safety-it’s in your
hands”2002 survey: comparisons to the 1999 benchmark survey and health
people 2010 food safety behaviors. Journal of the American Dietetic Association,
103,1115e1125.
Cox, P., Niewohner, J., Pidgeon, N., Gerrard, S., Fischhoff, B., & Riley, D. (20 03). The
use of mental models in chemical risk protection: developing a generic work-
place methodology. Risk Analysis, 23,311e312.
Cummings, K., Barrett , E., Mohle-Boetani, J. C., Brooks, J. T., Farrar, J., Hunt, T.,
et al. (2001). A multistate outbreak of Salmonella enterica serotype Baildon
associate d with domes tic raw tomatoe s. Emerging Infectious Diseases, 7,
104 6e1048.
D’Souza, G., Cyphers, D., & Phipps, T. (1993). Factors affecting the adoption of
sustainable agricultural practices. Agricultural and Resource Economics Review, 2,
159e165.
DeWaal, S., Tian, X., & Plunkett, D. (2009). Outbreak alert! Analyzing foodborne
outbreaks 1998e2007. Center for Science in the Public Interest. Available at
www.cspinet.org.
Dillman, D. A. (2007). Mail and Internet surveys-the tailored design method (2nd ed.).
New Jersey: John Wiley & Sons, Inc.
Dunn, O. J. (1964). Multiple comparisons using rank sums. Technometrics, 5,
241e252.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465464

Ericksen, J. A., Ericksen, E. P., Hostetler, J. A., & Huntington, G. E. (1979). Fertility
trends and patterns among the old order Amish. Population Studies, 33,
255e276.
Food and Drug Administration (FDA) and US Department of Agriculture, Food Safety
and Inspection Service (USDA FSIS). (2000). Food safety survey: Pathogen
reduction/HACCP rule evaluation report: Focus group study on food safety.
messages and delivery mechanisms. Available at www.fsis.usda.gov/oa/
research/fsmessages.pdf.
Francis, C. A. (2009). Conventional research on controversial issues: an exercise in
futility? Renewable Agriculture and Food Systems, 25,3e7.
Franz, E., Semenov, A. V., & van Bruggen, A. H. C. (2008). Modelling the contami-
nation of lettuce with Escherichia coli O157:H7 from manure-amended soil and
the effect of intervention strategies. Journal of Applied Microbiology, 105,
1569e1584.
Freimuth, V., Linnman, H. W., & Potter, P. (2000). Communicating the threat of
emerging infections to the public. Emerging Infectious Diseases, 6,337e347.
Galli, N. (1978). Foundations and principles of health education. New York: Wiley.
International Food Information Council Foundation. (2010). 2010 Food and health
survey: Consumer attitudes toward food safety, nutrition and health. Available at
www.foodinsight.org.
Jungerman, H., Shutz, H., & Thuring, M. (1988). Mental models in risk assessment:
informing people about drugs. Risk Analysis, 8,147e155.
Kraybill, D. B., & Nolt, S. M. (2004). Amish enterprise; from plows to profits (2nd ed.).
Baltimore: Johns Hopkins University Press.
Kruskal, W. H., & Wallis, W. A. (1952). Use of ranks in one-criterion variance
analysis. Journal of the American Statistical Association, 47, 583e621.
Kruskal, W. H. (1952). A nonparametric test for the several sample problem. Annals
of Mathematical Statistics, 23, 525e540.
Lavis, K. R., & Blackburn, D. J. (1990). Extension clientele satisfaction. Journal of
Extension, 28, Available at http://www.joe.org/joe/1990spring/rb4.php Article
1RIB4.
Leifert, C., Ball, K., Volakakis, N., & Cooper, J. M. (2008). Control of enteric pathogens
in ready-to-eat vegetable crops in organic and ‘low input’production systems:
a HACCP-based approach. Journal of Applied Microbiology, 105,931e950.
Lundgren, R. (1994). Risk communication: A handbook for communication of envi-
ronmental safety and health risks. Columbus: Battelle Press.
Lynch, M. F., Tauxe, R. V., & Herberg, C. W. (2009). The growing burden of foodborne
outbreaks due to contaminated fresh produce: risks and opportunities. Epide-
miology and Infection, 137,307e315.
Marik, M., & Fischhoff, B. (1992). The risk of using nuclear energy sources in space:
some lay activists’perceptions. Risk Analysis, 12, 383e392.
Mead, P. S., Slutsker, L., Dietz, V., McCaig, L., Bresee, J. S., Shapiro, C., et al. (1999).
Food related illness and death in the United States. Emerging Infectious Diseases,
5, 607e625.
Meadowbrook, A., & Fletcher, R. L. (1988). It’s worth the effort: understanding our
clients. Journal of Extension, 26, Available at http://www.joe.org/joe/1988fall/a6.
php Article 3FEA6.
Mood, A. M. (1954). On the asymptotic efficiency of certain nonparametric two-
sample tests. Annals of Mathematical Statistics, 25,514e522.
Morgan, M. G., Fischhoff, B., Bostrom, A., & Atman, C. J. (2002). Risk communication:
A mental models approach. Cambridge: Cambridge University Press.
Ng, P. J., Fleet, G. H., & Heard, G. M. (2005). Pesticides as a source of microbial
contamination of salad vegetables. International Journal of Food Microbiology,
101,237e250.
Ngathou, I. N., Bukenya, J. O., & Chembezi, D. M. (2006). Managing agricultural risk:
examining information sources preferred by limited resource farmers. Journal
of Extension, 44, Available at http://www.joe.org/joe/2006december/a2.php
Article 6FEA2.
Oliveira, M., Usall, J., Vinas, I., Anguera, M., Gatius, F., & Abadias, M. (2010). Micro-
biological quality of fresh lettuce from organic and conventional production.
Food Microbiology, 27, 679e684.
Orlich, S. (2000). Multiple comparisons with a MINITAB macro: Dunn’s test. State
College, Pennsylvania: Minitab, Inc.
Palma, M. A., Ribera, L. A., Paggi, M., & Knutson, R. (2010). Food safety standards for
the U.S. fresh produce industry. Policy Issues,P18.
Parish, M. E., Beuchat, L. R., Suslow, T. V., Harris , L. J., Garrett, E. H., Farber, J. N.,
et al. (2003). Methods to reduce/eli minate pathogens from fresh and fresh-
cut produce. Comprehensi ve Reviews in Food Sci ence and Food Safety, 3,
161e173.
Parker, J., Wilson, R., Le Jeune, J. T., Rivers, L., III, & Doohan, D. (2012). An expert
guide to understanding grower decisions related to fresh fruit and vegetable
contamination prevention and control. Food Control, 26,107e116.
Patil, S. R., Cates, S., & Morales, R. (2005). Consumer food safety knowledge, prac-
tices, and demographic differences: findings from a meta-analysis. Journal of
Food Protection, 68, 1884e1894.
Powell, D. A., Jacob, C. J., & Chapman, B. J. (2010). Enhancing food safety culture to
reduce rates of foodborne illness. Food Control, 22,817e822.
Prokopy, L. S., Floress, K., Klotthor-Weinkauf, D., & Baumgart-Getz, A. (2008).
Determinants of agricultural best management practice adoption: evidence
from the literature. Journal of Soil and Water Conservation, 63,300e311.
Redmond, E. C., & Griffith, C. J. (2003). Consumer food handling in the home:
a review of food safety studies. Journal of Food Protection, 66,130e161.
Redmond, E. C., & Griffith, C. G. (2006). Assessment of consumer food safety
education provided by local authorities in the UK. British Food Journal, 108,
732e752.
Renn, O. (2006). Risk communication: consumers between information and irri-
tation. Journal of Risk Research, 9,833e849.
Riessenberg, L. E., & Obel Gor, C. (1989). Farmers’preferences for emthods of
receiving information on new or innovative farming practices. Journal of Agri-
cultural Education.30. Available at doi:10.5032/jae.1989.030 07.
Rimal, A., Fletcher, S. M., McWatters, K. H., Misra, S. K., & Deodhar, S. (2001).
Perceptions of food safety and changes in food consumption habits: a consumer
analysis. International Journal of Consumer Studies, 25,43e52.
Rodriguez, J. M., Molnar, J. J., Fazio, R. A., Sydnor, E., & Lowe, M. J. (2008). Barriers to
adoption of sustainable agriculture practices: change agent perspectives.
Renewable Agriculture and Food Systems, 24,60e71.
Saltiel, J., Bauder, J. W., & Palakovich, S. (1994). Adoption of sustainable agricultural
practices: diffusion, farm structure and profitability. Rural Sociology, 59,
333e349.
Scallan, E., Hoekstra, R. M., Angulo, F. J., Tauxe, R. V., Widdowson, M. A., Roy, S. L.,
et al. (2011). Foodborne illness acquired in the United States: major pathogens.
Emerging Infectious Diseases, Available at http://www.cdc.gov/EID/content/17/1/
7.htm.
Schnitkey, G., Batte, M., Jones, E., & Botomogno, J. (1992). Information preferences of
Ohio commercial farmers: implications for extension. American Journal of
Agricultural Economics, 74, 486e496.
Sela, S., Nestel, D., Pinto, R., Nemny-Lavy, E., & Bar-Joseph, M. (2005). Mediterranean
fruit fly as a potential vector of bacterial pathogens. Applied and Environmental
Microbiology, 71, 4052e4056.
Slovic, P. (1993). Perceived risk, trust, and democracy. Risk Analysis, 13,82e675.
Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. G. (2004). Risk as analysis and
risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk
Analysis, 24,311e322.
Slovic, P., Fischoff, B., & Lichtenstein, S. (1981). Perceived risk: psychological factors
and social implications. In Proceedings of the royal society, risk report A376 (pp.
17e34). London: Royal Society.
Soil Association. (2001). Organic food and farming-myth and reality. Bristol: The Soil
Association and Sustain: the alliance for better food and farming.
Spearman, C. (1904). The proof and measurement of association between two
things. Journal of Psychology, 15,72e101.
Stephenson, J. (1997). New approaches for detecting and curtailing foodborne
microbial infections. Journal of the American Medical Association, 277,
1337 e1339.
Suvedi, M., Lapinski, M. K., & Campo, S. (2000). Farmers’perspectives of Michigan
State university extension: trends and lessons from 1996 and 1999. Journal of
Extension, 38, Available at http://www.joe.org/joe/2000february/a4.php Article
1FEA4.
Teplitski, M., Barak, J. D., & Schneider, K. R. (2009). Human enteric pathogens in
produce: un-answered ecological questions with direct implications for food
safety. Current Opinion in Biotechnology, 20,166e171.
Tversky, A., & Kahneman, D. (1983). Extensional versus intuitive reasoning: the
conjunction fallacy in probability judgment. Psychological Review, 90, 293e315 .
Tyler, H. L., & Triplett, E. W. (2008). Plants as a habitat for beneficial and/or human
pathogenic bacteria. Annual Review of Phytopathology, 46,53e73.
United States Department of Agriculture-National Agricultural Statistics Service
(USDA-NASS). (2007). 2007 Census of agriculture report. Available at http://
www.agcensus.usda.gov/Publications/2007/Full_Report/index.asp.
Vanclay, F., & Lawrence, G. (1994). Farmer rationality and the adoption of envi-
ronmentally sound practices; a critique of the assumptions of traditional agri-
cultural extension. The Journal of Agricultural Education and Extension, 1,59e90.
Wachtel, M. R., Whitehand, L. C., & Mandrell, R. E. (2002). Prevalence of Escherichia
coli associated with a cabbage crop inadvertently irrigated with partially
treated sewage wastewater. Journal of Food Protection, 65,471e475.
Warner, P. D., Christenson, J. A., Dillman, D. A., & Salant, P. (1996). Public perception
of extension. Journal of Extension, 34, Available at http://www.joe.org/joe/
1996august/a1.php Article 4FEA1.
Warnock, P. (1992). Surveying client satisfaction: Florida counties study their
information services. Journal of Extension, 30, Available at http://www.joe.org/
joe/1992spring/a1.php Article 1FEA1.
Warriner, K., & Namvar, A. (2010). The tricks learnt by human enteric pathogens
from phytopathogens to persist within the plant environment. Current Opinion
in Biotechnology, 21,31e136.
Wilson, R. S., Parker, J., Kovacs, D., Doohan, D., & LeJeune, J. T. (2009). Contamination
prevention and response related to fresh and fresh-cut produce: an expert
perspective on the farmer decision making process. Food Protection Trends, 29,
16e20.
Wilson, R. S., Tucker, M. A., Hooker, N., LeJeune, J. T., & Doohan, D. (2008).
Perceptions and beliefs about weed management: perspectives of Ohio grain
and produce farmers. Weed Technology, 22,339e350.
Wittwer, J., Nuckles, M., & Renkl, A. (2005). What happens when experts over-or
underestimate a laypersons knowledge in communication? Effects on
learning and question asking. In Proc. CogSci 2005.
Yosie, T. F., & Herbst, T. D. (1998). Managing and communicating stakeholder-based
decision making. Human and Ecological Risk Assessment, 4,643e646.
Young Center for Anabaptist and Pietist Studies, Elizabethtown College. (2010).
Amish population by state. Available at http://www2.etown.edu/amishstudies/
Population_by_State_2010.asp.
Zaksek, M., & Arvai, J. L. (2004). Toward improved communication about wildlife
fire: mental models research to identify information needs for natural resource
management. Risk Analysis, 24, 1503e1514.
M.L. Lewis Ivey et al. / Food Control 26 (2012) 453e465 465