Food Protection Trends - May/June 2017 - 163

Canadian food industry and governmental organizations
on the current food recall system. Understanding what
does and does not make our food recall system effective
is important for establishing, maintaining, and improving
practices and procedures. Ultimately, the collection of
personal accounts and suggestions may help to generate
new streams of thought within the food industry and
inspire the modification of current recall procedures
of food manufacturers and/or shape future government
processes and regulations.
MATERIALS AND METHODS
Prior to the beginning of data collection, this project
was approved by the Research Ethics Board at the University
of Guelph.
Instrumentation
To determine what makes a food recall effective from
industry and government perspectives, in-depth, semistructured interview guides were created. The interview
guide for industry members was similar to that for government officials. The in-depth interview questions were
created based on the literature, in conjunction with
suggested changes from the pretesting process.
To pretest, two individuals (one industry, one government) of similar background to that of the target population
(i.e., those involved in the semi-structured interviews) were
interviewed with the interview draft guide, which had been
adjusted to a grade six readability level. The pretesting was
carried out in a manner similar to that of the regular interviews,
with the pretesting participants answering each question.
During a debriefing session immediately following the
interview, the respondent and the interviewer discussed
each question, as well as the process of data collection.
This process was used to help detect and reduce any issues
regarding sensitive or offensive questions, misinterpretation
of questions, ambiguity of wording, double-barreled
questions, or leading questions (i.e., questions that reflect the
researcher's personal values, blind spots, or biases) (1).
Minor changes were suggested for both the government
and industry interview guides, and were made accordingly
prior to the interviews being conducted. The final interview guide used for participants in industry roles differs
slightly from that used for government officials. The
semi-structured interviews involved the pre-determined
questions but also allowed the interviewer to digress with
unscheduled probes (1). The interview schedule consisted
of essential questions that concerned the purpose of the
study, extra questions that were worded slightly different
to the essential questions but were roughly equivalent,
"throw-away questions" that helped develop rapport between
the interviewer and interviewee, and probing questions that
helped elicit more information (1).

Subject recruitment
Participants were sought from the network of professional connections of the research team. Those who had a
current or significant past role involving food recalls, either
in a government or industry position, were sent the Letter
of Information and consent form by email and were asked
to participate. Twenty-one email invitations were sent from
the research team to members of industry (n = 17) and
government (n = 4).
Conducting and analyzing interviews
Of the 21 direct invitations, 9 led to interviews, which were
conducted between July and November 2015. According
to McCracken, interviewing 8 respondents is "perfectly
sufficient"; it is better to work for a longer amount of time
with a few people than superficially with many (19). The
demographic profiles of the 9 interviewees, 6 from industry
and 3 from government, are presented in Table 1.
Because of the range of geographical locations of
participants, interviews were conducted over the telephone
or via Skype video call. The interviews, which ranged from
25 to 65 minutes, were audio-recorded and transcribed.
A second party verified the verbatim transcripts, and the
transcripts were then qualitatively content-analyzed using
the NVivo 11 Pro software, which helps to organize, store,
and retrieve unstructured data and to uncover any trends
existing within the data.
The credibility of the data was assessed through
confirmation of the interpretation of the data with the
interview participants and through discussion with the
food safety specialists involved during the pretesting
stage. The combination of in-depth interviews from both
industry and government, as well as the inclusion of a
detailed methodology, improves the dependability of the
findings (22). Additionally, individuals with food safety
experience or knowledge assessed the confirmability/
objectivity of the findings by reviewing the methodology
and the interpretation of data. The completed manuscript was sent to each of the interviewees for review
and comment, which further validated the interpretation
of the transcripts.
RESULTS
The broadness of the interview questions allowed for
open-ended and varied participant responses, which led
to a number of recurring topics.
Internal preparation
Many of the industry participants (4 of 6) stressed the
importance of having a trained and prepared core recall team.
One participant (IND-6) expressed that the biggest mistake
that the participant's company had made in past recalls is not
engaging the core recall team right from the trigger of a recall.
Upon correction of this issue, the participant believed the
company's recall response time was improved dramatically.
May/June Food Protection Trends

163

Table of Contents for the Digital Edition of Food Protection Trends - May/June 2017