Food Protection Trends - January/February 2015 - 15

to external hard drives for viewing on a computer monitor.
For ease of tracking, the video recordings were labeled with
the name of the researcher recording the meal preparation,
followed by the subject ID, which is a numerical number
assigned to the subject (e.g., Eyob_01).

subject failed to wash his/her hands, the second researcher
would also have to identify between 18 to 22 failures. If
divergence in excess of 10% between reviewers' recorded
observations was found, the recorded videos were reviewed
jointly and differences harmonized.

Administering questionnaires
The written questionnaires (Attachment A) were
administered after the video recordings were completed, to
minimize the impact of the questionnaires on the observed
behaviors. The questionnaires consisted of 51 questions
in six distinct sections: (1) consumption/preparation
frequencies, (2) product transportation, (3) storage/
handling temperatures, (4) poultry handling/sanitation,
(5) hand washing and (6) final cook temperatures. Each
questionnaire was labeled with the same codes used for the
video recordings.

Data analysis
Microsoft Excel (Microsoft Corporation, Redmond,
Washington) was used to compile data, summarize data
points, and calculate basic descriptive statistics (e.g.,
mean, standard deviation, percentages of responses and
observations). Fisher exact tests were used to evaluate
significant differences between reported and observed
food handling practices (GraphPad Calculator, GraphPad
Software, Inc., San Diego, CA). T-tests were performed to
evaluate the differences in cross-contamination episodes
relative to side dish preparation (GraphPad Calculator,
GraphPad Software, Inc., San Diego, CA).
RESULTS
Questionnaire results
Demographics
Between November 01, 2011 and December 15, 2012,
56 subjects were recruited. None of the study subjects
were students, faculty, or staff from the academic
department of the researchers, and more than half of the
subjects had no affiliation with the university. The racial
distribution of subjects in the study was 7% (4) African
American; 36% (20) Asian; 54% (30) Caucasian; and 3%
(2) others. The ages of subjects in the study ranged from
18 to 65 years old, with a median age of 21 years. Fiftythree percent of subjects were female. Thirty-five percent
(17/48) of the participants possessed a bachelor's degree;
62 percent (30/48) reported some college education,
while one subject reported only a high school education.
Some participants left a few questions unanswered so
that the total number of respondents per question was
less than 56. Because of the sampling method used (i.e.,
convenience sampling), the participants may not represent
the general public. The educational level of the study
subjects is higher than that of the general public, and
the average age is less than that of the general public;
therefore, our data might be more reflective of the food
handling practices of the younger generation than of the
general population.
Poultry consumption
All study subjects reported that they regularly buy
chicken, with 14% reporting also buying ground turkey
products. Chicken breasts were the most commonly
reported items, with 86 percent of subjects reporting
buying them. Study participants reported that they
consumed meals with poultry from once a month to
daily. On average, participants reported preparing meals

foodprotection.org

Notational analysis
To prepare the records of food handling activities for data
analysis, a modified version of a notational analysis form
(Attachment B) was used to annotate subject videos (25).
In brief, notational forms were used to keep a detailed track
of each activity a subject preparing a meal performed from
the beginning to the end of the recording. The information
recorded included the type of activity (e.g., hands washed),
how the activity was done (e.g., soap, hot water, and paper
towel used), how long the activity lasted (e.g., 15 seconds),
and how hands were dried (e.g., cloth towel). Each step of
the food handling procedure was written in the first column
of the notational form. In addition, the notational form
had columns for tracking specific cross-contamination
activities (e.g., cloth towels contaminated), activities related
to sanitation of food contact surfaces (e.g., cutting board
washed, rinsed and sanitized before re-use), and activities
that could compromise the final cooked product (e.g., tasted
food with unwashed hands) or contamination activities
that could potentially compromise future food handling
(e.g., salt shaker contaminated). For each recorded step, a
determination of whether or not an activity was a proper
food handling practice (e.g., whether or not hand washing
was required before or after the step, or if and what kind of
cross contaminations occurred) was made, based on predetermined criteria adapted from the United States Food
and Drug Administration, Retail Food Code, 2009
(Table 1) (38).
Each video recording was reviewed by two researchers
and their observations recorded on separate notational
analysis forms. To ensure consistency in coding between
researchers (i.e., using FDA guidelines), a margin of 10%
was set as the allowable maximum divergence in measured
activities (e.g., total number of required hand washes). For
example, if after watching the video of subject "01", one of
the two researchers documents 20 separate occasions the

Food Protection Trends

http://www.foodprotection.org

Food Protection Trends - January/February 2015

Table of Contents for the Digital Edition of Food Protection Trends - January/February 2015