Seven students participated in various aspects of the studies.
These projects were challenging and time intensive for the students
who participated. For instance, in the 2018 study, Phase 2 involved
a total 17 hours of field observations. Note, however, that the design
of the project can be modified to provide opportunities for larger
groups of students to be involved, by dividing up the research tasks
accordingly. For instance, a class of 24 can be divided into 12 teams
of two researchers or eight teams of three researchers. With additional student researchers, time-intensive tasks can be divided up,
the number of observations per student team is more manageable,
and the project can be conducted as a unit within a formal course.
Students can be involved in all aspects of the study or be assigned
tasks focusing on fieldwork or lab work. As long as inter-rater reliability is considered, larger numbers of students can conduct fewer
or shorter periods of observation. At the end of the study, students
can pool their data and work together as a larger group to analyze,
discuss, and present the data.
In the first study (Basch et al., 2016), students completed field
observations at 25 street food vendors in close proximity to five
major medical centers in New York City, designated locations A–E
(see Table 1). Food vendors near hospitals were chosen for this study
because such vendors are typically patronized by hospital workers
and visitors who bring food into the hospital, as well as by outpatients
with compromised immunity, who may be particularly vulnerable to
foodborne illness. The data collection process included documenting
the number of monetary transactions and the number of subsequent
glove changes following a monetary transaction.
The first step of each contact with a vendor was to collect a dollar
bill for testing. Students worked in pairs, and while one student
made a purchase with a higher-denomination bill ($5, $10, or
$20), the second student donned individually wrapped, single-use
sterile examination gloves. The gloved student received the change,
extracting a single dollar bill, which was immediately placed inside a
sterile tube containing 40 mL of sterile tryptic soy broth. In order to
maintain the best practice of aseptic technique possible in a public
setting, the ungloved student handled and opened each tube only
long enough for the gloved student to insert the sample. Samples
were stored on ice in an ice chest for transport to the laboratory
where they were tested for bacterial contamination. The process of
collection of currency was repeated using a fresh pair of sterile gloves
for each dollar bill sample collected, for a total of 25 separate trans-
actions (25 dollar bill samples) from five vendors.
In the lab, a 100 µL sample was removed from each test tube,
plated on tryptic soy agar, and incubated at 37°C overnight, or at
35°C for up to four days. Further dilutions of samples were prepared and plated as necessary to facilitate colony counting.
Table 1 summarizes the data, showing that in 495 observed monetary transactions a mere seven glove changes were completed by food
workers. Note that 11 of 34 food workers wore no gloves at all while
handling money and food (Basch et al., 2016). Nineteen of the 25 bills
collected (76%) generated bacterial colonies, with from 400 to 42,000
colony-forming units (CFUs) of varied morphology and size. Thirteen
of the samples with the highest CFUs were chosen for coliform bacteria testing using the Colilert test (Idexx Laboratories, n.d.), and 10 of
those samples tested positive for coliform bacteria.
It should be noted that in another study (Basch et al., 2015), observations of food vendor behavior alone were conducted in a similar fashion, but that study lacked the important element of a microbial analysis.
Hence, we were only able to tell part of an important story. In other
words, we observed that gloves were not often changed by food vendors after collecting money, but the fact that money can carry microbes,
which could be transferred to the customer, was not documented.
The addition of microbiological analysis to this project enabled our students to appreciate the link between disciplines, and it made the conclusions of the study more compelling – that food vendors potentially
endanger customers when they ignore health codes for food handling.
The second interdisciplinary study (Basch et al., 2018) focused on
17 mall food vendors in New Jersey. This study involved two phases:
Phase 1, which consisted of 10-minute observation periods capped off
by a transaction and collection of dollar bills; and Phase 2, which
involved 60-minute observation periods. Table 2 shows that in 175
transactions observed in Phase 1 there were four glove changes
(2.3%). In 1193 transactions observed in Phase 2 there were 40 glove
changes (3.35%). Seventeen dollar bill samples were collected from
the 17 vendors in Phase 1, and all the samples generated bacterial
CFUs. Six of the 17 samples (35.3%) were coliform positive.
The Institutional Review Board/Human Subjects Committees
at William Paterson University and Teachers College, Columbia
Table 1. New York City mobile food vendors study: summary of food cart data and biological testing by
Total no. of
Total no. of
Total no. of
times the food
Percent of tested
were positive for
A/5 120 115 0 100% 400–13,600 100%, 4 of 4
B/5 130 126 4 100% 800–42, 400 25%, 1 of 4
C/5 113 112 0 40% 0–2000 100%, 1 of 1
D/5 122 94 3 60% 0–2400 100%, 1 of 1
E/5 49 48 0 80% 0–22, 400 100%, 3 of 3
Total 534 495 7 0–42, 400
aFrom Basch et al. (2016); reprinted with permission.