. wkvea: ..
n1 ‘ ,3

. IK
é , n .
fit. a. Q «1 ..
HM... .ufi.u?1n.m~d.u .h
11:3!

eat
1.7". .

.v’i n...
2.

. I

‘ . r
nun"...
«WI-.13. l

as}! if

- ‘I‘- ‘t
s. .
{in}
‘ .

3....

.
a: I

5.515.. I I! t V .
éufihaw. ..
L‘I\
? y . ur .
fir". ., ‘ y
g
:

hf...

 

 

 

5:33....
I”... ($1911 9
)I‘! é ..

 

 

3'! .
1.: 4

by" 3:1...

 

 

 

“'3 r- LIBRARY
Michigan State

University

 

L.

This is to certify that the
thesis entitled

IDENTIFICATION AND PRELIMINARY EVALUATION OF
THE MOST VIABLE SHORT AND LONG TERM SOLUTIONS
REGARDING TAMPER-EVIDENCE FOR AN ASEPTIC i
PUREE PACKAGED IN PLASTIC :

presented by
JULIA CORINNA BREISINGER

has been accepted towards fulfillment
of the requirements for the

MS. degree in Packaging

 

 

 

    

7, 20 M

Date

 

MSU is an Affirmative Action/Equal Opportunity Employer

r I...

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:lProleoc&Pres/CIRCIDateDue.indd

 

IDENTIFICATION AND PRELIMINARY EVALUATION OF THE MOST VIABLE
SHORT AND LONG TERM SOLUTIONS REGARDING TAMPER-EVIDENCE
FOR AN ASEPTIC PUREE PACKAGED IN PLASTIC

By

Julia Corinna Breisinger

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE

Packaging

2010

ABSTRACT

IDENTIFICATION AND PRELIMINARY EVALUATION OF THE MOST VIABLE
SHORT AND LONG TERM SOLUTIONS REGARDING TAMPER EVIDENCE
FOR AN ASEPTIC PUREE PACKAGED IN PLASTIC

By

Julia Corinna Breisinger

Tamper evident (TE) features have been mandatory for over-the-counter drug
products since 1982. For the vast majority of food products no such legislation
exists, and most efforts made are voluntary. Presented research evaluates
current commercial TE technologies and promising technologies relevant to
tamper evidency and explores the feasibility of a gas sensor as a possible
solution. To do so, headspace gas concentration of an aseptic puree packaged

in plastic was measured using gas chromatography. The analysis defines the
initial headspace gas concentration (02, C02) of four types of puree and the gas
concentration after two different tampering methods; the opening of the container
and the prick of a needle. Results suggest that a sensor can be used to detect
tampering that is similar to opening of the container. Although the gas
composition of the unopened product was significantly different from the product
that had its lid removed and reaffixed (P= 0.0004 for 02 and P= 0.0046 for C02),
the unopened product compared with a sample penetrated with a needle was not
significant (P= 0.95 for 02 and P= 0.29 for 002). This suggests that oxygen and
carbon dioxide sensors may be a plausible design in gross tampering, but will not

provide protection against ingresses which are minimally invasive.

Copyright by
JULIA CORINNA BREISINGER

2010

Acknowledgments

I wish to acknowledge Dr. Laura Bix, my major professor, for her never-
ending effort in assisting me with this research. Her help was invaluable and

appreciated.

My appreciation is extended to Dr. Eva Almenar, Jerry Holycross and Kelby

Thayer for their guidance and support with the GO.

A special thank you goes to the GEM team, especially to Cindee Wilcox and

Dr. Evangelyn Alocilja, for their support and input on this research.

I would also like to thank the Statistical Consulting Center at Michigan State

University for their kind assistance with the analysis of my data.

Furthermore, I would like to thank all the faculty, staff and graduate students

of the School of Packaging for their constant support.

And last but not least, I want to thank everyone who contributed to this work

and all my friends and family who had to keep up with me during that time.

Table of Contents

List of Tables ................................................................................................... xii
List of Figures ................................................................................................. xii
Key to Symbols or Abbreviations ................................................................... ix
1 Introduction ............................................................................................. 1
2 Literature Review .................................................................................... 4
2.1 History of tampering ........................................................................... 4
2.1.1 Level and Types of Tampering .............................................. 4
2.1.2 Reasons for product tampering .............................................. 5
2.1.3 How product tampering occurred in the past ......................... 6
2.2 Research on Tampering ................................................................... 12
2.3 Government Reaction to Tampering - US Legislation and
Regulation ........................................................................................ 20
2.3.1 Liquor ................................................................................... 21
2.3.2 Drugs ................................................................................... 22
2.3.3 Milk ...................................................................................... 26
2.4 Reactions of the Public: Post-tampering ........................................... 27
2.5 Evaluation Methods .......................................................................... 29
3 Literature Review: Identification and Analysis of Promising New
and Established Technologies Relevant to TE .................................. 32
3.1 Authentication ................................................................................... 33 .
3.2 Antitheft and Track-and-Trace .......................................................... 34
3.3 Tamper evidence .............................................................................. 35
3.3.1 Current commercial solutions .............................................. 36
3.3.1.1 Closures ...................................................................... 36
3.3.1.2 Labels, films and polymers .......................................... 37
3.3.1.3 RFID ............................................................................ 39
3.3.2 Commercially available incremental systems ...................... 40
3.3.2.1 Dyes and inks .............................................................. 40
3.3.2.2 Gas Detectors ............................................................. 41
3.3.2.3 Biobased Sensors ....................................................... 43
3.3.3 Potential “leap” solutions ...................................................... 43
4 Focus group .......................................................................................... 46
4.1 Focus Group characterization ........................................................... 46
4.2 Procedure ......................................................................................... 48
4.3 Background and experiences of participants .......... . .......................... 50
4.4 Reactions .......................................................................................... 53

5 Characterizing the product: Headspace Analysis .............................. 55

5.1 Gas Chromatograph Trace GC Ultra ................................................ 55

5.2 Calibration ........................................................................................ 58

5.3 Materials and Methods ..................................................................... 59

5.4 Null hypotheses ................................................................................ 62

5.5 Results and analysis ......................................................................... 63
5.5.1 Broad Scope ........................................................................ 67

5.5.1.1 Oxygen ........................................................................ 67

5.5.1.2 Carbon dioxide ............................................................ 70

5.5.2 Narrow Scope ...................................................................... 73

5.6 Discussion ........................................................................................ 75

5.7 Conclusions ...................................................................................... 77

6 Recommendations for future research ............................................... 78
Appendices ..................................................................................................... 79
Appendix A. History of Tampering ............................................................ 80
Appendix B. Field Study on Baby Food Packaging .................................. 85
Appendix C. Rosette Protocol .................................................................. 89
Appendix D. IRB Approval and IRB approved Moderator Guide .............. 93
Appendix E. Consent Form .................................................................... 101
Appendix F. Data Collection Sheet and Results .................................... 104
Appendix G. Matrix Development ........................................................... 105
Appendix H. Details of tested product, Raw Data ................................... 107
Bibliography .................................................................................................. 116

vi

List of Tables

Table 1. Focus group panelist’s background on tampering ................................. 48
Table 2. Definition of “tamper evident“ and “tamper resistant“ ............................ 51
Table 3. Calibration gas concentrations .............................................................. 58
Table 4. Legend for Headspace Tables and Graphs .......................................... 64
Table 5. Headspace Gas Data for 02 and C02 ................................................... 65
Table 6. 002 concentrations (means) for Sample Type x Expiration Date ......... 74
Table 7. Acidity and pH for the tested purees ..................................................... 76
Table 8. Tampering incidents of food and pharmaceutical products ................... 80
Table 9. Safety categories for protection against tampering ............................... 85
Table 10. Current baby puree packages evaluated with method used in the
1984 field survey [34] ............................................................................... 87
Table 11. Summary: Average % per safety categories ....................................... 87
Table 12. TE Testing Procedure: Values accorded for Factors [37] .................... 91
Table 13. Data Collection of participants in focus group ................................... 104
Table 14. Expiration dates of tested product ..................................................... 107
Table 15. Legend for raw data tables and graphs ............................................. 107
Table 16. Raw Data for Puree 1: Applesauce ................................................... 108
Table 17. Raw Data for Puree 2: Green Beans ................................................ 110
Table 18. Raw Data for Puree 3: Prunes .......................................................... 112
Table 19. Raw Data for Puree 4: Bananas ....................................................... 114

vii

List of Figures

Figure 1. Various Tampering incidents after 1982 ................................................ 7
Figure 2. Packaging examples complying with the FDA TE requirements [47-
50] ............................................................................................................ 26
Figure 3. Bericap Galileo ll, 2-pieces flip top pourer closure [67] ........................ 36
Figure 4. Tamper-evident container with tear band [71] ...................................... 37
Figure 5. TE folding carton (22: front flap, 30: window, 40: tamper indicating
seaD ......................................................................................................... 37
Figure 6. Diagram of GC flows for split injection [97] .......................................... 56
Figure 7. Calibration Curve: Oxygen ................................................................... 58
Figure 8. Calibration Curve: Carbon dioxide ....................................................... 59
Figure 9. Sample characterization ...................................................................... 61

Figure 10. Comparison of oxygen and carbon dioxide concentrations (means
of all puree types) for each tampering method per shelf life stage and

tampering time (if available) ..................................................................... 67
Figure 11. Comparison of 02 concentrations (means) for Tampering Method

x Tampering Time .................................................................................... 69
Figure 12. Comparison of CO; concentrations (means) for Tampering Method

x Tampering Time .................................................................................... 71
Figure 13. Comparison of 002 concentrations (means) for Tampering Method

x Expiration Date ...................................................................................... 72
Figure 14. Control group: Comparison of C02 concentrations (means) for

Sample Type x Expiration Date ................................................................ 74
Figure 15. Changes in 02 and 002 concentrations for ‘Lid removal’ tampered

samples .................................................................................................... 75
Figure 16. Tamper-Evident Feature Rating Index for a sealed plastic tub .......... 90
Figure 17. Initial IRB Application Approval .......................................................... 93
Figure 18. Matrix: Tampering Methods vs. Technologies ................................. 106
Figure 19. Headspace Variation: Applesauce ................................................... 109

viii

Figure 20. Headspace Variation: Green Beans ................................................ 111
Figure 21. Headspace Variation: Prunes .......................................................... 113

Figure 22. Headspace Variation: Bananas ....................................................... 115

Key to Symbols or Abbreviations

cGMP

FATA

FDA

GC

MAP

MEMS

OTC

OTR

TCD

TE

TI'I

RFID

Current Good Manufacturing Practice
Federal Anti-Tampering Act

Food and Drug Administration
Gas Chromatograph

Modified Atmosphere Packaging
Micro-Electro-Mechanical System
Over-the-counter

Oxygen Transmission Rate
Thermal Conductivity Detector
Tamper Evidence

Time Temperature Integrator

Radio Frequency Identification

1 Introduction

Tampering incidents of packaged consumer goods have been a regular
occurrence, especially since 1982, when Tylenol, an over-the-counter drug, was
tampered in Chicago. The 1982 event was highly publicized, and helped to spur
copy-cat events in varied products in the time since. The concern is present not
only for OTC drugs, but for all products that are meant to be ingested. Certain
products, such as baby food, incite particular concern. Babies don’t have a
strong immune system and can be harmed very easily. As recently as 2009,
baby food, manufactured by Earth’s Best and Gerber Products, was involved in
tampering incidents. In both cases, glass jars were tampered with at the store
(see Appendix A, Table 8. Tampering incidents of food and pharmaceutical
products). On the other hand, tampering threats also occurred in combination
with extortion. In 2005, a tampered Snickers bar was sent to Masterfoods in

Australia along with extortion threats.

If concerns like these cannot be eliminated, the popularity of eg. homemade
baby food will increase and the growth of prepared baby food, such as purees,
will decrease [1]. For baby food manufacturers, this trend would be critical and
drop their sales drastically. Food manufacturers, therefore, try to secure
themselves by implementing new technologies that help consumers to identify a
tampered package more easily or remove consumer involvement by applying

techniques that work at the retail level.

Literature review on the history of tampering and government regulations
proposes that the implementation of tamper evident regulations for OTC drugs
did not prevent tampering from happening (see 2.1 History of tampering). As
mentioned earlier, baby food and OTC drugs were targeted by tamperers in the
past (see also 2.1 and Table 8). Food products, however, are not required by
law to have tamper evident features. Although, they have tamper evident
features integrated voluntarily. Their implementation is good business practice
and utilizing a dedicated approach to protect and secure the product can create

consumer confidence.

For OTC drugs, FDA recommends several tamper evident packaging designs
in the Compliancy Policy Guide (CPG) 7132a.17 (see 2.3.2 Drugs, p. 22). Many
of these recommended tamper evident features can also be employed for food
products. However, there is no compliance guide of tamper evident technologies
for food products. This enables researchers to look into new and incremental
tamper evident solutions and their possible implementation (see 3 Literature
Review: Identification and Analysis of Promising New and Established
Technologies Relevant to TE). There has been little research on tamper evident
technologies after FDA published the CPG for drugs. Ohio State University
researchers evaluated a new tamper evident closure system in 2006, which is

most likely designed for the use with beverage containers (see 2.2).

This thesis covers the state-of-the-art in tamper evidence with a close look at
commercial tamper evident packaging solutions and provides a basis for

incremental and “leap” solutions for an aseptic puree packaged in plastic. The

current tamper evident feature of the aseptic puree can be described as a blister
pack according to the CPG (see p. 24). Here, a blister pack is defined as one
having individually sealed dosage units in clear plastic sealed with foil or paper

baking that must be broken or torn to obtain the product.

One tamper evident design that is further analyzed in this research is the use
of a gas sensor as a new tamper evident solution. Gas detectors are already in
use for food packaging, however, not as tamper evident feature (see 3.3.2.2 Gas
Detectors). To examine the feasibility of a gas sensor as tamper evident
technology, a product (puree) with a modified atmosphere in its headspace is
evaluated. Headspace analysis provides information which has the potential to
garner insights for future tamper evident features (see 5 Characterizing the

product: Headspace Analysis).

2 Literature Review

2.1 History of tampering

2.1.1 Level and Types of Tampering

Tampering is defined in the Federal Anti Tampering Act (FATA) of 1983 as
“tainting any consumer product or rendering materially false or misleading the
labeling of, or container for, a consumer product with intent to cause serious

injury to the business of any person” [2].

There are three points at which a foreign object or other contaminants can get
into a product; during manufacturing, while the product is in distribution, and after
purchase. Tampering during manufacturing is generally the most harmful
scenario, in that large amounts of product can be accessed at this point, possibly
by angry employees. By contrast, tampering that occurs within the distribution
chain, is more likely to be committed by extortionist threatening the government
or a company or by individuals wanting to harm a single victim. Tampering after
purchase, on the other hand, is frequently related to “false report cases”, such as

the Pepsi tamperings in 1993 (see 2.1.3) [3].

“Modifying Criminogenic Products“ [4], indicates four main types of intrusion

to guard against:
. terrorism or random attacks (such as the 1982 Tylenol case),

. pilfering or damaging of items enclosed in the packaging,

. tampering during manufacturing and
. counterfeiting

This project, limits its focus to random attacks on the consumer unit-level in

post-production environments.

2.1.2 Reasons for product tampering

Park Elliott Dietz, a forensic psychiatrist and expert on criminal behavior, did
an analysis of tampering incidents in the US. which led him to the conclusion
that most tampering offences spring from “greed, anger, and hatred among
immature and antisocial people”. Dietz also indicates “real product tampering is

usually done by political terrorists or by people who are mentally ill” [5].

There are three different types of random acts of tampering that companies
must concern themselves with; initial acts, copycat incidents and threats of

tampering. All three types present problems that companies must manage [6].

The initial act is usually a true tampering, where an offender places a
contaminated product back into the distribution system. In faked tamperings,
someone contaminates a product in their household and makes it look as though
they are the victim of random tampering (false reports). Threats of tampering are
mostly coupled with product extortion, threatening to tamper with a product if the
demands (usually money) are not met [7]. Extortion tampering can be political or
social, where terrorist call the news to make threats to reduce the confidence in

either the government or the company. Another category is “politically motivated

and malicious tampering”, where terrorism groups were behind the threats [8]

(see 2.4 - Bioterrorism Act).

2.1.3 How product tampering occurred in the past

“Reports of tainted goods causing harm and resulting in litigation date back to
the late 19th century” [9]. Tampering of packaged goods, however, became
prominent after the 1982 Tylenol incident in Chicago, Illinois. Prior to this, “the
phenomenon of product tampering was virtually unknown and no homicides as a
result of product tampering has happened” [7]. Seven people died after ingesting
Extra Strength Tylenol capsules that had been laced with potassium cyanide.
The perpetrator placed the capsules in the original package and back on the
shelves [10]. This case is still unsolved, but interest was reignited in February
2009 when the Federal Bureau of Investigations (FBI) searched the condominium
of a leading suspect in Cambridge, Massachusetts. Still seen as the prime

suspect, James W. Lewis denies committing any crime [11].

Since the Tylenol poisonings, there have been periodic product tamperings
(see Appendix A, Table 8, p. 80). The problem, however, peaked in the United
States in 1986, after a second tampering of Tylenol and other consumer products

occurred, reigniting media attention (see Figure 1).

 

 

 

1 989
“i 2004
1982 [I’m]: Dairy Crest
MEMO! “ml 1 993 8
Chilean grapes I @ pepsi GGS'riiEr
1 986 1 991 1 997

 

Gage, Gerber
Contac, Dietac, Teldrin

WM

(35;. Lipton:
S ' Cup-a-Soup

Figure 1. Various Tampering incidents after 1982

 

In February 1986, a 23 year old woman died after ingesting Tylenol that had
again been laced with cyanide [12]. Tampering complaints reached an all-time
high of “1,900 during the six weeks following the death”, the largest increase
recorded since FDA started tracking tampering reports in the late 19708 [13; 14].
During this same period of history, Gerber received over 200 complaints of glass
shards in their baby food jars. This marked the' third time Gerber had been
alarmed by such reports. In 1984, they recalled baby food twice after receiving
unconfirmed reports of glass in the packages. During the 1986 scare, the US
Food and Drug Administration (FDA) inspected 36,000 unopened jars and failed
to discover anything that would justify a recall [15]. Gerber was sued for
$800,000 “for the pain and suffering caused to an Ohio mother whose child might

or might not have swallowed glass” [16].

One month later, in March, the SmithKline products: Contac, Dietac and
Teldrin were immediately recalled after traces of the rat poison warfarin were
detected in nine capsules [16]. In the same year, two people died after taking
adulterated Excedrin, manufactured by Bristol-Myers. In the Excedrin case, the
wife of the second victim, Stella Nickell, placed cyanide in the extra strength
Excedrin capsules, repackaged them, placed three of the bottles in area stores
and kept two bottles to use as a means to kill her husband. She was found guilty
of murder and sentenced to 90 years. She was also the first person to be tried
and convicted for committing murder by means of product tampering [10]. In
September 1986, another person died after eating Lipton Cup-A-Soup that was
also tainted with cyanide [17]. The tampering complaints during 1986 most often
cited Tylenol, Gerber baby food and SmithKline products. However, over “800
complaints involved a multitude of other products, from soft drinks to peanut
butter” [13]. It was suggested that nationally publicized tamperings correlated
with complaint rates (see 2.4 Reactions of the Public: Post-tampering) [4; 7; 13].
For instance, prior to the 1982 Tylenol tampering, “all recorded tampering
complaints totaled 37” and in the three years after, the number of complaints was

around 800 [13].

But the problem of tampering is not limited to the US. In the United Kingdom,
for example, there were two cases of food tampering in 1989 alone. Dairy Crest,
a large butter producer, had to take all its butter off the shelf after traces of '

mercury were discovered in one pack [18]. During the same year, a single jar of

Heinz baby food was identified to contain caustic soda and two large metal

drawing pins and others slivers of glass [4; 19].

“Political and social causes” have acted as a motive for product tampering
around the world [8]. One such threat occurred in the form of imported Chilean
grapes that were intended for wine and table use in the US in 1989. Terrorists
phoned the US Embassy in Santiago de Chile and claimed that all Chilean
grapes had been poisoned with cyanide. The threat was related to a protest over
unspecified policies of the Pinochet Government. After it appeared that
American consumers were at risk, food stores pulled tons of grapes and other
fresh fruit from their shelves. Investigators found two grapes that had a white
ring of crystalline dust around a puncture hole, where cyanide was “injected”.
However, each grape contained 0.003 mg cyanide, whereas the lethal dose is

considered to be 200 mg for an adult [20].

By the late 803, many documented tamperings employed the use of
chemicals, like Cyanide. Recognizing this, the FDA changed its Elemental
Analysis Research Center to focus on forensic research with the primary focus
on “what happens when poisons are added to food and drugs” [3]. This center
became the FDA Forensic Center, which was intended to “provide the agency
with a team of forensic science experts who can respond immediately to all
tampering incidents and provide expert advice and scientific evidence to FDA

officials” [21].

Despite the new tools being embraced by the Agency, the tampering of

products with cyanide continued, and in 1991 and 1992 the OTC drugs Sudafed

and Goody’s headache powder were targeted. At least six packages of the cold
remedy "Sudafed 12-Hour" were identified as adulterated. The capsules were
laced with cyanide, resulting in two deaths and a national recall [22]. This
incident happened after 1989, the year in which changes were made to the
tamper evidence packaging regulations of hard gelatin capsules (see 2.3.2
Drugs). In the case of Goody’s Headache Powder, an over-the-counter
Headache remedy, the product was recalled after the death of a 51 year old man
who ingested the tampered product. This poisoning occurred despite the

presence of a tamper evident feature (a plastic seal) [23].

Food products were targeted again in 1993 when Pepsi had to cope with
more than 60 reports of objects (like syringes and even a mouse) in their Diet
Pepsi cans. It started with a report from Tex (82) and Mary (78) Triplett, who
claimed to find a syringe in their Pepsi can. After four days of publicity the
complaints had increased to a total of 9. Even so, Pepsi’s crisis counselors
“decided to fight the media crisis with media”. They set up a crisis command
center in the company's TV room to make the statement that cans were 99.9%
safe and explained that in 50 ways. In the end it turned out, the syringe found by

the Tripletts likely belonged to a diabetic relative [21; 24].

In 2004, Gerber’s Banana Yogurt Dessert was tainted with non-lethal
amounts of ricin. Two jars contained the poison and warnings inside that the jars
were contaminated. As a result, Gerber added an additional tamper-evident

feature, a plastic seal, to the jars [25].

10

As mentioned earlier, the UK baby food market also suffered after a
tampering incident in 1989. In 1994, bottles of UK Safeway’s tonic water were
found to be laced with atropine, a form aptly named deadly nightshade. Eight
people became ill after consuming the product, with four of them being quite
seriously ill. Consumers didn’t see that the seal on the bottle was broken. The
genesis of the situation was a man who attempted to murder his wife by
administering the poison in the tonic water. The overall cost for the recall added

to at least £44,000 [4].

Nestle Germany has been the target of several extortionists. In early 1996,
police found tubes of Thomy mustard and mayonnaise that had been laced with
cyanide. In 1998 and 1999 jars of "Alete" brand baby food were laced with
pesticide and placed on several grocery stores. Fortunately, no injuries resulted
from any of the events. A German-Romanian who confessed to the
contamination of the Thomy products, which were tampered between August

1996 and September 1998, was sentenced to 11 years in prison [14; 26].

Like other countries throughout the world, Australia’s food and pharmaceutical
companies have faced big recalls of their products as the result of extortion
attempts. In 1997, Arnott recalled its biscuits after extortionists threatened to

poison the product with a pesticide [27].

Another Australian recall in 2000 involved Herron pharmaceutical group.
Herron capsules were removed from shelves after 2 people were poisoned in an
extortion attempt. The capsules, packaged in tamper-evident packaging, had

been laced with strychnine, a highly toxic poison used to kill rats [28]. Five years

11

later, Masterfoods recalled all Snickers and Mars Bars in New South Wales,
when they received a contaminated Snickers bar, laced with a substance similar

to pest poison, along with extortion threats [29].

2.2 Research on Tampering

There was never again a wider influence on tamper evidence packaging than
after the Tylenol tamperings in 1982. One year later, Sneden published “Testing
of tamper-resistant packaging: Consumer attitudes and perceptions on tamper
evident packaging” [30]. For this work, the researcher employed an attitude
survey designed to determine the effect of the Tylenol incident on consumer
awareness of TE packaging. Survey findings and a performance evaluation of
eleven tamper evident packages suggested that consumers were aware of

tamper evident packaging, but unsure or unable to detect tampered packages.

A subsequent study examined a specific tamper evident closure that was
introduced in 1983 by TBL development (Livingston, NJ) [31; 32]. The closure
changed color when twisted, indicating that the container had been opened. The
plastic cap, which was applicable for container closure systems that requires a
rotational motion to remove the closure, such as jars and bottles, featured a
“transparent, modified standard cap and a laminated liner”. Small tines in the cap
rip the upper foil away and expose the red paper liner under the foil, which is
intended to protect and warn the consumer of a prior opening [32]. When this
container/closure system was tested, a majority of survey respondents (57.4%)

indicated that they were unsure that they would be able to detect tampering.

12

Because many tamper-evident features are only effective if the consumer is
aware of the feature, attends to it and understands its meaning this result was

concerning [31].

Another survey, conducted in 1983 for DuPont Co. (Wilmington, DE),
questioned 500 consumers about their preference between two types of tamper
evident packaging [33]. The study compared two packaging systems in two

different configurations,

. cartons covered with an overwrap were compared to the same cartons sealed

with glue and

. bottles with an overwrap were compared to bottles sealed with shrink bands

around the neck.

The survey revealed that 95% of the people preferred the ovenivrap to the
glued carton. The perception was that an ovenrvrap provided more indication of
tampering attempts and was a higher barrier to tampering, with the rationale
being that special machinery needed to be used in order to reseal. When
solutions focused on applying TE technology to the bottle itself, 69.5% chose the
ovenNrap compared to 27.5% who indicated a preference of shrink neck bands

(the remaining 3% indicated no preference).

The magazine ‘Package Engineering’ published the results of two focus group
sessions focused on TE technologies during this same timeframe. One of the two
focus groups consisted of 10 women and the other 9 men. The study differed

from DuPont’s study, which provided a quantitative sample of consumer opinion,

13

because it qualitatively observed the perception of consumers regarding tamper

evident packaging [33].

The female panelists in this research felt that carded packages (see also
Figure 2, p. 26) were the most secure and most expensive package, and also the
one they would be willing to pay more for. The group also felt somewhat safe
with inner seals, provided the seal left behind a residue. Feelings about shrink
bands were mixed, with one person of the opinion that the band shouldn’t be
transparent to increase the noticeability. Plastic or paper pouches made the
consumers feel “fairly comfortable”, but blister packs were generally approved by

the female panelists.

The focus group consisting of men shared the women’s disapproval of that
tape seals. The observation of one man was that a package must be completely
destroyed in order to open it and that the “self-destruction” secures it. Clear
overwrap was discussed as needing some printing to prevent duplication and
was accepted, with care. Glass bottles were mentioned as among the best
tamper evident packages, because “you can’t stick a needle through glass” [33].
As with the women’s panel, men expressed favorable opinions of blister
packaging. However, offering pharmaceuticals in a blister instead of a bottle, one

man mentioned that consumers might rather stay with a package they know.

Hotchkiss and Gravani surveyed commercially available packaging for the
purposes of assessing the security of food products in commerce in 1984 [34].
They surveyed supermarket shelves, in order to categorize different food

products and determined the level of tamper evidence for various food groups.

14

The purpose of the field survey was to provide information to the food packaging
and manufacturing industry “on the status and vulnerability of food packaging to
tampering" [34]. One goal was to quantify the different types of packaging used
at the time to assign a “relative rating on the degree of protection against
tampering afforded by each product” [34]. Another goal was to create a ranking
of susceptibility to tampering. The survey was conducted by collecting data in 5
“super” retail grocery stores owned and operated by different companies. Their
conclusion was that less than 1% of the food surveyed in these markets had
tamper evident packaging which would meet the FDA regulation for OTC drug
products. To compare the safety features of baby food packages with findings of
this study, the present study repeated the work for this category in a single store

(see Appendix B, p. 85).

The effectiveness of tamper evident packaging was evaluated by researchers
at Michigan State University in 1989. The MSU researchers evaluated the
overall effectiveness of current, selected tamper resistant packages by
measuring the consumer’s ability to detect tampering and by surveying
packaging professionals involved with tamper evident packaging. The consumer
study included about 200 people from various age groups and educational
backgrounds that were given four tampered and four untampered packages for
observation. The study revealed that the participants did “a poor job overall in
correctly identifying packages which had been tampered” [35] or in differentiating
between tampered and untampered packaging. The opinion of packaging

professionals surveyed during the course of the study regarding tamper evident

15

packaging was split. Some indicated the current TE features are good indicators
of possible tampering, while others felt that many TE features can be repaired
without detection and, therefore, recommended better consumer education.
Ultimately, the research team recommended that new designs be evaluated
using two approaches: consumer testing (80%) and the professional evaluation
of the TE design feature (20%) “In order to cover all possible effectiveness
criterion in an appropriate proportion of importance” [35]. Therefore, they
combined both industry concerns and the consumer’s ability to detect tampering

in their evaluation procedure.

By contrast, Rosette (1985), recommended that design effectiveness for
tamper evidence be evaluated based on industry concerns. He started the
evaluation of tamper evident features in his Master Thesis “Development of an
Index for Rating the Effectiveness of Tamper-evident Packaging” in 1985 [36]
and continued studying on this topic in his dissertation “Improving the
Effectiveness of tamper-evident packaging” completed in 1989 [37]. The system
that he developed is now published as the “Rosette Protocol” (see 2.5 Evaluation
Methods) and is a ranking protocol that intends to provide the users and
producers of TE packaging an objective technique for evaluating protective

measures.

Closely tied to the previously mentioned studies conducted at Michigan State
University, another thesis “An Evaluation of Tamper-resistant Packaging: A
method for measuring tamper-evidence” was published by lwaszkiewicz in 1991

[38]. For this study, researchers presented three packages that incorporated five

16

different tamper evident technologies (film ovenNrap and blister, foil membrane
seal and plastic shrink band, vacuum button) in a tampered and untampered
condition to 96 consumers. Overall, they found out that the tamper-evident
packages failed to protect the consumer, because consumers didn’t know how to
use the tamper-evident packaging. The emphasis of the experiment, however,
was placed on the reaction of the subjects during their decision making process
of “tampered” or “untampered” package and the developed methodology
provides “the degree to which tampering is apparent to the observer” [38]. The
researchers studied the consumer behavior and showed the consumers’ inability
to detect tampered packages. As result, they suggested that the store be used
as a “control point” in the prevention of tampering. Specifically, they suggested
the use of a magnetized strip or ink that is demagnetized after purchase be used
to alert the retailer or consumer that “the package had been previously

purchased” [38].

In 1996, two years before packages for OTC drugs offered in capsules (see
also 2.3.2) were required to have at least two tamper evident features, another
study examined the consumer preference for solid oral dosage forms for OTC
pharmaceuticals [39]. Three hundred and eighty-eight students, of which 16.5%
were pharmacy students, were questioned about their usage or perception of
tablets, capsules and caplets. The questionnaire concluded with questions that
focused on tamper evidency. Data suggested that respondents believed
capsules were not tamper evident or safe, but effective. However, female

pharmacy students thought that tablets were not tamper evident and capsules

17

were of high quality. Studies such as this, which suggested a lack of confidence
in capsule technology, were likely considered as the regulation 21CFR211.132

was changed.

In the time since 1998, no regulation changes have been made to the tamper
evidence requirements for OTC drugs, but further studies on tamper evident
technologies have been conducted. In 2006, a focus group study, conducted at
the Ohio State University (OSU) explored consumer understanding and
experiences with tamper evident packaging devices [40]. The study was used as
guidance for the development of a new tamper evident packaging technology that
changes color. The device is not clearly described, but the TE feature is most
likely employed in the closure as a “flagging device” (see also ONE LOOK® in
3.3.1 Current commercial solutions). OSU researchers provided 130 female
participants with a questionnaire which asked about their experiences with TE
packaging. Among other things, the research team asked if participants usually
checked for an intact TE feature before buying or using a food product, or if they
had ever eaten food from a package with a broken TE device. The participants
answered that they usually check for TE features on a package at the store, but if
these are hidden or they are in hurry or shopping with their children, they would
examine the package before usage at home. In cases of an urgent need for a
product, however, they would also buy a product with a broken TE feature if no
indication of spoilage was apparent. Participants also mentioned that they did
not examine every product in the same way, but paid more attention to liquids or

semisolids, such as milk or baby food, with the reason that spoilage can occur to

'18

the product after opening and that especially infants are at a higher risk regarding

illnesses.

The OSU research found that consumers associate 3 TE device with food
spoilage, instead of the concept of product security. Regarding the development
of the new, color changing TE device, the consumers were skeptical about its
necessity and how to interpret a triggered device. Their suggestion was a “point—
of-purchase signage” to help the consumer understand the features properly [40].
After an educational session on the device’s use and meaning, participants that

were first doubtful about a color changing TE features changed their mind.

The OSU study, as well as those mentioned previously, indicates the
importance of consumer education regarding TE device design and use (see also
2.4). OSU researchers go further, suggesting, “consumer education is often
lacking in the product development cycle” [40]. In conclusion, the researchers
mentioned that their participants could identify a triggered TE device, even

though they did not check the products they buy consistently.

Most consumers are familiar with at least some level of tamper evident
packaging, but how can they know what a tamper evident device is, if it is not
indicated on the package? “Labeling and instructions to consumers are
frequently unclear” and information is often “buried in the fine print or is
completely missing”, Roger Johnston states in his publication on tamper-
indicating seals [41]. He says the reason for this is that manufacturers don’t want

their products to be associated with product tampering.

19

Even though, OTC drug products require a label statement (see 2.3.2)
referring to the TE feature, the warning is often unheeded. The warning is, by
regulation (21CFR 211.132) required to be “prominent” so that consumers are
encouraged to read the warning at the point-of-purchase [42]. A study published
in the Proceedings of the National Academy of Science of the United States used
eye tracking to examine the prominence of warning labels, among them the
required TE warning, on OTC pain relievers of 5 different drugs containing
acetaminophen [43]. Dependent variables analyzed included standardized time
(time per typographical character) spent on varied portions of the label (brand
name, tamper evident warning, child resistant warning, claims statement and
drug facts box). The study presents evidence that two required warnings, the TE
warning and child-resistant warning are not prominent when compared with other
elements of tested labels. The TE warning was particularly problematic;
researchers indicate that “more than 80% of study participants failed to record
time in the TE warning gaze zones across all packages tested” [43]. This means
that just 20% of participants registered any time in the “warning zone” despite the

fact that regulations require it to be prominent.

2.3 Government Reaction to Tampering - US Legislation and
Regulation

In October 2007 the FDA issued an updated “Guidance for Industry Food
Producers, Processors, and Transporters: Food Security Preventive Measures
Guidance” indicating tampering of products as a major threat to consumer health

and safety (Section 4). The document recommends actions in cases of

20

tampering or other malicious, criminal, or terrorist actions to minimize their risk.
Companies should have a strategy regarding the appropriate response to product
tampering, which also implies that the evidence of tampering should be easily

visible or obvious.

Liquor, drugs and milk are required to have tamper evidence devices that
show that provide evidence to the consumer that the container has been
tampered with. The existing tamper evident regulations for these products are

discussed in the following subchapters.

2.3.1 Liquor

Legislation and regulation of tamper-evident packages is nothing new, it has
been used ever since the repeal of Prohibition in the United States [30; 44].
During the early years of Prohibition Repeal (after 1933), when the sale of
alcohol was again legalized, whiskey distillers developed tamper-resistant
packages in the form of “foil-covered, metal-end fibre cans” to reassure the
consumers that their product was “the real thing” [45]. The primary purpose of
the distilled spirits regulation (27 CPR § 19) at its inception, however, dealt with
revenue tax stamps. The current regulation also mandates distilled spirits to be
closed with a tamper-evident device (§ 19.662). Liquor, in containers having a
capacity of one gallon (3.785 liters) or less, are required to have an “Affixed
closure”. The closure (or other device) shall be securely affixed to the
containers, to leave a portion remaining on the container when the container is
opened and the closures shall be constructed in such a manner as to require that

they be broken to gain access to the contents of the containers [46].

21

2.3.2 Drugs

The first Tylenol incident led to the massive recall of 31 million containers of
Tylenol. Following the recall, Johnson & Johnson (J&J) conducted a $150 million
“advertising blitz to restore consumer confidence”. This action is still considered

to be a model public relations response to a product harm crisis.

Widespread public concern motivated the creation of tamper evident
regulations for OTC products. In 1982, the FDA implemented 21CFR211.132,
“Tamper-evident packaging requirements for over-the-counter (OTC) human drug
products”. The rule required, among other things, tamper-resistant packaging for
selected cosmetic products such as oral hygiene products and many over-the-
counter (OTC) drugs. According to this regulation, “each manufacturer and
packer who packages an OTC drug product for retail sale shall package the
product in a tamper-evident package, if this product is accessible to the public
while held for sale.” A tamper-evident package is defined as “one having one or
more indicators or barriers to entry which, if breached or missing, can reasonably
be expected to provide visible evidence to consumers that tampering has

occurred.”

In the year following the poisonings (1983), Congress passed the Federal
Anti-Tampering Act (FATA), which made it a crime to tamper with consumer
products. Certain other behaviors, such as falsely reporting that a consumer
product has been tainted, also became chargeable under the provisions of the
Act. These felonies are punishable by three years to life imprisonment and a fine

of up to $250,000.

22

In the wake of several other publicized incidents of tampering (e.g. Excedrin —

see Figure 1), the regulations were revised. Capsules, consisting of two or more

pieces, were now required to have at least two tamper evident features, or to

have at least one tamper evident packaging feature if a tamper evident capsule

seal was employed. The main regulation, 21CFR211.132, includes the

requirement of tamper evident packages for most OTC pharmaceuticals and was

amended in 1989 and 1998. In 1998, 21CFR211 changed the packaging

requirements for OTC drugs in an attempt to further decrease the risks posed

from product tampering. Two highlights of the final rule are:

Sealing of all two-piece, hard gelatin capsules that were regulated as OTCs
(1989) are mandated. Since capsules were the choice of many tamperers, the
agency initiated this rulemaking to reduce the potential for tampering with
vulnerable two-piece, hard gelatin capsules. However, Johnson & Johnson
phased out capsules entirely after the second Tylenol tampering and replaced
them with either capsule-shaped tablets (caplets) or gelatin-covered tablets
(gelcaps) [22]. Another requirement for the packages of two-piece, hard
gelatine capsules was the use of at least two tamper-resistant packaging
features or at least one tamper-resistant packaging feature if a tamper-evident

capsule seal was employed.

Another change to the regulation was the terminology used throughout the
agency’s regulatory program. The term “tamper-resistant” was changed to
“tamper-evident” to more accurately characterize the role of tamper-evident

packaging in protecting consumers. Tamper resistance in this case, can be

23

defined as “the degree to which it is difficult to tamper and repair a packaging
without leaving evidence”. On the other hand, Tamper evidence can be

defined as “the degree to which tampering is apparent to the observer” [38].

The amendments to 21CFR211.132 also required labels to reference all TE
features employed by the packaging, including those present in the secondary
packaging. Tamper-evident warnings must be “prominently placed” on the
package, so they are unaffected if the tamper-evident feature of the package is
breached or missing. Dermatological products, denitrifies, insulin and lozenges
are exempted from the TE requirements and, therefore, the labeling as well. For
products that did require TE features, the features were dictated to be “distinctive
by design.” This means that they “cannot be duplicated with commonly available
materials”, or distinctive by the use of at least one indicator like eg. a registered

trademark, logo or pattern (21CFR211.132 (b)).

The FDA Compliance Policy Guide (CPG) 7132a.17, part of the current good
manufacturing practice (cGMP), “Tamper-Resistant Packaging Requirements for
Certain Over-the-Counter (OTC) Human Drug Products” indicates that these

designs meet the requirements of 21CFR21 1 .132 (see Figure 2):

1. Film Wrappers: Transparent film wrapped around the entire product that
must be cut or torn to open the container and that employs an identifying

characteristic.

2. Blister/ Strip packs: Dosage units individually sealed in clear plastic with foil

or paper baking that must be broken or torn to obtain the product.

24

. Bubble Pack: Product and container are sealed in plastic and mounted in

or on a display card to be torn or broken to remove the product.

. Heat shrink bands/ wrappers: Band or wrapper, heat shrunk applied to a
portion of the container must be cut or torn to open the container and
employing an identifying characteristic. The use of a perforated tear strip

can enhance tamper-evidence.

. Foil, Paper or Plastic Pouches: Sealed individual pouch that must be torn

or broken to obtain the product.

. Container inner mouth seal: Film, foil, or a combination thereof, sealed to

the mouth of a container (e.g., bottle) under the cap that must be torn or
broken. Polystyrene (PS) foam container mouth seal are not considered

tamper-evident.

. Tape seals: Rely on an adhesive to bond them to the package and are only
considered capable as TE feature if an identifying characteristic that cannot

be readily duplicated is employed.

. Breakable Caps: Plastic or metal cap with a band that either breaks away
completely when the cap is removed from the container or leaves part of

the cap attached to the container.

. Sealed metal tubes or plastic blind-end heat sealed tubes: Bottom of tube
is heat sealed and mouth or blind-end must be punctured to obtain the
product. Crimped ends can be used if they cannot be breached by

unfolding and refolding without visible evidence of entry.

25

10. Aerosol containers: Tamper resistant by design with the preference of
direct printing.

11. Metal and Composite Cans: Top and bottom must be joined to the can
walls such that they cannot be pulled apart and reassembled without

visible evidence of entry with the preference of direct printing.

 

 

r __ '17- “ . ”up .__- \x,

. "1: ”,3; ' " ._. W": “""‘—-~' \

\K‘x, - if» ‘ I I i i \ “ A
' I‘gi'.’ ._ 1‘. 1

Film Wrapper Blister Pack Bubble Pack Shrink bands Pouches

/’ \.

   

I
,/

-_.. r- 3:,1 /"C
-r \ "A _/ r, \
. . ‘,' ‘1; I," ‘.
u z v.»
\ .—-" {1. . 'flz'x-
\Ixf

Mouth Inner Tape Seal Breakable Caps Sealed metal Aerosols
Seal tubes and Cans

   

 

 

 

Figure 2. Packaging examples complying with the FDA TE requirements [47-50]
2.3.3 Milk

The requirement for tamper evidence has not been limited to drugs and
alcohol. In January of 2006, the Center for Food Safety and Applied Nutrition
(CFSAN) at FDA and the US Public Health Services (USPHS) began enforcing a
requirement for tamper-evident packaging for plastic containers of fluid milk.

Item 19p of the Grade “A" Pasteurized Milk Ordinance (Grade “A" PMO) states,

“The cap or closure shall be designed and applied in such a manner that the
pouring lip is protected to at least its largest diameter and, with regard to fluid

product containers, removal cannot be made without detection."

26

A major driver for the TE requirement was the fact milk has the potential to
serve as a vehicle of disease transmission and has, in the past, been associated
with disease outbreaks of major proportions. Item 19p of the PMO “help(s) to
assure the consumer that the milk and milk products have not been contaminated

after packaging” [51].

2.4 Reactions of the Public: Post-tampering

As mentioned (2.1.3), a publicized tampering event is frequently followed by a
surge in complaints for all types of products and copy-cat events. Complaints
range from “strange-tasting items to swollen cans, discoloration or insect parts in

packages” [13].

According to a consumer survey [30] conducted at Michigan State University
shortly after the 1982 Tylenol Poisonings, 75% of respondents (n=270) believed
that the new Anti-tampering regulation (21CFR 211.132) should also apply to
food and beverage products (see also Figure 2 and 2.2 Research on Tampering).
The FDA, however, did not go further in introducing TE features for food
packages because of the much greater diversity of the food products as
compared to those in the pharmaceutical sector [4]. Despite the fact that the
regulation didn’t apply to food packages, the food industry actively reviewed its
packaging and recall procedures during this period of history. Many food
companies voluntarily introduced TE packages in order to make their products
less vulnerable to tampering [52], to protect their consumers and to shield

themselves from liabilities associated with “foreseeable risks”.

27

When the TE system requires information processing on the part of the
consumer, companies have a duty to educate the consumer and to routinely
warn them to inspect packages and tamper evident indicators before using the
contents [9]. The theory is that “if consumers inspect their packages before use,
incidents of injury from product tampering will decline” [8] when consumers notice
that distinctive, tamper evident features are missing. When consumers are
asked about TE systems, some suggest that they prefer TE packages and are
“willing to pay slightly more” for “products that are resistant to tampering and

have shelf-visible features” [30; 31].

Protecting consumers from potential tampering is not just the right thing to do
in an ethical sense; it is also good business practice. Packaging must provide
protection against foreseeable events under product liability laws [37]. Although,
a federal product liability law doesn’t exist, but the United States Department of
Commerce has promulgated the Model Uniform Products Liability Act (MUPLA)
for voluntary use by the states (for foreseeable risk see also [9]). Manufacturers
and their insurers, however, favor one federal law that replaces the 50 state

products liability laws [53].

In addition, a tampering event is often followed by great public scrutiny that
can distress companies, who have to expend tremendous resources on a variety
of things. Publicity causes a resurgence of consumer anxiety about product
tampering and publicity surrounding legitimate cases of tampering is documented
to encourage copycat incidents [7]. To prohibit unnecessary publicity,

responsible reporting is crucial and begins with obtaining accurate Information

28

[8]. In cases of extortion, the extortionists seek media attention to increase the
pressure on the business. But the journalist’s duty is to inform the public with

precise information and should not be blinded by the willingness to create a story.

Heightened concerns for threats have resulted since the terrorist events of
9/11. To help companies facing a tampering threat, the- Food Marketing Institute
released an advisory guide [54], with step-by-step instructions regarding how to
examine and deal with threats or possible hoaxes. In the time since, the
Bioterrorism Act of 2002 was enacted. The intention of the Act was to improve
the ability of the US to prevent, prepare for, and respond to bioterrorism and
other public health emergencies. The Bioterrorism Act developed “a crisis
communications and education strategy with respect to bioterrorist threats to the

food supply” [55].

2.5 Evaluation Methods

An index developed by Jack Rosette and published as the “Rosette Protocol”
is a tool for evaluating the effectiveness of TE features of a package [37]. He
rates the factors that are involved in package tampering and detectability on a
basis of their effectiveness. The test is constructed to serve both package
suppliers and industry users. For the test at least three participants should be

used for reliable results. The TE features of a package are then rated by:
c The knowledge level of the participant (violator),
o The TE feature material (how feature was reused, reproduced, or replaced),

. The quickest required time for a successful violation,

29

. The equipment used to violate the package and
o The visibility of the TE feature (determined by a panel of 3-5 people).

For the rating, one of the participants has to violate the package successfully.
This is achieved if an inspector doesn’t detect evidence of tampering. The
inspector (e.g. manager, who leads the test) is given three packages, one of
them violated. Out of these three, the violated package has to be identified within
five minutes [37]. The test was carried out by the present researchers to rate the

effectiveness of an aseptic puree packaged in plastic (see Appendix C, p. 89).

If a security feature isn’t already present, the ASTM Standard F 1448 can
provide guidance in determining a suitable security technology. The ASTM
standard F 1448 is a “Guide for the Selection of Security Technology for the
Protection against Counterfeiting, Alteration, Diversion, Duplication, Simulation,
and Substitution (CADDSS) of Products or Documents” [56]. The guidance
provides a procedure to accomplish the proper selection of a security system,

and encompasses the six following steps:
0 Define the CADDSS problem
. Determine requirements using the CADDSS matrix

. Compare the user’s matrix and the technology matrix in order to make the

appropriate technology selections

. Test for effectiveness (using a test such as the Rosette protocol, referenced

above)

0 Implement the technology

30

c Institute educational programs so that the technology is used effectively

This procedure can be repeated with every available technology and can be
helpful in evaluating features of technologies available for use with the
application. The guidance also assists in determining the requirements by
completing the user's specific CADDSS versus parameters matrix included in the

standard [56].

In this research a similar approach was done. A matrix was created to
compare various commercial and incremental tamper evident solutions and their
effectiveness against different tampering methods (see Appendix G Matrix

Development, p. 105).

31

3 Literature Review: Identification and Analysis of
Promising New and Established Technologies
Relevant to TE

The market of security technologies specific to consumer products is
flourishing, and many technologies have been developed as a result. The food
industry, with its great diversity of products, is not regulated in terms of tamper-
evident packaging and therefore needs more attention, because their products
may be involved in “criminal activities such as tampering, relabeling,
unauthorized diversion, and counterfeiting” [57]. This chapter includes a short
introduction to technologies other than tamper evident technologies; some
technologies discussed are commercially available at the time of writing, others

require refinement and research prior to any implementation.

Securing the supply chain against threats is a comprehensive issue. Tools for
enhanced security include: authenticators, anti theft tags, and track and trace
technologies, as well as tamper evident packaging. Such tools make a package
more intelligent. Intelligent packaging can sense and inform [58] and can be
described as packaging containing external or internal indicators to provide
information about the product and/or package [59]. Intelligent packaging has the
potential to work on issues like product quality and safety, package integrity and
tamper evidence. They can also serve as authenticators, anti-theft and track-

and-trace technologies [59].

32

3.1 Authentication
Authenticators are tools or techniques that are used to verify that the
packaged product is genuine, and sometimes provide the added benefit of being

difficult to reproduce because of their complexity.

Authentication technologies are generally divided into three categories: overt,
covert and forensic. Overt authentication features do not require special readers
or detectors; they are perceptible in nature (generally visible) so that the general
public can identify them. Examples include: embossed holograms or color
changing inks as well as retroreflectives. Since Glaxo first used a tamper-evident
hologram to seal packages of Zantac in 1989, holograms have been utilized
frequently by the pharmaceutical and medical industry. They are used in the form

of labels, seals, hot-stamped patches and blister-pack foils [60].

Covert authentication encompasses technologies “that are not readily
apparent, but instead require a simple reader or verifier to detect” [61]. To notice
UV ink, for instance, a UV lamp is required, while the detection of microprinting
mandates a source of magnification. The most secure authentication features,
however, are at the forensic level. They are extremely covert and provide
information on a “need to know” basis. Forensic technologies include “unique
Taggants or other imperceptible changes that require sophisticated equipment to

read and recognize” [61].

33

3.2 Antitheft and Track-and-Trace

To prevent theft, retail shops currently employ electronic article surveillance
(EAS) systems. EAS tags are particularly attractive for high-cost goods, such as
cosmetics and electrical and entertainment products, and are not as common for
food products [57]. They are attached to the product they are intended to protect
and can set off an alarm, when the active device is passed through an EAS
detection system. Current developments include anti-theft tags that can be
directly integrated in the packaging material during manufacturing [62] or that are
“paper-thin” and the size of a post stamp [59]. Built-in anti-theft tags also have

the advantage that a possible thief can’t visualize and remove them.

A study conducted by The Freedonia Group Inc., a Cleveland-based industrial
market-research firm, suggests that the US smart-label industry, including EAS,
RFID and interactive packaging labels, however, will no longer be dominated by

EAS labels but rather by RF ID and smart labels1 (“Smart Labels”, 2006). [64].

Radio frequency identification (RFID) surged in popularity after Walmart
mandated RFID tags on the pallet level for its suppliers. RFID is a wireless
automatic identification and data capture technology. RFID tags can be
programmed with unique information for identification and tracking. An RF ID tag
can also include data about a product such as its color, size and model number

and act as Electronic Product Code (EPC). WaI-Mart, which is already using

 

1 According to the ‘Adhesives & Sealants lndustry’, smart labels include RFID, EAS and
interactive packaging [63]

34

EPCs, states that an EPC is nothing more than an “electronic version of a bar

code, but with greater capacity” [65]

With dropping cost for radio frequency identification tags, they can be the key
to electronic item level track-and-trace packaging. “Traceability is the ability to
track a product’s flow or attributes throughout the production process and supply
chain” [66]. Especially in cases of recalls or as anti-counterfeiting technology,

item level tagging can be a solution.

3.3 Tamper evidence

As mentioned previously, tamper evident features are defined as an indicator
or barrier to entry which, if breached or missing, can reasonably be expected to
provide visible evidence to consumers that tampering has occurred
(21CFR211.132). Tamper evident features should be “distinctive by design” so
that duplication is challenging. In other words, packaging that cannot be
duplicated with commonly available materials or through commonly available
processes. Examples include features with identifying characteristics (e.g., a
pattern, name, registered trademark, logo or picture). Tamper evident packaging
may involve an immediate container and closure system or secondary container
or any combination thereof and provides a visual indication of package integrity.
Features are designed to and shall remain intact when handled in a reasonable

manner during manufacture, distribution and retail display.

35

3.3.1 Current commercial solutions

Some technologies recommended by FDA that meet the requirements of
21CFR211.132 (TE for OTC products) (see 2.3.2) are also very common in the
food industry. Shrink bands around the necks of tubs and breakable caps on

bottles (e.g. milk or colas) are examples (see Figure 2, p. 26).

3.3.1.1 Closures
A two-piece, flip-top neck closure with two independent TE features,
introduced by Bericap in 2007 for the vinegar market, is another example (see

Figure 3).

 

 

Figure 3. Bericap Galileo ll, 2-pieces flip top pourer closure [67]

To open the cap, the consumer must first break the over-cap seal before
being able to remove the pull-up ring to break the second seal [68]. Another TE
closure invention launched by Heinlein, Germany is the ONE LOOK®. After the
first opening, four open windows on top of the closure change their color and
snap into position. As a result, they cannot be moved back when the cap is

retightened [69].

Fresh food products have been packed in tamper evident packages, like the

"Safe-T-Fresh” design. The tear strip lock keeps the container leak resistant and

36

eliminates the need for shrink banding. After removing the strip, the container
stays intact and is reclosable [70]. A similar technique for a tamper evident
closure equips a molded plastic container with a manually removable tear band
(see Figure 4). The band is constructed such that after tearing off the band, one

side of the cap still remains attached to the container (see Figure 4) [71].

 

 

Figure 4. Tamper-evident container with tear band [71]
3.3.1.2 Labels, films and polymers

A tamper evident package patented by Rexham Corporation is the invention
of a TE folding carton [72] with a “flagging device” incorporated in the carton (see
Figure 5). The carton shows a message or a change in window color at the
carton end if the flaps have been opened. A blank panel Changes upon opening

e.g. opened by the use of tamper indicating labeling materials.

 

 

 

Figure 5. TE folding carton (22: front flap, 30: window, 40: tamper indicating seal)

37

A tamper evident label offered by 3M presents a “film that fractures when
attempting label removal from many surfaces” [73]. The indication can appear as

text or shape or just as destructible facestock.

As mentioned previously, printed overwrap film around a carton is used as
tamper evident element (see Figure 2), but wrapping the entire product in printed
film can increase the package price substantially. One solution to this is the use
of preprinted tape, eliminating the need for preprinted ovenrvrap film. The tape
can also bear standard or custom holograms [74]. An easy tearing strip of film
over the opening of a package can also serves as anti-counterfeit and tamper-
evident attribute. The film retains its tensile strength and, therefore, is suitable

for automatic processing but fragments if removal is attempted [75].

Resealable, flexible packaging has grown popular in recent years and these
technologies can also provide tamper evidence. One example is the “Zip Pak”
(Zip Pak: division of ITVV) used by the manufacturers of fresh meat, dry and
frozen food as well as pet food. A similar invention, with the same closing
mechanism, is the “tamper-evident, reclosable flexible package” [76]. The
package consists of an “inner, hermetic peel seal and a reclosure seal comprised
of interlocking closure strips” as well as an outer peelable seal. The TE feature

has to be visibly disrupted to gain access to the reclosable seal.

A new material manufactured by Evonik (Essen, Germany) utilizes a polymer
compound called XT 375, which can be used in extruded, therrnoformed and
blow molded packages. This acrylic-based, multipolymer compound is

advertised as having the ability to exhibit evidence of tampering such as tearing,

38

cutting, or needle pricking and seal integrity [77]. A test with the mentioned
material showed a shift in color (transparent blue to opaque) around the hole that
was drilled through the plastic plaque. A therrnoformed tray, for example, can
change from a transparent blue tint to a highly visible, opaque white in the
location of a breach. Separating the lid stock from the tray also changes the
optical properties of the therrnoforrned compound and makes seal separation

evident by an apparent color change [77; 78].

3.3.1.3 RFID

Pliant Corporation offers a pallet-level solution for tamper evidence. Pliant
has constructed a tamper evident, stretch film that utilizes silver ink
(nanoparticles). This creates an electrically conductive trace that is embedded in
the stretch film. The conductive trace is printed in two parallel lines, 1 to 1.5
inches wide and 6 inches apart. After the load is fully unitized, the end points of
the conductive trace are attached with a conductive adhesive, creating a circuit to
which a battery-powered circuit board is electrically connected that contains a
passive EPC Gen 2 tag. The GEN2 tag is only readable when the circuit is
intact. As such, an unreadable tag becomes an indicator of possible tampering
[79].

Like the Pliant solution, manufacturer MIKOH also uses an RF ID tag to detect
tampering. Smart8tSecureTM is a physical security technology for RFID tags that
detects tampering or rather the removal of the tag. MIKOH offers two
configurations, differing from each other in their price and construction. The

simpler feature detects tampering, when the tag is deactivated and the more

39

expensive configuration continues to operate after tampering, but alerts the next
reader that it has been tampered with. The TE feature is a "tamper-release
layer", printed next to the conductive ink antenna layer, causing the antenna to
break when the tag is removed [80]. Breakage of the TE layer is guarantee by a
special multi-Iayer adhesive design. The tag can be used to seal a single

product, eg. its opening or even a whole case.

3.3.2 Commercially available incremental systems
Some technologies are commercially available, but require modifications to be

effectively employed as tamper evident packaging solutions.

The company ArjoWiggins, a leading banknote manufacturer, developed a
synthetic tamper evident security substrate (STES), which is susceptible to
delamination. A tampering attempt will delaminate the facestock (biaxially

oriented film of HDPE), making repositioning and reuse impossible [81].

3.3.2.1 Dyes and inks

Other technologies have the potential to be used in a TE fashion, but require
further refinement or research before doing so. One such technology
incorporates a mix of conventional polymers and small amounts of tailored,
fluorescent dyes. The dyes function as “natural molecular sensors, creating light-
emitting polymer blends” that show mechanical stress like tearing, pin-holing or
deformation of the formed polymer container by changing their color of the
fluorescence. For example, the fluorescence of the package might change from

orange to green or from yellow to blue after being exposed to the mechanical

4O

stresses induced by tampering. This technology is covert, because an ultraviolet
light is required to detect the change. The amount of fluorescent dyes added to
the resin is 0.2 percent, similar to the quantity of additives such as ultraviolet
stabilizers or antioxidants. This development, led by Christoph Weder at Case
Western Reserve University, has been indicated not change the polymer’s

characteristics [82; 83].

Another indicator is the Food Sentinel System® by SIRA Technologies
(Pasadena, US). The indicator is a barcode that is printed with irreversible
thermochromic ink, turning the barcode into a time temperature integrator (TTI)
[84; 85]. The thermochromic ink barcode is coupled with a standard ink barcode
and becomes unreadable after the thermochromic ink is exposed to a
temperature change. This activates the thermochromic ink and results in a
thermally-induced color change. Changes in the ink could make this technology
a tamper-evident feature. Consider, for instance, if the ink were formulated to
detect oxygen at a certain level. By printing with this ink inside the package
behind the barcode, exposure to oxygen could render the barcode unreadable. A
label containing a gas-sensitive dye has already been designed, where different
gas concentrations lead to different colors. The label could eg. be inserted into a
carbon dioxide flushed package and would detect the decline of the carbon

dioxide content if the dye color changes from blue to a permanent yellow [84].

3.3.2.2 Gas Detectors

Ageless Eye, 3 product of the Mitsubishi Gas Chemical Corporation, is an

oxygen indicating sachet, which includes an oxygen sensitive tablet [84]. In the

41

absence of oxygen, the tablet has a pink color (5 0.1 %) and turns blue after

about 5 minutes of exposure to oxygen (2 0.5 %) [86].

Similar technologies exist to measure the carbon dioxide level inside a
package. Although these detectors have the potential to measure 002 and are
used to do so in modified atmosphere packages that have been flushed with
carbon dioxide, [84] and offer a potential technology for TE, some products could
be problematic if this approach were employed. Many perishable products
respire, resulting in the production of carbon dioxide. The release of carbon
dioxide could lead to false positives of TE indicators based on this type of

approach, as such, the usage of such indicators needs careful consideration.

A fluorescence-based oxygen sensor, called OxySentry by OxySense, Inc
(Dallas, TX) is an oxygen monitoring and control system for modified atmosphere
packaging (MAP) [87]. It consists of a control unit and a passive oxygen sensor
based on the “fluorescence quenching of a dye immobilized in a gas permeable
hydrophobic polymer” [88]. The presence of oxygen results in a change in the
emitted intensity of the dye. In this case, the control system analyzes the optical
signal of the passive oxygen sensor and converts the signal into oxygen
concentration data. The OxySense system could be used as TE technology, if
the package that needs to be secured doesn’t contain of oxygen in the
headspace. The detection of violation can then possibly be determined by
changes in the oxygen content of the packages headspace gas. An oxygen
sensor present in the headspace would trigger when air from the atmosphere

made contact with the sensor if the package was destroyed. However, this

42

technology would need a separate control unit to check the passive detector, and
at least semi-transparent package and also depends on the oxygen variation of
the packaged product (see 5 Characterizing the product: Headspace Analysis).
Nevertheless, a control unit could be directly integrated in the check-out system

at the retailer.

3.3.2.3 Biobased Sensors

A commercially available solution for the detection of pathogens, Toxin
Guard, has the potential to be modified as a tamper evident indicator. This
diagnostic system detects pathogenic bacteria in food and indicates them by
displaying a visual signal to alert the consumer. When the bacterial toxin is in
contact with the immobilized antibodies integrated into the packaging material, it
will be bound to these antibodies, which are then identified by a printed
characteristic pattern. The Toxin Guard test is printed on polyethylene-based
packaging material and can be used for detecting freshness degradation, the
presence of specific food hazards, pesticides, or indicators of genetic

modification [84; 89].

3.3.3 Potential “leap” solutions
Leap technologies are ideas for tamper evidence that will require significant

effort to make them commercially feasible.

Solutions suggested by Food Science Australia utilize a visible change of
color to alert consumers to potential product tampering that is triggered from the

photochemical oxidation that occurs in the presence of light and air reacting with

43

the red-colored flexible film used as package. The technology acts like the

bruising of an apple and is a form of intelligent packaging. [90].

Newly developed by scientists at the University of Southampton (UK) and the
Deutsches Kunststoff-lnstitut (DKI) (Darmstadt, Germany), is a color changing
film based on photonic crystals. This film is a “class of photonic crystals that
change color in various chemical conditions” [91]. It is made of arrays of spheres
that reflect the light and also contains tiny carbon nanoparticles wedged between

the spheres that scatter off the light.

Putting nanotechnology "fingerprints" into products is a new method that can
be applied to (drug) packages or the pills themselves (see Draft Guidance for
Industry on “Incorporation of Physical-Chemical Identifiers into Solid Oral Dosage
Form Drug Products for Anticounterfeiting" [92]). Nanoscale materials can be
embedded to distinguish the medicines from counterfeits or to detect hazardous
material such as bacteria. For the package, nanomarkers are mixed with inks or
coatings that change color and are applied onto labels, cartons or closure seals.
The markers can also be highlighted by shining an ultraviolet ray on the package.
For a single pill, the markers are added to the components and confirmed by

“simple field-testing kits” [93].

Nanotechnology can also be used inside fibers or garments embedding
magnetic nanoparticles to “create a unique but invisible signature”. Remarkable
fibers are potentially capable of filtering out viruses, bacteria, and hazardous
particles. Fabricating a textile that could look like a sponge and act as a sensor

could detect the presence of hazardous bacteria by wiping the surface of the

product. If a contaminant is detected, the fibers would capture it and alert the

user by changing color or becoming fluorescent [94].

Since the use of radio frequency identification (RFID) technology is increasing
[64], there are also possibilities of connecting oxygen and pressure sensors with
RFID tags. RFID-based sensors can be micro-sized and enable the storage of
data during shipping. One method for the detection of oxygen could be the
integration of an oxygen sensitive conducting polymer incorporated in a RFID
tag. Changes in the packages pressure due to tampering could be detected with
MEMS-based (Micro-Electro—Mechanical System) pressure sensors that are also

integrated in a RFID tag.

45

4 Focus group

A focus group is a small group of people discussing a specific topic and
whose response is studied to “determine the response that can be expected from
a larger population” [95]. To garner insights regarding the topic of package
tampering, a focus group was conducted in accordance with procedures
approved under IRB 09-225 (see Appendix D). The focus group was conducted
on the campus of Michigan State University and consisted of subject-matter

experts in the areas of packaging, food processing and manufacturing.

4.1 Focus Group characterization
A group of Michigan State University faculty, involved in packaging, food
industry representatives and students was composed for a single session. The
group of nine participants, aged 21 to 80, shared their experiences and attitudes
related to tampering (see Appendix F. Data Collection Sheet and Results).
Three of the nine participants were students in their senior year at School
of Packaging and one student was in the graduate program of the biosystems
engineering department at the time of the meeting (1 female, 3 male; average

age, 22.5 years; SD, 1.3 years).

The group had notably different backgrounds with regard to tampering
incidents and tamper evident packaging (3 female, 6 male; average age, 36.4

years; SD, 21.4 years, two ages missing; see Table 1, p. 48).

46

One of the packaging students mentioned his awareness of the issue of
product tampering comes from the media. Another packaging student stated that
his parents warned him to examine Halloween candy, and he also worked on a
study where tamper-evident research was Involved. Most participants mentioned
that they had heard of, or were actively engaged in issues regarding the Tylenol
tampering. One participant, a former professor at the School of Packaging, was
working on research regarding tampering in 1982, when the Chicago poisonings
occurred. He presented to the media at that time, and that his experience
suggested that even a “knowledgeable lab technician with a background on
tampering can be fooled”. His feeling was also that peoples’ overall knowledge
abouttampering is limited. Another faculty member participating in the focus
group had conducted his Master’s research in tampering and consumer
awareness in the early 19905. As such, he too, was familiar with the history of
tampering, regulations, tamper evident features and the techniques that have
been used. Yet another member of the focus group represented the company for
the studied product. She indicated experience with glass and poison in a
packaged product and addressed the importance of tamper evident features and
the need of a company to be a good steward by addressing such issues. Faculty
working with packaging stated their awareness of the Tylenol incident. One of
them taught a class during the Tylenol incident, where students did a series of
projects; one group focusing on consumer education. That mentioned, he stated

that the understanding of tampering for average consumer is limited and that

47

Halloween candy provided to children within the household are inspected for

breaks in the wrapper.

Table 1. Focus group panelist's background on tampering

 

Gender

Age

Background with regard to tampering

 

Male

80

Performed research at the School of Packaging in 1982 when
the Chicago poisonings occurred. This created the 1st
National presence of the issue and a media and public frenzy.
Happened again in 1986. Asked “Could a knowledgeable lab
technician with a background be fooled? Answered yes. Feels
people don’t know much about tampering

 

Male

22

Tylenol

 

Male

43

MS Thesis focused on tampering of packaging.

 

Male

23

Parents warned to examine Halloween candy. Performed a
studyof the OneTouch pharmaceutical strips.

 

Female

42

Has experience with tampered products such as glass and
ricin in product. Addresses the importance for a company to be
a good steward by addressing the issues. Is familiar with a
2004 tampering incident and a recent unsubstantiated intemet
threat from a blogger

 

Female

24

Aware of the Tylenol poisonings and of tampering in
pharmaceuticals.

 

Male

21

His awareness of the issue comes from the media.

 

Female

Aware of the Tylenol Incident

 

Male

 

 

 

Mentions that Halloween candy provided to children within
household are inspected for breaks in wrapper. But the
sensitivity to tampering is not very high. During the Tylenol
incident, he taught a class, where students did a series of
projects; one group focused on consumer education. He
doesn’t believe that the average consumer understands the
issue.

 

 

4.2 Procedure

Before the beginning of the focus group, consent forms, approved under IRB

09-225/ APP# i032704, were signed and collected (see Appendix E.

48

Consent Form Consent Form). The focus group lasted about two hours and a
haIf and participants were asked a series of questions from an IRB approved
moderator guide (see Appendix D, p. 93) about product tampering. During the

session, researchers took notes.

As a warm-up activity, participants were asked to introduce themselves and
provide the group with any background or experiences that they had had with
tampered products or tamper evident packaging. Following the initial warm-up, a
series of questions were introduced to establish the group’s baseline

understanding of the state-of-the-art in tamper evidency.

After approximately 30 minutes of discussion around the guided questions,
the group was split into pairs; each pair was moved to varied locations that were
isolated from other groups. Each pair was provided with: . basic household
tools, . the food product, a plastic, aseptically processed puree and . a liquid
(sports drink), . solid (bbs) and . powder (a powdered juice drink). Participants
were tasked with four objectives for this portion of the focus group: (1) to find the
least detectable way to introduce the solid (bbs) (2) to find the least detectable
way to introduce the powder (a single serving drink mix) (3) to find the least
detectable way to introduce the liquid (a brightly colored energy drink), and,
finally (4) to find the most ways to get into the packages provided. One hour was

allotted and then the group was reconvened.

After reconvening the group, they discussed the methods they used to

introduce the solid, powder and liquid into the provided product. The debriefing

49

lasted about 20 minutes followed by general questions about tamper evident

packaging features.

4.3 Background and experiences of participants

In order to benchmark current protective features of a given product and
derive insights into the full range of techniques that might be attempted to tamper
this product, a group of experts was recruited for the focus group. The
fundamental principle of the concept was that people with significant
backgrounds in packaging, food production and tampering would provide rich
insights into the techniques that could be used by the general public with regard

to tampering.

Using the moderator guide, the focus group moderator led the group in

discussion around a series of questions related to tampering (see Appendix D).

o What comes to your mind when you hear the phrase “tamper evident”

packaging?

When asked for their thoughts and impressions on the phrase “tamper evident
packaging”, one group member mentioned, “things” that are put into place to help
the consumers to identify whether or not tampering has possibly occurred.
Another theme that emerged as a result of this question was that it was the
consumers’ responsibility to check for intact tamper evident packaging. A
recurrent theme with this question and others was that the involvement of the

consumers in recognizing tampering should be minimized.

50

Discussing the terms “tamper evident” and “tamper resistant” revealed that
everyone had heard both terms, but did not necessarily know the differences
between them. One former faculty member indicated that although the names
were changed in the laws and regulations, the employed technologies were not
(see 2.3 Government Reaction to Tampering - US Legislation and Regulation).
He felt that the terminology change was largely a marketing move on the part of
the OTC industry to say that they are making things evident as opposed to
resistant. Following the discussion of the two terms, the participants were asked
for their knowledge about their difference. In conclusion, one team member

came up with the following definitions:

Table 2. Definition of “tamper evident“ and “tamper resistant“

 

Tamper evident How obvious something is once tampered

 

 

 

 

Tamper resistant How difficult something is to tamper

 

. How would a consumer know if a product has been tampered?

To explore the self reported behaviors of these individuals, we asked them
about whether or not they had ever knowingly consumed a product that showed
potential signs of tampering (e.g. a popped seal). Because all participants were
selected as subject matter experts from three areas (packaging, food processing
or manufacturing), it was expected that they were sensitized to the issues and
outcomes associated with product tampering as well as the packaging features
employed to mitigate or detect it. Nevertheless, the group consensus was that

thorough observation of the package before consumption doesn’t always happen.

51

Additionally, several members of the group indicated that even when they do
notice something is awry (e.g. a missing induction seal), that they would
frequently not want to be troubled by taking the product back to the store.
Further discussion revealed that at least two group members had rationalized
incomplete seals or tears as production problems or damage and consumed a
product that they thought would be safe anyway (e.g. mustard, cereal). They
also expressed that they further rationalized the consumption of the questionable
goods with thoughts related to the fact that a fatal tampering would not happen to

them and that they were extremely rare.

When asked about the regulations and requirements for tamper evident
features, limited participants discussed food products, which are, with the noted
exceptions of milk and alcohol, not required to have tamper evident features in
the US (see 2.3, p. 20). Discussion then moved to pharmaceuticals, specifically

OTC drugs.

As previously discussed, OTCs are required to have at least one tamper
evident feature and their packages are required to be prominently labeled so that
the product is not consumed in the event that the feature is not intact. A broken
feature should therefore be visible to the consumer, so the consumer knows if a
product has been tampered. For the package (category: blister with glued
paperboard wrap) that was used as “tampering object”, the observation was
made that when the blister was squeezed and the lidstock bulged, tampering
could be detected when the sound of gas escaping was heard. Participants were

asked about their familiarity with tamper evident features. The group provided a

52

list of commonly used technologies which included: shrink bands and sleeves,
printed foils over seals, induction seals, tear tapes (e.g. with color that sticks and

won’t go when removed) and jars with vacuum up closures.

4.4 Reactions

The general consensus of the group was that a tamper evident technology
generally has to be detected by the consumer in order to work. However, as
noted previously, several group members (of this sensitized team) had admitted
that they had ignored or completely disregarded products with broken or missing
TE features due to rationalization (it won’t happen to me) or laziness (I don’t want

to hassle with taking this back and I need to use it).

During the experimental phase of the focus group (1 hour), the different teams
were tasked with tampering a specific package with the common tools provided
by the research team. The group felt that they were able to adequately tamper
numerous packages in less than an hour, and felt that it was highly unlikely that
their work would be detected. The importance of a tamper evident device on a
food package was discussed, and the group expressed the importance of TE
features, not only for the consumer, but also for the brand. Any injury or fatality is
devastating to the brand, but TE solutions also need to strike a balance, as they
can’t be too costly. Nevertheless, for manufacturers it is important to comply and

to show an effort is made to protect the consumer.

As discussion around the guided questions continued, one of the recurrent

themes that emerged from the group was that tamper evident designs should

53

minimize the need for consumer intervention, and that tamper evident strategies
should use the consumer as a “last step.” Several members mentioned the
possibility of auto-identification technologies as tamper evident strategies.
Specifically discussed was the strategy of blocking the bar code if a package is
tampered so as to render it “unscanable” and, thus, “unsalable”. The group
further discussed the idea of rendering the bar code “unscanable” at purchase in

order to prevent items from being purchased, tampered and then returned.

Focus group results suggest that a layered approach to tamper evidence,
which does not rely solely on the consumer, is warranted. Not every consumer
attends to the employed tamper evidence features and, even when attended, at
least occasionally, they are disregarded. Technologies which can leverage auto
identification or the development of smart materials or sensors are further

recommended (see also 3.2 Antitheft and Track-and-Trace).

54

5 Characterizing the product: Headspace Analysis

Headspace gas analysis was conducted to evaluate the headspace gas from
samples of the same product used in the focus group analysis. The product
headspace gases were tested at beginning and end of shelf life at three levels of
tampering. The levels of tampering, which were informed by focus group findings
were: lid removal, syringe, and control (no tampering). The purpose of the
analysis was to characterize the amount of two gases: oxygen and carbon
dioxide of tampered and untampered product. This was done with the purpose of
exploring the feasibility of gas-based detection as a way to detect tampering of
two extremes. The researcher asked whether or not a gas-based sensor could
be developed that would identify leaks or possible tampering attempts that did
not result in false positives. Understanding the inherent variation of these gases

is necessary in order to develop such a sensor.

Four different kinds of purees were chosen for the analysis; Applesauce,
Bananas, Green Beans and Prunes. A gas chromatograph (T raceGC Ultra,
ThermoScientific) was employed to analyze headspace data. Initial tests and
calibration of the gas chromatograph (GC) was necessary for accurate and

precise measurement of the two gases (see 5.2).

5.1 Gas Chromatograph Trace GC Ultra
The gas chromatograph “TraceGC Ultra” (Manufacturer: ThermoScientific)

was used to measure the oxygen and carbon dioxide content in the headspace of

55

the four different purees. The gas chromatograph (GC) is equipped with a
thermal conductivity detector (TCD) and a Carboxen 1010 GC Plot Capillary
Column (Manufacturer: Supelco) and uses helium as carrier gas. The porous
layer open tubular (PLOT) column that is used as a stationary phase is a type of
packed column. This column is more effective in separating oxygen and
nitrogen, bewuse its pore structure is larger than other columns and can be used

up to temperatures of 250 °C [96].

 

 

 

 

 

 

 

 

 

  
 
  

 

 

 

 

 

 

 

 

 

 

 

 

injection needle
flow controller port valve
carrier gas j —-.= septum
inlet . -m purge vent
septum nut . H -
o-ring or / , _ l't t
Spl ven
heater block . . 3-way solenoid
injector /-‘ valve
liner
backpressure
regulator ‘
analytical
column to
detector

Figure 6. Diagram of GC flows for split injection [97]

The GC injector, which is attached with a nut to the heater block (see Figure
6), consists of an inlet tube for the carrier gas, the heater block to maintain the
injector temperature, an inlet (injector) liner, a septum, a septum nut, outlet tubing
and a fitting to hold the column in place. The carrier gas is introduced to the
head of the column into an unpacked open space above the restraining frit and

beneath the septum. The septum seals the top of the column through which the

56

syringe needle is inserted. The most common way of introducing a sample into a
GC is with a syringe. The GC needle needs to be very narrow in order to not
destroy the septum disc and to minimize the volume of sample remaining in the
syringe. For the injection of the headspace gas as well as the gases for the'
calibration, a gas tight syringe (1MDR-V—GT, Manufacturer: SGE) with a
maximum volume of 1 ml was used. The detector analyzing the gases is a
thermal conductivity detector (TCD), which is generally the least sensitive
detector, but applied for the analysis of permanent gases [97]. The TCD
responds to differences in thermal conductivity of analytes in comparison to a
carrier gas, helium, which is highly thermally conductive. The high thermal
conductivity of the carrier gas allows the detection of most compounds, which are
less thermally conductive. The presence of even a small amount of analyte
reduces the thermal conductivity significantly and, thus, the filament with a
constantly applied voltage will decrease in thermal conductivity (or increase in
resistance), changing the temperature of the detector. This change in

conductivity gives a signal which is then represented in the chromatogram [98].

For the instrument method, different modes of injections are available; split
and splitless injections. The split injection method (see Figure 6) is used for the
analysis. The main purpose of this injection method is to reduce the amount of
sample that is injected onto the column. Split ratios can be “as high as 1000 to 1”
and were set here to a ratio of 30 to 1. The split ratio refers to the ratio of the gas

flow through the inlet versus the gas flow that goes through the column [98]. For

57

this study, the gas flow was held constant at a flow of 3 ml/min and with a split

ratio of 30, resulting in a split flow of 90 ml/min.

5.2 Calibration

To calibrate the GC, different percentages of each gas were injected in order
to obtain a calibration curve for each gas of interest (02 and 002) (see Table 3).

The amount of gas injected is limited to 100 pl, because initial tests showed that
an injection of >100 pl resulted in overlapping of peaks. For the calibration the

following concentrations of each gas were used:

Table 3. Calibration gas concentrations

 

Oxygen 0% 5% 30% 50% 100%
Carbon dioxide 0% 5% 30% 50% 100%

 

 

 

 

 

 

 

 

 

Oxygen
Y = 225241*x R"2 = 09993 W: Equal
_.’ ,X/
1 I
200000009} /
15000000; '
(B '. //
e j //
< 4 /I
10000000 ] //
3: /'/
5000000;
. ,1"
0 20 40 60 80 100

%

Figure 7. Calibration Curve: Oxygen

58

Carbondioxide
Y= 310674*X R"2 = 0.9984 W: Equal

30000000? A

1”"
j l’/
/',

250000007? /”

2000000043] //

Area
\
\

1500000051 /1"

100000001 7/{”

,4 1’,’

5000000
./l
’11
,/’
/
0 Ti‘f‘YV'FWT‘TYT‘VT'TI'T'Y‘VTTT"I"Y‘T'Y'Y"’T"I.r'T-Y‘P—VT‘I‘TTYT'EI r r. rlr ”I TTiT . I1 Yrv_v TTI'.YV vwr TY_T

0 20 40 60 80 100
%

 

Figure 8. Calibration Curve: Carbon dioxide

The R-Value, or correlation coefficient, of the curve is the quantitative value of
the calibration curve and computed as the fraction of the total variation of the Y
values of data points that is attributable to the assumed model curve. Possible
values of R lie between 0 and 1, with 1 indicating the best fit. In this case,
correlation coefficient, or R-value, was 0.9975 or greater with a forced origin (see

Figure 7 for oxygen and Figure 8 for carbon dioxide).

5.3 . Materials and Methods

The purees tested with this experiment were aseptically packaged in a
multilayer plastic cup that incorporated an oxygen barrier layer. The defined
oxygen transmission rate (OTR) of the molded container was 0.006
cc/package/day. The cup is sealed with a high barrier lid which contained a layer
of aluminum. The labeled net-contents, by volume, were 2.5 oz for Applesauce,

Green Beans and Prunes and 3.5 oz for Banana puree. The approximate

59

headspace volume of the purees was approximately 28 ccs, depending on the

density of the product. The area of puree in contact with the headspace was
approximately 35 cm2. The maximum amount of oxygen expected for all purees

at its ‘beginning of shelf life’ stage was expected to be 52% based on
specifications provided by the product manufacturer. It is likely that oxygen

content can increase with shelf life, due to permeation.

Each puree type (four levels) was tested at the beginning and the end of shelf
life (see Appendix H. Details of tested product, Raw Data, p. 107). Additional
factors that were investigated included “tampering method” at three levels, either
‘lid removal’ or ‘syringe’ or no tampering (see Figure 9, p. 61). For the ”tampered
samples”, tampering was either conducted three weeks before testing (delayed)
or immediately preceding the analysis of headspace gas (i.e. the same day). All
samples selected for the first tampering method were tampered by peeling back
the lid halfway and exposing the contents for 30s (+/- 39) so that the headspace
was exposed to the ambient air. Aftenrvards, the lid was secured again to ensure
a closed environment. The samples for the second tampering method were
tampered by puncturing the cup wall in the headspace area with the needle of a
sterile syringe. Nothing was injected or removed from the headspace, and
immediately after puncturing the puncture point was closed. The rationale for the
two methods was that they represented extremes in exposure conditions. In
contrast to the ‘lid removal’, the syringe provided the product with a relatively
modest exposure to the ambient environment. After every tampering process,

the package was gently squeezed to ensure that the lidstock was properly

60

resecured. Each treatment was repeated three times (triplicates, n = 3) (see
Figure 9). Thus, the total number of samples tested was n=120. Run order was

randomized across all treatments.

L-d I /7 Immediately
emo a
/ I r v \ Delayed
Immediately
. . /
Beginning of S -
shelf life ' ymge \ Delayed
/ \ None —> None
Puree 1/2/3/4 Ed I /' 'mmed'ate'y
l remova
/ \ Delayed
S . / Immediately
n
Iizfgd of shelf —> yri ge \ Delayed
\ None —> None

Figure 9. Sample characterization

The dependent variables of interest, oxygen and carbon dioxide content of the
headspace, were assayed by gas chromatography (GC; TraceGC Ultra,
ThermoScientific; Milan, Italy) using split injection (see 5.1) and a 7 A PLOT
column under the following GC conditions: oven temperature 35°C to 190°C at
25 °C/min.; detector 200°C; injector 125°C. Aliquots of 100 pl were withdrawn
using a gas-tight syringe (1MDR-V-GT, SGE; Victoria, Australia) from the
headspace of the purees via a rubber septum (Illinois Instruments, Johnsburg, IL)
fitted to the composite film (lid of container) in order to eliminate the entry of air

into the package.

61

Statistical analysis

The data obtained were statistically analyzed using a two-way analysis of
variance (ANOVA) in a mixed model with the Statistical Analysis System (SAS)
software (SAS Institute, Cary, NC). The objective of the analysis was to
Investigate the effect of tampering on headspace gas concentration (see also 5.4
Null hypotheses) at a confidence level of 95%. Significant two-way interactions
represent differences of Least Squares Means. The normality assumption was
checked with ‘proc univariate’ using normal plots (visual assessment) and the
equality of variances assumption was checked with ‘proc sort” using side-by-side
box plots (visual assessment). The responses, oxygen and carbon dioxide, were
analyzed separately with unequal variances in the broad model (see 5.5.1) and
the narrow one (see 5.5.2). The broad scope analysis accords the factor ‘sample
type’ as a random factor, because the purees were not chosen with great intent,
but due to their availability. However, within the available purees there was an
objective to include both fruit and vegetable purees. Therefore, the data was

also analyzed in a narrow scope according ‘sample type’ as a fixed effect.

5.4 Null hypotheses

Researchers tested the following null hypotheses:

0 There is no effect of tampering method (lid removal, syringe, no tampering) on

the headspace gas concentration (Oz, 002)

c There is no effect of tampering time (delayed and immediate test) on the

headspace gas concentration (Oz, 002)

62

c There is no effect of sample type (applesauce, green beans, prunes and

bananas) on the headspace gas concentration (02, COZ) and

. There is no effect of expiration date (beginning of shelf life and end of shelf

life) on the headspace gas concentration (Oz, 002)

5.5 Results and analysis
For each treatment, triplicates were tested and descriptive statistics were

caICUIated (599 Table 5, for raw data see Appendix H)

63

Table 4. Legend for Headspace Tables and Graphs

 

 

 

 

 

 

 

 

 

 

 

A: Applesauce P: Prunes

G: Green Beans B: Bananas

SL: Shelf life Beg: Beginning of shelf life
, End: End of shelf life

T: Tampered NT: Not tampered

LR: Lid removal Syr: Syringe

Delay: Tested after 3 weeks Imm: Immediate testing

 

64

 

Table 5. Headspace Gas Data for 02 and 002

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Abbreviation Label Average oz 3‘33" “@5329" 3‘33?"
A, SL: Beg, T: LR, Delay 7.3 5.1 4.2 1.1
A, SL: Beg, T: LR, Imm. 14.2 11.3 1.9 1.7
A, SL: Beg, T: Syr, Delay 0.3 0.6 4.2 0.3
A, SL: Beg T: Syr, Imm. 0.7 0.4 4.1 0.1
A, SL: Beg, NT 0.6 0.8 4.0 0.4
A, SL: End, T: LR, Delay 8.9 8.0 4.6 3.4
A, SL: End, T: LR, Imm. 19.7 1.0 2.2 0.3
A, SL: End, T: Syr, Delay 0.3 0.2 7.1 0.3
A, SL: End, T: Syr, Imm. 0.1 0.2 7.2 0.3
A, SL: End, NT 1.1 0.6 6.8 0.2
G, SL: Beg, T: LR, Delay 5.6 1.0 5.4 0.6
G, SL: Beg, T: LR, Imm. 15.5 2.3 3.1 1.0
G, SL: Big, T: Syr, Delay 0.9 0.8 4.5 0.3
G, SL: Beg, T: Syr, Imm. 1.2 0.8 4.2 0.4
G, SL: BeiNT 0.6 0.6 4.3 0.3
G, SL: End, T: LR, Delay 18.4 2.5 0.7 0.4
G, SL: End, T: LR, Imm. 21.2 0.6 1.9 1.1
G, SL: End, T: Syr, Delay 0.1 0.2 4.9 0.1
G, SL: End, T: Syr, Imm. 0.5 0.8 4.8 0.3
G, SL: End, NT 0.4 0.4 4.7 0.2
P, SL: Beg, T: LR, Delay 8.8 2.1 2.5 0.2
P, SL: Beg, T: LR, Imm. 19.8 1.4 0.7 0.1
P, SL: Beg, T: Syr, Delay 3.6 5.8 2.0 0.3
P, SL: Beg, T: Syr, Imm. 0.6 1.0 2.2 0.1
P, SL: Beg, NT 0.5 0.6 2.2 0.1
P, SL: End, T: LR, Delay 10.7 4.5 2.7 0.8
P, SL: End, T: LR, Imm. 20.3 0.5 1.0 0.3
P, SL: End, T: Syr, Delay 0.4 0.8 3.5 0.4

 

65

 

 

Table 5. cont.

 

 

 

 

 

 

 

 

 

 

 

 

 

P, SL: End, T: Syr, Imm. 0.0 0.0 3.5 0.1
P, SL: End, NT 0.4 0.7 3.7 0.3
B, SL: Beg, T: LR, Delay 7.1 9.8 7.0 2.9
B, SL: Beg, T: LR, Imm. 19.4 1.8 1.2 0.3
B, SL: Beg, T: Syr, Delay 0.2 0.3 3.6 0.1
B, SL: Beg, T: Syr, Imm. 0.0 0.0 3.8 0.4
B, SL: Beg, NT 1.1 0.5 3.3 0.5
B, SL: End, T: LR, Delay 12.8 9.6 5.1 4.9
8, SL: End, T: LR, Imm. 15.9 4.1 4.1 1.7
B, SL: End, T: Syr, Delay 1.0 0.9 8.6 0.7
B, SL: End, T: Syr, Imm. 0.6 0.7 8.9 0.4
B, SL: End, NT 0.3 0.6 8.2 0.5

 

 

 

 

 

66

 

Figure 10 depicts the average and standard errors associated with oxygen

and carbon dioxide concentrations for each treatment combination. Graphs for

each type of puree are presented in the Appendix (Appendix H, p. 107).

%

25.0

20.0 .

15.0 -.

10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HELL..—

002‘ l
LICOZJ

_ __ Lid Removal Tampering -- Svrinqe Tampering No Tampering . __ _-_

c". .3

z__iu. " _:: i __ _ §

1" F“ “I
313i iii

SL: SL: SL: SL: SL: SL: SL: SL: SL: SL:

Beg, End, Beg, End, Beg, End, Beg, End, Beg, End,

T: LR, T: LR, T: LR, T: LR, T: Syr, T: Syr, T: Syr, T: Syr, NT NT

 

Delay Delay Imm.

 

Imm.

Delay Delay Imm.

Imm.

 

 

 

 

 

 

Figure 10. Comparison of oxygen and carbon dioxide concentrations (means of all puree types)
for each tampering method per shelf life stage and tampering time (if available)

Two dependent variables, percentage of oxygen and percentage of carbon

dioxide, were analyzed using the broad model and the narrow one.

5.5.1 Broad Scope

5.5.1.1

Oxygen

The mixed model design ‘Broad scope’ treated oxygen as the response

variable (dependent variable), ‘Sample Type’ as a random effect and the

following factors as fixed effects:

67

o Expiration Date (beginning of shelf life and end of shelf life)

. Tampering Method (lid removal, syringe, no tampering)

o Tampering Time (samples tested immediately and 21 days after tampering)
. Expiration Date x Tampering Method

0 Expiration Date x Tampering Time

0 Tampering Method x Tampering Time

0 Expiration Date x Tampering Method x Tampering Time

The first analysis of oxygen content provided no evidence to support
significance in any three-way interaction, and only one signifiCant two-way
interaction, ‘Tampering Method x Tampering Time’ (p<0.0001) (see Figure 11).
The factors “Tampering Method’ (p<0.01) and ‘Tampering Time’ (p<0.05)
indicated a significant effect at a: 0.05. The factor ‘Expiration Date’ did not
provide evidence of a significant effect on the percentage of oxygen as a

response.

68

 

 

 

Oxygen: Tampering Method x Tampering Time

 

 

 

 

__ l

. j b Tampering Method
.320 l —— — *— ~ " * l - Lidremoval ,
(I) ;.__ _. _ _ - 1 I
§ 15 I o Syringe
E10-’--—————r — r

l 8 ‘ f x Control

I 5 777 _§‘” 7 (Non-

' 0 IE- .4. _- . ~ ' tampered)

Control 0—day 21-day

{ (lrnmediately) (Delayed)

’ Tampering Time

h l
* means followed by the same letter are not significantly different (d<0.05) I

l Error bars show confidence intervals

 

 

Figure 11. Comparison of 02 concentrations (means) for Tampering Method x Tampering Time

The analysis indicated oxygen concentrations to be significantly different
when the packages with their lids removed were compared with those that were
tampered using a syringe and those that were not tampered at all. Within the
samples that had their lids removed, a significant difference was also indicated
for tampering time. The samples that had their headspace gas tested
immediately following tampering (tampered on test day) had almost double the
concentration of oxygen of those that had this analysis delayed (tampered 3
weeks before headspace testing) and approximately 10 times the concentration
of ‘Syringe’ tampered samples and not tampered samples. Analysis provided no
evidence of difference when the control group and the syringed were compared

(in either the ‘delayed’ or ‘immediate’ categories).

69

5.5.1.2 Carbon dioxide
The mixed model design ‘Broad scope’ was repeated using carbon dioxide as
the response variable (dependent variable). As before, ‘Sample Type’ was a

random effect and the following factors as fixed effects:

0 Expiration Date (beginning of shelf life and end of shelf life)

0 Tampering Method (lid removal, syringe, no tampering)

. Tampering Time (samples tested immediately and 21 days after tampering)
. Expiration Date x Tampering Method

0 Expiration Date x Tampering Time

0 Tampering Method x Tampering Time

0 Expiration Date x Tampering Method x Tampering Time

For carbon dioxide, two two-way interactions, ‘Tampering Method x
Tampering Time’ (p<0.01) (see Figure 12) and ‘Tampering Method x Expiration
Date’ (p<0.05) (see Figure 13), were significantly different. As with the oxygen
analysis, there was no evidence of significance examined for the three-way
Interaction tested when the dependent variable was the percentage of carbon
dioxide present in the headspace. The factors ‘Tampering Method’ (p<0.01) and
‘Tampering Time’ (p<0.05) indicated a significant effect at on: 0.05, but tests of

the factor “Expiration Date’ did not indicate evidence of significance.

70

 

l Carbon dioxide: Tampering Method x Tampering Time

 

 

 

 

 

 

 

 

 

 

 

 

l 8
,.. I J Tampering Methodl
‘ 9: 6 I“ — ——I—’—* ”4 t3____-m.£’_ I Lid removal 1
m ._
' C c, c: -
| ‘ ‘
8 2 I A _ _ 1 X Control 1
O b , (Non- \
I 0 I __ 'a , J tampered) |

5 Control 0—day 21 -day
I (Immediately) (Delayed) -
l

Tampering Time l

I * means followed bythe same letter are not significantly different (q<0.05)
I Error bars show confidence intervals

 

 

Figure 12. Comparison of 002 concentrations (means) for Tampering Method x Tampering Time

The ‘Lid removal” tampered samples, tested immediately after tampering, had
the lowest average concentration of carbon dioxide (see Figure 12). Their
concentration is about 2.5 times less than the average C02 concentration of
samples that were tested after being tampered with a syringe and those not
tampered at all. ‘Lid removal’ samples, tampered 3 weeks before testing
(delayed), didn’t provide evidence of a significant difference in their carbon
dioxide concentration compared to syringe tampered samples or the control
group. However, their average carbon dioxide concentration was twice as much
as that of the ‘Lid removal’ samples that were tampered on test day
(immediately). Analysis of the ‘Syringe’ tampered samples did not provide

evidence of a difference in C02 concentration when compared with the control

group. This was the case for both the delayed and immediate testing.

71

Multiple mean comparisons using the Least Squares Means also indicated
significance of the interaction term ‘Tampering Method x Expiration Date’
presented In Figure 13. The main difference for this interaction is at the ‘End of
shelf life’ stage. At the ‘Beginning of shelf life’, all treatments show a similar

concentration of carbon dioxide; about 3.3%.

 

 

, Carbon dioxide: Tampering Method x Expiration Date I
l 8 l

 

 

   

 

 

 

 

 

 

 

 

_—e—b l . i
r; ; _e_ ,, ]+b Tampering Method
€61W~~- a ~--r——» ____
2 ‘ 7“” +le removal :
i l: l
m .1- n l
L __
GE) 4 i L. -9—Synnge
N I
O 2 l j____ +Control
o l +3 +a . (Non- d ,
I 0 id——— 7 —_— l ____.__ A tampere ) I
I 1 2 .

Expiration Date
j (1 =Beginning , 2=End of shelf life)

’* means followed by the same letter within each Expiration Date are not ,
significantly different (d<0.05) .
° there is no significant difference between the means within Tampering I
Methods at a<0.05 ‘
Error bars show confidence intervals

 

 

Figure 13. Comparison of 002 concentrations (means) for Tampering Method x Expiration Date

Data trends are consistent and a comparison of the averages indicates at the
end of shelf life there is about 1.7 times the concentration of carbon dioxide in the

product’s headspace that there is at the beginning of shelf life. ‘Lid removal’
tampered samples show about half the concentration of 002 when compared

with ‘Syringe’ tampered samples and those that had been not tampered. ‘Lid

72

removal’ tampered samples did not provide evidence of a significant difference in

their carbon dioxide content between beginning and end of shelf life samples.

5.5.2 Narrow Scope

When the model assumed that the products (applesauce, green beans,
prunes and bananas) Were chosen purposefully (varied fruits and vegetables with
differing acidities), they were considered a fixed effect of the model for analysis,

making the scope narrower.

The mixed model design ‘Narrow scope’ analyzed the control grOup and
treated carbon dioxide as the response variable (dependent variable) and the

following factors as fixed effects:

0 Expiration Date (beginning of shelf life and end of shelf life)

0 Sample Type (applesauce, green beans, prunes and bananas)
0 Sample Type x Expiration Date

The narrow scope analysis for all treatments for oxygen and carbon dioxide

resulted in no evidence of significance for ‘Sample Type’.

The analysis of the control group (not tampered samples), however, provided
evidence of a difference for the factOrs ‘Sample Type’ (p<0.001) and ‘Expiration

Date’ (p<0.001) and their interaction ‘Sample Type x Expiration Date’ (p<0.001).

73

7 cgeel’QEE
Carbon dioxide: Sample Type x Expirafion Date
10.0 L , ~ ”LL

 

l

 

8.0
as Applesauce

l 3
RN :2: Green Beans I

gull.- - Ill Prunes

‘ I ‘ “c 77M,
L _ s

002 means
0)
O

P
o

l
l
l
l
.

           

2.0 \ g

0.0 - Si '3 '
Beginning of shelf life End of shelf life
1 Error bars show confidence intervals

 

 

 

Figure 14. Control group: Comparison of C02 concentrations (means) for Sample Type x
Expiration Date

Table 6. 002 concentrations (means) for Sample Type x Expiration Date

 

A plesauce Green Beans Prunes Bananas
COZ Conf. COZ Conf. COZ Conf. CO2 Conf.

 

 

 

 

 

 

 

 

 

 

 

 

 

(%) Interval (%) Interval (%) Interval (%) Interval
Beginning
ofshelflife 4.0 0.40 4.3 0.31 2.2 0.14 3.3 0.56
Endof
shelflife 6.8 0.25 4.7 0.27 3.7 0.29 8.2 0.61

 

In the control group (not tampered samples), ‘Green Beans’ differ from the
other purees in that the analysis does not provide evidence for a significant
difference when pain/vise comparisons were conducted. The samples,
‘Applesauce', ’Prunes’ and ‘Bananas’ show a highly significant difference. The
carbon dioxide content of these purees is rising from the beginning to the end of

shelf life.

74

5.6 Discussion

Tampered samples varied significantly in their composition of headspace gas
when they were compared to control samples. Finer analysis of this result, using
the Least Squares Means method indicates that only samples with their lids
removed during the tampering process can be distinguished from the control
group. High standard deviations for some of the ‘Syringe’ tampered samples led
to the assumption that this type of tampering cannot be reliably discerned from
the control group and could lead to ‘false negatives’. ‘Syringe’ tampering did not

result in a sufficient change In the headspace gas concentration.

 

02, C02: Lid removal x Tampering Time

   
  

 

 

 

A ----- ‘__ “I Lid removal ,
l 2\: l ,

3 + 02 :

g f
, 2 —9— C02

 

Control 0-day 21 -day
(lrnmediately) (Delayed)

Tampering Time
. Error bars show confidence intervals

P W ___ ___

 

 

Figure 15. Changes In 02 and 002 concentrations for ‘Lid removal’ tampered samples

The likely reason for the consumption of oxygen and evolution of carbon
dioxide in ‘Lid removal’ tampered samples after three weeks (see Figure 15) Is

most likely due to aerobic microorganisms particularly molds. They consume

75

available oxygen and convert it into carbon dioxide. After three weeks, the
oxygen concentration is almost half of what was measured right after tampering

and the carbon dioxide concentration approximately doubled.

The carbon dioxide content found in the control group samples is probably
related to the pH and the chemistry of the puree. The concentration in carbon
dioxide is higher for the vegetable puree (Green Beans) than for the fruit purees
at the ‘Beginning of shelf life’ stage. Thus, it is likely that the production of
carbon dioxide is related to the pH of the food, because the vegetable puree
(higher pH, see Table 7) shows a higher concentration of carbon dioxide than the
fruit purees. The low acidity and the lower content of glucose (reducing sugars)

in green bean puree likely Inhibit the increase of carbon dioxide over time.

Table 7. Acidity and pH for the tested purees

 

 

 

 

 

 

 

 

 

Puree Type Acidity pH
Green Beans Low acid 5.6 - 6
Bananas Medium acid 4.5 - 5.2
Prunes Acid 3.6 - 4.3
Applesauce High acid 3.1 - 3.6

 

Fruit purees (Applesauce, Prunes, Bananas), however, have a higher carbon
dioxide concentration at the ‘End of shelf life’ stage (see also Figure 14). Visual
assessment also showed that browning occurred in the fruit purees, which was
especially noticeable in ‘Applesauce’ and ‘Banana’ samples. This could be
related to a chemical reaction such as the Maillard reaction (non-enzymatic
browning). The Maillard reaction is a complex set of reactions initiated between

carbonyl (e.g. reducing sugar, ascorbic acid) and amino compounds (e.g.

76

proteins, amino acids). Carbonyl derivates of the non-enzymatic browning
sequence react with free amino acids resulting in the degradation of the amino
acid to carbon dioxide, also known as the ‘Strecker degradation’ [99]. The
observed browning confirms to some extent that the food product degraded. The
observation of accumulating carbon dioxide in the fruit purees seems right,
because the amount of glucose Is probably higher than for the vegetable puree.
Higher acidity, provided by the fruit purees, is with a higher glucose content a

better starting point for the Maillard reaction.

5.7 Conclusions

Results suggest that the development of a gas sensor intended to detect
tampering in a nitrogen gas-flushed aseptic puree packaged in plastic would be,
to some extent, reliable for oxygen and carbon dioxide. A carbon dioxide
sensor, however, would not be as practical, due to a higher variability in the Initial
carbon dioxide content for different puree types. An oxygen sensor would be
more efficient due to the less variability in the concentration and the low initial
concentration. However, a needle prick through the plastic package did not
indicate an increase In oxygen. Thus, tampering with a needle cannot be reliably
be detected with an oxygen sensor. Tampering via ‘Lid removal’, on the other
hand, shows a great difference in the oxygen content after tampering. Therefore,
tampering that Involves the opening of the lid, or that has a similar amount of gas
exchange could be discerned from packages that maintained their integrity (for

these four products) by an oxygen sensor if it was designed properly.

77

6 Recommendations for future research

The evaluation of tamper evident packaging demonstrates that research in
this area stagnated in the last twenty-five years. To ensure state-of-the-art
technology, new innovations of tamper evident features are important. Tamper
evident features are required for OTC drugs. his review, however, suggests that
those in use commercially haven’t changed tremendously since the
implementation of the law (1982). Solutions of future tamper evident features
that don’t rely on the consumer are beneficial and recommended. The retailer as

control point should also be considered.

The headspace gas data (oxygen and carbon dioxide) obtained for fruit and
vegetable purees and visual assessment of the color of the purees illustrates that
scientific color measurement can give more insights regarding the fruit and
vegetable puree behavior. Color measurement is recommended, when analyzing

headspace data of fruit and vegetable purees.

The headspace analysis of an aseptic puree packaged in plastic indicates that

future research should focus on the detection of minimal invasive tampering.

78

Appendices

79

 

 

 

 

 

 

 

 

 

 

 

 

_No: .00. 00.0 :00.00 050:0 ”0:0003:
05.00E0. E0205 .0: =2 0. 003 0. 00.000 0.5. .000. 0:0 00.00 00.0
2:3 .025 0.0000 5 00.000 0:. .0 00.... 0000.0 0:0 E05 000020000. 5.00008 00
0:50:00 800:00 00.30000 5 0000.0 0. 02:08 0000.0 __0v.0_z 0:05 <0: 5000.0 0.06 .000?
05.00E0. E0205 .0...
2:3 .025 0.0000 00.30000 05: 5 000000 00.5 5.0.002, :00.00 .0. 5.0.0... mo
.0E30:00 E0000 .0 0000.. .000 00:000. 200.0085 0.0 00300.0 <0... 00.05 00.000 .000?
Im:_.00E0. E0205
E: 2...: genre .0: .000 cases we
.0E30:00 E0000 0.0 .00>-m~ 02:00 5.2. 0000. 5000 .0:0_>... <0: _0:0_>... .000?
:0: .E0300 00.5 5 0.0:; 0.00
05.0060. E0205 0:. E2. 00080. 0. 0300.0 .000 0.... 5 0.00:0
2:3 .025 0.0000 055060. 0:. 5 0:30.. .00. 0.0.5 050 .0990
.0E30:00 E00:00 0.5... 0.00:0 00005 0 5 50 02.0..030 E0030 (w: 0.00:0 0009.0 N00?
05.00E0. E0205
5:3 .025 0.0000 .00 3 00.0 0000....0 0:. 5 00.0 0.0000 0 ”02:00 _0:0.>._.
.0E30:00 500:00 .22. 0050. 0.0.5 00.30000 .0:0_>._. 50:0..0 0.0m. (m: 50:0..0 0.0m. N00?
0509.000 0:00.05
.0 .06.. “.0 00>... 00.5000 05.0055 .50.... 5:300 00300.0 00>

 

 

08300.0 .00..3000E.0:0 0:0 000. .0 0E0205 05.00E0... .m 0.00...

0:..00E0... .0 E30... .< x.0:000<

80

 

 

 

 

 

 

05.00.00. E0205 .00... .:00 20-5:0E-m.
5:3 .055 0.0050 .00 0. 0.3.:0000 .0.0>00 00. :00. 0:0 000. >000 .0 000.
.0E30:00 0.00000 .0. 0 5 000.0 .0 0000.0 om... :00. 0.0:. .30 :0Eo>> <0... >000 .0000 .00?
_NN. 000.0 0.0. 2000 000030 :00.
0:0 0.000 00:30 0:0 .30.. :000 000 2000 .0.0..0
0:..00E0. E0205 .0:.0.:. 05:0 _.0. 00... 5000. .0550: 0 0:0 05000
5:3 .055 ”0.0050 0.5. 5 005300. .00.:0>0 5.3 0000. 0.0.5 00.30000 .30..
0:50:00 0.00:0”. ...30...-NF 00.0030= .0 00002000 .00 .000. .< <0: .0. 00.0030 .00..
:0...0.x0
0:0
05.00:.0. E0205
5:3 .055 SN. 0003000 000:... .0 0.000 0: .0000 .050.
0:50:00 HE0..0..0.r -:0: .00.:0>0 5.3 00:00.00 0.03 0000.0 :00_.00 N 0:00 053.. :00_.00 000.
05.00:.0. E0205 .0: 000.0 .0
5:3 .055 “0.0000 0.050 00:.0E00 0.050 .050 0530.0 _0.0:. 00.0.
.0E30:00 0.00000 03. 0:0 0000 5.0300 0050.000 000. >000 .0 .00 v5 000. >000 N501 000.
05.0090. E008. .0 0
5:3 _05:_ 0.0000 .2000 0:0 5 00.0>000.0 0.0.5 >.30.0_.: .0 0000.. .0530
0:50:00 0.00:0... .0..0 002000 00. ..0 0020. 0. .0530 00.0.00 ..< 0.... .00.0 >500 000.
05.00:.0. E0205
5:3 .055 ”0.0050 F: 00.0 00:. >00.0.. 302 0.0 0300 00
.0E30:00 E00000 -.00>-0N 0 .00.:0>0 5.2, 00.50. 00.5 .00260 0300 <0... -<.030 00.0.. .0005

 

 

 

 

 

 

 

0:8 .0 0305

 

 

81

 

 

 

 

 

 

:0.to.x0 E0205 .0: 00.3.5 0:0 0: .000. 2.00 0050 .0.:0...0.x0
5:3 .00>000 :0 >0 00.000. :000 000 >000... .00.:0>0
0:50:00 ”0.0.0.05 5.2. 0000. 00.0::0>0:. 0:0 0.0.03.0 >E005 >:0:.00 $5002. >Eo05 000.
05.0.0 0500:.0.
5:3 .0 0.00.0 .5 EN. 05300.0 00:000. 00:.< 022.000 0
0:50:00 0.0.0.05 5.2. .0300.0 05 000.00 0. 00000.5 0.0.:0...0.xm. 0._0..03< 05300.0 ..0:.< 000..
0500.00. E0205 :0 .>_030..00 E00. .0 .30. .5 =0.
5:3 .0...:. “200.0 0.0000 .00.m. 0000050.: >_0000 5.2. 00.05.:0E00 .0.0>>
0:50:00 E0000 0.5 0.0.5 0.:0. 00.000. 0.3. 0.>0..>0.00 2: 0.005 02.0.00 0.00.
.0. 0:8 .800 .0... c. 8000.. 02. 0.00.8 0....
00.0.0 0. 0050.0 0.0000 0002. 00.000. :000 000 00000
5:: E0205 oo :00. 0.0.: .:00 .0000 500. 5 00:..>0 0 0:...
0:50:00 .00>000 0. 00:50.0 002. 0.0.00.5 05.0 .000. 0 052. 00.0.0 <0: .0000 0005
.>00E0.
00000000
05.85... 5.000... .00. 050. .0260
5:3 .055 ”20.0 00.000. 0. .0300... 00.0 :0... 0.0 00>-B .00.:0>0 00000001
0E30:00 0.0000 53.000 0...... 0000.00. 002. 002.00 00000000 00 0. <0: 0.>0000 N00.
.00 E 002000 .020... .0030
E0... 0._0..03< :.0.00>> 5 0.0.. 000. >000 000.0
:0...0.x0 E0205 __0 00>0E0. N50... .00_00:< 00.. 5 0000 050000
5:3 .055 ”200.0 05v. 0:000 05 0. .03000 0 5 02.00 500
0:50:00 0.00:0... 0. .:00 :000 000 02:0>0 5.2. 0000. 0.. 000. >000 0..0..03< 000. >000 ~50... N005

 

 

 

 

 

 

 

....00 .0 0.00...

 

 

82

 

 

 

 

 

05.0.0
000
05000.0. .:00.05 Km. 000.00
0.03 .0...:_ 0.0.0 0. 0:0 0: ”00.00.00.000 0.02. 0.0.. 00. .00. 00.0.02. 000.
0.030000 0.0000". 00.05 00.0: 0..0..0 0:0 50.. 000.00 000.0.000 0.0.. <03 >000 0000 0000
.00 3 .0030. 00 002. 00:00.0
0: .00000 :0 5 0500.000 0.... 5 00.3.0030 .0
0.00. .0. 0.00. 0. 0050.030 000.00 020.00 0.0000
00.00.00 05000.0. 0...... 0.00. 00.0.5 000 0>00..3. 00.0.. .0000
0.03 .0 0.00.0 .. 002.00. 0>00 0. 05.0.0.0 030.0 0.00.. .0500
00.30000 ”0.0.0.0... :0 00.. 00.0. 00.00.00.0..00>.000. 020.00 000000 020.00 meow
00.00.00
000 EN. 00:00.00 0.02. 0.0000 N 0.0
00.0000. .:00.05 00.000. 0. 0300.0 0.0. :0. 0. 0003 000.00 0.06.
0.03 .0...:_ >000 0 .00.:00>..0 0...... 0000. 0000 000 05000000
.0050000 00.0.0... 0... 5 0003000 005000 0003005000 00.0: 0..0..03< 00.0... 0000
05.0.00 EN. .00>000 0.0.. 00000. 0000
0:0 000 0.03000 000000030 =0 .00 .30 002. 0000.00
00.0000. .:00.00. 0000 000 00000 00. .00. 0000.030 .0. 05:02.
0.03 .0...:_ 0.0.0 0 0...... 00000 0 00.0 000.0 >000.0 .0000 0:002. 00
0050000 00000m 5 0000.0 000 00.0..000 0...... 0000. 000. >000 .0 00.. >000000 000. >000 0.0.4. .000.
.02.
000000 0.0 .00. >000000 00. 000.03 002000
00.00.00 00.00.00. :0 00. 000. 00:00.00 000.0 0. 0500.000. 0.00000
0.03 .0 0.00.0... 00. 00000000. 00.0 $00.00 00. 0. 00. 000.
00.30000 ”0.0.0.0... 00.00.00.000 .:00 0.: 0.0000003... 0.0.00>-om >00..E00 00k 000. 00002 000.

 

 

 

 

 

 

 

.08 .0 0......

 

 

83

 

.mo : ..5> 55.502 :555 ..:55: 5.. 55.0

55... .5555 55265 5.22. 555 5:5 5.5, 5559.5.
:555 555 .55. 5.. 5 5:50. 5:5 55.5 w>o 55.

o. .:53 5505.0 .500. >555 5550 5555.. 5555
55:50.0 5.. 5 5 :55. .55 5:5 5555.5. .0 5.5.55.
5.» 553.55 555520 55 5553 .055 55552. 02,.
55.5000 55.5 _500_ 55.05 .5 .:5505 055555.

 

 

 

555555. .:5505 55.05.. .500. >55 5 E55. 555:5 5:5 5.535.
5:: _5_._:_ 505.5 5555.4 5: 555520 552, 5: .552. :5w 5 w>o 5 .5 500.
5:55:00 80551 500. >55 .0 5.. 5 553 555565. 55.5555 :55 < <m: >55m 5550 meow
.8: 555.5
553 5:0 02 55.5 55. 5552 55.55 .5950
553 5.5 .55.. 5:55> 0. 555.500 55555 505
5.05.5 5.55.05 5 ES. 55 55055:: :5 5.55:.
555585. .:5505 0. 5555.5 0555:. 5 555: 55>» 55w .5 5555050
5:: _5_._:_ 5.05.5 .5055; 55.5 5555 5.55525 5 5555 55..
5:55:00 505:5m 500. >555 5:5:55 .55m 5.5m. 555555. 55:52, < <5: .55m 5.55m meow
.3: <55 2. o. 8:89 £555 5 53.5 .0 $82
0: :555 555 555. 5:5 553 555555. 553
555555. 555.05 >555 505535 0: 552, 555.. 5555 55:5.
.0 5.59.: 52 .55 52:56 5; 58. 32 .o 55.85 85.5 58.
”E5595... 55:055.. 555 0. 55550 >555. :5E :55... <5: >55m 5550 woom
555055 555555. EN. .5555. 55555 :52, 5:05
5:: .0 5.55:... 55an0 55. o. .:55 553 5:5 50555 .555 o. 5:55 550055.555.
5:55:00 ”55:95.. 5055555 5 5.5, 5505_ :555 555 55 550.055 555.53. 550.055. meow

 

 

 

 

 

 

 

5:8 .5 535.

 

 

84

Appendix B. Field Study on Baby Food Packaging

To compare the safety features of baby food packages from a survey
conducted in 1984 (see 2.2 Research on Tampering) with currently available
commercial packages, the same study was performed on a single category,
prepared baby food for a single store, Meijer, Lansing (see Table 10). l repeated
the work of Hotchkiss and Gravani, but felt that the use of a single store was
justified in the repeat study, because the food category tends to be dominated by
national brands, which do not tend to vary packaging formats widely in different
stores. The degree of protection against tampering is ranked with safety

categories, defined by Hotchkiss and Gravani defined in the following:

Table 9. Safety categories for protection against tampering

 

TE container would likely meet current FDA regulations for OTC

Category I drugs and has a label statement describing the TE feature

 

TE container would likely meet current FDA regulations for OTC

Category ” drugs, but there is no label statement describing the TE feature

 

TE container would likely not meet current FDA regulations for
Category III OTC drugs, but the inherent nature of the package gives some
degree of tamper evidency.

 

Containers have no tamper evident feature and can be easily

Category “I reclosed without detection.

 

 

 

 

The 1984 field survey was done on many more categories than just baby
food, but our repeat comparison was limited to baby food, especially baby puree.
Package surveys were conducted at the store. In Table 10, the current baby
puree packages are evaluated with the same methodology as in 1984. In total,

five different brands offered baby puree at the store surveyed. It should be noted

85

that brands sometime offer the puree in different package forms. Therefore, the
number of brands per container type is mentioned separately (see column
“Container Types”). The five brands offer their puree in three different container
types (glass jars, plastic tubs and plastic pouches) and used different closures
and tamper evident features. These are enumerated in the column entitled
”Closure typesfl' E feature”. Some tamper evident features, however, are the
same (e.g. sealed thermoform), but they incorporate a different closure type or
material (see row entitled “plastic tub”). The number preceding the closure
material doesn’t represent the amount of puree packages, but the amount of this
type of closure and TE feature for a given brand. For purees packed in glass
jars, there was one brand that had a TE feature in addition to the vacuum button;
a shrink band was incorporated around the closure, for some, but not all
packages. The column entitled safety category aggregates the information
presented in columns 1-3 and uses the system established by Hotchkiss and
Gravani to evaluate the containers that were surveyed. (Purees packages in
plastic tubs, for instance, are offered by two brands. One brand offers it in two
different closure types, but integrates the same TE feature. The puree closed
with a heat sealed closure, however, has a label statement that describes the TE

feature leading to safety category I.

86

Table 10. Current baby puree packages evaluated with method used in the 1984 field survey [34]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Brand Container Closure Types/ TE Safety % i" 93°“ category
Types Types feature category I II III IV
5 glass jars 4 Lug thread-Metal/ III - - 80 -
(4 brands) TE: Vacuum button
1 Continuous Thread, ll - 20 - -
TE: Plastic- Shrink band
plastic tub 1 Heat Sealed- I 33.3 - -
(2 brands) Composite/ TE: Sealed
Thermoform
2 Induction sealed- II - 66.7 - -
Metal/ TE: Sealed
Thermoform
plastic 1 Heat Sealed- II - 100 - -
pouch Composite/ TE: Sealed
(1 brand) bag/pouch
Glass jars fared poorly in the safety ratings (Safety category lll — see Table

10), largely due to the fact that the vacuum button in closure is not considered as

TE feature for OTC drugs (see 2.3.2 Drugs). On the other hand, a sealed plastic

tub or pouch is considered having TE features and therefore listed under safety

category II.

Table 11. Summary: Average % per safety categories

 

 

 

 

 

 

 

 

 

 

Average # Average % in each
Product Group Brand types Category
I II III IV
Baby Food (prepared, dry), 1984 - 61 39 -
Baby Food (prepared), 2009 5 11 62 27 -

 

 

87

 

Table 3 compares the results of the original study conducted by Hotchkiss
and Gravani (1984) when their results for the baby food category was compared
to our brief survey of this category in a single grocery store. Above findings
suggest that the use of tamper evident packaging has somewhat increased over
the last 15 years, however, most of current the packages surveyed do not include

a labeling statement on the existent TE feature.

88

Appendix C. Rosette Protocol

The effectiveness of the employed tamper evident feature of our test
packages (aseptic puree in a plastic container) was determined by the use of the
Rosette protocol (see 2.2 Research on Tampering and 2.5 Evaluation Methods).
For this purpose, three single aseptic puree packages were violated with usual
household tools. Each violated package and two control packages (i.e.,
unviolated) were presented to a panel that consisted of company executives.
The inspector had to detect evidence of tampering within five minutes, without

using special devices of any sort.

The accuracy of the rating (established by use of the test) is dependent on
reliable and objective answers to all questions. Any violated package, where
violation was obvious to inspectors under the previous instructions, cannot be

used for obtaining a rating for the tamper evident feature [37].

For the rating of the TE features (see Figure 16), the factors of violation time,
degree of knowledge of packaging, utilized equipment, material of TE feature and
TE feature visibility are considered. For a successfully violated test package, the
required time to tamper was 10-15 minutes, which earns points according to the
factor values (see Table 12, p. 91) a 2.0. The person, who violated the package
successfully, was highly specialized in packaging or research training which
resulted in the value 10.0. The utilized equipment was defined as “required small
hand tools”, where no special equipment purchase was required, which resulted

in a 2.0. The TE feature could be reproduced by using locally available materials

89

and therefore resulted in a 4.0. The visibility of the feature (blister) is in
widespread use and associated with consumer awareness (C), which resulted in
the value 4.0. This value is in this case multiplied with factor E or every other
visibility factor that accounts. E asks for “Destroys package upon opening”,
which in this case is applicable, because tampering with the blister pack causes

destruction of the TE feature (value 10.0).

Tamper-Evident Feature Rating Index-1989 Revision

 

 

 

 

Product . . . Date 5/30 M79
TestCode i; (£3 l'ff A FeaMWW(€L/y€53
Index
Time Requrred (lowest level) - 10 mm /j§m /\ 2. 0
Knowledge Level (lowest level) 40 .0
Equipment Used (highest level) 2 - 0
Feature Material (all that apply)
0 + 9'0 + + = Lf J

 

Feature Visibility (all that apply)
(BWLE

 

 

 

 

33’3me xE_Q_0= (_o' .0
Index Rating 6??
Supervisor “(.109 WIUIW Date 3/6 ( Oq

U 5555

 

 

Figure 16. Tamper-Evident Feature Rating Index for a sealed plastic tub

As result, all three tampered packages successfully passed the inspection.
The “Rating lndex” (according to the Rosette Protocol) for determining the
effectiveness of the TE features ended in the value 58. This method, however,

has to be repeated with any new implemented TE feature for comparison.

90

Table 12. TE Testing Procedure: Values accorded for Factors [37]

 

Deflee of Knowledge of Packagi_ng Required

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No special knowlegqe or ability required 2.0
Basic knowledge of manufacturing process 3.0
Less than six months experience in packaging 5.0
More than six months experience in packagi_ng 6.0
Specialized training in packaging field 9.0
Highly specialized in packaging or research training 10.0
Feature Material

Feature can be reused 0.0
Replaceable by material normally found in homes 2.0
Feature reproduced using locally available materials 4.0
Generic feature, not available from most outlets 6.0
Printed feature, limited distribution 6.0
Printed feature, local source, small quantity 7.0
Feature not available locally, in small quantity 8.0
Printed feature, large order guantity required 9.0
Proprietary feature, available without order 9.0
Proprietary feature not available 10.0
Time Required to Violate Successfully

Under five minutes 1.0
Five to thirty minutes 2.0
Thirty minutes to one hour 3.0
One to three hours 4.0
Three to six hours 5.0
Six to twenty-four hours 6.0
One to seven days 7.0
Seven to fifteen days 8.0
Fifteen to sixty days 9.0
Over sixty days 10.0

 

 

91

 

 

Equipment Utilized

Table 12. cont.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No special equipment needed 0.0
Requires small hand tools 2.0
Requires household appliances 4.0
Requires special tools not found above 6.0
Special Equipment Purchase Required

Under$50 6.0
$51-100 7.0
$101-500 8.0
$501-800 9.0
Over $800 10.0
Feature Visibility

A. Not visible if package is sealed 1.5
B. Requires close inspection 2.0
C. Widespread use with consumer awareness 4.0
D. Prominent feature, high visibility 7.0
E. Destroys package upon opening 10.0

 

 

92

 

Appendix D. IRB Approval and IRB approved Moderator
Guide

MICHIGAN STATE Initial IRB
March 30. 2009 Approval

To: Laura BIX
153 Packaging Building

 

Re: IRB# 09-225 Category: EXPEDITED 2-7
Approval Date: March 30, 2009
Expiration Date: March 29, 2010

Title: Tampering in aseptic purees packaged in plastic

The Institutional Review Board has completed their review of your project. I am pleased to advise you that your
project has been approved.

The committee has found that your research project is appropriate in design, protects the rights and welfare of
human subjects, and meets the requirements of MSU's Federal Wide Assurance and the Federal Guidelines (45
CFR 46 and 21 CFR Part 50). The protection of human subjects in research is a partnership between the IRB
and the investigators. We look forward to working with you as we both fulfill our responsibilities.

Renewals: IRB approval is valid until the expiration date listed above. If you are continuing your project, you
must submit an Application for Renewal application at least one month before expiration. If the project is
completed, please submit an Application for Permanent Closure.

Revisions: The IRB must review any changes in the project, prior to initiation of the change. Please submit an
Application for Revision to have your changes reviewed. It changes are made at the time of renewal, please
include an Application for Revision with the renewal application.

Problems: If issues should arise during the conduct of the research, such as unanticipated problems, adverse
events, or any problem that may increase the risk to the human subjects, notify the IRB office promptly. Forms
are available to report these issues.

Please use the IRB number listed above on any forms submitted which relate to this project, or on any
correspondence with the IRB office.

Good luck in your research. If we can be of further assistance, please contact us at 517-355-2180 or via email
at IRB@m§u.edu. Thank you for your cooperation.

Sincerely,

272mg.

Dan Ilgen, Ph.D.
SIRB Chair

Figure 17. Initial IRB Application Approval

93

Focus group moderator guide
GOALS
We are trying

1. To gather information about the various forms of tampering for an aseptic

puree packaged in plastic

2. To gather information about the different mechanisms by which tampering

can be effectively detected
3. To generate ideas for tamper evident features and sensors
Note

We will provide a liquid, solid and powder to expert teams, and ask that they
get them into the packages without being detected. We will also challenge teams
to come up with as many different ways in as they possibly can. Goals are only

for the moderator, they are not intended to be told to the participants

Time allotted for each section is as follows:

 

Section Minut
Intro 5
Warm up 10
Baseline 20
Tampering 60
Group Debrief 20
Tamper Evident 20
Total 135

94

iNTROQQCTIQN (5 minuteS)

Hello everyone. My name is Julia. Welcome to our focus group discussion. A
focus group is just a group of people who get together to a talk about a specific
topic. For a portion of the focus group, you will be divided into subgroups, and
asked to tamper these packages (show the aseptic puree). Your teams have two

challenges:

1. To get the three materials we are giving you (bbs, Gatorade and crystal

light) into the packages in a way that cannot be detected

2. To come up with as many different ways as you possibly can to get each

of the materials into the packages

To do this, we have provided some common household tools, but please feel
free to use your imagination; you may use anything in the building to accomplish
your task. One group can use this room; another can use 167, and the third can

use room 173.

I’ll be your moderator today and with me are other members of the team,
Cindee Wilcox, Evangelyn Alocilja and Yun Wang and Laura Bix; we will be
taking notes as your teams start to do your work. After one hour, we will call you
all back in here to see what you have come up with, and to discuss a few
questions about product tampering. I have to confess that I am not expert in this
topic. During the focus group portion of the day, my job will be to lead the

discussion and make sure that everybody gets a chance to talk. Basically, the

95

research team is here to _find out what you think, to listen to your opinions and to

gather insights from your actions.

We want everyone to feel comfortable talking about their ideas and opinions

so here are some typical rules for focus group discussion:
1. There are no right or wrong answers here.
2. Everything you have to say is very important for us.

3. Feel free to disagree or agree with other’s opinions. We expect people to

have different opinions
4. Please try not to interrupt each other.

5. I might skip over you if have talked a lot or I might call on you if you

haven’t talked at all. My goal is to try to get everyone to talk.

Ok. Whatever is said in this room will only be used for research purposes and
presentations in conferences. We will be transcribing our discussion today and
some of the dialogue that occurs during the tampering session today. Please be

aware that specific thoughts and discussion will not be attributed to individuals.

Any questions? All right, let’s get started!

WARM-UP i1 0 minutes)

I’d like to begin by having each of you tell us your name, and a bit about your

background regarding tamper-evident packaging.

96

BASELINE (20 minutes)

1. What comes to your mind when you hear the phrase “tamper evident”

packaging? PROBES

a. Are you familiar with the terms “tamper evident” and “tamper

resistant”?
b. What do each of these terms mean to you?

0. Do you investigate packages for tamper evident features or potential

tampering prior to using products?
d. Do you know that what products require tamper evident features?
2. How would a consumer know if a product has been tampered? PROBES
a. What types of tamper evident features are you aware of?
. Shrink bands
. Printed foil over seals
0 Tear tapes

b. How do you feel about these technologies and the current approach to

tamper evidence?
LAMERING (60 minutes)
(Break participants into groups and dismiss them to the other 2 rooms)
ACTIVITY: Please try to tamper these packages with the following objectives

1. Find the least detectable way to introduce the bbs you have been provided

97

2. Find the least detectable way to introduce the powder

3. Find the least detectable way to introduce the liquid

4. Find the most ways into these packages (detectable or undetectable)
RECONVENE GROUP to room 159 with tampered examples

GROUP “DEBRIEF” (20 minuteS)

 

(Set the powder, liquid and solids onto the table for participants to view)

1. Please comment on the experience that your group had trying to get the

liquid into the products provided. PROBES

a. Each team briefly present your “best work” that involved getting the

liquid into the package.
b. Were there any challenges that were unique to the liquid?
c. Was there anything about the liquid that made it easier?

d. What were the varying ways that you were able to put the liquid into

the package?

2. Please comment on the experience that your group had trying to get the

bbs into the products provided. PROBES

a. Each team briefly present your “best work” that involved getting the bbs

into the package.
b. Were there any challenges that were unique to the solid?

c. Was there anything about the solid that made it easier?

98

d. What were the varying ways that you were able to put the solid into the

package?

3. Please comment on the experience that your group had trying to get the

powder into the products provided. PROBES

a. Each team briefly present your “best work” that involved getting the

powder into the package.
b. Were there any challenges that were unique to the powder?
c. Was there anything about the powder that made it easier?

d. What were the varying ways that you were able to put the powder into

the package?
TAMPER EVIDENT FEATURES (20 minutes)
(Set some samples of the aseptic puree on the table)

1. What features of the packaging would you consider to be tamper evident?

PROBES

a. What approach did your team use to defeat the features that you felt

were tamper evident?
2. How do you feel about these features? PROBES
a. Are the features effective?
b. Would you notice the tampering that was shown by the teams?

c. Do you think that the average consumer would notice the tampering

that was done by these teams?

99

3. What suggestions do you have for tamper evident features? PROBES
a. Should the tamper evident features rely on consumers?

b. What approaches would you take with design?

Thank you very much for participating in our focus groups! Your input will help

us to improve the tamper evidence of package design. Have a good day!

100

Appendix E. Consent Form

Research Study Consent Form

Tampering and Asefl Pgree

You are being asked to participate in a research study. We are conducting a
study that investigates the different ways that product packaging can be

tampered. Ultimately, this study attempts to:
0 Gather information about the different mechanisms of tampering

0 Gather information about the different mechanisms by which

tampering can be effectively detected
. Generate ideas for tamper evident features
To participate you must:
. Be willing to try to tamper packages

. Be willing to share your thoughts regarding the best way to tamper

the packages (provided by the research team)

. Be 18 years of age or older

You will be assigned to a team with other participants (teams will consist of 3-
4 people). Your team will be provided with some basic household items, a case
of aseptic puree product and a liquid, powder and solid. We ask that your team

find a way to insert the liquid, powder and solid into the puree product (to tamper

101

them). In addition, we are curious about the number of ways into the package;
we are looking not only for the most difficult to detect methods of tampering, but

also the largest variety of techniques.
You are free to withdraw from the study at any time for any reason.

The “tampering” session will be followed by a focus group that is meant to
synthesize the insights that you have garnered during the course of the
tampering“ session. During the course of both the tampering session and the
focus group meeting, members of the research team will transcribe thoughts
around this issue in an attempt to gather insights for the research team that can
be used to develop and refine future tamper evident designs. All information
provided in the transcripts will not be tied to your name, and only members of the

research team will have access to them.

You will not directly benefit from your participation in this study, although your
participation in this study may contribute to an understanding of the mechanisms

of tampering, serving to inform designers about this issue.

Possible risks include the chance that you could injure yourself as you try to
open packages. (This will be particularly true if you choose to use tools to try to
open or close any of the packages). The research team will have first aid

supplies on hand should you need them.

If you are injured as a result of your participation in this research project,
researchers from Michigan State University will assist you in obtaining

emergency care, if necessary, for your research related injuries. If you have

102

insurance for medical care, your insurance carrier will be billed in the ordinary
manner. As with any medical insurance, any costs that are not covered or in
excess of what are paid by your insurance, including deductibles, will be your
responsibility. The University’s policy is not to provide financial compensation for
lost wages, disability, pain or discomfort unless required by law to do so. This

does not mean that you are giving up any legal rights you may have.

You may contact Laura Bix at 517-355-4556 with any questions or to report
an injury. If you have any questions or comments regarding this study, please
contact Dr. Laura Bix, Assistant Professor of Packaging, at 153 Packaging

Building Michigan State University 48823, 517-355-4556 or pixlaura@msu.ed_u

If you have questions or concerns about your role and rights as a research
participant, would like to obtain information or offer input, or would like to register
a complaint about this study, you may contact, anonymously if you wish, the
Michigan State University's Human Research Protection Program at 517-355-
2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail at 202 Olds Hall,

MSU, East Lansing, MI 48824.
I voluntarily agree to

Participate in this study:

 

Date:

 

You will be provided with a copy of this form.

103

Appendix F. Data Collection Sheet and Results

Data Collection Sheet - Focus Group
“Tamper Evidence in the Plastic Packaging of an Aseptic Puree”
Female Male Other

Age

 

Race

Profession

 

Results of Data Collection Sheet

Table 13. Data Collection of participants in focus group

 

Gender Age Race Profession

 

Male 80 Caucasian Packaging Professor Emeritus

 

 

Male 22 Caucasian Packagiiwfitudent

 

 

Male 43 Caucasian Packagfljg Educator

 

 

Male 23 Caucasian Packagigg Student

 

 

 

 

 

 

 

 

 

 

 

Female 42 Caucasian Packaging Margger

Female 24 Asian Biosystems @gineerim Student
Male 21 Caucasian PackaginLStudent

Female - - Biosystems Engineering faculty
Male - - Biosystems Engineering faculty

 

104

Appendix G. Matrix Development

Based on the different TE categories, evaluated through literature search,
patent review and brainstorming, a Matrix with tamper evident solutions (x-axis)
versus tampering methods (y-axis) was developed. The tampering methods
were identified with help of the focus group (see 4 Focus group). The table
includes the rating of technologies identified to be the most promising solutions
for the aseptic puree packaging. The rating of the incremental solutions,
however, is based on the existing technologies and not based on the assumption
of turning them into a TE solution. The many unknowns regarding the leap
solutions made a clear rating impossible and therefore the rating was left blank.
To find out the security level the current TE feature bears, the effectiveness of
the TE feature is determined (see Appendix C: Rosette Protocol). But “the best
feature for a product is the one that provides the greatest resistance to violation

for the product in its current form and size” [31].

105

SJOSUBS

SUBQ GIISOdLUOQ DUB IBISW '91,
SJBUIBIUOD |OSOJeV ‘3],

999190 I?

SJGddBJM UJ|I_—_l '|,

106

products

0 No Detection

 

Figure 18. Matrix: Tampering Methods vs. Technologies

Appendix H. Details of tested product, Raw Data

Table 14. Expiration dates of tested product

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type Samme # Expiration Date Exp. Date

Applesauce 1-1 5 4-Oct-10 Beg.

Applesauce 16-30 16-Nov-09 End

Green Beans 31 ~45 18-Sep-10 Beg.

Green Beans 46-60 12-Feb-1 0 End

Prunes 61 -75 1 1-Oct-1O BeL

Prunes 76-90 9-Jul-09 End

Bananas 91-105 26-Nov-10 Beg.

Bananas 106-1 17 14-Jun-09 End

Bananas 1 18 10-Mar-09 Eg.

Bananas 1 19, 120 29-Aug-09 End
Table 15. Legend for raw data tables and graphs

Exp. Date: Expiration Date Beg: Beginningof shelf life

End: End of shelf life

T: Tampered NT: Not tampered

LR: Lid removal Syr: Syringe

Delay: Tested after 3 weeks Imm: Immediate testing

 

 

 

 

 

107

 

 

Table 16. Raw Data for Puree 1: Applesauce

 

N00 >06 .95

1.11

 

3.. «co .0 252:4

3.80

 

 

3.29
0.98

 

 

3.89
4.39

 

 

4.38
3.95

 

 

4.17
3.78

 

4.39 0.35
3.78
6.39

 

 

 

6.82
2.33

 

1.87 0.25

2.28

 

 

NO >06 .Uuw

5.12 5.41

0.60 3.78 0.35

8.03 0.65 3.44

 

3.. «o ._o 2:22

6.30
2.78
12.86

21.21

20.21 11.31 0.76 1.75

1.14
0.00
1.03
0.00

1.19

0.47 0.39 4.23 0.15

0.58
0.39

0.00 0.81
1.56
3.17
18.11

5.53

19.12
20.80 0.97

19.12

 

an 2.02 .N« EE. ._. u
>530 35.an8 Ecru 053 «no...

1

2
2

2

3

1

2
2
2

 

an 932 .NH Sm .wu m...
60522 actanu...

 

NH ._.2 .vu ._.

 

an Em .r... sum .83 .em.

 

an 9.02 .w u >& «no...
:0 .T ES .88 acreage

 

 

mu OFRNN inn SEEN mu oEmfi
.mn cvsvufi .Vn OEMNN .nn 2.3 EN
.Nu 2.3—kw ._.u SB EN .33. «we.

5

 

u Eoucwm

14
59
86

85
92
94
37
76
31

73
58
60
98
1 1

47

87
89
27

15

 

 

% oEEmw

 

 

 

 

 

 

 

 

 

 

10
11
12
13
14
15
16
17
18
19
20

 

 

 

 

 

 

 

 

 

 

 

21

 

 

108

 

 

 

 

 

 

 

 

 

 

 

 

 

 

t. Noo >6:.Bm_ &. m. n.

L m 0 0 0
6. 330.3525. m. u... m .6. w. M a w. w
.:IV 6 7 7 7 6 7 6 6 6
m
_m «0 >865. "a. m. :76.

O O 0
§Z93§o£< 0% m % m m m m M :7.
0 0 0 O 0 0 1 1 0

 

nu ocoz.Nu EE_.Fu
>360"actonfiscotaofizuno... 1 1 1 2 2 2 3 3 3

 

an 232 .NH Sm ._.u m.—
fiofics. 55.2.86... 2 2 2 2 2 2 3 3 3

 

N".—-Z.F":—:111111222

 

Nnvcménéomfiwaodxm 2 2 2 2 2 2 2 2 2

an 0.32 N" no “no... m

57:53.88 859:2. 1 1 1 2 2 2 3 3 3 _
mu 2:3 .:u ozmuaimn SEEN
.muotfia.:uoramaauozfia 5 8 6 7 5 6 1 1 8
.Nu :55 .fi :55. .83 “3:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Eocm 100041 3
% _u ¢5H795862H
_
#2155 2345678902 %
222222223,

 

End,
NT

End,

T: LR, T: LR, T: Syr, T: Syr, NT T: LR, T: LR, T: Syr, T: Syr,

Delay lmm. Delay Imm.

End,
Delay lmm. Delay Imm.

109

 

A, SL: A, SL: A, SL: A, SL: A, SL: A, SL: A, SL: A, SL: A, SL: A, SL:
Beg, Beg, Beg, Beg, End, End,

Beg.
Figure 19. Headspace Variation: Applesauce

Table 17. Raw Data for Puree 2: Green Beans

 

N00 >03 .uum

 

3.. «co .6 .525:

5.57

 

4.80 0.59
5.97
3.89

 

 

1.96 1.04

3.59
4.38

 

 

 

4.34
4.29

 

 

3.72
4.04

 

 

4.54
0.30

 

 

1.16

3.14

 

1.16 1.12

1.26

 

 

NO >2. .Em

1.01

 

3.. No .6 2.8.5

5.89
6.46
4.50
14.46

18.07 2.28
13.86
1.24

0.00 0.77 4.84 0.28

1.40
1.88

0.37 0.76 4.60 0.44

1.31
1.18

0.73 0.59 4.47 0.27

0.00
20.51

18.96 2.50 0.50 0.45

15.62
21.66

20.44 0.64
21.38

 

an ego: .Nn EE. .w
>a_on ”acted—:3 Late 2:: «no...

2
2
2

1

2

3

1
1
1

2
2
2

 

an 262 .NH im ._.u m...
”co—=05. acted-:3.

 

N" ._.z .ru ._.

 

an new .wu .mcm demo .nxm

 

 

an 262 um n >Mol amp...
:o .9" BEN .98 6:26:56:

 

an 255 .2." 26$ .6" 25%.
.mn 23a .4" 23% .nu 235
.N... 265 .71. Sara 62...: Zoe

4

 

a. £0.55.

80

62
1 04

52
26

29

72

57
45

25

10
109

67
105

117

21

111

56

 

 

u oEEww

 

31

 

32
33
34
35

 

 

 

 

36 40

 

37
38
39
40

 

 

 

 

41

 

42

 

43

 

 

45

 

46
47

 

 

48

 

49

 

50
51

 

 

 

110

 

 

moo >66 .36)

 

3.. «co .6 5:325:

4.91

 

 

4.92
4.45

 

 

4.88
4.43

 

 

4.80

 

Table 17. cont.

NO >06 .Em

 

 

 

0.19 4.80 0.07

0.80 5.05 0.31

0.36 4.88 0.24

 

3L «0 .6 5:325

0.32
0.00

1.39
0.00
0.00
0.71
0.00
0.48

 

an 262 .NH EE. J. .1.
>22”. "actanS Esta 6E: amok

1

 

an 262 .NH Sm ._.u m...
"cop—ups. actanmh

 

N" ._.z ._.u._.

 

an new .Fn .mom 6qu dxm

2

 

 

an 932 N .1. no «wok
:o .71. BEN .38 656:5:

1

 

 

mu OENEN hn OFBNN on oEmNN
.mn etewfi .Vu OEMNN .mu 33 EN
.Nu ctm EN .rn 02m EN .320 «wok

 

% Eoccmm

78
30
65

13

101

118

48

68

83

 

 

 

u. anwJ

52
53
54
55
56
57
58
59

 

 

 

 

 

 

 

 

60

 

 

 

 

 

 

End,

End,

End.

 

End,
Delay lmm. Delay Imm.

Beg, End,

 

Beg, Beg, Beg, Beg,
T: LR, T: LR, T: Syr, T: Syr, NT T: LR, T: LR, T: Syr, T: Syr, NT

G, SL: G, SL: G, SL: G, SL: G, SL: G, SL: G, SL: G, SL: G, SL: G, SL:

Delay lmm. Delay lmm.

Figure 20. Headspace Variation: Green Beans

111

Table 18. Raw Data for Puree 3: Prunes

 

N00 >0“. .Em

 

3.. «co co 352:1

2.30

 

 

2.74
0.75

 

 

0.63
1.74

 

 

2.13
2.31

 

 

2.17

2.11

 

 

2.23
3.01

 

1.74 0.85

3.34

1.05

 

 

0.77 0.27

1.31

 

 

«0 >3. 3m

1.37 0.67 0.06

5.83 2.22 0.25

0.51

 

g «o .6 356:2

10.18

9.88 2.07 2.38 0.23

6.46
21.29
19.57
18.59
10.35

0.00

0.53

0.00

0.00 0.99 2.26 0.07

1.71
1.12

0.00 0.59 2.36 0.13

0.26
8.98

15.86 4.52
7.33
20.75
20.51

19.78

 

an 262 .NH EE. ._. 1
>260 Huston—:5 Leta 6E: “not

1
1

2
2
2
1
1

2

3

1

2
2
2

 

an 932 .NH Sm ._.u m...
"vogue—2 gran—Eur

 

N" ._.z ._.u ._.

 

«u Em .T .63 .33 .93

 

 

an 2.02 .w n 2% «web
:o .3" BEN .33 65.3.53

 

an 25a s... 26$ .6" 23%.
.mn 233 .:u :28“ .nu SEE
.Nu :55 .T :55 .33 .35

4

 

a. Eoucum

35

103

17
91

112

43

79
49

116

108

12
44

100
75

38
84

77

19
102
55

 

 

a. 2955

 

61

 

62

 

63

 

64

 

65

 

66
67

 

 

68

 

69

 

70

 

71

 

72

 

73

 

74
75

 

 

76
77

 

 

78

 

79

 

80
81

 

 

 

112

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. moo >o:._3m_ %. w ”.26
L W. 0 0 0
C
8. 5333:3652: w. M m % M. m. w M. %
4nlu 3 3 3 3 3 3 3 3 3 ,
m W
m «0 >339 m. m w 2
O O 0 W
.éuococcsoel‘ m v.6 m m m. m w m m 2
0 1 0 0 0 0 1 0 0 2
an ccoz.un EE_._.u W
>230“actonESLctmcEzumo... 1 1 1 2 2 2 3 3 3

 

 

on 952 .NH Sm .9" w...
605.22 mctanfl. 2 2 2 2 2 2 3 3 3

 

NHL-zn—‘Hnr111111222

 

~uu:m€n.mcm.oumodxm 2 2 2 2 2 2 2 2 2

mn ocozdu>aoumch

5.? SEN .33 6:28.53. 1 1 1 2 2 2 3 3 3
an SEEN Kn otwwfi .mu BEEN

.mn OF\VN\N .QH O=flN\N .Mfl OEwEN 8 4 7 5 5 8 4 8 7
.Nu oEw EN J." 2.3 EN .32... «we.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

quucmm42616m/6952
H 4 9 6 4 1 3 H 8

#03an 2 3 4 5 6 7 8 9 0 /o
8 8 8 8 8 8 8 8 9 o

 

 

 

 

 

 

 

 

 

 

End,

End,

 

End, End, End,
Delay lmm. Delay lmm.

T: LR, T: LR, T: Syr, T: Syr, NT T: LR, T: LR, T: Syr, T: Syr, NT

113

Beg, Beg,

P, SL: P, SL: PTSL: P, SL: P, SL: P, SL: P, SL: P, SL: P, SL: P, SL:

Beg, Beg,

 

Delay lmm. Delay lmm.

Beg.
Figure 21. Headspace Variation: Prunes

Table 19. Raw Data for Puree 4: Bananas

 

N00 >66 .65

1.75

 

3.. «co co 353:2

3.63

 

 

8.89

1.03

 

1.56 0.27

1.15
3.57

 

 

 

3.51

3.50

 

 

3.57
3.24

 

 

3.75

10.2

 

4.65 4.86

0.48
2.96

 

 

3.22

 

NO >66 .65

9.82 8.40 2.91

1.76

0.00 4.23 0.41

 

 

g no .6 358:;

18.43
1.71
1.16
18.57
18.24

21.44

0.57

0.00 0.33 3.65 0.07

0.00
0.00
0.00
0.00

1.1

5

1.59 0.50 2.76 0.50

0.60
1.86

16.44 9.61
19.98
18.28

11.15 4.14 6.11

18.37

 

an 232 .Nfl 8:: J. n
>260 65.6088 .666 656 “66...

1

2
2

1

2

3

1
1

2
2
2

 

an 6.32 .Nu L>m ._.n m.—
u6o562 0:660E6...

 

N" ._.z ...u...

 

an 6.5 ._.n .060 .6660 dxm

 

 

nu 6:02 .w u >wmam60.
:o .9" BEN .33 6:23.53

 

 

 

mu OFRNN .hu OEmNN on otmfi
.mn OEVNN .vu OEMNN .nu 020 EN
.NH 020 EN ._.u 02m EN .366 “660.

8
7
7

3
7
5
8
2
6
4

 

% £3.50

50
63

120
28

32

1 8
33
74
66

95

16

 

 

a 6.9560

 

91

 

92

 

93

 

94
95

 

 

96
97

 

 

98

 

99

 

100 119

101

 

 

102 97
103

 

 

104 36

 

105 88

 

106 53

 

107 106

108 22

 

 

109 69

 

110 41

 

111

 

 

114

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m. moo 5: .362 m u... M. 2 22 m. .2
w 0 0 0 20 C2
11 . 2 20 I22
.61. 23.36.35.664 .42. w. w. ._./c m... w :2. M. .4. -l -i
TI. .0 _ _ . _
a 2 .
T NO >66 .652 w. % mew. 1
0 0 0 2
2 _
23:23:35.4 m. m n. w m .2... m. m m 2
0 1 1 0 0 1 0 1 0 2
2
_

 

an 6.32 .Nu EE. ._. n.
>260 "65.6685 .666 656 «66...

 

 

 

 

 

1
1
1
2
2
2
3
3
3
i

B, SL: B, SL: B, SL: B, SL: B, SL: B, SL: B, SL: B, SL: B, SL: B, SL:

 

 

nu 6coz.~n ._>m.—.u m.— 2 2 2. 2
"6966s. 05.6686... 2 2 2 2 2 2 3 3 3 2

 

N".—-Z.P".—-111111222

 

 

 

 

NH6:M.-.n.06m.3uodxm 2 2 2 2 2 2 2 2 2

on 6coz.wu>.6i0 66... 2
5.335333 6:23:23 1 1 1 2 2 2 3 3 3 _

an :52: .:u 21% an 26" 2
.6" 23a .:n 233 . u :55 . _
.uu 6:65 .T 6:65 .33 .3: 2 2 2 ..

23.121

 

 

 

 

 

6
2
2
2
7
1
7
1
6

 

 

 

fl

 

 

 

 

 

 

 

 

 

 

 

 

#:5650443 309 1 |.-.
2211
3 5 7890
u6_nEmmu1M1Mm1112~ %

 

End, End,

End,
T: LR, T: LR, T: Syr, T: Syr, NT T: LR, T: LR, T: Syr, T: Syr, NT

Delay 'mm Delay lmm.

End, End,

 

Beg, Beg, Beg,
115

Beg. Beg.

Delay lmm helm: Imm.

Figure 22. Headspace Variation: Bananas

Bibliography

116

Bibliography

10.

Euromonitor International. 2008. Baby Food - US. Country Sector Briefing.

Federal Anti-Tampering Act. 1983. Tampering with consumer products 18
United States Code 1365.

Muth, A. 1999. Ed. Stehlin, l.B. Forensic medicine sourcebook Health
reference series. 1st ed. Detroit, MI: Omnigraphics.

Clarke, R.V.G. and GR. Newman. 2005. Designing out crime from
products and systems Crime prevention studies,. Monsey, N.Y.: Criminal
Justice Press.

Anonymous. 1987. Expensive, deadly threat: Everyone pays a price to
prevent product tampering. The Financial Post, Toronto, Canada, June 29.

Godar, SH. 2003. A Model for Communication Locus In International
Product Crisis Management. In AlB-SE Annual Meeting. Cleanrvater.

Dietz, P. 2000. Product Tampering: The Foundation for American
Communications

h_ttp://www.facsnet.org/index.php?option=com content&view=section&lavo
ut=blog&id=5&ltemid=100003.

Doeg, C. 2005. Crisis management in the food and drinks industry a
practical approach. New York, NY. ; [Great Britain]: Springer.

Morgan, PW. 1988. Tampered Goods: Legal Developments and
Marketing Guidelines. Journal of Marketing: 86-96.

Bell, R. 2008. The Tylenol Terrorist: Death in a Bottle. In Tenorists &
Spies, trutv.com.
http://www.trgtv.com/Iibrarv/crimelterrorists spies/terrorists/tylenol magic;
slindex.html.

117

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Saltzman, J., J.R.E. Milton, J. Staff and A. David. 2009. Fatal tampering
case is renewed FBI searches a condo in Cambridge. The Boston Globe,
February 5, metro section.

Anonymous. 1986. Tylenol not poisoned in plant , probe finds. The Globe
and Mail, Canada, February 12, news section.

Russell, C. 1986. Putting a Lid on Product Tampering ; FDA ls Buried
Under Complaints After Consumer Scares. Washington Post, May 23,
First Section; A13 section.

Anonymous. 1997. Extortionist again targets German food products
company Thomy. Deutsche Presse - Agentur, August 12.

Oldham, R. 1998. Gerber Management’s Response to the Glass Scare of
1986: An Ethical Analysis of the decisions of Gerber’s Management Team:
Oklahoma Baptist University.
h_ttp://www.pillowrock.com/ronnie/gerber.htm.

Anonymous. 1986. Poisonings; Copycat killers. The Economist, March 29.

Barron, J. 1986. Jerseyan is Poisoned by Cyanide in Soup; Tampering is
Cited. The New York Times, September 4.

Anonymous. 1989. Police to investigate mercury in butter. The Guardian,
London, February 11.

Young, J. 1989. Poisoned baby food bought by shopper ; Heinz. The
Times, London April 11.

Shenon, P. and S.T.T.N.Y. Times. 1989. Chilean Fruit Pulled From
Shelves As US. Widens Inquiry on Poison. The New York Times, March
15.

Stehlin, LB. 1995. FDA's Forensic Center: Speedy, Sophisticated
Sleuthing. FDA Consumer: The magazine of the US. FDA 29 (6).

118

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Gladwell, M. and W.P.S. Writer. 1991. Sudafed Poisonings Lead FDA To
Reconsider Use of Capsules ; Tamper - Resistant Packages Seen to Lull
Consumers. The Washington Post, March 6. -

Anonymous. 1992. Headache Remedy Recalled Over Fatal Tampering.
The New York Times, December 25.

Doug Kirby, K.S., Mike Wilkins. 1996. The Pepsi Product Tampering
Scandal of 1 993. Novato: RoadsideAmerica.com.
http://www.roadsideamerica.com/rant/pepsipanic.html.

Pollack, A. 2004. FDA Finds Traces of Poison in 2 Jars of Baby Food in
California ; No One Is Harmed. The New York Times, July 29, A; National
Desk section.

Anonymous. 1998. Extortionist targets Nestle food concern. Deutsche
Presse - Agentur, February 3.

Anonymous. 1997. Australian firm in product - tampering nightmare.
Deutsche Presse - Agentur, February 15.

Doneman, P., D. Murray and S. Parnell. 2000. Extortion sparks tablet
recall. Courier Mail, Queensland, Australia, March 17.

Anonymous. 2005. Pesticide found in Snickers bar. The Advertiser, July 7.

Sneden, J. 1983. Testing of tamper-resistant packaging, [1], vii, 142
[i.e.50] leaves: Michigan State University. School of Packaging, 1983.

Rosette, J.L. 2005. Packaging, Cosmetics and Pharmaceuticals: Forensic
Packaging Concepts, Inc.
http://mrw.interscience.wilev.comlemrw/9780471238966/kirk/article/cosmr
ose.a01/current/Ldf.

Anonymous. 1983. New tamper-evident closure changes color when its
twisted. Food Engineering 55 (3): 71-71.

Anonymous. 1983. Consumer influence. Package Engineering 28 (7): 91.

119

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

Hotchkiss, J.H. and RB. Gravani. 1984. The current status of tamper
evident packaging for foods: a field survey.

Hewartson, LA. 1989. Evaluation of the overall effectiveness of selected
current tamper resistant packages, xi, 152 leaves: Michigan State
University. School of Packaging, 1989.

Rosette, J.L. 1985. Development of an Index for Rating the Effectiveness
of tamper-evident packaging Master of Science Thesis. Santa Ana,
California: California Coast University.

Rosette, J.L. 1992. Improving tamper-evident packaging : problems, tests,
and solutions. Lancaster, Pa., U.S.A.: Technomic Pub. Co.

lwaszkiewicz, RA. 1991. An evaluation of tamper-resistant packaging : a
method for measuring tamper-evidence. In School of Packaging, 235 East
Lansing: Michigan State University.

Reisenwitz, TH. and G.J. Wimbish, Jr. 1996. Over-the-counter
pharmaceuticals: Exploratory research of consumer preference toward
solid oral dosage forms. Health Marketing Quarterly 13 (4): 47.

Pascall, M.A., K. Lee, A. Fraser and L. HaIim. 2009. Using Focus Groups
to Study Consumer Understanding and Experiences with Tamper-Evident
Packaging Devices. Journal of Food Science Education 8(2): 53-59.

Johnston, RG. 2006. Tamper-Indicating Seals. American Scientist 94 (6):
515.

Food and Drug Administration. 2005. Guidance for the Industry: Labeling
OTC Human Drug Products Questions and Answers.

Bix, L., N.M. Bello, R. Auras, J. Ranger and MK. Lapinski. 2009.
Examining the conspicuousness and prominence of two required warnings
on OTC pain relievers. Proceedings of the National Academy of Sciences
of the United States of America 106 (16): 6550-55.

Addington, P. Tamper-Evident Packaging: The Food Club International
h_ttp://www.thefoodclub.org.uk/TEPackagLnggpj.

120

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

Anonymous. 1950. Packaging's Hall of Fame: Four Roses. Modern
Packaging 11: 84.

Federal Bureau of Alcohol Tobacco and Firearms (Batf). 2005. Affixed
Closures. 27 Code of Federal Regulations 19.662

Australian Government - Department of Health and Ageing. 2003. Code of
Practice for the Tamper-Evident Packaging (TEP) of Therapeutic Goods.

Stock.Xchng. 2009. Pills (Image ID: 866422).
h_ttp://www.sxc.hu_/photol86642_2_.

Stock.Xchng. 2009. Aerosol Can (Image ID: 444031).
mpzllwww.sxc.ht_rlphotol444031 .

Soroka, W. 2009. Fundamentals of Packaging Technology.

Food and Drug Administration. 2007. Standards for Grade "A"
Pasteurized, Ultra-Pasteurized and Aseptically Processed Milk & Milk
Products, Grade "A" Pasteurized Milk Ordinance (PMO). 19p. Capping,
Container Closure and Sealing and dry milk product storage.

Lewin, T. 1982. Worries of Food Tampering. The New York Times,
December 27.

Henry Cohen, NE. 2005. Products Liability: A Legal Overview.

Food Marketing Institute. 1997. Product Tampering - Newsroom Advisory.
h_ttp:I/www.fmi.org/gocs/medialproduct tampering1pdf.

 

Food and Drug Administration. 2002. Public Health Security and
Bioterrorism Preparedness Response Act.

American Society for Testing and Materials. 2006. F 1448 - 93a: Standard
Guide for Selection of Security Technology for Protection Against
Counterfeiting, Alteration, Diversion, Duplication, Simulation, and
Substitution (CADDSS) of Products or Documents.

121

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

Jotcham, R. 2005. Authentication, Antitamper, and Track-and-Trace
Technology - Options To Protect Foods. Journal of Food Protection 68 (6):
1314, 4.

Kinrvan, M.J., D. Mcdowell, R. Coles and Knovel (Firm). 2003. Food
packaging technology. Oxford; Boca Raton, FL: Blackwell; CRC.

Robertson, G.L. 2006. Food packaging : principles and practice Food
science and technology. 2nd ed. Boca Raton, FL: Taylor & Francis/CRC
Press. h_ttp://cat<ir.Iogqov/caJLdir/enhancements/fv0648/2005043706-
d.html

International Hologram Manufacturers Association (lhma). 2008.
Holograms Battle Counterfeiting. In Packaging Digest.
http://www.packagingdigest.com/article/CA6583507.html?g=Holograms+B
attle+Cognterfeiting.

Bix, L. 2008. Illicit Activity and the Supply Chain. In Medical Packaging
PKG 455 Lecture Notes: Michigan State University.

Gasnier, Y. 2000. Food product packaging with a hidden built-in anti-theft
device and a method for preparing such packaging. US: Labeyrie S.A.

Anonymous. 2006. Smart labels: RFID labels to account for more than
85% of smart labels demand in 2014.(market research)(radio frequency
identification). Adhesives 8. Sealants Industry,
mpJ/www.accessmvlibrarv.com/comsglsgmmarv 028645340730 lTM
(accessed January 14, 2010).

Fontelera, J. 2006. RFID Demand Heats Up - Opportunities open for
converters as more end-user customers clamor for the tracking
technology. In Converting Magazine.
http://www.convertingmaga_zine.com/article/CA6399487.html.

Walmart. 2009. Electronic Product Codes (EPC) — Helping Walmart Help
Customers and Suppliers. EPC Fact Sheet,
h_ttp://walmartstorescom/cjownload/2449.pdf (accessed.

Roth, A., A. Tsay, M. Pullman and J. Gray. 2008. Unraveling the food
supply chain: Strategic insights from China and the 2007 recalls, 22.

122

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

_lLttpzllezproxv.msu.edu:2047/Iggin?grl=httg://Qroqgest.umi.com/ggdweb?di
g=143661771 1 &Fmt=7&clientld=3552&RQT=309&VNfl1e=PQ_Q.

Marra, N. 2009. BERICAP Galileo II (Cap for Brasil). In E-Mail

Anonymous. 2007. Bericap introduces tamper-evident closure for vinegar.
In FoodProductionDai/y.com.
_h_t_tg:/lwww.foogproc_luctiongailv.com/content/view/print/2_2150.

Heinlein Plastiktechnik. 2008. ONE LOOK®: New Dimensions in closure
technology. h_ttp:/lwww.heinlein-
plastik.de/09 sprachen/rgssisch/07 news/ON§_I_.OOK 07 eng Logos R

Spdf.

Inline Plastics Corp. 2008. Safe-T-Fresh.
h_ttp:/lwww.inlineglastics.com/newsme/safe t fresh.html.

Schwarz, S. 2009. Tamper-evident container with tear band. US: Letica
Corporation.

Johnston, A. 1992. Tamper evident folding carton. US: Rexham
Corporation.

3m. 2009. 3MTM Tamper Evident Label Materials.
mg/Isolutions.3m.com/wp§lporta|/3Mlen LS/Converter Solgtionflome/
Prod Info/Prod Catalog/?PC 7 RJH9U5230GE3E02LECIE20GER4 nid
=0WMGBM7OSYbeFLW8KCNRGle.

Daphne Allen and K. Marouf. 2001. Printed Tear Tape Catches the Eye
and Protects the Consumer. In European Medical Device Manufacturer.
h_ttp://www.devicelink.com/emdm/archive/O1/05/007.html.

Clarifoil. 2009. Anti-counterfeit, tamper evident films. In Packaging Digest.
http://www.packagingdigest.com/article/CA6640997.html?fiAnti-
_c_<_)_t_rnterfeit%2C+tamper+evident+fiIms.

Thompson, B.J. 1993. Tamper-evident, reclosable flexible packages. US:
Oscar Mayer Foods Corporation.

123

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

Colburn, P. 2003. Tamper-Evident Then'noformed Packaging. In
Pharmaceutical & Medical Packaging News.
http://www.devicelink.commrnpn/archive/03l09/005.html.

Evonik. 2008. Product Information: XT® polymer 375 acrylic-based
multipolymer compound. h_ttp://www.cvro.com/NR/rdonlvres/CAE9031E-
0630-4EBD-ADBO-E80C5B1 9A060/0l351 5XTjolvmer375techdata.pdf.

O'connor, MC. 2007. Packaging Maker Offering Tamper-Evident RFID
Film. In RFIDJouma/.com.
h_ttp:l/www.rfidiogrnal.com/articleLarticleview/2959/1/1/.

Anonymous. 2005. MIKOH's Smart&SecureTM RFID Tags Detect
Tampering When it Happens. In RFID Product News Update.
http://www.rfidprodgctnews.com/issues/2005.1 1/newprod/updates.php.

Reynolds, G. 2006. Banknote security transformed for food safety. In
FoodProductionDaily.com. h_ttp:l/www.foogproductiondaily.com/Supply-
Chain/Banknote-security-transformgl-for-food-safetv.

Bertrand, K. 2005. Making tampering evident. In FoodProcessing.com.
h_ttp://www.foogprocessing.com/articles/2005/550.html.

Bureau of International Information Programs. 2005. New Sensors Can
Reveal Package Tampering in Food, Medicines. In Americagov.
mp://www.america.gov/st/washfile-
Q9lish/2005/Febwary/20050202140627lcnirellepo.8408014.html.

Han, J.H. 2005. Innovations in food packaging. San Diego, Calif.: Elsevier
Academic.

Sira Technologies. 2010. Food Sentinel System.
http://cms.siratechnologies.com/index2.phg.

Mitsubishi Gas Chemical Company. 2007. AGELESS EYE Oxygen
Indicator h_ttp://www.mgc.co.ip/eng_/prodgcts/abc/ageless/eve.html.

Oxysense Inc. 2009. OxySense OxySentry System.
h_ttp://www.oxvsense.comlwel_>/file.axd?file=2009%2F4%2Foxvsentrvp<fl.

124

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

Almenar, E. 2009. Active and Intelligent Packaging. In PKG 891 Lecture
Material: Michigan State University.

Elamin, A. 2006. Company develops food safety indicators for packaging.
In FoodProductionDaily.com.
http://wwwfoodproductiongailvcom/PackaginglCompanv-develops-food-
gfetv-indicators-for-packaging.

Anonymous. 2002. Food Science Australia moves intelligent packaging
ahead. In FoodProductionDai/y. com.
mtg://www.foodproductiondgilv.com/Qualitv—ngetv/Foog-Science-
fistralig-moves—intellkgent—packaging-ahead

Dorneanu, L. 2007. Polymer Opal Films Identify Counterfeit Money and
Rotten Food. In Softpedia. h_ttp://news.softpegia.com/news/Polvmer-Ogl-
F ilms-Igentifv-go_unterfeit-Monev-afil-Rotten-F cog-61 126.8html.

Food and Drug Administration: Department of Health and Human
Services. 2009. Draft Guidance for Industry on “Incorporation of Physical-
Chemical Identifiers into Solid Oral Dosage Form Drug Products for
Anticounterfeiting". Federal Register

Anonymous. 2007. Counterfeit Drugs Can Kill - Nanotechnology
Fingerprints Can Certify Authenticity. In ScienceDaily.com.
h_ttg://www.sciencegailv.com/videos/2007l0209-

flinterfeit drugs can kill.htm.

Ulrich, C. 2006. Nano-Textiles Are Engineering a Safer World. Human
Ecology - Cornell University.

Merriam-Webster Online Dictionary. 2010. Focus Group.
mpzllwww.merriam-wepstercom/dictiongrv/focus+grom).

Supelco. 2003. Carboxen GC PLOT Capillary Columns. Product
Information.

Robinson, J.W. 2005. Undergraduate instrumental analysis. 6th ed. New
York: M. Dekker.

125

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

Trsl. 2007. Instrumental Chemistry.
mtg/lwww.instrgmenglchemistrvcom.

Damodaran, S., K.L. Parkin and OR. Fennema. 2008. Fennema's food
chemistry Food science and technology. 4th ed. Boca Raton: CRC
Press/Taylor & Francis.

Mcfadden, RD. 1982. Poison Deaths Bring U.S. Warning on Tylenol use.
The New York Times, October 2.

Anonymous. 1982. THE CITY: Girl swallows Pin in Cheese Snack. The
New York Times, November 9.-

Turner, W. 1986. Cyanide Death halts the sale of Excedrin capsules. The
New York Times, June 18.

Stehlin, D. 1992. Mother goes to jail for baby food tampering. FDA
Consumer: The magazine of the US. FDA 26 (7).

Anonymous. 1992. Poison in jar of baby food. Courier Mail, Perth,
Australia, May 5.

Anonymous. 1998. Turkish-born man gets five years in jail for attempted
extortion. Deutsche Presse - Agentur, December 22.

Anonymous. 2003. RCMP probe turkey-tampering claim. The Gazette,
Montreal, Quebec, December 26.

Zambito, T. 2008. Harlem man nabbed for poison rant on YouTube. Daily
News, New York, August 1.

Anonymous. 2009. Woman seen tampering with baby food denied bond:
WSVN.com. http:l/www.wsvn.com/newslarticles/local/Ml1 132_2_8/.

Anonymous. 2009. Police Warn Man Tampered With San Jose Stores'
Baby Food. KTVU.com, http://www.ktvu.com/news/20611866/detail.html
(accessed May 25, 2010).

126

will

36

Ll
Vfl

3 1293 03063 6

 

H
I
I"
II
I
|
.l H
u
|..
A”
H
H
H