: .
.
.. €1.23 :4.

.m.

h ‘ ..

a 2
v: _ RA...
3...... 1 v 1..»
1 :91... ..
A .x. N,

,6 .

2:3..k

Sean”.

aufl ..
(t.

i.
.95
J.

I
by"
. “Ht: . 1

a. .2.
g 'H.
amp?“ .
“$.01!
5 .3» .u Pu.
Latwwu
.Prwfi
V .1 ‘1’ .23,»- I;
. :b.:km.hu #4..
12?... .1.‘
E4! :2 5h 511.3"; 9
. .3 fix», .

K.
\

 

win.
Luna. v

(1.3:

 

. .. 1
mormrfld‘x
k.-

 

I
.I
II. \uKh!
>O.):9 t2 KL. iii-=3. .
~ 1 .: .55.: . V
20..“— IzvVZI 4i 2““: B. ‘f‘t
3.7-7.5}.3:
in .‘:\Y .1...
e x P 355.33..
tr $.19..I:r.,3~\n.:: 3:.
£4. 39.1.92) .313
II! n)...

p.33:

7..

a.“ z...)

“unwittml

#04

n 5
Lfll
5W 2.
a2
.
VI)”. 3.?!
#2.": .2:
.o-K5'IY’C. .0. 2
. ate: .. ss
Fig-7V lit) V
231..

0.1

‘7:
i! t.
4‘
Q:
t!
v

3:3? ,

.fr”!

7. ‘ 5

a .39”
. x

Ail} ‘Hm‘

3

iv: ,. I’lgl‘tf:tltvt5l.:
J 1x..l.;:|\5\cxl§!l.
u

3-.

1

{Gill 7:
.l)

 

 

r...l¢h.l

5"

{II (3.1?)

a. . 5......
Gil: l)... »
mnVl}:1)..l-EAAVL.7¢I£..
v31.;ls.$|e‘.l.lt:
. ysit.
, Cliovflit ‘
731-}: 4!. u 5\l\ 1.x
‘VAX ..
1.». ‘ nut. I
.. . . tn u. inn-UV. .
..2..$1. . 3!. .15.}
I311: .- author‘thf}. I.
.6» up L5. is ‘
C’uotlr
1. Q. .(Luuu
I ,
V v . ‘
.535935 r1333?

 

THrszs

2,001,

 

Michlgan State
University

 

 

 

This is to certify that the

dissertation entitled

TOWARD A PERFORMANCE STANDARD FOR
TYPEFACE LEGIBILITY: THE LOCKHART
LEGIBILITY INSTRUMENT

presented by

Laura L. Bix

has been accepted towards fulfillment
of the requirements for

PhD . Packaging
degree in '

 

 

 

 

Major professor

 

Date (pecan/m3) a? UV I

MS U Lt an Affirmative Action/Equal Opportunity Institution

*wd‘h~v—“v

.- 7 1.4 W“ —-‘u—-.' W" ,

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

 

N Whom

 

AUG 0 3 2qu

 

W

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c/CIFIC/DateOmpGS-nts

TOWARD A PERFORMANCE STANDARD FOR TYPEFACE LEGIBILITY:
THE LOCKHART LEGIBILITY INSTRUMENT

By

Laura L. Bix

A DISSERTATION

Submitted to
Michigan State University
In partial fillfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
School of Packaging

2001

 

ABSTRACT

TOWARD A PERFORMANCE STANDARD FOR TYPEFACE LEGIBILITY:
THE LOCKHART LEGIBILITY INSTRUMENT

By

Laura L. Bix

On March 17, 1999 the Food and Drug Administration (FDA) issued a regulation
mandating the design and font size of over-the-counter (OTC) drug labels. The FDA used
a prescriptive approach in an attempt to ensure label legibility. Research presented here
proposes that this not the best way to ensure the legibility of labels. A performance
standard, one that incorporated the Lockhart Legibility Instrument (LLI) for instance,
would be a better approach.

This research tests two hypotheses using the LLI: (l) noncompliant labels can be
created that have equal or greater legibility than labels that comply with FDA’s regulation;
and (2) labels that contain a familiar drug message will be easier for subjects to read than
those that are not familiar (a nonsense message).

Hypothesis 1, noncompliant labels can be created that have equal or greater
legibility than compliant labels, was found to be true at a statistical level of or=0.01. This
result is especially pertinent because one of the two compliant designs tested did not just
meet FDA’s 6 point minimum, it exceeded it. Noncompliant messages created using a
typeface that was 5.5 points were found to be more legible than the compliant designs,
which were created using a typeface that was 9.0 points in size. This difference
was highly significant, p= 5.0*10’7, vividly illustrating the flaws of the prescriptive

approach.

Hypothesis 2, labels that contain a familiar drug message will be easier for subjects
to read than those that are not familiar, was not found to be true at a statistical level of
(1:005. There are two possibilities with regard to this result. The first is that subjects
arrive at the same measurement on the instrument, regardless of their level of familiarity
with message; a desirable outcome for the LLI.

The second possibility is that subjects were no more familiar with this common
drug message (a decongestant) than the message that was created using random words.
This conclusion would support the idea that, despite potential dangers of OTC misuse,
consumers are not highly involved with OTC products (Reisenwitz and Wimbish, 1997;
Sansgiry and Cady, 1995; Robinson and Stewart, 1981); uninvolved consumers are less
likely to seek information, use complex rules when evaluating alternatives, and devote
focal attention and controlled comprehension to the product (Rifon, 2000). In other
words, the finding supports the idea that consumers are not reading OTC labels.

This is consistent with the findings of a survey conducted by Dr. Janet Engle,
Professor of Pharmacy at the University of Illinois, Chicago. At a news conference in
December 1998 she indicated, “47% failed to always read the product label before starting
a pain medication, and one-third were unaware that over-the-counter (OTC) drugs carry
risk” (Norton, 1999). Given the significant ramifications of improper OTC use, and the
difficulty in changing consumer/product involvement, this second possibility is a

fi'ightening, but real, risk.

Copyright by
LAURA L. BIX
2001.

Dedication

This dissertation is dedicated to my wingman. In the words of Bob Dylan, you
give me “shelter from the storm.” I love you.

Acknowledgements

I would like to take this opportunity to officially thank my major professor for
his dedication to both my research and professional development (past and fiiture).
Dr. Lockhart serves as an example both professionally and personally. He treats
students with respect and dignity. He is my role model for both the academic
community and the general community; the world is a better place as a result of his
presence.

I would also like to recognize Dr. Susan Selke. Dr. Selke balances her
research, grant writing, teaching, publishing and family life like no one that I have
ever seen. It is my hope that I can mnage my time as effectively and fi'uitfully as
this wonderfirl member of my committee.

Dr. Diana Twede has been another tremendous mentor to me while I have
attended MSU. Her financial assistance and interest in me personally made me feel
comfortable fi'om the beginning of my graduate career (long, long ago). She is not
only a mentor, but a fiend.

Dr. Nora Rifon has opened my eyes to research beyond the narrow scope of
legibility. She has helped me to see the cognitive and psychological aspects of
labeling, and has brought me to a place where I can see multiple avenues for future
research. Her enthusiasm and different viewpoints have opened doors that I never
even realized were there. A scholar approaches topics in ways that leap beyond
incremental improvements into innovation, and then conveys that information to

others in ways that energize them to act. She has done this for me.

vi

The final two members of my committee, Dr. Michael Fanizza and Dr. Joe
Kuszai, have also been tremendous assets to my worlc Like Dr. Rifon, they bring a
completely different perspective to the problem, enriching the solution. Their
impressive knowledge of the design of typeface, message layout and legibility are the
backbone of this entire document (from the formulation of the hypothesis, to the

design of the labels, to the results that we obtained). Thank you.

“A teacher affects eternity; [slhe can never tell where his [her] influence stops.”

Henry Books Adams

I would also like to thank a newly found resource, the Statistical Consulting
Center of the CANR Biometry Group. Specifically, I would like to thank my
consultant, Fernando Cardoso. Fernando worked exhaustively with me to ensure that
my statistics were solid, despite my fear and lack of ability with regard to statistical
methods. Fernando was patient, kind and went well beyond the call of duty.

Finally, but not least, I would like to acknowledge the MSU Writing Center at
Bessey Hall. My writing team, Rosamari Feliu-Baez, Catherine Fleck and Michelle
Ryan reviewed my writing every other Monday without fail (unless I failed to write).
It is because of these tremendous people that I have gotten this far this fast. Thank

you for your advice, your grammatical expertise and, most of all, your fiiendship.

“No matter what accomplishment you make, somebody helps you.”

Althea Gibson Darhen

vii

 

TABLE OF CONTENTS

LIST OF TABLES ......................................................................... viii
LIST OF FIGURES ........................................................................ ix
INTRODUCTION .......................................................................... 1
CHAPTER 1
LITERATURE REVIEW .................................................................. 5
Why examine OTC drug labels? ................................................... 5
Why has the aging of the population brought attention to OTC label
design? ....................................................................... 7
Financial Benefit ................................................... 7
Drug Mismanagement in the Elderly Population .............. 8
Changes in Vision .................................................. 9
Cognitive Changes ................................................ 11
Reductions in Working Memory ....................... 11
Difficulties with Language Comprehension .......... 13
Decreasing Prospective Memory ....................... 13
Problems with Symbol Comprehension ............... 15
Pharmacokinetic and Pharmacodynamic changes ............. 16
Social Changes ...................................................... 18
Design and Legibility ................................................................ 19
Letter Design and Legibility ...................................... 19
Message Design and Legibility ................................... 35
CHAPTER 2
MATERIALS AND APPARATUS ......................................................... 45
The Lockhart Legibility Instruments .............................................. 46
Legibility Cards (Preliminary Studies Only) ..................................... 52
Labels (Primary Study Only) ....................................................... 53
Dow Corning Ophthalrnics Card (All Studies) ................................... 56
AW Sperry Light Meter (All Studies) ............................................. 57
CHAPTER 3
PROCEDURES ............................................................................... 59
Subject Orientation .................................................................. 59
Collection of Subject Related Information ....................................... 60
Preliminary Study 1 .................................................................. 61
Subject Demographics: Preliminary Study 1 .................. 63
Preliminary Study 2 ................................................................. 63
Subject Demographics: Preliminary Study 2 .................. 65
Preliminary Study 3 ................................................................. 65

viii

 

Subject Demographics: Preliminary Study 3 .................. 67

Primary Study ........................................................................ 67
Subject Demographics: Primary Study ......................... 70
CHAPTER 4
RESULTS .................................................................................... 72
Preliminary Studies 1 and 2 ....................................................... 72
Preliminary Study 3 ................................................................ 75
Primary Study ....................................................................... 78
CHAPTER 5
CONCLUSIONS ............................................................................. 89
Suggestions for future research .................................................... 91
APENDICES
Appendix 1- Summary Tables of FDA’s Activity Regarding Labels ........ 94
Appendix 2- A Variety of Typefaces in the Same Type Style ................ 101
Appendix 3- Legibility and Color Contrast ...................................... 107
Appendix 4- Primary Study Labels ............................................... 126
Appendix 5- Examples of a Variety of Compliant Labels Designs
(Drug Messages) .......................................................... 133
Appendix 6- Examples of a Variety of Noncompliant Label Designs
(Drug Messages) .......................................................... 142
Appendix 7- Consent Form ...................................................... 153
Appendix 8- Data Recording Sheet: Primary Study ......................... 155
Appendix 9- Data Recording Sheets: Preliminary Studies 1 and 2 ......... 157
Appendix 10- Data Recording Sheet: Preliminary Study 3 ................. 162
Appendix 11- Tukey-Kramer Pair Wise Comparisons ...................... 165
BIBLIOGRAPHY ......................................................................... 168

LIST OF TABLES

1. Preliminary Study Label Messages ........................................... 53
2. Factors and Levels of Primary Study Labels ................................ 68

3. Treatment Combinations for Each Label and Average Legibility
Index for Each Treatment Combination ...................................... 79

4. Results of the Restricted Form of the Mixed Model Using

Sattherwaite’s Method ......................................................... 82
5. Results of the Second Model ................................................... 84
6. Tukey-Kramer Pair Wise Comparisons ...................................... 85
7. FDA OTC Legibility Publications ............................................. 95
8. Content Requirements for OTC Labels (Section201.66(c)) ............... 97

9. Format Requirements:201.66 (d) (for Presenting the Title, Headings,
Subheadings and Information Set Forth in 201 .66(c)1 through

201 .66(c)9) ........................................................................ 98
10. Legibility Messages Used in the Color Contrast Study ..................... 108
ll. Contrast Treatments .............................................................. 109
12. Treatment Combinations for the Color Contrast Study ..................... 109
13. Color Contrast Results: Examining Groups 1-3 for an Effect

of Operator ........................................................................ 112
14. Partial Analysis of the Complete Data Set ................................... 113
15. Analysis of the Complete Data Set (Simplified Model) .................... 114

16. Tukey-Kramer Pair Wise Comparisons (of contrast) within Age
Group For Significance ......................................................... 1 16

LIST OF FIGURES

l. The Structure of the Eye ........................................................ 10
2. Font, Typeface and Family ................................. h ..................... 2 1
3. Comparing the x-heights of Various Fonts ................................... 22
4. Comparing the x-heights of Various Sans Serif Fonts ...................... 23
5. Diagram of a Block of Type .................................................... 25
6. Varying Fonts of the Same Type Size Are Not Equal in Height ........... 26
7. Ascenders, Descenders and x-height ............................................ 27
8. Counter Forms of Various Typefaces .......................................... 28
9. Small Counter Forms Vs. Large Counter Forms .............................. 29
10. Letters with Varying Weights .................................................... 30
11. An Evolution of Contrast in Letter Weight ..................................... 31
12. The Serifs of a Variety of Typefaces ............................................ 32
13. The Evolution of Serif Style ...................................................... 33
14. Sans Serif in a Variety of Typefaces ............................................. 34
15. Keming (Negative Letter Spacing Between Specific Letter Pairs) .......... 36
16. A Comparison of Typefaces with Difi'erent Set Widths ...................... 38
17. Examining Differences in Leading ............................................... 39
18. Lockhart Legibility Instrument Easels .......................................... 47
19. Circularly Polarized Light Becomes Linearly Polarized Afier Being

Passed Through a Single Polarizing Filter ...................................... 48
20. Polarizer and Analyzer: Axes Crossed (0° of Rotation) ...................... 49
21. Polarizer and Analyzer: Axes Parallel (90° of Rotation) ..................... 49

22. The 1993 Lockhart Legibility Instrument, Built by James Pietrowski.

23. The 1999 Lockhart Legibility Instrument, Built by Sycamore Technical
Services ..............................................................................

24. Factors of Label Design .........................................................
25. COmparing the x-height of Gill Sans, Verdana and Lucida Fax Typefaces
26. Dow Corning Ophthalmics Near Point Visual Acuity Card ................

27. AW Sperry Light Meter and Sensor ............................................

28. Graphical Interpretation of the Procedure Used for Preliminary
Study 1 ...........................................................................

29. Graphical Interpretation of the Procedure Used for Preliminary
Study 2 ...........................................................................

30. Graphical Interpretation of the Procedure Used for Preliminary
Study 3 ...........................................................................

31. Graphical Interpretation of the Procedure Used for the Primary Study.
32. Coefficients of Variation for Preliminary Study 1 by Subject ............
33. Coeflicients of Variation for Preliminary Study 2 by Subject ............
34. Residual Analysis: Preliminary Study 1 Vs. Preliminary Study 2 .......
35. Coefficients of Variation: Preliminary Studies 1 and 2 by Message.
36. Coefficients of Variation for Preliminary Study 3 ........................

37. Average Legibility Indices for All Eight Labels ...........................
38. Box Plots of Several Treatment Combinations ............................

39. Residuals Vs. Subject .........................................................

40. Box Plots of the Factors Message and Design .............................

41. Average Legibility Index by Age Group (Color Contrast Study) ......

42. Average Legibility Index by Color Treatment ............................

xii

50

52

54

57

58

62

66

73

73

74

76

77

80

81

83

88

115

115

Introduction

A major function of packaging is communication. Graphics, labels, inserts, tags,
time and temperature indicators (T'I‘Is) and barcodes are just a few of the tools used to
communicate information about a packaged product. These tools serve a variety of
purposes, fi'om motivating sales to keeping consumers safe. For products like over-the-
counter (OTC) drugs, the information on the package, usually in the form of a printed
label, is essential for the safe and effective use of the product and is fi'equently the only
form of information used by the consumer (Tennesen, 1999; Sansgiry and Cady, 1996A;
Wogalter et al., 1996A; Braus, 1993; Discenza and Ferguson, 1992; Lumpkin et al.,
1990). If an OTC label fails to communicate vital information about the drug,
consequences can range in severity fiom the inefficacy of the product to the death of the
user.

Exacerbating the potential for label failure is the aging of the population
“Chronic diseases and disorder in old age are multiple and synergistic. They bring on
functional decline and require multiple medication - a recipe for noncompliance” (Fulrner
et al., 2000-2001). The elderly represent a “worst case” scenario with regard to OTC
products, and as the population ages, the potential for drug related complications
increases. It is predicted that 20% of the US population will be age 65 or older by 2030,
and that by 2050 people over 85 will constitute 5% of the population (Federal
Interagency Forum on Aging, 2000). Policy makers, researchers and manufacturers,
aware of the increasing potential for medication-related complications and the financial

clout of the up-and-coming baby boomers, have begun to investigate ways that label

informtion can be made accessible to elderly consumers. This effort will not only
benefit the elderly, but the general population as well.

When a label is effective, consumers successfully accomplish four steps of
consumer/label interaction (Rousseau et al, 1998). Labels must be (1) noticed, (2)
encoded [read], (3) comprehended and, finally, consumers must (4) comply with the
information given. Failures at any of these steps diminish the label’s ability to fulfill the
communication fimction. The consumer reading the label, the label itself and the
environment in which the interaction takes place determine the success or failure of each
of these steps.

The Food and Drug Administration (FDA) has concentrated much of its effort on
regulating Labels so that consumers can successfully accomplish step two of Rousseau’s
model (1998). They recognized that consumers had difficulty reading OTC drug labels in
1990, when The Pharmacist Planning Service of Sausalito, CA. and several citizens’
groups petitioned them in an attempt to get standards set for the size and style of text
used on OTC labels (Sansgiry et al., 1997; Wechsler, 1991). “That same year the
California legislature enacted a bill (AB. 2713) requiring manufacturers to assess and
improve label readability” (Sansgiry et al., 1997).

The Nonprescription Drug Manufacturers Association (NDMA) responded
swiftly to the legislative and regulatory action. In 1991 they convened a task force that
produced “The NDMA Label Readability Guidelines”. It contained a set of
recommendations that drug manufacturers could voluntarily adopt to improve the

legibility of their products’ labels.

One study of OTC drug labels on the market in 1996 indicated that manufacturers
were not adopting the voluntary guideline, and that there was still reason to be concerned
about encoding. “Study results indicate tlmt not all of the recommendations stated in the
NDMA Label Readability Guidelines have been adopted by manufacturers.” Perhaps
more alarming than the high rate of noncompliance with the voluntary standard was the
finding that “font size of warnings did not increase with an increase of package size, but
remained constant at a level that might be difficult for certain patients to read” (Sansgiry
et al., 1997).

FDA conducted research studies, sought comment, reviewed existing knowledge
and, ultimately, imposed a new regulation for OTC label design. In the Federal Register
published March 17, 1999, the FDA published a final rule entitled “Over-the-Counter
Human Drugs; Labeling Requirements” (see Appendix 1 for regulation details). This
rule established,

“a standardized format and standardized content requirements for
the labeling of OTC drug products. This final rule is intended to
assist consumers in reading and understanding OTC drug product
labeling so that consumers may use these products safely and
effectively. This final rule will require all OTC drug products to
carry the new, easy-to-read format and the revised content
requirements .. .” (Food and Drug Administration, 1999)

The standardized, prescriptive approach directed by FDA is problematic.
Previous research cautions against the use of standardized warnings because of a
phenomenon referred to as “habituation”.

“Over time and repeated exposure, a warning will attract less
attention. . .There are many ways to retard habituation, however.
One way is to alter the characteristics of an existing warning fiom

time to time so that it looks different”(Wogalter and Laughery,
1996B)

Altering the appearance of OTC warnings is not possible under the current rule; label
format is explicitly prescribed by the regulation.

Prescribing the details of OTC label design may not be the most effective
alternative for ensuring accessibility to the information they provide. FDA’s approach
does not directly address consumers’ abilities to read informtion, but assumes a legible
solution has been found and sweepingly applies that solution to all labels (Bix and
Lockhart, 2000). This approach oversirnplifies the complex interactions that can occur
when all the elements of design come together to convey a message.

The research presented in this document challenges the idea that a prescriptive
approach to regulation addresses the true issue of importance: the consumer’s ability to
visually read the label. The Lockhart Legibility Instrument (LLI), an instrument
developed at the MSU School of Packaging, provides a direct measure of a subject’s
ability to read text. The numerous elements of letter, word and message design, and the
complicated interactions that can occur between these elements are evaluated for
legibility with a single test. A performance standard for legibility, utilizing an instrument
like the LLI, would better serve consumers by ensuring that they could effectively read
essential information provided on drug labels, helping to ensure their safe and effective
use.

The fundamental hypothesis of this research is that by manipulating the rmny
components of letter, word and message design, a noncompliant label can be created that
is at least as legible as one tlmt complies with the 1999 regulation, as measured with the
LLI. The ability of people to effectively read OTC label information should be the focus

of FDA’s regulation, not compliance with a prescribed standard.

Literature Review
Why Examine OTC drug labels?

Although creating labels that can be successfully read by consumers will benefit
all products, effective OTC label design is paramount for a variety of reasons.

When prescription drugs are sold to consumers, it is typical that two sources will
act as “learned intermediaries” to inform them of possible adverse consequences, or risks,
associated with product use (Alsobrook, 1992). The physician who prescribes the drug
has the opportunity to provide consumers with information, as does the pharmacist who
sells the drug.

“. . .consumers generally cannot hold manufacturers of prescription drugs

liable provided the manufacturer has adequately warned the prescribing
physician of pertinent side effects and manufactured the drug properly”

(Alsobrook, 1992).

Guidance provided by the physician and the pharmacist augments the printed label
information that appears on prescription drug packaging, and “buffers” prescription drug
manufacturers fi'om liability (Alsobrook, 1992).

This is not the case with OTC drug products. Although consumers can seek
information fiom other sources, in the majority of cases, the label is the sole provider
(Wogalter et al., 1996A). In fact, the number of consumers that actually seek advice
fiom a health-care professional while selecting OTC medications is very low (Tennesen,
1999; Sansgiry and Cady, 1996; Braus, 1993; Lumpkin et al., 1990). The courts have
taken notice of the importance of the need for sufficient OTC labeling, and it is reflected

in the jurisprudence. “Although many theories are available to an injured consumer of an

OTC, proof of inadequate warning is most effective in obtaining relief according to the
restatement and jurisprudence” (Alsobrook, 1992).

The potential for manufacturer liability is increasing as the use of OTC
medication continues to rise.

“Switching drugs from prescription to over-the-counter (OTC) status is

proceeding at an ever-increasing pace. In the 10 years fiom 1984 to 1994,

9 drugs were switched from prescription to OTC status - an average of

about 1 a year. But in 1995 alone, 7 agents were moved to OTC status,

and last year (1996), 13 drugs were switched to OTC” (Marwick, 1997).
It is no wonder that the popularity of OTC drugs is increasing: self-medicating offers an
average savings of $84 per illness (Tarlach, 1998) when compared with the average cost
of a trip to the doctor and the purchase of a prescription medication. “Although 60% of
the drugs purchased by consumers in the United States are OTC, they account for less
tlmn 2 % of the US health care dollar” (Marwick, 1997; People’s Medical Society, 1997).
In addition to saving money, self-medicating is convenient. Over-the-Counter (OTC)
drugs can be purclmsed at almost any grocery store, rmking them attractive to consumers
who are increasingly pushed for time.

Although self-medication offers freedom, flexibility and cost savings to
consumers, it has risks.

“Researchers and the trade press have reported cases of inappropriate

consumer use of nonprescription medicines (OTCs), primarily involving

misinterpretation and misuse. Some consumers are unable to interpret

label information correctly and others delay medical treatment for a more

serious underlying disease, overdose, or use nonprescription drugs

chronically. . .Non prescription drugs also are known to interfere with

laboratory test values, cause adverse interactions with prescription drugs,

and sometimes even render prescription drugs ineffective” (Gore et al.,
1 994).

Further increasing consumer risk is the fact that stronger and stronger drugs are being
switched fiom prescription status to OTC. The Consumer Healthcare Products
Association, a trade group that represents nonprescription drug makers, indicates, "more
than 600 OTC drugs contain ingredients and dosages that 20 years ago were available
only by prescription" (N ordenberg, 1999).

“While information alone will not ensure a secure patient, the corollary is more
difficult to refirte. . .. It is essential that individuals understand the nature of OTC
medications, the consequences and benefits of compliance” (Lumpkin et al., 1990). This
is especially true for the elderly population, who are at increased risk for adverse
reactions fiom OTC drugs for various reasons.

Why has the Aging of the Population Brought Attention to Label Design?

Financial Benefit
Designing more accessible OTC labels as the population ages is not just “the right

thing to do”; it makes good business sense. Recent American Association of Retired
Persons (AARP) research indicates that 8 in 10 baby boomers expect to work at least part
time during their retirement (Hignite, 2000). Couple this with the recently passed
“Senior Citizen's Freedom to Work Act of 2000” and the result is a substantial segment
of the population with increasing financial clout.

Not only do older adults work longer and keep more of the money they earn, they
spend more on drugs then other segments of the population. "People 65 and older are
more likely to require multiple medications, both prescription and OTC. Consumers over
age 50 used an estimated $41 billion in prescription drugs last year (1998). . .twice the
per-capita consumption of the rest of the population” (Woolley, 1999), and elderly

consumers spend even more on OTC drugs. A study conducted at the University of

Missouri Columbia’s School of Nursing (Conn, 1992) indicated “elderly subjects
reported using almost twice as many OTC as prescription medications.”

Manufacturers are beginning to take notice of this financially attractive population
by creating products and packages that target their needs and desires. John Bitner (2000),
Packaging Manager for Pharmacia Corporation, explains, “Older Americans are
becoming less tolerant of inconvenience and having their needs generally ignored... in
not too many years, this population segment will have the leverage to demand
performance.”

Drug Mismanagement in the Elderly Population
Even more compelling than the financial benefits, are the safety issues. The most

readily available, economically attractive source of information about a product is on its
label. Accessible information is important for all products, but can be of life and death
importance for products like OTC drugs. A number of factors place the elderly
population at particular risk for misuse.

“People over age 65 are especially vulnerable to medication-

related problems because of the number of medications that they
take and the biological changes of aging and disease. Older people
are the greatest consumers of prescription and OTC medications,
and they are more likely to be taking multiple medications at the
same time. . .Changes in physiology place older people at greater
risk of adverse reactions. .. Visual or cognitive impairment can
combine with psychological, social and care giving needs to
interfere with the proper use of medications” (Beers, 2000-2001).

The statistics of misuse confirm the risk to the elderly population. “Older people
are the group most likely to take medicine improperly, according to a study reported in
the New England Journal of Medicine” (Braus, 1993). “Americans aged 60 and older
account for 40% of adverse drug reaction cases” (Braus, 1993) an “. . . ..51% of deaths

from drug reactions” (Holt and Hall, 1990).

Older adults are less able to effectively navigate the four steps of consumer/label
interaction (notice, encode, comprehend and comply) (Rousseau, 1998) than their
younger counterparts because of changes in their perceptual and cognitive abilities.
Combine this with changes in the psychological, physiological and social states of the
elderly and you begin to see why they account for large percentages of drug related
hospitalizations and deaths.

Changes in Vision
To effectively notice and encode (read) printed label information, consumers must

physically perceive the text using their vision. As people age, ocular declines interfere
with their ability to successfully complete the first two steps of Rosseau’s (1998) model,
putting them at increased risk for improper use. Age-related changes in the pupil, cornea,
lens, vitreous humor and retina (see Figure 1) combine with an increased propensity for
ocular disease to diminish visual function and fuel the need for additional light.

The sclera and the cornea form the outermost portion of the eye (see Figure l).
The sclera is a fibrous coating that protects and shapes the eye, providing an anchor for
muscles. The cornea bulges from the sclera and is transparent, allowing light to enter the
eye (Marieb, 1992). As light enters the eye, it is refi'acted by the cornea, which provides
approximately 75% of the eye’s focusing power (Kelly, 1993). It then passes through the
aqueous humor to the lens, where it is refocused and passed through the vitreous humor,

3 clear fluid, to the retina (Marieb, 1992).

 

 

 

 

 

 

 

Iris

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cornea

 

 

 

 

Aqueous
Humor

 

 

 

 

 

 

 

 

 

 

 

 

Nerve

 

 

Vitreous
Humor

 

 

 

 

 

 

 

FIGURE 1- THE STRUCTURE OF THE EYE

One of the cornea’s most significant age-related changes is a “flattening” that
results in astigmatism. Degenerative changes also occur in the sclera, which result in a
loss of clarity that affects the quality of vision of older adults. These changes result in a
loss of accommodation, the ability to focus at a variety of distances, in aging adults.

Aging also affects the retina, the vitreous humor and the lens of the eye. As aging
takes place, the retina, the vitreous humor and the lens yellow and thicken. These
changes reduce the short-wave lengths of light (the blue color) entering the eye and

scatter the light that does enter. The resultant efl‘ect is that colors such as violet, blue and

10

green are filtered out, reducing the contrast sensitivity of the eye and causing glare which
interferes with the visual image.

Visual quality of older adults is also affected by changes in the pupil. “The iris is
unable to dilate as much as in youth under all light conditions, but this is especially
evident and troublesome under dark-adapted conditions” (Kelly, 1993). As a result of the
iris’s inability to dilate, the size of the pupil decreases with age. This creates a kind of
chain reaction. The iris is unable to dilate, keeping the pupil small and, ultimately,
decreasing the amount of light reaching the retina, which, as a result of age-related
thickening, is not as sensitive as in younger subjects.

Not only do older adults have to contend with changes that result in diminished
visual function, they are also more likely to develop certain ocular pathologies.
Cataracts, glauconm, senile macular degeneration (SMD), diabetic retinopathy, and
presbyopia, a decrease in the ability to focus, are diseases that are more prevalent in the
elderly (Kelly, 1993; Holt et al., 1990).

Cognitive Changes
In addition to perceptual losses, aging consumers frequently experience several

cognitive changes that can affect their ability to successfully use labels. Unlike
perceptual changes, cognitive changes generally impact the consumer’s ability to
comprehend and comply with label information, the final two steps of Rousseau’s (1998)
consumer/label interaction. Cognitive changes include decreases in working memory,
language comprehension, prospective memory, and the comprehension of symbols.

Reductions in Working Memory
One age related phenomenon that affects the ability of consumers to comprehend

and comply with label information is a reduction of working memory. Working memory

11

refers to a person’s capacity to store and process information (Rousseau et al., 1998;
Morrow and Leirer, 1999). "Memory load" is the amount of information that individuals
have to process. Higher "memory loads" exacerbate problems associated with decreases
in working memory. Older adults experience a reduction in working memory, yet have
higher memory loads related to medication instructions due to their propensity to use
multiple rmdications (Morrow and Leirer, 1999; Woolley, 1999; Sansgiry and Cady,
1996).

Additionally, “older adults may have trouble understanding instructions that
require inferring or reorganizing information, which impose heavy demands on working
memory” (Morrow and Leirer, 1999). Despite this knowledge, many warning labels still
use instruction statements that require consumers to make inferences and reorganize the
information presented. One example is a label that requires consumers to integrate time
(“take twice daily”) and dosage information (“take two pills”) (Morrow and Leirer,
1999). Clear dosage clmrts complete with prescribed amount and explicit time of day
reduce the requirement for working memory.

Explicit directions that do not require inferences will do no good if consumers do
not remember to read them. “Older adults benefit fi'om environments or contexts that
provide meaningful cues for remembering information” (Rousseau et al., 1998). One
way to apply this to OTC drug labels is to place the label information on the primary
package, the package that contains the product, rather than on the secondary package,
which is typically an outer wrap or folding carton. When packages are designed in this
fashion, the relevant information is part of the immediate environment, available to the

older consumer as they are using the product.

12

Difliculties with Language Comprehension
Reduction of working memory also creates problems with language

comprehension. In 1978 Kintsch and van Dijk proposed one of the most prominent
language comprehension models for text processing:
“According to their [Kintsch and van Dijk] model, text is read and
brought into memory as propositions, which are information
packets containing a predicate and an argument (e.g., a verb and a
noun). The maintenance of these propositions in working memory
depends upon their relevance to the text. Important (high
involvement) information is maintained and propositions of less
importance (low involvement) fade fi‘om memory” (Rousseau et
al., 1998).
Light (1990) proposed that several age-related effects impact the Kintsch and van Dijk
model. Light suggested that the decreased working memory capacity of older adults
resulted in faster decay of propositions fiom memory. He also hypothesized that older
adults have more difficulty associating propositions that occur farther apart, and that
older adults have a harder time “keeping extraneous, irrelevant information out of the
working memory, further reducing its capacity” (Light, 1990). The implications of
decreased language comprehension for the labels of OTC drug products are fairly clear.
Label information should be simple, direct and explicit.

Decreasing Prospective Memory
Another problem faced by elderly consumers is a decrease in ‘prospective

memory.” According to cognitive psychologists, “prospective memory” entails
“remembering to perform an action in the firture” (Rousseau et al, 1998). Prospective
memory is related to the task prescribed by the label. Rousseau indicates that elderly
consumers have less difficulty remembering event-based tasks. “Take two pills after

breakfast” is an example of an event-based task. Elderly consumers have been shown to

13

have a harder time complying with time-based tasks, such as “take two pills every eight
hours” due to losses in prospective memory (Rousseau et al., 1998).

The introduction of environmental cues alleviates the age-related decline in
prospective memory. “Interactive warnings” are a relatively new type of design that is
meant to take advantage of environmental cues, increasing label effectiveness.

“This format (interactive warnings) requires the product user to
physically manipulate the warning when using the product, and
researchers have found that these types of warnings increase the
likelihood of the user noticing and complying with the
information. . . the interactive labels serve as an event-based cue to
recall the appropriate safety procedures” (Rousseau et al., 1998).
One field where interactive labels have proven successful is in the firmiture industry.
Directions for assembly are placed in such a way that the consumer must physically
destroy the label to accomplish the directed task.

Few studies have applied the use of interactive labels to packaging. Wogalter et
a1. (1996A) investigated the effect of an additional label on understanding of elderly
consumers by adding a printed “easy open fin” to the caps of OTC drug bottles that
contained motion sickness tablets. “The information on the cap label (the fin) was
extracted fi'om the main label text and was chosen, based on consultation with a
pharmacist, to reflect the most important cautions and directions for proper, safe use of
the product.” Consistent with Rousseau’s findings, Wogalter’s research team found that
this simple change in package design and label format improved knowledge acquisition.
Interactive warnings and their application to OTC drug packages is a rich area for firture

research.

14

Problems with Symbol Comprehension
A final type of cognitive deterioration that occurs during the course of aging is a

decrease in symbol comprehension.

“Age-related changes in text comprehension would seem to

indicate that symbols would have particular utility for improving

older adults’ understanding of warning information. However,

symbols may not always facilitate comprehension for everyone,

and this is particularly the case for older adults. In addition, some

working memory limitations that are known to affect language

comprehension may also affect symbol comprehension”(Rousseau

et al, 1998)
A study conducted by Morrell et al. (1990) examined the effect of age on the ability to
comprehend text-only labels and labels that contained a combination of text and symbols.
While younger adults benefited fiom the combined format, older adults performed best
when text-only formats were supplied. Researchers did not suggest a reason for this
effect.

Sansgiry and Cady (1996) also evaluated the efl‘ect of age and pictorial
inforrmtion (symbols) on the comprehension of OTC drug product labels. A study they
conducted in 1996 was divided into two parts.

In the first part subjects were asked to interpret eight pictures (symbols), six of
which were taken fiom actual OTC labels. Subjects were allowed as much time as they
needed to view labels and make interpretations. Numbers of responses and percentage of
correct responses were tabulated. Results of the 1996 study supported Rousseau’s
position that the comprehension of symbols decreases with increasing age. “Older adults
had greater difficulty in interpreting correct responses” (Sansgiry and Cady, 1996); the
elderly group’s score was below that of the young adults for all eight pictures. At least

80% of the younger group paired symbol with affliction correctly for five pictures. By

15

contrast, the older group had 80% of its participants make the correct pairing for only 2
of the eight pictures.

The second part of the experiment dealt with the effect of pictorial information on
subject understanding. Four OTC products intended for medicating headaches, sinus
conditions, menstrual pain and sleeplessness were investigated using four treatments.
The first treatment was a picture only label. The second treatment was a text only
presentation. The third and fourth treatments were a combination of text and pictorial
information; one depicted an incongruent picture/text combination and the other
displayed a congruent message. It was hypothesized,

"Congruent picture-verbal label design (picture rmtched with
written information) will enhance understanding of information in
both younger and older adults, compared to verbal only, picture
only, and incongruent picture-verbal (picture and written
information does not match and the written information represents
a distracter) label designs" (Sangsgiry and Cady, 1996).

Sansgiry and Cady found the congruent picture/text combination to be the most
easily interpreted, followed by text-only designs. However, when the results of these two
designs were compared, the difference was found to be insignificant. This result is
consistent with those of the study conducted by Morrell et a1. (1990). “Addition of a
picture complementing written information did not enhance understanding significantly”
(Sansgiry and Cady 1996). When the results were compared between age groups,
significant differences occurred in all but one label design at p<0.001. Older adults had
lower understanding scores for all label designs when compared with the younger group,

suggesting that symbol comprehension does decrease with age.

16

Pharmacokinetic and Pharmodynamic Changes
Not only do perceptual and cognitive changes make consumers less likely to

effectively use drug labels, physiological changes negatively impact how the body
processes the drug, and what the drug does to the body. All of these changes combine to
put elderly consumers at even more risk from the ill effects of mismedication.

Pharmacokinetics refers to how the body processes medication. According to
Beers (2000-2001) there are three components of pharmacokinetics. The first component
is absorption, how the body gets the drug. The second component is distribution, “where
and how the medication goes once it is in the body” (Beers, 2000-2001). The last
component, elimination, refers to the system’s ability to clear the drug.

Although aging significantly impacts how the body processes medications, the
first component, absorption, is affected very little by age. Distribution, the second
component of pharmacokinetics, is substantially impacted as people age.

“Most medications are distributed to either body fit or body water. With

aging, there is an increase in the percentage of body fat. The typical older

person has about 25 to 30 percent more fit than the typical younger

person; the percentage is higher for older women. With the increase in

percentage of body fit, there is a corresponding decrease in the percentage

of the body consisting of water. In older people, blood levels of water-

soluble medications will be higher than would be expected in younger

people because there is less body water to distribute into. Fat-soluble

medications stay in the body much longer because there is more fit in

which to be stored” (Beers, 2000-2001).

The final component, elimination, is even more dramatically impacted by age.

Drugs are eliminated from the body either through the kidneys or metabolized in
the liver. “There is about a 50 percent decline in the renal (kidney) clearance of
medications by the time people reach age 75 to 80”. Drugs that are metabolized

by the liver are also affected. Beers (2000-2001) indicates that hepatic (liver)

17

blood flow in a person between the ages of 75-80 is about halfthat of younger
adults, which substantially impacts drug clearance. Additionally, the P450 system,
the major enzymatic system by which the liver metabolizes medication, is rapidly
saturated in old age.

Pharmacodynamics, the way that drugs affect the body, also changes with
age, firrther compounding the problems that the elderly have with drug products.
“As we get older we are more sensitive, rather than less sensitive to most drugs”
(Beers, 2000-2001). Age-related changes in pharrnacodynamics result in
increased potential for toxicity. This is particularly true of drugs that act on the
central nervous system; as a result, the “list of medications that cause confirsion or
cinnges to the central nervous system in older people is very long” (Beers, 2000-
2001).

Social Changes
The elderly not only have to contend with perceptual, cognitive and physiological

changes that put them at increased risk for nrismedication, but many of them live in social
conditions that exacerbate the risk further. The number of elderly people that live alone
is on the rise. In 1989 the US. Senate Special Committee on Aging estimated that 47%
of people 85 and older lived independently. According to the US. Senate Special
Committee on Aging, they are one of the most vulnerable groups in America. 60% of the
elderly living alone are reported to have chronic problems with vision or hearing (U .8.
Senate Special Committee on Aging, 1991). As a result, a significant portion of older
adults have no one to assist them with their medical regimen, are suffering fiom
afflictions that require polypharmacy, and do not have sufficient vision to read printed

information.

18

 

Design and Legibility

The aging of the population, combined with the risks of self-medicating, its
increasing prevalence, and the trend to switch drugs from prescription to OTC status have
all contributed to the need for printed OTC labels that are easily read and understood.
Recognizing these trends, FDA published a regulation entitled “Over-the-Counter Human
Drugs; Labeling Requirements” (1999). The regulation is an attempt to ensure that OTC
labels are sufficiently legible. It dictates, among other things, a minimum font size,
sufficient contrast, minimum leading, a maximum number of letters per inch, and
strongly recommends that manufacturers use a specific style of type.

FDA’s approach looks at the elements of design as distinct entities that can be
isolated and manipulated one at a time to improve label legibility. Reality does not match
the simplicity of this approach. Legibility is the overall goal in a complex system of
interrelated elements (letter weight, letter compression, counter form shape, stress, type
Style, type size, message layout, leading, keming, ink, substrate, and printing process)
that come together to create a message. Researchers who recognize the complexity of the
elements of typography understand that changes in design do not occur in a vacuum. In
Changing one element, it is likely that numerous others, all of which impact legibility,

Will be affected (e.g. increasing stroke weight will decrease counter form size).

Letter Design and Legibility
When creating messages, designers must be careful to not affect basic letters, thus

weakening communication (Craig, 1980). The challenge is to make the most effective
use Of the enormous flexibility that is inherent in typographic design (Bigelow and Day,
1983) by creating designs tlfit are both interesting and practical. Effective designers

develop a high level of awareness of font in order to construct messages that not only

19

attract readers (notice), but allow them to easily read (encode) and understand

(comprehend) the message they have created. This awareness begins with a basic

understanding of the “anatomy” of letters.
Textual messages are usually constructed of words that are made of upper and

lower case letters that are set in a single font. “A font consists of all the characters (upper
and lowercase, figures, fiactions, reference marks, etc.) of one size of one particular
typeface” (Craig, 1980). Typefice (see Figure 2) is defined as the full range (size) of
type of the same design (Department of Mathematics, University of Utah, 2001). In other
words, a typeface consists of all characters, in all sizes, of a particular design.
“Typefaces are usually available in 6- to 72-point [one point is equal to 1/72”], with a

complete font in each size” (International Paper, 1997). A family of type encompasses

all related typefaces (see Figure 2).

20

 

 

See Spot run. (Ariel Alternative 12 point)

 

A font consists of all of the characters used for normal composition in a
SINGLE SIZE of a specific design.

 

 

 

 

See Spot run.
(Arial Alternative 12 point)

See Spot run.
(Arial Alternative 20 point)

See Spot run.
(Arial Alternative 23 pt)

A typeface consists of all of fonts of the same design (in this case Arial) in
different sizes.

 

 

 

 

 

 

See Spot run.
(Ariel Alternative 12 points)

See Spot run.
(Arial Alternative 20 pts)

See Spot run.
(Arial Alternative 25

points)

80. Spot run. (Arlal Black 10 polnts)

See Spot run. (Arial Black 18 point)
See Spot run. (Arial Blck 23 pt)

See Spot run. (An'al Narrow 15 points)
See Spot run. (An'al Narrow 22 points)

See Spot run. (Arial Narrow 25 pt)

A fimily consists of all typefaces of the related designs. In this case we are
looking at a few examples of typefaces that belong to the Aria] family. This by
no means represents the entire Arial family.

 

 

 

 

 

FIGURE 2- FONT, TYPEFACE AND FAMILY

21

 

There are several common elements of letters that can be examined. These
include x-height, ascenders and descenders, cormters (also called counter forrm) serifs (or
lack of serifs, referred to as sans serif), and stroke weight (thick and thin). The terms x-
height, ascender and descender refer only to lower-case letters, while counters, serifs, and
stroke weight apply to both upper and lower case letters.

X-height refers to the height of the body of a lowercase letter. It is called the x-
height because it is equal to the height of the lowercase x (see Figures 3 and 4).

“Although the x-height is not a unit of measurement, it is significant

because it is the x-height - not the point size - that conveys the visual

impression of the size of the letter. Typefaces of the same point size may

appear larger or smaller because of variations in the x—height” (see
Appendix 2 and Figure 3) (Craig, 1980).

   

. II. .1! OX. .1! III .11 Ill-LC in

49 Point 49 Point 49 Point 49 Point 49 Point
Garamond Baskerville Modern 20 Century Helvetica Light

   

 

 

 

 

 

 

 

 

FIGURE 3- COMPARING THE X-HEIGHTS OF VARIOUS FONTS

Despite the fact that it is the x-height that conveys the visual impression of letter
size, not the point size, the FDA regulation mandates a minimum type size of 6 points,
and makes no direct reference to x-height. FDA does, however, strongly recommend the
use of a sans serif font. These fonts are characterized by larger x-heights, indirectly
addressing the issue. FDA suggests that adhering to their standard will produce legible
results, “the agency [FDA] believes it has selected type sizes [6 point minimum] and

styles [sans serif] that are consistent with the need for readable OTC drug product

22

labeling by a majority of OTC drug consumers” (FDA, 1999). Despite the urging of
FDA, companies may dismiss the recommended typefices in favor of others; it is not
mandated that they comply. Even if manuficturers strictly adhere to the recommendation
and use sans serif type styles, the x-heights from family to family vary dramatically (see
Figure 4). The result is that 6 point fonts with varying x-heights (see Figures 3 and 4),

and presumably varying legibility, are allowed by the regulation.

 

 

      

It III IL! 010 DA! iii;

 

 

 

 

 

 

Gill Sans (a sans Helvetica (a sans Verdana (a sans
serif type) serif type) serif type)
79 point sample 79 point sample 79 point sample

 

 

 

 

 

 

 

FIGURE 4- COMPARING THE X HEIGHTS OF VARIOUS SANS SERIF FONTS

Type size is, perhaps, the letter clmracteristic that is most frequently manipulated
to improve legibility; a common perception is that increasing type size will automatically
improve message legibility. To some extent, this is true. However, to say that type size
determines legibility is an oversirnplification. The design elements of letters, and the
waytheyarepresented, canhave agreaterimmct on legibilitythan sizeofthe type.

A study conducted at the New England College of Optometry (Watanabe, 1994)

found elements other than type size had a more significant impact on legibility. “Type

23

 

size alone may not be responsible for poor readability. Other factors that may be
contributing to this difficulty include letter and line spacing, letter contrast, print and
background color, and type style” (Watanabe, 1994). The study concluded, “horizontal
letter compression had a greater effect on readability than vertical letter height.”

An experiment conducted at the Michigan State University School of Packaging
in July 1997 (Lockhart and Bix, 1997) also suggests that more fictors influence legibility
than font size. A message in 4.5 point type with black on white contrast was more easily
read than the same message printed in 6 point type with yellow on red contrast. These
results indicate that color contrast can have a greater impact on legibility than type size.

Regulating the 6 point minimum is also problematic because measuring type size
is not straightforward. The size of a given font is based on the now-antiquated system of
setting metal type. Metal type setting was the system used when letterpress, a type of
relief printing, was the only way to print text. In letterpress printing, each letter is raised
from the surfice of a metal block (see Figure 5). The block is referred to as the body; the
printing surface (the letter) is referred to as “the fice” (Craig, 1980). Type size is based
on the size of the block from which the letter is carved and is not directly related to the
height of the letter.

The difficulty occurs because different typefaces utilize different areas of the
block, and even though type is now created using computer programs, type size is still
based on the letterpress system. As a result, a type size of 6 points does not equal a 6
point letter height. Typefaces, like Verdana (see Figure 4 and 6 and Appendix 2), that
utilize a large percentage of the block are close to 6 points tall. Typefaces that do not use

as much of the block are much shorter, but they are still referred to as 6 point type.

24

 

 

 

 

 

Point Size
(Type Size) or
Block Size

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FIGURE 5- DIAGRAM OF A BLOCK OF TYPE

As a result, “the fice of any letter is not the hill point size. Corresponding letters in the
same size type may vary in height” (see Figure 6 for a limited sample and Appendix 2 for
a more extensive example) (International Paper, 1997). “No type face fills the amount of
space allowed in its measure, e.g. a type face in 10 point may print a letter only 6 points

high; another type fice in 10 point will print a letter 8 points high” (Ralph, 1982).

25

 

X X X X X

 

 

 

100 point
Verdana Capital
X

 

100 point Helvetica
Light Capital X

 

100 point Times
Capital X

 

100 point Play
Bill Capital X

 

100 point Tw Cen
MT Condensed
Capital X

 

FIGURE 6-VARYING FONTS OF THE SAME TYPE SIZE ARE NOT EQUAL IN

HEIGHT

Occasionally, individuals will measure type size from the bottom of the descender

to the top of the ascender or cap-line (see Figure 7). Ascenders refer to any portion of the

letter extending above the x-height; a lower-case “h” contains an ascender (see Figure 7).

Letters with descenders contain portions that fall below the x—height. A lower-case “p”

contains a descender (see Figure 7). Ascenders and descenders are only found in lower

case letters. International Paper’s Pocket Pal: A Graphic Arts Producti_on Handbook

advises againstmeasuringtype inthisfishion. “Type size cannotbemeasured fi'omthe

top of an ascending letter to the bottom of a descending letter. The face of any letter is

not the full point size” (International Paper, 1997).

26

 

 

 

 

Counter Forms

I \

 

 

 

Capline

  

 

 

Ascender

61 Way/5 III iI liI: _

, Descender
FIGURE 7- ASCENDERS, DESCENDERS AND X—HEIGHT

 

 

 

 

 

 

Even if FDA does find a way to accurately determine the point size of letters on
OTC labels, scientists and manuficturers continue to debate what constitutes thg
minimum legible type size. Many authors indicate that for type to be legible it should be
well above the 6-point size decreed by FDA. Hauptrmn (1979) recommends a minimum
of 7 points, while Jewler (1981) suggests sizes no smaller than 10 points. If visually
limited persons are considered (as previously noted, 60% of the elderly living alone are
reported to have chronic problems with vision [US Senate Special Committee on Aging,
1991]), it is suggested that a minimum of 12 points be used (Ralph, 1982).

Others argue that smaller type sizes can be used to create sufficiently legible
messages (CHPA, 1999; NDMA, 1991). They indicate that 6 points is too large for small
medication bottles, and not feasible fiom a production standpoint (CHPA, 1999; NDMA,
1991). In a statement to FDA, R. William Soller of the Consumer Healthcare Products
Association (CI-IPA), argues against the 6 point minimum,

“. . ...[there is a] notable kick of convincing objective support that 6-point

type adds an advantage in legibility over smaller types sizes down to and

including 5.0 point type... These are important observations, and it is

important to find ways to address them. They are important because they

suggest a ripple in what has been up to the Final Rule a fiirly reasonable
and productive partnership on labeling” (CHPA, 1999).

27

This is where the prescriptive approach that is being used by FDA fails.
Specifying a minimum type size does not ensure legible labels. Wide variance in the
heights of different type styles is likely to yield designs with varying degrees of legibility.
Even if with the use of a 6-point minimum guaranteed legibility, it is difficult to
determine the type size from the height of printed letters. This reality makes regulating
the current mandate a formidable task. These issues provide finther evidence that a
performance standard for the legibility of OTC products is in the best interest of the
public and the FDA.

Other design elements that impact legibility include counters (counter forms), the
presence or absence of serifs, and variations in thickness. These elements all apply to
both upper and lower case letters. Although most readers do not have a conscious
awareness of the negative spaces within letters, also called counter forms or counters (see
Figures 7 and 8), the design of these spaces significantly impacts letter identification.
Both the negative and positive spaces of each letter work in concert to allow viewers to

identify letters at a glance.

L 9112

Helvetica Light Century Modern 20 Baskerville Garamond
ll II H I! “X" II I! u ll

 

 

 

 

 

 

 

 

 

FIGURE 8—COUNTER FORMS OF VARIOUS TYPE FACES
(Negative spaces are in black while strokes are in white)

28

AA . . ..

 

 

 

 

 

 

 

 

 

Helvetica Helvetica Helvetica Helvetica Helvetica
Condensed Condensed Condensed Condensed Condensed
Black Black Black 25 Black Black
200 Points 100 Points Points 10 Points 6 & 4.5 Point
A A A A
Helvetica Helvetica Helvetica Helvetica Helvetica
Light Light Light Light Light
200 Points 100 Points 25 Points 10 Points 6 &4.5 Point

 

 

 

 

 

 

FIGURE 9 —SMALL COUNTERFORMS VS LARGE COUNTERFORMS

A corrrparison of the two typefaces in Figure 9 reveals that a typeface with large counters,

like Helvetica Light, is easier to read at smaller sizes when compared with a typeface that

29

 

contains smaller counters, like Helvetica Condensed Black. This is because the counters
of the letter are not “swallowed up” as letter size decreases; readers are able to use both
positive and negative spaces to identify the letter.

Letters are produced in a wide variety of stroke weights (see Figure 10). Possible
weights, arranged from lightest to heaviest, are: hairline, extralight, light, book, regular,
medium, dernibold, semibold, bold, extrabold, heavy, black, ultra and poster (weights
that appear in bolded type are pictured below) (Department of Mathematics, University
of Utah, 2001). Letters with thinner strokes are characterized by more open counter
forms than their thicker counter parts (see Figure 9), allowing readers to use the positive

and negative spaces for easy letter identification at small type sizes (Craig, 1980).

X X X X

 

 

 

 

 

Helvetica Helvetica Helvetica Helvetica
Condensed Condensed Condensed Condensed
Light Regular Bold Black

 

 

 

 

 

 

 

 

 

 

 

FIGURE 10- LETTERS WITH VARYING WEIGHTS (Within letter thickness is
uniform.This sample includes only four possible stroke weights.)

Stroke thickness is not always uniform (as pictured in Figure 10); thickness can
vary within a single letter (see Figure l 1). A prominent characteristic of the Old Style
faces, those designed between 1450 and 1693, is little contrast in weight between the

thick and thin strokes of the character (Craig, 1980). Transitional faces, developed by

30

 

designers between 1693 and 1784, had “a tendency toward refinement and greater
contrast between the thicks and thins” (Craig, 1980). Typefaces designed between 1784
and 1815, the Modern era, show an even greater contrast in thick and thin. The Slab
Serif, or Egyptian, (1815-1930) era of typography marked a return to less contrast. This
trend was continued in the Contemporary era (1816—present); contemporary designs are

clmracterized by uniformity in thickness (see Figure 11). They have no contrast in thick

 

and thin.
100 point 100 point 100 point 100 point Century 100 point
Garamond, Baskerville, Modern No. 20, Schoolbook, a Helvetica Light, a
an “Old Style” a “Transitional” a “Modern” refined version of an “Contemporary”
Typefice Typefice Typeface Egyptian, or slab typeface
1617 1757 1788 serif typefice 1957
1894

 

 

 

 

 

 

 

FIGURE 11- AN EVOLUTION OF CONTRAST IN LETTER WEIGHT
(Thick and thin)

Another design element utilized to classify typefaces is the use (see Figure 12),
or exclusion (see Figure 14) of serifs. Serif fonts have terminal strokes that are short
cross lines at the end of the main stroke (International Paper, 1997). “Serifs originated
with the Roman masons who terminated each stroke in a slab of stone with a serif to

correct the uneven appearance made by their tools” (Craig, 1980).

31

 

T/E-Ni

FIGURE 12- THE SERIFS OF A VARIETY OF TYPEFACES -Serifs are in white. “T”
appears in Garamond. “m” appears in Modern No. 20. “N” appears in Century. “W”
appears in Baskerville, and “g” appears in Times New Roman.

Serifs vary in their weight and design. The appearance of serifs, like the contrast of thick

and thin, can be used to identify periods of type design.

32

I I I I

 

 

Garamond,
an “Old Style”
Typeface
l 6 1 7

 

Baskerville,
a
“Transitional”
Typefice
1 757

 

Modern No. 20,
a “Modern”
Typeface
1788 (Bodoni)

 

Century
Schoolbook, a
refined version of
An “Egyptian”, or
slab serif typeface
1 894

 

Helvetica Light, a
“Contemporary”
typeface
1 957

 

FIGURE 13- THE EVOLUTION OF SERIF STYLE

“Over the centuries type became more and more refined; that is, the
contrast between the thick and thin strokes became greater (see Figure 13)
and the serifs became finer (see Figure 10). This refinement was possible
because of the development of smoother papers, better inks, and more
advanced printing methods. The ultimate refinement was attained in the
late 1700’s (the Modern era) when Bodoni reduced the thin strokes and
serifs to fine hairline strokes” (Craig, 1980).

Literature reviewing how serifs impact legibility is divided. Many works indicate

that serifs positively contribute to message legibility, while others indicate that sans-serif

fonts are more easily read.

Researchers who believe serifs contribute positively to legibility (Rehe, 1990;

Craig, 1980; McLean, 1980; Vanderplas and Vanderplas, 1980; Wright et. al, 1977;

Perles, 1977; Tinker, 1963; Burt, 1959) generally provide two reasons for the

improvement of legibility when using serif types: (1) “They (serifs) contribute effectively

to the horizontal movement of the reading eye and thus help in combining separate letters

33

 

into word-wholes”(Perles, 1977) (2) Letters with serifs are more easily differentiated by

readers than letters without serifs (sans serif: see Figure 14).

 

FIGURE 14- SANS SERIF IN A VARIETY OF TYPEFACES (Fonts that do not have
the short terminal strokes required to hide imperfections caused by the early printing
processes).

Researchers who support the legibility of sans serif types (Food and Drug
Administration, 1999 and 1997; Bix, 1998; Pietrowski, 1993; NDMA, 1991) generally
provide the following explanations for improved legibility in the absence of serifs. “Sans
serif type is free of visual distractions” (Garcia, 1981) which improves legibility.
Additionally, the x-heights of sans serif fonts are frequently greater than the x-heights of

serif fonts of equal point size (see Figure 3 and Figure 11); this increase allows for more

34

open counter forms, filling more of the space provided by the type size measure,
improving legibility.

Message Design and Legibility
The preceding discussion involves the elements that come together to create

letters. However, messages are not merely letters. Letters must be integrated into words
to be used to convey meanings through messages. Legibility is affected not only by the
design of the letters, but also by the way that they are presented. Several elements of the
presentation, or layout of the letters and words, can impact the reader’s ability to read the
information.

“Letter spacing is the amount of space used between letters, negative or positive,
either for readability, aesthetics or to fill a certain area” (International Paper, 1997).
Historically, in letterpress printing, which used “. . .metal type, letter spacing is [was]
accomplished mechanically by inserting pieces of metal between the type” (Craig, 1980).
Currently, letter spacing is accomplished by using computer programs to adjust the
distance between letters. Because designers no longer have the physical limitations
imposed by a metal block, negative spacing between letters is now possible. “Negative
letter spacing involves the removal of space between letters individually (keming) or
between all letters equally (white space reduction or tracking)” (International Paper,
1997). Letter combinations that typically allow keming (negative spacing between pairs
of individual letters) include: we, We, yo, Yo, wa, Wa, Ta, To, ye, Ye, wo, Wo, va, Va,
WA, VA (International Paper, 1997). The first letter in each of these two letter
combinations provides a negative space that allows for the “overlap” of the two letters in

the form of kerning (see Figure 15).

35

 

200 point Helvetica Regular 200 point Helvetica Regular

 

 

 

 

 

 

0 spacing (no keming) 0 spacing (no keming)
200 point Helvetica Regular 200 point Helvetica Regular
-150 spacing (kemed) -150 spacing (kemed)

 

 

 

 

 

 

FIGURE 15- KERNING (NEGATIVE LETTER SPACING BETWEEN SPECIFIC
LETTER PAIRS)

Although FDA’s current regulation does not directly address the use of kerning,
they are aware that manufacturers try to exploit label space by using negative letter
Spacing and typefaces with minimal widths. The width of a typeface is dependent on the
Compression or expansion of the font. Arranged from narrowest to widest, the various
Widths for type include: ultra compressed, extra condensed, compressed, narrow,
colldensed, regular, expanded and wide (widths that appear in bolded type are pictured
in Figure 16) (Department of Mathematics, University of Utah, 2001).

FDA indirectly regulates typefice width and negative spacing by mandating “type

Styles which ensure letter compression of no more than 39 characters per inch” (FDA,

36

1999). This decision was largely based on research conducted at the New England
College of Optometry (Watanabe, 1994). Watanabe’s research indicates that the elderly
population had a “great degree of sensitivity to small changes in horizontal letter
compression”. His work suggests that compression has a greater impact on legibility than
letter height. The FDA followed his recommendation that “39 characters per inch is

sufficient to allow good readability” (Watanabe, 1994).

37

 

Helvetica- Narrow (190 points) T]

 

 

Helvetica— Condensed (190 points)

 

 

Helvetica— Regular (190 points) J

 

 

FIGURE 16- A COMPARISON OF TYPEFACES WITH DIFFERENT SET WIDTHS

38

Leading, the amount of space between lines of type, is also directed by the 1999
regulation. Leading is measured fiom baseline to baseline (see Figure 7) and is expressed
in points or fiactions of a point. “The amount of space or leading used in printing is
usually 0 to 2 points depending on the typeface used” (Ralph, 1982). 50-point type with
no lead is written as 50/50; the type size is 50 and the distance between baselines (see
Figure 7) is 50. 50-point type with 10 points of leading is written 50/60 (see Figure 17).

The type size is 50 and the distance between base lines is 60.

IChampion Champiori
Worlds Worlds

50 point Helvetica Regular with
O-leading (SO/50). This is . . .
referred to as type that is set 50 pornt Helvetica R3891?" Wlth
“soli ”. The small difference 10' point leading (SO/50)-
between the ascender (l) and the
descender (p) is the type size
discrepancy discussed earlier.

 

 

 

 

 

 

 

 

 

 

 

 

 

FIGURE 17- EXAMINING DIFFERENCES IN LEADING

Leading is known to contribute significantly to legibility, although “there is no set
rule to follow (with regard to appropriate lead)” (International Paper, 1997).
“Too much leading can sometimes be as bad as not enough. Typefaces
with long ascenders and descenders require more leading. Also, the wider
the measure of text composition, the more leading is required for good
readability” (International Paper, 1997).
Ascenders and descenders are not the only aspect of typefice that dictate differences in

leading, “serif type calls for less leading than sans serif type because the serifs reinforce

39

the horizontal eye flow. Bolder typefaces require more leading than lighter fices” (Rehe,
1990). Typographical researchers Becker et a1. (1970), agree that optimal leading is
dependent on a variety of design fictors, “different typefices need different amounts of
leading.”

FDA does provide a “set-rule” for the designers to follow on OTC labels,
regardless of typeface or message presentation. The FDA regulation specifies a
minimum of 0.5 point leading “to ensure readability” (FDA, 1999). The minimum 0.5
point requirement is less than the l-point minimum that was specified in the proposed
rule (FDA, 1997). This change was made in response to comments that indicated that if
graphical features (i.e., type size, leading, kerning, and highlighting) were required (many
comments indicated that they should not be), that minimum type size and leading should
be reduced to maximize label space.

Research indicates that 0.5 point leading may be problematic. It is likely that the
manufacturers of OTC drugs will use the minimum as a “standar ” element of design
because it will result in smaller labels. However, using the minimum as a standard will
not necessarily produce the most legible labels. As mentioned previously, optimal line
spacing is dependent on typefice and layout. Even if the impact of typeface and layout is
ignored, Ralph’s (1982) findings suggest that FDA is not using sufficient leading for
elderly consumers. His “Publishing Guidelines for Geriatric Visual Concerns” indicates
“no less than two points of leading should be used with smaller than 11 point type”
(Ralph, 1982).

Another fictor to consider when discussing legibility is color contrast, the color of

the text and the background on which it is printed. FDA’s regulation does address the

40

issue of color contrast. “The type must be all black or one dark color, printed on white or
other light, neutral color, contrasting background” (FDA, 1999).

FDA’s contrast mandate is consistent with the vast majority of research findings.
A study conducted at Michigan State University (Lockhart and Bix, 1996) examined the
legibility of 6 color combinations: black type on a white background, blue type on a
yellow background, white type on a blue backgrormd, blue type on a white background,
yellow type on a red background and black type on a red background. Black type on a
white background proved the easiest combination to read for all age groups tested (six
age groups ranged in age from 19 to 81; see Appendix 3 for a more complete review of
data). Research conducted by Sorg (1985) concurs that black on white is the easiest
combination to read. The work of the Institute of Grocery Distribution (IGD) (1994),
Arnold (1972), and Summer (1932) further supports the FDA’s use of a light colored
background paired with a dark type; Arnold and Summer found dark ink printed on
yellow paper to be highly legible, while the IGD supports “dark print on a light
backgroun .” Bradley et al. (1994) concur that black text on a white or yellow
background provides good legibility; they also suggest that these combinations avoid
difficulties associated with red/ green color blindness so that messages are accessible to a
large percentage of the population.

Substrate (paper) color not only is a factor in color contrast, it also affects the
color of the text and graphics printed on the package surface. International Paper (1997)
advises, “Type is more easily read against a soft (yellowish) white, while process colors
reproduce most accurately on neutral white paper.” As a result, the optimal printing

substrate for a textually oriented design may be quite different than one that is graphically

41

loaded; in the case of pharmaceutical messages, this would suggest a soft (yellowish)
white is optimal.

Substrate smoothness, or surfice consistency, also affects message legibility.
Smoothness can be the result of calendering, a processing technique where paper is run
between a stack of rollers. Papers are fiequently classified as “uncalendered, machine
calendered, and supercalendered” (International Paper, 1997), depending on the process
used during their manufacture. Papers produced using these processes vary greatly in
surface consistencies.

Surfice treatments, called coatings, also lmve a significant impact on smoothness.
Coated papers were developed in response to the demand for high quality reproduction of
photographs.

“[They] reproduce much finer halftone screens with sharper definition,

improved density and greater color fidelity than can be reproduced on

uncoated papers. Coated paper finishes range fi'om dull to very glossy,

have a greater affinity for printing inks, greater smoothness, higher opacity

and better ink hold out than uncoated papers” (International Paper, 1997).

In a recent interview, Tom Michalsen of Web Marking Systems emphasized the
importance of using smooth substrates for drug packaging. Michalsen believes that his
pharmaceutical customers are his most demanding customers due to the paramount
importance of legibility for drug labels, and the small fonts that are common in the drug
industry. Because of these stringent requirements, Michalsen indicates extremely smooth
substrate surfices are required for improved ink release for pharmaceutical applications
(Mateo, 2000). Ink release refers to the transfer of the inked image area of the plate onto

the substrate during the printing process. Smoother surfices result in a more complete

transfer of the ink.

42

Substrate-ink affinity, like smoothness, substantially impacts ink release, which
impacts legibility.

“If a paper (the substrate) absorbs too much ink (as in newsprint) the

images appear weak, desaturated and flat (no gloss). If absorbency is low

(as in coated paper) the ink sets near the surface and dries with a

reasonable gloss. This is holdout. If holdout is too high it can cause set—

ofl (transfer to the back of adjacent sheet) in the paper pile” (International

Paper, 1997).

Either end of the absorbency spectrum can negatively impact the inked image and,
therefore, legibility.

Both the substrate and ink have the ability to affect message legibility. The
process by which the package is printed can also have a significant impact. Packages are
usually printed using one of three techniques: gravure, flexography or lithography. These
three processes differ in the type of plate that transfers the ink to the substrate, the way
thattheinkistransferred,thetypeofinkused,andtheappearance oftheprintedirnage.
Within each of the three processes, quality can vary greatly, depending on the skill of the
press associates, the speed that is required by the production schedule, the maintenance
schedule of the press, the quality of incoming materials and even the climate of the
production environment. As a result, like the elements of letter and message design, the
manufacturing process is a series of fictors that affect the printed image, and therefore,
the legibility. There is no mention of either material quality (substrates and inks) or
process requirements in the FDA’s prescriptive standard for legibility.

It is tempting to believe that legible labels can be guaranteed by mandating certain

elements of design; at a glance, the study of textual elements appears simple. After all,

visual recordings have been around for 30,000 years (International Paper, 1997). In truth,

43

even if we exclude the variability introduced by the user and the reading environment,
there are still numerous elements that contribute to design eflicacy. In order for
designers to create accessible designs, they must develop sensitivity to the multitude of
design elements that make up letters, words and messages, the manuficturing methods
and materials used to produce the package, and the complex relationships between these
elements, all of which contribute to legibility.

It may appear reasonable to conclude that sufficient legibility can be obtained by
using a certain size and style of type, with sufficient contrast, a minimum leading and
limited compression. FDA has handed down a prescription for OTC labels in exactly this
fashion. Reality does not match the simplicity of this approach. Legibility is the overall
goal in a complex system of interrelated elements (letter weight, letter compression,
counter form shape, stress, type style, type size, message layout, leading, kerning, ink,
substrate, and printing process); it is not easily prescribed.

A good performance standard for legibility accounts for all of the elements of
design and manufacturing and the interactions of these elements while it is measuring
what is truly important, the legibility of the label. It allows designers flexibility in
design, provides manufacturers with defensible proof of message accessibility, provides
consumers with designs that have been tested to be legible and gets FDA out of the
business of “micro managing” label design. A performance standard for legibility would

better serve industry, regulators and, most importantly, the consumers of OTC drugs.

Materials and Apparatus

This research was divided into four separate studies. Three studies, termed
“preliminary studies”, were designed to examine the effect of different procedures and
instruments on the reproducibility and repeatability of test data. Results fi'om the
preliminary studies aided researchers in determining which procedure to use for the 4th
study, referred to as the “primary study”.

The primary study compared a set of labels that complied with the FDA
regulation to a set that did not. Several of the elements of design previously discussed
were manipulated in an attempt to create noncompliant labels that were more easily read
than four labels that complied with the FDA regulation. All of the noncompliant label
designs contained messages that were created in 5.5 point type, type that did not meet the
6 point minimum set by FDA. This is of interest, because many manuficturers have
argued that the 6 point minimum does not ensure, or even appreciably improve, legibility
over smaller type (CI-IPA, 1999; NDMA, 1991).

If successful, this research answers the call fi'om industry issued by R. William
Soller. Soller, Senior Vice President and Director of Science & Technology at the
Consumer Healthcare Products Association, believes there is a need for an objective
method to measure legibility and argues against the 6 point minimum set by FDA.

“. . ...[there is a] notable lack of convincing objective support that 6 point

type adds an advantage in legibility over smaller types sizes down to and

including 5.0 point type. .. These are important observations, and it is

important to find ways to address them” (CHPA, 1999).

Not only does this research provide Dr. Soller with objective evidence, it demonstrates

the ability of the Lockhart Legibility Instrument (LLI) to account for the effect of

45

multiple elements on legibility in a single test, and also provides insight into the complex
interactions of the elements of design.

The Lockhart Legibility Instruments (LLIs):
Legibility of printed material can be measured objectively, using instruments and

a test protocol being developed at the Michigan State University (MSU) School of
Packaging. Instruments referred to as “Lockhart Legibility Instruments” (LLIs) quantify
the legibility of textual messages with a single test. Legibility is measured in terms of the
degrees of rotation of an analyzing filter. The degrees of rotation is referred to as the
“Legibility Index”; the higher the Legibility Index value, the more difficult it is for a
subject to read the message being tested.

The LLI has evolved since its creation in the 19603. Initially, the concept was
developed as a way to measure label impact (step one in Rousseau’s 1998 model, termed
notice); it was primarily used as a marketing device. Dr. Hugh Lockhart, a professor at
Michigan State University’s School of Packaging, recognized that the instrument was not
fulfilling its potential and began using it as a way to measure legibility.

The LLI is essentially a large box with two light sources inside. Printed items to
be tested are placed inside the box on an easel (see Figure 18). Two 25-watt
incandescent floodlights illuminate the test material. Power to the floodlights, and,
ultimately, the light level inside the LLI, is controlled by a rheostat. Once the light level
inside the instrument has been adjusted, subjects are asked to look through a viewing
screen that is located at the front of the instrument and adjust a handle to their right until

the first point that they can easily read the printed message without straining their eyes.

46

  
 
     
 
 
       

  

Easel
illuminated by
two

      
     

  

  

incandescent , .
Easel. floodlights (not
rllumrnated pictured
by two ' A, .
incandescent

floodlights
FIGURE lS-LOCKHART LEGIBILITY INSTRUMENT EASELS (Older LLI,
built in 1993, is on the left. New model, built in 1999, is pictured on the right.)

The viewing screen that the subject looks through is made of a pair of polarizing
filters. Subjects begin by rotating a handle on their right; as they rotate the handle, the
first filter inside the viewing screen also rotates. Rotation of the first filter controls the
amount of light that reaches the test subject’s eyes. Messages that are difficult to read
require more light; therefore, difficult messages require the subject to rotate the filter
further, resulting in high legibility indexes.

The filters are Polaroid HN22 Linear Polarizing Filters that are 0.030 inches thick.
James Pietrowski (1993) indicated that he chose the HN22 filters because they had a
1miform level of light transmission through the portion of the spectrum to which the eye
responds, 440-750 nm wavelength. The unique properties of Polaroid lend themselves to
application within the viewing screen of the LLI.

“Polaroid represents a class of materials that absorbs light oscillations in
one direction but not the component oriented at right angles. These

47

materials often contain long particles, rods or plates, aligned parallel to
each other in a regular arrangement. These aligned particles transmit one
plane of polarized light and absorb the perpendicular one. . The Polarizer
can transform circularly polarized light into linearly polarized light” [see
Figure 19] (Department of Physics and Astronomy, ASU, 1999).

 

 

 

 

 

 

 

Polarizer
Y
Y
t VA
Circularly polarized Lineorly polarized

FIGURE 19- CIRCULARLY POLARIZED LIGHT BECOMES LINEARLY
POLARIZED AFTER BEING PASSED THROUGH A SINGLE POLARIZIN G
FILTER (Department of Physics and Astronomy, ASU, 1999)
Unlike Figure 19, the LL] uses a pair of filters (see Figures 20 and 21). The filter
closest to the subject is referred to as the analyzer. Subjects rotate the analyzer as they
adjust a handle to their right. A second filter, placed directly behind the analyzer, is

referred to as the polarizer. The polarizer is fixed in place; it does not move as the

subject adjusts the handle.

48

 

 

Polarizing ,
Filter Analyzmg
Frlter

 

 

 

 

 

/

 

 

FIGURE 20-POLARIZER AND AN ALYZER- AXES CROSSED (0° of rotation)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Polarizing Analyzing
Filter Frlter
«
/
PW
Linearly
Polarized Light

 

 

 

FIGURE 21- POLARIZER AND ANALYZER- AXES PARALLEL (90° of rotation)

Subjects begin at 0° of rotation; the handle to the subject’s right is turned
clockwise until it comes to a physical stop. At 0° of rotation the axes of the filters are
crossed (see Figure 20); when the axes are crossed the analyzer will absorb all the light
transmitted by the polarizer. As a result, when testing begins, the viewing screen is
black. Subjects are asked to rotate the handle using a counterclockwise motion until the

first point that they can easily read the text without straining their eyes. This action

49

rotates the analyzer (relative to the polarizer) to a maximum of 90°. At 90° the two filters
have their orientation axes aligned parallel to each other. The first filter, the polarizer,
transmits linearly polarized light (see Figure 19). This linearly polarized light is then
transmitted, without absorption, through the second filter, the analyzer (Department of
Physics and Astronomy, ASU, 1999). The more difficult the text is for the subject to see,
the more light they require. For more light, subjects have to rotate the analyzer further.
This results in a higher Legibility Index value for items that are difficult to read.

Two LLIs were used in the research presented here; one was constructed in 1993
and a second was created in 1999. James Pietrowski (1993), a graduate student,
developed the older instrument (see Figure 22). Coefficients of variation are large for
this instrument, typically ranging fi'om 30-50%. In an attempt to try to reduce the
variability of data, Dr. Hugh Lockhart directed the production of a second instrument in

. 1999 (see Figure 23).

   
 

 

 
  
 

Printed text is placed inside the
instrument where it is
illuminated by two incandescent
floodlights

Rotating
Analyzing Filter

  
 

 

 

  
  

 
 

Analog Readout of Legibility
Index (Degrees of Rotation of
Analyzing Filter)

 
 

 

  

FIGURE 22- THE 1993 LOCKHART LEGIBILITY INSTRUMENT (LLI), BUILT BY
JAMES PIETROWSKI
(Preliminary study #3, which investigated the variability of data collected by each of the
two LLIs was the only study that used the older instrument)

50

The 1999 model features several improvements over Pietrowski’s model. Many
of the improvements were added in an attempt to reduce variability. A light meter is now
contained in the easel, allowing researchers more precise control of the intensity of light
illuminating the test material. Digital readout of the degrees of filter rotation has
replaced an analog system; the digital readout has a sensitivity of 0.1 degrees of rotation,
as compared with 1 degree of rotation for the analog. The viewing area has been reduced
in size, and a shielded view port has been added in an attempt to block distracting
reflections and ambient light. The view port was created using a pair of oxyacetylene
welder go ggles with the lenses removed.

Another new feature of the 1999 model is the adjustable easel. The new LLI was
built so that subjects are able to manipulate the distance that they use to view the test
material. The easel inside the new model is mounted to a track (see Figure 18). Subjects
can move the easel along the track by adjusting a hand crank at the front of the
instrument (see Figure 23). This enables subjects to alter the viewing distance, allowing
them to adjust to their “natural” reading distance rather than reading test material fiom a
distance chosen by researchers.

It was hypothesized that the new features of the 1999 instrument, which allow
more precise control, would result in smaller coefficients of variation of measurements.
Preliminary study 3, one of three studies that examined the variability of data produced
by LLIs, tested this hypothesis. The Procedures and Results Chapters of this document

detail the experiment and findings, respectively.

51

 

 
 

 

    
 
 
  

Light meter directly
:. mounted to easel for
. more precise control
of internal flood

 
 
 
 
 

Shielded view port

blocks ambient light
from interfering with
subject’s View

 

 

  
    
 
 
  

Digital readout
of both intenal
light and
degrees of filter
- .. . it-.. .‘w
Hand wheel allows subject
to more precisely control
the degrees of rotation

 

  
 
 

 
 

   

Hand crank allows subject
to manipulate the viewing
distance

 

FIGURE 23-THE 1999 LLI, BUILT BY SYCAMORE TECHNICAL SERVICES
Legibility Cards (Preliminary Studies Only):

All three preliminary studies used six cards printed in a single font, 10 point
Helvetica Light. Label designs were created using Microsoft Word and were printed
using a Hewlett Packard 722 Ink Jet printer. Each card contained a different message
(see Table 1). Messages were shown in previous studies to have a statistically marginal
(Bix, 1998) or insignificant effect (Lockhart and Bix, 1996) at or=0.05. Cards were
labeled one through six so that they could be easily identified (see Table 1).

After messages were printed, they were mounted onto 3” x 4” cards, with a
message centered horizontally on each. All messages were printed using a black on
white contrast. Black on white was chosen because it provides a high degree of contrast,
is generally recognized as highly legible (Lockhart and Bix, 1996; Sorg, 1985; Paterson

and Tinker, 1935) and does not pose problems for people with color vision deficiencies.

52

 

 

 

 

 

 

 

 

 

 

 

 

 

Font Size Meisage Font Message— As it appeared on the card
. H 1 t- It may help most of them to work today. She
10 Pomts 1 if}? works In lhlS club after midnight. The order to
g gowilbe done after two. .
. H 1 t' She works in this club after midnight. The .
10 P011118 2 ighica order to go Will be done after two. There Will
be some sugar in the kitchen.
. H 1 t' The order to go will be done after two. There
10 P011118 3 ighica will be some sugar in the kitchen. Here IS a
copy of lunch hours for taday.
. H 1 t' There Will be some sugar in the kitchen. Here
10 P011118 4 :Yehica IS a copy of lunch hours for today. From here
lg tothere flowers cannot grow;
_ Helvetica Here is a copy of lunch hours for today. From
10 Pomts 5 Light here to there flowers can not grow. It may
helnmostotthemlmrktoday
. H 1 t- From here to there flowers can not grow. It
10 POMS 6 11:11:“ may help most of them to work today. She
works in

 

 

thisclunaiteLmidnight.
TABLE l-PRELIMINARY STUDY LABEL MESSAGES

Labels (Primary Stuay Only):
The primary study used eight labels created using Adobe Illustrator 9.0 (see

Appendix 4 for a complete set of labels). Label fictors included design, compliance and
message. Design contained four levels: Univers Ultra Condensed in 9.0 points, Gill Sans
in 6.0 points, Lucida Fax in 5.5 points and Verdana in 5.5 points. Design was nested
within compliance; the Univers Ultra Condensed and the Gill Sans labels were always
compliant with the FDA regulation, while the Lucida Fax and Verdana designs were
always noncompliant (see Figure 24). Compliance indicates that labels complied with
the FDA’s regulation; noncompliance indicates that labels violated the regulation. The
body text of noncompliant labels was only 5.5 points. They did not meet the 6-point
minimum prescribed by FDA.

Two levels of message, “drug” and “nonsense”, were used to test for an effect of

familiarity with message (see Appendix 4 for specific messages). It was hypothesized

53

that the drug labels would require fewer degrees of rotation than their nonsense
counterparts because subjects would be familiar with their contents. The wording for the
drug label message was taken fi'om a sample label published in the final rule (FDA,
1999). This message was then converted into the “nonsense” message by replacing each

word with a random word of equal length. For example, the word “ ” may have been

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

replaced with the word “golf”.
Univers 1 Univers 1 Gill Sans 1 Gill Sans
Ultra Ultra
2 Condensed 2 Condensed 2 2
3 Nonsense 3 3 Nonsense 3
Message Drug Message Message Drug Message
1 Lucida Fax“ E 1 Verdana . ll 1 Verdana;
l 2 j ’ 2 2
3 l . 3 Nonsense 3
Drug Message l Message Drug Message ‘

 

 

 

 

 

 

 

 

 

 

Designs pictured in white were always compliant with regard to the FDA regulation,
while designs pictured in gray were not compliant. As a result, compliance is nested
within design in the analysis of data. The numbers 1, 2 and 3 indicate the general
label location where subjects were asked to read text

 

 

 

FIGURE 24- FACTORS OF LABEL DESIGN

As mentioned previously, several elements of design were manipulated in an
attempt to create noncompliant labels that were at least as easily read as labels that

complied with the FDA regulation (see Appendices 5 and 6 for designs that were not

54

used in the study). The rmjor element of design that determined which labels were used
in the study was x-height (see Appendix 4 for Primary Study Labels).

Gill Sans was used to create one of the two compliant label designs. Gill Sans is,
asFDArecommends,asansseriftypeface. Itwaschosenbecauseitis firmlywithinthe
guidelines set by FDA but has a small x-height for a sans serif font (see Figure 25). By
contrast, Verdana, one of the two noncompliant designs, has a much larger x-height (see
Figure 25), providing a more Open counter form at small sizes and giving the reader a
better visual impression of letter size. This is also true of the other noncompliant design,
Lucida Fax. Lucida Fax was chosen not only for this reason, but also because it runs
contrary to FDA’s recommendation of sans serif typefaces. It does contain serifs, but

unlike many serif fonts, has a large x-height (see Figure 25).

    

i i-“Ii
‘1 A ‘ I; A

 

 

 

 

Gill Sans Verdana Lucida Fax
70-point sample 70-point sample 70-point sample

 

 

 

 

 

 

 

 

 

FIGURE 25- COMPARING THE X—HEIGHTS OF GILL SANS, VERDANA AND
LUCIDA FAX TYPEFACES (UNIVERS ULTRA CONDENSED IS NOT PICTURED)

Although the other compliant typeface, Univers Ultra Condensed (UUC), does
have a mirly large x-height, it also has a large amount of letter compression (see
Appendix 4). This letter compression makes it a difficult font to read, despite the fact

that it has a fairly large x-height, illustrating the point that the elements of design cannot

55

9S

‘ov/oz ‘OZ/OZ 5" 891008 ‘SPJOM Jame III (100: ‘HIOMISN “011901193 [901mm 9101401)

“on“ [sures] our peer urea 95: [mason e qorqm re oouersrp oqr sereorpur .roruunnouep

eql '[pasn sfiemle st 199} oz] can oqr pear nee roofqns oqr qorqm ‘rmqo aqr (110.9

(1993 “1) ”WP 9‘11 SGWOIP“! .roreramnu 9‘11. “WW 9 59 13°le 9! 41!”? Imam

'ueueug ,(q pedoreAap sum ruqr srlnsor Arman .10} realm} preparers e arzmrn pres Krrnoe
[Bust/t rurod mu sorurreqrqdo Sunnog moq eqr urog sruernamseem Krrnoe [ensr A

parser sem uorsrA men noqr 9mm ‘sesserfi Surpeer .ro sraoogrq

89 [tons ‘sasuar poquosald 119m mm 01 perms“! 919M 810°!an '(1861 ‘Smmoo moo)

“1qu1 p008, u; safia naqr urog seqom 91,, pm aqr ploq or are sroel’qns reqr arers pm aqr

no momma! mu '(9z amfitd 905) pm mm [mm mod mu sownmIquo finance

moq eqr Sursn pamseetn sem roefqns qoea JO Krrnoe [ensrA eqr ‘Burrsar or mud
(sazpms 11 V) we sozwzmrurdo Burma Moo

'Krmqrfiel 9023on or prefix qum sruanrfipnl' exam or sroufirsep
sxsa rnq ‘eArrdrroseJd st 1; fuorrelnfier men oqr qrrm Annomrp our JO trad st snu 'peer
'01'5999 10 19°19 10“ 9! POSUQPUOC) 9111.0 819mm ““11 ”“319 9‘1 PW” 11 '(6661 ‘VCI:1)
“elfirsedfir ‘peer-or-Ksee ‘maro ‘9[3U!S,, flue eq prnoqs eoegadfir our reqr steerpu! op Kelp
‘aoeredfir Agroods ron saop Kouefie our 91% °rou st rr reqr penflm aq pluoo rr ‘uorralnfier
6661 ‘L1 mm 9m 01 918391 mm Invndtnoo 88 can DWI—191 am am finoqnv
'perunoo cram sleqel $.19me am no srods poroalos Anuopum nor uaqm 9,; sum qour
rod sroroemqo JO runoo 939.19,“; eqr srurod 6 1V va,; ,(q ponnbor qour red sraroemqo 65
9m MOPQ 993mm 939889111 9m mun 99mm! 8% 100} can awe 92:8 am ‘Kpms Sim
u; peruesard $19qu eqr Burufirsep :11 1pm red sreroemqo 6g JO umurrxeur e 30 rueurerrnbor
9111 m0 S! 1! 95mm WPIIdWOOUOU 8! 1! 1991mm1 ummwrw 9m) smod 9 19 mm sacrum

srqr u! poounouord os 51 uorssarduroo 191191 11092105; In ‘pererorp .IO ‘paroprsuoo sq

20/200, etc., indicate that the subject being tested at 20 feet canjust discriminate letters
that a person with normal vision, the average person, can see at 20 feet, 40 feet, 200 feet

and so forth (Kelly, 1993).

 

FIGURE 26- DOW CORNING OPHTHALMICS NEAR POINT VISUAL ACUITY
CARD (Card is reduced fi'om actual size)

A. W. Sperry Light Meter, Model SLM~110 (A 11 Studies)
A light meter (See Figure 27), manufactured by A.W. Sperry, was used in two

ways. In all studies, the light meter was used to record the ambient light at the time of
testing. It was also used to measure and control the intensity of light inside the older
instrument; preliminary study #3 was the only study that utilized this instrument. To
measure the light level inside the 1993 instrument, the sensor was placed on the lower
easel and the lid was closed. The light inside the LL] Was adjusted using a rheostat until

the light meter reached a level of 25 foot-candles i one foot-candle. Because the new

57

LLI has a sensor that is built into its ease], it was not necessary to use the Sperry Light

meter to ensure the proper illumination levels inside the new instrument.

      
 

Light Level Readout (in
Foot-Candles)

 

FIGURE 27— AW. SPERRY LIGHT METER AND SENSOR

58

Procedures

As mentioned previously, this research was divided into a series of four
experiments. Preliminary studies one and two were repeatability/reproducibility studies
that tested for an effect of procedure on the variability of data collected. Preliminary
study three was also a repeatability/reproducibility study; it examined the effect of
instrument on the variability of data collected. The final study, termed the primary study,
examined several effects of design on legibility, as measured by the LLI.

The procedures of each experiment will be presented in this chapter separately so
that readers can develop a clear idea of each experimental design. Subject orientation
and the collection of subject-related information, which was the same for all experiments,
is presented below.

Subject Orientation
Prior to testing, the level of ambient light was measured by placing the sensor of

the A.W. Sperry light meter on the table to the right of the LLI, facing the ceiling. The
ambient lighting conditions were recorded prior to testing, but could not be adjusted. All
testing was conducted at the School of Packaging under florescent lights during daylight
hours in an attempt to maintain consistency.

Before data was collected, subjects were provided with a brief one-on-one
orientation with the researcher. The researcher explained,

“This is an instrument that quantifies how easy or difficult a label is to

read. It does this by measuring the amount of light a subject requires to

read a given message. The harder a message is to read, the more light is

required; the easier it is to read, the less light is required. If you choose to

participate in this study you will be asked to fill out some information

regarding your education, eyewear and age. This information will be
anonymous; your name will not be recorded on any documents. Your

59

visual acuity will be measured and recorded. You will be asked to read a

series of labels that are placed inside the instrument. If you normally use

corrective eyewear to read you should use it for this experiment. You will

read the labels by looking into the viewing field on the instrument. As

you look through the viewing field rotate the handle to your right counter-

clockwise until the first mint where you are able to easily read all of the

words you are directed to read without straining your eyes. Testing will

not take any longer than 25 minutes. It is important to remember that this

is a test of the printed labels and not of your eyes. There is no need to

worry about how your results compare with the results of other test

subjects. We are concerned about how the labels compare to one

another.”

In order to protect the test subjects’ rights, subjects were asked to review and sign
a written consent form (IRB # 01-292; expiration May 15, 2002). Subjects signed the
consent form, signifying either acceptance or rejection of the invitation to participate,
before testing began (see Appendix 7 for written consent form).

Collection of Subject-Related Information
After the subject signed the consent form, a data-recording sheet was used to

record subject-related information, which included: gender, educational background and
age group (see Appendix 8 for the Primary Study’s recording sheet, Appendix 9 for the
recording sheet used in preliminary study one and two, and Appendix 10 for the
recording sheet used in Preliminary Study 3). These sheets were also used to identify the
order in which the subjects participated. The sequential order of testing was recorded as
“subject number”; in other words, the first subject in each study was labeled subject one,
the second subject two, etc.

After demographic information (gender, educational background and age group)
and subject number were recorded, subjects were given the Dow Corning Ophthalrnics
Near Point Visual Acuity card (see Figure 26). While seated in fi'ont of the LLI, they

were asked to hold the card approximately 16 inches in front of them and read the

60

smallest print that they could read. The researcher aided subjects with the l6-inch span
by marking the distance with a measuring tape. Each subject’s Snellen visual acuity was
recorded on the appropriate data-recording sheet (See Appendices 8-10). Data was
collected from all subjects willing to participate in the study; however, only the results of
subjects with measured visual acuities of 20/30 or better were used in the data analysis.

In a previous study (Bix et al., 1997), an analysis of residuals revealed that initial
readings tended to be higher than readings that followed. This suggested that subjects go
through an adjustment period as they get used to using the LLI. As a result of this
information, it was decided that two “dummy cards” would be used before any data was
recorded. Subjects read the dummy cards in the same manner that they did the test
material, but these readings were not recorded or analyzed, allowing subjects to adjust to
the instrument without affecting test results. Two dummy readings were taken before
testing, and any time that the distance between the subject and the message was changed
during testing. Dummy cards were used in all studies presented in this document.
Preliminary Study One

In preliminary study one, the ease] was positioned at the m of the instrument
(see Figures 18 and 23); subjects were asked to adjust its distance until it was at the most
comfortable reading distance for them. They viewed two dummy cards and each of the
six messages (see Table 1) twice, for a total of 12 recorded readings (dummy cards were
not recorded) fiom the distance that they chose. Data was also collected fi'om a fixed
distance of approximately 18.5”. From this distance, subjects read a set of two dummy
cards and each of the six messages twice, for a total of 12 recorded readings fi‘om a fixed

distance of 18.5”. To counteract any effect of learning as subjects became familiar with

61

the messages, the procedure alternated between subjects; if subject one began the study
by adjusting the distance of the easel to the distance of their choosing, then subject two

began fi'om a fixed distance, etc. (see Figure 28 for a graphical description of the

procedure).

 

 

Preliminary Study One-
Manipulated Distance from the
Front of the Instrument Vs Fixed Distance

 

 

 

 

1“ Fixed distance 2:18.5” Read two dummy cards
and 6 messages twice each

 

 

Subject Manipulated Distance

(begin at instrument front)
2'“I W Read two dummy cards
Subject 1 and 6 messages twice each

 

 

 

 

 

Subject Manipulated Distance
(begin at instrument front)
Read two dummy cards

In W and 6 messages twice each

 

Fixed distance x 18.5”

 

 

 

 

 

 

 

 

 

2"d Q3 3 Read two dummy cards
Subject 2 and 6 messages twice each
Fixed distance a: 18.5”
1't . . Read two dummy cards
and 6 messages twice each
Subject Manipulated Distance
(begin at instrument front)
2nd Read two dummy cards
Subject 3 and 6 messages twice each

 

 

 

 

 

 

FIGURE 28-GRAPHICAL INTERPRETATION OF THE PROCEDURE USED FOR
PRELIMINARY STUDY ONE

62

Subject Demographics: Preliminary Study One

Ten people were tested using the aforementioned procedure to examine whether
allowing subjects to choose their own viewing distance created less variable data than
imposing a fixed distance. Of the 10 subjects, seven were between the ages of 19-28 and
three were between the ages of 29-38.

Visual acuity, another factor tint was statistically examined, also varied within
the subject population. Seven of the ten subjects had visual acuities of 20/20; three had
measured visual acuities of 20/30. Two of the three 20/30 acuities were in the younger
age group so that acuity was not confounded with age in the analysis.

The gender and educational level of each subject were recorded, but not used in
the statistical analysis. Of the ten people tested, six were female and four were male. Six
were enrolled in a doctoral program, one was completing a master’s degree, two were
undergraduate students and one had completed high school.

Preliminary Study Two

In preliminary study two, the easel was positioned at the back of the instrument
(see Figures 18 and 23) and subjects were asked to adjust its distance until it was at the
most comfortable reading distance for them. Please note the change in the beginning
position of the easel, which makes this study different from preliminary study one. As in
preliminary study one, each subject viewed six messages (see Table l) twice from the
distance they chose and twice fi'om a fixed distance of approximately 18.5”, for a total of
24 readings; dummy readings were performed, but not recorded. Again, the procedure

alternated between subjects to counteract any effect of learning that occurred as subjects

63

became familiar with the test cards (see Figure 29 for a graphical representation of the

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

procedure).
Preliminary Study Two-
Manipulated Distance from the Back of the
Instrument Vs Fixed Distance
1’t Fixed distance 318.5” Read two dummy cards
and 6 messages twice each
Subject Manipulated Distance
(begin at instrument back)
2nd fll-fi—Q-é— Read two dummy cards
Subject 1 and 6 messages twice each
Subject Manipulated Distance
(begin at instrument back)
Read two dummy cards
1’t W and 6 messages twice each
Fixed distance 2: 18.5”
. 2'“l $3 3 Read two dummy cards
Subject 2 and 6 messages twice each
Fixed distance a: 18.5”
1’t . . Read two dummy cards
and 6 messages twice each
Subject Manipulated Distance
(begin at instrument back)
2"d W++ Read two dummy cards
Subject 3 and 6 messages twice each

 

 

 

 

 

 

FIGURE 29-GRAPHICAL INTERPRETATION OF THE PROCEDURE USED FOR
PRELIMINARY STUDY TWO

Subject Demographics: Preliminary Study Two

Ten people were tested using the aforementioned procedure to examine whether
allowing subjects to choose their own viewing distance created less variable data than
imposing a fixed distance. Of the 10 subjects, five were between the ages of 19—28 and
five were between the ages of 29-3 8. Eight of the ten subjects tested had visual acuities
of 20/20. Two had measured visual acuities of 20/30. Both subjects that had visual
acuities measured to be 20/30 were from the youngest age group.

Gender and educational level were recorded, but not used in the statistical
analysis of the data. Seven of the ten tested were female; three were male. Seven
subjects had begun working on doctoral degrees, one was working toward a master’s
degree and two were pursuing undergraduate degrees.

Preliminary Study Three
Preliminary study three compared the variability of data collected using the older

instrument (see Figure 22) with the variability of data collected with the new instrument
(see Figure 23). It was hypothesized that rmny of the features of the new model, which
were discussed in the Materials and Apparatus Chapter of this document, would decrease
the variability of data.

In preliminary study three, distance was permanently fixed at approximately 17.5”
and the light level was set to 25 i 1 foot candles for both instruments. Ten subjects used
the 1993 model (see Materials and Apparatus Chapter) to view two dummy cards and
each of the six messages (see Table l) twice, for a total of 12 recorded readings from the
older instrument. subjects also read two dummy cards and each of the six messages
twice using the 1999 model. There were a total of 12 recorded readings using the newer

instrument. To counteract any effect of learning as subjects repeatedly read the cards,

65

each subject began testing with a different instrument than the previous subject. That is,
if subject one began the study by using the 1999 model (see Materials and Apparatus
Chapter), then subject two began testing using Pietrowski’s 1993 model (see Figure 28

for a graphical description of the procedure).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Preliminary Study Three-
Effect of Instrument on Data Variability
1’t 1993 Model (Older) Read two dummy cards
and 6 messages twice each
1999 Model (Newer)
2“d Read two dummy cards
Subject 1 and 6 messages twice each
1999 Model (Newer) Read two dummy cards
1" and 6 messages twice each
1993 Model (Older)
. 2"‘1 Read two dummy cards
Subject 2 and 6 messages twice each
1993 Model (Older)
I" Read two dummy cards
and 6 messages twice each
1999 Model (Newer)
2'“I Read two dummy cards
and 6 messages twice each
Subject 3

 

 

 

 

 

 

FIGURE 30-GRAPHICAL INTERPRETATION OF THE PROCEDURE USED FOR
PRELIMINARY STUDY THREE

66

Subject Demographics: Preliminary Stuay Three

Ten people were tested using the aforementioned procedure to conrpare the
variability of data collected using the older instrument to that collected with the newer
model . Of the 10 subjects, nine were between the ages of 19-28 and one was between
29-38. Eight of the ten subjects tested had visual acuities of 20/20. Two had measured
visual acuities of 20/30. Both subjects that had visual acuities measured to be 20/30 were
from the younger age group.

Gender and educational level were recorded, but not used in the statistical
analysis of the data. Six of the ten tested were male; four were female. Seven subjects
had begun working on master’s degrees and three were pursuing doctoral degrees.

Primary Study
In the primary study, fifty subjects read eight labels at the top, middle, and bottom

of the label to test two hypotheses: (1) A noncompliant label can be created that is at least
as legible as one tlmt complies with the FDA regulation; (2) Labels that contain a familiar
message (a drug label) will require fewer degrees of rotation than a message that is
unfamiliar to subjects (a nonsense label).

After the ambient light level had been recorded, and signed consent had been
obtained (see Appendix 7), subjects were asked to read eight labels (see Appendix 4)
using the new LLI (see Figure 23). Label factors included: design (4 levels), compliance
(2 levels), and message (2 levels) (see Figure 24). An additional factor, position (3
levels), was tested in the ANOVA; subjects read all labels at the top, the middle and the
bottom position. Due to requirements of the statistical program (SAS), different levels
within each factor were assigned numbers (see Table 2). For example, labels created

using Univers Ultra Condensed were considered design level one, Gill Sans labels were

67

design level two, Lucida Fax labels were called level three and all Verdana designs were
referred to as level 4. Labels were assigned a number, one through eight, so that the

researcher could easily and quickly identify the combination of factors and levels that

 

 

 

 

 

 

 

 

 

 

 

made up each label (see Table 2).
Message . .7
Label # (Level #FifrnAnalysis) (Level # in (Leiltillpirlrlmysis)
Analysis)
9 point Univers Ultra .
l Condensed (l) Nonsense (1) Complrant (l)
2 6 point Gill Sans (2) Drug (2) Conmliant (1)
3 6 point Gill Sans (2) Nonsense (1) Compliant (l)
4 5'5 pomt(15;rc1da Fax Nonsense (l) Noncompliant (2)
5 5.5 point Verdana (4) Drug (2) Noncompliant (2)
6 9 point Univers Ultra Drug (2) Compliant (1)
Condensed (l)
7 5'5 pomtégcrda Fax Drug (2) Noncompliant (2)
8 5.5 point Verdana (4) Nonsense (l) Noncompliant (2)

 

 

 

 

TABLE 2- FACTORS AND LEVELS OF PRIMARY STUDY LABELS
Before testing began, the researcher positioned the easel so that it was at the front
of the instrument, and placed one of the two dummy cards on it. With the analyzer at 0°
of rotation (total darkness), the light level inside the machine was adjusted to 25 foot
candles i 1. Once the light level had been adjusted, the researcher rotated the analyzing
filter to a total of 90° of rotation (total light). Subjects were then instructed,

“Look through the viewfinder. You should see an easel holding a card.
Turn the hand crank in the center of the machine (see Figure 23). This
will adjust the distance between you and the card. Make the adjustment
until the card is at the most comfortable reading distance for you. You
will probably be able to read the card fiom any distance, but choose the
distance that is most comfortable for your eyes, just like you might adjust
the distance of a book for comfortable reading.”

68

The light level inside the LLI was adjusted as subjects moved the easel in an attempt to
maintain a constant level of 25 foot candles falling on the surface of the easel.

After the easel distance lmd been adjusted, and the subject had completed two
readings using the dummy cards, data collection began. As mentioned, 50 subjects read
each label (see Appendix 4) in three different places (a line near the top, the middle and
the bottom of the label). The labels were randomly grouped and then assigned the
numbers one through eight (see Table 2). Subject one read the message on labels one
through eight first at the top, position one. The same subject then read labels one through
eight a second time, but this time at the label’s middle, position two. For the third and
final reading, subject one again read labels one through eight, but this time they were
asked to read the label at the bottom, position three. For subject two a single label was
rotated to the bottom of the pile (label one was placed behind label eight). As a result,
subject two read labels two through eight followed by label one. Additionally, subject
two began the test by reading the message in the middle of the label, position two, first.
Subject two read the labels a second time, again labels two through eight followed by
one, but this time they read the message at the bottom of the label, position three. Subject
two finished testing by again reading labels two through eight followed by label one, but
this time read the top position of the label, position one. This rotation of label and

position continued throughout the testing for all 50 subjects.

69

 

lSt Subject is presented the labels in the
following order:

12345678

 

 

Position 1

 

 

2“d Subject is presented the labels in the
following order:

12345678

 

Position 2

 

 

 

 

Subject 1

 

 

3rd Subject is presented the labels in the
following order:

12345678

 

 

Position 3

 

 

lSt Subject is presented the labels in the
following order:

23456781

 

 

Position 2

 

 

2nd Subject is presented the labels in the
following order:

23456781

 

 

Position 3

 

 

 

Subject 2

 

 

 

3rd Subject is presented the labels in the
following order:

23456781

 

 

Position 1

 

 

FIGURE 31-GRAPHICAL INTERPRETATION OF THE PROCEDURE USED FOR

Fifty subjects tested for the primary study were included in the analysis. Subjects

THE PRIMARY STUDY

Subject Demographics: Primary Study

70

with measured visual acuities lower than 20/30 were tested, but their results were not
used in the statistical analysis of the data. Subjects tested who were older than 48 were

also eliminated fi'om the analysis. This data was eliminated because previous studies

 

 

 

 

 

 

have shown that variability of data collected using the LLI increases with increasing age
(Bix, 1998) and decreasing visual acuity. Since there is potential to use this instrument as
part of a performance standard for legibility, keeping variation to a minimum is
paramount.

Originally, it was proposed that only subjects age 19-28 would be used for this
study. However, it proved difficult to find 50 qualified subjects within this age range
who were willing to participate. As a result, two more age groups (29-38 and 39-48)
were included in the analysis. It is important to note that the number of subjects in the
two older age groups is relatively small, and that in the analysis there was no significant
difference attributable to age group. Thirty-five subjects were 19-28; twelve subjects
were between the ages of 29-38 and three were 39-48. Of the 50 subjects that were
included in the analysis, thirty-eight had a measured visual acuity of 20/20 or better and
twelve were measured to be 2030.

The educational level and gender of subjects were also recorded, but these were
not used in the analysis of data. Of the 50, thirty-four were male and .sixteen were
female; twenty-four subjects were currently pursuing undergraduate degrees, 17 master’s

degrees and nine doctoral degrees.

71

Results

Preliminary Studies One and Two:

Figures 32 and 33 summarize the coefficients of variation for the data collected
during preliminary studies one and two, respectively. Coefficients of variation are
presented because the first two preliminary studies were concerned with the effect of
procedure on the variability of data.

Preliminary study one used a group of 10 people to compare the variability that
resulted when researchers fixed the reading distance at 18.5” to the variability that
resulted when subjects adjusted the easel to the most comfortable position for them. In
preliminary study one, subjects began manipulating the easel distance fiom the front of
the instrument. Preliminary study two used a second group of 10 people to compare the
variability that resulted when researchers fixed the reading distance at 18.5” to the
variability that resulted when subjects adjusted the easel to the most comfortable position
for them, with distance adjusted from the back of the instrument.

Results were tested for statistical significance using a mixed model Analysis of
Variance (ANOVA).

Response = p, + Distance + Study + Message + Subject (Study) + Visual

Acuity + Distance“ Subject (Study) + Residual
Subject, an effect that was nested within study, and residual were considered random
effects. All remaining effects were treated as fixed effects. Using this approach, it was
determined that the variability of the results produced by the 10 people from study one
differed significantly fi'om the variability of the results obtained fi'om the 10 people tested

in study two. An analysis of the residuals shows these differences graphically (see Figure

72

34). Subjects in preliminary study two produced data that was significantly more
variable than subjects who participated in preliminary study one. This difference is also

evident in the coefficients of variation (see Figures 32, 33 and 35).

 

 

Coefficients of Variation .Distance Fixed at
Preliminary Study 1 18.5"

 

 

 

30

25 D Subjects Adjust
the Distance
(Begin from

 

 

 

Instrument Front)

0 .1 fl fl SUDjOCt

FIGURE 32- COEFFICIENTS OF VARIATION FOR PRELIMINARY STUDY ONE
BY SUBJECT (Subjects begin adjusting distance fi'om the front of the instrument)

 

 

 

Percentage
...L .3 N
0| 0 01 O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

Coefficients of Variation 'aissgnce Fixed at
0 Preliminary Study 2 '
25 U Subjects Adjust the

0 Distance (Begin from
g 20 the back of the
= Instrument)
o 15
2 , I ll
g 10 ‘ 5

5 d I] l1 I1 ll ll — -

 

 

 

 

 

 

 

 

 

 

FIGURE 33- COEFFICIENTS OF VARIATION FOR PRELIMINARY STUDY TWO
BY SUBJECT (Subjects begin adjusting distance from the back of the instrument)

73

Residuals Versus Distance Procedure

 

60

59..
401
30"
211-2

O
8
z i
10* ‘

.31 ---------------------------- t ------------- l

-204

F6 P1 50' P1 Fl) P2 SDIPZ

0 ”00m 0 O. .0

Residual

 

 

 

FIGURE 34: RESIDUAL ANALYSIS PRELIMINARY STUDY ONE
VERSES PRELIMINARY STUDY TWO
(FD P1 refers to Fixed Distance, Preliminary Study 1; SD P1 refers to
Subject Manipulated Distance, Preliminary Study 2; FD P2 refers to Fixed
Distance, Preliminary Study 2, etc.)

Because subject is nested within study, and the subjects were shown to differ
significantly, results fiom study one cannot be compared with study two. However, the
effect of procedure can be examined by limiting analysis to within study comparisons.

Within study comparisons were made using a likelihood ratio test. Each study
was analyzed using two statistical models. Model one assumes equal variances of
treatment; ifthedistancewasfixed, variabilityoftheresultantdataisno differentthan
the variability produced when subjects adjust the distance (otz=022). Model two assumes

there is an effect of treatment on the data’s variability (otzmzz); fixing the distance

74

between the subject and the reading material produces more (or less) variability than
occurs when subjects choose the viewing distance. The -2 residual log likelihood value
of Model 2 is subtracted from the —2 residual log likelihood value of Model 1 to get a test
statistic variable that, under the null hypothesis, has a chi-squared distribution with one
degree of freedom. From this distribution a p-value is obtained to determine whether the
two models return significantly different results. Statistical significance indicates that
there is difference in the variability that occurs when different procedures are employed.

The —2 residual log likelihood values for preliminary study one were 1048.7
(Model 1) and 1047.4 (Model 2). The chi-squared value was 1.3; this resulted in p-value
of 0.25421. There was no statistical difference in the variability that occurred when
researchers fixed the distance and the variability that occurred when subjects chose their
own reading distance by beginning easel adjustment fi'om the fiont of the LLI.

The —2 residual log likelihood values for preliminary study two were 1417.8
(Model 1) and 1415.1 (Model 2). This resulted in a chi-squared value of 2.7. The p-
value was 0.10035. Like preliminary study one, there was no statistically significant
difference when researchers fixed the easel distance and when subjects chose the distance
by beginning easel adjustment fiom the back of the instrument.

A simple comparison of within subject variation and between subject variation
was made by calculating several coefficients of variation. Figures 32 and 33 reveal
information about within subject variation because they depict CVs for each subject by
treatment. These values can then be compared with coefficients shown in Figure 35,

which include variability tlmt results fi'om differences in subjects.

75

 

 

Percentage

 

 

 

 

 

 

 

FIGURE 35- COEFFICIENTS OF VARIATION: PRELIMINARY STUDIES ONE
AND TWO BY MESSAGE
A comparison of Figure 35 with Figures 32 and 33 reveals something of interest.

Coefficients of variation are much smaller when they are examined on a per subject basis
(see Figures 32 and 33). When the coefficients are measured between subjects, as in
Figure 35, they are much larger. Our research examines the effects of procedure and
instrument on the variability of data collected using the LLI. However, it is important to
note that much of the variability of observations is attributable to the differences in the

subjects themselves, something that is beyond our control.

Preliminary Study Three:
Figure 36 summarizes the coefficients of variation for the data collected during
preliminary study three. Coefficients of variation are presented because preliminary

study three examined the effect of instrument on the variability of data collected.

76

Specifically, study three aimed to determine if the new instrument resulted in less

variable data than the older model.

 

Coefficients of Variation-Preliminary

Study 3 I 1993 Instrument
D 1999 Instrument

 

 

A
O

 

 

 

 

(a)
O

 

 

  

 

 

 

Percentage
re
0

.3
O

 

 

 

 

 

 

 

 

 

 

Ojifllnfln I I'

subject12 3 4 5 6 7 8 910

 

 

 

FIGURE 36: COEFFICIENTS OF VARIATION FOR PRELIMINARY STUDY
THREE

As with preliminary studies one and two, comparisons for preliminary study three
were made using a likelihood ratio test. Two statistical models were employed to
determine if the instrument has an effect on the variability of data collected. Model one
assumes equal variances of treatment; there is no difference in the variability collected
using the 1993 model when it is compared with the variability of data collected using the
1999 model (012=022). Model two assumes there is an effect of treatment on data
variability (01215022); the variability ofthe data collected with the 1993 model is assumed

to be different than that collected with the 1999 model.

77

The —2 residual log likelihood value of Model 2 is subtracted fi'om the —2 residual
log likelihood value of Model 1 to get a test statistic variable that, under the null
hypothesis, has a chi-squared distribution with one degree of freedom. From this
distribution a p-value is obtained to determine whether the two models return
significantly different results. Statistical significance indicates that there is difference in
the variability that occurs when different instruments are used.

In addition to comparing the two models by way of a p—value, the variability of
data collected can also be examined using the residual values that were calculated as part
of model two. The residuals, a measure of variability, attributable to the older instrument
were 16.755; this was nearly eight times greater than the residuals attributed to the 1999
model, which were only valued at 2.3304. Although this gives an indication that the
treatments were different, it is not a formal comparison.

A formal comparison was made using the likelihood ratio test. The —2 residual
log likelihood values for preliminary study three were 1190.8 (Model 1) and 1152.5
(Model 2). The chi-squared value was 38.3, which resulted in p-value of 6.0663* 10'").
This indicates a highly significant difference in the variability of the data collected with
the 1993 and 1999 LLIs.

It is tempting to tout this significant difference as a momentous accomplishment.
After all, many of the new instrument’s features were added in an attempt to reduce
variability. However, it is important to temper this enthusiasm by reviewing the
graphical representations of the variability of data (see Figure 36). Four out of the ten

people tested actually had more variable data when they used the newer instrument. The

78

large magnitude of significance is at least partially attributable to just one individual,

subject 7, who had much more variable results when using the 1993 LLI.

Primary Study:

Table 3 summarizes the average legibility index for each of the 8 labels, and

provides readers with information about the treatment combinations that make up each

 

 

 

 

 

 

 

 

 

 

one.
Label Design . 9 Message Drug or , , ,
# (Point Size) Compliant. Nonsense? Average Legibility Index
Univers
Ultra
1 Condensed Yes Nonsense 29. 1
(9.0)
Univers
6 Ultra Yes Drug 28.8
Condensed
Gill Sans
3 (6.0) Yes Nonsense 27,8
Gill Sans
2 (6.0) Yes Drug 28.1
Lucida Fax
4 (5.5) No Nonsense 26.9
Lucida Fax
7 (5.5) NO Drug 26.7
Verdana
8 (5.5) NO Nonsense 26.8
Verdana
5 (5. 5) N0 Drug 26.4

 

 

 

 

 

TABLE 3- TREATMENT COMBINATIONS FOR EACH LABEL AND AVERAGE
LEGIBILITY INDEX FOR EACH TREATMENT COMBINATION

Figure 37 visually summarizes the average legibility index of each label tested in the

primary study. All four compliant messages required a larger average legibility index

value than the four noncompliant designs; larger values are indicative of text that is more

difficult to read.

79

 

Three out of the four designs, all but the Gill Sans, had a nonsense message that
resulted in a greater average legibility index than the identical design presented as a drug
message. These differences, however, were quite small for all of the four designs tested;
the largest difference, 0.4 degrees of rotation, is attributed to the Verdana design. Its

nonsense message had an average legibility index of 26.8 and its drug message an

 

 

 

 

 

 

average of 26.4.
35
30 Average Legibility Index for Each Label UUC
Nonsense
UUC Drug
25 " .°.°::
‘ I111 Gill Sans
20 _ . . Nonsense
. EGill Sans Drug
15 ‘ ""L I Lucida Fax
.. Nonsense
10 _ B Lucida Fax
Medical
aVerdana
5 ‘ . 1 Nonsense
. mVerdana Drug
0 _

 

 

 

 

FIGURE 37- AVERAGE LEGIBILITY INDEX VALUES FOR ALL EIGHT LABELS

Figure 37 presents a variety of treatments as one average, a grand average, for
each label. Information about the results of subjects with varying visual acuities is

presented in Figure 38. Figure 38 presents the median and data spread of each label

80

when subjects with visual acuities of 20/20 viewed them versus the median and data

spread when people with visual acuities of 20/30 viewed them.

Treatment Combinations Across all Designs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

C
3100— NH DH NH DH NM DH NH DH
39‘“! i i i t i i t"
0811—: * g N *
[Elli—X- ; i i * * i
L t t
050— 1 g" ' 8
$50_ # aL uh
gin—
r:3330— - " "‘ - I
29" I T T T T T
in—
n
. . I l l T I Tl IN! T] l l l l
‘l’flz’fijzlnmgrfg‘lzlzizlzi2121212
2= 211/31] uuc cs LF Var

 

FIGURE 38- BOX PLOTS OF SEVERAL TREATMENT COMBINATIONS
(UUC indicates Univers Ultra Condensed designs, GS indicates Gill Sans designs, LF
indicates Lucida Fax designs and Ver represents Verdana designs; NM indicates
nonsense messages while DM indicates drug messages)

Results were tested for statistical significance (See Table 4) using a mixed model
Analysis of Variance (ANOVA).

Response = p, + Subject (Age Group) + Compliance + Design

(Compliance) + Position + Message + Visual Acuity + Age Group +

Conlpliance*Message + Message'Design (Compliance)
+ Position‘Design (Compliance) + Residual

81

Subject, an effect that was nested within age group, and residual were considered random
effects. All remaining effects were treated as fixed effects. Design was nested within the
factor compliance due to the fact that designs 1 and 2 (Univers Ultra Condensed in 9.0
points and Gill Sans in 6.0 points) were always compliant, and designs 3 and 4 (Lucida

Fax in 5.5 points and Verdana in 5.5 points) were always non-compliant with regard to

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

the FDA regulation.
Fixed Effects
Num Den
Def)?” Defies F Value Pr > F Significant?
Freedom Freedom
Compliant 1 1135 16.17 (.0001 YES
Desigl (Compliant) 2 1 135 1.38 .2532 No
Position 2 1135 71.80 (.0001 YES
Message 1 1 135 0.14 .7053 No
Visual Acuity 1 46 1.34 .2359 No
Age Group 2 46 1.59 .2154 No
Compliant * Message 1 1135 0.13 .7162 No
. . *
D3313“ (“mph“) 2 1135 0.17 .8413 No
Message
. . *
Des‘g“(c‘?‘.m’lm) 6 1135 1.74 .1089 No
Posrtron
Random Effects
Cov Parrn Estimate Smd Z Value Pr Z Sigmf' lcant?
Subject (Age Group) 454.75 95.3254 4.77 (.0001 YES
Residual 58.0118 2.4352 23.82 (.0001 YES

 

 

 

 

 

 

TABLE 4- RESULTS OF THE RESTRICTED FORM OF THE MIXED MODEL
USING SATTHERWAITE’S METHOD (Bolded effects indicate statistical significance
a= 0.01. Italicized Efi‘ects are significant at a= 0.05).

After the data had been analde (see Table 4), an analysis of the residuals
revealed the normal probability assumption was not supported when the raw responses
were examined. Gill (1978) suggests that failing to meet this assumption is not critical in

many cases. “The f test of the hypothesis of treatrrrent effects is known to be robust, i.e.,

82

 

the probabilities of errors of Type I and Type II are little effected [sic] by moderate
departures from normality” (Gill, 1978). Although the f-test applied in the analysis
presented here is type III, Gill’s comments with regard to robustness still apply (Cardoso,
2001).

Nonetheless, a second model was used to reanalyze the data (See Table 5). The
second model divided the 50 subjects into two groups based on their residual variability.
Subjects with estimated residuals that fell outside of i 3 a were considered group one
(See Figure 39). Eight subjects produced data that was ten times more variable than the
other forty-two subjects; the forty-two subjects with less variable data were considered
group two.

Residuals Versus Stbjecl

(mam)

 

8
J
an.

 

V V"

"v'v-v v r‘ vv v

 

 

Residual
B 8
l l

 

 

P o
’ O
l ‘ o
It! - -!--E-:~ij
-‘D—l l ||. ,
“1' I {1511 ‘ tll ! i; .

00
-“O

 

 

 

Stbjed

FIGURE 39- RESIDUALS VERSUS SUBJECT (Subjects were divided into two groups
based onthe variability ofthedatathattheyproduced. Members ofthe groupwith
highly variable data can be recognized because a line has been drawn through their

residuals)

83

This second model accounted for the differences in the two groups’ residual variability,

without errantly attributing these differences to a factor in the model.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fixed Effects
Effect Degrees of Degrees of F -Value PR>F
Freedom Freedom
Compliance 1 983 30.07 (.0001
Design 2 983 3.05 .0477
(Compliance)
Position 2 983 82.26 (.0001
Message 1 983 .85 .3562
Visual Acuity 1 45.4 1.38 .2468
Age Group 2 45.6 1.60 .2122
Complaince * l 983 .10 .7512
Message
Design
(Compliance)* 2 983 .64 .5254
Message
Design
(Compliance) * 6 983 298 .0069
Position
Random Effects
9"“ ‘3'th Stnd. Estimate Standard Error 2 Value PRZ
Subject
(Operator) 450.09 94.8297 4.76 (.0001
Residual 245.39 25.6831 9.55 (.0001

 

 

 

 

 

TABLE 5- RESULTS OF THE SECOND MODEL (Bolded effects indicate significance

at a= 0.01. Italicized Effects are significant at a= 0.05.

The second analysis does not negate the first treatment of the data, but uses a

model that is better suited to the data set. After data was completely analyzed using both

models, comparisons were made between the results to see how closely each estimated

84

 

significance. The models produced the same results with respect to significance with two
exceptions: the factor design (compliant), which was not Significant in the first model,
changed to significant at a=0.05 when the second was applied, and the significance level
for the interaction between design (compliance) and position changed fi'om insignificant
to significant at a = 0.01.

Using the second analysis, the factors compliance, design and position are
statistically significant. Compliance and position are significant at a=0.01 and design is
significant at (1:0-05- This is informative, but does not give a high level of detail in the
results. (We know that design is a significant factor, but is there a significant difference
between the Gill Sans and the Lucida Fax designs?) To achieve more detail, pair-wise
comparisons of each possible combination of compliance, design (compliance), and
position were tested for significance using a Tukey-Kramer test. Because of the large
number of possible combinations when the interactions of all three factors are considered,

limited results are presented here (see Table 6; for a complete set of results, see Appendix

 

 

 

 

 

 

 

 

 

1 1).
A Compliant Design 'Position‘ Compliant Design Position p Value Significant
UUC ‘ LF 3 .3

Yes GS All , No Ver All 5.2 10 YES
Yes UUC All Yes GS All 0.0819 No
Yes UUC All No LF All 1.5* 10" YES
Yes UUC All No Ver All 5.0*10?’ YES
Yes GS All No LP All 0.0870 No
Yes GS All 7 No Ver All 0.01366 YES

“No LF All ~ No Ver All 0.9085 No

 

 

 

 

 

 

 

 

TABLE 6- TUKEY-KRAMER PAIR WISE COMPARISONS
Tukey-Kramer tests break the results into a series of tests for statistical
significance. Table 6 examines each possible pair of designs using the Tukey-Kramer

method. Each row represents a comparison of treatments; the first design, shaded in

85

 

gray, is tested against a second design, which is not shaded, to determine if the pair are
statistically significantly different from one another.

From Table 6 it is apparent that Univers Ultra Condensed (UU C) is significantly
more difficult to read than all other designs, with the exception of Gill Sans (GS). This
supports the idea that letter compression has a greater impact on legibility than type size
(Watanabe, 1994). The comparison of GS and UUC produced a p-value of 0.0819, which

is not significant at a=0.05. A comparison of GS and Lucida Fax (LF) designs also
failed to produce statistically different results, p—value = 0.0807. A comparison between
LF and Verdana (V er) yielded another insignificant p-value (the value was 0.9085).
These results add an important dimension to the findings. One of the main
purposes of the study was to Show that noncompliant designs could be created that were
more easily read than designs that complied with the regulation. It is tempting to report
the effect of compliance as significant at a=0.01, and the effect of design as significant at
a=0.05 (see Table 5). However, to report only this information would not provide

readers with a thorough examination of the results. Although the results were highly
significant for the entire group of compliant designs versus the noncompliant designs, as
we examine pair wise comparisons, it is evident that the designs are close in their
legibility measurements.

Nonetheless, the research did achieve its objective. Noncompliant labels were
created that were statistically easier to read than labels that complied with the regulation.
The Verdana label was shown to differ fiom the Gill Sans label with a p-value of
0.01366; the same label differed from the Univers Ultra Condensed label with a p-value

of 5.0‘10'7. Although the Lucida Fax label did not significantly differ fi'om the Gill Sans

86

label (the p-value was only 0.0870), it did differ significantly from the Univers Ultra
Condensed label with a p-value of 1.5*10'5.

Another goal of this research was to address the issue of message familiarity and
legibility (see Figure 40). It was hypothesized that messages tlmt subjects were familiar
with, the drug message, would require fewer degrees of rotation than messages that
subjects had not viewed before, “nonsense” messages. Three out of the four designs, all
but the Gill Sans, had a nonsense message that had a greater average legibility index
value than the identical design presented as a message typical of a drug label. When the
results were analyzed for significance, they were not found to be significant at a=0.05.
The insignificance of the differences in messages is shown graphically in Figure 40.

There are two possibilities with regard to this result. The first is that subjects
rotate, as they are instructed, to the first point that they can easily read the text without
straining their eyes, regardless of whether they are familiar with the message or not. This
is a desirable outcome for the LLI.

The second possibility is that subjects were no more familiar with this common
drug message than the message that was created using random words. This conclusion
would support the idea that, despite potential dangers of OTC misuse, consumers are not
highly involved with OTC products (Reisenwitz and Wirnbish, 1997; Sansgiry and Cady,
1995; Robinson and Stewart, 1981); they do not read labels. This is consistent with the
findings of a survey conducted by Dr. Janet Engle, Professor of Pharmacy at the
University of Illinois, Chicago. At a news conference in December 1998 she indicated,
“47% failed to always read the product label before starting a pain medication, and one-

third were unaware that over-the-counter (OTC) drugs carry ris ” (Norton, 1999). Given

87

the significant ramifications of improper OTC use, and the difficulty in changing

consumer/product involvement, this second possibility is a fiightening, but real, risk.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Box Plots- Message and Design
:12: _ UUC * * GS * LF §Ver
s ,0 _. i z i f
E i * il- -rt- 1*
g 70 _ * g * t * t
..5 60 __ ll if
8 5° “
l— 40 ——t
3’ so ——
20 fi
10 -+ l I I I I I I I
0 '1
1= Nonsense Message ' ' l l ' I I I
2: Drug Message 1 2 1 2 1 2 1 2
Compliant Noncompliant

 

FIGURE 40- BOX PLOTS OF THE FACTORS MESSAGE AND DESIGN

88

Conclusions

This research pr0posed two testable hypotheses: (1) A non-compliant label can be
created with equal (or greater) legibility than one that complies with the FDA regulation,
a regulation that is intended to ensure legibility and (2) Drug labels, which contain
messages that subjects are familiar with, will require fewer degrees of rotation than non-
sense labels, which subjects have not seen before. Both of these hypotheses were tested
using data obtained by the LLI, an instrument that provides an objective measure of
legibility. The higher the degree of rotation, termed the legibility index, the more
difficult a message is to read.

Although the statistical analysis (see Table 5) reveals that the compliant label
designs are more difficult to read than the noncompliant designs at a level of a=0.01 , the
difference in means is not practically significant. The largest difference in treatment
means occurs when the designs created using Univers Ultra Condensed are compared
with designs that utilize the Verdana typeface. The difference in the means of these two
designs is 2.5707 degrees of rotation when the nonsense labels are compared and 2.338
degrees when the labels containing drug messages are compared.

From a practical standpoint, any comparisons made between Univers Ultra
Condensed designs and the noncompliant label designs represent an exaggeration of
results. It is an exaggeration because it is unrealistic to assume that drug manufacturers
would choose to use Univers Ultra Condensed. Although we have termed it as compliant
with regard to the March 17, 1999 regulation, it could be argued that it is not. Even

though the agency does not specify typeface, they do indicate that the typeface should be

89

any “single, clear, easy-to-read, typestyle” (Food and Drug Administration, 1999). It
could be argued that Univers Ultra Condensed is not clear or easy-to-read. This is part
of the difficulty with the new regulation; it is prescriptive, but asks designers to make
judgments with regard to the legibility of a typeface.

A comparison between the means of the Gill Sans and Verdana designs provides a
more realistic comparison of compliant versus noncompliant labeling. Gill sans is a sans
serif font, and sans serif fonts are strongly encouraged by the regulation. Gill Sans is
clear and easy-to-read. Although these designs have been shown to be statistically
different (see Table 6), the difference in the mean reading for each of the two designs
(see Table 3) is very small (1.0 degree of difference for the nonsense messages, and 1.7
degrees for the drug messages).

Although the practicality of this difference can be debated, the results do
accomplish the goal of the research; noncompliant labels were created that were at least
as legible as labels that complied with FDA’S regulation. The results also demonstrate
the LLI’s ability to detect small differences in legibility and show the complexity of the
interrelated variables that determine how easy, or difficult, a message is to read.

Dictating these variables one by one is not the best approach to ensure the
legibility of OTC labels, but it is the approach taken by FDA in their 1999 regulation.
Research presented here challenges the regulation and this approach; different design
variables were carefully examined and manipulated in an attempt to create compliant
labels that would be difficult to read and noncompliant labels that would be easily read.
Although the objective was successfirlly accomplished, noncompliant designs were

statistically easier to read than their compliant counterparts, the differences were very

9O

small (see Table 3 and Figure 37). The creation of noncompliant labels that were more
legible than labels that complied with FDA’s regulation proved a formidable task; FDA
did a good job at specifying the variables of design that aid readers.

FDA did, however, fail to dictate several variables that were not studied here, that
undoubtedly impact legibility. In the FDA regulation there is no indication of quality
requirements for materials or production methods; the sole focus of the regulation is
design and format. Just as the elements of letter and layout have a multitude of factors
that impact legibility, so do the materials and production techniques used to create labels
and packages.

A performance standard for legibility, utilizing a measurement tool like the LLI,
not only accounts for the production issues, but takes into account the various elements of
both letter design and layout, while measuring what is important, the consumer’s ability
to read the label. The performance standard approach allows designers flexibility in
design, provides manufacturers with defensible proof of message accessibility, gives
consumers designs that have been tested to be legible and gets FDA out of the business of
“micro managing” label design. A performance standard for legibility would better serve
industry, regulators and, most importantly, the consumers of OTC drugs.

Suggations for Future Research

1. Rousseau’s model (notice, encode, comprehend and comply) illustrates that there are
more aspects to effective label design than legibility. For labels to be effective
consumers must (1) notice them (2) encode them (3) comprehend them and, finally (4)
comply with them. Study into the other three aspects of Rousseau’s model and ways

(like the LLI) to quantify a label’s success or failure at each of the steps is needed. It is

91

anticipated that this team will be investigating consumers’ capacity to notice label
elements using eye-tracking technology.

2. Another issue that is in need of study is the involvement of elderly consumers with
OTC products. “Product involvement refers to consumers’ knowledge about the personal
relevance of the products in their lives” (Peter and Olson, 1999). This personal, or self,
relevance can be either intrinsic or situational and can vary in intensity. Consumers who
are more involved with a product will devote more of their resources to it. These
resources are not necessarily monetary, but may take the form of time and effort.
Involved consumers are more likely to seek information (fiom the label and fi'om other
sources), use complex rules when evaluating alternatives, and devote focal attention and
controlled comprehension to the product (Rifon, 2000).

Involvement is very important when we examine consumer behavior relating to
the proper use of OTC drug labels. Research is split on whether elderly consumers have
a high level of involvement when purchasing OTC drugs (Sansgiry and Cady, 1996; Gore
et. al, 1994) or a low level of involvement (Reisenwitz and Wirnbish, 1997; Sansgiry and
Cady, 1995; Strutton and Tanner, 1994; Robinson and Stewart, 1981) Further research
into the involvement level that elderly consumers have with OTC drugs is needed.

3. FDA’S failure to address production-related issues and their impact on legibility was
mentioned several times in this work. The labels produced for this study were created
using a laser printer; although production issues were discussed, it was not a focus of this
work. Research into various aspects of printing (materials and production) and their

effect on legibility is needed. Of particular interest is the surface reflectance of

92

packaging materials and the impact that this has on the elderly population’s ability to read
labels.
4. It was suggested in this document that x-height is a better indicator of a typeface’s
legibility than type size. A study to explore this hypothesis is advised. It was also
suggested that previous studies my have errantly attributed increased legibility to
incorrect elements of design. Bix (1998), for example, suggested that sans serif fonts
were more easily read than serif fonts; this may have been an incorrect conclusion based
on the fact that the typefaces being compared had unequal x-heights. Explorations in x-
height, as it relates to legibility, using the LLI are advised.
S. A sampling of OTC labels on the market and elderly consumers is advised. Can
elderly consumers effectively decipher the 6 point type size unaided?
6. Work to further the use of this instrument as part of a performance standard for
legibility is advised. The first step is to define what “legible” is, in terms of degrees of
rotation. This definition should be created with a particular concern for the elderly, who
are at particular risk for drug mismanagement for a variety of reasons.
7. “Interactive warnings” are a relatively new type of design that is meant to take
advantage of environmental cues, increasing label effectiveness.
“This format (interactive warnings) requires the product user to
physically manipulate the warning when using the product, and
researchers have found that these types of warnings increase the
likelihood of the user noticing and complying with the
information... the interactive labels serve as an event-based cue to
recall the appropriate safety procedures” (Rousseau et al., 1998).
The application of interactive warnings to packaging applications is a ripe area for

future research.

93

Appendix l-Summary Table of FDA’s Activity Regarding Labels

94

How to use these tables:

Table 7 presents a general summary of the regulatory activity occurring at FDA regarding
OTC labels. Table 8 presents a summary of the headings and subheadings required under

section “c” of the final rule, “Over-the-Counter Human Drugs; Labeling Requirements.”
Table 9 summarizes label-formatting requirements set forth in section “d” of the

regulation.

 

Table 7-FDA OTC Legibility Publications

 

Date-Action-Title

Summary

 

February 27, 1997-
Proposed Rule

The rule attempted to improve the legibility and ease of use
of OTC labels by requiring: a specific font (Helvetica), a
minimum font size, standardized headings and
subheadings, bullet points, pictograms, specified words
and an exact format for all required information

 

March-December 1997-
Study A, “Evaluation of
Proposed Over-the-
Counter Label Format
Comprehension Study,”
is conducted

Examined the influence of variation formats on the
communication of directions for use and required warnings
(March 17, 1999 Federal Register). 1,202 consumers were
randomly surveyed at malls in 8 states. Consumers were
directed to “view examples of OTC label designs.
Respondents were asked questions designed to measure
knowledge and attitudes about OTC drug products, as well
as decisions about proper use of the products” (FDA. Study
A Table of Contents). Before consumers were
interviewed, information regarding site location, past
participation, subject age, corrective eyewear and gender
was recorded. Subjects who had previously participated in
the study, were younger than 18 or did not have eyewear
that they normally required for reading were dismissed.
Interviewers did not measure or record subjects’ visual
acuity.

 

March-December 1997-
Study B, “Over-the-
Counter Label Format
Preference,” is conducted

Examined “examples and variations of current OTC label
designs. Respondents were asked to indicate their
preference for various designs. Also, consumers were
asked to evaluate labeling terminology and graphics to
investigate how they interpret various ways of
communicating drug safety and efi’ectiveness” (FDA.
Study B Table of Contents).

 

December 30, 1997

Comment period on Study B is announced. It closes

February 13, 1998

 

 

February 13, 1998

 

Comment period on Study A is announced. It closes March
30, 1998.

 

95

 

 

Table 7- Continued

 

 

March 17, 1999- Final Rule establishes a “standardized format and standardized
Rule- “Over-the-Counter content requirements for the labeling of OTC drug
Human Drugs; Labeling products. This final rule is intended to assist consumers in
Requirements” reading and understanding OTC drug product labeling so
that consumers may use these products safely and
effectively. This final rule will require all OTC drug
products to carry the new, easy-to read format and the
revised content requirements within prescribed
implementation periods.

 

 

96

 

 

Table 8-Content Requirements for OTC Labeb (Section 201.66 (c))

 

March 17, 1999- Final

Rule- “Over-the-Counter

Human Drugs; Labeling
Requirements”

Rule establishes a “standardized format and standardized
content requirements for the labeling of OTC drug
products. This final rule is intended to assist consumers in
reading and understanding OTC drug product labeling so
that consumers may use these products safely and
effectively. This final rule will require all OTC drug
products to carry the new, easy-to read format and the
revised content requirements within prescribed
implementation periods” (FDA, 1999).

 

Section

Summary

 

201.66c 1 Heading

“Drug Facts” is required

 

201.66c 2 Heading

“Active Ingredients”- established name and quantity of
each active ingredient/dosage unit follow this heading

 

201.66c 3 Heading

“Purpose” or “ ses” - general pharmacological
category(ies) of drug or of each active ingredient follow
this heading

 

201.66c 4 Heading

“Use” or “Uses”- the indications for use of a product
follow this heading

 

201.66c 5 Heading

“Warning” or “Warnings” subheadings (where applicable)
are specified in sections 201.66(95i-201.66(c)5x

 

201.66c 5i Subheading

“For external use only”, “For rectal use only”, “For vaginal

use only”, “Allggy alert”

 

 

201.66c Sii Subheading “Do not”
201.66c 5iv Subheading “Ask a doctor before use if you have” followed by pre-
existingonditions

 

201.66c 5v Subheading

“Ask a doctor or pharmacist before use if you are”
followed 11y cautions about potential drug/ food interactions

 

201.66c 5vi Subheading

“When using this product” followed by side effects
consumers may experience substances/activities to avoid

 

201.66c 5vii Subheading

“Stop use and ask a doctor it” followed by signs of toxicity
and other serious reactions

 

201.66c 5viii Subheading

This section directs the placement of other warnings not
covered previously

 

201.66c 5x Subheading

Reference to Poison Control Centers

 

201.66c 6 Heading

“Directions” followed by applicable directions for use

 

201.66c 7 Heading

“Other Information” followed by information that doesn’t
fall within any of the other categories in 201.66(c) but is
required or made optional under other OTC drug regulation
or an approved drug application

 

 

201.66c 8 Heading

 

“Inactive ingredients” followed by ingredients listed in
alphabetical order

 

97

 

 

Table 8- Continued

 

 

201.66c 9 Heading

 

“Questions?” or “(hlestions or Comments?” followed by
telephone number printed in a minimum of 6-point bold
font. It is also recommended that the days of the week and
times when someone is available to respond to questions
be included.

 

 

Table 9- Format Requirements: Section 201.66d (for Presenting the Title,
Headings, Subheadings and Information Set Forth in 201.66(c)l through

201.66(c)9)

 

201.66d 1 Capitalization
Justification of Headings
and Subheadings

The first level of each word in the title in 201.66c1 must
appear in upper case. Only the 1”t letter of the 1St word of
each heading and subheading of c2-c9 appear in upper
case. Title headings and subheading set forth in cl-c2 and
c4-c9 must be left justified

 

201.66d 2 Type Size
(Title, Headings,
Subheadings and Text)

“Drug Facts” must appear in a type size greater than the
largest type size used within the “Drug Facts” area. This
title must be no smaller than 8-points. Headings in
paragraphs c2-c9 must be at least 2-point sizes larger than
the text (8 point or greater type). The subheadings and all
information described in 201.66c2-c9 must appear in at
least 6-point type. Format exceptions for small packages
require text no smaller than 6 points. “The agency chose to
require a minimum type size of 6-point and type styles
which ensure letter compression of no more than 39
characters per inch”.

 

201.66d 3 Font, Leading,
Kearning, Contrast and
Highlighting

Any “single, clear, easy-to-read type style” is allowed.
The agency believes that san serif type styles are the most
likely to be considered clear and easy to read. They note
that Helvetica and Univers have consistent and uniform
stroke weight characteristics and are commonly available.
The title “Drug Facts” must appear in bold italic print. At
least .5 point leading is needed to ensure readability. The
type must be all black or one dark color, printed on white
or otherght, neutral color, contrast'mg background.

 

 

201.66d 4 Bullet Point
Style and Format to
Introduce and Highlight
Informative Statements

 

Solid square or circles of 5-point type size must be
presented as the same shape and color throughout the
labeling. Bullets and bulleted statements under each
heading subheading must be vertically aligned to ensure
visual separation and adequate white space between
discrete information chunks. Two bulleted statements are
allowed on a Single line, however each statement must be

separated by at least 2 square “m”s.

 

98

 

 

 

Table 9 Continued

 

201.66d 5 Multiple
Panels

Provides that headings, subheadings and information
required under 201.66c including the warnings section,
may appear on more than one panel. Appropriate visual
cues must be provided so that the flow of information is
retained. The arrow, directing the consumer to the
continuation of information on the next panel. The
continuation of the required content and format onto
multiple panels must retain the required order and flow of
headings, subheadings and information. The UPC symbol
may appear on the same panel as some of the information,
but must be outside the box or enclosed.

 

201.66d 6 Active
Ingredients

Listing of the active ingredients. The established name,
the quantity or proportion and the “purpose” of each active
ingredient is listed.

 

201.66d 7 Graphical
Images

Graphical images (such as UPCS) and any information not

set forth in section 201.66c must not interrupt the required

information panel or panels. The UPC symbol may appear
on the same panel as the required information but must not
be outside the box or enclosure

 

201.66d 8 Placement and
Style of Lines

Lines partition the information set forth in 201 .66c1-c9. A
bar line must be used to form a box or similar enclosure to
separate the sections (sections begin with a heading).
Hairlines separate subsections (subsections begin with a
sub heading). Hairlines must extend to within two spaces
on either side of the “Drug Facts” box while bar lines
extend to each end of the “Drug Facts” box.

 

 

201.66d 9 Directions

 

Requires that dosage directions, when provided for 3 or
more age groups or populations must be presented in a
table format. A text format may be used when there are
less than 3 dosage directions.

 

99

 

 

Table 9- Continued

 

 

201.66d 10 Small
Package Format

 

Products are considered small when 60% or more of the
total surface area available to bear labeling on the entire
outside container or wrapper, or immediate container or
wrapper would be needed to present FDA required
labeling. For products that are sold with an outer package,
the FDA is encouraging, but not requiring, the use of the
modified small package format on the immediate
container. Font size for body text does not change (6
points /no more than 39 characters/inch). Headings must be
minimum 7 points. Leading may be less than 0.5 points.
(It can be adjusted so that the ascenders and descenders of
the letters do not touch). Bulleted statements may continue
to the next line and need not be vertically aligned. The box
required in 201.66d8 may be omitted if the headings,
subheadings and information in 201.66c1-c9 are set off
from the label by color contrast

 

100

 

Appendix 2- A Variety of Typefaces in the Same Type Size

101

Champion
L'hampiun
Champion
Champion
Champion
Champion
Champion

Champion
Champion
Champion

Champion
Champion
Champion
Cllillllllloll
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion
CHAMPION

(Abadi MT Condensed 20 points)
(Agency FB 2|] paints)

(Arial 20 points)

(Aria1 A1ternative)
(Arial Black 20 points)

(Arial Narrow 20 points)
(Arial Rounded MT 8 20 pts)

(Arial Unicode MS 20 points)

(Baskerville Old Face 20 points)
(Batang 20 points

(Iouhour 20 poinlr)
(Bell MT 20 Points)

(Berlin Sans FB 20 Points)
(Bernard 1" condenser! 20 Milli!)
(Book Antiqua 20 points)
(Bookman Old Style 20 pts)
(Britannic Bold 20 points)
(Californian FB 20 points)
(Calisto MT 20 points)

(Centaur 20 points)

(Century 20 points)

(Century School Book 20 pts)

(Comic Sans MS 20 points)
(Cooper Black zo points)
(COPPERPLATE GOTHIC BO)

102

CHAMPION (COPPERPLATE GOTHIC LT)
Champion (Courier New 20 points)

Champion (Elephant 20 points)
CHAMPION (EN GRAVERS MT 20)
Champion (Eras Bold ITC)

Champion (Eras Medium lTC 20 points)
Champion (Eras Demi ITC 20 points)

Champion (Eras Light ITC 20 points)
Champion (Franklin Gothic Book 20 points)
Champion (Franklin Gothic Demi 20 points)
Champion (Franklin Gothic Demi Condensded)
Champion (Franklin Gothic Heavy 20 pts)
Champion (Franklin Gothic Medium 20 pts)
Champion (Franklin Gothic Medium Condensed)
Champion (Garamound 20 points)

Champion (Georgia 20 points)

Champion (Gill Sans MT 20 points)

Champion (Gill Sans Condensed 20 points)

(infill (Glimlllnlnimeilinifl)

Champion (Gill Sans Illtra Bold)
Champion (Gill Sans Ultra Bold Condensed)

Champion (Glouster MT ExhaCondensed 20 points)
Champion (Goudy Old Style 20 points)
CHAMPION ( GOUDY ST)
W Wfllfllfl

Champion (Helvetica 20 points)
Champion (Helvetica Narrow 20 points)

103

Champion
Champion
Champion
Champion
Champion
Champion
lilillllllioll

C H AM P IO N
Champion
Champion
Champion
Champion
Champion

Champion
Champion
Champion
Champion
Champion
Champion
Champion
ilnmpinn
Champion
Champion
Champion
(la-pin
Champion

(Helvetica Black 20 pts)
(Helvetica Condensed 20 points)
(Helvetica Condensed black 20 pt)
(Helvetica Condensed Light 20 pts)
(Helvetica Light 20 points)
(High Tower 20 points)

limIIacl 20 Mints]

(LITHOS REGULAR 20 PTS)

(Lucida Bright 20 points)
(Lucida Console 20 pts)

(Lucida Fax 20 points)
(Lucida Sans 20 points)
(Lucida Sans Typ wrt)

(Lucida Sans Unicode 20 pt)
(Minion Condensed 20 points)
(Modern # 20 20 points)

(MS Mincho 20 points)

(Myriad Roman 20 points)

(Myriad Tilt 20 points)

(News Gothic MT 20 points)
[Niaum Solid le mints]

(Nueva Bold Extended 20 p
Nueva Roman 20 points)

(OCR A Extended 30 pts)
(MI 20 pills)

(Fapgrus 20 points)

104

Champion
CHAMPION
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion

Champion

 

CHAMPION
Champion

Champion
cnrmmon

Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion
Champion

(Perpetua 20 points)

(PERPETUA TILTING MT 20)
(Mill 20 point)

(PmingLiU 20 points)

(Poor Richard 90 points)
(Rockwell 20 points)

(Rockwell Condensed 20 points)

(Rockwell Extra Bold 20
(Sanvito Light 20 points)
(Sanvito Roman 20 points)

(Swim .Ma‘ 93am 20 painto)

(snowman con-Inc 20 m)

(SimSun 20 points)

(Snap ITC 20 Points)

(STENCIL 20 P0INTS)

(T ahoma 20 points)

(Tekto MM 20 points)

(lekto MM_lOO LT250 on 20 points)

(TektoMlelOOLT 564 No 20 pointe)

(TektolvlleiOO LT 850 EX 20

(Tekto MM_24O KG 250 CN 20 points)

(Tekto MM_24O KG 564 NO 20 pta

(Tekto MM_24O KG 550 EX)

(Tekto MM_505 ED 250 CN 20 points)

(Tekto MM_503 ED 455 N0 20pt

(Tekto MM_503 ED 550
EX 20 points)

105

Champion (Tempus Sans lTC 20 Points)

Champion (Times New Roman 20 points)
Champion (Trebuchet MS 20 points)
Champion (Tw Cen MT 20 points)

Champion (Tw (en MT Condensed 20 points)

Champion (Tw Con M'i Condensed Extra Bold 20)

Champion (Utopia 20 points)
Champion (Verdana 20 points)
Champion (Verdana Ref 20 points)
Champion Mva Bold

Extra lE-Xlt )
(Champion (Vin/a1 Regular 2(0) points)

6’hampzan (071°27ch 20 pom)
draw fl/am 'SM' 20 PM /
Bhompim {lLIeatminster an prints}

ChamPion (Wide Latinzo
(lllllll dillntln‘nil

106

Appendix 3- Legibility and Color Contrast

107

The following represents the analysis of an ongoing study directed by Dr. Hugh

Lockhart. The experiment investigates the effect of varying color contrast combinations

on legibility. The experimental design, data collection and a preliminary analysis of the

results of this study were conducted in 1996. A more thorough treatment of the 1996

data is presented here.

36 cards (6 messages x 6 color combinations, or contrasts) were created in order

to examine how different contrast combinations affect the legibility of messages. 6

messages (see Table 10) were centered on cards approximately 3” x 5” cards; text was

 

 

 

 

 

 

 

 

justified.
Table 10- Legibility Messages Used in the Color Contrast Study

Message 1 It may help most of them to work today. She works in this club
after midnight. The order to go will be done after two.

Message 2 She works in this club after midnight. The order to go will be
done after two. There will be some sugar in the kitchen.

Message 3 The order to go will be done after two. There will be some sugar
in the kitchen. Here is a copy of lunch hours for today.

Message 4 There will be some sugar in the kitchen. Here is a copy of lunch
hours for today. From here to there flowers can not grow.

Message 5 Here is a copy of lunch hours for today. From here to there
flowers can not ggow. It may help most of them to work today.

Message 6 From here to there flowers can not grow. It may help most of

 

them to work today. She works in this club after midnight.

 

Each message was printed in 6 color combinations (see Table 11) for a total of 36

treatments (see Table 11).

108

 

 

Table 11- Contrast Treatments

 

 

 

 

 

 

 

Color 1 Blue text/white background
Color 2 Yellow text/red backgron
Color 3 Blue text/yellow backgron
Color 4 White text/blue background
Color 5 Black text/red background
Color 6 Black text/white backgron

 

 

 

Table 12- Treatment Combinations for the Color Contrast Study

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Card # Color # Message # Card # Color # Message #
1 1/ l 19 4/1
2 1/2 20 4/2
3 1/3 21 4/3
4 1/4 22 4/4
5 1/5 23 4/5
6 1/6 24 4/6
7 2/1 25 5/1
8 2/2 26 5/2
9 2/3 27 5/3
10 2/4 28 5/4
11 2/5 29 5/5
12 2/6 30 5/6
13 3/1 31 6/1

 

 

 

 

109

 

 

 

Table 12- Treatment Combinations for the Color Contrast Study (Continued)

 

 

 

 

 

 

14 3/2 32 6/2
15 3/3 33 6/3
16 3/4 34 6/4
17 3/5 35 6/5
18 3/6 36 6/6

 

 

 

 

Near point visual acuity was tested using a Dow Corning Opthalmics Near Point
Visual Acuity Card. These values were recorded. Subjects that had visual acuities of
20/20 were coded as “1”, subjects with measured values of 20/30 “2”, 20/40 “3” and so
on. Subjects were also tested for color blindness; results were recorded as “normal” or
“red/ green color blind”. Other information that subjects provided included age group,
gender, eye wear and highest level of education completed. Eye wear was coded as 1
through 4 for the purpose of statistical analysis. Eye wear of 1 indicated that subjects did
not wear use any kind of correction, 2 indicated that they wore glasses with a single lens,
3 indicated bifocals and 4 trifocals.

Researchers asked 6 age groups (19-28, 29-38, 39-48, 51-60, 61-70, and 71 and
older) to read 12 cards (a third of the total treatments) using the polariscope. Subjects
were first asked to rotate the polariscope’s filter until the first point that they could read
the message. Researchers recorded this number. Subjects were then asked to continue
rotating until “the first point that they could easily read the words on the card without
straining their eyes”. Researchers also recorded this number. The results from the second
data set, where subjects rotated the filter until the first point that they could read the cards

without straining their eyes, are presented here.

110

 

Although only 6 age groups were tested, data presented here shows a total of 8
groups. The reason for this is two-fold. The first age group (age 19-28) contained three
times as many subjects as any other age group; a total of 45 people were tested that were
19-28, while only 15 were tested in the other five age groups. As a result of the large
number of subjects age 19-28, 3 different operators were used to collect this group’s data;
it was determined that data collected by each operator would be reported as a separate
group so that the effect of operator could be examined. As a result, data reported as
groups 13 represent readings from people age 19-28 collected by three different
operators; data reported as group four represent readings made by people age 29-38, data
reported as group five represent readings made by people age 39-48, data reported as
group six represent readings made by people age 51-60, data reported as group seven
represent readings made by people age 61-70 and data reported as group eight represent
readings made by people age 71 and older.

An analysis was conducted on groups 1-3 (all age 19—28) using the restricted form
of the mixed model. Operator, visual acuity, eye wear, gender and color were tested as
fixed effects while message, subject (nested within operator), and the interaction of
subject (nested within operator) and color were treated as random effects. Using SAS 8.1

the following results were obtained (see Table 13).

111

 

Table 13- Color Contrast Results: Examining Groups 1-3 for an Effect of Operator

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fixed Effects
Effect Degrees of Degrees of F -Value PR>F
Freedom Freedom
Operator 2 38 2.25 .1189
Visual Acuity 2 38 5.38 .0088
Eye Wear 1 38 .54 .4649
Gender 1 38 .26 .6155
Color 5 220 81.53 (.0001
Random Effects
Covariance Stnd. Estimate Z Error Value PRZ
Parameters
Message 1.19 E-18
Subject 8.1749 1.9729 4.14 (.0001
(Operator)
Residual 2.9065 .2502 11.62 (.0001

 

 

 

 

 

Bolded effects indicate a high level of statistical significance (a= .01 ).

Once it was determined that the effect of operator was not significant, the entire
data set was analyzed for an effect of Age Group. Data from all three of the groups was
used in the second analysis. Leaving in the first three groups gives a conservative
estimate of the effect of age. Researchers first attempted to use the following model
Response = Age Group + Visual Acuity + Eye Wear + Gender + Color+
Age Group*Color + Subject (Nested within Age Group) + Color‘Subject (Nested within
Age Group) + Message*Subject (Nested within Age Group) + Color*Message*Subject
(Nested within Age Group) + Color*Message + Age*Message + Age
Group*Color*Message + Residual

Age group, visual acuity, eyewear, gender, color and age*color were treated as

fixed effects while subject (nested within age group), color*subject (nested within age

112

 

group), message*subject (nested within age group), color*message* subject (nested

within age grouP), color*message, age groupflnessage and age group‘color*message and

residual were all treated as random effects. Due to the large number of interaction terms

and random effects, the computer did only a partial analysis of the model. The results

that were completed follow (see Table 14)

 

Table 14: Partial Analysis of the Complete Data Set

 

 

 

 

 

 

 

 

 

 

 

 

Random Effects
Covariance Estimate Error Value PRZ
Parameters
Message .09980 .1200 .83 .2029
Subject (Age) 66.3943 9.3353 7.11 (.0001
Subject (Age) 1 . I 621 5. 730 2.03 .0213
* Color
Subject (Age) 0
* Message
Subject (Age) * -n
,, Color * 6.91 10
Message
Color .08067 .1340 .60 .2735
* Message
Age Group .07609 .1472 .52 .3026
" Message
Age Group
* Color * .2877 .3685 .78 .2175
Message
Residual 12.2947 .7481 16.43 (.0001

 

 

 

 

 

Bolded effects are highly significant (a=.01). Effects that appear in italicized typeface

are moderately significant (a=.05).

In the third analysis a simplified model was used so that a complete analysis could

be obtained (see Table 15). The simplified model used was:

113

 

Response = Age Group + Visual Acuity + Eye Wear + Gender + Color+
Age Group‘Color + Message + Subject (Nested within Age Group) + Subject (Nested
within Age group)*Color + Residual

 

Table 15: Analysis of the Complete Data Set (Simplified Model)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fixed Effects
Effect Degrees of Degrees of F Value Pr > F
Freedom Freedom
Age Group 7 105 5.70 (.0001
Visual Acuity 3 105 21.13 (.0001
Eye Wear 3 105 1.17 .3236
Gender 1 105 .01 .9274
Color 5 561 177.28 (.0001
Age Group 35 559 6.57 (.0001
* Color
Random Effects
Covariance Estimate Error Value PrZ
Parameters
Message .1286 .1166 1.10 .1351
Subject (Age) 66.520 9.3501 7.11 (.0001
Subject (Age) .9647 .5555 1. 74 .0412
* Color
Residual 12.813 .6764 18.94 (.0001

 

 

 

 

 

 

Effects that appear in bolded type are highly significant (a=.01) while those that appear
in italicized type are moderately significant (a=.05). The efl‘ect of both color and age can
be examined visually when the data is broken into bar graphs (see Figure 41 for the
Effect of Age and Figure 42 for the Effect of Color). Every age group found black type
on white paper the easiest to read, and all but the two oldest groups (group 7, age 61-70,
and group 8, age 71 and older) found black type on a red background to be the most
difficult combination to read. The two oldest groups found yellow text on a red
background the most difficult to read, possibly because of the physiological changes
discussed earlier in this document combined with poor contrast provided by this

combination.

114

 

I Black
White

'3 Blue
Yellow

lWhite
Blue

Blue
White

IYellow
Red

. Black
Red

FIGURE 41- AVERAGE LEGIBILITY INDEX BY AGE GROUP (COLOR
CONTRAST STUDY)

Figure 41- Average Legiblity index by Age
Group

 

 

 

 
    

‘5 \A

 

 

  
 

«(a I

 

 

. 4” I

 

 

Average Legibility Index

  

‘

5»'>‘\‘\‘»\‘)\ S \S {A
'22)).')))i’22))?»”2222?))))))))))))))))))))))))”2222)”?IW222).’2.’23)).22).:

.”,3;))))))))))S))))§)»)))§»»»»));)))))‘j‘))3-,y.y;))\‘...1
- .91? .44 “adult 1' .‘l. ..hx' -

«: («<«<2«<«««««««««<<<a:««

‘ ‘ ..
W/dd’flflt’flfi’lt

 

 

Group4 Groups Groups

Gnome

 

 

 

 

 

 

Figure 42- Average Legibility Index
.Group 1

 

 

by Color Treatmgent

 

 

 

 

 

 

'3 Group 2
‘I Group 3

 

 

 

v
i..

  
     

‘. ~ ~ D. s s '-

Java-"9"!”
...........................................
...........................................

. wan,a./.rwaauriru

a ,s‘.~.‘...---s‘-~‘~ s‘s
........-........................--.-u.......................
......u-u......-.....................o.-......................

—
///////// I
"~’2‘~’1'~'1.‘ bi"; .

~.,~ -._~.\.'4.‘~. I. ~;\

"H’l’v'u’r’r' “'4'."
..no-n-c-oo-lo-oenoo-u
Itlgnlllllillolllciil.

—
’..////////

 

 

White
Blue

Black Red Black
White

Blue
Yellow

Yellow
Red

Blue
White

   

- Group 4
. Group 5
m Group 7

I Group 8

 

 

 

FIGURE 42- AVERAGE LEGIBILITY INDEX BY COLOR TREATMENT

The statistical implications of the Age/Color interactions were examined by

 

performing Tukey-Kramer test, and an analysis was performed on all possible pairs of

age*color to see which pairs had significant differences. Pair-wise comparison of each

115

possible combination of age and color were tested for significance using a Tukey-Kramer

test. Results of the pair-wise comparisons within each age group are reported below (see

Table 16).

Using the first pair wise comparison to illustrate what this means, when the results

of color 1 (blue text on white background) and color 4 (white text on blue background)

(both observed by 19-28 year olds in group 1) are compared an adjusted p of1.00 is

obtained. This indicates that there is no statistical difference between these two color

samples for this group.

 

Table 16: Tukey-Kramer Pair Wise Comparisons (of contrast) within Age Group For

 

 

 

 

 

 

 

 

 

 

Significance
. . . . Adjusted P Significant
Age Group Color Combmatron Color Combmatron Value Difference?
Blue Text/ Yellow Text!
19-28 (1) White Red Background .9983 No
Backggmdfl) (2)
Blue Text/ Blue Text/
19-28 (1) White Yellow 1.00 No
Background(l) Background (3)
Blue Text/ White Text/
19-28 (1) White Blue 1.00 No
Background(l) Background(4)
Blue Text/ Black Text/
19-28 (1) White Red Background .5027 No
Backgroundfl) (5)
Blue Text/ Black Text/ 9616
19-28 (1) White White Background ' No
Background(l) (6)
Yellow Text/ Blue Text/
19-28 (1) Red Background Yellow .1508 No
(2) Background (3)
Yellow Text/ White Text/
19-28 (1) Red Background Blue Background .3675 No
(2) (4)
Yellow Text/ Black Text/
1928 (1) Red Background Red Background 1.00 No
(2) (5)

 

 

 

 

 

116

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . Adjusted P Significant
Age Group Color Combmatron Color Combmatron Value Difference?
Yellow Text/ Black Text/
19-28 (1) Red Background White .0100 YES
(2) Background (6)
Blue Text/ White Text/
19-28 (1) Yellow Background Blue Background 1.00 No
(3) (‘1)
Blue Text/ Black Text/
19-28 (1) Yellow Red Background .0024 YES
Background (3) (5)
Blue Text/ Black Text/
19-28 (1) Yellow Background White Background 1.00 No
(3) L6)
White Text! Black Text/
19-28 (1) Blue Background Red Background .0112 YES
14) 15)
White Text/ Black Text/
19-28 (1) Blue Background White Background 1.00 No
(4) (6)
Black Text! Black Text/
19-28 (1) Red Background White (.0001 YES
(5) Background (6)
Blue Text/ Yellow Text/
19—28 (2) White Red Background .9185 No
Backgundfl) (2)
Blue Text/ Blue Text/
19-28 (2) White Yellow 1.00 No
Background(1) Background (3)
Blue Text/ White Text/
19-28(2) White Blue Background 1.00 No
Backgroundfl) (4)
Blue Text/ Black Text/
19-28 (2) White Red Background .3745 No
Background(1) (5)
Blue Text/ Black Text/
19-28 (2) White White Background 1.000 No
Backgroundfl) (6L
Yellow Text/ Blue Text/
19-28 (2) Red Background Yellow .0949 No
(2) Background (3)

 

 

 

 

 

117

 

 

Table l6- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . Adjusted P Significant
Age Group Color Combmatron Color Combmatron Value Difference?
Yellow Text/ White Text/
19-28 (2) Red Background Blue Background .3500 No
(2) (4)
Yellow Text/ Black Text/
19-28 (2) Red Background Red Background 1.000 No
(2) (5)
Yellow Text/ Black Text/
19-28 (2) Red Background White Background .1023 No
(2) (6)
Blue Text/ White Text/
19-28 (2) Yellow Background Blue Background 1.00 No
(3) (4)
Blue Text/ Black Text/
19-28 (2) Yellow Red Background .0061 YES
Background (3) (5)
Blue Text/ Black Text/
19-28 (2) Yellow Background White Background 1.00 No
(3) (6)
White T ext/ Black T ext/
I 9-28 (2) Blue Background Red Background .0413 Moderate
( 4) (5)
White Text/ Black Text/
19-28 (2) Blue Background White Background 1.00 No
(4) (6)
Black Text/ Black Text/
19-28 (2) Red Background White .0067 YES
(5) Background (6)
Blue Text/ Yellow Text]
19-28 (3) White Red Background .7445 No
Background(1) (2)
Blue Text/ Blue Text/
19-28 (3) White Yellow 1.00 No
Background(l) Background (3)
Blue Text] White Text/
19-28 (3) White Blue Background 1.00 No
Background(l) (4)
Blue Text/ Black Text/
19-28 (3) White Red Background .1082 No
Background(1) (5)
Blue Text/ Black Text/
19—28 (3) White White Background .9994 No
Background(1) (6)

 

 

 

 

 

118

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . Adjusted P Significant
Age Group Color Combmatron Color Combmatron Value Difi‘erence?
Yellow Text/ Blue T ext/
19.28 (3) Red Background (2) Yellow Background '02 74 Moderate
(3)
Yellow Text/ White Text/
19-28 (3) Red Background Blue Background .0697 No
(2) (4)
Yellow Text/ Black Text/
19-28 (3) Red Background Red Background 1.00 No
12) 15)
Yellow Text] Black Text/
19-28 (3) Red Background White .0023 YES
(2) Background (6)
Blue Text/ White Text/
19-28 (3) Yellow Background Blue Background 1.00 No
(3) (4)
Blue Text/ Black Text]
19-28 (3) Yellow Red Background .0005 YES
Background (3) (5)
Blue Text/ Black Text/
19—28 (3) Yellow Background White Background 1.00 No
(3) £61
White Text] Black Text/
19-28 (3) Blue Background Red Background .0017 YES
(4) (5)
White Text/ Black Text/
19-28 (3) Blue Background White Background 1.00 No
(4) (61
Black Text/ Black Text/
19-28 (3) Red Background White (.0001 YES
(5) Background (6)
Blue Text/ Yellow Text/
29-38 (4) White Red Background 1.00 No
Background(1) (2)
Blue Text! Blue Text/
29-38 (4) White Yellow .9763 No
Background(1) Background (3)

 

 

 

 

 

119

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age Group Color Combination Color Combination Adillfltjg P 31322322?)
Blue Text/ White Text/
29-38 (4) White Blue Background 1.00 No
Backgroundfl) (4)
Blue Text/ Black Text/
29-38 (4) White Red Background .9828 No
Background(l) (5)
Blue Text/ Black Text/
White White Background .9145
29-38 (4) Background(l) (6) No
Blue Text/
2933 (4) Red’gélé’kzgz’; (2) Yellow B(a;jkground .1512 No
Yellow Text/ White Text/
29-38 (4) Red Background Blue Background .5846 No
Q) (4)
Yellow Text/ Black Text/
29-38 (4) Red Background Red Background 1.00 No
(2) (5)
Yellow Text/ Black Textl
29-38 (4) Red Background White .0757 No
(2) Background (6)
Blue Text/ White Text/
29—38 (4) Yellow Background Blue Background 1.00 No
(3) (42
Blue Textl Black Text/
29-38 (4) Yellow Red Background .0045 YES
Background (3) (5)
Blue Text/ Black Text/
29-38 (4) Yellow Background White Background 1.00 No
(3) (6)
White Text/ Black Text/
29-38 (4) Blue Background Red Background .05 I 4 Moderate
(4) (5)
White Text/ Black Text/
29-38 (4) Blue Background White Background 1.00 No
Bla k4')l‘ t/ Bla k6')l‘ t/
c ex c ex
”'38 (4) Red Background White '00" YES
(5) Background (6)

 

120

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . Adjusted P Significant
Age Group Color Combmatron Color Combination Value Difference?
Blue Text/ Yellow Text/
39-48 (5) White Red Background 1.00 No
Background(l) (2)
Blue Text/ Blue Text/
39-48 (5) White Yellow .4517 No
Backgroundfl) Background (3)
Blue Text/ White Text!
39-48 (5) White Blue Background .5498 No
Background(1) (4)
Blue Text/ Black Text/
39-48 (5) White Red Background .2499 No
Backgroundfl) (5)
Blue Text/ Black Text/
39-48 (5) White White Background .2297 No
Background(l) (6)
Yellow T ext/ Blue T m/
39-48 (5) Red Background Yellow .0043 YES
(2) 3005me (3)
Yellow Text/ White Text/
39-48 (5) Red Background Blue Background .0070 YES
(2) (4)
Yellow Text/ Black Text/
39-48 (5) Red Background Red Background .9983 No
(2) (5)
Yellow Text/ Black Text/
39-48 (5) Red Background White .0011 YES
(2) Background (6)
Blue Text/ White Text/
39-48 (5) Yellow Background Blue Background 1.00 No
(3) (4)
Blue Text/ Black Text/
3948 (5) Yellow Red Background (.0001 YES
Background (3) (5)
Blue Text/ Black Text/
39-48 (5) Yellow Background White Background 1.00 No
(31 (92
White Text/ Black Tcxt/
39-48 (5) Blue Background Red Background (.0001 YES
(4) (5)

 

 

 

 

 

121

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . Adjusted P Significant
Age Group Color Combmatron Color Combmatron Value Difference?
White Text/ Black Text/
3948 (5) Blue Background White Background 1.00 No
(4) (6)
Black Text/ Black Text]
3948 (5) Red Background White (.0001 YES
(5) Background (6)
Blue Text/ Yellow Text/
51-60 (6) White Red Background (.0001 YES
Background(l) 42)
Blue Text/ Blue Text/
51-60 (6) White Yellow .3835 No
Backgroundfl) Background (3)
Blue Text/ White Text/
51-60 (6) White Blue Background .9699 No
Background(l) (4)
Blue Text/ Black Text/
51-60 (6) White Red Background (.0001 YES
Backgroundfl) (5)
Blue Textl Black Text/
51-60 (6) White White .0110 YES
Background(1) Background (6)
Yellow Text/ Blue Text/
51-60 (6) Red Background Yellow (.0001 YES
(2) Backgpund 13)
Yellow Text/ White Text/
51-60 (6) Red Background Blue Background (.0001 YES
(2) (4)
Yellow Text/ Black Text/
51-60 (6) Red Background Red Background .5277 No
(2) (5)
Yellow Text/ Black Text/
51-60 (6) Red Background White (.0001 YES
(2) Background (6)
Blue Text/ White Text/
51-60 (6) Yellow Background Blue Backgron 1.00 No
(3) (4)
Blue Text/ Black Text/
51-60 (6) Yellow Red Background (.0001 YES
Background (3) (5)
Blue Text/ Black Text/
51-60 (6) Yellow Background White Background 1.00 No
(3) (6)

 

 

 

 

 

122

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age Group Color Combination Color Combination Adi/1281:: P 3%223222
White Text/ Black Text/
51-60 (6) Blue Background Red Background <.0001 YES
(4) (5)
White Text/ Black Text/
51-60 (6) Blue Background White Background .9979 No
(4) (6)
Black Text/ Black Text/
51-60 (6) Red Background White <.0001 YES
(5) Background (6)
Blue Text/ Yellow T ext/
61- 70 (7) White Red Background .0582 Moderate
Background(1) (2)
Blue Text/ Blue Text/
61-70 (7) White Yellow .0193 YES
Backgroundfl) Background (3)
Blue Text/ White Text/
61-70 (7) White Blue Background .2386 No
Baclgroundfl) (4)
Blue Text/ Black Text/
61-70 (7) White Red Background .9996 No
Background(l) (5)
Blue Text/ Black Text/
61-70 (7) White White .0003 YES
Background(1) Background (6)
Yellow Text/ Blue Text/
61-70 (7) Red Background Yellow <.0001 YES
(2) Background (3)
Yellow Text/ White Text/
61-70 (7) Red Background Blue Background <.0001 YES
(2) (4)
Yellow Text/ Black Text/
61-70 (7) Red Background Red Background .9973 No
:2) (5)
Yellow Text/ Black Text/
61-70 (7) Red Background White <.0001 YES
(2) Background (6)
Blue Text/ White Text/
61 -70 (7) Yellow Background Blue Background 1.00 No
(3) (4)

 

 

 

 

 

123

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . . Adjusted P Significant
Age Group Color Combmatron Color Combination Value Difference?
Blue Text! Black Text!
61-70 (7) Yellow Red Background <.0001 YES
Background (3) (5)
Blue Text! Black Text!
61-70 (7) Yellow Background White Background 1.00 No
(31 (6)
White Text! Black Text/
61-70 (7) Blue Background Red Background .0001 YES
(4) (5)
White Text! Black Text!
61-70 (7) Blue Background White Background 1.00 No
(4) (6)
Black Text! Black Text!
61-70 (7) Red Background White <.0001 YES
(5) Background (6)
Blue Text! Yellow Text!
71+ (8) White Red Background <.0001 YES
Background(l) (2)
Blue Text/ Blue Text!
71+ (8) White Yellow ‘ <.0001 YES
Background(1) Background (3)
Blue Text/ White Text!
71+ (8) White Blue Background <.0001 YES
Background(l) (4)
Blue Text! Black Text/
71+ (8) White Red Background <.0001 YES
Background(l) (5)
Blue Text/ Black Text!
71+ (8) White White <.0001 YES
Bacflroundu) Background (6)
Yellow Text/ Blue Text!
71+ (8) Red Background Yellow <.0001 YES
(2) Background (3)
Yellow Text! White Text!
71+ (8) Red Background Blue Background <.0001 YES
(2) (4)
Yellow Text! Black Text!
71+ (8) Red Background Red Background 1.00 No
(2) (5)

 

 

 

 

 

124

 

 

Table 16- Continued

 

 

 

 

 

 

 

 

 

 

 

Age Group Color Combination Color Combination Adjusts: P Sifigfifig
Yellow Text! Black Text/
71+ (8) Red Background White <.0001 YES
(2) Background (6)
Blue Text! White Text!
71+ (8) Yellow Background Blue Background 1.00 No
(3) (4)
Blue Text! Black Text!
71+ (8) Yellow Red Background <.0001 YES
Baclground (3) (5)
Blue Text/ Black Text!
71+ (8) Yellow Background White Background 1.00 No
(3) (6)
White Text! Black Text!
71+ (8) Blue Background Red Background <.0001 YES
(4) (5)
White Text! Black Text/
71+ (8) Blue Background White Background 1.00 No
(4) (61
Black Text! Black Text!
71+ (8) Red Background White <.0001 YES
(5) Background (6)

 

 

 

125

 

Appendix 4- Primary Study Labels

126

Drug Facts
Active Ingredients (in each 5 ml)

 

 

 

Bromgheoiamioe maleate 2 mg Aolihistamioe
llertromcthorohan llllr II) on tough suggressaol
Pseudoeuhedrine llll iii) no Ilasal decongestant
Use Iemgorarily relieves:

0 sheeting 0 runny nose 0 nasal congestion 0 cough
Warnings

do out use iI you are now taking a urescriution monoamioe oridase inhibitor
(llAlll) (certain drugs in degressioo. usychratric or emotional conditions or
Parkinson's disease), or lot 2 weeks alter slugging the MAIN drug. llyou do
not know it your gresctigtton drug contains an llAlil. ask a doctor or
ghaooacist belore taking this groduct

Ask a doctor belore use it you have 0 diabetes 0 glaucoma

0 thyroid disease 0 much that occurs with too much ghlegm (mucus)
0 trouble urinating due to an enlarged otustate gland 0 bean disease
0 a breathing groblem or chronic cough that lasts or as occurs with smoking

asthma. chronic bronchitis or emphysema '

Ask a doctor or gharmacist belore use il you are taking sedatives or traoguilirers

When using this groduct 0 do not use more than directed
0 drowsiness may occur 0 avoid alcoholic drinks

0 alcohol sedatives aod traogurlitets may increase drowsiness

0 be carelul when driving a motor vehicle or ogetating machinery
0 excitability may occur, esuecrally in children

Slog use and ask a doctor ii 0 you get nervous dizzy, or sleegless

0 cough lasts more than Idavs comes back or occurs with lever, rash or headache
that lasts. Ihese could he srgos ol a serious condition.

0 symgroms do out get benet within 1 days til occur with a lever

ll gregnant or breastleediog ask a health orolessional belore use.
leeg out ol reach ol children. In case ol overdose. get medical helg or contact a
Poison [ontrol [enter right away.

 

 

Directions 0 take every 4 to 6 hours; out more than 4 doses in 24 hours
_12 years god over it) ml

dtollyears __-.-_ _ _Eiml _._ ___. 9

goder byears ) askadoctor

 

 

Inactive Ingredients citric acid lddl blue ll. glycerin umvlleoe glycol.
’ ourtlied water. sarchario sodium, sodium beoruate. sorbitol

 

. log I ghases esrage out go oil gone

 

ill right sin tool; --__ _ i i ii

 

Compliant Label
Drug Message
Univers Ultra Condensed

 

 

(Body text is 9.0 points)

' GoIprpIe

deluxe destination (it word 5 in)

 

 

 

fliscrrmioalioobrrdgesiol Meteorologist

Polycoodeosatioo Alld II) is [ouch collections

llosuccesslully Hlilllillar lhinkgarticularly
" Ndl lemgerament gorgeous:

0 deathbed 0 badly rose 0 count conscience 0 baggy

Sgleodor

lt are buy is wow lar (or length a unsuccesslul hangings: ouietly gessimisl
(PddH) Iuetlecl truly mom moderation intertuuled by increment limitation or
Progagaoda hormone), it mag 2 shade slows eleuhaot the PllllH tree. lo cal so
gem teed to (our luodameotals bear chicken on klAllI, too I lawlul no
dictionary animal gaiama hum nobody

o gharmacy 0 maniage
o gutting disease 0 cough that royally legless sale bed gage easier (being)
- mrdwrle criticism (in an in weakness absolute right 0 thumb another

0 I gowerless several by bargain dream what skies or at colors hand singing,

babies billion hislortaos or eguigmeol

Are 3 notice it comguoeots assume war is we who wonder cumgooeot or international

(his desks that reader 0 to log sir mill warg darabank
0 crbersuace cat zebra 0 could ergaosion others

0 medical. increased. mug understanding but sgerilic urogeosity
- be auueou: hank barcode lwbrch several at recycling treatment
0 sigorlrcaoce (or groom auueodices il identity

Were sun son are a larger Ill 0 can hit degrees level, in magazines

0 event louod glut (out 9 race. chain ages to design inks horse. rash, or soerilic
mane sroce. laced sheet be (root in 2 tltrouuh cromgliog.

0 smallest in her get result either 4 cone in (root back I label

1. ll occeuted io suggly-subiects two a gadget reguiremeols comgly reo.
' Your (yo I" white be contrast In home to standard gin clarity some at counter a

Poor desklou Harder words that

y
i EXDEllmElll 0 like house 8 to 2 mouse; wio sign wine l chose it IS until

_ i -“ifQQiiitfiflj _
lolldmake_s____.. ___ j: 5i),
ltghtbllogr,___. -. _. __ _. j ”_binlviston

Averages [lgtbalmics normal loud llildll llies ll, distance, grescribe reduce
subiects which, urovided reader. visual educated, whatever

 

Compliant Label
Nonsense Message
Univers Ultra Condensed

 

 

(Body text is 9.0 points)

 

 

127

- Asit a doctor belore use if you have
- - thyrord disease

Drug Facts

Active ingredients (in each 5 mL)
Brompheniamine maieate 2 mg
Dextromethorphan HBr l0 mg
Pseudoephedrine HCL 30 mg.

.. Antihistamrne
...Cough suppressant
Nasal decongesunt

 

 

~ USE temporarily relieves:

° sneezing - runny nose - nasal congestion

- cough
Warnings

Do not use if you are now taking a prescription monoamine oxidase inhibitor
(MAOI) (certain drugs for depression. psychiatric or emotional condruons. or
Parkinson's disease). or for 2 weeks after stopping the HAOI drug ll you do
not know il your prescription drug contains an MAOI. ask a doctor or
pharmacist belore taking this product

' diabetes ' glaucoma
‘ ° cough that occurs with too much phlegm (mucus)
' trouble urinating due to an enlarged prostate gland - heart disease

. ° a breathing problem or chronic cough that lasts or as occurs with smoking.

asthma. chronic bronchitis. or emphysema

Ask a doctor or pharmacist before use if you are taking sedatives or tranquilizers

When using this product ' do not use more than directed

' drowsiness may occur ' avoid alcoholic drinks

- alcohol. sedatives. and tranquilizers may increase drowsiness

' be careful when drrvrng a motor vehicle or operating machinery
- excitability may occur. especially in children

Stop use and aslt a doctor if ° you get nervous. dizzy. or sleepless

- cough lasts more than 7 days. comes back. or occurs with lever. rash. or headache
that lasts. These could be signs of a serious condition. .

- symptoms do not get better within 7 days or occur with a lever

ll pregnant or breast—feeding. ask a health professional before use.
Keep out of reach oi children. in case oi overdoseget medical help or contact a
Poison Control Center right away.

Directions ' take every 4 to 6 hours; not more than 4 doses in 24 hours

 

127;“. 5213927 _ . . 10 mL _
6 col] years_ _ -_____ __ .5511:—
. 335‘. ““97. .. ,

under 6 years _
inactive Ingredients citric acid. FD8C blue #1. glycerin. propylene glycol.
purified water. saccharin sodium. sodium benzoate. sorbrtol

 

. Golf Apple

Deluxe Destination (it word 5 in)
Discrimination bridges 2 0‘
Polycondensation ANd 10 is...............
Unsuccessfully HOW 30 at

Meteorologist
-..........Couch collections
Think par trcuiariy

Nor temperament gorgeous:
'deathbed °badly rose 'count conscience ~happy
Splendor

. it are buy is wow far for length a unsuccessful happiness quretiy pessimist
(POOH) (perfect truly mom moderation. interrupted by increment limitation. or
Propaganda hormone).it map 2 shade slows elephant the POOH tree. In cat so
gem feed to four fundamentals bear chicken an MAOI. ton l lawlui no

dictionary animal pajama from nobody

 

Top 4 phases escape out go oil gone ° pharmacy 0 marriage
-putting disease .cough that royally legless sale bed page easier (being)
-midwile criticism tin an in weakness absolute right 'thumb another

l -7 powerless several by bargain dream what skies or at colors hand singing.
blbleS. billion historians. or equipment

Are a notice if components assume wax is cue who wonder component or international

This desks that reader °to top six mill warp databanlt
'cyberspace cat zebra °couid expansion others

-medical. increased. mop understanding bat specific propensity
°be aqueous bank barcode l which several at recycling treatment
’Signlflcancc lor group. appendices if identify

'i Were sun son are a larger in «nan bit degrees. level. in magazines
‘ «event found plot lour 9 cana.chain ages, in design inks horse. rash. or specific
‘ mane since. Laced sheet be lront in 2 through crumpling.

'smailest to bet gel result either 4 cone in front back I label

ll accepted in supplyosubjectsxwo a gadget requirements comply zen.
Your fin in white be contrast. in home to standardgin clarity some at counter a
Prior Desktop Harder words that

 

Experiment alike house 8 to 2 mouse; win sign wine 4 chose it 45 until
. 0| right finTQEjW—flé IT
:iOl i9makeg- —_--'--.+— S W...

l't'". 91‘2“ " barrage."-

.. -.l---._ - _

Averages Opthalmics normal lood.WH&O llies #i.distance. prescribe reduce.
subjects which. provided reader. visual educated. whateyer

 

 

Compliant Label
Drug Message
Gill Sans
(Body text is 6.0 points)

 

 

 

Compliant Label
Nonsense Message
Gill Sans
(Body text is 6.0 points)

 

 

 

128

Golf Apple

Deluxe Destination (it word 5 in)

 

 

 

Discrimination bridges 2 of - Meteorologist
Polycondcnsation ANd 10 is .. Couch collections
Unsuccessfully HOW 30 at .. Think particularly
Nor temperament gorgeous:

- deathbed . badly rose - count conscience - happy
Splendor .

lt ate buy is wow far for length a unsuccessful happiness quietly pessimist
(POOH) (perfect truly morn moderation. interrupted by increment limitation. or
Prepaganda hormone). it map 2 shade slows elephant the P00" tree. In cat so
gem feed to four fundamentals bear chicken an MAOl. ton l lawful no
dictionary animal pajama from nobody

Top 4 phases escape out go off gone . pharmacy . marriage
- putting disease - cough that royally legless safe bed page easier (being)
-midwife criticism tin an in weakness absolute right - thumb another

- 7 powerless several by bargain dream what skies or at colors hand Singing.
babies. billion historians. or equipment

Ate a notice if components assume wax is cue who wonder component or international

This desks that reader - to top six mill warp databank
- cyberspace cat zebra . could expansion others

- medical. increased. mop understanding bat specific propensity

- be aqueous bank barcode [which several at recycling treatment

- significance for group. appendices if identify

Were sun son are a larger in - man bit degrees. level. in magazines

. event found plot four 9 cane. chain ages. in design inks horse. rash. or specific
mane since. Laced sheet be front in 2 through crumpling.

- smallest to bet gel result either 4 cone in front back I label

lf accepted in supply-subjects. two a gadget requirements comply zen.
Your fin in white be contrast. In home to standard. gin clarity some at counter a
Prior Desktop Harder words that.

Experiment - like house 8 to 2 mouse; win sign wme 4 chose it 45 until
atriiiiiisiiiw’ _ '
l of 19 make

light 6 floor .. v

  
 

' ' bin l vismn

Averages Opthalmics normal food. WHGO flies ll. distance. prescribe reduce.
subjects which. provtded reader. visual educated. whatever

 

Non-compliant Label
Nonsense Message
Lucida Fax
(Body text is 5.5 points)

 

 

 

129

Drug Facts

Active Ingredients (in each 5 mL)

 

 

Brompheniamine maleate 2 mg .. Antihistamine
Dextromethorphan llBr 10 mg ............. Cough suppressant
Pseudoephedrine HCL 30 mg ................ Nasal decongestant

 

Use temporarily relieves:
- sneezing - runny nose - nasal congestion - cough

Warnings

Do not use if you are now taking a prescription monoamine oxidase inhibitor
(MAO!) (certain drugs for depression. psychiatric or emotional conditions. or
Parkinson's disease). or for 2 weeks after stopping the MAOl drug. ll you do
not know if your prescription drug contains an MAOI. ask a doctor or
pharmacist before taking this product

Ask a doctor before use if you have - diabete - glaucoma
- thyroid disease - cough that occurs with too much phlegm (mucus)
- trouble urinating due to an enlarged prostate gland - heart disease

- a breathing problem or chronic cough that lasts or as occurs with smoking.
asthma. chronic bronchitis. or emphysema

Ask a doctor or pharmacist before use if you are taking sedatives or tranquilizers

When usmg this product - do not use more than directed

- drowsiness may occur - avord alcoholic drinks

~ alcohol. sedatives, and tranquilizers may increase drowsiness

- be careful when driving a motor vehicle or operating machinery
- excitability may occur. especially in children

Stop use and ask a doctor if - you get nervous. dizzy. or sleepless

- cough lasts more than 7 days. comes back. or occurs with lever. rash. or headache
that lasts. These could be Signs of a serious condition.

- symptoms do not get better within 7 days or occur with a fever

If pregnant or breast-feeding. ask a health professional before use.
Keep out of reach of children. In case of overdose. get medical help or contact a
Poison Control Center right away.

Directions - take every 4 to 6 hours; not more than 4 doses in 24 hours
1.2" yia’ris ifidfigizsr " L110 Inf-
6t012 years-.-“ ., - ,5 mL-
und_er_6 year. . _ i. ,aslsadpstgt _.

Inactive Ingredients citric acid. FD&C blue ll. glycerin. propylene glycol.
purified water. saccharin sodium. sodium benzoate. sorbitol

 

Non-compliant Label
Drug Message
Lucida Fax
(Body text is 5.5 points)

 

 

 

130

Drug Facts

Active Ingredients (lin each 5 mL)

 

 

Brompheniamine maleate mg ............. Antihistamine
Dextromethorphan HBr 10 mg ................................................ Cough suppressant
Pseudoephedrine HCL 30 mg ............ Nasal decongestant
Use temporarily relieves:

. sneezing . runny nose - nasal congestion - cough
Warnings

Do not use if you are now taking a prescription monoamlne oxidase inhibitor
(MAO!) (certain drugs for depression, psydiiatric or emotional conditions, or
Parkinson's disease), or for 2 weeks after stopping the MAC! drug. If you do
not know if your prescription drug contains an MAO], ask a doctor or
pharmacist before taking this product

Ask a doctor before use if you have 0 diabetes 0 glaucoma ‘

o thyroid disease - cough that occurs with too much phlegm (mucus)

o troUble urinating due to an enlarged prostate gland . heart disease
0 a breathing problem or chronic cough that lasts or as occurs with smoking.
asthma, chronic bronchitis, or emphysema

Ask a doctor or pharmacist before use if you are taking sedatives or tranquilizers

When using this product 0 do not use more than directed

o drowsiness may occur 0 avoid alcoholic drinks

0 alcohol, sedatives, and tranquilizers may increase drowsiness

. be careful when driving a motor vehicle or operating machinery
o excitability may occur, especially in children

Stop use and ask a doctor if 0 you get newous, dizzy, or sleepless

- cough lasts more than 7 days, comes back, or occurs with fever, rash, or headache
that lasts. These could be signs of a serious condition.

0 symptoms do not get better within 7 days or occur with a fever

If pregnant or breast-feeding, ask a health professional before use.
Keep out of reach of children. In case of overdose, get medical help or contact a
Pooson Control Center right away.

Directions 0 take every 4 to 6 hours; not more than 4 doses in 24 hours
12 years and over _1 . _‘ 10 mt f
6t012 years 7_ . _ SmL ,,
under 6 years ask a _doctor _ .

Inactive Ingredients citric acid, FD&C blue # 1, glycerin, propylene glycol,
purified water, saccharin sodium, sodium benzoate. sorbitol

 

Non-compliant Label
Drug Message
Verdana
(Body text is 5.5 points)

 

 

 

131

Golf Apple

Deluxe Destination (it word 5 in)

Discrimination bridges 2 of ................................................... Meteorologist
Polycondensation ANd 10 is.... ......................................... Couch collections
Unsuccessfully HOW 30 at ............................................... Think particularly
Nor temperament gorgeous:

odeathbed obadly rose ocount conscience ohappy
Splendor

it ate buy is wow far for length a unsuccessful happiness quietly pessimist
(POOH) (perfect truly mom moderation, interrupted by increment limitation, or
Propaganda hormone), it map 2 shade slows elephant the POOH tree. In cat so
gem feed to four fundamentals bear chicken an MAOI, ton l lawful no

dictionary animal pajama from nobody

Top 4 phases escape out go off gone . pharmacy 0 marriage
oputting disease ocough that royally iegless safe bed page easier (being)
omidwife criticism tin an in weakness absolute right othumb another

.7 powerless several by bargain dream what skies or at colors hand singing,
babies, billion historians, or equipment

Ate a notice if components assume wax is cue who wonder component or international

This desks that reader oto top six mill warp databank
ocyberspace cat zebra ocould expansion others

omedical, increased, mop understanding bat specific propensity
ebe aqueous bank barcode l which several at recycling treatment
osignit‘icance for group, appendices if identify

Were sun son are a larger in oman bit degrees, level, in magazines
oevent found plot four 9 cane, chain ages, in design inks horse, rash, or specific
mane since. Laced sheet be front in 2 through crumpling.

osmallest to bet gel result either 4 cone in front back i label

If accepted in supply-subjects, two a gadget requirements comply zen.

Your fin in white be contrast. In home to standard, gin clarity some at counter a
Prior Desktop Harder words that.

Experiment 0 like house 8 to 2 mouse; win sign wine 4 chose it 45 until

 
 

61' right‘si‘n leak“ " 13 iT
l of 13 mak_e_ __ 5 iF_
light6 floor bin (vision

Averages Opthalmics normal food, WH&0 flies #1, distance, prescribe reduce,
subjects which, provided reader, visual educated, whatever

 

Non-compliant Label
Nonsense Message
Verdana
(Body text is 5.5 points)

 

 

 

132

Appendix 5- Examples of a Variety of Compliant Label Designs (Drug
Messages)

133

CM
U bol

 

 

 

Drug Facts

Active Ingredients (in each 5 mL)

Bromphcniamlno maloctsz In: Aetihlstamlnc
Dcsuomcthorphsn liar 10 me PIN-2|- suppressant
Psoudocphcdrino HCl. 30 me Nasal docongostant
USO temporarily relieves:

esnoczing orunny nose encscl congestion ecough
Warnings

Donotuscllyouaronowtakingoproscrbtionmonoandnooxldosolnhibitor
(MAOlllcortsindrugsfordoprosslon. psychiatrlcoromotional conditions."
Parkinson' s disease). orfor 2 weeks after stopping the W drug. If you do
notknowifyowpnscrlptlondmgcontalnsanm, askadoctoror
pharmacist before taking this product

Askadoctorboforcusclfyouhavo edlabotos egloocoma
ethyroiddisesss scoughdrotoccurswlthtoomuchphlogmlmucus)
Otroublourinatlngduotoanonlsrgsdprootaugland Ohesrtdisoaso

osbrcathlngproblomorduonlccoughthstlasuorasoccuswlthsmokhg.
asthma, chronlcbronclritlsmr emphysema

Asksdoctororphcnnodstbdonusollyonantaklngsododvcsortranquuiaus

thnuslngthlsproduct-donotuscmorcdrandroctod
edrowsincss may occur eavold alcoholic drinks

alcohol. sedatives. and tronqulliacrsmoyincroosochowslncss
ebocsrefulwhondrivingamotorvchlclooroporatingmachlncry
0 excitability may occur, especially in children

~Stopusoandasksdoctorl eyougotnorvous.dlzzy.orslccplcss
ecoughbstsmondran7dsys.connsbodt.oroccuswlthfcvw.nsh.orhosdsdro
thstlasts. Thosecouldboslgnsofssorlouscondltlon.
esymptomsdonotgotbcttsrwlthln7daysoroccurwlthafovcr

llpngnsntorbnast-bcrlngsskshoaldiprofcsclonclbcforcuso.
Kcopoutofroschofchildron. lncosoofovcrdosc.gotmodlcslhclporcontacta
PoisonControlContorrlglrtmy.

 

 

Directions 'ubmy4b0hownotmonthon4doscsln24hom
Igycorssndovcr lomL

‘Qtetzvsen Sml
undersyoars askadoctor

 

 

Inactive Ingredients cltrlcocld, FDICbluofl. mmgw,
Wurst-r. ssccharlnsodurn. www.mw

 

Compliant Label
Drug Message
Univers Bold

(Body text is 6.0 points)

 

 

 

134

CM
U 55R

Drug Facts

Active Ingredients (in each 5 mL)

 

 

 

Brompheniamine maleato 2 mg ‘ntihistamino
Dextromethorphan HBr 10 mg Cough suppressant
Pseudoephedrine HCL 30 mg Nasal decongestant
Use temporarily relieves:

0 sneezing 0 runny nose 0 nasal congestion 0 cough
Warnings

Do not use if you are now taking a prescription monoamine oxidase inhibitor
(MAOII (certain drugs for depression, psychiatric or emotional conditions, or
Parkinson's disease), or for 2 weeks after stopping the MAOI drug. If you do
not know if your prescription drug contains an MAOI. ask a doctor or
pharmacist before taking this product

Ask a doctor before use if you have 0 diabetes 0 glaucoma
0 thyroid disease - cough that occurs with too much phlegm (mucus)
0 trouble urinating due to an enlarged prostate gland 0 heart disease

0 a breathing problem or chronic cough that lasts or as occurs with smoking,
asthma, chronic bronchitis, or emphysema

Ask a doctor or pharmacist before use if you are taking sedatives or tranquilizers .

When using this product 0 do not use more than directed
0 drowsiness may occur 0 avoid alcoholic drinks
0 alcohol, sedatives, and tranquilizers may increase drowsiness

0 be careful when driving a motor vehicle or'operating machinery

0 excitability may occur, especially in children

Stop use and ask a doctor if 0 you get nervous, dizzy, or sleepless

0 cough lasts more than 7 days, comes beds, or occurs with fever, rash. or headache
that lasts. Thesacould be signs of a serious condition.

0 symptoms do not get better within 7 days or occur with a fever

lf pregnant or breast-feeding, ask a health professional before use.
Keep out of reach of children. In case of overdose, get medical help or contact a
Poison Control Center right away.

 

 

Directions 'tako every 4 to 8 hours: not more than 4 doses in 24 hours
12 yam m

@912, veers 5 till-___

under 6 years ask a doctor

 

 

Inactive Ingredients citric acid. FD&C blue 01. glycerin, prOpylone glycol,
purified water, saccharin sodium. sodium benzoate. sorbitol

 

Compliant Label
Drug Message
Univers 55 Regular
(Body text is 6.0 points)

 

 

 

135

[Ill

Drug Facts UlUlle
Milli Ingledmls in emit 51ml)

Bantam mm 7 gm... wMilitia-t
flatware Hfli iii”_M._-.._-._._..__~.....W was:
Putnam: ill ill I; “..-.--m... .............h We!
littmli the

"Mt 'mm °dmim “all
Walnuts

Diwalmnmtinimtmmmm

Ml tuna lupin him. Man: II mind Will”

Palmslmmmmsmmwmmm
ubulwuummimmallllldamll
washrlhgllwdu

Minimalism 0m Oils:
mum ofihmmmmmm
Omitmlqlsltaahmlmtgi 'Illh
simianiblrtkismimmlamniuh;
«culminates!

Hater-mmumnmmim

Manama: seen-attained
“mum Ouihhtlmi:
Omeletailmefinqnmhm
"cadmium-emulsifier
minimum-m

liquidate-l Opp-militants
owe-nuimmmmmmmw
Ihhiharihmdtarsmfln
”ecumenical-limousine

lpqntuiiwteindakaimdirtn

hedwiefiarhmmhewmmt
MWdMqu

[inches 'tatmlttiietumihliaullim

 

12w“ Ill-l
Wm _ - _... , 5.9L.
rifle: was;
mummdmwimmem
plenum-summits:

 

 

 

 

Compliant Label
Drug Message
Univers Light Ultra Condensed
(Body text is 9.0 points)

 

136

 

CM

 

 

FMed
Drug Facts
Active Ingredients (in each 5 mL)
Brompheniamine rnoleate 2 mg ......... Antihistamine
Dextromethorphon HBr l0 mg ............................................. C h suppressant
Pseudoephedrine HCl. 30 mg Nosa decongestant
Use tempomr'i reli .
'sneezing ly Omrynose Onasdcongestion Occugl'i

Warning? toll idase h bi
Donotuseiyouarenaw ingaprescriptionmonoamineox' ini'tor
(MAOll (certain drugs ior depression, sychiatric or emotional conditions, or
Parkinson's disease), or ior 2 wealts stoppi the MO drug. Ii you do
notknawilmrprescriptiondmgcontoinsan Ol,askadoctoror

pharmacist tollingthisprodud
Askadoctorbeioreuseilyou 0didaetes Ogloucoma
0 thyroid disease 0 cat-d1 Moccurs with toomud'i phlegm lrrtucus)

Otroubleurinati duebcrtenlarged stateglond Oheartdisease
Oobreathing ordironiccougtrlhotlastsorasoccuswimunolting,
osd'imo, chronic bronchitis, or amhysema

Askadocbrorphomudstbeloreuseilyoumbkingsedafivesorhonquilizen

Manusinglhisproduct Odonotusemorelhandirected

0 drowsiness may occur 0 avoid alcoholic drinks

0 alcohol, sedatives and tranquilizers may increase drowsiness

0 be corelul when driving a motor vehicle or operating machinery
0 excitability may occur, especially in children

Stopuseondaslradoctoril 0 nervous dizzy,or s
chghbshmemm7days,mback,y:ruogcdwnwiiilevu,msh,mod~e
lhotlosts. lhese couldbesignsoiaseriouscondition.
'symptomsdonotgstbetterwifltin7daysoroccurwithalever

ll torbreosrleed ,mi health - below so.
tgwsmsmuifi. heinoioverdosamgjmedicaluhelporcontoda
PoisonControlCanterrightaway.

 

 

Directions 'kiteaveryltoohoursmatmoreiianldosesinuhours
JZyeanondw 10ml
goto12years 5ml
underdyeors Jodoctor

 

 

lnactivel redientSciincodd,FD&Cbluoal, ', ,
purifiedwu:rtle.r9socchar1'nsodium,sodiumbenzoate.ggigetdlm”0”anleCOI

 

Compliant Label
Drug Message
Futura Medium
(Body text is 6.0 points)

 

 

 

137

CM
GSL

Drug Facts

Active Ingredients (in each 5 mL)

Brompheniamine maleate 2 mg. . e__ ....... Antihistamine
Dextromethorphan HBr l0 mg -..”... m... Cough suppressant
Pseudoephcdrine HCL 30 mg. ..-.We . . ..,..._..Nasal decongestant

Use temporanly relieves:

- sneezing ° runny nose ° nasal congestion - cough
Warnings

Do not use if you are now taking a prescription monoarnine oxidase inhibitor
(MAOI) (certain (huge for depress-on. psychiatric or emotional conditions or
Parkinson's disease). or for 2 weeks after stopping the MAOl drug. if you do
not know if your prescription drug contains an MAOI. ask a doctor or
pharmacist before talung this product

Ask a doctor before use it you have - diabetes 0 glaucoma
- thymid disease . cough that occurs wrth too much phlegm (mucus)
° truible urinating due to an enlarged prostate gland 0 heart disease

' a breathing problem or chronic cough that lasts or as occin wrth smoking.
asthma. chronic bronchitis. or emphysema

Ask a doctor or pharmacist before use if you are taking sedatives or tranquilizers

VVhenussngthisproduct -donotusemorethmdrected

° drowsuness may occur - avOid alcoholic drinks

° alcohol. sedatives. and tranquilizers may increase drowsiness

0 be careful when driving a motor vehicle or operating machinery
' excitability may occur. especially in children

Stopuseandasliadoctorif smugetmaimarsleepless
-coughhstsmorethan7dayscanesbackoroccurswnhlmrashcrheadache
thatlasts Thesecouldbesgnsofasenouscondition.
~symptomsdonotgetbetterwrttm7daysoroccuthl'iafever

If pregnant or breast-feeding. ask a health Wessional before use.
Keepoutofreacholchildren. hcaseofoverdose.getrriecbcalhelporcoritacta
Paison Control Center right away.

 

Directions ’talceevery4t06h01n;notmorethan4dosesnl4 hours
Myers and Over _lO mL—

6 toll years, _4__ ,__,A5,"‘L_

under_ __6 years_‘_____ askadoctor

 

 

Inactive Ingredients citric acid, FD&C blue “glycerin. propylene glycol.
purified water. sacchann sodium. sodium benzoate. sorbitol

 

Compliant Label
Drug Message
Gill Sans Light
(Body text is 6.0 points)

 

 

 

138

CM
HReg

Drug Facts

Active Ingredients (in each 5 mL)

Brompheniamine maleate 2 mg Anllhlslamlne
Dextromethorphan Hat to mg Cn-gh suppressant
Pseudoephedrine HCL 30 mg Nasal

Usetemporarflyreleves:
'sneezlng -runnynose masalcongestlon ~cough

Warnings
Donotusellyouerenowlsklngspresalpuonmonosmlneoxldasehflbltor
(MAOI) (certaindrugs lordepresslon, psychiatric oremotional conditions. or
Parkinson's disease). orlor2weekssltet stopping the MAOI drug. llyou do
notkmwllyourprescmtlondmgconlalnsanMAOImsksdoctoror
phsnnaclstbeloretaldngthispmduct

Askadoctorbetonusellyouhsve diabetes ~g|sucoma
-lhyroiddisease -cwghlhstoccursw‘nhloomuchphlegn(mucus)
~lroubleuinalingduetoansnlorgedproetaleglmd -heartdisease
~sbreattungpmblemordimniccoughlhsthstsorssocunwlthsnnklng.
unimadvuucbruididgoremphysema

Asksdoctororphumsdstbsbreuseflyouuetaldngsedalhesormimzers

Whenuslngthlsprodud -donotusemrelhmdrected
~drowsinessmayoccur ~avoidslcohollcdrinks
-.alcolnlsedalives.mdlranuilizersmayltmssedrowshess
-becarelulwhendnvingsmvehlclsoropentingmschlnery
- excitability may occur. especially 'Ii children

Stopusssndaskadoctorl -yougstnervous.dzzy.orsleepless
~coughhslsme0nn7deys.cormback.oroccunwldilsva.mh.uheedm
lhallasts. Thesecouldbeslgnsolasenouscondltlon.
-syrnplomsdonotgelbetlsrwltlh7dsysoroccurwlthsfever

NWaWMIMWmm.
Keepwtolrescholchlld'en. lncssoolaverdosemstmedlcalhelporcoructs
PoleonControlCenterrlflilswsy.

Directions Wemy4b6hmmnolnmtlnn4dosssh24hours
,IZVQW

___lQmL_
QQLZJQM A
underayears sskadoctor

Inactive Ingredients citric acid. mac blue 01. glycerin. propylene glycol.
purified water. sscchsln sodium. sodium benzoate. sorbltol

 

 

 

 

 

 

 

Compliant Label

Drug Message
Helvetica Regular

(Body text is 6.0 points)

 

 

 

139

CM
TNR

 

 

 

Drug Facts

Active Ingredients (in each 5 mL)

Bromphenismine maleste 2 mg Antihistsmine
Dextromethorphan HBr l0 mg Cough suppresssnt
Pseudocphcdrinc HCL 30 In; Nssal dccongestsnt
U86 temporarily relieves:

wheezing 'runnynose °msslcongcstion °cough
Warnings

Do not use if you are now taking s prescription monounine oxidase inhibitor
(MAO!) (certain drugs for depression. psychiatric or emotional conditions. or
Parkinson's disease). or for 2 weeks after stopping the MAOI drug. If you do
not know if your prescription drug contains an MAOI. ask s doctor or

pharmacist before taking this product

Ask s doctor before use if you have - diabetes - glaucoma

- thyroid disease - cough that occurs with too much phlegm (mucus)
° trouble urinating due to an enlarged prostate gland - heart discos:

- sheathing problcmorchronic coughthst lutsorssoccurswith smoking.
asthma. chronic bronchitis. or en'physcms

Asksdoctororphunucistbeforeuseifyoumnkhgscduivcsamuilizen

Whenusingthisproduct vdonolusemorethsndirocted

. drowsiness may occur - avoid alcoholic drinks

- alcohol. sedatives. and mnquilizcn may increase chewsincss

r be careful when driving I motor vehicle or operating machinery
' excitability may occur. especially in children

Stopusesndssltsdoctorif ~yougctncrvous.dizzy.orslccplcss
~coughlsstsmorethan?dsys.comcsbsct.oroccmswithlevcr.tulLorhesdsche
that lull Mcouldbcsipis of s serious condition.
oxymptomsdonotgetbettawithin7dsysoroccmwithst‘evcr

lfpregrisntorbrcsst-fccdingsskshcslthprofcssionslbeforem.
chpoutol'reschofchildrcn. lncsssol'overdose.gctmodicslhelporconucts
Poison Control Center right sway.

 

 

Directions ‘tsksevay4todhouunotmthsn4doscsh24hotn
mum 51L.
undcroycsrs uhsdoctor

 

 

Inactive Ingredian citric acid. FNC blue fl. glycerin. propylene glycol.
pmificdwstuawchsrinsodimnnodiunbcnmstsnorbitol

 

Compliant Label
Drug Message
Times New Roman
(Body text is 6.0 points)

 

 

 

140

CM

 

 

 

FBld
Drug Facts
Active Ingredients (in each 5 ml.)
Dexhomthgnhzz‘omoz mg 1"- Amsuppresaant
Pseudoephednne "CI. 30 mg” 5.13“ decongestont
U
0 WWII... 0 nasal congestion 0 cough

Donduuflyoummhkhgoammhnmm

(MOI) (certain buffer deg-essiondasyc ernotronal,eonditions
”my” :........""""' mmmmfl3
pharmacist mop" talclngthi mem '

Ashadoctorheforeuseityouhave 0°diabetes

disease 0 coughthatoccurswithtoornuch Imus)
:troulsie ' dueto taonenlargedprestategland heart
Oahreathingp orchroniccoughthatlastsoraaoccurswithsnrolcing,
asthrna,chronrchronchitis,arensphyserna

Askadactororpharnsadstheioreuuiiyouaretahhsgsedativeswhanquiiurs

Whenu thisproduct Odonotusernorethan
drewsfirnne'ss mayoccur 0 avoid alcoholic drinlcs
0,alcohol sedatives,and

hecarefulwhendriving rnotorvehicle operatingrnachinery
Oerccitalsilityrnayoccur,e:pecialyineh ilge'n

Stopuseandaslcadoctarit 0 nervoua,¢uy,orsleoplees
0coudrlastsmorethan7days,carney?b?clt', oroceurswithiever, rash, orheadache
thatlasts. Ihesecauldhesignsataseriousconition.
Osylnptornsdanatgethetterwithin7daysoroccurwithafever

snowm“ .. amwmm-w.
PoisonControICenterrightaway.

Directions ‘talceeveryltobhaurs;natrnorethan4doaesin24haurs
gleam—13L...

“sermon urea“...
Wflmxmwwr‘omacwnw ,dyeerhpropylenedyeol,

 

 

 

 

Compliant Label
Drug Message
Futura Bold
(Body text is 6.0 points)

 

 

 

141

 

Appendix 6- Examples of a Variety of Noncompliant Label Designs
(Drug Messages)

142

NCM
GSL 5.5

Drug Facts

Active Ingredients (in each 5 mL)
Bromphen

amine maleate 2 mg ._,._...“sauna---”----..Anthlstamine
Dextromethorphan HP! 10 mg .. .__ -- .....Wm--. -.- , Cough suppressant
Pseudoephedrne HCL 30 mg”--._.....--_.,..-_..s_NasaJ decongestant
Use temporarily relieves
- sneezing - many nose ' nasal (digestion 0 cough
Warnings

Do not use if you me now (along .1 prescnpton monoamtne omdase inhibitor
(MAOI) (certain drugs lor depressom, psychutnr or emotional conditions or
Parkinson‘s disease). or (or 2 weeks after stopping the MAOI drug If you do
not know it your prescriptm drug contains at MAOI. ask a doctor or
pharmacist belore tailing this product

Askadortorbeloreuseifyouhm 0diabetes -gtaucoma
- thymcd disease 0 cough that occurs wrth too much phlegm (mucus)
- trouble urinating clue to an enlarged prostate gland ' heart disease

' a breathing problem or rhronnc cough that lasts or as occurs wrfli smoking
astl'irna,chrom( bronchttrs. or W3

Madoctororphamacmbeloreusedyouzetakhgsedatwesorumqulizers

When using this product - do not use more than directed
- drowsiness may 0((t1‘ - avoid alcoholic (kinks

- alcohol. sedatwes. and tranqmlrzen my Increase drowsiness

- be careful when drmng .1 motor Whide or opening machinery
' excitability may occur. cspecally in children

Stopuseandasiadoctorrf -yougetnervous.dzzyorsleepless

- cough lasts wore 0m 7 days. comes backororcurs wrth leverxashorheadache
thatlastsThesecmildbesognsolasenouscondltm
-syrnptansdonotgetbetterwnlm7daysoroccuwrthaie~er

l1 pregnant or breast-feeding, ask a health prolessnonal before use.
Keep out of reach of children In case of overdose. get med-cal help or contact a
W Control Center nght sway.

 

Directions ’takeeverytto6hmnotmorethan4ttosesm24horn
1.2 racism QISL._4L_.__LQL_
6tofl31L_ _, _km 5 mL_,_
1199c.§.z§§.__w_-maksaios19r_

 

Inactive ingredients citric acid. FDR-C blue ti I . giycem. propylene glycol.
purified water. sacchann Sodium sod-um benzoate. sorbitol

 

 

Non-compliant Label
Drug Message
Gill Sans Light
(Body text is 5.5 points)

 

 

 

143

 

NCM
TNR 5.5

Drug Facts

 

 

 

Active Ingredients (in each 5 mL)

Brompheniamine msleatc 2 m: Antihistamine
Dextromcthorphan HBr IO m- Cough suppressant
Pseudoephednne HCL 30 m- Nasal decongest-a
Use temporarily relieves:

~sneez'arg -nrnnynoae 'nasalcongestion ocough
Warnings

Do not use if you we now taking a prescription monoaminc oxidase inhibitor
(MAO!) (certain drugs for depression. psychiatric or emotional conditions. or
Parkinson's disease). or for 2 weehs after stopping the MAOI drug. If you do
notknow ifyourprescriptrondrugcontainssnMAOl.askadoctoror

pharrnacislbel’oretakingthisproduct

Ask sdoctorbeforeuse if you have diabetes 'glsueoma
'thyroiddiaeaae -eoughthatoccurswrthtoornuchphlegrn(mucus)
'troublewinatingduetoanenlargedprostategland °heartdiaeaae

osbresthingproblemorclaoniccmrghthatlastsorasoccuswithsmokhrg.
sstlunachronicbronchimaemphyaema

Asksdoctotorph-macistbeforeoaeifyouueukmgsedanvesoruanquiliaen

Whenusingth'aprothict -donotusemorethsndirocted
- drowsiness may occur . avoid alcoholic drinks

0 alcohol. sedatives. and tranquilizers may uicrease drowsiness

- be careful when driving a motor vehicle oroperating machinery
- mutability may occur. especially in chilthen

Stopureandsaksdoctorif -yougetnervous.dizzy.orslecpleas
'coughlastsmorethan‘ldoyncmnesbeehoroceurswithl’evernashorheadache
thatlasts. 'l'heaccouldbesignsofaseriouseonditm
~rymptonldonotgetbetterwithm7daysoroccrlwithalcver

If W or breast-feeding. ask a health professional before use.
Keepoutot'reachofchrldren. Incaaeoloverdosc.getmedicalheborcontacts
Poison Causal Center ngbt sway.

 

 

 

 

Directions ‘takeevery4t06hours;notrnorethsn4dosesin24hours
Hm ml. _

Molly-II Sail.—

mgm asksdoctor

 

Inactive Ingredients citric acid. mac blue at. glycerin. Wlaie glycol.

 

Non-compliant Label

Drug Message
Times New Roman

(Body text is 5.5 points)

 

 

 

144

 

NCM
Tahoma 5.5

Drug Facts

Active Ingredients (in eadi 5 mL)
Brompheniamlne maleate m9. ........................

Wamrngs
Donotuseitywuenmtaldngsprescnwmnionoaminemddaalnhibkor
(MAOI)(certaindmgsfudepre$ion.psydtiancaemofionalwidmons.a
Parkinson'sdisease).orfor2weelcsatterstoppin9theHAOldru9. Ifyoudo
notknowitywrprescnptiondrugcontalnsanMAOLaskadoctoror

plunnadstbeforetakingthisproouct

Asksdodorbeforeuselyouteve odabets cglaucorna
.Wdisease .mghmarocmwmmomnhleqmtmm)
otroubleurinatingduetoanenlaroedprostztegtand -henrtdisease

oebreattfinq problemadrroniccoughthatlafiorasomrrswlflrsmoklng.
“Illa, CINGK' ”Nth.“ “ UIIDIIIEE I

Mamuwmmlmnmmamflm

- readability may noun. espeddly

Stopusearidaskadoctorll' ongetmrfiaymrsteepless
oowghlastsmethan7days.annesbadcaocossmmmaheadade
thatlasts. Theoecouldbesignsofasertwswtdltion.
osyrrmmsdonotoetbetterwithin7daysorocurrwlthslever

"madam-Muaskaheommbdueme

Keepoutotreadiotdiildren. lnasedovadmgetnmhelporcontacta
PocsonControlCenternghtmy.

DII’ectIOt'lS 'taleeveryttoShoursmotmorethansdoseshZ‘ttuts

 

 

 

 

Alwanttova ton-L
_6t°-,12vears ' Smt.
underByear askadoctor

 

Inactive IDQFEdIentSclrlcedeDthhrelfigycernprowleneglycol.
Wmmmmm.m

 

Non-compliant Label
Drug Message
Tahoma
(Body text is 5.5 points)

 

 

 

145

NCM
LS 5.5

Drug Facts

Active Ingredients (in each 5 mL)

 

 

 

Brompheniarnine maleate 2 mg Antlhlstamine
Dextromethorphan HBr l0 mg rough suppressant
Pseudoephedrine HCI. 30 mg Masai decongestant
USE temporarily relieves:

- sneezing - runny nose - nasal congestion . cough
Warnings

Do not use if you are now taking a prescription monoamlne oxidase inhibitor
(MAOI) (certain drugs for depression. psychiatric or emotional conditions. or
Parkinson's disease). or for 2 weeks after stopping the MAOI drug. If you do
not know If your prescription drug contains an MAOI. ask a doctor or
pharmacist before taking this product

Ask a doctor before use if you have - diabetes - glaucoma

- thyroid disease ~ cough that occurs with too much phlegm (mucus)
- trouble urinating due to an enlarged prostate gland - heart disease

- a breathing problem or chronic cough that lasts or as occurs with smoking.
asthma. chronic bronchitis. or emphysema

Ask a doctor or pharmacist before use If you are taking sedatives or tranquillzers

When using this product - do not use more than directed

. drowsiness may occur - avoid alcoholic drinks

- alcohol. sedatives. and tranquilizers may Increase drowsiness

- be careful when driving a motor vehicle or operatlng machinery
- excitability may occur. especially in children

Stop use and ask a doctor if - you get nervous. dizzy. or sleepless

- cough lasts more than 7 days. comes back. or occurs with fever. rash. or headache
that lasts. These could be signs of a serious condition.

- symptoms do not get better within 7 days or occur wlth a fever

If pregnant or breast-feeding. ask a health professional before use.

Keep out of reach of chlldren. In case of overdose. get medical help or contact a
Poison Control Center right away.

Directlons 'take every 4 to 6 hours; not more than 4 doses In 24 hours
Ilium—1.1L
___5 mL_

itolLyears.__,
mgr 6 ygar ask a doctor

 

 

Inactive Ingredients citric acid. rosc blue rt. glycerin. propylene glycol.
purified water. saccharin sodium. sodium benzoate. sorbitol

 

Non-compliant Label
Drug Message
Lucida Sans
(Body text is 5.5 points)

 

 

 

146

NCM
Censb 5.5

Drug Facts

Active Ingredients (in each 5 mL)

 

 

 

Brompheniamine maleate 2 mg ‘ "" ‘ '
Dextromethorphan HBr 10 mg Cough suppressant
Paeudoephedrine HCL 30 In: Nasal decongestant
Use temporarily relieves:

sneezing ' runny nose . nasal congestion ' cough
Warnings

Do not use if you are now taking a prescription monoamine oxidase inhibitor
(MAOl) (certain drugs for depression. psychiatric or emotional conditions. or
Parkinson's disease). or for 2 weeks after stopping the MAOI drug. 11' you do
not know if your prescription drug contains an MAOI. ask a doctor or
pharmacist before taking this product

Ask a doctor before use if you have ‘ diabetes ° glaucoma

' thyroid disease cough that occurs with too much phlegm (mucus)
trouble urinating due to an enlarged prostate gland ° heart disease

- a breathing problem or chronic cough that lasts or as occurs with smoking.

asthma. chronic bronchitis. or emphysema

Askadoctorwpharmscistbefonuuifyouantakingeedativuortranquilisers

Whenusingthisproduct - donotusemorethandirected

' drowsiness may occur ° avoid alcoholic drinks

° alcohol. sedatives. and tranquilizers may increase drowsiness

' be careful when driving a motor vehicle or operating machinery
- eacitability may occur. especially-in children

Stop use and ask a doctor if you get nervous. diary. or sleepless

cough lasts more than 7 days. comes back. or occurs with fever. rash. or headache
that lasts. These could be signs of a serious condition.
° symptoms do not get better within 7 days or occur with a fever

If pregnant or breast~feeding. ask a health professional before use.
Keepoutofreachcfchildren. lncaseofoverdose.getmedicalhelporoontacta
Poison Control Center right away.

Directions 'takeeveryttofihoumnotmorethansdoeesinflhours

12 yemndch____1iLmlc_
6 lumen—__fi L

 

 

Inactive Ingredients citric acid. mac blue s1. glycerin. propylene glycol.
purified water. saccharin sodium. sodium benaoata.eub

 

Non-compliant Label
Drug Message
Century Schoolbook
(Body text is 5.5 points)

 

 

 

I47

NCM
Cengot 5.5

Drug Facts

Active Ingredients (in each 5 mL)

 

 

 

Brompheniamine maleale 2 mg Anlhislamino
Dextromethorphan HBr 10 mg Cough suppressant
Pseudoophedrine l-iCL 30 mg Nasal docongeslanl
Use lempororiy reieves:

- sneezing orunnynoso onasolcongeslion ~cough
Warnings

Do not use if you are now taking a prescription monoamine oxidase inhibitor
(MAOI) (certain drugs for depression. psychiatric or emotional conditions. or
Parkinson's disease). or for 2 weeks after slopplng lhe MAOI drug. If you do
not know ii your prescription dug conlolns an MAOI. ask a doclor or
pharmacist beiore taking this product

Askadociorbeioreuseilyouhave ~diabeles oglaucoma

. thyroid disease - cough that occurs with too much phlegm (mucus)
. trouble urinating due lo an enlarged prostate gland - heart disease

0 a breathing problem or chronic cough thal lasls or as occurs with smoking.
asthma. clvonic bronchitis. or emphysema

Ask a doctor or pharmacist before use it you as loiting sedatives or tranquilizers

When using this product 0 do not use more lhan drecled

- dows'lness may occur r avoid alcoholic drinks

0 alcohol. sedatives. and lranquflizers may increase dowsiness

0 be careful when Giving a molar vehicle or operator; moch'nory
- excitability may occur. especialy in children

Slop use and psi: a doctor if - you gel nervous, dizzy, or sleepless

. cough lasts more than 7 days. comes bocic or occurs with lever. rash. or headache
that lasts. These could be signs of a serious condition.

. syrnplomsdonol gel bellerwilhin 7doysoroccurwilhalever

ll pregnanl or breast-loodng. ask a health professional before use.
Keep out at reach al children. In case ol overdose. gel medical hob or contact 0
Poison Conlrol Center right away.

 

Directions 'ialte every 4 lo 6 hours; nol more than 4 doses in 24 hours
J2 years ondover_ 10m L

6 IoIZ years“ ArS ml. _

meter 6 m. Josh 0. glacier. _

 

Inactive Ingredients citric acid. rosc blue l I. glycerin. propylene glycol.
purified water. saccha'n sodium. sodium benzoate. sorbilol

 

Non-compliant Label
Drug Message
Century Gothic

(Body text is 5.5 points)

 

 

 

148

NCM
BMOS 5.5

Drug Facts

Active Ingredients (in each 5 mL)

 

 

Brompheniamine maleate 2 mg .Antihistamine
Dextromethorphan HBr l0 mg .......... Cough suppressant
Pseudoephedrine HCL 30 mg ............................................... Nasal decongestant
Use temporarily relieves:

- sneezing - runny nose - nasal congestion - cough
Warnings

Do not use if you are now taking a prescription monoamine oxidase inhibitor
(MAO!) (certain drugs for depression, psychiatric or emotional conditions. or
Parkinson's disease), or for 2 weeks after stopping the MAOI drug. If you do
not know if your prescription drug contains an MAO]. ask a doctor or
pharmacist before taking this product

Ask a doctor before use if you have - diabetes 0 glaucoma

- thyroid disease 0 cough that occurs with too much phlegm (mucus)
- trouble urinating due to an enlarged prostate gland 0 heart disease

0 a breathing problem or chronic cough that lasts or as occurs with smoking.
asthma. chronic bronchitis. or emphysema

Ask a doctor or pharmacist before use if you are taking sedatives or tranquilizers

When using this product 0 do not use more than directed

- drowsiness may occur - avoid alcoholic drinks

- alcohol. sedatives. and tranquilizers may increase drowsiness

- be careful when driving a motor vehicle or operating machinery
- excitability may occur. especially in children

Stop use and ask a doctor if 0 you get nervous, diuy, or sleepless

0 cough lasts more than 7 days, comes back. or occurs with fever. rash, or headache
that lasts. These could be signs of a serious condition.

- symptoms do not get better within 7 days or occur with a fever

lf pregnant or breast-feeding, ask a health professional before use.
Keep out of reach of children. In case of overdose. get medical help or contact a
Poison Control Center right away.

Directions ‘talte every 4 to 6 hours; not more than 4 doses in 24 hours

lZyeanandm, _ 1;)me
6t012yeara ___r _ ital.—
undgom gadoctgr

 

 

Inactive Ingredients citric acid. mac blue it. glycerin. propylene glycol.
purified water. saccharin sodium. sodium benzoate. sorbitol

 

Non-compliant Label
Drug Message
Bookman Old Style
(Body text is 5.5 points)

 

 

 

149

NCM
Hbld 5.5

Drug Facts

Active lngrodlonts (In each 5 ml.)

Brornphsnlsmlns wists-to: Animist-ulna
Dextromsthorphsn no: to my... ..................... Cough slipprssssut
Pssudosphsdrlns HCL so me Need decongestant

USOt-npor-ttyrolfsvss:
she-am nirmynoss -nssslcongsstloo -coooh

 

 

Wamlngs
Donotussflyouaronowtsklngaproscrlptlonmonosulnsoaldasslm
(MAOMcsrtslndrugstorMpsyehlstrtcor orouuohorsstcondltlons,"
Puklnson'sdssssstorforzvoslrsaflsrslopplngthsmm tfyoudo
nolknoivtfyotuprsscrfptfondrugcornalnssnmmasksdocloror
phamsctsthstorstsldngfhlsproduct

Ashadoctorhdorousolfyouhavo -d&dss ~glsucolha
-thyrolddlsosss ~cooghthstoccurswlthtoormchphlsgmmucus)
Mudflatlmduotosnsnlsrgsdproshtsglsnd Heart
-shmthlngprohlunorchronleeooghlhsthsuorasoccwswlmm

“WMMaor-nphym

Mamuwmmmflmmmmum

thnuslngthlsprodoct donotusornorothsndlroclsd
-aowslnsssmsyoccur -svoldslcohollc¢l'lnfts
ssdstlvss.lncressssndtranulllesrsmoy mm

-slcohol.
-hscsrolulvrhsndrlvlngsulotorvohfclooropsrsthgmschlnsry
'WWMW.WIIylnchm

prusosndssltsdoctorll ~yougstnsrvoos,dluy.orslospfsss

°syrlmtomsdonotgslbattuivlttilh7dsysoroccurwtthatovw

uwummmsmmmm
Kospoiitofroscholchlldren. hcsssolovuoosagdmsdcalhstporcontscts
mausoleum”

Dlroctlons 'hhsovuyltoIMnotmmsnldossshuhou-s
fimmLJOL
stolzyssrs Suit. _
Wfim

lnoctlvolngrodlentsmmroscwn, Mount-mm
Mmmmmm

 

 

 

 

Non-compliant Label
Drug Message
Helvetica Bold

(Body text is 5.5 points)

 

 

 

150

NCM
HL 5.5

Drug Facts

Active ingredients (in each 5 mL)
Bromphoniarnino maleato 2 m9

 

 

 

. .I.I .
Dextrornelhorphan HBr 10mg mingh suppressor!
PsoudoophsdrlnoHCLSOme NMW
Uselomporarllyrslioves:

mooning -runnynoso -nasalcongsstion -cough
Warnings

Donolusslyousronowlakmsproscrlpfionmonoamhoaxldasshhbhor
(MAOl)(cenaiidmgslordeprossion.psychiaaicoremohondcondlm.or
Partriwon'sdiuase).orlor2weeksalterstopphglheMAOldug. llyoudo
nolltnowilyourprescnptiondnigcontainsanMAOl.askadoctoror

pharmacistbotorelsliinglhlsproduci

Asksdoctorboforousoityou -dfabotss -glsucorris
-ttryroidaisoaso woughthachrswithbonuchphlongm)
-troubfourinsthgduetosnonlargedprostategland -hoarlalsoaso

'sbrssmingproblemmchruiccoughlhstlastsorasoccinmm.
sstllrlactuoriicbrorictfiuoromphysems

Asksdocioraphannsclslbslorsusslywmtakhossdslvosoremlflzon

thnusirigthlsprodud -donotusornorethan&sclsd
-drowslnossmayoccur -avoidalcoholicdrinks

- alcohol. sedatives. andtranquitizers rnsy lr'icrsasodromshoso
~bocarstulwhondrivingsmotorvohicleor

- sscltabilitymsyoccur. ospociaftyin children

Stopusoandasksdoctorl -yougotnervous.dlzzy.orsloeploss
-cwghlastsmoroman7days,comosbadcorocwnwlhbvor.rodkuhoadodio
Ihallasts. Thosocouldbssignsotsserlooscondition.
osyrnplomsdonotgetbotlsrwlthirndsysoroccuraMafovsr

lfpregnmtorbresst-toeding.astlshsslhprolessbnalbotorouso.

Keoporrlofroscholchildron. Incassotovsroosomslmsdcalhohorcontacts
PoisonControlentorrlghtavrsy.

Dil’OCthf‘lS ’lalrsovsryuoohowsmolmoroltmldososhflhours

i ___an.
smifififi ml.—

lindyw“ # Madge”

 

 

Inactive Ingredients elite sold. me on. n. glycsrh. propylene glycol.
Winter. ssccharlnsodfum. sedimbsnzosts. sarhllol

 

Non-compliant Label
Drug Message
Helvetica Light
(Body text is 5.5 points)

 

 

 

151

NCM
LSTW 5.5

Drug Facts

Active Ingredients (in each 5 mL)

Irowphenisnine aaleate 2 lg ........................... Antihistaline
Dextronethorphsn ”Or 10 lg ........................ (ou suppressant
Pseudoephedrine HCL 30 wg... ............ . ...... ..Nasa decongestant

Use teaporsrily relieves:

- sneezing - runny nose - nasal congestion - cough
Warnings

Do not use if you are now taking a prescription sonoawine oxidase inhibitor
(HAD!) (certain drugs for depression. psychiatric or ewotionsl conditions. or
Parkinson's disease). or for 2 weeks after stopping the HAO! drug. If you do
not know if your prescription drug contains an HADI. ask a doctor or
pharaacist before taking this product

Ask a doctor before use if you have ~ diabetes - glaucoma

- thyroid disease - cough that occurs with too such phlega (wucus)

- trouble urinati due to an enlarged prostate gland - heart disease

- a breathing prob ea or chronic cough that lasts or as occurs with sacking.
ssthws. chronic bronchitis. or eaphyseaa

Ask a doctor or phsrwacist before use if you are taking sedatives or tranquilizers

when using this product - do not use were than directed
- drowsiness say occur - avoid alcoholic drinks

- alcohol. sedatives. and trsnouillzers say increase drowsiness

- be careful when driving a actor vehicle or operating wachinery
- excitability aay occur. especially in children

Stop use and ask a doctor if - you get nervous. dizzy. or sleepless

- cou . lasts sore than 7 days. cases back. or occurs with fever. rash. or headache
that ssts. These could be signs of a serious condition.

- sywptoas do not get better within 7 days or occur wdth a fever

If pregnant or breast-feeding. ask a health professional before use.
Keep out of reach of children. In case of overdose. get wedicsl help or contact a
Poison Control Center right away.

 

 

 

Directions 'tslie every 4 to 6 hours; not sore than 4 doses in 24 hours
1; ygsrs and ovgr 19 !L
_§ to 12 ears 5 It

under year sii a doctor

 

 

Inactive Ingredients citric acid. FDIC blue 01. glycerin. propylene glycol.
purified water. ssccharin sodiua. sodiua benzoate. sorbitol

 

Non-compliant Label
Drug Message
Lucida Sans TW
(Body text is 5.5 points)

 

 

 

152

Appendix 7- Consent Form

153

Label Legibility— Consent to be Tested

You are being asked to read printed cards using an instrument called the Polariscope.
The Polariscope is a tool tint provides a numerical value for how easy or difficult a
message is to read. The purpose of this research is to develop the Polariscope as a
performance standard for legibility. This would ensure that the label information
provided on a variety of packages is sufficiently legible. This is becoming increasingly
important as the population ages.

Prior to testing, your visual acuity will be measured by reading a card. The lowest line
that you can read on the card will determine your visual acuity (20/20, 20/30, etc.). This
information will be recorded. Information about your gender, educational background
and age will also be recorded.

You will read a card placed inside the grey box. Look into the box through the screen on
the front. As you look through the screen turn the knob on the right side of the box until
you can easily read the words on the card in the box without straining your eyes. The
operator will record the value you get for each card. Once the value is recorded, turn the
knob back to its starting position so that the screen is dark again The operator will put a
different card in the box for you to read.

You will not be identified by name in any records of this testing; testing is anonymous.
Your participation will be protected to the maximum extent of the law. You may
discontinue participation at any time without penalty.

Testing will take no more than 30 minutes.

I choose to participate in the label legibility study.

 

Date:

 

I decline participation in the study. (Declining to participate will not reflect negatively in
any way on subjects).

 

Date:

 

If you have any questions or concerns about this study, please feel free to contact Dr.
Hugh Lockhart, Michigan State University School of Packaging at 517-355-3604 or
Laura Bix at 517-333-9967.

If you have questions regarding your role or rights as a research subject, contact David
Wright, PhD Chair, University Committee Involving Human Subjects, at 517-355-2180.

You will be provided with a copy of your signed consent form.

154

Appendix 8- Data Recording Sheet: Primary Study

155

Data Recording Sheet

 

 

 

 

Primary Study
Subject #
Inside Light Level Ambient Light Level
Male Female Visual Acuity
8th Grade High School Undergraduate Graduate Doctorate
19-28 29-38 39-48 49-58 59-68 79-88

Number of times .
label has been Required Degrees

viewed of Rotation

 

l
l
l
l
l
1
l
1
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3

156

Appendix 9-Data Recording Sheets: Preliminary Study 1 and 2

157

Data Recording Sheet
The Effect of Distance on Variability Data (from Machine Front)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Subject #
Inside Light Level Ambient Light Level
Male Female Visual Acuity
8th Grade High School Undergraduate Graduate Doctorate
1 9-28 29-38 39-48 49-58 59-68 79-88
Part A: Subjects Choose Reading Distance
Number of .
. Required
Label # times label has Degrees of Time Required
been Viewed by R .
. otation
subject
D
D

l

1
Distance (Wall 1

to Easel) 1

l

1

2

2

2

2

2

2

 

 

 

 

 

158

 

Part B: Distance Fixed
Number of
times label has Required Degrees of

Label # been viewed by Rotation

Time Required

D
D

 

1
l
l
l
1
1
2
2
2
2
2
2

159

Data Recording Sheet
The Effect of Distance on Variability Data (from Machine Back)

 

 

 

 

Subject #
Inside Light Level Ambient Light Level
Male Female Visual Acuity
8th Grade High School Undergraduate Graduate Doctorate
1 9-28 29—38 39-48 49—58 59-68 79-88

 

Part A: Subjects Choose Reading Distance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Number of Required
Label # gfisvliztveédh: Degrees of Time Required
. Rotation
subject
D
D
1
1
Distance (Wall 1
to Easel) 1
1
1
2
2
2
2
2
2

 

 

 

 

 

160

 

Part B: Distance Fixed
Number of
times label has Required Degrees of

Label # been viewed by Rotation

Time Required

D
D

l
l
l
1
l
l
2
2
2
2
2
2

 

161

Appendix 10- Data Recording Sheet: Preliminary Study 3

162

 

Data Recording Sheet
The Effect of Instrument on Variability Data

 

 

 

 

Subject #
Inside Light Level Ambient Light Level
Male Female Visual Acuity
8th Grade High School Undergraduate Graduate Doctorate
19-28 29-38 39-48 49-58 59—68 79-88

 

Part A: Older Instrument

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Number of Required
Label # $322332: Degrees of Time Required
. Rotation
subject
D
D
1
1
1
1
1
1
2
2
2
2
2
2

 

 

 

 

 

163

 

Part B: New Instrument
Number of
times label has Required Degrees of

Label # been viewed by Rotation

Time Required

D
D

 

l
l
l
l
l
l
2
2
2
2
2
2

164

Appendix 11- Tukey-Kramer Pair Wise Comparisons

165

Effect
Compllant
Design(Compllant)
Design(Compllant)
Design(Compllant)
Designmompllant)
Design(Compl|ant)
Design(Compllant)
Element
Element
Element
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compli)
Deslg*Elemen(Compll)
Desig*Elemen(Gompll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compli)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Gompll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Gompll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Oompll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Oompll)
Desig*Elemen(Compll)
Desig*Elemen(Gompll)
Deslg*Elemen(Gompll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*EIemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Desig*Elemen(Compli)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)
Dedg*E|emen(Compll)

Compllant

NHHHHHH

NNHHHHHHHHHHHHHHHHHHHHHHHHHHHHH#HHHHHHHHHHHHHHHHHHHHH

Design

UNNHHH

wwNNNNNNNNNNNNNNNNNNNNNHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

Element

HwawwwwNNNNNNNHHHHHHHHuwwwwwwwwNNNNNNNNNNo-bHHHHHHHHHHNHH

166

_Compllant

NNNNNHN

NNNNNNNNNNNNNNHNNNNNNHHNNNNNNHHHNNNNNNHHHHNNNNNNHHHHH

_Deslgn

ccwnwu

DUN-bA-hwuw-hA-bWNWN-h-h-hWUWNN-h#AWNWNNN-hA-hWUWNNNH-h-hANNUNNNHH

_Element Adlp

WNWNHWNHWNHWNHUUNHWNHWNWNHWNHWNHWNHWNHUNHWUNHUNHWNHWNWWN

0.00020811
0.88291162
0.03416144
0.00579328
0.19701401
0.05316928
034048957
2.3505506
4.3537510
1.254508
0.20197274
5.049509
0.99999987
0.48554091
6.8459509
1
0.84716174
0.01518079
0.99991935
0.7703398
0.1013974
0.00075223
0.0761 1 167
0.99999887
0.00449697
0.13005174
0.99751379
0.99926878
0.02803901
0.99935702
1
4.9556509
8.6784505
0.99999952
4.9763509
5.2777506
0.02362419
4.9499509
1.102505
0.00243802
0.24376451
5.0905509
1
0.60558057
0.00363642
039999998
0.50479818
0.03275156
0.00064177
0.35990085
039999413
037251583
0. 1 121 7919
039999973
039986459
5.5053509
4.9163505
0.08853269
4.963509
9.6968505
0.01272543
0.74262107
0.00782038

Desig*Elemen(Compli)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compll)
Deslg*Elemen(Gompll)
Deslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Gompll)
Deslg*Elemen(Compll)
Deslg*Elemen(Compli)
Dslg*Elemen(Compll)
Desig*Elemen(Compll)
Deslg*Elemen(Compll)

NNNNNNNNNNNNN

Abawwuwwwwwww

NHwawNNNNHHH

167

NNNNNNNNNNNNN

A-A-bA-b-b-b-b-hw-h-b-h

WWNUNHWNHWWNH

039999621
0.64868479
0.0605106
0.76434068
0.37457078
1
038091364
0.00094177
0.84227808
039998197
0.28895838
0.01071318
039239672

Bibliography

168

BIBILIOGRAPHY

Alsobrook, Henry B. , “An Overview of Liability for OTC Drugs,” Drug Information
Journal, Vol. 26, Pp.317-328, 1992.

Arnold, Edmund, Ink on Paper, Harper and Row Publishers, New York, NY,
1972.

Beers, Mark, “Age-Related Changes as a Risk Factor for Medication-Related Problems,”
Generation, Pp 22-48, Winter 2000-2001.

Becker, D; Heinrich, J .; von Sichowsky, J. and Wendt, D, “Reader Preferences for
Typeface And Leading,” The Journal of Typographic Research, Vol. 4 .(1): 61-
66, 1970.

Bigelow, Charles and Day, Donald, “Digital Typography,” Scientific American, Volume
249(2), August. 1983.

Bix, Laura; Gilliland, D; Chen, B. and Sung, H. from unpublished file, “Using the
Polariscope as a Measure of Legibility,” December, 1997.

Bix, Laura, The Effect of Subject Age On Legibil'mLLMaster’s Thesis, Michigan State
University, 1998.

Bix, Laura; Lockhart, Hugh “Why Investigate the Legibility of OTC Drug Labels?”
Proposal for Funding- Center for Food and Pharmaceutical Packaging Research.
October, 2000.

Bradley, B.; Singleton, M. and Li Wan Po, A, “Readability of Patient Information
Leaflets On Over-The-Counter Medicines,” Journal of Clinical Pharmacy
Therapeutics, Vol. 19(1): 7-15, February, 1994.

Braus, Patricia, “Did You Take Your Medicine?,” American Demographics, Pp. 14-15,
January 1993.

Burt, C., A Psychological Study of Typography, Cambridge University Press, London,
England, 1959.

Cameron, Kathleen and Richardson, Anne, “A Guide to Medication and Aging,”
Generations, Pp. 8-20, Winter 2000-2001.

Caranasos, George, MD; Stewart, Ronald, MS and Cluff, Leighton, MD,

“Drug-Induced Illness Leading to Hospitalization”, Journal of the American
Medical Association, Vol. 228 (6): 713-717, May 6, 1974.

169

Cardoso, Fernando, Personal Communication, October, 2001.

Conn, VS, “Self-Management of Over-The-Counter Medications by Older Adults,”
Public Health Nursing, Vol. 9 (1): 29-36, March, 1992.

Consumer Healthcare Products Association (CHPA), “CHPA Statement to FDA at
September 17, 1999 Labeling Feedback Meeting,” Comments from R. William
Soller, Ph.D. Senior Vice President and Director of Science and Technology,
Posted to:www.chpa-info.org. September 1999.

Craig, James. Designing with Type: A Basic Course in Typogr_apl_iy. Watson-Guptill
Publications. New York. 1980.

Department of Mathematics, University of Utah, “Fontname,”
http://www.math.utah.edu/docs/info/fontname_toc.html, August 15, 2001.

Discenza, Richard and Ferguson, Jeffery, “The Instrumental Role of Product Information:
A Study of Warning Labels for Non-Prescription Drugs,” Health Marketing
Quarterly, Vol. 10 (1/2): 155-168, 1992.

Department of Physics and Astronomy, Arizona State University, “Patterns in Nature:
Light and Optics,” http:/hcept.asu.edu/PII‘I/rdg/polarize/polarize.shtml
December, 1999.

Dow Corning Ophthalmics, “Near Point Visual Acuity Card Instructions,” Norfolk,
VA, 1981.

Federal Interagency Forum on Aging-Related Statistics, “Older Americans 2000: Key
Indicators of Well-Being: Population-Indicator 1: Number of Older Americans,”
www.agingstats.gov/chartbookZOOO/population.html, May, 2000.

Food and Drug Administration, “Final Rule- Over-the-Counter Human Drugs;
Labeling Requirement,” Federal Register, March 17, 1999.

Food and Drug Administration, “Over-The-Counter Human Drugs; Proposed Labeling
Requirements,” Federal Register, February 27, 1997.

Fulmer, Terry and Kim, Tae Sock; Montgomery, Kristen and Lyder, Courtney, “What the
Literature Tells Us About The Complexity of Medication Compliance in the
Elderly,” Generations, Pp. 43-48,Winter 2000-2001.

Garcia, M., Contemporary Newspaper Design, Prentice-Hall, Inc., Englewood Cliffs,
NJ, 1981.

Gill, John, Design and Analysis of Experiments in the Animal Sciences, The Iowa State
University Press, Ames, Iowa, Pp.154, 1978.

170

Gore, Prasanna; Madhavan, Suresh; McClung, Gordon and Riley, David, “Consumer
Involvement in Nonprescription Medicine Purchases,” Journal of Health Care
Mareketing. Vol. 14 (Summer): 16-23, 1994.

Hauptman,D., “Effective Direct Response Typography,” Direct Marketing, 42 (8):
24-28, 1979.

Hignite, Karla, “The Accessible Association,” Association Management,
Vol.52 (13): 36-43, December, 2000.

Holt, Gary and Hall, Edwin, “The Self-Care Movement,” Handbook of Nonprescription
Drugs, 9th Edition. American Pharmaceutical Association, Pp.1-9, 1990.

Holt, Gary; Hollon, James; Hughes, Scott and Coyle, Rebecca, “OTC Labels: Can
Consumers Read and Understand Them?,” American Pharmacy, N830 (11):
51-54, November, 1990.

Illumination Engineering Society of North America. Recommended Practice for Lg’ htmg'
and the Visual Environmgt for Seflr Living. IESNA RP-28-98. 1998.

International Paper, Pocket Pal: A Graphic Arts Production Handbook,
International Paper, Seventeenth Edition, 1997.

Institute of Grocery Distribution, “Packaging Design: Improving Legibility”
Consumer Concern Programme, 1994.

Institute of Packaging Professionals, “New Packaging Challenges Presented by Aging
Population, Safety,” Pack Info Newsletter, Vol. 12 .(4): 4-5 2000.

Jewler, J ., Creative Strategy in Advertising, Wadsworth Publishing Co., Belmont, CA.
1 981 .

Kelly, Maryann, Visual Impairment in the Elderly and its Impact on Their Daily Lives,
P.h.D. Dissertation,Texas Woman’s University College of Nursing, 1993.

Light, LL., “Interactions Between Memory and Language in Old Age,” In Birren, JE and
Schaie, KW.(Eds.), Handbook of the Psychology of Aging, 3'“ Edition, New York.
Pp.275-290.

Lockhart, Hugh and Bix, Laura, fi'om unpublished file report, “Color Contrast Studies,”
1996.

Lockhart, Hugh and Bix, Laura, from unpublished file report, “Comment to FDA on
Proposed Rule,” July, 1997.

171

Loyola Medical Education Network, “http://wwwmeddean.luc.edu/lumen/MedEd
/medicine/pulmonar/pd/pstep9.htm,” September 11, 2001.

Lumpkin, James; Strutton, David; Lim, Christina and Lowrey, Shelia, “A Shopping
Orientation Based Prescription for the Treatment of OTC Medication Misuse
Among the Elderly,” Health Marketing Quarterly, Vol 8(1/2): 95-110, 1990.

Marieb, Elaine N., Human Anatomy and Physiology, Benjamin/ Cummings Publishing
Company, Inc., 1992.

Marwick, Charles, “From Rx to OTC: More Drugs Make the Switch,” Journal of the
American Medical Association (JAMA), Vol. 278 (2):]03, July 9, 1997.

Mateo, Lisa, “Paper Substrates,” Label and Narrow Web, July/August, 2000.

McLean, Ruari, The Thames and Hudson Manual of Typography. Thames and Husdon.
Ltd, Pp.42-48, 1980.

Montgomery, Douglas, “Factorial Experiments with Random Factors,” Design and
Analysis of Experiments, Wiley and Sons, 1997.

Morrell, RW; Park, DC and Poon, LW., “Effects of Labeling Techniques on Memory and
Comprehension of Prescription Information in Young and Old Adults,” Journal of
Gerontology, Vol. 45: 166-172, 1990.

Morrow, Daniel and Leirer, Von, ‘Designing Medication Instructions for Older Adults,”
In Park, Denise; Morrell, Rodger and Shifren. Processing of Medical Information
In Aging Patients, Lawrence Erlbaum and Associates Publishers, London, 1999.

Nonprescription Drug Manufacturers Association, Label Readability Guidelines,
November, 1991.

Nordenberg, Tamar, "New Drug Label Spells It Out Simply," FDA Consumer Magazine,
Washington, DC, Publication #99-3232, July-August 1999.

Norton, Amy, “OTC Drug Labels Ofien Unread: Public Largely Ignorant of
Non-prescription Pain Relievers, Survey Finds,” Medical Tribune, 40(1): 1, 5,
January 7, 1999.

People’s Medical Society, “Useful OTC Drug Labels,” People ’s Medical Society
Newsletter, Vol. 16 (4):6, August 31, 1997.

Perles, P., “Readability, Has It Gone? Or, Pity the Poor Reader,” Direct Marketing,
39(10): 32-40, 1977.

172

Peter, J Paul and Olson, Jerry C, Consumer Behavior and Marketing Strategy, Irwin
McGraw-Hill, 1999.

Pettit, Glenn. Conversation. September, 2001.

Pietrowski, James, Development of a Methodology to Quantify Package and Label
Legibility, Master’s Thesis, Michigan State University, 1993.

Prince, J ., “Printing for the Visually Handicapped,” Journal of Typographic Research,
1 January: 31-47, 1967.

Ralph, Jack, “A Geriatric Visual Concern: The Need for Publishing Guidelines,” Journal
of The American Optometric Association, Vol. 53 (1): 43-50 January 1982.

Rehe, Rolfe, “Newspaper Typography Some Do’s and Don’ts,” Step-by-Step Graphics:
Designers Guide, Pp.116-121, 1990.

Reisenwitz, Timothy and Wirnbish, Joesph, “The Purchase Decision Process and
Involvement of the Elderly Regarding Nonprescription Products,” Health
Marketing Quarterly, Vol. 15 (1): 49-68, 1997.

Rifon, Nora, “Decision Making Lecture,” ADV 823, Michigan State University,
September 28, 2000.

Robinson, J Daniel and Stewart, Ronald B., “Elderly: Understanding Their Prescription
Needs,”.American Pharmacy, Vol. N821 (1 l): 48, November 1981.

Rousseau, Gabriel; Lamson, Nina and Rodgers, Wendy, “Designing Warnings to
Compensate For Age-Related Changes in Perceptual and Cognitive Abilities,”
Psychology and Marketing, Vol.15 (7): 643-662, 1998.

Sansgiry, Sujit and Cady, Paul, “The Effect of Label Content and Placement on
Consumers’ Understanding of OTC Product Label Information,”.JournaI of
Pharmaceutical Marketing and Management, Vol. 9 (3):55-67, 1995.

Sansgiry, Sujit and Cady, Paul, “Can Picture Use Effectively Enhance the Understanding of
Nonprescription Drug Labels in Older Adults?,” Journal of Geriatric Drug Therapy,
Vol. 10 (4): 51-69, 1996.

Sansgiry, Sujit; Cady, Paul and Shubhada, Patil. “Readability of Over-the Counter
Medication Labels,”.Journal of the American Pharmaceutical
Association, Vol. N837. (5): 522-528, September/October, 1997.

Sorg, Judith, An Exploratory Study of Typeface, Type Size and Color Paper Preferences
Among Older Adults,. Master’s Thesis, Pennsylvania State University, 1985.

173

Strutton, David and Tanner, John, “Older Consumers’ Perceptions of Health Care
Information Sources: The Effect on Search Behavior,” Health Marketing Quarterly,
Vol. 11 (3/4): 175-199, 1994.

Summer, F., “Influence of Color on Legibility of Copy,” Journal of Applied Psychology,
16:201-204, April, 1932.

Tarlach, Gemma M., “Using OTC Drugs Makes Good Cents, Study Finds,”_Drug Topics,
February 2, 1998.

Tennesen, Michael, “Before you Play Doctor,” Health, Pp 98-100, 102-103
January/February 1999.

Tinker, M., Legibility of Print, Iowa State University Press, Ames, Iowa, 1963.

Tinker, M; Paterson, D., “Studies of Typographic Factors Influencing the Speed of
Reading VII. Variations in Color of Print and Background,” Journal of Applied
Psychology, Pp.471-479, 1931. 15. October.

US. Senate Special Committee on Aging; The American Association of Retired Persons;
The Federal Council on Aging, Aging America- Trends and Projections, Printed
By the US. Department of Health and Human Services, Washington, DC. 1991.

Vanderplas, J and Vanderplas, J ., “Some Factors Affecting Legibility of Printed
Materials for Older Adults,” Perceptual and Motor Skills, Vol. 50 (1): 923-932,
1980.

Watanabe, Ronald k., OD, “The Ability of the Geriatric Population to Read Labels on
Over-the-Counter Medication Containers,” Journal of the American Optometric
Association, 65. Pp 32-37, 1994.

Wechsler, Jill, “Information Please: Patient Education as 3 Cost Cutter,” Pharmaceutical
Executive, Pp.16-l9, July, 1991.

Wogalter, Michael; Magumo, Amy; Scott, Kevin and Dietrich, David, “Facilitating
Information Acquisition for Over-the-Counter Drugs Using Supplemental
Labels,” The Human Factors and Ergonomics’ Society, Proceedings. 40"I
Annual Meeting. 1996A.

Wogalter, Michael and Laughery, Kenneth, “WARNING! Sign and Label Effectiveness,”
Current Direction in Psychological Science, The American Psychological
Society, V015(2): 33-37, April, 1996B.

Woolley, Scott, “The Quest for Youth,” Forbes, Vol.163 (9): 146-151, May 3, 1999.

Wright, J; Warner, D and Winter, W; Zeigler, 8., Advertising. McGraw-Hill Book
Company, New York, NY, 1977.

174