Content uploaded by Bernard Roullet
Author content
All content in this area was uploaded by Bernard Roullet on Oct 30, 2015
Content may be subject to copyright.
Bernard Roullet, Olivier Droulers (2005), PHARMACEUTICAL PACKAGING COLOR AND
DRUG EXPECTANCY, in Advances in Consumer Research Volume 32, eds. Geeta Menon and
Akshay R. Rao, Duluth, MN : Association for Consumer Research, Pages: 164-171.
Advances in Consumer Research Volume 32, 2005 Pages 164-171
PHARMACEUTICAL PACKAGING COLOR AND DRUG EXPECTANCY
Bernard Roullet, CREM, Universite de Rennes 1 B France
Olivier Droulers, CREM, Universite de Rennes 1 B France
[Both authors equally contributed to this work.]
ABSTRACT -
Pharmaceutical markets are expanding considerably due to the aging population, higher development
costs and also direct-to-consumer advertising which entails more demands from consumers and
prescriptions from physicians. Pharmaceutical packaging as a visual communication tool is promised
to a mounting importance, because of growing blister packaging, safety standards upgrading,
expansion of OTC drugs and developing television advertising. This study examines the impact of
packaging color on consumers’ expectancies towards the drug and seeks determining if prototypical
color codes exist for drug categories. Results show a significant influence of color and darkness on
perceived drug potency. Gender differences are discussed.
"You take the blue pill and the story ends. You wake in your bed and
believe whatever you want to believe. You take the red pill and you stay in
Wonderland and I show you how deep the rabbit-hole goes." ["Matrix",
written by Andy & Larry Wachowski]
These words, pronounced by the character Morpheus in the "Matrix" movie, illustrate the evocative
power or potency of a pill color. During the last three decades, a dozen studies have dealt with the
evaluation of a drug’s expectancy or potency according to its intrinsic color. But quite a few have
addressed the subject of prescription drug package color effects on consumers’ judgments and
attitudes. The subject is not superfluous because direct-to-consumer (DTC) advertising now
represents an important means of promotion for pharmaceutical companies, knowing that the color
and form of a packaging or a medicine are the only visual aspects or hints presented in print or on
television (see for instance Nexium, "the purple pill" from AstraZeneca). Moreover, mail order
(direct sales) distribution which now represents the second retail distribution channel behind
drugstores (independent and chain outlets combined) relies heavily on the visuals of drugs promoted
on-
line. We propose, in this paper, first to give some key facts about the pharmaceutical industry and
state why drug packaging is assuming a growing role in product promotion. Then we shall review
the few previous studies pertaining to the effects of drug color. Due to space limitations, color
psychology and its applications to other fields of consumer research (print advertising, retail
atmospherics) will not be reviewed in this paper. Finally, we shall present a laboratory experiment
studying attitudes towards prescription drugs, according to the dominant color of a drug package.
The results of this experiment will be discussed and expanded.
THE PHARMACEUTICAL US MARKET
In 2003, about 466.3 billion dollars were spent worldwide on medications (IMS Health, 2004). This
figure includes ethical (prescription
-
only medicine; POM), semi
-
ethical and OTC drug sales. Nearly
Page
1
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
half of this sum (49%) was spent in the USA, 25% in Europe and 11% in Japan. During the same
year, the 291 million US residents spent 203 billion dollars on prescription drugs alone (+ 11.2% vs.
2002), representing 3.22 billion scripts (+ 2.4%; NACDS, 2004). These figures have more than
tripled within ten years (IMS Health; NACDS Economics Department). This 2003 global spending
corresponds to an average of 698 dollars per capita in the USA. Demographic (aging of population),
social (Medicare reform), economic (increasing R&D costs, high levels of domestic prices) and
marketing factors (growing share of mail-order channel and increasing direct promotion
investments) explain this rapid evolution to some extent. Another cause of this escalation is imputed
to direct-to-consumer advertising (Findlay, 2000; Kaiser Family Foundation [KFF], 2003), which is
forbidden in Europe (Cozens, 2002). In this general context, several reasons explain why packages
(for both prescription and OTC drugs) and especially printed color on them are becoming a major
issue in pharmaceutical marketing and justify the present study:
Growing direct-to-consumer advertising (DTCA): in 2003, $3.23 billion were spent on DTC
advertising (Lawrence and Zaugg, 2004). Since the FDA authorized less stringent regulations about
communication in 1997 (Sumpradit, Ascione, and Bagozzi, 2004), television has become the major
medium for drug communication (66% of DTCA in 2003 vs. 13% in 1994). Most observers
recognize that DTCA works: "every additional $1 the industry spent on DTC advertising in 2000
yielded an additional $4.20 in sales" (KFF, 2003) and that it influences physicians’ prescriptions
(Mintzes, 2003). The drug appearance or package on TV (i.e. packshot) then becomes the main
visual and the principal means of differentiation between suppliers.
Drug samples to consumers: sampling is already considered as an efficient promotion tool (Joseph
and Mantrala, 2003) but DTC sampling is about to develop; the packaging could have a greater
impact than standard compulsory information notices.
Packaging standards: although 80% of solid medicines in US are conditioned in bottles (less than
15% in Europe), experts forecast a fast growth for blister packaging (Pilchik, 2000). While the issue
is not prevalent today in the US for prescription drugs, it could soon become, as total world demand
for blister packs should exceed that of bottles in 2007 (Packaging Digest, 2004). A blister is
composed of a thermoformed plastic with a sealing foil. This latter packaging implies a secondary
cardboard package, displaying a larger printable surface.
Changes of drug status: a substantial number of drugs, previously prescribed by physicians, became
free-access OTC medicines. Now some studies report that nearly 3 OTC purchasing decisions out of
4 are made in-store. Packaging has naturally a role to play.
New FDA regulations on OTC drugs: in March 1999 new regulations (effective April 2002) urged
manufacturers "to standardize their presentation of such information as active ingredients, directions,
uses, warnings, and other data. [] The information must be boxed and cannot include logos, graphics,
or bar codes." (Canale, 2001). Measures against tampering are also being upgraded. All these
dispositions entail a need for larger packaging.
Weapon of massive differentiation: knowing that the color or the form of a medicine per se cannot
be automatically patented or constitute a trademark (for instance, see INTA, 1996; Steele, 2002), a
specific packaging may convey and reinforce brand equity (especially for OTC drugs on drugstores’
shelves).
Economic worries: more and more counterfeiting and smuggling cases (low-priced export drugs
which are reintroduced into the US) are reported. Losses for the pharmaceutical industry appear
significant. For instance, GlaxoSmithKline felt compelled to change the color of its Combivir7
tablets (from white to red; Murray-West, 2003) aimed at developing countries. Special printing
techniques (bright stocking, nude labeling) are encouraged for limiting illegal practices.
Page
2
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Health issues: experts recognize that the colorful presentation of a medicine (compound or package)
may improve compliance (adherence to treatment) and reduce the risks of confusion (Carter, Taylor,
and Levenson, 2003; Elwyn, Edwards and Britten, 2003) and medication errors (Hethcock, 1978)
which are counted yearly by the thousands.
Order mail growth: Web-based drug selling activities display the highest growth rates (+21.7%
between 2001 and 2002) in the sector. Drug and package visuals are paramount in this medium.
THEORETICAL FRAMEWORK AND HYPOTHESES
Since the very first days of medicine, the power of a drug appearance and its attached expectancy has
played a significant role on therapeutic success. The placebo effect ("I will please" in Latin) has been
recognized since Socrates (Moerman and Jonas, 2002). The very belief in the physician and/or the
treatment contributes to the cure. Among the manifest signs constructing this belief, the drug
appearance (color and form) and its packaging, may have some influence in efficacy of treatment
(Buckalew and Coffield, 1982a). In the same way, the very brand name of a drug will have
differential therapeutic effects according to its consonance (Klink, 2002) or its notoriety
(Branthwaite and Cooper, 1981).
Previous studies pertaining to prescription drugs and color
Surprisingly, to our knowledge, no specific study addressed the packaging color for pharmaceutical
products as a predictor of drug expectancies. On the other hand, a few studies have been conducted
on the influence of pill color on attitudes towards medication or treatment. We reviewed eight
studies, conducted between 1970 and 1991, to wich we added two extra studies found and reviewed
by de Craen et alii (1996). Most reviewed studies employed either patients or students with limited
sample sizes. Results are sometimes divergent or weakly significant. It nevertheless appears that
color does influence perception of medicine potency or expectancies towards it.
A first group of studies addressed the relationships between color and form of medicine and their
perceived potency and efficacy. Schapira et al. (1970) showed that anxiety was reduced with green
pills and depression with yellow tablets. Cattaneo, Lucchelli and Filippucci (1970), along with
Lucchelli, Cattaneo and Zattoni (1978) demonstrated that blue pills induced quicker and longer sleep
than orange pills. Similar reports were given by Blackwell, Bloomfield and Buncher (1972),
indicating sedative effects of blue capsules. Huskisson (1974) showed that a red placebo is as
efficient as a real analgesic drug. Sallis and Buckalew (1984) demonstrated that the perceived
potency of a drug decreased in function of the following pill color order: red, black, orange, yellow,
green, blue and white.
A second group of studies dealt with the relationships between drug colors and perceived therapeutic
classes. Jacobs and Nordan (1972) showed that red and yellow placebo pills were classified as
stimulants, while a blue placebo was classified as a depressant or a tranquilizer. In two successive
studies, Buckalew and Coffield (1982a; 1982b) demonstrated that some ethnic and cultural factors
could alter general color classifications of medicines. Significant differences between African
American and European American samples were also noted for pill size-strength relationships.
Finally, Buckalew and Ross (1991) revealed that only a few colors presented some obvious link with
therapeutic classes: beige and orange for skin treatment and red for heart condition.
A comprehensive review of literature on the placebo effect of color was also conducted by de Craen
et alii (1996) that covered about three decades. The main conclusions of these authors were that "the
colour of drug seems to influence its effectiveness, but consistent trends are not apparent". They
concluded by writing that further research "contributing to a better understanding of the effect of the
colour of drugs" was warranted. Some authors explain the color placebo effect by the physiological
effects of color (e.g. Jacobs and Hustmyer, 1974) while others merely associate it with idiosyncratic
Page
3
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
color preferences (Schindel, 1962) or learned cultural symbolism (Adams and Osgood, 1973).
Hypotheses
Given the absence of previous works on packaging color and perceived drug expectancy and
potency, we shall set forth our hypotheses by relying on previous studies exploring relations between
these perceptions and the drug color. We therefore hypothesize:
H1: a "warm-colored" (red or yellow) packaging will be perceived as
containing a more potent drug than "cool-colored" (blue or green) ones.
This would impact various expectancies related to a given medicine.
Indeed, some authors like Berlyne (1960) or Jacob and Hustmyer (1974)
have stressed the arousing nature of long visible wavelengths.
H2: a dark packaging will be assessed as containing a more potent drug
than a light one. This would impact various expectancies related to a given
medicine. As early as the 1950’s, authors underlined the potency effect of
dark or saturated colors (Osgood, Suci and Tannenbaum, 1957).
H3a: "warm color" packaging will be more often associated with stimulant
therapeutic classes. Recurrent studies indeed showed associations between
red and stimulant qualities or drugs (Jacobs and Nordan, 1972).
H3b: conversely, "cool color" packaging will be more often associated with
sedative/soothing therapeutic classes (Lucchelli, Cattaneo and Zattoni,
1978).
METHOD
The main objective was to assess the impact of a drug packaging color on related perceptions and
expectancies. A pilot study revealed that general beliefs about prescription drugs as a whole were
likely to moderate perceptions and attitudes toward a specific packaging. A specific measurement
instrument for these beliefs was consequently warranted. Horne, Weinman and Hankins (1999) have
constructed a specific scale, the 'Beliefs about Medicines Questionnaire’ (BMQ) that was primarily
intended for patients with heavy chronic pathologies (psychoses, renal dialyses and heart conditions),
while Perrien et al. (1998) used a general involvement scale for an analgesic. Therefore, a more
general index targeting consumers (ICOMED) was created in another study and used in the present
one. This experiment consisted in exposing 150 European participants to a prescription drug
packaging bearing a specific color and asking them to make several judgments about the drug’s
perceived qualities.
FIGURE 1
PACKAGING USED IN EXPERIMENT (BROWN CONDITION)
Independent variables
Independent variables comprise packaging color (hue and brightness), the pre-test mood, gender and
general beliefs towards medicines.
Stimuli selection
Packaging form and appearance. In Europe, most solid medicines (pills, tablets, etc.) are packaged in
Page
4
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
blisters and boxes rather than in bottles (about 85% against 20% in the US; Pilchik, 2000). This
blister, composed of PVC and foil, is usually kept by the consumer in its box. This form of
packaging is now increasing significantly in the US (+16% annually vs. 6% for bottles). Also, for
reasons of ecological validity and for the sake of practicality, an actual drug cardboard packaging
was tested, i.e. a rectangular box (Cf. Figure 1), with a fictitious brand name.
The packaging of an actual generic analgesic drug was used for this experiment. Once scanned, the
box image was altered in several ways: the original brand name was erased and replaced by a
fictitious but credible name (cf. infra); the "paracetamol" mention was erased because of its popular
notoriety; the background colorsBexcept for the white or gray partsBwere modified and switched to
a condition color.
Choice of color conditions. In this experiment, 7 conditions were applied: 6 hues and an achromatic
color (medium gray). The colors displayed on a LCD computer screen are showed in Table 1, along
with their RGB and HSL (hue, saturation and lightness/brightness) references. Absolute lightness
levels are also indicated. The RGB reference displays the phosphors’
intensity values (between 0 and
255) for the three primary colors red, green and blue. The second HSL norm specifies hues in
degrees (0ƒ to 360ƒ) on the chromatic wheel while saturation and lightness are expressed in
percentages.
A given color can be defined by its three dimensions hue, brightness and saturation (Mounts and
Melara, 1995). The authors opted here for "natural" colors, i.e. hues that are often seen on actual
pharmaceutical packages. Thus, a strict control for color saturation levels was not implemented and
only hue and lightness levels were taken into account for results.
Choice of the drug brand name. So as to assess strictly the chromatic effect of a pharmaceutical
packaging, an unknown drug brand name was warranted to avoid any familiarity effect (Kent and
Allen, 1994). For additional reasons of intellectual property and trademark protection, we decided to
use a fictitious but credible brand name (see Perrien et alii, 1998). A preliminary approach, resorting
to an expert group (physicians, pharmacists and nurses), a documentary research [MedScape Drug
Info: http://www.medscape.com/druginfo/;
DMOZ:http://dmoz.org/Health/Pharmacy/Drugs_and_Medications/] and a quantitative survey
among students have allowed determining that drug brand names were often perceived or imagined
with rare consonants in the native language and with diphthongs which evoke a foreign country or
intrinsic qualities. For example, at least 66 drug brand names start with the letter Z. This latter seems
to communicate and connote a concept of efficacy (Erlich, 1995; Klink, 2003). These statements led
the authors to invent the brand name "
Zolgan" which recapitulated the various characteristics evoked
supra. Verification on official Web sites helped to ensure the inexistence of such a brand name. We
shall observe however that this fictitious brand appears realistic to the extent that a consultation at
the Drugs@FDA [http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm] site allowed
detecting 211 brand names (plus 264 generic appellations) which contained the syllable "
Zol" and 42
brand names (plus 5 generic names) which contained the syllable "Gan". Eventually, the modified
packaging comprised the fictitious brand name (see Figure 1), an upper-right pictogram symbolizing
a capsule, a central hexagon, the active molecule "dextropropoxyphene", the mention "20 capsules"
and the corporate name of the manufacturer (Irex, a subsidiary of Synthelabo). The height of
package was diagonally halved, the left part remaining untouched (white) and the right part showing
the experimental hue.
Mood before test. Mood of participants was assessed before exposure to stimuli by the Self-
Assessment Manikin (SAM) scale, comprising pleasure, arousal and dominance 9-point sub-scales
(Morris, 1995).
Attitudes and beliefs towards prescription drugs. A specific scale
B
developed and validated in a
Page
5
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
European context (submitted paper)Bwas used in this study. The ICOMED index (for Index of
COnfidence toward MEDicines) is composed of two opposed 7-item sub-scales: the first one
measures the individual’s defiance (distrust) level towards prescription drugs and the second
subscale measures the individual’s reliance (trust) level towards prescription drugs. Each item
consists in a 7-point scale.
Dependant variables
Two main variables, likely to be influenced by color, mood and general drug attitude, were measured
in this experiment: the expectancies towards the displayed drug packaging and its possible
attribution to a specific therapeutic class (resultant mood is also a dependent variable).
Drug expectancies. Ten items were successively proposed, displayed as 7-point semantic differential
scales (Osgood, Suci and Tannenbaum, 1957). Half of the scales were score-inverted. These items
comprised: medicine power (gravity of condition), drug activity duration, required precautions of
use, rapidity of action, perceived dearness (price), therapeutic efficacy, potential side effects,
prescription drug (as opposed to OTC), type of treatment (symptomatic or curative) and brand
identity ("genericness").
Attribution to therapeutic classes. Referring to previous studies about drug expectancies according to
the color of pill or capsule (see supra), we tried to confirm some relationships between specific hues
and ailments or therapeutic classes. Eight major categories were proposed to participants:
heart/blood pressure (cardiac), digestion/liver (heartburn etc.), inflammation/fever (antipyretics),
pain/migraine (analgesics), respiratory system, depression/anxiety (psychotropic/stimulant),
insomnia (hypnotics/sleep pill) and skin. Participants could make only one choice.
TABLE 1
COLOR REFERENCES FOR EXPERIMENTAL CONDITIONS
TABLE 2
ICOMED SCORES FOR SAMPLE
Measures and procedures
Experimental sessions took place with small groups composed of 2 to 4 individuals (totaling 150
participants; 53.7% female; X=19.96 years), to whom a flat 17" LCD screen was displayed at a
distance of about 1.2 meters. To avoid any discrepancy in color display, the same screen was used
for all groups. A 6-page booklet was given to each participant. In a first stag, subjects self-evaluated
their current mood with the Self-Assessment Manikin (SAM) iconic scale, composed of three 9-
point
scales. The experimenter then explained: "a pharmaceutical company wants to market a new drug
within a few months. Several packaging layouts have already been selected. You are going to assess
one. Look at the screen displaying the packaging while answering the following items regarding this
medicine". The packaging image was then displayed on-
screen (only one color per condition). While
looking at the packaging, participants evaluated it and rated the associated drug expectancies.
Participants would afterwards assign the package to one possible therapeutic class among eight. A
second mood evaluation was accomplished with the SAM iconic scale before subjects expressed
their beliefs towards drugs in general (ICOMED index). As a conclusion, Ishihara chromatic plates
were displayed to control for individuals’ color vision. Demographics were finally recorded.
RESULTS
Manipulation check
Page
6
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Mood. Participants’ mood significantly changed after exposure to stimulus and questionnaire items.
Further covariance analyses did not show any specific effect of mood and its dimensions on drug
attitudes and judgments. Pleasure: before 6.18, after 5.81 (p<.000); Activation: before 5.35, after
4.85 (p<.000) and dominance: before 5.72, after 5.83 (p<.000).
ICOMED index. The participants’ general attitude towards medicines was assessed by the ICOMED
index. Reliability for the defiance scale showed an alpha of .68; alpha for the reliance scale
reached .73. The overall mean score for participants was .65, knowing that the index may vary from
-
6 (high defiance towards drugs) to +6 (high trust towards drugs). A significant gender difference for
the defiance scale [F (1, 149)=10.67; p<.001] and the global index [F (1, 149)=4.12; p<.044) is noted
(see Table 2). Female participants trusted drugs less than male participants did.
Relations between expectancy items. Partial correlations have been calculated (i.e. controlling for
color) so as to assess the strength and direction of putative links between the 10 expectancy items.
Thus, we notice that a drug construed as designed for a benign illness will also be associated with
brief action (r=.47; p=.000; two-tailed), low risk (r=.60; p=.000), low price (r=.52; p=.000), limited
efficacy (r=.39; p=.000), limited side effects (r=.58; p=.000), OTC status (r=.40; p=.000) and
symptomatic treatment (r=.50; p=.000). Correlations with rapidity (vs. delayed) of action (r=.10) and
"
genericness" (r=.11) are not significant (p>.16). A further factor analysis showed that 8 items out of
10 loaded on the same factor. The rapidity of action and brand identity ("genericness") were
considered as different constructs. The remaining 8 items indicated an eigenvalue of 3.975,
explaining 49.7% of variance, with all loadings above .5. The aggregation of these 8 items
constituted a "drug potency" index, which showed a good reliability with a Cronbach alpha of .8507.
TABLE 3
MEAN POTENCY SCORE BY COLOR CONDITION
Color effects on dependent variables
Six hues were used in the experiment, along with a neutral condition (medium grey). Another
independent variable was constituted post hoc in a dual way: the stimulus brightness. Two categories
were constituted after agreement between experts: light (yellow, green, orange and grey) and dark
(red, blue and brown). This choice was validated by a metric measure of color brightness which was
taken with the help of PhotoShop 7.0 imaging software. The respective effects of hue and brightness
were analyzed for the three main dependent variables: drug expectancies (10 items), the drug
potency (8 aggregated items) and the attribution to a specific therapeutic class.
Drug expectancies. An analysis of variance showed a main effect of color hues on some drug
expectancies. Main effects of hues were significant on three expectancies: "medicine power" [gravity
of illness; F (6, 149)=2,635; p=.019], "required caution" [F (6, 149)=2,558; p=.022] and "drug
dearness" [drug price; F (6, 149)=3,210; p=.005]. Red, brown and grey packages are perceived as
designed for serious illnesses vs. yellow or green packaging. Brown, red and orange packages
require some precaution of use, compared to blue, green and yellow. Brown- and red-
packaged drugs
are perceived as more costly than orange or yellow boxes.
As far as package brightness is concerned (light vs. dark hues), significant effects were detected for
the following expectancies: "drug delayed action" [F (1, 149)=4.05; p=.046], "drug dearness" [F (1,
149)=9.49; p=.002], "side effects severity" [F (1, 149)=3.8; p=.05] and "drug curative value" [F (1,
149)=5.65; p=.019]. Dark-hued packages (red, blue and brown conditions) are considered as acting
more rapidly, more expensive, more susceptible of side effects and more curative than light-hued
packages (yellow, green, orange and grey conditions). H2 is validated.
When effects of color screen brightness (metric values) on drug expectancies were assessed by linear
Page
7
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
regressions, it appeared that absolute screen color lightness (varying from a minimum of 0BblackB
to
a maximum of 255Bwhite; see Table 1) impacted significantly drug power (beta=-.171; p=.037),
drug action duration (beta=-.175; p=.033), activity delay (beta=+ .172; p=.036), drug dearness
(beta=-.23; p=.005) and curative value (beta=-.163; p=.047).
Perceived drug potency (8-item summated score). The packaging color hue had an impact on the
global perception of the drug we called potency [F (6, 149)=2.35; p=.034]. Brown and red packages
entail greater potency scores compared to green or yellow hues (see Table 3). H1 is validated.
When experimental hues were rearranged in two brightness categories, a significant positive effect of
brightness on perceived potency was also apparent [F (1, 149)=7.27; p=.008]. Brightness absolute
levels were also employed as a metric variable. A linear regression shows a significant relationship
between brightness and perceived drug potency (R2=.052; beta=-.228; p=.005).
Therapeutic class attribution. Participants had been given no particular indication regarding the
precise nature of the proposed drug. Eight therapeutic classes were proposed, knowing that only one
could be chosen. Overall, the most often chosen classes were analgesics, drugs for heart condition,
antidepressants and hepatic drugs. Chi-square tests did not reveal any specific relationship between
hues and therapeutic applications. Red is mostly attributed to the heart condition drug (32%) and
analgesics (20%). Yellow is attributed mostly to heart (23%) or dermatologic (23%) drugs. Green is
related to analgesic (25%) and hepatic (21%) medicines. The blue package is affected to an analgesic
drug (26%), while the brown one is related to a heart condition drug (33%). The neutral package
(gray) is partly attributed to an analgesic drug (25%). Regarding the attribution of light or dark
packages, one association appears significant: dark vs. light packages are related to heart condition
drugs (z=2.794; p=.005). Another relation approaches significance: light packages are more related
to antipyretics (z=-1.83; p=.06). Although non-significant, results show trends compatible with H3a
and H3b.
DISCUSSION
Given the scarcity of empirical research on pharmaceutical packaging and its growing importance in
global drug companies’ communication, and given the known visual impact of color, an exploratory
approach on drug packaging color seemed warranted and justified.
Indeed, previous works about design were mainly conceptual papes pertaining to product design as a
whole (Bloch, 1995), dealing with product form rather than packaging color per se (e.g. Underwood,
Klein and Burke, 2001). Other authors addressed the subject of which design attributes (e.g.
prototypicality, unity) influence consumers’ cognitive and affective responses (Veryzer and
Hutchinson, 1998). Specific studies on packaging color are rare (Gordon, Finlay and Watts, 1994;
Garber and Hyatt, 2003) and thus warranted, even limited to an occidental context.
The main objective of this study was to assess the presumed effect of the color of a drug packaging
on subjective evaluations pertaining to that drug. The present results indicate that packaging color
does have an effectBin a European contextBon some expectancy items related to a given drug
(strength, safety measures and price) and especially on the "potency" construct (which is composed
of the 8 specific items abovementioned) which presents a good reliability (alpha=.85). Color warmth
also seems to imply drug potency, as being comparable to a color arousing quality (Berlyne, 1960;
Jacobs and Hustmyer, 1974).
The brown and red hues appear to signify and connote gravity in a supposed treatment, with a high
potency score (respectively 38.3 and 36.8) as opposed to green and yellow hues which are more
associated to trivial or limited effects (both 31.2 for potency score). Similarly, dark tones generally
induce more potent considerations: the darker the package, the more potent the drug; this seems
compatible with previous findings on color perceptions and meanings (Osgood, Suci and
Page
8
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Tannenbaum, 1957; p 299-302) and color effects on emotions (Valdez and Mehrabian, 1994). This
lightness effect is corroborated by results of regressions on metric lightness values of employed
package colors. Although not always reaching significance levels, some trends regarding therapeutic
class attribution by color are similarBeven partlyBto those obtained previously (Buckalew and
Coffield, 1982a; Schapira et al., 1970). Thus, reddish hues (brown and red) are found to be
associated with heart condition drugs, while yellow is related to skin medicines (Buckalew and Ross,
1991). Nevertheless, we did not find any confirmation of white (i.e. achromatic) association with
analgesics and green/blue with sedative drugs (Buckalew and Coffield, 1982a; 1982b; Blackwell,
Bloomfield and Buncher, 1972). A gender effect also appears in our results. Women appear to
distrust medicines more than men do. Color preferences among gender were not controlled for in this
study. Yet, specific preferences might play a role, similar to those exposed by Cattaneo, Lucchelli
and Filippucci (1970; study 2), where men preferred orange (warm) capsules and women preferred
blue (cool) ones. The ICOMED index does not seem to influence the participants’ ways to evaluate
color packages. However, the relative youth (and health condition) of participants (mean age=20)
may explain an absence of mediation. We may add that ICOMED was also tested elsewhere on
adults (30-50 years old) and seniors (65 to 80) and that it presented higher reliability levels and
better fitness indices in confirmatory factor analyses. Links between medicine beliefs and health
involvement should be explored (e.g. Perrien et al., 1998). But again, it should be noted that
significant differences in ICOMED appear according to gender: female participants scored higher on
the defiance scale (p<.001) and lower on the global index (p<.05) than their male counterparts did.
Implications for future research. Some additional packaging features have not been addressed in this
experiment and some experimental extensions seem warranted. For instance, package graphics or
picture (Underwood, Klein and Burke, 2001), letter fonts and styles, cardboard texture, along with
brand name sounds (Klink, 2003; Yorkston and Menon, 2004) would deserve a further exploration.
A similar experiment is to be submitted to healthy adults and senior patients. Recurrent treatments
for chronic illnesses or ailments are likely to impact attitudes towards specific medicines and their
packaging. Also, individual characteristics such as aesthetic sensitiveness (Bloch, Bruneland Arnold,
2003) or style of processing (Childers, Houston and Heckler, 1985) should be included as
moderating variables in future studies. Another question is whether cultural color meanings impact
drug expectancies in line with packaging hues (Madden, Hewett and Roth, 2000).
Implications for business research. Pharmaceutical businesses should heed the form and color of
their new packages before marketing, be they prescribed or OTC. The impact of color on drugstores’
shelves is obvious (see for example Celebrex or Prilosec), but the likely development of blister-
packaged prescribed drugs, along with FDA compulsory notices, will also stress the growing
importance of boxes’ general outlay, either in physical outlets or on-line (189 million scripts by mail
order in 2003). The expanding R&D costs and shorter ROI periods imply a maximization of market
shares and profits during the market life of a drug; an attractive and meaningful package can then
make the difference against a more ordinary and conventional container.
REFERENCES
Adams, Francis M., and Charles E. Osgood (1973), "A Cross-Cultural Study of the Affective
Meanings of Color," Journal of Cross Cultural Psychology, 4, 2, 135-157.
Berlyne, Daniel E. (1960), Conflict, Arousal and Curiosity, New York: McGraw Hill Co.
Blackwell, Barry, Saul S. Bloomfield, and C. Ralph Buncher (1972), "Demonstration to Medical
Students of Placebo Responses and Non-Drug Factors," The Lancet, 19, 10 (June), 1279-1282.
Bloch, Peter H. (1995), "Seeking the Ideal Form: Product Design and Consumer Response", Journal
of Marketing, 59, 16
-
29.
Page
9
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Bloch, Peter H., Brunel Frederic F. and Arnold Todd J. (2003), "Individual Differences in the
Centrality of Visual Product Aesthetics: Concept and Measurement", Journal of Consumer Research,
29, 4, 551-565.
Branthwaite, A., and P. Cooper (1981), "Analgesic Effects of Branding in Treatment of Headaches,"
British Medical Journal, 282, 16 (May), 1576-1578.
Buckalew, Louis W., and Kenneth E. Coffield (1982a), "An Investigation of Drug Expectancy as
Function of Capsule Color and Size and Preparation Form," Journal of Clinical
Psychopharmacology, 2, 4, 245-248.
Buckalew, Louis W. (1982b), "Drugs Expectations Associated with Perceptual Characteristics:
Ethnic Factors," Perceptual and Motor Skills, 55, 915-918.
Buckalew, Louis W., and Sherman Ross (1991), "Medication Property Effects on Expectations of
Action," Drug Development Research, 23, 101-108.
Canale, Colleen M. (2001), "Regulations Spark Package Redesign", Pharmaceutical & Medical
Packaging News Magazine, February.
Carter, Sarah, David Taylor and Ross Levenson (2003), "A Question of Choice: Compliance in
Medicine Taking", Medicines Partnership.
Cattaneo, Angelo D., Paolo E. Lucchelli, and Giorgio Filippucci (1970), "Sedative Effects of
Placebo Treatment," European Journal of Clinical Pharmacology, 3, 43-45.
Childers, Terry L., Michael J. Houston and Susan E. Heckler (1985), "Measurement of Individual
Differences in Visual and Verbal Information Processing", Journal of Consumer Research, 12, 2,
125-134.
Cozens, Claire (2002), "Europe rejects Drug Advertising," The Guardian, Wed. October 23,
Guardian Newspapers Limited, London.
De Craen Anton J.M., and alii (1996), "Effect of Colour of Drugs: Systematic Review of Perceived
Effect of Drugs and of their Effectiveness," British Medical Journal, 313, 1624-1626.
Elwyn Glyn, Adrian Edwards and Nicky Britten (2003), "Doing Prescribing: How Doctors Can Be
More Effective," British Medical Journal, 327 (October), 864-867.
Erlich, John (1995), "Giving Drugs a Good Name: When it Comes to Marketing Pharmaceuticals,
The Placebo Effect Extends Even to the Label," New York Times Magazine, September 3, SM 36-
37.
Findlay, Steven (2000), "Prescription Drugs and Mass Media Advertising," Research Brief, NIHCM
Foundation, September (Washington, DC).
Garber Lawrence L. and Eva M. Hyatt (2003), "Color as a Tool for Visual Persuasion", in Scott
Linda M. & Batra Rajeev (eds), Persuasive Imagery: A Consumer Response Perspective, Hillsdale,
NJ: Lawrence Erlbaum Associates Inc.
Gordon Ariane, Karen Finlay and Trevor Watts (1994), "The Psychological Effects of Colour in
Consumer Product Packaging", Canadian Journal of Marketing Research, 13, 3
-
11.
Page
10
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Hethcock, James M., (1978) "Similarities in Color-Coded Drug Product Packages", American
Journal of Hospital Pharmacy, 35, 8, 901.
Horne, Rob, J. Weinman, and M. Hankins (1999), "The Beliefs About Medicines Questionnaire:
Development and Evaluation of A New Method for Assessing the Cognitive Representation of
Medication," Psychology and Health, 14, 1-24.
Huskisson, E.C. (1974), "Simple Analgesics for Arthritis," British Medical Journal, 4, 196-200.
Lawrence, Mary Beth and Ana-Maria Zaugg (2004, "IMS Health Report, Bruised but Triumphant",
Medical Marketing & Media, May 2004, 37-50.
International Trademark Association (1996), Protection of Color Trademarks, November, INTA:
New York, NY.
Jacobs, Keith W., and Frances M. Nordan (1979), "Classification of Placebo Drugs: Effects of
Color," Perceptual and Motor Skills, 49, 367-372.
Jacobs, Keith W. and Frank E. Hustmyer Jr. (1974), Effects Of Four Psychological Primary Colors
on GSR, Heart Rate And Respiration Rate, Perceptual and Motor Skills, 38, 3, 763-766.
Joseph, Kissan, and Murali K. Mantrala (2003), "Prescription Drug Promotion: The Role and Value
of Physicians’ Samples under Competition," Working paper, School of Business, University of
Kansas.
Kent, Robert J., and Chris T. Allen (1993), "Does Competitive Clutter in Television Advertising
'Interfere’ with the Recall and Recognition of Brand Names and Ad Claims ?," Marketing Letters, 4,
2, 175-184.
K.F.F. (2003), "Impact of Direct-to-Consumer Advertising on Prescription Drug Spending",
publication Nƒ 6084 (June), Menlo Park, CA: The Henry J. Kaiser Family Foundation.
Klink, Richard R. (2001), "Creating Meaningful New Brand Names: A Study of Semantics and
Sound Symbolism," Journal of Marketing: Theory and Practice, 9 (Spring), 27-34.
Lucchelli, Paolo E., Angelo D. Cattaneo, and Jacomo Zattoni (1978), "Effect of Capsule Colour and
Order of Administration of Hypnotic Treatments," European Journal of Clinical Pharmacology, 13,
153-155.
Madden Thomas J., Kelly H. Hewett and Martin S. Roth (2000), "Managing Images in Different
Cultures: A Cross-National Study of Color Meanings and Preferences", Journal of International
Marketing, 8, 4, 90-107.
Melliship, Neil P. (2003), "Color and Shape of Pills Not a Trademark," INTA Bulletin, Law and
Practice, 58, 16 (September).
Mintzes Barbara et alii (2003), "How Does Direct-To-Consumer Advertising (DTCA) Affect
Prescribing? A Survey in Primary Care Environments With and Without Legal DTCA," Canadian
Medical Association Journal, 169, 5, 405-412.
Moerman, Daniel E., and Wayne B. Jonas (2002), "Deconstructing the Placebo Effect and Finding
the Meaning Response," Annual Internal Medicine, 136, 471
-
476.
Page
11
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Morris, Jon D. (1995), "ObservationsBSAM: The Self-Assessment ManikinBAn Efficient Cross-
Cultural Measurement of Emotional Response," Journal of Advertising Research, 35, 6, 63-68.
Mounts, Jeffrey R.W., and Robert D. Melara (1995), "Classification of Color Dimensions in
Multiple Contexts," Journal of Experimental Psychology: Human Perception and Performance, 21, 2,
257-274.
Murray-West, Rosie (2003), "Glaxo Takes on Smugglers by Changing Drug’s Colour," The
Telegraph, December 27, Telegraph Group Ltd.
NACDS (National Association of Chain Drug Stores) (2004),
"
2002 Community Pharmacy Results",
[http://www.nacds.org/, last accessed September 22, 2004].
Osgood, Charles E.., George J. Suci, and Percy H. Tannenbaum (1957), "The Measurement of
Meaning", Urbana IL, University of Illinois Press.
Packaging Digest (2004), "Packaging diagnosis: growth through 2007", Interphex supplement,
February issue [http://www.packagingdigest.com/articles/200402/intsup2-1.php; last accessed
August 8
th
, 2004].
Perrien, Jean et alii (1998), "Exploring The Persuasive Effects Of A Commercial For A
Pharmaceutical Product: The Elderly Vs. Young Adults", Advances in Consumer Research, 25, 513-
517.
Pilchik, Ron (2000), "Pharmaceutical Blister Packaging, Part IBRationale and Materials,"
Pharmaceutical Technology, (November), 68-77.
Sallis, Robert E., and Louis W. Buckalew (1984), "Relation of Capsule Color and Perceived
Potency," Perceptual and Motor Skills, 58, 3, 897-898.
Schapira, Kurt, and alii (1970), "Study on the Effects of Tablet Colour in the Treatment of Anxiety
States," British Medical Journal, 8, 1, 446-449.
Schindel, Leonard E. (1962), "Placebo in Theory and Practice", Antibiotica et Chemotherapia
Advances, (Basel), 10, 398-430.
Steele, Alexandra (2002), "The colour and shape of a pharmaceutical tablet deemed not distinctive
enough to warrant trade-mark registration, federal court rules", Leger, Robic and Richard Lawyers,
Montreal, Quebec (http://www.robic.com/publications/Pdf/142.134.pdf).
Sumpradit, Nithima, Frank J. Ascione, and Richard P. Bagozzi (2004), "A Cross-Media Content
Analysis of Motivational Themes in Direct to Consumer Prescription Drug Advertising," Clinical
Therapeutics, 26, 1, 135-154.
Underwood Robert L., Noreen M. Klein and Raymond R. Burke (2001), "Packaging
communication: Attentional effects of product imagery", The Journal of Product and Brand
Management, 10, 6/7, 403-422.
Valdez, Patricia, and Albert Mehrabian (1994), "Effects of Color on Emotions," Journal of
Experimental Psychology: General, 123, 394-409.
Veryzer, Robert W. and J. Wesley Hutchinson,
"
The Influence of Unity and Prototypicality on
Page
12
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
Aesthetic Responses to New Product Designs", Journal of Consumer Research, 24, 1998, 374-394.
Yorkston, Eric and Geeta Menon (2004), "A Sound Idea: Phonetic Effects of Brand Names on
Consumer Judgments," Journal of Consumer Research, 31, 1, 2004, 43-51.
----------------------------------------
Page
13
sur
13
07/06/2006
http://www.acrwebsite.org/volumes/display.asp?id=9064&print=1
