The n e w e ng l a n d j o u r na l of m e dic i n e

H e a l t h L a w, E t h i c s , a n d H u m a n R i g h t s

Why Do the Same Drugs Look Different? Pills, Trade Dress,

and Public Health
Jeremy A. Greene, M.D., Ph.D., and Aaron S. Kesselheim, M.D., J.D., M.P.H.

Generic drugs are widely considered to be cost- Principle s of Tr ade Dre ss

efficient substitutes for brand-name medications.
They make up about 70% of the total number of A trademark is a name, logo, or imprint that
U.S. prescriptions but less than 20% of the total designates a particular company or product in
prescription-drug costs.1 Although brand-name the minds of consumers. Trade dress describes
drugs and generic drugs are both approved by the any material quality of a product’s packaging or
Food and Drug Administration (FDA) and may physical appearance that serves a branding func-
be interchangeable with respect to their clinical tion.5 Although the name “Coca-Cola” is a word
effects, they can differ substantially in their ap- mark, the design of the iconic Coca-Cola bottle
pearance. Consumers of brand-name medications is an example of trade dress. Trade dress sup-
receive identical-appearing batches of pills with ports a manufacturer’s investment in establish-
each refill, whereas consumers of generic drugs ing a product as distinct from other, similar
must be prepared to receive pills of a different products. Legal recognition of trade dress is eco-
size, color, and shape, depending on which man- nomically efficient for the marketplace because it
ufacturer is supplying their pharmacies.2 For ex- allows consumers to identify a product’s unique
ample, at least 10 generic versions of fluoxetine source and therefore recognize its established
exist (4 of which are shown in Fig. 1) that are quality and reliability. Unlike the 20-year term of
pharmacologically equivalent to the innovator drug a patent, the term of protection for trade dress
but vary in their color patterns. This variation in can be indefinite.
the appearance of generic drugs has its roots in To qualify as trade dress an attribute must
U.S. intellectual-property law. In the past, drug meet three criteria: it must be nonfunctional, it
manufacturers successfully claimed exclusive own- must lead to confusion (or deception) if imitated,
ership of the physical aspects of their products and it must have a secondary association with
— including the size, shape, color, texture, aroma, the product for the consumer. A product’s func-
and flavor — as private property under a subset tional attributes are essential to the use or pur-
of trademark law called “trade dress.”3,4 Such a pose of the product or must affect the cost or
practice constrained the ability of generic-drug quality of the product.6,7 Functionality is key in
manufacturers to design follow-on products that pharmaceutical-related trade dress, because if a
reproduced the physical appearance of the inno- company with a brand-name drug owned exclu-
vator brands. sive rights over a functional attribute of that drug,
In this article, we review the legal basis of a competitor could not offer a truly equivalent ge-
trade dress as it has applied to pharmaceutical neric version.
products and consider the public health implica- For pharmaceutical products, the most obvi-
tions of variations in pill appearance. We then ous functional attributes include their efficacy,
discuss how a system of more uniform drug ap- safety, and quality. In the 1920s, the Supreme
pearance could be designed to reduce medical Court ruled on a trade-dress case that involved
error and promote patient adherence to treat- Eli Lilly’s Coco-Quinine, a formulation of the
ment regimens that involve generic drugs. unpatented antimalarial drug quinine mixed

n engl j med 365;1 nejm.org july 7, 2011 83

The New England Journal of Medicine
Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

drugs in place of brand-name prescriptions, charg-

Prozac (40-mg capsule)
ing the consumer for the brand-name drug while
pocketing the price difference.10 During this era,
antisubstitution laws in over 40 states made it
illegal for a pharmacist to fill a prescription for a
brand-name drug with a product from a different
manufacturer.11
Protection of intellectual property covering
the physical attributes of pills therefore served
Generic forms of fluoxetine two primary purposes. One purpose of trade-
dress protection was to reduce the practice of
palming off. Premo Pharmaceuticals was sued
for trade-dress infringement when it marketed
its generic version of the diuretic hydrochloro-
thiazide/triamterene with a maroon-and-white
capsule identical to that of brand-name drug
Dyazide, produced by Smith, Kline and French.
In SK&F v. Premo, the Third Circuit Court of Ap-
peals upheld trade-dress protection because near-
identical pills would facilitate the practice of
“unscrupulous pharmacists” in “substituting less
expensive generic drugs for the brand name
drugs prescribed without informing their cus-
Figure 1. Differences in the Appearance of the Innovator Drug (Prozac, Lilly) tomers and without passing along the benefit of
and Four Generic Versions of Fluoxetine (40-mg capsules). the lower price.”12 The court also found that the
Images courtesy of Cerner Multum. color scheme was nonfunctional because it did
not help patients identify the drug, pointing to
other maroon-and-white capsules that were not
with chocolate syrup, and a subsequent product diuretics.
with a similar color, taste, and name (Quin-Coco) A second purpose, the courts rationalized,
marketed by a competitor. The Court ruled that was to allow trade-dress protection to serve a
in spite of these similarities, Quin-Coco did not public health function by preventing the substi-
represent trade-dress infringement and could re- tution of a drug that was similar but not identical
main on the market because the chocolate flavor to another.13 In SK&F v. Premo, the two diuretic
did “not merely serve the incidental use of iden- products were chemically equivalent, but their rate
tifying the respondent’s preparation” but rather of absorption into the bloodstream (bioavailability)
“supplie[d] the mixture with a quality of palat- differed. In another case, a federal district court
ability for which there [was] no equally satisfac- in Michigan enjoined a competitor from produc-
tory substitute.”8 Similarly, in the 1950s, the Sec- ing a version of the diet pill phentermine that was
ond Circuit Court of Appeals held that the similar in appearance to a brand-name version
“soothing” pink color of Pepto-Bismol provided because the efficacy of the hydrochloride salt of
“therapeutic value” in treating upset stomachs phentermine in the generic manufacturer’s version
and was therefore not protectable as trade dress.9 did not necessarily match the efficacy of the brand-
name manufacturer’s phentermine resin complex,
Pharmaceu tic al Tr ade Dre ss so the two drugs were not interchangeable.14 No-
tably, both these arguments upholding pharma-
Legal protection for pharmaceutical trade dress ceutical trade-dress rights were meant to protect
expanded in the mid-20th century amid increasing consumers from deception by the producers of
concerns about counterfeit drugs, which copied look-alike drugs.
the appearance, packaging, and name of a brand- An alternative view was offered by a New York
name drug. Corrupt pharmacists deceptively sub- federal district court in Ives v. Darby, a case that in-
stituted (“palmed off”) poor-quality counterfeit volved the peripherally acting vasodilator cyclan-

84 n engl j med 365;1 nejm.org july 7, 2011

The New England Journal of Medicine

Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
 health law, ethics, and human rights

delate (Cyclospasmol). After the patent for cyclan­ which served to identify the brand of Viagra
delate expired, competing products mimicked the with both the color (pale blue) and the shape
brand-name drug’s color schemes: blue for the (diamond) of the tablets.
200-mg capsule, and blue and red for the 400-mg Concurrent Supreme Court rulings in other
capsule. The district court ruled that the compet- industries promised even broader pathways for
ing cyclandelate products could remain on the trade dress. In one ruling that protected the fes-
market because the colors were functional to pa- tive décor of a Mexican restaurant, the Court sug-
tients and their physicians. One example of such gested that some trade dress could be inherently
functionality was that “some patients co-mingle distinctive, despite the lack of evidence that con-
their drugs in a single container and then rely on sumers associated the décor with the restaurant.20
the appearance of the drug to follow their doc- Soon after, in the case of Qualitex v. Jacobson, the
tors’ instructions.”15 In addition, the cyclandelate Court held that color alone could constitute a
competitors could demonstrate biologic equiva- defensible trademark if it had acquired a sec-
lence of their products to Cyclospasmol. When ondary meaning.21 Nonetheless, even under the
the case rose to the Supreme Court, the decision ruling in Qualitex, the color of a pill might not
was upheld on a different basis, although the be protectable if the attribute served “a signifi-
Court noted “the petitioners offered a legitimate cant nontrademark function.”
reason for producing an imitative product.”16
Despite the outcome in Ives, trade dress re- The Rise of Generic Drugs
mained integral to the sales strategies of some
manufacturers of brand-name drugs. For exam- Meanwhile, new developments in the pharma-
ple, AstraZeneca’s omeprazole (Prilosec) was ceutical marketplace began to undercut the legal
widely promoted as “the purple pill” after its foundations previously used to support trade
launch in 1989. As Prilosec’s market exclusivity dress in products such as the purple pill and the
was ending, AstraZeneca launched the prescrip- blue, diamond-shaped tablet. First, in the 1960s
tion-only follow-on product esomeprazole (Nex- and 1970s, a wave of patent expirations for wide-
ium) as “the new purple pill” in 2001 to encour- ly used drugs expanded the potential for gener-
age patients accustomed to taking Prilosec to ic-drug sales among hospitals, pharmacies, and
switch to Nexium. Notably, when AstraZeneca government purchasing agencies. Generic-drug
began to sell omeprazole without a prescription manufacturers evolved from a shadowy group of
as Prilosec OTC, the company changed the color firms pressing tablets in small, nondescript ware-
of its product to salmon pink. Conversely, as houses into a set of legitimate pharmaceutical
Lilly’s green-and-cream capsule fluoxetine (Pro- firms regulated by the FDA.22
zac, 20 mg) faced generic-drug competition in Second, as state and federal government in-
2001, the company repackaged fluoxetine in pink- surance programs took on a larger burden of
and-purple capsules and marketed it as a new drug costs, the value of cost containment through
drug, Sarafem (20 mg), which was approved by the substitution of therapeutically equivalent generic-
FDA in 2000 for the treatment of a new indication drug products led to a reexamination of the an-
— perimenstrual dysphoric disorder.17 In this case, tisubstitution laws. During the 1970s, state laws
the change in color was designed to discourage prohibiting generic substitution were largely re-
physicians from prescribing the less expensive versed and were replaced by laws mandating the
generic fluoxetine in place of Sarafem. interchangeability of brand-name drugs and ap-
The 1997 FDA guidelines for expanding direct- proved generic drugs. This was aided by the
to-consumer (DTC) advertising of prescription 1978 publication of the Approved Drug Products with
drugs further enhanced the power of pharma- Therapeutic Equivalence Evaluations (known as the
ceutical trade dress as broadcast campaigns be- Orange Book) by the FDA, a compendium of all
gan to include images of the pills themselves.18,19 products approved by the FDA that were found
One of the first drugs to be promoted heavily to to be bioequivalent and available for generic in-
consumers after its approval in 1998 was Viagra terchange. Finally, the federal Hatch–Waxman
(sildenafil), Pfizer’s drug for treating erectile Act of 1984 established an Abbreviated New Drug
dysfunction. The company included a picture of Application (ANDA) pathway for FDA approval of
the drug in nearly all the advertisements for it, generic drugs based on bioequivalence to the

n engl j med 365;1 nejm.org july 7, 2011 85

The New England Journal of Medicine
Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

brand-name version. In the aftermath of this ers of so-called authorized generics, which adver-
legislation, the number of prescriptions for ge- tise the characteristic of similar appearance as a
neric drugs began to rise quickly in the United reason for consumers to use these products.
States.23,24 Proponents of trade-dress protection argue
These alterations had substantial implications that consumers need to be able to identify the
for the legal status of pharmaceutical trade dress. brand by the appearance of the pills because the
Concerns about palming off were mitigated after drugs are dispensed in a standard prescription
FDA protocols ensuring bioequivalence of generic bottle.19 Yet this argument is undercut by the
drugs undercut the public health risk that courts success of the FDA’s bioequivalence protocols,
had used to justify trade dress for pill attributes. which have been consistent on a pharmacologic
The new landscape was formally recognized in level27 and translate into comparable clinical ef-
2003 by the Third Circuit Court of Appeals in fectiveness for nearly all brand-name and generic
Shire v. Barr, a case involving the prescription stim- drugs.28 Instead, the existence of generic drugs
ulant mixture of dextroampheta­mine and am- that look different from the brand-name version
phetamine (Adderall), marketed principally for can have important negative effects on patient
the treatment of attention deficit–hyperactivity outcomes in three key areas — prescription er-
disorder (ADHD). When Shire Pharmaceuticals ror, medication adherence, and the contribution
brought this drug to market in 1996, its promo- of the placebo effect — as described below.
tional materials highlighted how differences in It is well known that prescription error can
the color, size, and shape of the various doses result from confusion regarding the appearance
of Adderall promoted the ability of children of a drug. Much of the research into this prob-
with ADHD to adhere to their regimens. Shire lem emerges from the inpatient setting,29 but
sued Barr, the first generic competitor, when it confusion borne by outpatients involving pills
copied Adderall’s distinctive dose–color scheme. with different attributes has been reported as
At trial, Barr argued that Shire’s own claims es- well.30,31 Although all pills are imprinted with
tablished the functionality of the purported trade an identifying code, these codes are indecipher-
dress. The court agreed, finding that because of able by most patients, as well as by many physi-
their functionality, the color, size, and shape of cians.32 Confusion about pills may be exacerbat-
Adderall were nonprotectable.25 ed as a result of regimen complexity or among
patients with limited health literacy.33 Patients
Tr ade Dre ss and Public He alth who take multiple medications are often elderly,
with higher rates of visual or cognitive impair-
The Shire v. Barr case unraveled the legal protec- ment, which increases the risk of errors.34-36 Re-
tion that previously supported pharmaceutical cently, color and shape differentiation have been
trade dress.18 Since then, certain Supreme Court shown to be important factors in patients’ cor-
cases have further demarcated its boundaries.26 rect identification of over-the-counter drugs.37
For example, the Court held that trade dress did The World Health Organization has estimated
not apply to functional aspects of temporary road that as few as half of all drug regimens for the
signs even when alternative designs existed that treatment of chronic diseases are optimally ad-
could perform a similar function.6 Thus, the hered to.38 As the public health problem of non-
Court clarified that competitors were indeed free adherence became increasingly recognized,39 the
to copy features that increased the utility of a use of DTC advertisements that included images
product. of pills were defended as increasing adherence
However, claims of trade dress remain vital to medication regimens for chronic diseases.40
in the pharmaceutical market.19 With increasing Although no direct research has been carried out
generic competition, trade-dress strategies are on this issue, claims linking pill color and shape
described in industry publications as ways for to adherence by patients with ADHD were central
innovator firms to retain market share for their to the case of Shire v. Barr. Particularly during the
products after their patents and market exclu- transition period after brand-name drugs lose
sivity expire.18 During at least the past 5 years, their market exclusivity, consistency in appearance
brand-name pharmaceutical companies have be- between brand-name drugs and generic drugs
gun to license their trade dress to manufactur- could help promote patient adherence.

86 n engl j med 365;1 nejm.org july 7, 2011

The New England Journal of Medicine

Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
 health law, ethics, and human rights

Finally, a resurgence of research on the place- have similar appearances. Where these drugs do
bo effect suggests that drug appearance can have differ (e.g., as in dyes, fillers, or excipients), phy-
a distinct functionality. For decades, studies have sicians or pharmacists could still locate manu-
shown that the efficacy of placebo pills varies facturer data from unique identifier codes em-
according to the size, shape, and color of the bossed on pills. Further public health benefits
pills.41-44 The placebo effect is particularly evi- could emerge if the reduction in trade dress
dent in the treatment of patients whose disorder helps to combat the physician’s persistent use of,
has potential psychosomatic components, such and the patient’s preference for, costly brands
as anxiety, depression, dyspepsia, impotence, obe- when generic equivalents are available.49
sity, and pain.45,46 Newer research suggests that The obvious limitation of this approach is that
placebo efficacy varies with external packaging it would apply only to newly introduced pharma-
and the perceived dollar value of the treatment ceutical products, leaving most of the existing
being applied.47 Although the classic logic of therapeutic armamentarium unaffected. There-
the randomized, controlled trial casts the place- fore, we suggest that a rational scheme be cre-
bo effect as a negative foil for measuring thera- ated for pharmaceuticals that have already been
peutic efficacy, in practice a drug’s effectiveness is approved whereby each distinct agent could be
still due, to some extent, to placebo effects. By identified by a combination of its size, shape,
not allowing a generic version to fully benefit and color. An example of such a scheme is the
from the functionality of such effects, differing successful introduction in the United Kingdom
appearances may reduce the ultimate effective- of color-coding for metered-dose inhalers. Pa-
ness of certain generic drugs. tients with asthma had frequently confused bron-
Literature that measures the magnitude of pre- chodilators with steroid inhalers,50 leading the
scription error, medication adherence, and pla- National Health Service to systematize inhaler
cebo effects attributable to generic or brand pill appearance: all short-acting inhalers (bronchodi-
appearance is limited; explicit outcomes research lators) became blue and all preventive agents
in this area is still lacking. No one knows how (steroids) became brown, orange, or burgundy.51
many medication errors are due to problems re- A similar color-coding scheme was piloted in
lated to visual cues. Nonetheless, it is clear that the United States for ophthalmologic products,
the external attributes of pills can have benefits in which the caps on generic preparations of
separate from the therapeutic effect of their ac- atropine, pilocarpine, and other drug products
tive ingredients. It follows that maintaining some having multiple strengths were color-coded to
consistency of these effects is important to ensure match those of the innovator-drug products.52
the overall equivalency of brand-name drugs and Introducing a color-coding scheme for the en-
generic drugs. tire pharmacopoeia may require additional legis-
lation to ensure full compliance and would place
P olic y Recommendations an additional cost on generic-drug manufactur-
ers to retool production of oral forms of their
If brand-name pharmaceutical manufacturers are drugs. However, to begin the process of policy-
no longer able to rely on trade dress to protect making, the FDA could formally recognize the
the attributes of their products, federal policies importance of pill appearance to the therapeutic
affecting this field need to be sharply reconsid- equivalence of generic drugs. Such a statement
ered. A first step toward reform would be to in- could help clear the way for generic-drug man-
clude FDA certification of pharmaceutical size, ufacturers to adjust their products to conform
shape, and color in the drug-approval process. with the original appearance of the innovator
For example, a pill’s attributes could be pro- products.
posed by the manufacturer during the original
New Drug Application. Currently, such a process Conclusions
occurs for the brand name of the medication48;
extending it to pill appearance should not re- The legal protection of pharmaceutical-pill attri-
quire additional legislation. This would create a butes emerged in part to prevent the sale of
clear path for generic manufacturers to declare counterfeit goods and to prevent the practice of
during the ANDA process that their products palming off. But the rise of the modern generic-

n engl j med 365;1 nejm.org july 7, 2011 87

The New England Journal of Medicine
Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

drug industry, the products of which are subject 17. Greenslit N. Pharmaceutical branding: identity, individuality,
and illness. Mol Interv 2002;2:342-5.
to FDA inspection and bioequivalence standards, 18. Fritch DM. Should “the purple pill” by any other drug com-
has largely obviated these concerns. With few pany still be as purple? The changing face of trade dress protec-
exceptions, consumers should be able to expect tion for pharmaceutical manufacturers. Intellectual Prop Law Rev
2006;51:171-201.
that a generic drug will treat their condition as 19. Matheson JA, Justus MR. Eye candy. Med Mark Media 2009;
well as the brand-name version. The benefit of 44:55-8.
having similar brand-name and generic-drug 20. Two Pesos v. Taco Cabana, 505 U.S. 763 (1992).
21. Qualitex Co. v. Jacobson Prods, 514 U.S. 159 (1995).
products that range widely in appearance now 22. Carpenter DP, Tobbell DA. Bioequivalence: the regulatory
seems negligible. Instituting a more consistent history of a scientific concept. Bull Hist Med (in press).
and organized system of pill appearance would 23. Frank RG. The ongoing regulation of generic drugs. N Engl
J Med 2007;357:1993-6.
increase patient adherence, reduce the complex- 24. Kirsche ML. Generic substitution spells big money for retail
ity of medical regimens, reduce medication error, pharmacy. Drug Store News. August 22, 2005.
and encourage the rational use of bioequivalent 25. Shire v. Barr, 329 F.3d 348 (3rd Cir., 2003).
26. Wal-Mart Stores v. Samara Bros., 529 U.S. 205 (2000).
generic drugs. 27. Davit BM, Nwakama PE, Buehler GJ, et al. Comparing ge-
Dr. Greene reports receiving a Junior Faculty Fellowship from neric and innovator drugs: a review of 12 years of bioequivalence
the Ewing Marion Kauffman Foundation for Entrepreneurship data from the United States Food and Drug Administration. Ann
Research. Dr. Kesselheim reports receiving a career development Pharmacother 2009;43:1583-97.
award (K08HS18465-01) from the Agency for Healthcare Research 28. Kesselheim AS, Misono AS, Lee JL, et al. Clinical equivalence
and Quality and a Robert Wood Johnson Foundation Investigator of generic and brand-name drugs used in cardiovascular disease:
Award in Health Policy Research. a systematic review and meta-analysis. JAMA 2009;300:2514-26.
Disclosure forms provided by the authors are available with 29. Kohn LT, Corrigan J, Donaldson MS, eds. To err is human:
the full text of this article at NEJM.org. building a safer health system. Washington, DC: National Acad-
We thank Jerry Avorn, Alfred Engelberg, Kevin Outterson, emies Press, 2000.
and Stacey Dogan for their comments on earlier drafts of the 30. Glaxo whistle-blower lawsuit: bad medicine. CBS 60 Minutes.
manuscript. December 29, 2010. (http://www.cbsnews.com/stories/2010/12/29/
60minutes/main7195247_page2.shtml.)
From the Division of Pharmacoepidemiology and Pharmaco- 31. Kenagy JW, Stein GC. Naming, labeling, and packaging of
economics, Department of Medicine, Brigham and Women’s pharmaceuticals. Am J Health Syst Pharm 2001;58:2033-41.
Hospital and Harvard Medical School, Boston. Address reprint 32. Schiff GD. Standardized pill imprint codes: a pharma fantasy.
requests to Dr. Greene at the Division of Pharmacoepidemiology J Med Syst 2004;28:1-7.
and Pharmacoeconomics, Department of Medicine, Brigham and 33. Ngoh LN. Health literacy: a barrier to pharmacist-patient
Women’s Hospital, Harvard Medical School, 1620 Tremont St., communication and medication adherence. J Am Pharm Assoc
Suite 3030, Boston, MA 02120, or at greene@fas.harvard.edu. 2009;49(5):e132-e146.
34. Levinthal BR, Morrow DG, Tu W, Wu J, Murray MD. Cogni-
1. Generic Pharmaceutical Association. Facts at a glance. tion and health literacy in patients with hypertension. J Gen In-
(http://www.gphaonline.org/about-gpha/about-generics/facts.) tern Med 2008;23:1172-6.
2. Cutler C, Kesselheim A, Gabardi S, et al. Generic immuno- 35. Cárdenas-Valladolid J, Martín-Madrazo C, Salinero-Fort MA,
suppressants in hematopoietic cell transplantation. Biol Blood Carrillo de-Santa Pau E, Abánades-Herranz JC, de Burgos-Lunar C.
Marrow Transplant 2011;17:285-90. Prevalence of adherence to treatment in homebound elderly peo-
3. Stinson N. Trademarks and “lookalike” drugs. Indiana Law ple in primary health care: a descriptive, cross-sectional, multi-
Rev 1982;15:733-64. centre study. Drugs Aging 2010;27:641-51.
4. Newman S. Kill the “mere color” rule: equal protection for 36. Wolf MS, Curtis LM, Waite K, et al. Helping patients sim-
color under the Lanham Act. Univ Chic Law Rev 1994;61:1595- plify and safely use complex prescription regimens. Arch Intern
626. Med 2011;171:300-5.
5. Lanham Act §43(a), 15 U.S.C. §1125 (2010). 37. Hellier E, Tucker M, Kenny N, Rowntree A, Edworthy J. Merits
6. TrafFix Devices v. Marketing Displays, 121 S. Ct. 1255 of using color and shape differentiation to improve the speed
(2001). and accuracy of drug strength identification on over-the-counter
7. Brunswick Corp. v. British Seagull, 35 F. 3d 1527 (Fed. Cir. medicines by laypeople. J Patient Saf 2010;6:158-64.
1994). 38. Adherence to long-term therapies: evidence for action. Gene-
8. Warner v. Eli Lilly, 265 U.S. 526 (1924). va: World Health Organization, 2003.
9. Norwich Pharmaceutical Company v. Sterling Drug, 271 F.2d 39. Greene JA. Prescribing by numbers: drugs and the definition
569 (2nd Cir., 1959). of disease. Baltimore: Johns Hopkins University Press, 2007.
10. Krieg M. Black market medicine. Englewood Cliffs, NJ: 40. Donohue JM, Cevasco M, Rosenthal MB. A decade of direct-
Prentice-Hall, 1967. to-consumer advertising of prescription drugs. N Engl J Med
11. Facchinetti NJ, Dickson WM. Access to generic drugs in the 2007;357:673-81.
1950s: the politics of a social problem. Am J Public Health 1982; 41. Huskisson E. Simple analgesics for arthritis. Br Med J 1974;
72:468-75. 4:196-200.
12. SK&F Co. v. Premo, 625 F.2d 1055 (3rd Cir., 1980). 42. Lucchelli PE, Cattaneo AD, Zattoni J. Effect of capsule colour
13. A.H. Robbins Co. v. Medicine Chest Corp., 1980 U.S. Dist. and order of administration of hypnotic treatments. Eur J Clin
LEXIS 14412 (E.D. Mo., 1980). Pharmacol 1978;13:153-5.
14. Pennwalt v. Zenith Laboratories, 472 F.Supp. 413 (E.D. 43. Harrington A. The placebo effect. Cambridge, MA: Harvard
Mich.1979). University Press, 1997.
15. Ives v. Darby, 455 F.Supp. 939 (E.D.N.Y., 1978). 44. Jacobs KW, Nordan FM. Classification of placebo drugs: effect
16. Inwood v. Ives, 456 U.S. 844 (1982). of color. Percept Mot Skills 1979;49:367-72.

88 n engl j med 365;1 nejm.org july 7, 2011

The New England Journal of Medicine

Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
 health law, ethics, and human rights

45. de Craen AJ, Roos PJ, Leonard de Vries A, Kleijnen J. Effect 49. Shrank WH, Choudhry NK, Agnew-Blais J, et al. State ge-
of colour of drugs: systematic review of perceived effect of drugs neric substitution laws can lower drug outlays under Medicaid.
and of their effectiveness. BMJ 1996;313:1624-6. Health Aff (Millwood) 2010;29:1383-90.
46. Khan A, Bomminayuni EP, Bhat A, Faucett J, Brown WA. Are 50. Horn CR, Cochrane GM. Colour coding for bronchodilator
the colors and shapes of current psychotropics designed to max- inhalers. Lancet 1986;1:165.
imize the placebo response? Psychopharmacology (Berl) 2010; 51. Gray S. Asthma inhaler devices. In: NHS evidence — respira-
211:113-22. tory. (http://www.library.nhs.uk/respiratory/viewresource.aspx?
47. Waber RL, Shiv B, Carmon Z, Ariely D. Commercial features resid=125010.)
of placebo and therapeutic efficacy. JAMA 2008;299:1016-7. 52. Berman A. Reducing medication errors through naming,
48. Lambert BL, Lin SJ, Tan H. Designing safe drug names. labeling, and packaging. J Med Syst 2004;28:9-29.
Drug Saf 2005;28:495-512. Copyright © 2011 Massachusetts Medical Society.

n engl j med 365;1 nejm.org july 7, 2011 89

The New England Journal of Medicine
Downloaded from nejm.org on January 14, 2016. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.