Not for Distribution

55 The Journal of Law, Medicine & Ethics Not for Distribution Challenges in the Federal Regulation
of Pain Management Technologies Lars Noah T Journal of Law, Medicine & Ethics, 31 (2003): 55–74.
© 2003 by the American Society of Law, Medicine & Ethics. hose who write about pain management have fo-
cused almost entirely on delivery issues, paying
essentially no attention to the federal regulatory challenges that affect the development of pain relief tech-
nologies — namely, pharmaceuticals and medical devices
indicated for analgesic uses. The academic literature is
strangely devoid of any sophisticated discussion of the diffi-
culties that attend, first, the product approval decisions of
the Food and Drug Administration (FDA) and, second, the
scheduling decisions made by the Drug Enforcement Ad-
ministration (DEA). If a “bottleneck” develops upstream, it
could have serious repercussions downstream — without
pain relief technologies, the issues of access that have preoc-
cupied previous commentators would have little practical
consequence. The modern pharmaceutical industry traces its origins back more than a century, around the time that the German
company Bayer first synthesized aspirin (acetylsalicylic acid)
and began marketing it as an analgesic. 1 Federal regulation of drug products in the United States began shortly thereafter, 2 and it has evolved alongside the growing sophistication of
the pharmaceutical industry. Although not specifically geared
toward the control of pain management technologies, these
various laws have had important consequences for the avail-
ability and use of analgesic products, at least in part because
of certain peculiar aspects of these pharmaceuticals. In par-
allel, Congress has imposed special requirements on narcotics
often used for analgesia because of concerns about addiction
and abuse. This article considers, in turn, the roles played by the FDA and the DEA in regulating pain management technolo-
gies. Whether a company wishes to market an over-the-counter pain reliever containing a well-known active ingredient, a
prescription drug containing a novel analgesic compound,
or a medical device intended for the treatment of pain, it
must satisfy a number of requirements designed to ensure
the safety and effectiveness of the product. In addition, if a
drug product contains a narcotic or other controlled sub-
stance, its availability will depend on the manner in which
the DEA has classified that substance. Although implemented
by two quite dissimilar agencies — the former preoccupied
with medical and scientific questions, while the latter fo-
cuses on law enforcement matters — these two regulatory
regimes operate in tandem and overlap in potentially impor-
tant ways. Several themes emerge from this discussion. First, both agencies have shown a marked resistance to making narcot-
ics available as analgesic products, though the FDA has better
appreciated the value of providing patients with a wide range
of options for treating pain. Second, notwithstanding its far
more limited understanding of clinical practice, the DEA
has demonstrated a greater enthusiasm for efforts to keep
physicians in line than has the FDA. These and similar ex-
amples of contrasting behavior by the two federal agencies in
this field — reflecting their distinctly different missions and
cultures — provide important lessons about comparative in-
stitutional competence and suggest that Congress might want
to reconsider this sometimes unstable division of regulatory
authority. Finally, the tools currently available to both agencies may be too blunt for sensibly addressing the conflict that
often arises between the needs of patients and efforts to pre-
vent irresponsible use. Once it decides to approve a new
drug, the FDA can restrict access to prescription-only and
then hope to influence physician use through recommenda-
tions in labeling. For the DEA, most of the important
restrictions on access to controlled substances represent an 56 Volume 31:1, Spring 2003 Not for Distribution outgrowth of initial decisions about appropriate scheduling,
which may cause the agency to overschedule substances in
order to serve law enforcement purposes. Providing more
refined regulatory options may allow for a more sensible
resolution of the perennial tension between patient access
and drug diversion. If federal agencies become excessively concerned about misuse, they may deprive patients of valuable new analgesic
agents. Indeed, depending on the stringency of the restric-
tions imposed, companies may be unable or unwilling to
develop such products in the first place, or health care pro-
fessionals may fail to make full use of them. 3 Already hesitant to approve powerful analgesics, the FDA and the DEA may
be forced to revisit their original clearance decisions due to
the growing problems with the theft and diversion of cur-
rently marketed painkillers containing ingredients such as
oxycodone and fentanyl. Any such regulatory actions —
whether expressed as a refusal to allow an analgesic product
to enter the market initially or the withdrawal of such a
product in the face of rampant abuse — would have to grapple
with the classic difficulty of choosing between the medical
needs of individual patients and the broader societal hazards
associated with the availability of such products. These ques-
tions do not admit of any simple answers, of course, but they
cannot be avoided. This article delineates the nature of the
various challenges that the federal government has faced in
regulating pain management technologies before tentatively
recommending a public health approach that attempts to
bridge the FDA’s predominantly clinical focus and the DEA’s
preoccupation with the potential adverse consequences for
third parties. FDA R EGULATION OF A NALGESIC P RODUCTS Before pain management technologies can reach patients,
the Food and Drug Administration must assess their safety
and effectiveness. If a product has not received marketing
approval (or an exemption) from the agency, then it cannot
be sold. Even if a company has surmounted the often diffi-
cult hurdle of proving that a product serves a therapeutic
purpose without posing an undue risk, the FDA’s decisions
about appropriate labeling may affect how readily patients
will be able to access it. Product approval requirements Over the course of the last century, federal regulation of
medical technologies has shifted from an emphasis on polic-
ing against economic frauds to a premarket approval system
mandating proof to support therapeutic claims. 4 The FDA has, for example, expressed long-standing and largely justi-
fied skepticism about “quack” medical devices, including
products indicated for pain relief. 5 As a result, legitimate articles used for analgesia may encounter significant regula- tory obstacles originally fashioned to protect consumers from
wasting their resources on worthless remedies. In particular, the FDA’s insistence on placebo-controlled clinical trials when evaluating the effectiveness of pharma-
ceuticals and medical devices means that firms seeking to
market pain management technologies shoulder a particu-
larly challenging evidentiary burden given the pronounced
placebo effect that researchers encounter in this context. 6 Moreover, because of difficulties in measuring a largely sub-
jective condition such as pain, 7 coupled with the significant variability in patient response, 8 the agency may struggle to interpret placebo-controlled clinical trials submitted as part
of an application for new drug approval (NDA). 9 Even if persuaded that a drug works, the agency will have to decide whether or not the inevitable side-effects pose
too great a hazard to justify granting product approval. In
making these risk-benefit judgments, “the FDA takes into
account the significance of a targeted health condition, or
the status of that condition as a treatable disease.” 10 Interven- tions such as analgesics that provide only symptomatic relief
may not fare as well in this process, 11 though some experts maintain that pain should qualify as a serious disease process
in its own right. 12 A similar dichotomy, between curative and palliative care, 13 may account for the persistent undertreatment of pain by health care professionals. 14 Notwithstanding this pair of potential obstacles — namely, the heightened difficulty of establishing efficacy and the pre-
sumption that symptomatic relief represents a less compelling
clinical endpoint for purposes of making risk-benefit judg-
ments — no one has accused the FDA of overcaution in
reviewing new analgesics. On the contrary, some observers
have criticized the agency for approving too many new non-
steroidal antiinflammatory drugs (NSAIDs) that offer no
particular advantage over existing, and typically less expen-
sive, drugs in the class. 15 The FDA usually does not, however, make judgments about comparative efficacy, 16 preferring to leave that task for physicians and patients based on the infor-
mation supplied in the labeling dictated by agency reviewers. In some instances, the management of side-effects asso- ciated with analgesics rather than any differential effectiveness
accounts for the need to have a range of therapeutic options.
In 1999, the FDA approved Vioxx ® (rofecoxib) and Celebrex ® (celecoxib), the first in a new class of painkillers called COX-
2 inhibitors. 17 NSAIDs represent peripherally acting analgesics, which means that they lessen the localized in-
flammation (and production of prostaglandins) that triggers
a sensation of pain. 18 They do this by blocking the enzyme cyclooxygenase. As it turns out, this enzyme has two isoforms,
one associated with inflammation (COX-2) and the other
one thought to protect the lining of the stomach (COX-1).
Older NSAIDs blocked both isoforms, which may explain
their association with the development of ulcers. 19 By selec- tively blocking only the form of the enzyme linked to
inflammation, COX-2 inhibitors promised to offer compa- 57 The Journal of Law, Medicine & Ethics Not for Distribution rable pain relief, especially in arthritis patients, without the
associated risk of gastrointestinal side-effects. 20 These drugs appear, however, to pose a heightened risk of cardiovascular
side-effects. 21 As may happen with any new drug product, serious side- effects associated with analgesics may become evident only
after approval and widespread use. 22 During the 1980s, the FDA received numerous adverse event reports for three
NSAIDs that led to their hasty withdrawal from the market:
Zomax ® (zomepirac), Oraflex ® (benoxaprofen), and Suprol ® (suprofen). 23 Zomax was, however, notable for another rea- son. Even though the FDA conceded that it knew at the time
of approval that the drug represented a potential human car-
cinogen (and that some patients would use it chronically),
Zomax received an NDA because agency reviewers thought
that it could substitute for narcotics used in the treatment of
severe pain. 24 This excessive concern about patient use of any controlled substances — so much so that it would dis-
place the FDA’s normal resistance to approving nonessential
products that create a risk of cancer — appears repeatedly in
other contexts. When patients mysteriously began dying from
anaphylactic reactions, even the supposed advantage of Zomax
as a substitute for narcotic analgesics could not prevent the
drug’s commercial demise. 25 In July 1997, the prescription analgesic Duract ® (bromfenac sodium) entered the U.S. market. Less than a
year later, the manufacturer withdrew the drug from the
market after it had been associated with liver toxicity result-
ing in at least four deaths and eight liver transplants. 26 During clinical trials, researchers had reported an unexpectedly high
incidence of elevated liver enzymes in patients who took
Duract for relatively long periods, 27 and one FDA reviewer voiced significant concerns about the drug’s potential hepa-
totoxicity. 28 Nonetheless, as happened with Zomax, the agency viewed the product as a substitute for narcotic anal-
gesics, so it decided to approve Duract, though only for
short-term use and with labeling information about the risk
of elevated liver enzymes. 29 Many physicians, however, pre- scribed Duract for longer than the 10 days specified by the
FDA, and the agency soon began receiving reports of liver
failure. As the FDA initially became aware of liver problems in patients taking Duract, it tried to notify physicians about the
emerging safety problems. 30 The agency required the addi- tion of a prominent boxed warning in the drug’s labeling,
and the manufacturer mailed out a “Dear Doctor” letter
emphasizing the drug’s dangers and providing guidelines for
proper use. 31 These efforts did not, however, completely pre- vent the inappropriate prescribing of Duract for long-term
use. After the FDA concluded that it could not impose effec-
tive restrictions to limit the duration of use, 32 the manufacturer voluntarily withdrew the drug from the mar-
ket in June 1998. 33 Duract may well have offered a net benefit over other drugs in some class of patients when used as rec- ommended, but the agency could not ignore the pattern of
physician misuse causing risks to other patients. Over-the-counter marketing One fundamental labeling question is whether to make a
product available only upon a prescription from — or
through direct administration by — a licensed health care
professional. Because analgesics relieve symptoms and do
not purport to treat any underlying disease process, they would
seem to represent natural candidates for nonprescription or
over-the-counter (OTC) marketing. 34 Even if most consum- ers would not need a physician’s diagnostic skills in order to
decide whether to select a particular pain reliever, however,
the safety profile of a product may justify some restriction on
access. At least initially, most new ingredients are available
only by prescription while the FDA collects additional ad-
verse event data. 35 In addition, the risk of abuse has, from the outset, represented one of the primary rationales for limiting
drugs to prescription-only sale. 36 The FDA may decide to authorize OTC marketing for drugs that do not require the supervision of a physician, have
a history of safe use, and present no abuse potential. This
may happen in a couple of different ways. First, a company
may sell an OTC drug if it abides by the terms of the appli-
cable “monograph,” which specifies for particular categories
of products the active ingredients and dosages that the FDA
has determined to be safe and effective, along with the pre-
cise labeling necessary to facilitate appropriate consumer use. 37 The agency’s OTC drug review for internal analgesics began
with a call for data in 1972. 38 Five years later, the review panel, which had considered forty-nine active ingredients,
issued its recommendations. 39 More than one decade later, the FDA published a tentative final monograph (TFM) for
this OTC drug category. 40 In brief, this proposed rule includes aspirin and acetaminophen as permitted active ingredients and
allows labeling “for the temporary relief of minor aches and
pains” with directions against taking the product for more
than 10 days along with an assortment of warnings. 41 After more than 30 years, the OTC monograph for internal anal-
gesics remains at least 1 year from finalization. 42 The second route to OTC marketing requires that a company secure supplemental NDA for a reformulation (in-
cluding revised labeling) of a product previously approved
for prescription use. 43 Among its most prominent Rx to OTC switches, the FDA authorized nonprescription sale of a lower
dose product containing ibuprofen (e.g., Motrin ® ). 44 It later switched a number of other NSAIDs, including ketoprofen
and naproxen, from prescription to OTC status. 45 Of course, it did not take long for consumers to realize that they could
self-medicate with a prescription strength simply by exceed-
ing the dose recommended in the OTC labeling. 46 Nonprescription marketing does not mean that a drug product entails no serious risks, 47 as again revealed by the 58 Volume 31:1, Spring 2003 Not for Distribution FDA’s experience with analgesics. For instance, in the early
1980s, the agency became aware of a link between Reye
syndrome and the use of aspirin by children suffering from
viral infections. 48 The labels of OTC drug products contain- ing aspirin now must include the following statement:
“WARNING: Children and teenagers should not use this
medicine for chicken pox or flu symptoms before a doctor is
consulted about Reye syndrome, a rare but serious illness
reported to be associated with aspirin.” 49 Notably, the FDA rejected suggestions urging “more drastic measures such as
banning use of aspirin in products for individuals under 21
years of age or limiting such products to prescription use.” 50 More recently, after it received reports of an association be-
tween acetaminophen and liver toxicity, the agency imposed
special warning requirements. 51 DEA R ESTRICTIONS ON A NALGESIC P RODUCTS Although NSAIDs dominate both the prescription and OTC
markets in terms of volume, 52 most of the truly significant pain management technologies used by physicians qualify as
“controlled substances,” primarily opioid analgesics. 53 (Less frequently, health care professionals may try behavioral
therapy, surgical interventions such as nerve blocks, or
medical devices such as transcutaneous electrical nerve
stimulators. 54 ) Unlike peripherally acting drugs, opioids relieve pain by acting directly on the central nervous sys-
tem, binding with the receptors that are involved in the
transmission of pain signals to the brain. These drugs must
undergo the same FDA premarket review process as any
other pharmaceutical product, but special authority over
controlled substances resides with a separate agency, the
Drug Enforcement Administration of the U.S. Department
of Justice. 55 Classification of controlled substances In 1970, Congress enacted the Controlled Substances Act
(CSA), which establishes different “schedules” of narcotics
and other substances prone to abuse or diversion. The most
restrictive classification, Schedule I, is defined as follows: (A) The drug or other substance has a high potential for abuse. (B) The drug or other substance has no currently accepted medical use in treatment in the United States. (C) There is a lack of accepted safety for use of the drug or other substance under medical supervision. 56 The statute then provides the following definition for
Schedule II: (A) The drug or other substance has a high potential for abuse. (B) The drug or other substance has a currently accepted medical use in treatment in the United States or a
currently accepted medical use with severe
restrictions. (C) Abuse of the drug or other substance may lead to severe psychological or physical dependence. 57 Schedules III, IV, and V, which all have “a currently accepted
medical use in treatment in the United States,” contain drugs
with progressively lower potentials for abuse and severity of
dependence. 58 Most states have adopted parallel legislation modeled on the federal statute. 59 By way of illustration, 60 Schedule I includes substances such as heroin (diacetylmorphine) and marijuana (cannabis).
Schedule II includes, for instance, drugs containing synthetic
forms of morphine sold as Dilaudid ® (hydromorphone hy- drochloride) and Demerol ® (meperidine hydrochloride). Schedule III includes products such as Tylenol ® (acetami- nophen) with codeine. Schedule IV includes products such
as Darvon ® (propoxyphene hydrochloride). Schedule V in- cludes products such as Robitussin ® , a cough syrup which contains a limited amount of codeine. Congress understood that “[m]any of the drugs included within this subchapter have a useful and legitimate medical
purpose and are necessary to maintain the health and general
welfare of the American people.” 61 This explains the central role, for purposes of distinguishing Schedule I from all other
controlled substances, of the criterion that asks whether the
drug has “a currently accepted medical use in treatment in
the United States.” 62 Congress did not provide further guid- ance on this score, 63 but it insisted that the Attorney General request and abide by recommendations from the Secretary
of Health and Human Services (HHS) when revising the
schedule. 64 In amending the CSA in 1978 to implement the United Nations Convention on Psychotropic Substances,
Congress announced its intent that the legislation not “inter-
fere with ethical medical practice in this country as
determined by the Secretary of [HHS] on the basis of a con-
sensus of the views of the American medical and scientific
community,” 65 which suggests a somewhat more restrictive standard than “a currently accepted medical use.” Under the statute, the DEA supervises the manufactur- ing and distribution of legal narcotics. 66 For Schedule II drugs, manufacturers must register their operations, and the DEA
assigns aggregate and individual production quotas. 67 Sched- ule II drugs must be produced in a secure facility, transported
with care, stored in a vault, tracked using a precise inventory
system, supplied in response to an order form completed by
a registered practitioner, and dispensed only upon a written
prescription and without refills. 68 Progressively weaker re- strictions apply to Schedule III, IV, and V drugs. Physicians
wishing to prescribe controlled substances first must register
with the DEA, which enjoys sweeping authority to suspend
or revoke certificates of registration. 69 The agency has, for 59 The Journal of Law, Medicine & Ethics Not for Distribution instance, threatened to use this authority to discipline physi-
cians who have done nothing more than recommend that their
patients consider taking a Schedule I substance as permitted un-
der limited circumstances in a growing number of states. 70 Legislative scheduling decisions Congress typically makes the initial decision about how to
schedule a controlled substance. In United States v. Oakland
Cannabis Buyers’ Co-operative, 71 the U.S. Supreme Court showed tremendous deference to the legislature’s judgment
about the appropriate classification of marijuana. The dis-
pute arose after California voters passed Proposition 215 (the
Compassionate Use Act of 1996), which created a limited
safe harbor from prosecution under the state’s controlled
substances act for the medical use of marijuana. 72 The state law could not, of course, protect critically ill patients or
their health care providers from the risk of federal prosecu-
tion for the medical use of marijuana. As part of its
multipronged response to the California initiative, the De-
partment of Justice sought to enjoin buyers’ clubs from
dispensing the drug to patients with severe pain and other
debilitating symptoms. The U.S. Supreme Court rejected these
groups’ argument that a medical necessity defense might be
implied under federal law. It held that persons manufactur-
ing or distributing Schedule I controlled substances for medical
uses could not avoid federal sanctions, 73 though three mem- bers of the Court wrote separately to emphasize that it had
not decided whether a seriously ill patient could raise a medi-
cal necessity defense if prosecuted for using marijuana. 74 The Supreme Court’s decision turned entirely on the notion that, in placing marijuana within Schedule I, Con-
gress necessarily had concluded that the drug lacked any
currently accepted medical use: “It is clear from the text of
the Act that Congress has made a determination that mari-
juana has no medical benefits worthy of an exception.” 75 The Court rejected the argument that, although placed within
Schedule I because its medical use had not yet achieved “gen-
eral acceptance,” a controlled substance may offer therapeutic
benefits for some narrow class of patients. It also rejected the
suggestion that Congress may have placed a controlled sub-
stance into Schedule I without strictly abiding by the criteria
(such as the lack of any currently accepted medical use) that
it had established to govern scheduling decisions made by
the DEA. 76 In fact, when it enacted the CSA, Congress had expressed a good deal of ambivalence about whether marijuana be-
longed in Schedule I. 77 After all, the United States Pharmacopeia (U.S.P.), which Congress has cross-referenced
in other statutes as a source for information about therapeu-
tic products, 78 had listed marijuana as a drug for almost a century (until 1941), and prominent physicians had endorsed
its use early in the twentieth century for treating maladies
such as migraine headaches. 79 Although neither one of these facts would establish that the drug offers genuine therapeutic
benefits, subsequent research has suggested that marijuana
may have analgesic and other clinically useful properties. 80 Within the last few years, both the National Institutes of
Health and the Institute of Medicine have recommended
further scientific work on this question. 81 Although more than half a dozen states, 82 along with Canadian health officials, 83 have concluded that marijuana may have legitimate medical uses, Congress has decided to
abide by its earlier, contrary conclusion: A resolution passed
in 1998 “oppose[d] efforts to circumvent this [federal sched-
uling] process by legalizing marijuana, and other Schedule I
drugs, for medicinal use without valid scientific evidence
and the approval of the Food and Drug Administration.” 84 This leaves an opening, of course, for revisiting the original
scheduling judgment administratively if new research emerges
to persuade regulatory officials that marijuana has medical
utility, an issue taken up more fully below. A similar controversy arose in the early 1980s in con- nection with the ultimately unsuccessful efforts by some in
Congress to reschedule heroin. Unlike marijuana, which may
have multiple therapeutic applications and an arguably exag-
gerated abuse potential, no one seriously doubts that heroin
causes addiction, but, in common with the Schedule II drugs
cocaine and morphine, it also offers a powerful analgesic
effect. In fact, some have argued that it has unique properties
as a pain reliever for terminally ill patients or, at the very
least, offers an alternative for those who do not respond well
to approved opioids. 85 For this reason, the United Kingdom continues to recognize the medical usefulness of heroin; 86 however, the CSA requires that a controlled substance have a
currently accepted use in the United States in order to avoid
classification in Schedule I. 87 In the end, fears of diversion and confidence in the effectiveness of already available opioid
analgesics — coupled with the understandable political im-
perative against appearing to be soft on drugs — scuttled the
effort to reschedule heroin. 88 In another instance, Congress decided to reclassify a drug as Schedule I even though it clearly enjoyed a currently ac-
cepted medical use. The FDA previously had approved
methaqualone for treating insomnia, which concededly gave
the drug a currently accepted use in treatment, but Congress
concluded that methaqualone offered no advantages over other
products that posed less of a risk of abuse. 89 For that reason, Congress directed the DEA to reschedule the drug and the
FDA to withdraw its NDA. 90 Although this did not alter the statutory criteria generally applicable to scheduling decisions,
the legislative rescheduling of methaqualone arguably set a
troublesome precedent. 91 In effect, Congress adopted a “one size fits all” approach, which fails to account for the possi-
bility that this drug might provide some unique benefit to a
small group of patients who are refractory to the drug of
choice, whether because of their physiologic or genetic de-
viation from the norm, progression of disease, heightened 60 Volume 31:1, Spring 2003 Not for Distribution susceptibility to side-effects, co-morbid factors, or concomi-
tant use of other medications. If aggregate risk-benefit
balancing of this sort becomes the standard for future sched-
uling decisions, then the needs of individual patients will
compete against the consequences of the irresponsible be-
havior of abusers, and the DEA may opt to sacrifice products
that offer insufficiently dramatic advantages over existing al-
ternative treatments. Administrative scheduling decisions In the case of newly synthesized chemicals, or in response to
new information about previously scheduled controlled sub-
stances, the DEA may have to make its own scheduling
decisions. As mentioned previously, Congress mandated that
the agency first consult with HHS, 92 which has subdelegated that task to the FDA, and then abide by any recommenda-
tions that the DEA receives from the Department. 93 In this way, Congress hoped to “strike[] a balance between the ex-
tent to which control decisions should be based upon law
enforcement criteria, and the extent to which such decisions
should be based on medical and scientific determinations.” 94 Given the long-running “war on drugs” in this country, it is
difficult to maintain this sort of balance. 95 Such cooperative arrangements between agencies or “split enforcement” models have posed challenges in other con-
texts. 96 In some instances, Congress has decided against consolidating authority in a single administrative agency to
counteract the tendency toward tunnel vision in regulatory
decisions, or it has established a separate watchdog group
for an agency, 97 as it did when assigning the responsibility for accident investigations to the National Transportation Safety
Board (NTSB), which often criticizes the Federal Aviation
Administration (FAA) for taking inadequate steps to improve
safety. 98 In other instances, the division of authority over a field between multiple agencies does not reflect any pur-
poseful design but instead an accident of history that
subsequently leads to calls for consolidation. 99 Apart from questions of efficiency, these organizational choices can have
significant impacts on substance, especially if two agencies
have different sorts of expertise and missions, and respond
to different constituencies. 100 Without meaning to exaggerate the cultural explanations for their contrasting approaches to
pain management technologies, the FDA probably would
have implemented the Controlled Substances Act differently
than the DEA has done. Immediately after passage of the Act, the National Or- ganization for the Reform of Marijuana Laws (NORML),
together with a couple of allied organizations, attempted to
persuade the DEA and its predecessor agency to reschedule
marijuana. Starting in 1972, the agency repeatedly denied
these rescheduling petitions, though on four separate occa-
sions the reviewing court remanded the dispute to the DEA
for further consideration. 101 In contrast, constitutional chal- lenges to the DEA’s refusal to down-schedule marijuana have
not fared as well in the courts. 102 Perhaps the refusal to down-schedule marijuana arises from a concern that, unlike chemicals synthesized by phar-
maceutical companies, the FDA could not effectively exercise
its regulatory authority to demand proof of safety and effi-
cacy over a raw product with variable composition that
individuals can grow in their homes. 103 After the FDA ap- proved an NDA for Marinol ® (dronabinol), an antiemetic drug containing synthetic tetrahydrocannabinol (THC), the
principal psychoactive component in marijuana, the DEA
placed this controlled substance into Schedule II. 104 In 1999, the DEA further down-scheduled that drug as the FDA began
to approve additional indications for its use. 105 In 1992, 20 years after NORML first petitioned the agency, the DEA Administrator opined that “no responsible
physician could conclude that marijuana is safe and effective
for medical use,” 106 echoing his predecessor’s earlier conclu- sion that it was “not recognized as medicine in generally
accepted pharmacopeia, medical references, journals or text-
books.” 107 This time around the federal appellate court upheld the agency’s decision. 108 Late in 1997, after receiving yet another petition for the rescheduling of marijuana, the DEA
once again referred the matter to HHS for a recommenda-
tion. 109 In 2001, the DEA appeared to soften its stance somewhat when it authorized use by researchers at the Uni-
versity of California to conduct clinical trials investigating
marijuana’s analgesic properties in multiple sclerosis and
AIDS patients with peripheral neuropathy. 110 The protracted effort to reschedule marijuana forced the DEA to elaborate on the meaning of the critical statutory
phrase “currently accepted medical use.” 111 The agency de- cided to demand that there be adequate safety information
coupled with adequate and well-controlled studies establish-
ing efficacy, which are widely available and accepted by
qualified experts. 112 These criteria closely track the FDA’s test for whether to exempt a product from new drug ap-
proval requirements. Indeed, after noting Congress’s failure
to define “currently accepted medical use,” the DEA looked
to the FDA’s enabling statute for guidance. 113 Under that stat- ute, a manufacturer need not secure an NDA for a drug that
is “generally recognized as safe and effective” (GRASE), which
the FDA has defined as requiring essentially the same proof
of safety and efficacy that it demands for new drugs. 114 In effect, a controlled substance could only avoid inclusion in
Schedule I if the FDA already had approved it or exempted it
from new drug approval requirements, a standard that seems
overly stringent and inconsistent with the statutory design. 115 Congress could have explicitly linked scheduling decisions
to a drug’s FDA regulatory status, but it did not do so, choos-
ing instead the arguably more flexible standard of “currently
accepted medical use.” This distribution of regulatory authority, between a tra- ditional law enforcement agency and one that focuses on 61 The Journal of Law, Medicine & Ethics Not for Distribution patient health, can generate incongruities. In some instances,
the DEA’s desire to facilitate prosecution of drug abusers by
placing a substance into Schedule I or II conflicts with the
FDA’s effort to promote the development of a drug poten-
tially valuable in the treatment of a legitimate class of users. 116 In other instances, at least where it has not already approved
a drug proposed for inclusion in Schedule I, the FDA has
done little more than “rubber stamp” DEA scheduling rec-
ommendations. 117 Once placed in Schedule I, of course, it becomes exceedingly difficult to conduct the sort of research
necessary to secure FDA approval and subsequent down-
scheduling by the DEA. Off-label prescribing The DEA has overlaid another “currently accepted medical
use” requirement in regulating prescriptions for narcotics.
Physicians may prescribe controlled substances only for “a
legitimate medical purpose.” 118 In turn, pharmacists may dispense controlled substances only pursuant to a valid pre-
scription, which might require a comparison between the
indications appearing in the FDA-approved labeling and the
patient’s condition. 119 In contrast, the FDA has long recognized the legitimacy of “off-label” drug prescribing, an outgrowth of Congress’s
admonition against federal interference with the practice of
medicine. 120 As the agency has explained, “[o]nce a drug product has been approved for marketing, a physician may,
in treating patients, prescribe the drug for uses not included
in the drug’s approved labeling.” 121 Apart from deferring to the congressional decision against undue interference with
the practice of medicine, this policy acknowledges the inevi-
tability of incomplete information, the lag time before widely
accepted new uses appear in revised labeling (if they ever
do), and patient variability. Thus, physicians routinely, and
often appropriately, deviate from the directions contained in
approved prescription drug labeling. 122 In the case of controlled substances, however, physi- cians may not have the freedom to engage in such off-label
prescribing. 123 In contrast to the FDA, which focuses its at- tention on the activities of commercial entities, the DEA
enjoys the power to supervise the activity of individual physi-
cians by virtue of its registration requirements. The agency also
has shown less deference to medical practitioners and the regu-
latory prerogative of the states, instead seeming to regard them
with some suspicion. For example, when it down-scheduled
THC, the DEA formally announced a policy threatening to
revoke (as inconsistent with the public interest) the registra-
tion of anyone “who engages in the distribution or dispensing
of dronabinol for medical indications outside the [FDA] ap-
proved [antiemetic] use associated with cancer treatment.” 124 For a variety of reasons, patients respond variably to opioids, 125 which explains the need for a range of alterna- tives and the interest in using powerful analgesics in different combinations. For instance, a certain genetic polymorphism
found in more than 5 percent of patients makes them poor
metabolizers of codeine. 126 In addition, patients may develop tolerance from chronic treatment. Would a physician or phar-
macist face DEA sanctions for prescribing or dispensing a
Schedule II drug approved only for nonanalgesic indications
to a patient with severe pain refractory to the other available
drugs? In recent testimony before Congress, the DEA Ad-
ministrator seemed to imply otherwise, 127 but the agency’s position remains unclear. Contemporaneously with the Administrator’s congres- sional appearance, and in an effort to undercut Oregon’s
Death with Dignity Act, the Attorney General threatened to
sanction physicians who assist in the suicide of terminally ill
patients using Schedule II drugs approved by the FDA for
other purposes (primarily sedatives such as secobarbital). 128 In effect, the DEA decided that, notwithstanding state law to
the contrary, physician-assisted suicide does not qualify as
legitimately within the scope of medical practice. 129 A fed- eral court subsequently invalidated the Attorney General’s
order, holding that Congress had never intended to grant the
DEA such sweeping power to define the contours of legiti-
mate medical practice. 130 If affirmed on appeal, this decision may have broader consequences for the DEA’s authority to
limit off-label prescribing and dispensing of other controlled
substances for analgesic uses. Formulation issues and distribution controls Dosage forms For the most part, scheduling decisions relate to the intrinsic
characteristics of active ingredients, paying little attention to
product formulation and dosage, 131 though the DEA’s differ- ential treatment of smoked marijuana and THC encapsulated
for ingestion may represent an exception. In contrast, the
FDA routinely addresses precisely these sorts of issues when
it reviews an NDA application, 132 and the choices that it makes may have important repercussions for the threat of
abuse and diversion. Fentanyl citrate is a Schedule II controlled substance. During the last decade, the FDA approved products contain-
ing this opioid analgesic in unusual dosage forms. First, it
authorized the marketing of a transdermal fentanyl patch
(Duragesic ® ). 133 Within a couple of years, the misuse of these patches — for instance, a few individuals died after sucking
on them — led the agency to demand stronger warnings to
physicians and patients. 134 More controversially, in 1994, the FDA approved a transmucosal form of fentanyl — a lollipop
intended for use by children (Oralet ® ) — notwithstanding objections that it had received from the DEA and others that
this would promote abuse. 135 In 1997, another manufacturer received approval to market a lollipop form of fentanyl
(Actiq ® ), though this time intended only for use by cancer 62 Volume 31:1, Spring 2003 Not for Distribution patients experiencing breakthrough pain and with special
packaging designed to minimize the risk of accidental poi-
soning by children. 136 Recent years have seen widespread misuse of other opioid analgesics, especially OxyContin ® (oxycodone hydrochlo- ride). 137 Introduced in 1996, shortly after securing FDA approval, OxyContin quickly became the most widely pre-
scribed narcotic painkiller, recording more than $1 billion
in sales last year. The drug’s active ingredient is a synthetic
form of morphine regulated as a Schedule II controlled sub-
stance. Older painkillers such as Percocet ® and Percodan ® also contain oxycodone, but OxyContin uses a time-released
formulation designed to offer sustained relief over a 12-hour
period to patients with chronic moderate-to-severe pain. In
contrast, the older drug products in this class (including the
related hydrocodone drugs such as Vicodin ® and Lortab ® ) may offer uneven relief over just a 3–4 hour period. It is difficult trying to quantify the benefits of this drug. Anecdotal reports from physicians testify to the effectiveness
of OxyContin in particular patients, but these give no sense
for the drug’s aggregate utility. One could use the number of
prescriptions written each year — now in excess of 6 million
— as a rough proxy. Even if some number of physicians
prescribed the drug to patients who did not actually suffer
from severe pain or to those for whom the older opioids had
offered satisfactory relief, the high volume of prescribing
suggests that OxyContin has helped to fill a significant unmet
need and that many “thousands” of patients have benefited
from its availability. 138 If nothing else, the slow-release fea- ture made OxyContin more convenient than older opioids,
which patients with severe pain would have to take every 4
hours throughout the day and night. The time-released formulation also seemed to make OxyContin less prone to abuse because it would not provide
a quick euphoric effect upon initial ingestion. As a result,
Purdue Pharma and Abbott Labs promoted their drug to a
broader group of physicians and as presenting a lower risk of
abuse and diversion. 139 The companies apparently failed to appreciate the creativity of drug abusers. To defeat the slow-
release feature, these individuals chewed, crushed, dissolved,
or scraped the coating off of the tablets, leaving stronger
dosages of oxycodone than found in individual Percocet or
Percodan tablets. They would then ingest, snort, or inject the
substance. Reports indicate that hundreds of people have
died after overdosing in this fashion, 140 usually as a result of acute pulmonary edema. Diversion occurs in several ways. Individuals might feign pain and shop for doctors willing to prescribe the drug, or
they might engage in prescription fraud and theft. These in-
dividuals then might use the drugs themselves or sell their
supplies to others. A few desperate addicts have committed
armed robberies at pharmacies, demanding OxyContin rather
than cash. 141 Most of the deaths and other injuries linked to the drug have occurred in persons other than legitimate patients. Although deaths resulting from OxyContin have received significant publicity, they should be put in context: The di-
version of other prescription controlled substances over
the years has resulted in numerous deaths among drug
abusers, and the toll pales in comparison to the injuries
associated with the lawful use of nonnarcotic drugs. NSAIDs
may represent a far more serious public health menace,
contributing to thousands of patient deaths each year. 142 The volume of use is also higher, but these comparative
statistics raise an interesting policy question: Should inju-
ries to third parties who misuse prescription drugs attract
greater concern from public officials than injuries suffered
by legitimate patients? Purdue recently announced plans to investigate the pos- sibility of including another ingredient (naltrexone) that might
counteract efforts to defeat the slow-release mechanism, but
it will have to conduct trials to determine the safety and
efficacy of this combination and then await FDA approval of
a supplemental NDA, which could take several years. 143 What if naltrexone reduces the effectiveness of OxyContin, as hap-
pened during clinical trials using a similar ingredient
(naloxone), 144 or else causes adverse reactions in some sub- set of users? From the perspective of the patient with cancer
or other type of intractable pain, such a new form of the drug
definitely would not represent an improvement. Moreover, what if naltrexone does not really help pre- vent misuse — should OxyContin never have been marketed
because it poses greater risks to nonusers than some of the
older opioids? Is the extended-release feature not a substan-
tial enough utility compared to other narcotic pain relievers
to justify continued marketing in light of emerging patterns
of diversion (much in the same way that Congress evaluated
relative risks and benefits when it decided to reschedule
methaqualone)? And, finally, who should make these sorts
of choices? Tentative answers to such questions appear be-
low, after first considering some of the other possible
regulatory responses. Marketing and distribution Just as happens with formulation issues, the abuse potential
of a drug may extend beyond the intrinsic characteristics of
the active ingredient to include how the manufacturer pro-
motes the drug product to health care professionals. Some
critics have alleged that Purdue Pharma overpromoted
OxyContin for the treatment of temporary or less serious
pain, arguing that this led to excessive prescribing and cre-
ated a larger supply for potential diversion. 145 Even though the DEA does not regulate the marketing of controlled sub-
stances (leaving that task to the FDA), it has castigated the
manufacturer for its aggressive promotion of OxyContin to
physicians. 146 Critics also object that Purdue and Abbott made OxyContin generally available for prescribing by any physi- 63 The Journal of Law, Medicine & Ethics Not for Distribution cian and dispensing by any pharmacy. The companies might
have decided to supply the drug only to hospital pharmacies
for dispensing, and only in response to a prescription by a
pain management (or similar) specialist, but such a restricted
distribution network would have been unprecedented and
perhaps even unlawful. 147 The FDA generally does not have the authority to restrict the distribution of drugs that it ap-
proves. 148 Although the DEA clearly enjoys the power to limit the channels of distribution for controlled substances by vir-
tue of its scheduling decisions, it does not usually impose
more precise restrictions tailored to a particular drug. 149 Ei- ther agency could attempt to persuade a manufacturer to
accept nominally voluntary limitations that they could not
mandate directly, 150 but that did not happen at the time of OxyContin’s approval. In whatever manner achieved, distribution restrictions would have important practical consequences for patients.
Although more stringent limitations may reduce the threat of
diversion and abuse, they also may complicate access for
legitimate users of these drugs. For instance, with just over
1,000 pain management specialists practicing in the United
States, 151 patients would have difficulty getting prescriptions for needed drugs if only such specialists were permitted to
prescribe them. Similarly, patients would find it inconve-
nient if they regularly had to fill their prescriptions at a hospital
pharmacy. Even without distribution restrictions, of course,
physicians hesitate before prescribing controlled substances, 152 and local pharmacists may fail to stock them. 153 One should not lose sight of the fact that the FDA’s decision to restrict
access to some analgesics on prescription-only creates an
important barrier, 154 both because physicians have ethical and legal obligations designed to limit inappropriate pre-
scribing and because of the practical (especially financial)
hurdles involved in visiting a physician. Schedule II, by fur-
ther restricting physician flexibility and by insisting on repeat
office visits (through the no-refill rule), enhances these barri-
ers to patient access. In July 2001, the FDA mandated labeling revisions for OxyContin to provide stronger warnings, 155 and, a few months later, it convened one of its advisory committees to provide
additional recommendations. 156 States have gotten involved as well: A few have limited Medicaid reimbursement for
OxyContin, 157 and members of the National Association of Attorneys General have discussed options for curbing abuse
of the drug with representatives from the manufacturer. 158 At the same time, the DEA urged Purdue Pharma to consider
restricting distribution to pain management specialists on
the theory that these physicians would know to use the drug only
as a last resort, if other pharmaceutical options did not help a
patient. 159 In testimony before Congress, the acting Administra- tor of the DEA even threatened to slash the company’s annual
production quota by approximately 95 percent. 160 Ultimately, such responses to widespread misuse may undermine recent efforts to provide patients with better pain management. It makes little sense to protect irresponsible
physicians and illegitimate users from their own bad judg-
ment if it means sacrificing the welfare of those in genuine
need. 161 Abuse and diversion remain unlawful, of course, and persons who violate the CSA may suffer serious legal
and other consequences, but agency initiatives that attempt
to restrict access by limiting supplies or channels of distribu-
tion would reflect an unfortunate pursuit of administrative
expediency or a response to the failure of more precisely
targeted law enforcement efforts. It is particularly challenging, of course, to target abusers when large supplies of a drug legally move through channels
of commerce. Unlike illicit substances that law enforcement
officials can attempt to interdict at the source or while still
moving through a distribution network, controlled substances
approved for medical use do not become a law enforcement
concern until diverted by drug abusers fairly late in the dis-
tribution process. Agencies will have a natural inclination to
reduce the undoubted difficulty of their task by restricting
supplies, but they must not lose sight of the other half of the
equation. 162 To its credit, the DEA recently took the unprec- edented step of issuing a public statement joined by numerous
public health groups to emphasize the importance of not
letting concerns about abuse interfere with the legitimate use
of OxyContin. 163 Although an important gesture, it remains to be seen whether this conciliatory rhetoric translates into
more enlightened regulatory responses by the agency when
confronted with calls for swift action to crack down on the
next wave of controlled substances abuse. The experience with antibiotics may offer an instructive contrast. Physicians continue to overprescribe these often
powerful prescription drugs with attendant risks to their pa-
tients’ health. 164 Patients also may abuse antibiotics, whether by disregarding dosage and duration of use instructions or by
passing them along to family members and friends. 165 Unlike the abuse and diversion of controlled substances, none of
this unwise behavior violates federal law. The same prob-
lems may, of course, arise with any pharmaceutical product, 166 but antibiotic misuse carries a societal risk as well — wide-
spread overuse has created drug-resistant strains of infectious
agents. 167 As with analgesics, this explains the need to con- tinue developing new and improved antimicrobial agents even
though the old stand-bys usually work well enough for most
patients with simple bacterial infections. 168 So far, public health agencies have responded by pleading with physicians
to exercise restraint in prescribing, 169 but some commenta- tors would go further and restrict access to the latest
compounds. 170 Because individual physicians and patients do not directly bear the diffuse societal risks associated with
the spread of resistance, and because they do not face any
real legal consequences for misusing antibiotics, a paternal-
istic strategy of limiting access has much to recommend it. 171 Because law enforcement tools already exist to deal with the
abuse and diversion of controlled substances, however, fed- 64 Volume 31:1, Spring 2003 Not for Distribution eral access restrictions that may interfere with legitimate use
seem far less defensible. Perhaps the central lesson from this brief discussion of antibiotics is that neither the FDA nor the DEA has ap-
proached pain management issues from the proper perspective.
Traditionally, the FDA has adopted a clinical (or individual-
istic) mindset, leaving most of the difficult risk-benefit
judgments in the hands of health care professionals and pa-
tients. Although such an attitude has much to commend it,
the agency may have placed excessive faith in the good sense
of physicians and the power of labeling to encourage proper
use and to limit the occasions for inappropriate prescrib-
ing. 172 Conversely, the DEA’s law enforcement mindset goes to the opposite extreme, giving perhaps undue weight to the
negative externalities associated with access to narcotics and
not trusting health care professionals. A public health per-
spective, which the Centers for Disease Control and
Prevention (CDC) has expressed in connection with antibi-
otics as well as vaccine programs, 173 might help to mediate between these two potentially incompatible perspectives. A public health approach to concerns about the overuse of narcotic analgesics might bring with it a variety of inter-
mediate regulatory responses. For one thing, the government
might limit access to those medical specialists who usually
encounter persons suffering severe or chronic pain — in-
cluding, for instance, oncologists and orthopedic surgeons
along with pain specialists — in the hopes that such special-
ists would better resist the tendency to prescribe Schedule II
analgesics for patients for whom milder agents would work
equally well. As mentioned previously, however, this would
risk creating serious access problems for legitimate patients,
at least if the range of specialists was defined too narrowly. The federal government also could try to limit promo- tional efforts. At present, the FDA does not permit
manufacturers of Schedule II drugs to advertise directly to
consumers, and it might restrict the distribution of free
samples to physicians. An outright prohibition on advertis-
ing directed to physicians could, however, run afoul of the
First Amendment. 174 Enhanced tracking of prescriptions of- fers still another — though controversial — option worth
exploring, and the DEA has encouraged more states to set up
prescription monitoring programs. 175 Perhaps enhanced agency efforts to educate health care professionals would offer the
best mechanism for achieving the ideal balance between pro-
moting appropriate use in patients and discouraging excessive
or otherwise inappropriate prescribing. 176 C ONCLUSION Federal agencies represent the first, but hardly the only, line
of defense against the misuse of pain management technolo-
gies. Their licensing decisions should not reflect an excessive
preoccupation with the potential for abuse unless the prod-
ucts genuinely have no value as therapeutic interventions. If an analgesic drug or medical device offers a relatively safe
and effective option for the treatment of pain in some group
of patients, then any concerns about misuse and diversion
need to balance the therapeutic benefit for legitimate users
against the risk that individuals who act unlawfully may in-
jure themselves and others. It would be unfortunate if an
inability to deal with the latter problem by other means (in-
cluding educational efforts as well as state and local policing)
led to regulatory decisions that denied effective relief to those
in pain. Federal officials must resist the temptation to place
law enforcement imperatives ahead of genuine medical need. A CKNOWLEDGMENTS This project was made possible by the generous support of
the Mayday Scholars Fund and the American Society of Law,
Medicine & Ethics. I would like to thank David Brushwood,
Barry Furrow, Tim Greaney, Diane Hoffmann, Sandra
Johnson, Ben Moulton, Barbara Noah, Christina Spellman,
Deirdre Schweiss, Nicolas Terry, and Steve Ziegler for their
comments. R EFERENCES 1. See J.A. Page, “Book Review,” Food & Drug Law Journal, 47 (1992): 459–81 (reviewing C.C. Mann & M.L. Plummer, The
Aspirin Wars: Money, Medicine, and 100 Years of Rampant Compe-
tition (New York: Knopf, 1991)); J.R. McTavish, “What’s in a
Name? Aspirin and the American Medical Association,” Bulletin
of the History of Medicine, 61 (1987): 343–66. See generally J.
Liebenau, Medical Science and Medical Industry: The Formation of
the American Pharmaceutical Industry (Baltimore: Johns Hopkins
Univ. Press, 1987). A few years earlier, Bayer had synthesized
heroin for possible use as an analgesic. 2. See R.A. Merrill, “The Architecture of Government Regu- lation of Medical Products,” Virginia Law Review, 82 (1996):
1753–866, at 1759–61; see also United States v. Doremus, 249
U.S. 86 (1919) (rejecting a constitutional challenge to the Harrison
Narcotic Act of 1914). See generally L. Noah & B.A. Noah, Law,
Medicine, and Medical Technology: Cases and Materials (New York:
Foundation Press, 2002). 3. See M.J. Field & C.K. Cassel, eds., Approaching Death: Improving Care at the End of Life (Washington, D.C.: National
Academy Press, 1997): at 190–98; H. McIntosh, “Regulatory
Barriers Take Some Blame for Pain Undertreatment,” Journal of
the National Cancer Institute, 83 (1991): 1202–04; see also A.M.
Gilson and D.E. Joranson, “Controlled Substances and Pain
Management: Changes in Knowledge and Attitudes of State
Medical Regulators,” Journal of Pain & Symptom Management,
21 (2001): 227–37 (finding some signs of improvement). 4. See Drug Amendments of 1962, Pub. L. No. 87-781, § 102, 76 Stat. 780, 781 (codified as amended in scattered sections
of 21 U.S.C.); M.P. van Huysen, “Reform of the New Drug Ap-
proval Process,” Administrative Law Review, 49 (1997): 477–99;
Note, “Drug Efficacy and the 1962 Drug Amendments,”
Georgetown Law Journal, 60 (1971): 185–224; see also Medical
Device Amendments of 1976, Pub. L. No. 94-295, 90 Stat. 539
(codified as amended in scattered sections of 21 U.S.C.); P.B. Hutt
et al., “The Standard of Evidence Required for Premarket Ap-
proval Under the Medical Device Amendments of 1976,” Food 65 The Journal of Law, Medicine & Ethics Not for Distribution & Drug Law Journal, 47 (1992): 605–28. 5. See, e.g., United States v. Universal Mgmt. Serv., Inc., 191 F.3d 750, 754–55 (6th Cir. 1999) (electric stimulators); United
States v. An Article . . . Acu-Dot, 483 F. Supp. 1311, 1313–15
(N.D. Ohio 1980) (magnets); United States v. Articles of Device
(Acuflex; Pro-Med), 426 F. Supp. 366 (W.D. Pa. 1977) (electric
acupuncture devices); see also P.B. Hutt, “A History of Govern-
ment Regulation of Adulteration and Misbranding of Medical
Devices,” Food Drug Cosmetic Law Journal, 44 (1989): 99–117.
Although initially it had proposed treating acupuncture needles
as investigational devices, see 44 Fed. Reg. 63,292, 63,299 (1979),
the FDA eventually issued a classification regulation, see 61 Fed.
Reg. 64,616 (1996) (codified at 21 C.F.R. § 880.5580 (2002)). 6. See A. Hróbjartsson & P.C. Gøtzsche, “Is the Placebo Powerless? An Analysis of Clinical Trials Comparing Placebo with
No Treatment,” N. Engl. J. Med., 344 (2001): 1594–602, at
1596–97; D.J. Rowbotham, “Endogenous Opioids, Placebo Re-
sponse, and Pain,” Lancet, 357 (2001): 1901–02; J.A. Turner et
al., “The Importance of Placebo Effects in Pain Treatment and
Research,” JAMA, 271 (1994): 1609–14; see also S. Hoffman,
“The Use of Placebos in Clinical Trials: Responsible Research or
Unethical Practice?,” Connecticut Law Review, 33 (2001): 449–
501, at 497 (defending the use of placebos in clinical trials of
investigational drugs to treat moderate pain, but only under
limited circumstances). 7. See J.T. Farrar et al., “Clinical Importance of Changes in Chronic Pain Intensity Measured on an 11-Point Numerical Pain
Rating Scale,” Pain, 94 (2001): 149–58; E.S. Pryor, “Compensa-
tion and the Ineradicable Problems of Pain,” George Washington
Law Review, 59 (1991): 239–306, at 246–57; S. Rich, “Seeking
to End Disability Fraud, Government Lays Hands on Pain: Ex-
perts Seek Ways to Measure True Extent of Impairment,” Washington
Post, Feb. 3, 1986, at A13; E. Rosenthal, “Chronic Pain Fells
Many Yet Lacks Clear Cause,” New York Times, Dec. 29, 1992, at
C1. See generally G.M. Aronoff, ed., Evaluation and Treatment of
Chronic Pain, 3d ed. (Baltimore: Williams & Wilkins, 1999). 8. See R.A. Moore et al., “Size Is Everything — Large Amounts of Information Are Needed to Overcome Random
Effects in Estimating Direction and Magnitude of Treatment Ef-
fects,” Pain, 78 (1998): 209–16. 9. Cross-over designs may facilitate comparisons against pla- cebo or active controls. See FDA Center for Drug Evaluation &
Research, Guideline for the Clinical Evaluation of Analgesic Drugs,
Docket No. 91D-0425 (Dec. 1992): at 7–8 (“The salient advan-
tage of crossover designs lies in the potential for reduction of
experimental intersubject variation….”), available at <http://
www.fda.gov/cder/guidance/old041fn.pdf>; see also M.B. Max
et al., Advances in Pain Research and Therapy: The Design of Analge-
sic Clinical Trials (New York: Raven Press, 1991); J.A. Forbes,
“Diflunisal: A New Oral Analgesic with an Unusually Long Dura-
tion of Action,” JAMA, 248 (1982): 2139–42; A.R. Jadad et al.,
“Morphine Responsiveness of Chronic Pain: Double-Blind
Randomised Crossover Study with Patient-Controlled Analge-
sia,” Lancet, 339 (1992): 1367–71; S. Mercadante et al., “Analgesic
Effect of Intravenous Ketamine in Cancer Patients on Morphine
Therapy: A Randomized, Controlled, Double-Blind, Crossover,
Double-Dose Study,” Journal of Pain & Symptom Management,
20 (2000): 246–52; D.E. Moulin et al., “Randomised Trial of
Oral Morphine for Chronic Non-Cancer Pain,” Lancet, 347
(1996): 143–47. 10. L. Noah, “Pigeonholing Illness: Medical Diagnosis as a Legal Construct,” Hastings Law Journal, 50 (1999): 241–307, at
261. 11. See id. at 263 (“If only regarded as a symptom, proposed new … treatments will fare less well in the agency’s risk-benefit calculus and product approval decisions in the future.”); cf. M.
Segal, Comment, “Overdue Process: Why Denial of Physician-
Prescribed Marijuana to Terminally Ill Patients Violates the United
States Constitution,” Seattle University Law Review, 22 (1998):
235–63, at 258 (crafting a constitutional argument for medical
use of marijuana in part by downplaying the state’s regulatory
interest because, “unlike other controversial drugs such as la-
etrile, [which patients foolishly may select in lieu of conventional
therapies,] marijuana is not being advanced as a cure for any of
the diseases in question, but merely as a painkiller”). 12. Cf. J.C. Liebeskind, “Pain Can Kill,” Pain, 44 (1991): 3–4.
13. See E. Fox, Editorial, “Predominance of the Curative Model of Medical Care: A Residual Problem,” JAMA, 278 (1997):
761–63, at 762 (“[P]alliative care medicine is often intensely
concerned with the treatment of pain, despite the fact that pain
cannot be definitively verified and at times cannot even be ex-
plained.”). 14. See B.A. Rich, “A Prescription for the Pain: The Emerging Standard of Care for Pain Management,” William Mitchell Law
Review, 26 (2000): 1–91; R.J. Oken, Note, “Curing Healthcare
Providers’ Failure to Administer Opioids in the Treatment of
Severe Pain,” Cardozo Law Review, 23 (2002): 1917–92. 15. See S.G. Stolberg, “New Painkiller Is Withdrawn After Four Deaths,” New York Times, June 23, 1998, at A1; see also M.
Kaufman, “Report Says Drugmakers Innovate Less, Modify
More,” Washington Post, May 29, 2002, at A5. 16. See E. Elhauge, “The Limited Regulatory Potential of Medical Technology Assessment,” Virginia Law Review, 82 (1996):
1525–622, at 1593; M.J. Mehlman, “Health Care Cost Con-
tainment and Medical Technology: A Critique of Waste Theory,”
Case Western Reserve Law Review, 36 (1986): 778–877, at 788
(explaining that the FDA “has occasionally, albeit rarely, denied
approval to market a drug on the basis that it was less safe or less
effective than an alternative already on the market”). 17. See S. Hensley, “FDA Approves Pharmacia’s Bextra Months Sooner Than Expected,” Wall Street Journal, Nov. 19,
2001, at B5 (adding that the FDA has begun to approve newer
versions). 18. See J.N. Cashman, “The Mechanisms of Action of NSAIDs in Analgesia,” Drugs, 52, suppl. 5 (1996): 13–23. 19. See M.J. Langman et al., “Risks of Bleeding Peptic Ulcer Associated with Individual Non-Steroidal Anti-Inflammatory
Drugs,” Lancet, 343 (1994): 1075–78; M.M. Wolfe et al., “Gas-
trointestinal Toxicity of Non-Steroidal Anti-Inflammatory Drugs,”
N. Engl. J. Med., 340 (1999): 1888–99. 20. See C. Bombardier et al., “Comparison of Upper Gas- trointestinal Toxicity of Rofecoxib and Naproxen in Patients
with Rheumatoid Arthritis,” N. Engl. J. Med., 343 (2000): 1520–
28; M.J. Langman et al., “Adverse Upper Gastrointestinal Effects
of Rofecoxib Compared with NSAIDs,” JAMA, 282 (1999): 1929–
33; L.S. Simon et al., “Anti-Inflammatory and Upper
Gastrointestinal Effects of Celecoxib in Rheumatoid Arthritis: A
Randomized Controlled Trial,” JAMA, 282 (1999): 1921–28.
But see P. Jüni et al., Editorial, “Are Selective COX 2 Inhibitors
Superior to Traditional Non Steroidal Anti-Inflammatory Drugs?,”
British Medical Journal, 324 (2002): 1287–88. 21. See D. Mukherjee et al., “Risk of Cardiovascular Events Associated with Selective COX-2 Inhibitors,” JAMA, 286 (2001):
954–59, at 958; see also R.A. Bonnel et al., “Asceptic Meningitis
Associated with Rofecoxib,” Annals of Internal Medicine, 162
(2002): 713–15; L. Neergaard, “Painkillers May Delay Bone
Healing,” Associated Press Newswire, May 27, 2002 (reporting on
research that suggests narcotic analgesics may avoid an NSAID
side-effect in one class of patients). 22. See U.S. General Accounting Office, FDA Drug Review: 66 Volume 31:1, Spring 2003 Not for Distribution Postapproval Risks 1976–1985, PEMD-90-15 (1990): at 25, 101–05. 23. See U.S. General Accounting Office, FDA Premarket Ap- proval: Process of Approving Ansaid as a Drug, HRD-92-85 (1992):
at 3; B.L. Strom & P. Tugwell, “Pharmacoepidemiology: Current
Status, Prospects, and Problems,” Annals of Internal Medicine,
113 (1990): 179–81, at 180; see also Deficiencies in FDA’s Regu-
lation of the New Drug “Oraflex”, H.R. Rep. No. 98-511 (1983):
at 3–4 (noting liver toxicity and other serious reactions); A.C.
Rossi et al., “The Importance of Adverse Reaction Reporting by
Physicians: Suprofen and the Flank Pain Syndrome,” JAMA, 259
(1988): 1203–04. 24. See FDA’s Regulation of Zomax, H.R. Rep. No. 98-584 (1983): at 7–8; see also Nichols v. McNeilab, Inc., 850 F. Supp.
562, 564–65 (E.D. Mich. 1993) (holding that the manufacturer
had a duty to inform the public directly of the withdrawal); D.
Ross-Degnan et al., “Examining Product Risk in Context: Mar-
ket Withdrawal of Zomepirac as a Case Study,” JAMA, 270 (1993):
1937–42; cf. D.A. Kessler et al., “Approval of New Drugs in the
United States,” JAMA, 276 (1996): 1826–31, at 1831
(“[K]etorolac, a parenteral analgesic, has been retained on
the market … despite [serious] adverse effects … [because
they] were not deemed to outweigh ketorolac’s benefits, which
include an absence of the respiratory depression normally
associated with narcotic analgesics, typically the alternative thera-
peutic choice.”). 25. See id.
26. See J. Schwartz, “New Painkiller Taken Off Market After Deaths,” Washington Post, June 23, 1998, at A2. 27. See R. Sharpe, “How a Drug Approved by the FDA Turned into a Lethal Failure,” Wall Street Journal, Sept. 30, 1998,
at A1. 28. See id.
29. See id.
30. The FDA alerted physicians that the drug was unsafe when used longer than the 10 days tested in the clinical trials. See
id.; FDA, Warning Label Changes for Pain Reliever Duract, Talk
Paper, No. T98-6 (Feb. 10, 1998), available at <http://
www.fda.gov/bbs/topics/ANSWERS/ANS00849.html>. 31. See Letter from Wyeth-Ayerst Laboratories to Health Care Professionals (Feb. 1998), available at <http://www.fda.gov/
medwatch/safety/1998/duract.htm>. 32. See B.A. Noah, “Adverse Drug Reactions: Harnessing Experiential Data to Promote Patient Welfare,” Catholic Univer-
sity Law Review, 49 (2000): 449–504, at 488–89. 33. See FDA, “Wyeth-Ayerst Laboratories Announces the Withdrawal of Duract from the Market,” Talk Paper, No. T98-36
(June 22, 1998), available at <http://www.fda.gov/bbs/topics/
ANSWERS/ANS00879.html>. 34. See United States v. Article of Drug Labeled “Decholin”, 264 F. Supp. 473, 482 n.9 (E.D. Mich. 1967) (noting that the
FDA would not limit aspirin to prescription use even though “at
the root of a headache may lie anything from nervous tension to
a malignant brain tumor”). 35. See L. Noah, “Informed Consent and the Elusive Di- chotomy Between Standard and Experimental Therapy,” American
Journal of Law & Medicine, 28 (2002): 361–408, at 395 & n.178. 36. See Food, Drug & Cosmetic Act, ch. 675, §§ 502, 503, 52 Stat. 1040, 1050 (1938) (codified at 21 U.S.C. §§ 352(d),
353(b)) (requiring prescriptions for all habit-forming drugs),
amended by Durham-Humphrey Amendments, Pub. L. No. 215,
ch. 578, § 1, 65 Stat. 648 (1951), amended by Drug Abuse Con-
trol Amendments of 1965, Pub. L. No. 89-74, § 4, 79 Stat. 226,
amended by Controlled Substances Act, Pub. L. No. 91-513, title
II(G), 84 Stat. 1242, 1281-82 (1970); see also P.B. Hutt, “A Legal
Framework for Future Decisions on Transferring Drugs from Prescription to Nonprescription Status,” Food Drug Cosmetic Law
Journal, 37 (1982): 427–40, at 428, 435, 440; cf. 21 U.S.C. §
829(d) (2000) (allowing the DEA to recommend to the FDA
prescription status for OTC drugs with an abuse potential). 37. See 21 C.F.R. pt. 330 (2002); see also Cutler v. Hayes, 818 F.2d 879, 883–85 (D.C. Cir. 1987) (describing the OTC
drug review process); P.R. Jones, Note, “Protecting the Con-
sumer from Getting Burned: The FDA, the Administrative Process,
and the Tentative Final Monograph on Over-the-Counter Sun-
screens,” American Journal of Law & Medicine, 20 (1994): 317–35. 38. See 37 Fed. Reg. 14,633 (1972); see also 37 Fed. Reg. 26,456 (1972) (call for data on topical analgesics). 39. See 42 Fed. Reg. 35,346 (1977) (concluding, for instance, that a few ingredients used in then-marketed analgesics (e.g.,
phenacetin) were not generally recognized as safe and/or effec-
tive); see also 44 Fed. Reg. 69,768 (1979) (panel report for
external analgesics). 40. See 53 Fed. Reg. 46,204 (1988); see also 48 Fed. Reg. 5,852 (1983) (TFM for external analgesics), amended, 51 Fed.
Reg. 27,360 (1986). 41. The TFM includes a number of warnings applicable to aspirin. See 53 Fed. Reg. at 46,256 (to be codified at 21 C.F.R. §
343.50(c)). In addition, with the OTC drug review for internal
analgesics still pending, the FDA promulgated a requirement
that any nonprescription products containing aspirin include a
special warning against use during pregnancy. See 55 Fed. Reg.
27,776, 27,784 (1990) (codified at 21 C.F.R. § 201.63(e)). 42. See 66 Fed. Reg. 61,555, 61,575 (2001) (semiannual unified regulatory agenda forecasting final action on this mono-
graph by Dec. 2002). The most recent unified regulatory agenda
does not provide any estimated date of finalization for this rule.
See 67 Fed. Reg. 33,058 (2002). 43. See 21 C.F.R. § 330.11 (2002); 65 Fed. Reg. 24,704, 24,704–05 (2000); Farquhar v. FDA, 616 F. Supp. 190, 192
(D.D.C. 1985); see also S.P. Mahinka & E. Bierman, “Direct-to-
OTC Marketing of Drugs: Possible Approaches,” Food & Drug
Law Journal, 50 (1995): 49–63; L.R. Rook, “Listening to Zantac:
The Role of Non-Prescription Drugs in Health Care Reform and
the Federal Tax System,” Tennessee Law Review, 62 (1994): 107–
39; P. Temin, “Realized Benefits from Switching Drugs,” Journal
of Law & Economics, 35 (1992): 351–69. 44. See I. Molotsky, “Agency Approves Painkiller for Over- the-Counter Sales,” New York Times, May 19, 1984, at 1. The
FDA recently proposed amending the internal analgesics TFM
to include ibuprofen, which would eliminate the need to con-
tinue filing applications for supplemental or abbreviated new
drug approval for future OTC drug products containing this
active ingredient. See 67 Fed. Reg. 54,139 (2002). 45. See G. Mays, “Pain-Killer Wars Can Be a Pain for Ailing Consumers,” Chicago Tribune, Nov. 24, 1995, at Bus. 1. 46. Cf. Bober v. Glaxo Wellcome PLC, 246 F.3d 934, 939–40, 942 (7th Cir. 2001). 47. See S.G. Boodman, “Painful Choices: Consumers Face a Baffling Wall of Choices — and a Surprising Number of Serious
Risks — When They Seek Relief from Minor Pains and Illnesses at
the Drug Store,” Washington Post, Feb. 11, 2003, at Z1. Apart
from the risks associated with the proper use of OTC products,
they also may pose hazards of misuse. For example, responding
to numerous instances of childhood poisoning from the acciden-
tal ingestion of aspirin and similar products, Congress enacted
special child-proof packaging requirements. See Poison Preven-
tion Packaging Act of 1970, Pub. L. No. 91-601, 84 Stat. 1670
(codified as amended at 15 U.S.C. §§ 1471–1476 (2000)); 16
C.F.R. pt. 1700 (2002). In addition, after seven people in the
Chicago area suffered cyanide poisoning in 1982 from ingesting 67 The Journal of Law, Medicine & Ethics Not for Distribution tainted Extra-Strength Tylenol ® capsules, the FDA swiftly im- posed tamper-resistant packaging requirements for most OTC
drugs. See 47 Fed. Reg. 50,442, 50,449–50 (1982) (codified as
amended at 21 C.F.R. § 211.132); see also Federal Anti-Tamper-
ing Act, Pub. L. No. 98-127, 97 Stat. 831 (1983) (codified at 18
U.S.C. § 1365 (2000)). Although one occasionally hears com-
plaints that such rules have made it difficult for elderly and arthritic
consumers to open containers, these controls have reduced in-
stances of dangerous misuse at relatively trivial additional cost to
users. 48. See 50 Fed. Reg. 51,400, 51,401 (1985); see also Public Citizen Health Research Group v. Comm’r, FDA, 740 F.2d 21, 34–
35 (D.C. Cir. 1984) (remanding for lower court to consider claim
of unreasonable delay by the FDA in acting on a citizen petition
urging it to require a warning of this risk); American Home Prods.
Corp. v. Johnson & Johnson, 672 F. Supp. 135, 137–41 (S.D.N.Y.
1987) (offering a detailed account of the history behind the early
Reye syndrome warning efforts). 49. 21 C.F.R. § 201.314(h)(1) (2002).
50. 53 Fed. Reg. 21,633, 21,635 (1988).
51. See 63 Fed. Reg. 56,789, 56,801–02 (1998) (codified at 21 C.F.R. § 201.322 (2002)) (warning against the use of internal
analgesics in combination with heavy alcohol consumption); see
also Benedi v. McNeil-P.P.C., Inc., 66 F.3d 1378, 1387, 1389 (4th
Cir. 1995) (sustaining a negligence claim and punitive damage
award against the seller of Tylenol where it had delayed submit-
ting adverse reaction reports — concerning liver toxicity resulting
from interactions between acetaminophen and alcohol — to the
FDA during the OTC monograph review process for internal
analgesics); C. Adams, “Makers of Rival Pain Relievers Trade Jabs
on Safety,” Wall Street Journal, Sept. 18, 2002, at B1. 52. See S.H. Roth, Editorial, “Nonsteroidal Anti-inflamma- tory Drugs: Gastropathy, Deaths, and Medical Practice,” Annals
of Internal Medicine, 109 (1988): 353–54; C.M. Wilcox et al.,
“Striking Prevalence of Over-the-Counter Nonsteroidal Anti-
Inflammatory Drug Use in Patients with Upper Gastrointestinal
Hemorrhage,” Archives of Internal Medicine, 154 (1994): 42–46,
at 42; J. Foreman, “Painkillers Often Take Toll on Stomach,”
Boston Globe, July 8, 1996, at 25; J. Weber & Z. Schiller, “Pain-
killers Are About to O.D.,” Business Week, Apr. 11, 1994, at 54.
These two categories do not, however, exhaust the range of
options. For example, the prescription drug tramadol (Ultram ® and Ultracet ® ) qualifies as neither an NSAID nor an opioid anal- gesic. See J.E. Edwards et al., “Combination Analgesic Efficacy:
Individual Patient Data Meta-Analysis of Single-Dose Oral
Tramadol Plus Acetaminophen in Acute Postoperative Pain,” Jour-
nal of Pain & Symptom Management, 23 (2002): 121–30. 53. See N.I. Cherny, “Opioid Analgesics: Comparative Fea- tures and Prescribing Guidelines,” Drugs, 51 (1996): 713–37; A.
Jacox et al., “New Clinical-Practice Guidelines for the Manage-
ment of Pain in Patients with Cancer,” N. Engl. J. Med., 330
(1994): 651–55, at 653 (“Of the many methods available to
manage pain in cancer, drug therapy is the cornerstone because it
entails relatively little risk, is usually inexpensive, and as a rule
works quickly.”); M.H. Levy, “Pharmacologic Treatment of Can-
cer Pain,” N. Engl. J. Med., 335 (1997): 1124–32; H. McQuay,
“Opioids in Pain Management,” Lancet, 353 (1999): 2229–32;
C. Ripamonti & E.D. Dickerson, “Strategies for the Treatment of
Cancer Pain in the New Millennium,” Drugs, 61 (2001): 955–
77; P.C. Crowley, Comment, “No Pain, No Gain? The AHCPR’s
Attempt to Change Inefficient Health Care Practice of With-
holding Medication from Patients in Pain,” Journal of Contemporary
Health Law & Policy, 10 (1994): 383–403, at 391–93. See gener-
ally C. Stein, ed., Opioids in Pain Control: Basic and Clinical Aspects
(Cambridge, England: Cambridge Univ. Press, 1999). 54. See J. Laurance, “Are We Really Born to Suffer?,” Times of London, Jan. 27, 1997, at 18; see also 21 C.F.R. § 882.5890
(2002) (FDA regulation classifying transcutaneous electrical nerve
stimulation devices for pain relief); M.I. Johnson et al., “An In-
Depth Study of Long-Term Users of Transcutaneous Electrical
Nerve Stimulation,” Pain, 44 (1991): 221–29; cf. R.A. Deyo et
al., “A Controlled Clinical Trial of Transcutaneous Electrical Nerve
Stimulation (TENS) and Exercise for Chronic Low Back Pain,”
N. Engl. J. Med., 332 (1990): 1627–34 (finding the treatment
ineffective); E.A. Ghoname et al., “Percutaneous Electrical Nerve
Stimulation for Low Back Pain: A Randomized Crossover Study,”
JAMA, 281 (1999): 818–23 (evaluating a method supplying
deeper stimulation). 55. See Controlled Substances Act, Pub. L. No. 91-513, title II, 84 Stat. 1242 (1970) (codified as amended at 21 U.S.C. §§
801–904 (2000)). 56. 21 U.S.C. § 812(b)(1).
57. Id. § 812(b)(2).
58. See id. § 812(b)(3)–(5).
59. See Uniform Controlled Substances Act of 1994; see also R.L. Brown, “Uniform Controlled Substances Act of 1990,”
Campbell Law Review, 13 (1991): 365–74. 60. These examples come from the lists of substances ap- pearing in each of the schedules. See 21 U.S.C. § 812; 21 C.F.R.
pt. 1308 (2002). The brand-name versions of analgesic products
come from the Physicians’ Desk Reference, 56th ed. (Montvale,
New Jersey: Medical Economics Co., 2002). 61. 21 U.S.C. § 801(1).
62. See Alliance for Cannabis Therapeutics (ACT) v. DEA, 930 F.2d 936, 938 (D.C. Cir. 1991); see also L. Scott, “The Pleasure
Principle: A Critical Examination of Federal Scheduling of Con-
trolled Substances,” Southwestern University Law Review, 29
(2000): 447–500, at 455 (“The purpose of the legislation is to
legalize the possession and use of certain drugs for medical pur-
poses, and to criminalize their possession and use for any other
purposes.”). 63. Congress did set out a number of factors to consider, but these relate primarily to the potential for abuse rather than what
qualifies as currently accepted medical use: (1) [A substance’s] actual or relative potential for abuse.
(2) Scientific evidence of its pharmacological effect, if known.
(3) The state of current scientific knowledge regarding the drug or other substance. (4) Its history and current pattern of abuse.
(5) The scope, duration, and significance of abuse.
(6) What, if any, risk there is to the public health.
(7) Its psychic or physiological dependence liability.
(8) Whether the substance is an immediate precursor of a substance already controlled under this subchapter. 21 U.S.C. § 811(c); see also National Organization for the Reform
of Marijuana Laws (NORML) v. DEA, 559 F.2d 735, 747–48
(D.C. Cir. 1977) (suggesting that the potential for abuse rather
than medical use distinguishes the schedules). 64. See 21 U.S.C. § 811(b) (directing HHS to consider the listed factors); 116 Cong. Rec. 33,300 (1970) (statement by Rep.
Springer) (emphasizing “that purely enforcement responsibilities
are placed with the Department of Justice while medical and
scientific judgments necessary to drug control are left where they
properly should lie and that is with the Department of Health,
Education, and Welfare”); NORML, 559 F.2d at 745–47. 65. Pub. L. No. 95-633, § 101, 92 Stat. 3768 (1978) (codi- fied as amended at 21 U.S.C. § 801a(3)(C)) (emphasis added). 66. See 21 U.S.C. §§ 821–829.
67. See id. §§ 823(a), 826; 21 C.F.R. pt. 1303; see also MD 68 Volume 31:1, Spring 2003 Not for Distribution Pharm., Inc. v. DEA, 133 F.3d 8, 10–11, 16 (D.C. Cir. 1998)
(rejecting methylphenidate manufacturer’s challenge to the reg-
istration of a competitor); Western Fher Lab. v. Levi, 529 F.2d
325, 330–32 (1st Cir. 1976) (affirming challenged production
quotas for phenmetrazine); L. Noah, “Sham Petitioning as a
Threat to the Integrity of the Regulatory Process,” North Caro-
lina Law Review, 74 (1995): 1–73, at 9 n.24, 69 (discussing the
DEA’s effort to combat the delays that result when competitors
routinely file objections to each other’s manufacturer registra-
tion and production quota applications). 68. See 21 U.S.C. § 829(a); 21 C.F.R. §§ 1301.72(a), 1301.73, 1301.74(e), 1304.11(e)(3)(i), 1305.03, 1306.11(a), 1306.12;
United States v. Poulin, 926 F. Supp. 246, 249–55 (D. Mass. 1996)
(concluding that a pharmacy had violated numerous require-
ments applicable to Schedule II drugs); D.J. Pisano, “Controlled
Substances and Pain Management: Regulatory Oversight, For-
mularies, and Cost Decisions,” Journal of Law, Medicine & Ethics,
24 (1996): 310–16, at 311–12. 69. See 21 U.S.C. §§ 822–824; 21 C.F.R. § 1301.36; see also Humphreys v. DEA, 96 F.3d 658 (3d Cir. 1996) (reversing the
revocation of a physician’s certificate of registration); Kirk v. Mullen,
749 F.2d 297, 298 (6th Cir. 1984) (noting that the DEA pro-
cessed more than half a million CSA registrations annually). 70. See 62 Fed. Reg. 6,164 (1997); see also Conant v. McCaffrey, 2000 WL 1281174 (N.D. Cal. 2000) (invalidating
one aspect of this policy as inconsistent with the CSA when read
to take the First Amendment into account), aff ’d, 309 F.3d 629
(9th Cir. 2002); G.J. Annas, “Reefer Madness — The Federal
Response to California’s Medical-Marijuana Law,” N. Engl. J.
Med., 337 (1997): 435–39. 71. United States v. Oakland Cannabis Buyers’ Co-operative, 532 U.S. 483 (2001). For some of the relevant academic com-
mentary that predated the Court’s decision, see M.N. Cohen,
“Breaking the Federal/State Impasse over Medical Marijuana: A
Proposal,” Hastings Women’s Law Journal, 11 (2000): 59–74; J.P.
Kassirer, Editorial, “Federal Foolishness and Marijuana,” N. Engl.
J. Med., 336 (1997): 366–67; A.J. LeVay, Note, “Urgent Compas-
sion: Prosecutorial Discretion and the Medical Necessity Defense,”
Boston College Law Review, 41 (2000): 699–753; E.R. Neusch,
Comment, “Medical Marijuana’s Fate in the Aftermath of the
Supreme Court’s New Commerce Clause Jurisprudence,” Uni-
versity of Colorado Law Review, 72 (2001): 201–55. 72. See Cal. Health & Safety Code Ann. § 11362.5 (West Supp. 2002) (decriminalizing only the cultivation and possession
for use based on a physician’s recommendation); People v. Mower,
49 P.3d 1067 (Cal. 2002). See generally A.W. Bock, Waiting to
Inhale: The Politics of Medical Marijuana (Santa Ana: Seven Locks
Press, 2000). 73. See Oakland Cannabis, 532 U.S. at 491–94.
74. See id. at 500–01 & n.2 (Stevens, J., concurring in judg- ment) (calling the majority’s suggestion to the contrary dicta). 75. Id. at 493.
76. See id. at 492–93.
77. See H.R. Rep. No. 91-1444 (1970), reprinted in 1970 U.S.C.C.A.N. 4566, 4577–79; NORML v. Ingersoll, 497 F.2d
654, 657 (D.C. Cir. 1974). 78. See 21 U.S.C. § 321(g)(1)(A) (2000); see also State v. Wakeen, 57 N.W.2d 364, 369 (Wis. 1953) (noting that the phar-
macy practice statutes in most states cross-reference the U.S.P.).
But cf. United States v. Article of Drug . . . Ova II, 414 F. Supp. 660,
665–66, 667–73 (D.N.J. 1975) (rejecting an FDA effort to assert
its drug authority over a product simply by virtue of its inclusion
in the U.S.P.), aff ’d mem., 535 F.2d 1248 (3d Cir. 1976). 79. See E. Russo, “Cannabis for Migraine Treatment: The Once and Future Prescription? An Historical and Scientific Re- view,” Pain, 76 (1998): 3–8. See generally L. Grinspoon & J.B.
Bakalar, Marihuana: The Forbidden Medicine, rev. ed. (New Ha-
ven: Yale Univ. Press, 1993); M.L. Mathre, ed., Cannabis in Medical
Practice: A Legal, Historical and Pharmacological Overview of the
Therapeutic Use of Marijuana (Jefferson, North Carolina:
McFarland & Co., 1997). 80. See I.D. Meng et al., “An Analgesia Circuit Activated by Cannabinoids,” Nature, 395 (1998): 381–83; E.M. Williamson
& F.J. Evans, “Cannabinoids in Clinical Practice,” Drugs, 60 (2000):
1303–14; R.L. Hotz, “Chemicals in Pot Cut Severe Pain, Study
Says,” Los Angeles Times, October 27, 1997, at A1; see also L.
Grinspoon & J.B. Bakalar, “Marihuana as Medicine: A Plea for
Reconsideration,” JAMA, 273 (1995): 1875–76. 81. See J.E. Joy et al., eds., Marijuana and Medicine: Assessing the Science Base (Washington, D.C.: National Academy Press,
1999): at 145 (concluding that cannabinoid drugs may have a
therapeutic potential for pain relief); id. at 179 (“Until a
nonsmoked rapid-onset cannabinoid drug delivery system be-
comes available, we acknowledge that there is no clear alternative
for people suffering from chronic conditions that might be re-
lieved by smoking marijuana, such as pain….”). 82. See Oakland Cannabis, 532 U.S. at 502 n.4 (Stevens, J., concurring in judgment) (noting, in addition to California’s law,
the passage of voter initiatives in Alaska, Colorado, Maine, Ne-
vada, Oregon, and Washington, along with legislative action in
Hawaii); M. Tiersky, Comment, “Medical Marijuana: Putting
the Power Where It Belongs,” Northwestern University Law Re-
view, 93 (1999): 547–95, at 551, 578–84 (describing and
defending these various state initiatives). 83. See E. Goodman, Editorial, “The Uses of Pot,” Washing- ton Post, Aug. 4, 2001, at A23. 84. Not Legalizing Marijuana for Medicinal Use, Pub. L. No. 105-277, Div. F, 112 Stat. 2681–760, 2681–761 (1998). 85. See A.S. Trebach, The Heroin Solution (New Haven: Yale Univ. Press, 1982): at 59–84; A. Mondzac, “In Defense of the
Reintroduction of Heroin into American Medical Practice and
H.R. 5290 — The Compassionate Pain Relief Act,” N. Engl. J.
Med., 311 (1984): 532–35, at 533; E.L. Shapiro, “The Right to
Privacy and Heroin Use for Painkilling Purposes by the Termi-
nally Ill Cancer Patient,” Arizona Law Review, 21 (1979): 41–59,
at 43–48. 86. See T. Bennett, “The British Experience with Heroin Regulation,” Law & Contemporary Problems, 51 (Winter 1988):
299–314. 87. In a related vein, the FDA at one time categorically re- fused to consider prior foreign use of an ingredient in food as
providing evidence of the safety of a substance, but a court invali-
dated the policy because it found “no basis for a purely
ethnocentric distinction of this kind, divorced from demographic
considerations.” Fmali Herb, Inc. v. Heckler, 715 F.2d 1385, 1390
(9th Cir. 1983); see also L. Noah & R.A. Merrill, “Starting from
Scratch?: Reinventing the Food Additive Approval Process,” Bos-
ton University Law Review, 78 (1998): 329–443, at 354–55; R.G.
Pinco, “Implications of FDA’s Proposal to Include Foreign Mar-
keting Experience in the Over-the-Counter Drug Review Process,”
Food & Drug Law Journal, 53 (1998): 105–22. 88. See S.M. Stoll, Comment, “Why Not Heroin? The Controversy Surrounding the Legalization of Heroin for Thera-
peutic Purposes,” Journal of Contemporary Health Law & Policy,
1 (1985): 173–94, at 190–93; see also id. at 179 (“Clearly, the
United States maintains a model of drug control more suited to
law enforcement than to medical concerns.”); P.W. Fitzgerald,
Comment, “Members of Congress as Medical Experts: Heroin
and the Compassionate Pain Relief Act,” St. Louis University Pub-
lic Law Review, 6 (1987): 371–90. 69 The Journal of Law, Medicine & Ethics Not for Distribution 89. See H.R. Rep. No. 98-534, at 5 (1984) (asking “whether the adverse health effects caused by diversion of a drug outweigh
its therapeutic usefulness,” and concluding that methaqualone
“has no unique therapeutic advantages over other available drugs
and has a significantly higher incidence of and potential for
abuse”), reprinted in 1984 U.S.C.C.A.N. 540, 543–44. 90. See Pub. L. No. 98-329, 98 Stat. 280 (1984).
91. Congress may have set a similar precedent when it origi- nally decided to classify cocaine as a Schedule II “narcotic” even
though pharmacologically the substance does not qualify as a
narcotic. See United States v. Whitley, 734 F.2d 1129, 1140–41
(6th Cir. 1984) (upholding the classification as rational in order
to promote law enforcement purposes); United States v. Alexander,
673 F.2d 287 (9th Cir. 1982). 92. See 21 U.S.C. § 811(b) (2000).
93. See id.; Touby v. United States, 500 U.S. 160, 162, 167 (1991) (noting that the parties characterized this provision as
giving a “‘veto power’ to the Secretary of HHS”); American Pharm.
Ass’n v. Weinberger, 377 F. Supp. 824, 831 n.16 (D.D.C. 1974)
(explaining that the DEA “must first call upon FDA for its recom-
mendation. The recommendations of FDA, insofar as they concern
‘scientific and medical matters’ relating to the ‘appropriate sched-
ule, if any, under which such drug or substance should be listed’
are binding on the Attorney General.”), aff ’d, 530 F.2d 1054
(D.C. Cir. 1976). 94. H.R. Rep. No. 91-1444 (1970), reprinted in 1970 U.S.C.C.A.N. 4566, 4589. 95. See D.F. Musto, The American Disease: Origins of Narcotic Control, 3d ed. (New York: Oxford Univ. Press, 1999); R.E.
Barnett, Book Review, “Bad Trip: Drug Prohibition and the Weak-
ness of Public Policy,” Yale Law Journal, 103 (1994): 2593–630
(reviewing S.B. Duke & A.C. Gross, America’s Longest War: Re-
thinking Our Tragic Crusade Against Drugs (New York: G.P. Putnam’s
Sons, 1993)); E.G. Luna, “Our Vietnam: The Prohibition Apoca-
lypse,” DePaul Law Review, 46 (1997): 483–568; E.A. Nadelmann,
“Drug Prohibition in the United States: Costs, Consequences,
and Alternatives,” Science, 245 (1989): 939–47; M.D. Newcomb,
“Substance Abuse and Control in the United States: Ethical and
Legal Issues,” Social Science & Medicine, 35 (1992): 471–79. 96. See L. Noah, “Divining Regulatory Intent: The Place for a ‘Legislative History’ of Agency Rules,” Hastings Law Journal, 51
(2000): 255–323, at 301; S. Shapiro & T. McGarity, “Reorient-
ing OSHA: Regulatory Alternatives and Legislative Reform,” Yale
Journal on Regulation, 6 (1989): 1–63, at 57–62. 97. See Inspector General Act, Pub. L. No. 95-452, 92 Stat. 1101 (1978) (codified as amended at 5 U.S.C. app. II (2000));
M.J. Gates & M.F. Knowles, “The Inspector General Act in the
Federal Government: A New Approach to Accountability,” Ala-
bama Law Review, 36 (1985): 473–513; see also L. Noah,
“Scientific ‘Republicanism’: Expert Peer Review and the Quest
for Regulatory Deliberation,” Emory Law Journal, 49 (2000):
1033–83. 98. See 49 U.S.C. § 1131 (2000); M. Wald, “Two Positions on Safety,” New York Times, Aug. 30, 1998, at A16; “The FAA
Should Inspect Itself,” Washington Post, May 23, 1996, at A20;
see also A.J. Wood et al., “Making Medicines Safer — The Need
for an Independent Drug Safety Board,” N. Engl. J. Med., 339
(1998): 1851–54, at 1852–53 (advocating the creation of a simi-
lar counterweight to the FDA in order to improve postmarket
surveillance of drugs approved by the agency). 99. See, e.g., R.A. Merrill & J.K. Francer, “Organizing Fed- eral Food Safety Regulation,” Seton Hall Law Review, 31 (2000):
61–173; M. Allen & J. Mintz, “Homeland Department May
Take a Year to Take Shape,” Washington Post, Nov. 21, 2002, at
A8. 100. See J. Mashaw et al., Administrative Law: The American Public Law System, 4th ed. (St. Paul: West Group, 1998): at 21–
28, 174–75; E. Meidinger, “Regulatory Culture: A Theoretical
Outline,” Law & Policy, 9 (1987): 355–86, at 360, 372–74. 101. See 37 Fed. Reg. 18,097 (1972), remanded, NORML v. Ingersoll, 497 F.2d 654, 660–61 (D.C. Cir. 1974), petition denied,
40 Fed. Reg. 44,164 (1975), remanded, NORML v. DEA, 559
F.2d 735, 757 (D.C. Cir. 1977), petition denied, 44 Fed. Reg.
36,123 (1979), remanded, NORML v. DEA, No. 79-1660 (D.C.
Cir. Oct. 16, 1980), hearing announced, 51 Fed. Reg. 22,946
(1986), petition denied, 54 Fed. Reg. 53,767, 53,784 (1989) (“The
administrative law judge’s conclusion that a ‘respectable minor-
ity’ of physicians is all that is necessary to establish accepted medical
use in treatment in the United States is preposterous.”), remanded,
ACT v. DEA, 930 F.2d 936, 940–41 (D.C. Cir. 1991). 102. See United States v. Greene, 892 F.2d 453, 455–56 (6th Cir. 1989); Pearson v. McCaffrey, 139 F. Supp. 2d 113, 120–23
(D.D.C. 2001); NORML v. Bell, 488 F. Supp. 123, 132–43 (D.D.C.
1980) (three-judge court); cf. Washington v. Glucksberg, 521 U.S.
702, 791 (1997) (Breyer, J., concurring in the judgment) (sug-
gesting that the Court might hold it unconstitutional “were state
law to prevent the provision of palliative care, including the ad-
ministration of drugs as needed to avoid pain at the end of life”);
R.A. Burt, “The Supreme Court Speaks: Not Assisted Suicide but
a Constitutional Right to Palliative Care,” N. Engl. J. Med., 337
(1997): 1234–36. A lower federal court once found a funda-
mental right to receive acupuncture treatment. See Andrews v.
Ballard, 498 F. Supp. 1038, 1047–57 (S.D. Tex. 1980) (invalidat-
ing a state’s licensing restriction). 103. Cf. United States v. McMahon, 861 F.2d 8, 11 (1st Cir. 1988) (noting the “organic-synthetic distinction in Schedule I”
between marijuana and THC). Of course, similar concerns did
not dissuade the agency from its ill-fated effort to assert regula-
tory jurisdiction over cigarettes containing variable quantities of
the drug nicotine from tobacco leaves. See FDA v. Brown &
Williamson Tobacco Corp., 529 U.S. 120 (2000); L. Noah & B.A.
Noah, “Nicotine Withdrawal: Assessing the FDA’s Effort to Regu-
late Tobacco Products,” Alabama Law Review, 48 (1996): 1–63. 104. See 51 Fed. Reg. 17,476, 17,478 (1986) (synthetic dronabinol in sesame oil encapsulated in soft gelatin capsules);
see also 47 Fed. Reg. 10,080 (1982) (announcing the FDA’s pro-
posed rescheduling recommendation pending approval of the
NDA); R.M. Cooper, “Therapeutic Use of Marihuana and
Heroin: The Legal Framework,” Food Drug Cosmetic Law Jour-
nal, 35 (1980): 68–82, at 76–79 (defending the FDA’s earlier
recommendation against down-scheduling). 105. See 64 Fed. Reg. 35,928, 35,930 (1999).
106. See 57 Fed. Reg. 10,499, 10,507–08 (1992) (adding that, “[b]y any modern scientific standard, marijuana is no medi-
cine”); id. at 10,503 (“Beyond doubt, the claims that marijuana is
medicine are false, dangerous and cruel.”). Although one can
understand the DEA Administrator’s exasperated tone in once
again denying the petition, the published explanation contains a
surprising note of glibness and sarcasm. 107. 54 Fed. Reg. 53,767, 53,784 (1989). On the pitfalls of relying on the biomedical literature in this fashion, see L. Noah,
“Sanctifying Scientific Peer Review: Publication as a Proxy for
Regulatory Decisionmaking,” University of Pittsburgh Law Review,
59 (1998): 677–717. 108. See ACT v. DEA, 15 F.3d 1131, 1135–37 (D.C. Cir. 1994).
109. See United States v. Cannabis Cultivators Club, 5 F. Supp. 2d 1086, 1105 (N.D. Cal. 1998). Four years later, HHS again
recommended against down-scheduling marijuana. See Gettman
v. DEA, 290 F.3d 430, 432 (D.C. Cir. 2002) (holding that the
petitioners lacked standing to challenge the DEA’s subsequent 70 Volume 31:1, Spring 2003 Not for Distribution rejection of their request). 110. See P.J. Hilts, “After Two-Decade Halt, Marijuana Re- search Is Set,” New York Times, Dec. 15, 2001, at A16; see also M.
Robichaux, “Would Marijuana Be OK by Prescription If You Didn’t
Get High?,” Wall Street Journal, Feb. 28, 2001, at A1 (“Recent
findings suggest that THC holds more potential as a painkiller
than anyone ever guessed.”). 111. See 21 U.S.C. § 812(b) (2000).
112. See 57 Fed. Reg. at 10,504–07 (also requiring that the drug’s chemistry be known and reproducible); id. at 10,505
(“When a drug lacks NDA approval and is not accepted by a
consensus of experts outside FDA, it cannot be found … to have
a currently accepted medical use.”). The DEA had first described
these factors a few years earlier, but in combination with a few
others, see 53 Fed. Reg. 5156, 5157–58 (1988) (classifying
methylenedioxymethamphetamine (MDMA), commonly known
as Ecstasy, as a Schedule I controlled substance), 54 Fed. Reg. at
53,783–84, which the reviewing court rejected as unworkable,
see ACT v. DEA, 930 F.2d at 940. 113. See 57 Fed. Reg. at 10,503–04.
114. See United States v. 50 Boxes More or Less, 909 F.2d 24, 26–28 (1st Cir. 1990) (sustaining an FDA enforcement action
against an unapproved prescription drug for the treatment of
vascular headaches); United States v. Seven Cardboard Cases . . .
“Esgic with Codeine Capsules”, 716 F. Supp. 1221, 1224–25 (E.D.
Mo. 1989). 115. See Grinspoon v. DEA, 828 F.2d 881, 886–91 (1st Cir. 1987) (rejecting the notion that the absence of FDA approval
demonstrated the lack of a legitimate medical use); NORML v.
DEA, 559 F.2d 735, 748–50 & n.65 (D.C. Cir. 1977); see also
Reckitt & Colman, Ltd. v. DEA, 788 F.2d 22, 24 (D.C. Cir. 1986)
(describing the DEA’s decision to move buprenorphine, an opiate
derivative, from Schedule II to Schedule V on the recommenda-
tion of HHS after the FDA approved the drug as an analgesic).
After the remand in Grinspoon, the DEA adhered to its decision
to place MDMA in Schedule I. See United States v. Carlson, 87
F.3d 440, 444–45 (11th Cir. 1996); cf. R. Weiss, “On Ecstasy,
Consensus Is Elusive,” Washington Post, Sept. 30, 2002, at A7
(reporting that the FDA now has approved research — pending
authorization from the DEA — into MDMA’s possible efficacy as
a treatment for post-traumatic stress disorder). 116. See D.D. Rohde, “The Orphan Drug Act: An Engine of Innovation? At What Cost?,” Food & Drug Law Journal, 55 (2000):
125–43, at 138–39 (discussing the disagreement between the
agencies over gamma hydroxybutyrate (GHB), which appears to
be an effective treatment for narcolepsy but also facilitates date
rapes); see also Pub. L. No. 106-172, § 3(a)(1), 114 Stat. 7, 8
(2000); R. Rubin, “Company Wants ‘Date Rape’ Drug Approved
for Sleep Disorder Treatment,” USA Today, June 6, 2001, at 10D
(describing compromise legislation that placed GHB into Sched-
ule I for most purposes but Schedule III when used in
FDA-approved studies); A. Zitner, “Date-Rape Drug OK’d to
Treat Sleep Disorder,” Los Angeles Times, July 18, 2002, at A12
(reporting that the FDA approved GHB subject to stringent re-
strictions on patient access). 117. See Grinspoon, 828 F.2d at 897. In connection with the DEA’s decision to up-classify methamphetamine to Schedule II,
the courts have rejected objections that HHS had done too cur-
sory a medical and scientific review. See United States v. Lafoon,
978 F.2d 1183, 1184–85 (10th Cir. 1992). 118. 21 C.F.R. § 1306.04(a) (2002); see also United States v. Moore, 423 U.S. 122, 141–42 (1975) (“[P]rovisions throughout
the Act reflect the intent of Congress to confine authorized medi-
cal practice within acceptable limits.”); id. at 126–27, 139–45
(allowing felony conviction of physician who prescribed metha- done in an unorthodox detoxification program that more closely
resembled the activities of a “pusher”); United States v. Betancourt,
734 F.2d 750, 757 (11th Cir. 1984) (“[T]he jury needed medical
testimony as to what the drug is, how it is properly used, how it
can be abused and the medical profession’s view of the drug.”);
Noell v. Bensinger, 586 F.2d 554, 557–58 (5th Cir. 1978) (up-
holding the revocation of a physician’s certificate of registration
notwithstanding the fact that the only expert who testified had
stated that the prescription of amphetamines to counteract fa-
tigue comported with accepted standards of medical practice);
United States v. Green, 511 F.2d 1062, 1069–70 (7th Cir. 1975)
(upholding the DEA’s regulation even though the statute did not
explicitly require that a controlled substance only be prescribed
for a legitimate medical use); D.J. Behr, “Prescription Drug Con-
trol Under the Federal Controlled Substances Act: A Web of
Administrative, Civil, and Criminal Law Controls,” Washington
University Journal of Urban & Contemporary Law, 45 (1994): 41–
119, at 61–65, 109, 112–13; Annotation, “Federal Criminal
Liability of Licensed Physician for Unlawfully Prescribing or Dis-
pensing ‘Controlled Substance’ or Drug in Violation of the
Controlled Substances Act,” 33 A.L.R. Fed. 220 (1977 & Supp.
2002). 119. See D.B. Brushwood & J.J. Carlson, “The Pharmacist’s Responsibility to Evaluate Suspicious Prescriptions,” Food Drug
Cosmetic Law Journal, 46 (1991): 467–85, at 481 (noting that
the DEA’s Pharmacist Manual lists as one indicia of an illegitimate
prescription “whether the purported prescription order con-
tains an indication other than one found in the package insert”);
id. at 475 n.45 (“Pharmacists would have to question the appro-
priateness of virtually every prescription that is out of the ordinary,
in a way that is inconsistent with the federal framework in which
physicians are allowed wide latitude in prescribing.”); see also
United States v. Leal, 75 F.3d 219, 223 (6th Cir. 1996) (upholding
conviction of pharmacist); United States v. Hayes, 595 F.2d 258,
261 n.6 (5th Cir. 1979) (“[A] pharmacist can know that prescrip-
tions are issued for no legitimate medical purpose without his
needing to know anything about medical science.”). 120. See 21 U.S.C. § 396 (2000) (medical device regulation); 42 U.S.C. § 1395 (2000) (“Nothing in [Medicare] shall be con-
strued to authorize any Federal officer or employee to exercise
any supervision or control over the practice of medicine or the
manner in which medical services are provided.”); 37 Fed. Reg.
16,503, 16,504 (1972) (“[I]t is clear that Congress did not in-
tend the [FDA] to regulate or interfere with the practice of
medicine….”). 121. 48 Fed. Reg. 26,720, 26,733 (1983); see also 21 C.F.R. § 312.2(d) (explaining that the FDA’s investigational new drug
requirements “do[] not apply to the use in the practice of medi-
cine for an unlabeled indication of [an approved] new drug”);
Buckman Co. v. Plaintiffs’ Legal Comm., 531 U.S. 341, 350–51 &
n.5 (2001); FTC v. Simeon Mgmt. Corp., 391 F. Supp. 697, 706–
07 (N.D. Cal. 1975), aff ’d, 532 F.2d 708, 717 (9th Cir. 1976)
(“FDA has specifically recognized the legality of using drugs for
purposes other than those for which they have been found safe
and effective.”); J.D. Archer, Editorial, “The FDA Does Not Ap-
prove Uses of Drugs,” JAMA, 252 (1984): 1054–55. 122. See L. Noah, “Constraints on the Off-Label Uses of Pre- scription Drugs,” Journal of Products & Toxics Liability, 16 (1994):
139–65, at 139–44. 123. See J.R. Cooper et al., “Prescription Drug Diversion Control and Medical Practice,” JAMA, 268 (1992): 1306–10, at
1308–09. 124. 51 Fed. Reg. 17,476, 17,477 (1986) (adding, by way of explanation, that the “DEA has encountered practitioners who
attempt to justify illegal or improper distribution or dispensing 71 The Journal of Law, Medicine & Ethics Not for Distribution by claiming unique knowledge of a drug’s effectiveness for a
broad range of medical indications”). One decade later, the DEA
announced a similar threat against any physicians in California
who simply recommended the use of marijuana. See 62 Fed.
Reg. 6,164 (1997). 125. See B.S. Galer et al., “Individual Variability in the Re- sponse to Different Opioids: Report of Five Cases,” Pain, 49
(1992): 87–91; R.K. Portenoy, “Opioid Therapy for Chronic
Nonmalignant Pain: Clinicians’ Perspective,” Journal of Law,
Medicine & Ethics, 24 (1996): 296–309, at 298–99; see also C.
Ripamonti et al., “An Update on the Clinical Use of Methadone
for Cancer Pain,” Pain, 70 (1997): 109–15. 126. See S.H. Sindrup & K. Brøsen, “The Pharmacogenetics of Codeine Hypoalgesia,” Pharmacogenetics, 5 (1995): 335–46,
at 343; see also S. Olivier, “Tailor-Made Drugs?,” Times of Lon-
don, Feb. 12, 2002 (describing the impact of hormonal differences
on the effectiveness of analgesics). See generally L. Noah, “The
Coming Pharmacogenomics Revolution: Tailoring Drugs to Fit
Patients’ Genetic Profiles,” Jurimetrics Journal, 43 (2002): 1–28. 127. See Departments of Commerce, Justice, and State, the Judi- ciary, and Related Agencies Appropriations for 2002: Part 10 —
OxyContin: Hearings Before a Subcomm. of the House Comm. on
Appropriations, 107th Cong., at 19 (2001) (statement of Asa
Hutchinson, Administrator, DEA) (“Federal laws and regulations
do not attempt to define or set standards as to what constitutes
‘legitimate medical purpose’ or ‘the usual course of professional
practice,’ the requisite elements of lawful prescriptions under the
CSA and DEA regulations. Instead, DEA relies upon the medical
community to make these determinations.”); see also D.E.
Joranson & A.M. Gilson, “Policy Issues and Imperatives in the
Use of Opioids to Treat Pain in Substance Abusers,” Journal of
Law, Medicine & Ethics, 22 (1994): 215–23, at 216 (describing
the dronabinol restriction as an aberration). 128. See 66 Fed. Reg. 56,607, 56,608 (2001) (adding, how- ever, that “[p]ain management, rather than assisted suicide, has
long been recognized as a legitimate medical purpose justifying
physicians’ dispensing of controlled substances”); see also R.
Steinbrook, “Physician-Assisted Suicide in Oregon: An Uncertain
Future,” N. Engl. J. Med., 346 (2002): 460–64; J. Cordaro, Note,
“Who Defers to Whom? The Attorney General Targets Oregon’s
Death with Dignity Act,” Fordham Law Review, 70 (2002): 2477–
514; A. Trafford, “Don’t Dismiss This as Physician Paranoia,”
Washington Post, Mar. 12, 2002, at Z1. 129. In contrast, when a group of death row inmates peti- tioned the FDA in the early 1980s to restrict the off-label use of
Schedule II drugs for lethal injection, that agency declined to
exercise its enforcement discretion in deference to the choices
made by state penal officials. See Heckler v. Chaney, 470 U.S. 821
(1985) (rejecting a challenge to the agency’s decision); see also L.
Noah, Letter, “Attorney General’s Intrusion into Clinical Prac-
tice,” N. Engl. J. Med., 346 (2002): 1918. 130. See Oregon v. Ashcroft, 192 F. Supp. 2d 1077 (D. Or. 2002), appeal pending, No. 02-35587 (9th Cir. 2003). 131. In the less stringent schedules, some of the listed sub- stances refer to particular formulations and dosage strengths.
See 21 U.S.C. § 812(c)(III)(d) & (V); see also id. § 811(g)(1)
(calling for the descheduling of any nonnarcotic substance used
in an FDA-approved OTC drug product); United States v. Martinez,
950 F.2d 222, 223–24 (5th Cir. 1991) (construing this provi-
sion); United States v. Caperell, 938 F.2d 975, 978–79 (9th Cir.
1991). 132. For instance, alternative modes of delivery for other phar- maceutical products may reflect attempts to reduce pain associated
with their administration. See T. Chea, “MedImmune’s Pain-
Free Ambitions: If Approved by the FDA, FluMist Would Become First Vaccine Delivered as a Nasal Spray,” Washington Post, Mar.
18, 2002, at E1. 133. See K.H. Mosser, “Transdermal Fentanyl in Cancer Pain,” American Family Physician, 45 (1992): 2289–94; see also W. Jeal
& P. Benfield, “Transdermal Fentanyl: A Review of Its Pharmaco-
logical Properties and Therapeutic Efficacy in Pain Control,”
Drugs, 53 (1997): 109–38; D.M. Neighbors et al., “Economic
Evaluation of the Fentanyl Transdermal System for the Treat-
ment of Chronic Moderate to Severe Pain,” Journal of Pain &
Symptom Management, 21 (2001): 129–43. 134. See “Deaths Are Followed by Pain Patch Restrictions,” Chicago Sun-Times, Feb. 6, 1994, at 54 (reporting that the manu-
facturer strengthened warnings after several deaths were
associated with misuse of the Duragesic patch); see also Erony v.
Alza Corp., 913 F. Supp. 195 (S.D.N.Y. 1995) (allowing an inad-
equate warning claim to proceed on behalf of a teenager who
died after sucking on his father’s used Duragesic patches). 135. See D. Brown & J. Schwartz, “The Good and Bad Sides of a Narcotic Lollipop,” Washington Post, Jan. 31, 1994, at A3;
P.J. Hilts, “U.S. Urged to Bar Narcotic Lollipop for Children,”
New York Times, Jan. 26, 1994, at A10 (describing the DEA’s
objections); see also J. Brody, “The Forgotten Child in Treating
Pain Is the Child,” New York Times, Oct. 25, 1995, at C11 (noting
“drug companies’ reluctance to develop pediatric analgesics,”
and explaining that doctors have resisted using the fentanyl lolli-
pop with children). 136. See S.G. Boodman, “Narcotic Lollipop Gets Approval by FDA Panel,” Washington Post, Sept. 23, 1997, at Z5; see also R.
Payne et al., “Long-Term Safety of Oral Transmucosal Fentanyl
Citrate for Breakthrough Cancer Pain,” Journal of Pain & Symp-
tom Management, 22 (2001): 575–83. 137. See Hutchinson statement, supra note 127, at 15–17; A. Kumar, “Prescription Drug Abuse Soars,” Los Angeles Times, Jan.
17, 2003, at A30. In 2001, the National Institute of Drug Abuse
(NIDA) issued a report documenting startling levels of prescrip-
tion drug abuse. See K. Fackelmann, “Health Campaign Takes
Aim at Prescription Drug Abuse,” USA Today, Apr. 10, 2001, at
D7; see also B. Vastag, “Mixed Message on Prescription Drug
Abuse,” JAMA, 285 (2001): 2183–84, at 2184 (“At the same time
that NIDA is raising alarm bells about abuse potential, new stud-
ies point to chronic underprescribing of appropriate pain relief
and a low risk of addiction to prescription drugs.”). 138. See N. Aoki, “Abusing Pain Pills: Is Maker to Blame?,” Boston Globe, July 4, 2001, at D1; Editorial, “Curbing the
OxyContin Scourge,” Pittsburgh Post-Gazette, May 30, 2001, at
A11 (“OxyContin has provided long-lasting pain relief for hun-
dreds of thousands of cancer patients and others suffering from
chronic pain.”); see also C. Adams, “Painkiller’s Sales Far Ex-
ceeded Maker’s Plans,” Wall Street Journal, May 16, 2002, at D2. 139. See P. Tough, “The Alchemy of OxyContin,” New York Times Magazine, July 29, 2001, at 32; cf. Crocker v. Winthrop Lab.,
514 S.W.2d 429 (Tex. 1974) (allowing a tort claim to proceed
against the seller of the prescription analgesic Talwin ® (pentazo-

Comments