Download Journal Of Manged Care Pharmacy

ORIGINAL RESEARCH
Gabapentin Use in a Managed Medicaid Population
ANN M. HAMER, PharmD; DEAN G. HAXBY, PharmD,
BENTSON H. MCFARLAND, MD, PhD; and KATHY KETCHUM, RPh, MPA:HA
ABSTRACT
OBJECTIVE: The use of gabapentin, an antiepileptic agent, in a primary care setting
was evaluated to determine (a) the conditions being treated, (b) specialties of the
prescribers, (c) dose ranges, and (d) the extent of documentation and follow-up.
METHODS: A retrospective review of both claims data and patient charts was performed by a clinical pharmacist. Patients were identified from CareOregon and
Oregon Medicaid fee-for-service drug claim databases. All patients were recipients
of the Oregon Medicaid program known as the Oregon Health Plan (OHP) and were
enrolled members of CareOregon, a contracted OHP managed care organization that
primarily serves Medicaid recipients. All patients received care at one of four Oregon
Health and Science University primary care clinics.
RESULTS: Of the 105 patients studied, 95% received gabapentin for off-label diagnoses. Chronic pain and mental disorders were the diagnoses associated with the
majority of prescriptions for gabapentin. Dose and dose intervals varied greatly.
Very few patients (12%) had a documented efficacious response to gabapentin therapy. Of the patients started on gabapentin, 40% of patients had no documented
follow-up.
CONCLUSIONS: Almost all patients in this sample from the Medicaid managed population received gabapentin for off-label indications. Evidence from clinical trials does
not support the use of gabapentin for many of the conditions treated with
gabapentin in this study. Most patients did not appear to benefit from gabapentin
therapy.
KEYWORDS: Gabapentin, Medicaid, Drug Use Review, Off-Label Use
J Man Care Pharm 2002:(8)4:266-71
G
abapentin is an antiepileptic agent approved by the FDA
in 1993 for the adjunctive therapy of partial seizures with
or without secondary generalization. Despite being indicated only for the adjunctive treatment of seizures, utilization of
gabapentin has increased markedly since its introduction.
The National Institute for Health Care Management reported
that gabapentin was the 17th-ranked drug in terms of expenditures in 2000 and the top-selling anticonvulsant. They also reported a 63.3% increase in annual sales from 1999 to 2000 and a
45.9% increase in prescription volume.1 These are among the
largest increases reported for the top 50 drugs. Similar trends are
noted in the Oregon Health Plan (OHP), Oregon’s Medicaid program. In 2001, gabapentin was listed in the top-10 drug products
by total cost for the fee-for-service Oregon Medicaid program.
The rise in the number of prescriptions and drug spending for
gabapentin is not attributed solely to its use as an anticonvulsant.
In fact, much has been published on the off-label use of
gabapentin, particularly in psychiatric disorders and chronic pain
syndromes. Many of these studies, however, are open-label or
case studies and do not offer adequate evidence for the off-label
use of gabapentin. A limited number of randomized controlled
trials are available and have been summarized in Table 1.
Currently, there is little published information available regarding how gabapentin is being used in clinical practice. Due to the
widespread and growing use of this drug, it is important to assess
utilization patterns. The objective of this study was to evaluate the
use of gabapentin in a primary care setting to determine the following: (a) conditions being treated, (b) specialties of the prescribers, (c) dose ranges, and (d) extent of documentation of follow-up care, monitoring, and evidence of effectiveness.
■■ Methods
Authors
ANN M. HAMER, PharmD, is a Clinical Pharmacy Specialist, Oregon State
University, College of Pharmacy, Oregon Health and Science University,
Portland, Oregon; DEAN G. HAXBY, PharmD, is an Associate Professor of
Pharmacy Practice and Director of Medicaid Programs, Oregon State
University, College of Pharmacy, Oregon Health and Science University,
Portland, Oregon; BENTSON H. MCFARLAND, MD, PhD, is Professor of
Psychiatry, Public Health, and Preventive Medicine, Department of Psychiatry,
Oregon Health and Science University, Portland, Oregon; KATHY KETCHUM,
RPh, MPA, HA, is Coordinator of Medicaid Programs, Oregon State University,
College of Pharmacy, Oregon Health and Science University, Portland, Oregon.
AUTHOR CORRESPONDENCE: Dean Haxby, PharmD, Oregon State
University, College of Pharmacy, GH 212, Oregon Health and Science
University, Portland, OR 97201. Tel: (503) 494-5778; Fax: (503) 494-8797;
E-mail: [email protected].
Copyright©2002 Academy of Managed Care Pharmacy. All rights reserved.
266
Journal of Managed Care Pharmacy
JMCP
July/August 2002
Study Design
This study involved a retrospective review of both reimbursement
claims data and patient medical charts. Subjects included in the
study were identified from CareOregon and Oregon Medicaid feefor-service drug claim databases. CareOregon is a contracted OHP
managed care organization that primarily serves Medicaid recipients. These databases capture basic demographic information as
well as drug claim information from prescriptions dispensed at
Oregon pharmacies. Chart review and data collection were conducted by a clinical pharmacist. A data-collection form was developed and tested on a sample population. Information gathered by
the form included patient demographics, prescriber information,
indications for gabapentin use, secondary diagnoses, duration of
use, reasons for discontinuation, and follow-up information.
Vol. 8, No. 4
www.amcp.org
Gabapentin Use in a Managed Medicaid Population
TABLE 1
Summary of Randomized Controlled Trials Using Gabapentin to Treat Off-Label Conditions
STUDY
Neuropathic Pain
Backonja M, Beydoun A, et al.
7/96-3/97
RCT
8-week trial
Morello CM, et al.
3/97-12/97
RCT, crossover
2-week, 6-week trials
Gorson KC, et al.
unknown date
RCT, crossover
12-week trial
Rowbotham M, et al.
8/96-7-97
RCT
8-week trial
Bipolar Affective Disorder
Frye M, Post RM, et al.
RCT, crossover
3-week, 6-week monotherapy
evaluations
Pande AC, et al.
RCT
Spasticity/Tremor
Gironell A, et al.
RCT, crossover and
placebo-controlled
26-day trial
Ondo W, et al.
RCT, crossover and
placebo-controlled
TREATMENT
POPULATION
RESULTS
REFERENCE
3,600 mg/day
(forced max)
67% achieved
Uncontrolled diabetics
(75% type 2)
n=84 gabapentin
n=81 placebo
JAMA. 1998;280:1831-36
gabapentin=
900-1,800 mg/day
amitriptyline:
25-75 mg/day
Controlled diabetics
n=19 veterans
completing both
tx arms of
cross-over trial
Diabetics; control
not mentioned
n=19 gabapentin
n=21 placebo
Gabapentin vs. placebo
Difference in mean pain score at
end point=-1.2 (P<.001).
Difference in mean sleep
interference score=
-1.47 (P<.001).
No statistically significant
difference in pain intensity
scores or in pain relief.
Statistical improvement in only
1 of 4 end points; the MPQ
(McGill Pain Questionnaire) with
a mean reduction of 8.9 pts.
compared to 2.2 pts. placebo (P=0.03).
Decrease in average daily pain
score=33% gabapentin,
7% placebo (P<0.001).
J Neurol Neurosurg
Psychiatry. 1999;
66:251-52
900 mg/day
vs. placebo
3,600 mg/day
(65% achieved
max dose)
vs. placebo
Postherpetic
neuralgia
n=113 gabapentin
n=112 placebo
gabapentin= 3,987
±856 mg/day
lamotrigine=274
±128 mg/day
vs. placebo
adjunctive use of
gabapentin
900-3,600 mg/day
Refractory BAD or
unipolar mood
disorder
n=31
BAD, type I, current
symptoms despite
ongoing therapy
gabapentin=
Essential tremor
1,200 mg/day
n=16
propranolol=
120 mg/day
gabapentin=
Essential tremor
1,800 mg/day or
n=20
3,600 mg/day vs. placebo
Cutter N, et al.
RCT, crossover
26-day trial
2,700 mg/day
vs. placebo
Spasticity in MS
n=21 (19M, 2F)
Panic/Anxiety
Pande AC, et al.
RCT
14-week trial
900-3,600 mg/day
vs. placebo
Social phobia
n=69
600-3,600 mg/day
vs. placebo
Panic disorder
n=103
2,400 mg/day
vs. placebo
Migraine prophylaxis
n=143
Pande AC, et al.
RCT
8-week trial
Migraine Prophylaxis
Mathew NT, et al.
RCT
16-week trial
Arch Intern Med.1999;
159:1931-37
Ann Pharmacother. 2000;
34:802-07
Gabapentin monotherapy is
equivalent to placebo. (P=0.70)
Primary outcome measure was
the Clinical Global Impressions
Scale (CGI).
Both decreased Young Mania
Rating Scale, placebo significantly
more than gabapentin (P<0.05).
No significant difference
between groups on HAM-D.
J Clin Psychopharmacol.
2000;20:607-14.
No significant differences found
between active treatments on
Tremor Clinical Rating Scale
(P=0.20).
Patient global assessments
(P<0.05), observed tremor scores
(P<0.005), water pouring scores
(P<0.05), and ADL scores (P<0.005)
significantly improved. Low dose
equivalent to high dose.
Comparing changes from gabapentin
with changes from placebo, there were
significant differences noted in spasm
severity, interference with function,
painful spasm, global assessment,
Modified Ashworth Scale, and plantar
stimulation response. There was no
significant change in Digit Span, Digit
Symbol, the adjective generation
technique, or the EDSS.
Arch Neurol. 1999;
56:475-80
Bipolar Disorders.
2000;2:249-55.
(abstract only)
Movement Disorders. 2000;
15:678-82.
(abstract only)
Arch Phys Med Rehabil.
2000; 81:164-69
A significant (P<0.05) reduction in sex
was seen in active vs. placebo groups.
Further studies are required to determine
dose-response relationship.
No overall drug/placebo difference was
observed in scores on the Panic and
Agoraphobia Scale. (P=0.606).
J Clin Psychopharmacol.
1999;19:341-48
A significant number of patients
receiving gabapentin showed at least a
50% reduction in migraine rate
over placebo; 24 of 98 gabapentin
patients discontinued prematurely due
primarily to adverse effects.
Headache. 2001;41:
119-28.
www.amcp.org Vol. 8, No. 4 July/August 2002
JMCP
J Clin Psychopharmacol.
2000;20:467-71.
Journal of Managed Care Pharmacy
267
Gabapentin Use in a Managed Medicaid Population
TABLE 2
Conditions Treated With Gabapentin
in the Study Population
INDICATION
N=105 (%)
Pain
Neuropathies
Back pain
Unspecified chronic pain
Fibromyalgia
Headache/Migraine
Neuralgias
Other
75 (71)
31 (30)
10 (10)
8 (8)
6 (6)
4 (4)
3 (3)
13 (12)
Bipolar affective disorder
11 (11)
Epilepsy
5 (5)
Restless leg
3 (3)
Multiple sclerosis
3 (3)
Anxiety
2 (2)
Behavior disorders
2 (2)
Other
4 (4)
TABLE 3
■■ Results
Demographics
A total of 105 patients met the eligibility criteria, and all were
reviewed. Patients ranged in age from 22 to 83 years (mean 50.3
years; SD 12.9) and were predominantly female (73% versus 27%
male); 81% were classified as Caucasian, 8.6% African American,
6.6% Hispanic, and 3.8% Asian.
Prescriber Information
Initial Prescriber
Maintenance Prescriber
Discipline
N (%)
N=105
N (%)
N=105
Internal medicine
17 (16)
28 (26)
Family practice
15 (14)
28 (26)
Nurse practitioner
12 (11)
24 (23)
Psychiatrist
18 (17)
7 (7)
Neurologist
17 (16)
3 (3)
Pain specialist
7 (7)
4 (4)
Physician assistant
2 (2)
2 (2)
Other
6 (6)
3 (3)
11 (11)
6 (6)
Undetermined
Study Population
The study population was limited to those who were continuously eligible for the OHP and enrolled in CareOregon during the
entire study period. A total of 4,638 patients were continuously
eligible and enrolled in CareOregon during this period. To be
included in the retrospective chart review, patients must have had
at least one drug claim for gabapentin in the 33-month period
268
Journal of Managed Care Pharmacy
JMCP
July/August 2002
between January 1, 1998, and September 30, 2000. In addition,
participants must have been patients at one of 4 Oregon Health
and Science University primary care clinics (3 family medicine,
1 internal medicine). A pool of 29 physicians, 42 residents, 3
nurse practitioners, and 1 physician assistant attend patients at
these clinics. Exclusion criteria included disenrollment from the
OHP and inability to access the patient chart or electronic medical
record.
Patient confidentiality was maintained as data were collected,
entered into an Access database, and verified by the clinical pharmacist. Descriptive statistics were used for analysis. The study was
approved by the Oregon Health and Science University
Institutional Review Board (Human Subjects Committee).
Utilization
Table 2 summarizes the indications for use of gabapentin within
this study population. Ninety-five percent of patients were receiving gabapentin for an off-label indication. Various chronic pain
conditions accounted for the majority of gabapentin use (71.4%).
Mental health disorders accounted for 15% of the gabapentin use,
with bipolar disorder being the most common. Seizure disorders
only accounted for 5% of gabapentin use.
Prescriber information is outlined in Table 3. Therapy was initiated by primary care providers 45% of the time compared to
46% started by specialists. While 59% of patients continued to
have involvement by the initial prescriber, often there was a shift
of care. Specialists appeared to be less involved with maintenance
prescribing of gabapentin. Dosing of gabapentin was inconsistent.
There was a wide range of prescribed doses both between and
within each indication, as demonstrated in Table 4. Only 47% of
patients received gabapentin 3 times daily, the usual dosing frequency. Other patients received the drug once daily (26%), twice
daily (18%), or 4 times daily (9%).
After the initiation of gabapentin therapy, 40% (n=42) of
patients had no documented follow-up related to the drug. Only
12% (n=13) of the study patients had a documented efficacious
response. Any positive statement, regardless of how subjective,
was included as a response to the medication. A summary of these
patients is presented in Table 5. The average dose for those
patients where a positive response was documented was 1,770 mg
(range=300 mg-4,800 mg). The average duration of therapy for
these patients was 24.8 months, with 11 of the 13 patients remain-
Vol. 8, No. 4
www.amcp.org
Gabapentin Use in a Managed Medicaid Population
TABLE 4
Dosing of Gabapentin By Indication
Diagnosis
For all indications
Pain
Bipolar affective disorder
Epilepsy
Restless leg
Multiple sclerosis
Anxiety
Average Daily Dose (mg) Range (mg)
975
100–4,800
1,080
100–4,800
645
200–1,200
1,320
300–2,700
100
100
1,000
900–1,200
550
200–900
ing on therapy at the end of the study period. As expected, most
patients were receiving ongoing treatment for either chronic neuropathic pain or epilepsy and were using at least one other medication for the same purpose.
By the end of the study period, 68 patients (65%) had discontinued therapy. The average duration of therapy for patients who
discontinued gabapentin was 4.2 months (range=1 day to 4 years).
The most prominent documented reason for discontinuation was
lack of efficacy (43%; n=29), followed by adverse effects (12%;
n=8), nonadherence (10%; n=7), and symptom resolution (1%;
n=1). In 34% (n=23) of patients, the reason for discontinuation
was not documented.
The total cost of drug therapy for all 105 patients during the
study period was $1,198,613 ($11,415 per patient). An average of
$1,123 was spent per patient on gabapentin alone, or about 10%
of the total cost of drug therapy. Patients who were labeled as
responders had a higher average gabapentin expenditure of
$3,095 per patient.
■■ Discussion
Little clinical trial data exist (Table 1) to support the use of
gabapentin in many of the conditions being treated in this study.
Due to a lack of clinical trial data, there are no dosing or therapeutic guidelines to direct the use of gabapentin. Information that does
exist is primarily based on case reports and a few open-label trials.
While such reports are valuable for directing further research, they
are generally not sufficient as the basis of treatment decisions.
In this study, gabapentin was largely used for a variety of pain
syndromes. Clinical trials of gabapentin support its efficacy in
selected neuropathic pain syndromes. Specifically, gabapentin is
efficacious for the treatment of diabetic neuropathy and postherpetic neuralgia, especially when high doses are used.2,3 Data are
lacking for other types of chronic pain conditions. To date, only
one controlled trial has been published comparing gabapentin to
tricyclic antidepressants in the treatment of diabetic neuropathy.
Tricyclic antidepressants are generally considered first-line therapy, although toxicity in overdose is considered a disadvantage.
Gabapentin and amitriptyline appear to be comparable in efficacy
and tolerability for the treatment of diabetic neuropathy,4 but the
effective dose of gabapentin has an average wholesale price of $191
per month compared to $7 per month for amitriptyline.
There are two published randomized controlled trials that studied
the use of gabapentin in migraine prophylaxis. In the first study,
gabapentin failed to demonstrate efficacy greater than placebo.5
A second study reported efficacy with gabapentin for the prevention
of migraine when compared to placebo.6 The American Academy of
Neurology cites both of these studies in its published guidelines for
the prophylaxis of migraines. The AAN guidelines maintain that
there is limited evidence for the use of gabapentin in clinical practice.7
Studies of gabapentin in psychiatric disorders have ranged
from anxiety and social phobia to affective disorders and drug
dependence. Only two randomized trials have been published
that evaluate the use of gabapentin monotherapy in bipolar disorder. Both failed to demonstrate efficacy superior to placebo.8,9
Similarly, Pande et al. evaluated the use of gabapentin in the treatment of panic disorder and failed to show a difference from placebo in the Panic and Agoraphobia Scale.10 For other psychiatric
indications (e.g., anxiety, obsessive-compulsive disorder, drug and
alcohol withdrawal), randomized controlled trials are lacking.
This retrospective chart review found that gabapentin was
being used almost exclusively for off-label indications in the population reviewed. Unfortunately, it appeared that relatively few
patients benefited from treatment based on information documented in the medical record. Although gabapentin was being
used for chronic conditions, only a small number of patients
remained on therapy at the end of the study period.
It is interesting to speculate as to why gabapentin was being
used for such a wide variety of off-label indications. First, marketing may have played a role. The manufacturer of gabapentin has
been accused of aggressively promoting the drug for off-label indications due to the limited market for seizure disorders.11 Second,
patients with chronic pain and/or mood disorders can be difficult
to manage. Gabapentin may be viewed as a treatment option that
has relatively few serious side effects and drug interactions, reasonable tolerance, and is fairly easy to prescribe. Clinicians may be
hoping for a positive response while minimizing the risk of harm.
The pattern of gabapentin use documented in this study does
not appear to represent the best use of health care resources. Health
plans might want to consider options to promote more cost-effective use of gabapentin. Prior authorization is one option. Criteria
could be developed to approve the use of gabapentin for off-label
diagnoses where clinical trial data support its efficacy. Mandatory
follow-up could also be considered to ensure an appropriate therapeutic response after 2 or 3 months of treatment, before long-term
use is approved. Educational interventions such as academic detailing might also help promote the appropriate use of gabapentin.
There are a couple of important limitations to interpreting this
www.amcp.org Vol. 8, No. 4 July/August 2002
JMCP
Journal of Managed Care Pharmacy
269
Gabapentin Use in a Managed Medicaid Population
TABLE 5
Summary of Patients Who Responded Favorably to Gabapentin Therapy
Maximum Daily
Dose (mg)
Duration of
Therapy
2,700
20 months
300
Secondary Diagnoses
Concurrent Therapy
Charted Efficacious Response
Depression, Fibromyalgia,
Anxiety, Diabetes
Vicodin, Relafen, Lodine
“Neurontin is helpful but prefer Vicodin.”
12 months
Celebrex, Acupuncture.
“Gabapentin helps her feet and hands
quite a lot.”
4,800
23 months
Tylenol #3, Tegretol,
Darvocet, Vicodin
“Has made a substantial impact on pain.”
Peripheral neuropathy
300
5 months
Prednisone
“Satisfied with the results of Neurontin.”
Hyperesthesia
900
6 months
APAP
“Vague abdominal discomfort gone with
Neurontin.”
1,200
24 months
Depression, Substance Abuse,
Anxiety
Percocet
“Noted some benefit from it.”
900
45 months
Substance Abuse
Unspecified
neuropathic pain
3,200
30 months
Diabetic neuropathy
1,800
36 months
Amputation pain
1,800
Indication For Use
Peripheral neuropathy
Fibromyalgia
Transverse myelitis
HIV-related
neuropathy
Epilepsy
Depression
“Good initial clinical response.”
IBU, MS Contin
“Had some relief; has decreased the
burning pain.”
Diabetes
Ultram, Tylenol #3,
Nortriptyline
“Apparently improved with Neurontin.”
23 months
Depression
Tylenol #3, Vicodin,
Naprosyn
“She has noted particular improvement in
her generalized pain since starting
Neurontin.”
300
9 months
Diabetes
Amitriptyline
“Able to walk with much less foot pain.”
Epilepsy
1,200
42 months
Diabetes
Depakote
“Have apparently stopped most of the
episodes with the slight increase in
Neurontin.”
Intercostal neuralgia
3,600
48 months
Depression
Vicodin, Imipramine,
Catapres
“Pain is 6/10, which is tolerable.”
Diabetic neuropathy
study. First, we studied a Medicaid population at one academic
institution. These results may not generalize to other settings.
Additional studies are needed to confirm our findings. Second,
data were obtained only from the review of patient medical
records and reimbursement claims data. We did not collect information directly from patients or providers. Thus, definitive conclusions regarding effectiveness of gabapentin for various conditions cannot be drawn from our data. There is a tremendous need
for clinical trials to better define the role of gabapentin in a variety
of conditions for which it is currently being prescribed.
■■ Conclusion
This retrospective review demonstrated that gabapentin is prescribed for a wide variety of conditions. Almost all patients (95%)
received gabapentin for off-label diagnoses. The use of gabapentin
in many of these patients is not supported by evidence from ran270
Journal of Managed Care Pharmacy
JMCP
July/August 2002
domized clinical trials. Most patients did not appear to benefit
from gabapentin therapy, and many patients did not appear to
receive adequate follow-up once gabapentin was started.
DISCLOSURES
Funding for this research was provided by a service contract with Oregon
Medical Assistance Program. It was obtained by author Dean G. Haxby.
None of the authors involved with the creation of this project have financial
interests or are affiliated with any company or product mentioned in the article.
Author Ann M. Hamer served as principal investigator and author of the
study. Study concept and design were contributed primarily by Hamer, Haxby,
and authors Bentson H. McFarland and Kathy Ketchum. Analysis and interpretation of data were contributed primarily by Hamer, Haxby, and Ketchum.
Drafting of the manuscript was primarily the work of Hamer and Haxby.
Critical revision of the manuscript was primarily the work of Hamer, Haxby,
McFarland, and Ketchum. Statistical expertise was contributed primarily by
Hamer, Haxby, and Ketchum.
Vol. 8, No. 4
www.amcp.org
Gabapentin Use in a Managed Medicaid Population
REFERENCES
6. Mathew NT, Rapoport A, Saper J, et al. Efficacy of gabapentin in migraine
prophylaxis. Headache. 2001;41:119-28.
1. Department of Health and Human Services Office of Inspector General.
Medicaid pharmacy actual acquisition cost of brand-name prescription drug
products. A-06-00-00023, August 2001. Available at: http://www.hhs.gov/progorg/oas/reports/region6/60000023.htm. Accessed September 17, 2001.
7. American Academy of Neurology: U.S. Headache Consortium. Evidencebased guidelines for migraine headache in the primary care setting: pharmacological management for the prevention of migraine. Available at:
http://www.aan.com. Accessed February 14, 2002.
2. Backonja M, et al. Gabapentin monotherapy for the symptomatic treatment
of painful neuropathy. A multicenter, double-blind, placebo-controlled trial in
patients with diabetes mellitus. JAMA.1998;280:1831-36.
8. Frye M, et al. A placebo-controlled study of lamotrigine and gabapentin
monotherapy in refractory mood disorders. J Clin Psychopharm. 2000;20:607-14.
3. Rowbatham M, et al. Gabapentin for the treatment of postherpetic neuralgia. A randomized controlled trial. JAMA. 1998;280:1837-42.
4. Morello CM, et al. Randomized double-blind study comparing the efficacy
of gabapentin with amitriptyline on diabetic peripheral neuropathy pain. Arch
Intern Med. 1999;159:1931-37.
5. Wessely P, Baumgartner C, Klinger D, et al. Preliminary results of a doubleblind study with the new migraine prophylactic drug gabapentin. Cephalalgia.
1987;7(suppl 6):477-78.
9. Botts SR, Raskind J. Gabapentin and lamotrigine in bipolar disorder. Am J
Health-Syst Pharm. 1999;56:1939-44.
10. Pande AC, et al. Placebo-controlled study of gabapentin treatment of panic
disorder. J Clin Psychopharm. 2000;20:467-71.
11. Petersen, M. Whistle-blower says marketers broke the rules to push a
drug. New York Times. March 14, 2002;sect C:1.
www.amcp.org Vol. 8, No. 4 July/August 2002
JMCP
Journal of Managed Care Pharmacy
271