A Publication of Western Missouri Mental Health Center,
Department of Psychiatry, University of Missouri - Kansas City
May 1998 Volume 3, Issue 5
INSIDE...
Jodi Worrel, Pharm.D.,
Psychopharmacy Resident,
Western Missouri Mental Health Center.
Aggression has been described as "a domineering, forceful, or assaultive verbal or physical action toward another person as the motor component of the affects of anger, hostility, or rage."1 Aggressive behavior may also be directed toward oneself or an object. Aggression may be preceded by an aura, headache, or anxiety and is often unprovoked or triggered by trivial events. Most episodes of aggression are unpredictable, short in duration, and directed toward staff.2 Although aggression is not a diagnosis in and of itself, aggressive behavior is a primary symptom or associated feature of numerous Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) diagnoses, including antisocial and borderline personality disorders, autistic disorder, conduct disorder, dementia, mental retardation, oppositional defiant disorder, posttraumatic stress disorder, and premenstrual dysphoric disorder.3 Aggression is also observed with acute intoxication or withdrawal from various agents such as alcohol, amphetamines, barbiturates, cocaine, PCP, hallucinogens, anabolic steroids, cannabis, and benzodiazepines. Various medical conditions such as thyroid disease, stroke, pain syndromes, and acute infection are also associated with aggressive behavior.
Aggression research, in general, is in the preliminary stages and has focused on genetic, environmental, neuroanatomical, and neurochemical factors in an attempt to delineate the underlying mechanisms responsible for aggressive behavior.4 Placebo controlled studies are difficult to conduct in aggressive patients and drug-free washouts are unsafe; therefore, most of the aggression literature consists of case reports and uncontrolled studies. Also, many of the patients studied are on concomitant medications making it difficult to assess whether decreased aggression is the direct effect of medication or the successful treatment of the underlying disorder.2
Although genetic and environmental factors appear to modulate the development of aggression, drug treatment is based on neuroanatomical and neurochemical research. Various areas in the brain have been associated with aggressive behavior. Patients who exhibit uncontrolled rage often present with a prior history of head trauma. Up to 70% of individuals with brain injury due to blunt trauma exhibit irritability and aggressive behavior.4 Both the amygdala and the temporal lobes are involved with the regulation of emotions and both are susceptible to kindling. Anti-epileptic drugs such as carbamazepine and valproic acid are believed to be useful for the treatment of aggression due to their anti-kindling effects. Lesions of the prefrontal cortex have also been associated with aggressive behavior. The prefrontal cortex is believed to be involved with the inhibition of aggression. Subsequently, damage to this area may interfere with the mechanism responsible for the inhibition of aggression and the likelihood of aggressive behavior is increased.4
Multiple neurotransmitters are believed to be involved in the modulation of aggression. The neurotransmitters serotonin (5-HT), gamma-aminobutyric acid (GABA), norepinephrine (NE), and dopamine (DA) have all been linked to aggressive behavior in both animal and human studies. 4 Specifically, 5-HT and GABA deficiencies and NE and DA excess are associated with increased aggressive behavior. These neurotransmitter systems are interconnected such that changes in one system may affect or lead to changes in another.
Before starting chronic pharmacological treatment for aggressive behavior, several factors should be addressed. First, identify and treat the underlying disorder. Often with successful treatment of the underlying disorder, associated aggressive behavior abates. Differentiation of agitation and anxiety from aggression is important when choosing treatment options. Non-pharmacological interventions should also be utilized. Various non-pharmacological interventions include the use of non-threatening body language, establishing eye contact, addressing the individual by name, and remaining calm while taking command. Limits should be set and the environmental stimulation responsible for escalation should be minimized.2 If pharmacological treatment is deemed necessary, keep in mind that sedation is usually not desirable long-term. Typical antipsychotics are most effective in controlling acute aggression, mainly due to their sedative effects. Long-term use of typical antipsychotics for aggression often results in tolerance to sedative effects and increased likelihood of tardive dyskinesia. Finally, objective tools such as the overt aggression scale (OAS) should be implemented to measure response.
At this time there are no pharmacological agents FDA-approved for the treatment of aggression; however, a number of agents are currently being used "off-label" for the treatment of aggressive behavior.
Beta-blockers:
The exact mechanism of action of beta-blockers in aggression is unknown. The anti-aggressive action of beta blockers is believed to be related to NE blockade or secondary to antiakathisia effects. The majority of clinical studies utilizing beta blockers for aggression have focused on patients with schizophrenia or organic brain damage. Propranolol and nadolol have been shown to be effective in decreasing aggression in chronic psychiatric inpatients independent of psychotic symptoms. Propranolol should be initiated at 20 mg TID and titrated slowly by 40-60 mg every 3 days up to 120-320 mg/day (some have gone as high as 640 mg/day). Maintenance doses of nadolol and pindolol are 40-160 mg/day and 40-60 mg/day, respectively. Clinical response may take up to 2-4 weeks and an adequate trial is 6-8 weeks. Before initiating beta blocker therapy, preexisting cardiac disease must be ruled out. Vital signs (heart rate and blood pressure) and lipid profiles should be monitored closely because beta-blockers may cause bradycardia, hypotension, and elevated triglycerides. Some other adverse effects of beta-blockers include sedation, depression, and masking of hypoglycemic signs and symptoms. Finally, use beta blockers with caution in patients with asthma, diabetes, COPD, and preexisting cardiac disease.
Mood Stabilizers:
Mood stabilizers have been studied for the treatment of aggressive behavior in patients with mental retardation – developmental delay (MRDD), head injury, bipolar disorder, conduct disorder, borderline personality disorder, schizophrenia, depression, and violent prisoners.5
Lithium:
The mechanism of action of lithium for the treatment of aggression is believed to be due to decreased beta-receptor function and NE availability and increased 5-HT. Lithium may be considered as first line therapy for patients whose aggression is related to mania or cyclic affective disorders.2 Lithium is usually initiated at 300 mg TID or 10-17 mg/kg/d divided in 2 or 3 doses. Dosage is usually titrated until therapeutic blood levels between 0.5-1.2 mEq/L are achieved. Because lithium is renally eliminated, dosage adjustments are required for patients with compromised renal function. A 6-8 week trial at a therapeutic concentration is necessary to evaluate efficacy of lithium for the treatment of aggressive behavior. Baseline labs include thyroid function tests, CBC, renal function tests, EKG, and electrolytes (especially sodium). Due to its narrow therapeutic window, lithium levels must be drawn routinely. Monitor for signs of toxicity such as sedation, confusion, tremors, and seizures. Some common adverse effects associated with lithium are polyuria, polydipsia, tremor, hypothyroidism, and gastrointestinal intolerance.
Carbamazepine:
The mechanism of action of carbamazepine in aggression is believed to be related to its antikindling properties. Carbamazepine has been shown to be equally as effective as propranolol for reducing aggression in patients with intermittent explosive disorder.2 Carbamazepine dosing should begin at 200 mg BID or TID for 5-7 days and then increased as tolerated by 200 mg every 5-7 days until therapeutic plasma concentrations of 4-12 mcg/ml are reached. Since carbamazepine induces its own metabolism after about 3-5 weeks, dosage changes may be necessary based on plasma concentration. Monitor CBC with differential (leukopenia, aplastic anemia), LFTs, plasma concentration, and sodium. Adverse effects of carbamazepine include drowsiness, dizziness, leukopenia, rash, and hyponatremia.
Valproic acid and derivatives:
The mechanism of action of valproic acid and derivatives in aggression is due to antikindling effects as well as GABAergic potential. Dosing may be initiated with 250 mg BID or TID or 20 mg/kg as a loading dose and then titrated to a therapeutic concentration of 50-125 mcg/ml. Monitor CBC, LFTs, and plasma concentration. LFT’s may be 3-5 times normal before discontinuing treatment. Adverse effects of valproic acid and derivatives include gastrointestinal upset, drowsiness, liver toxicity, alopecia, tremor, weight gain, and thrombocytopenia.
Buspirone:
Buspirone is a partial 5-HT1A agonist. The mechanism of action of buspirone in aggression is believed to be due to increased serotonin. Buspirone has been studied clinically in patients with MRDD and head injury.6 Transient worsening of aggression initially has been reported. This worsening of aggression is thought to be due to a temporary decrease in brain 5-HT initially. Buspirone 15-60 mg/day for at least a 12 week trial is necessary to determine efficacy. Results from clinical studies suggest that higher doses (45-60 mg) are more effective than lower doses for the treatment of aggression. Some common adverse effects associated with buspirone include nausea, dizziness, and headache.
Selective Serotonin Reuptake Inhibitors (SSRIs):
The mechanism of action of SSRIs in aggression is believed to be due to increased 5-HT. The best efficacy of SSRIs for the treatment of aggression has been shown in patients with MRDD and personality disorders. Child and adolescent data suggest worsening of aggression with SSRI treatment. Fluoxetine is dosed at 20-80 mg/day and sertraline at 50-200 mg/day. Adverse effects include nausea, insomnia, tremor, agitation, sexual dysfunction, and anorexia. Citalopram is a new SSRI estimated to be released in the summer of 1998. One double-blind cross-over study in 19 violent schizophrenic patients studied the efficacy of citalopram 20-60 mg/d added to concurrent neuroleptic treatment for 24 weeks. Results from this study indicated a significant decrease in aggression with few adverse effects during concurrent citalopram treatment.7
Atypical Antipsychotics:
Clozapine:
The mechanism of action of clozapine in aggression is due to increased 5-HT and decreased DA. Clinical studies have shown a significant decrease in assaults, seclusion, and mechanical/chemical restraints with clozapine treatment. A dramatic decrease in aggression and agitation has been shown with clozapine independent of its effect on psychosis. Clozapine is initiated at 25-50 mg/day and then increased by 25-50 mg/day as tolerated up to 300-750 mg/day. Due to the risk of agranulocytosis, WBC must be monitored weekly. Orthostatic hypotension should also be monitored. Adverse effects associated with clozapine include sedation, orthostasis, agranulocytosis, hypersalivation, and weight gain. At this time, clozapine is reserved for the treatment of aggression in patients with co-morbid psychosis refractory to other antipsychotics.2
Risperidone:
Risperidone, in low doses (0.5-2.5 mg/day), appears to be useful for the management of pathologic aggression in demented patients.8
Serenics:
Eltoprazine is the main representative of a new class of 5-HT1A and 5-HT1B agonists known as serenics. These agents are not available in the United States but are currently being tested for the treatment of aggression. Serenics have been tested in animal models and have shown antiaggressive properties with very little sedation. Clinical studies have been conducted in patients with dementia, MRDD, psychosis, and personality disorders. However, results from these studies are mixed and at this time optimal efficacy of these agents appears to be in refractory patients.
References:
1-Stedman’s Medical Dictionary. 26th ed. Baltimore: Williams
and Wilkins, 1995:37.
2-Pabis DJ, Stanislav SW. Pharmacotherapy of Aggression. Ann of
Pharmacother 1996;30:278-87.
3-Diagnostic and Statistical Manual of Mental Disorders: DSM-IV.
4th ed. Washington, DC: American Psychiatric Association, 1994.
4-Kavoussi R, Armstead P, Coccaro E. The Neurobiology of Impulsive
Aggression. Psych Clin N Am 1997;20:395-403.
5-Citrome L, Volavka J. Psychopharmacology of Violence. Psych Ann
1997;2:696-703.
6-Stanislav SW, Fabre R, Crismon ML, Childs A. Buspirone’s Efficacy
in Organic-Induced Aggression. J Clin Psychopharm 1994;14:126-30.
7-Vartiainen H, Tiihonen J, Putkonen A, et al. Citalopram, A SSRI
in the Treatment of Aggression in Schizophrenia. Acta Psychiatr Scand.
1995;91:348-351.
8-Goldberg RJ, Goldberg J. Risperidone for Dementia-Related Disturbed
Behavior in Nursing Home Residents: a clinical experience. International
Psychoger., 1997;9:65-68.
Psychotherapy and Pharmacotherapy for Depression: United They Stand?
This exploratory study examined utilization and costs
among depressed patients in two treatment models--integrated treatment,
in which psychotherapy and pharmacotherapy were provided by a psychiatrist,
and split treatment, in which pharmacotherapy was provided by a psychiatrist
and psychotherapy by a nonphysician psychotherapist. A retrospective design
was used to compare claims data from a national managed mental health care
organization for 191 patients in integrated treatment and 1,326 in split
treatment. During the 18-month study, patients receiving integrated
treatment used significantly fewer outpatient sessions and had significantly
lower treatment costs, on average, than those in split treatment.
Integrated treatment appeared to be associated with a pattern of utilization
characterized by frequent treatment episodes in contrast to that of split
treatment, which was characterized by more sessions with fewer breaks of
90 days or more. The results do not support the prevailing assumption
that integrated treatment is more costly than split treatment in a managed
care network. Despite limitations in the study methods, the strength
of these preliminary findings poses a powerful challenge and invites further
study.
William Goldman, M.D., Joyce McCulloch, M.S., Brian
Cuffel, Ph.D., Deborah A. Zarin, M.D.,
Ana Suarez, M.P.H., Barbara J. Burns, Ph.D. (Psychiatric
Services 49: 477-482, 1998)
Risperidone versus Clozapine for Treatment-Resistant Schizophrenia
Studies have shown that both clozapine and risperidone
are effective in managing positive and negative symptoms of schizophrenia
and have few Extra-pyramidal side effects. Bondolfi et al. (p. 499) extended
these findings by studying the effects of the two drugs in 86 patients
with chronic schizophrenia who had not responded to treatment with conventional
neuroleptics or could not tolerate them. Both drugs significantly reduced
the severity of psychotic symptoms--67% of the group given risperidone
and 65% of the group given clozapine showed clinical improvement--although
risperidone appeared to have a faster onset of action. There were few adverse
effects in both groups, and those that did appear were mild.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 4, April 1998, page A70.
Euphoria, Dysphoria, and Cerebral Blood Flow
Evidence from several types of studies suggests that
effects of opioid drugs are mediated by different receptor subtypes. Schlaepfer
et al. (p. 470) used single photon emission computed tomography (SPECT)
to study of the effects of the mu-receptor agonist hydromorphone and the
mixed agonist/antagonist butorphanol (which has a kappa agonist component
of activity) in nine nondependent opioid abusers. They found that hydromorphone
produced more subjective "good effects" and an increase in cerebral blood
flow (CBF) in three distinct brain regions belonging to the limbic system,
while butorphanol led to more subjective "bad effects" and fewer increases
in regional CBF, which were mainly in the area of the temporal lobes. The
findings show that opioids with different subjective effects produce changes
in CBF in different brain regions and also that SPECT functional neuroimaging
can be used to study medications that have the potential for abuse.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 4, April 1998, page A70.
Alternative Care That Meets Patients More Than Half Way
Current efforts to control medical costs have moved
onto center stage the search for effective non-hospital alternatives for
severely ill psychiatric patients requiring acute care. Crisis residential
care is one such alternative. Crisis residential care is not a halfway
house, but a neighborhood-based home setting in which acutely ill patients
are treated as an alternative to being admitted to a general or psychiatric
hospital. In a study by Fenton et al. (p. 516), adult patients with severe
and persistent mental illness under the care of a public mental health
system were randomly assigned to the acute ward of a general hospital or
to a community residential home. Symptom reduction and patient satisfaction
were comparable for the two settings, as were psychosocial functioning
and use of acute care in the six months following admission. Overall, 87
percent of patients receiving alternative care were successfully treated
and returned to the community. For patients who do not require intensive
general medical intervention and are willing to accept voluntary treatment,
the alternative program model provides outcomes comparable to those of
hospital care, and at a lower cost.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 4, April 1998, page A70.
Men (and Women) Behaving Badly
Adolescents with conduct disorder who also have alcohol
and drug problems may persist in their antisocial behaviors in adulthood
and later receive a diagnosis of antisocial personality disorder. Myers
et al. (p. 479) did a longitudinal study of 137 adolescent inpatients with
conduct disorder at two alcohol and drug treatment facilities to determine
factors that might lead to the continuation of pathological behavior. When
they examined the subjects a second time, four years after treatment, 61%
of the group met the DSM-III-R criteria for antisocial personality disorder
and were still functioning poorly in many areas of life. Factors predicting
this progression to the adult disorder included onset of deviant behavior
at age 10 or earlier, greater diversity of deviant behavior, and more extensive
use of drugs before treatment. In contrast, those whose deviant behavior
occurred primarily in the context of drug or alcohol use were more likely
to cease their antisocial activities after being treated for substance
abuse. The authors conclude that careful assessment of conduct disorder
history at the time of substance abuse treatment may be valuable in planning
effective interventions for these adolescents.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 4, April 1998, page A70.
The Faith of the Elders
When asked what helped them cope with illness and
other depressing events in their lives, one-third of 87 medically ill hospitalized
older patients with depression spontaneously gave replies such as "God,"
"my faith," or "prayer." The subsequent study of these patients by Koenig
et al. (p. 536) yielded a similar finding: during the 47 weeks after hospital
discharge, the depressed patients with higher scores on intrinsic religiosity
had more rapid remissions than depressed patients with lower scores. Church
attendance and private religious activities, however, were not related
to time of remission. Intrinsic religious commitment appears to exert a
therapeutic effect in older patients with disabling physical health problems.
These beliefs may bring comfort and facilitate coping.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 4, April 1998, page A70.
Exercising Choice for Panic?
Despite a popular belief among the public that exercise
has anxiolytic and antidepressant properties, surprisingly few studies
have examined the therapeutic potential of exercise for psychiatric disorders.
To compare the effect of exercise on panic disorder with the effects of
placebo and a drug with proven efficacy, Broocks et al. (p. 603) randomly
assigned 46 outpatients with panic disorder to 10 weeks of regular running,
placebo, or clomipramine. Both exercise and clomipramine led to significant
decreases of symptoms according to all measures, although drug treatment
improved anxiety symptoms earlier and more effectively. Since regular aerobic
exercise alone is associated with significant clinical improvement, it
may be valuable for patients who are unable or unwilling to take medication.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 5, May 1998.
Who Recovers From Bipolar Disorder? It Depends on How You Ask
The course of bipolar disorder is unfavorable for
many patients. In determining predictors of recovery, it is important to
distinguish various aspects of recovery so that factors differentially
contributing to recovery can be clarified. Syndromic recovery is the resolution
of a constellation of symptoms such that diagnostic criteria are no longer
met. Symptomatic recovery refers to the magnitude of symptom improvement.
Functional recovery is the return to previous levels of work and psychosocial
function. Keck et al. (p. 646) compared the 1-year outcomes for patients
with manic and mixed episodes of bipolar disorder. There was no difference
in outcome between the manic and mixed types, and 48% of patients had syndromic
recoveries, but only 24% had functional recoveries. Predictors of syndromic
recovery included shorter duration of illness and full treatment compliance.
Symptomatic and functional recovery occurred more rapidly and more frequently
in patients from higher social classes.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 5, May 1998.
***
Looking for a performance measurement system that will meet the
needs of a psychiatric hospital? Check out the National Association of
State Mental Health Program Director’s Performance Measurement System at
the following Web-Site: http://dmhmrs.chr.stateky.us/nripms/
***
Long–Term Lithium Limits Morbidity
Evidence supporting the relative value of lithium
in the subtypes of the Bipolar-I and Bipolar-II syndromes and for bipolar
depression versus mania or hypo-mania remains sparse and inconclusive.
Tondo et al. (p. 638) studied patients with Bipolar-I (N=188) or Bipolar-II
(N=129) disorder before lithium treatment and during lithium maintenance
(average, six years) under naturalistic conditions. The type I/II distinction
was strongly supported, and the benefits of lithium were even greater for
Bipolar-II than Bipolar-I disorder. Lithium treatment yielded striking
long-term reductions of both depressive and manic symptoms in both subtypes,
particularly for patients whose lithium treatment began early, and markedly
reduced the need for rehospitalization. Lithium did not completely prevent
affective episodes, but it was remarkably effective in limiting long-term
morbidity.
Reprinted by permission of the American Psychiatric
Association: From "In this issue" of the American Journal of Psychiatry,
Vol. 155, No 5, May 1998.
Modafinil,
the First Drug Indicated for Narcolepsy
with a Lower Abuse Potential
Michelle A. Gravlin, Pharm.D.,
Psychopharmacy Fellow,
Western Missouri Mental Health Center.
Modafinil, a central (alpha-1)a1 post-synaptic agonist, was recently granted approvable status by the Food and Drug Administration (FDA) for the treatment of narcolepsy. It is expected to be available in mid-1998 to be marketed by Cephalon Incorporated. The mechanism of action for use in narcolepsy is unknown, however central a1-antagonists block its stimulant effects and exacerbate cataplexy. Modafinil also decreases central GABA which may play a role in its activity in narcolepsy.1
Two phase-III trials were completed in the United States for approval. Both studies were placebo-controlled, double-blind, randomized parallel-group in their design. The randomization was 1:1:1 modafinil (MOD) 400 mg, modafinil 200 mg or placebo (PBO). The maintenance of wakefulness test (MWT), a sensitive test to medication response, was used to assess efficacy. Subjects sit in a comfortable chair and are told to try and stay awake for five 20 minute time periods spaced two hours apart while hooked to polysomnography equipment. The sleep latency during these times is the result of the MWT. MWT results found greater sleep latency for both modafinil groups in both trials at a p< 0.0001 level. The clinical global impression of change was greater in both modafinil groups as well compared to placebo (MOD 400 - 61%, 74%, MOD 200 58%, 65%; PBO 38%, 37) (p< 0.05). 2, 3 Although no head-to-head comparisons of modafinil and stimulants are known to be published, modafinil is thought to be slightly less effective.4 Adverse effects included headache, nervousness, and nausea. Modafinil offers benefits in its side effect profile and abuse potential, hopefully offering more convenience to the patient and prescriber with less prescribing controls.
Modafinil is thought to have a lower abuse potential based on animal studies and at least one human trial in substance abusers possibly due to its minimal effects on dopamine. Twenty-four male substance abusers received single doses of methylphenidate 45 and 90 mg, modafinil 200, 400 and 800 mg and placebo. Assessments include the Morphine Benzedrine Group (MBG) scale, the Amphetamine Scale, subjective reports, vital sign, appetite, and sleep measures. No amphetamine scale effects or euphoriant effects on the MBG scale were found with modafinil, although subjects could discriminate between placebo and modafinil. Modafinil did not suppress appetite and affected vital signs less than methylphenidate.5 At this time, modafinil has not been studied in disorders beyond narcolepsy and sleep apnea.
Modafinil is a welcome alternative for those suffering from narcolepsy, an orphan disease affecting less than 0.2% of the population. All currently FDA approved treatments for narcolepsy are Schedule-II controlled substances requiring new written prescriptions monthly with no refills or telephone prescriptions. One recent study found that only 41% of narcolepsy patients take the prescribed amount of Schedule-II stimulants while 88% of those prescribed pemoline, a Schedule-IV agent, are compliant.6 The likelihood of modafinil being a lower schedule allows for easier access and hopefully improved compliance with treatment.
References:
1-Moachon G. Kanmacher I, Clenet M, Matinier D. Pharmacokinetic
profile or modafinil. Drugs Today 1996;32(4)327-37.
2-Richter RW. A clinical trial of modafinil in patients
with narcolepsy: safety and improvement in illness [abstract]. Abstracts,
Posters and Presentations. NCDEU 37th Annual Meeting, Boca Raton,
FL: NCDEU, 1997. NIMH, 1997.
3-Lamberg L. Narcolepsy researchers barking up the
right tree. JAMA 1996;276(10):765-6.
4-Mitler MM, Aldrich MS, Koob GF, Zarcone VP. ASDA
standards of practice: narcolepsy and its treatment with stimulants. Sleep
1994; 17(4):352-71.
5-Jasinski DR. An evaluation of the abuse potential
of modafinil using methylphenidate as a reference [abstract]. Abstracts,
Posters and an Presentations. NCDEU 37th Annual Meeting, Boca
Raton, FL: NCDEU, 1997. NIMH, 1997.
6-Rogers AE, Aldrich MS, Berrios AM, Rosenberg RS.
Daytime sleepiness: Compliance with stimulant medications in patients with
narcolepsy. Sleep 1997;20(1):28-33.