European Neuropsychopharmacology 8 (1998) 5566

Guidelines for depot antipsychotic treatment in schizophrenia

a
b
c
d
John M. Kane , Eugenio Aguglia , A. Carlo Altamura , Jose Luis Ayuso Gutierrez ,
e ,n ,
f
g
h
Nicoletta Brunello *, W. Wolfgang Fleischhacker , Wolfang Gaebel , Jes Gerlach , Julieni
j
k
l , Julien Mendlewicz m ,
D. Guelfi , Werner Kissling , Yvon D. Lapierre , Eva Lindstrom
n
o
Giorgio Racagni , Luis Salvador Carulla , Nina R. Schooler p
a

Department of Psychiatry, Hillside Hospital, Division of Long Island Jewish Medical Center, Glen Oaks, NY 11004, USA
b
Institute of Psychiatric Clinic, University of Trieste, Via S. Cilino 16, 34126 Trieste, Italy
c
Institute of Psychiatric Clinic, Via Liguria, 13, Cagliari, Italy
d
Psiquiatria, Alcale 152, 28028 Madrid, Spain
e
Dept. Pharmaceutical Sciences, University of Modena, Via Campi 183, 41100 Modena, Italy
f
Dept. of Biological Psychiatry, Innsbruck Univ. Clinics, Anichstrasse 35, A-6020 Innsbruck, Austria
g
Rheinische Landesklinik, Psychiatrische Klinik der Heinrich-Heine-Univ. Dusseldorf, Bergische Landestrasse 2, 4000 Dusseldorf, Germany
h
SCT Hans Hospital, Dept. 2, 4000 Roskilde, Denmark
i

Hopital
Paul Brousse, 12 Av. P.V. Couturier, 94804 Villejuif, France
j
Munchen, Ismaningerstrasse 22, 81675 Munchen, Germany
Dept of Psychiatry, Technische Universitat
k
Dept of Psychiatry, Institute of Mental Health Research, University of Ottawa, 1145 Carling, Ottawa, Ontario, Canada KIZ 7 K4
l
Department of Psychiatry, University Hospital, S-751 85 Uppsala, Sweden
m
Dept. of Psychiatry, University of Brussels, Erasme Hospital, 808 Rue de Lennik, 1070 Brussels, Belgium
n
Center of Neuropharmacology, Institute of Pharmacological Sciences, University of Milan, Via Balzaretti 9, 20133 Milan, Italy
o
Urb. El Bosque E /Cycas 78 C, 11405 Jerez de la Frontera, Cadiz, Spain
p
Western Psychiatric Inst. and Clinic, 3811 O Hara St., Pittsburgh, PA 15213, USA
Received 1 December 1996; received in revised form 30 April 1997; accepted 16 May 1997

Abstract
These guidelines for depot antipsychotic treatment in schizophrenia were developed during a two-day consensus conference held on
July 29 and 30, 1995 in Siena, Italy.
Depot antipsychotic medications were developed in the 1960s as an attempt to improve the long-term treatment of schizophrenia (and
potentially other disorders benefiting from long-term antipsychotic medication). Depot drugs as distinguishable from shorter acting
intramuscularly administered agents can provide a therapeutic concentration of at least a seven day duration in one parenteral dose.
The prevention of relapse in schizophrenia remains an enormous public health challenge worldwide and improvements in this area can
have tremendous impact on morbidity, mortality and quality of life, as well as direct and indirect health care costs. Though there has been
debate as to what extent depot (long-acting injectable) antipsychotics are associated with significantly fewer relapses and rehospitalizations, in our view when all of the data from individual trials and metaanalyses are taken together, the findings are extremely compelling in
favor of depot drugs. However in many countries throughout the world fewer than 20% of individuals with schizophrenia receive these
medications.
The major advantage of depot antipsychotics over oral medication is facilitation of compliance in medication taking. Non-compliance is
very common among patients with schizophrenia and is a frequent cause of relapse. In terms of adverse effects, there are not convincing
data that depot drugs are associated with a significantly higher incidence of adverse effects than oral drugs. Therefore in our opinion any
patient for whom long-term antipsychotic treatment is indicated should be considered for depot drugs.
In choosing which drug the clinician should consider previous experience, personal patient preference, patients history of response
(both therapeutic and adverse effects) and pharmacokinetic properties.
In conclusion the use of depot antipsychotics has important advantages in facilitating relapse prevention. Certainly pharmacotherapy
*Corresponding author. Tel: 139 2 20488331; Fax: 139 2 29403673;
e-mail: brunello@imiucca.csi.unimi.it
0924-977X / 98 / $19.00 1998 Elsevier Science B.V. / ECNP. All rights reserved.
PII S0924-977X( 97 )00045-X

56

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

must be combined with other treatment modalities as needed, but the consistent administration of the former is often what enables the
latter. 1998 Elsevier Science B.V. / ECNP.
Keywords: Depot antipsychotics; Schizophrenia; Treatment; Relapse

1. Introduction
These guidelines were developed during a two-day
consensus conference held on July 29 and 30, 1995 in
Siena, Italy. The need for this effort was based on the
recognition that the prevention of relapse in schizophrenia
remains an enormous public health challenge world wide
and that improvements in this area can have tremendous
impact on morbidity, mortality and quality of life, as well
as direct and indirect health care costs. Despite overwhelming evidence that non-compliance in medicationtaking is a major contributing factor to unnecessarily high
relapse rates, there is still inadequate attention being paid
to strategies which can enhance medication acceptance and
adherence, ranging from better psychoeducation directed at
patients and families to the more extensive use of depot
antipsychotic administration. Estimates suggest that in
many countries throughout the world fewer than 20% of
individuals with schizophrenia receive long-acting injectable medication. It is hoped that the promulgation of clear
and concise guidelines for the use of depot drugs will help
to remedy one important aspect of this problem.

1.1. Schizophrenia and the need for long-term treatment

Schizophrenia is a chronic illness usually beginning in
late adolescence or early adulthood. The condition is
characterized by remissions and exacerbations, though a
proportion of patients remain persistently ill. There is
evidence that after 10 to 20 years some patients may
improve in terms of their overall level of psychopathology
and community adjustment. The disease affects 1% of
most populations of the Western World, but consumes a
disproportionate share of health care costs. A large number
of persons with schizophrenia are permanently disabled,
and in many countries, homeless.
The risk of suicide may be as high as 1 in 10,
particularly in the early years after illness onset and among
males (Miles, 1977; Drake et al., 1984). Mortality is also
higher due to accidental deaths and other causes (Bland et
al., 1976). Patients suffering from this illness often receive
sub-optimal general medical care and frequently have
undiagnosed comorbid medical conditions.
The treatment of schizophrenia requires an integration of
biologic, psychologic and psychosocial perspectives. There
is increasing evidence that early diagnosis and appropriate
treatment can improve long-term outcome (May et al.,
1981). Antipsychotic drugs are a critical modality in
managing this disease in all phases acute, stabilization

and relapse prevention. These drugs cannot only alleviate

or improve psychopathology, but may also enhance psychosocial and vocational adjustment and improve subjective well being. Although medications can be highly
effective, response varies and some patients derive considerably less benefit than others. Despite heterogeneity in
drug responsiveness, antipsychotic drugs are indicated for
all patients with schizophrenia.
Long-term treatment with medication is critical in
optimizing outcome and is the focus of these guidelines.

1.2. Definition and measurement of relapse

The participants defined relapse as the appearance,
reappearance or exacerbation of symptoms (typically psychotic) of schizophrenia which may require a change in
clinical care.
When a relapse is observed, clinicians should make a
differential diagnosis and assess possible contributing
factors, e.g. natural course of the illness, non-compliance,
underdosage (or drug discontinuation), stress, comorbid
conditions, drug abuse, medical illness, adverse effects,
etc.
Alternative clinical interventions should be considered
such as increased surveillance or intervention of a psychotherapeutic / psychosocial or pharmacotherapeutic nature
(e.g. reinstitute drug treatment, increase dosage, prescribe
adjunctive pharmacotherapy, change antipsychotic).
Given the fluctuating course of this illness, an important
aspect of treatment focuses on the maintenance of therapeutic gains and the prevention of clinical exacerbation,
relapse and rehospitalizations. This is a concern not only
because of the immediate personal and psychosocial
disruption, but also because frequent relapses can increase
the likelihood of poorer long-term outcome.
The definition of relapse has been an important variable
in studies of long-term treatment in schizophrenia. The
basis on which we strongly recommend continued antipsychotic drugs is the significant reduction in relapse rate
(despite variability in definition) across numerous studies
with treatment as compared to untreated cases or placebotreated controls. The manner in which relapse is defined,
however, takes on critical importance in understanding the
clinical implications of those findings as well as making
comparisons across studies. The efficacy of specific treatment strategies may vary depending upon what definitions
of relapse are applied.
Gilbert et al. (1995) recently reviewed 66 studies
involving neuroleptic withdrawal. They found that 22

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

studies did not provide any definition of relapse. In 11

studies relapse was defined as a return to active medication. The remaining 33 studies defined relapse as either
the emergence of behavioral worsening (with agitation,
aggression, insomnia, anxiety, hallucinations, delusions, or
assaultive or suicidal behavior). Some of these investigations utilized a specified change seen on particular
items on a clinical rating scale such as the Brief Psychiatric Rating Scale (BPRS). In one large scale study
(Schooler et al., 1995), psychotic relapse was defined by a
rating of moderate or greater representing an increase of
at least two scale points on any of five psychotic items of
the BPRS (conceptual disorganization, grandiosity, suspiciousness, hallucinatory behavior and unusual thought
content). This increase in psychotic symptoms had to
persist for two successive scheduled ratings separated by
four weeks or a scheduled rating and an unscheduled rating
associated with the initiation of open (non-blind) active
medication.
Similar criteria were employed by Kane et al. (1983) in
a previous study. Marder et al. (Marder et al., 1984, 1987)
defined three levels of unfavorable outcome that would
lead to an antipsychotic dosage increase. When patients
had an increase of 3 or more points on the BPRS cluster
scores for thought disturbance or paranoia they were
considered to have had a psychotic exacerbation. These
exacerbations were relatively mild and seldom led to
rehospitalization. Clinicians were allowed to essentially
increase the dosage by up to 100%. If symptoms could not
be adequately controlled within this range, they were
considered to have had a relapse. The third level of
outcome was rehospitalization. Those criteria for relapse,
therefore, not only involved an objective measure of
worsening psychopathology, but also failure to respond to
a specified clinical intervention.
As can be seen from these examples in defining relapse,
a number of key factors need to be considered:

1. Absolute degree of increase in psychopathology

2. Nature of psychopathology increasing (i.e. psychotic or
non-psychotic)
3. Degree of increase in psychopathology relative to the
baseline state of the patient
4. Duration of the exacerbation
5. Response of the exacerbation to treatment intervention
(which may be pharmacologic and / or non pharmacologic)
The critical question in attempting to define relapse is the
desired balance between specificity and sensitivity. This
judgment will in turn be influenced by the relative risk
associated with acting on the basis of a false positive and
not acting on the basis of a false negative. The potential
consequences of a relapse for that given individual based
on history, current life situation, etc., must be factored into

57

the ultimate clinical judgment. Though at present we are

not aware of significant risks associated with treatment of a
false positive relapse, there is some reason to be
concerned about unnecessary increases in antipsychotic
drug dosage in relation to the development of tardive
dyskinesia (Kane, 1995).
Definitions of relapse which can be used by clinicians in
routine practice will never be a substitute for experienced
clinical judgment, but can provide a useful frame of
reference for organizing and systematizing the decision
making process.

1.3. Benefits and risks of neuroleptic maintenance

treatment
Several extensive reviews have appeared in recent years
summarizing the data on the impact of continued antipsychotic medication on rates of relapse in schizophrenia
(Davis et al., 1989; Gilbert et al., 1995). There is overwhelming evidence that the use of medication can have a
significant (clinical and statistical) benefit in improving
outcome. The consequences of relapse are diverse and
often unpredictable ranging from loss of confidence and
self-esteem, disruption in psychosocial and vocational
adjustment and family burden to risk of suicide or aggressive behavior. There is no question that relapse is associated with substantial increase in the costs associated with
the illness (both direct and indirect). In addition, there is
some suggestion that with each subsequent episode time to
recovery and degree of recovery are not as good previously. It is possible that this reflects the natural course of the
disease as well, but prevention of relapse is a goal which
may have long-term impact on the ultimate course of the
disease.
The risks associated with long-term neuroleptic treatment are largely those related to a variety of adverse
reactions, particularly neurologic effects such as tardive
dyskinesia or tardive dystonia. Other side effects such as
drug-induced parkinsonism, akathisia, weight gain and
sedation can also pose problems to some patients.
Those adverse reactions which are of most concern in
terms of potential-seriousness and persistence are the
abnormal involuntary movement disorders associated with
long-term neuroleptic treatment. Although prevalence estimates vary widely, on average 1520% of patients
chronically-treated with neuroleptic medication manifest
some degree of tardive dyskinesia (Kane et al., 1992).
Incidence studies (Kane, 1995; Glazer and Kane, 1992)
suggest that approximately 5% of young to middle-aged
adult patients develop some evidence of abnormal involuntary movements with each year of neuroleptic treatment.
The majority of these cases are mild and nonprogressive
and a substantial proportion can in fact improve or remit
entirely if neuroleptic dosage is reduced (Kane et al.,
1992) or they are switched to a drug such as clozapine
(Lieberman et al., 1991).

58

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

In the case of elderly patients, the incidence rate (at least

in the first year or two of neuroleptic treatment) appears to
be six-fold higher than among younger adults (Saltz et al.,
1991; Jeste and Caligiuri, 1993).
As the epidemiology of tardive dyskinesia became better
understood, an extensive reevaluation of the benefit-to-risk
ratio took place and a new generation of maintenance
treatment studies were initiated to determine if alternatives
to continuous neuroleptic treatment had any substantive
role to play in the long-term treatment of schizophrenia.

be improved. At present, however, continuous medication

prophylaxis is the treatment of choice and targeted treatment has little to recommend it, except perhaps in the rare
situation where continuous treatment is contraindicated or
the patient adamantly refuses maintenance antipsychotic
drug treatment. Even when tardive dyskinesia develops, it
is not at all clear that intermittent treatment is preferable to
continuous (preferably low dose) treatment.

1.4. Are there alternatives to neuroleptic maintenance

treatment?

Although non-compliance is very common among patients with schizophrenia, detection is often difficult. In
addition, though non-compliance is a frequent cause of
relapse, not all relapses are the consequence of noncompliance. Even full medication compliance, as insured
by the regular administration of depot neuroleptics is not a
guarantee of a relapse free state. As reviewed by Schooler
(1985), a series of studies that compared depot and oral
neuroleptic administration found relapse rates on depot
drugs that ranged between 8 and 40% during 10 to 24
months. These figures suggest that relapse cannot be taken
as a reliable marker for non-compliance because relapse
occurs in patients for whom compliance can be documented by depot medication administration. Depot neuroleptics do not completely prevent non-compliance. What
depot administration insures is that patients have received
medication and that non-compliance, defined as failure to
return for a scheduled injection, is immediately detected.
With oral medication, failure to take medication may not
be detected until some time after it has first occurred and
often not until the resulting relapse.
Another problem with the use of relapse to define
non-compliance is that even if schizophrenia patients are
non-compliant, relapse may not occur for some time after
medication has been discontinued. Hogarty et al. (1974)
reported a cumulative relapse rate of 80% on placebo over
24 months that is linearabout 3% a month. McEvoy et al.
(1984) documented the differences between relapsers who
are compliant with medication and those who are not.
Non-compliant relapsers have more classic psychotic
symptoms of schizophrenia and show a gradual onset of
symptoms as would be expected from data regarding
placebo relapse. Compliant patients show a more rapid
onset of symptoms and prominent affective symptoms.
Steingard et al. (1994) found that compliant patients who
relapsed improved during hospitalization regardless of
whether they received additional medication or not.
All of these factors relapse among compliant patients,
the variable time course of relapse among non-compliant
patients, the potential differences in the nature of relapse
for compliant and non-compliant patients and the fact that
response in compliant relapsers may not be a function of
medication changes makes the use of relapse as the sole
marker for non-compliance unwise although it is clearly
the outcome variable of greatest interest. The compliance

As a potential alternative to continuous maintenance

treatment, a strategy of intermittent or early intervention, time-limited targeted pharmacotherapy has been
investigated by several groups in the US and Europe. This
strategy is an outgrowth of observations that relapses do
not occur immediately, even following complete neuroleptic drug discontinuation, and that many patients can
maintain remission for many months without medication
(Carpenter et al., 1982; Carpenter and Heinrichs, 1983). In
addition, Herz and Melville (1980) observed that psychotic
relapses are often preceded by a variety of prodromal
symptoms the identification of which may be used to
facilitate early intervention.
This strategy has been investigated in a number of
large-scale studies in the US, England and Germany (Herz
et al., 1982, 1991; Carpenter et al., 1987, 1990; Jolley et
al., 1989, 1990; Pietzcker et al., 1986, 1993; Schooler et
al., 1993).
Five of these studies involved a comparison of targeted
treatment with continuous treatment over a two-year
period. In all of these investigations, the targeted treatment
was associated with significantly less cumulative drug
exposure over the course of the trial, however, only some
of the trials found advantages in terms of adverse effects.
On the other hand, the rate of relapse was significantly
higher in the targeted treatment groups than in the continuously treated groups during the second year of treatment in
all of the studies and in the first year of treatment in four
out of the five. There were no counterbalancing advantages
associated with the intermittent treatment such as reduced
rates of tardive dyskinesia or overall improvements in
psychosocial adjustment.
Only the Pietzcker et al. (1993) study investigated
whether apparent prodromal symptoms are actually valid
predictors of subsequent relapse. In terms of relapse and
rehospitalization rates, this study was similar to the others
in demonstrating clear superiority for continuous treatment, but also found no significant relationship between
prodromal symptoms and relapse (Gaebel et al., 1993).
Perhaps if more valid predictors of relapse (either clinical
and / or biologic) could be developed (Gaebel and Awad,
1994), the results of intermittent / targeted treatment could

1.5. Compliance: Assessment and determinants

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

literature in schizophrenia is complicated by the confusion

of independent variables that may predict compliance such
as attitudes towards medication, demographic characteristics, experience of side effects and psychopathological
symptoms and the dependent variables that define medication taking behavior such as pill counts, plasma and urine
levels, self report, etc. Finally there is actual treatment
outcome which is determined by multiple factors, those
mediated by compliance and others as well.

1.5.1. Definitions of compliance

It is generally agreed that compliance is not an all or
nothing phenomenon. Schizophrenia patients, like those
with other medical illnesses may vary in their compliance
from taking all medication as prescribed to complete noncompliance. Many studies of compliance do not include
precise operational definitions. A recent study by
Radomsky (1995) represents a notable exception. She used
plasma determinations of haloperidol and defined compliance with oral haloperidol by a combination of the
absolute plasma concentration and variability over multiple
assays
1.5.2. Measures of compliance
These can be classified as Direct and Indirect (Sleator,
1985). Direct methods include measures such as plasma or
urinary measures of the parent drug or its metabolites
(Perel, 1988), the use of markers such as riboflavin and for
phenothiazines the very old Forrest Test (Forrest et al.,
1961). A problem with direct measures is that they assume
that a level of medication taking behavior is required to
define compliance. Particularly in schizophrenia it is
unclear how high a level of compliance with prescribed
medication is needed. Since there is a fairly wide range
over which most antipsychotic medications are effective,
skipped doses or taking less than prescribed may represent
non-compliance but may not have any effect on long-term
outcome.
Indirect measures include relapse, self report and clinician judgment. Young et al. (1986) reviewed 23 studies of
compliance in schizophrenia. Six of these studies used
direct methods, 17 used interviews or staff reports and only
two included both direct and indirect methods.
Predictors of non-compliance are classified as standard and subjective risk factors by Weiden et al.
(1994). Standard risk factors include such variables as
complexity of the medication regime, structure provided to
insure / enhance medication taking and symptoms of psychopathology.
Subjective risk factors include perceptions and attitudes
such as lack of insight, denial of illness, family beliefs and
beliefs regarding the value of medication. There has been a
substantial and growing interest in this area. Weiden et al.
(1994) describe three instruments that look at subjective
factors: the Drug Attitude Inventory (Hogan et al., 1983);
the Neuroleptic Dysphoria Scale (vanPutten and May,

59

1978); and the Rating of Medication Influences (Weiden et

al., 1994). Radomsky (1995) developed a questionnaire
that assesses both attitudes toward medication and medication taking behaviors. Radomskys study highlights an
important problem. She found that when a direct measure
of compliance (plasma concentrations of haloperidol) was
used there was very little relationship between attitude and
actual compliance. Further, patients substantially overestimated their compliance as defined by actual medication
taking behavior.
As indicated above, because measures of attitudes
toward medication-taking and medication-taking behaviors
are often included in the same assessment instrument, it is
sometimes difficult to distinguish the predictors of compliance (attitudes, beliefs and personal predictions of
medication taking) from the outcome-self report of actual
medication ingested. The patients self-reported attitudes
are also likely to influence the clinicians judgments of
medication taking. Patients who complain about medication may be seen as non-compliant even when they are.
A model of compliance and treatment outcome in
schizophrenia should include the following variables:

1. Subjective predictors: personal attitudes and beliefs,

family / significant other attitudes and beliefs; perceived
side effects.
2. Objective predictors: symptoms of schizophrenia (particularly delusions regarding medication and insight
into presence of illness, cognitive problems); side
effects; environmental supports or deficits (someone to
monitor medication taking, forgetting).
3. Medication taking behavior: for patients receiving depot
preparations this variable is highly reliable; for patients
taking oral medications its measurement depends on
self report, reports of others (nurses, care-givers, family
members) and pill counts.
4. Direct measures of medication concentration in blood
or urine or use of biological markers.
5. Measures of treatment outcome, including relapse but
also measures of psychopathology, subjective distress
and side effects.

1.6. Quality of life

Quality of life issues are receiving increasing attention
in health care. Patients suffering from schizophrenic
disorders struggle with numerous quality of life issues
ranging from subjective comfort with pharmacotherapy
(mainly influenced by medication side effects and general
attitudes toward taking drugs) (Awad and Hogan, 1994;
Lauer and Stegmuller-Koenemund, 1994) to problems
around housing and psychosocial adaptation (Meise et al.,
1994). It is becoming increasingly accepted that improvement of psychopathology is not the only relevant outcome

60

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

criterion and other variables related to changes in quality

of life (Collins et al., 1991) must be assessed. Although
pharmacologic studies are just beginning to adequately
address these problems, there is reason to believe that a
significant reduction in rates of relapse and rehospitalization (Pikney et al., 1991) has considerable impact on
quality of life. The potential advantages of one treatment
strategy over another (e.g. depot vs. oral antipsychotic)
need to be demonstrated in controlled clinical studies in
which these issues are given relevant attention. However,
while we eagerly await such studies, it is important that we
consider a wide range of quality of life issues in our
ongoing treatment decisions and evaluations (Angst et al.,
1994; Awad, 1992).
There are no universally accepted definitions of quality
of life. In general, the concept includes the individuals
perception of their own life within the framework of the
values of their culture and their own standards, goals and
interests. Quality of life assessments should probably
include four basic factors: psychosocial; somatic; functional and social. A number of instruments have been
utilized to assess quality of life and measure changes over
time. However, there is still controversy as to how best to
define and measure this domain. There is no question that
in the case of a schizophrenic illness numerous factors
besides the specific nature and severity of the diseaserelated psychopathology will influence overall quality of
life. A partial list would include personality variables,
family factors, environmental and cultural influences and
conditions. Specific attention has been given to family
factors such as expressed emotion (Brown et al., 1972),
but a family environment where resignation dominates can
also be deleterious (Schulze-Monking and Buchkremer,
1995). These variables might influence quality of life and
also have an effect on medication compliance or rates of
relapse (Brown et al., 1972; Kreisman et al., 1988).
Adverse effects of medication can also impact on quality
of life and covert non-compliance or overt refusal of
medication frequently results from such effects (e.g.
akathisia, sexual dysfunction, weight gain). In this context,
there are a number of levels on which quality of life issues
can have considerable importance. It is appropriate that
these issues are beginning to receive the attention that they
deserve.

1.7. Poor responders

As discussed previously, there is a substantial subgroup
of patients who experience psychotic relapse despite
definite adherence to a regimen of prophylactic antipsychotic medication. Results from a series of controlled trials
involving depot medication (Kane, 1996) indicate that on
average 15% of such patients will relapse within one year.
The advantage of having such patients on a depot drug is
that non-compliance can be ruled out as a causative factor

and other issues can receive appropriate attention (e.g.

reduction of environmental stress, substance abuse, comorbid conditions, etc.). As discussed previously (Steingard et
al., 1994), there is evidence that increasing antipsychotic
drug dosage in such patients may not be necessary and
having such patients on depot drugs may, therefore, result
in avoidance of unnecessary dosage increases.
Though we may not consider patients who break
through maintenance antipsychotic medication non-responders in the traditional sense, they do experience a type
of non-response. The other category of non-responders
includes those patients who fail to derive sufficient benefit
from an acute trial of an antipsychotic medication.
These patients frequently undergo dosage increases or
are switched to a different class of conventional antipsychotic drug, though there is limited evidence that either of
these strategies are associated with substantial improvement in the majority of such patients (Kinon et al., 1993;
Shalev et al., 1993). It has been suggested that some
patients are poor responders because of inadequate absorption and / or extensive first pass metabolism. Depot drug
administration largely overcomes this problem and one
study (Adamson et al., 1973) of chronically institutionalized patients with relatively low blood levels following an
oral test dose showed superior efficacy after depot treatment in comparison to those receiving oral treatment in a
28 day, double-blind design. Therefore, it is reasonable to
suggest a trial of depot drugs in some patients who are
poorly or only partially responsive to oral antipsychotic
medications.

1.8. Cost of schizophrenia treatment

Schizophrenia is a costly disease due to its early onset
and chronicity, the high utilization of resources used for
care, the lost productivity and the enormous burden
imposed on patients and relatives. Studies both in US and
Europe show that schizophrenia costs comprise a significant proportion of the costs of all mental disorders (Rice
and Miller, 1996; Smith et al., 1995). In addition, most of
the total life costs are confined to the group with poorer
outcome (Davies and Drummond, 1994). This group
shows the highest rates of intensive community treatment
and hospitalization, which accounts for nearly fifty percent
of all the direct costs attributed to schizophrenia. The
economic analysis of antipsychotic drugs is a field of
growing interest, particularly on such high-cost patients,
who may benefit more from some of the serotonindopamine antagonists (e.g. clozapine) (Frankenburg and Hegarty, 1996), and depot antipsychotics which may increase
compliance and reduce relapse rates. A recent modeling
study calculates the net treatment saving of switching from
oral to depot formulations ten percent of all patients with
schizophrenia living in the community in UK (Hale and
Wood, 1996).

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

2. Depot antipsychotics
What are depot drugs? So-called depot antipsychotic
medications were developed in the 1960s as an attempt to
improve the long-term treatment of schizophrenia (and
potentially other disorders benefiting from long-term antipsychotic medication). Depot drugs as distinguishable from
shorter acting intramuscularly administered agents can
provide a therapeutic concentration of at least a seven day
duration in one parenteral dose. The depot antipsychotics
are esters formed between the alcohol group of the drug
and a long chain of fatty acids (enantic, decanoic, palmitic,
and undecylenic acid), and this ester is dissolved in oil
(sesame oil or Viscoleo). Antipsychotics in themselves are
fat soluble, but the esterification increases this fat solubility a thousand fold, the longer the carbon chain, the
more solubility is increased.
The ester is released from the oil phase by slow
diffusion into the body water phase. The higher the fat
solubility, the slower the release. Upon release from the
oil, the ester is either immediately hydrolysed or distributed to other tissues and then hydrolysed. The goal is to
obtain a sufficiently constant delivery of the drug from the
depot, so that the serum level is kept as constant as
possible between injections.
The type of esters and oil vehicles of available depot
antipsychotics is seen in Table 1.

2.1. Pharmacokinetics
Table 1 (Barnes and Curson, 1994) provides a summary

61

of dosage and pharmacokinetic parameters of depot drugs.

Although estimates of dosage, equivalence are provided it
is important to recognize that these are estimates and not
necessarily based on comparative trials. It is also possible
that dosage equivalencies at the low end of the dosage
spectrum may not necessarily be the same at the higher
end of the dosage spectrum, and that even for wellestablished widely used medications, systematic information on dose response relationships is often lacking.
In addition, plasma concentrations resulting from depot
drugs are often near the limits of sensitivity for some
analytic methods making pharmacokinetic studies more
difficult. Even when investigations have been carried out
with oral medications examining the relationship between
therapeutic effect and plasma levels, the results are far
from consistent. In the case of depot drugs, far less
information is available. As Barnes and Curson (1994)
point out, all of the pharmacokinetic data available in the
literature is based on depot injections given into striatal
muscle. Much less is known about the pharmacokinetic
properties of subcutaneous injections. Given the fact that
the depth of gluteal fat is often more than 3.5 cm, it is
likely that many injections are not truly intramuscular.
After a depot drug is injected, it is slowly released from
the site. The initial rate limiting kinetic step is diffusion of
free drug resulting from rapid enzymatic hydrolysis from
the ester. As a result the rate of elimination is controlled by
this release rate rather than the rate of hepatic metabolism
(Ereshefsky et al., 1984). Because absorption (release) is
generally slower than drug elimination, the time to reach
steady-state is a function of the absorption rate while the

Table 1
Dosage and pharmacokinetic parameters of depot antipsychotic drugs
Preparation

Antipsychotic
group

Vehicle

Equivalent
doses a (mg)

Dose q2w
equivalent to
chlorpromazine
100 mg per day b

Test dose

Recommended
dose (mg)

Maximum
dose (mg)

t max

t ]1 c

Fluphenazine
decanoate d
Flupenthixol
decaonate e
Zuclopenthixol
decaonate
Pipothiazine
palmitate
Haloperidol
decaonate
Fluspirilene

Phenothiazine
(piperazine ring)
Thioxanthene

Sesame oil

25 q2w

10 mg

12.5100 q25w

See dose range

824 h

14 d

Viscoleo

40 q2w

16 mg

12.5 (6.25 in
elderly)
20

20300 q24w

400 qlw

37 d

17 d

Thioxanthene

Viscoleo

200 q2w

80 mg

100

200400 q24w

600 qlw

47 d

19 d

Phenothiazine
(piperidine ring)
Butyrophenone

Coconut oil

50 q4w

20 mg

25

50100 q4w

200 q4w

1224 h

1516 d

Sesame oil

100 q4w

30 mg

50

5300 q4w

See dose range

39 d

21 d

Phenybutylpiperidine

Aqueous
suspension

NA

NA

28 q1w

20 q1w

2472 h

over 72 h

British National Formulary (1993) recommendation.

Schulz et al., 1989.
c
Multiple doses. Elimination half-life relates to the rate of release from oil depot and not to elimination of the drug.
d
Fluphenazine enanthate and perphenazine enanthate have been discontinued.
e
Flupenthixol palmitate and pipothiazine undecanoate are available in some countries but used less commonly.
Abbreviations: d5days; h5hours; qxw5every x weeks; t ]1 5plasma elimination half-life (with multiple doses); t max 5time to peak plasma concentration;
2
NA5not available.
From Barnes and Curson, 1994 (with permission).
b

62

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

plasma concentration once steady-state is achieved is a

function of the elimination rate. For example, the time to
reach peak concentrations following a single injection of
fluphenazine decanoate is 0.3 to 1.5 days, for haloperidol
decanoate 3 to 9 days and for flupenthixol decanoate 3 to 7
days. After multiple injections during several months, the
elimination half-life of fluphenazine decanoate is approximately 14 days, haloperidol decanoate 21 days and flupenthixol decanoate 17 days.
Since it can take 23 months to achieve steady-state, it
might be necessary in some patients (depending upon prior
dosage, clinical state, initial dose of the depot preparation,
etc.) to supplement temporarily with oral medication
(Altamura et al., 1990a). Even though there is a better
correlation between dosage administered and plasma levels
achieved following depot as compared to oral antipsychotics, there is still wide individual variability in
plasma levels achieved after a particular dose of depot
drug. Therefore, it is not possible to provide well-validated
guidelines for oral supplementation requirements.
Ereshefsky et al. (1990) have recommended a loading
dose strategy with depot drugs which can reduce considerably the time required to achieve steady state, however, this approach requires relatively high initial doses
and has not been widely studied or utilized.
The effects of renal and hepatic disease on pharmacokinetics of depot drugs have not been thoroughly
studied. Decreasing renal function is not likely to have
significant effect since clearance is largely dependent on
hepatic function. With mild to moderate hepatic dysfunction, little dosage adjustment of depot drugs should be
necessary given the slow release of the drug from the
injection site. Substantial hepatic impairment, however,
may require dosage alteration and may also alter protein
binding complicating matters further.
It has been recommended that dosage reductions of
approximately 50% be considered in the elderly when
either oral or depot drugs are employed (Altamura et al.,
1990b) Ereshefsky et al. (1985) have suggested that
smoking can increase fluphenazine decanoate clearance by
two fold, therefore, higher doses may be required in
smokers to achieve the same plasma level.
Table 1 provides some recommendations for test doses
and starting doses of various depot drugs. Table 2 provides
a summary of controlled trials comparing different fixed
doses of some depot antipsychotic drugs which lasted at
least one year. Though as discussed previously, these
investigations did not employ identical definitions of
relapse the results are relatively consistent and provide
some guidance as to dosage requirements in stable outpatients attending well-staffed clinics with a variety of ancillary services and careful monitoring.

2.1.1. Advantages
Depot antipsychotics have several advantages over oral
medication, the major one being facilitation of compliance

Table 2
Controlled trials comparing different fixed doses of some depot antipsychotic drugs
Drug / Dose (mg)

Relapse rate (percent)

during one-year treatment periods

FD 1.255
FD 2.510
FD 12.550

56%
24% Kane et al., 1983; Kane and Borenstein, 1985
14%

FD 510
FD 2550

22%
20% Marder et al., 1987; Marder, 1994

FLUD mean512
FLUD mean525

32%
10% Johnson et al., 1987

FD mean53.8
FD mean525

20%
14% Hogarty et al., 1988

HD
HD
HD
HD

60%
25% Kane, 1986
23%
15%

25
50
100
200

FD 2.510
FD 12.550

29%
20% Levine et al., 1980

FD5fluphenazine decanoate.
FLUD5flupenthixol decanoate.
HD5haloperidol decanoate.

in medication-taking. Mann (1986) suggested that roughly

one in three patients with schizophrenia taking oral medication fail to comply reliably, though higher estimates
have been reported as well. As emphasized previously, one
aspect of compliance enhancement associated with depot
drugs is the fact that non-compliance when it occurs is
overt, immediately detectable and therefore can be responded to productively.
Other advantages of depot drugs include more predictable and stable plasma levels of active drug because the
variability associated with absorption and first-pass metabolism are avoided. In addition, the treating clinician has
much better control over the drug management and is
therefore in a better position to titrate to optimum levels.
Another advantage is that if a patient misses an injection
for whatever reason there is not an abrupt discontinuation,
therefore, early relapse and adverse withdrawal effects are
less likely. Although some have argued that relying on
long-acting injections may reduce the frequency of contact
with mental health professionals, this is not determined by
the mode of medication administration. At the same time,
an injection schedule can help to establish a pattern of
regular contact with the health care system.
Though there has been debate as to what extent depot
drugs are associated with significantly fewer relapses and
rehospitalizations, in our view when all of the data from
individual trials and meta-analyses are taken together, the
findings are extremely compelling in favor of depot drugs.
Davis et al. (1994) have recently reviewed the doubleblind studies comparing oral and depot antipsychotic

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

administration in outpatients. Six studies lasting at least 40

weeks in duration were included in a meta-analysis which
demonstrated a highly significant reduction in relapse rate
with depot medication ( p50.0002, MantelHaenszel
Test). The value of depot medication is also supported by
the results of mirror-image studies which contrast relapse
rates within the same patients over a specified time before
and after the initiation of depot antipsychotic medication in
their treatment. Davis et al. (1994) reanalyzed data from
six mirror-image studies and found for all of them a
statistically significant reduction in the number of days in
hospital during the period of depot administration.
It is important to recognize that the population of
patients most likely to benefit from depot medication are
those who are uncooperative and non-compliant. These
patients are probably least likely to participate in controlled clinical trials where they are randomly assigned to
one of two conditions, but end up having to take pills and
receive injections (in order to maintain the double-blind)
as well as undergoing more extensive clinical evaluations
than under routine circumstances. In addition, if patients
are initially compliant, non-compliance may take some
time to develop and the consequences of complete noncompliance (i.e. discontinuation of oral medication) may
not be seen (in terms of relapse) for several months.
Therefore, studies which last only one year may be
inadequate to accurately estimate the potential superiority
of depot drugs (Kane and Borenstein, 1985). Most such
studies lasted only one year.
Even if depot drugs were associated with a very modest
10% reduction in rates of relapse and rehospitalization, this
has enormous public health and health care cost implications.

2.1.2. Disadvantages
The potential disadvantages of depot drugs include
patient reluctance to accept injections or a sense of being
overly controlled. In addition, there is a fear on the part of
clinicians and sometimes patients that if adverse effects do
occur, they will be more difficult to manage because of the
inability to rapidly discontinue the medication.
Some clinicians and patients also continue to believe
that certain adverse effects may be more common with
depot drugs than with oral drugs. In the experience of the
panel, these disadvantages are exaggerated or insignificant.
If clinicians are patient in explaining the potential benefits
of depot drugs and can work with the patient to at least try
one or two injections (and if the injections are given
skillfully), the overwhelming majority of patients can
overcome any initial trepidation. The perceived loss of
control is more imagined than real and there are many
other areas of every day life where patients have ample
opportunity to become more autonomous without risking
psychotic relapse and its consequences.
In terms of adverse effects, there are not convincing data
that, when comparisons are made of equivalent dosages

63

and taking into consideration compliance issues, depot

drugs are associated with a significantly higher incidence
of adverse effects than oral drugs.
We specifically include tardive dyskinesia and neuroleptic malignant syndrome in this analysis. In the case of the
latter, there is no evidence that mortality is any higher
(despite the inability to rapidly discontinue) with depot as
opposed to oral drugs (Glazer and Kane, 1992).
Some patients experience pain or discomfort at the
injection site. In some cases (Haman et al., 1990), the
injection site can become edematous, and tender or pruritic
with a palpable mass being present for up to three months.
These authors estimated such reactions to be occurring in
approximately 8% of their patients receiving haloperidol
decanoate. Haloperidol decanoate is probably the most
viscous of the depot preparations and, therefore, should not
be given in volumes exceeding 3 ml. Local reactions can
be reduced by rotating the injection site, limiting the
volume of the injection and assuring deep intramuscular
injections. The z-track technique has been recommended to
reduce leakage (Belanger and Chouinard, 1982).

2.2. Clinical dose titration or plasma drug monitoring

It may be difficult to find the optimal dose, when our
titration endpoints depend to some extent on symptom
exacerbation or adverse effects.
In long-term maintenance treatment, the optimal or
lowest effective dose is best obtained by means of gradual
dose reduction until the appearance of mild prodromal
symptoms. Especially when using depot neuroleptics, this
may be done without great risk of severe relapse and
rehospitalization. Furthermore, by close observation of the
patient and close contact with the relatives, it is possible to
intervene immediately and prevent a manifest relapse and
rehospitalization.
The plasma concentration corresponding to the lowest
effective dose shows a wide range, varying by a factor of
10. This means that the plasma concentration cannot be
used as a guideline in these patients, although 6075% of
the patients have plasma concentrations within a relatively
narrow range, varying by a factor of 4. Regardless, this is
too wide a variation to be useful in routine monitoring. A
dose 4 times higher than necessary may be potentially
harmful and may produce severe side effects. Vice versa,
there are patients who require very high doses and plasma
concentrations and in whom a dose reduction to within the
therapeutic window could cause symptomatic worsening.
There are, however, a number of situations where
measurement of the plasma drug concentration is useful.
Such situations include cases of insufficient therapeutic
effect, unexpected serious side effects, suspicion of noncompliance (in case of oral medication), drug interactions
(e.g. with antidepressants and carbamazepine) and certain
somatic diseases involving the liver or the kidney.

64

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

2.3. Who should be considered for depot drugs?

Any patient for whom long-term treatment is indicated
should be considered. Those patients who are irregular in
taking medications are particularly good candidates. Even
in patients who initially refuse this option, clinicians
should work with them (through the therapeutic alliance)
to help the patients understand the potential advantages.

2.3.1. When should the treatment start?

As soon as possible and feasible, after the improvement
of acute symptoms (oral or short or intermediate-acting
i.m. medication is preferable for acute treatment as long as
flexibility of dosage is desirable).
2.3.2. Which drug?
In choosing which drug the clinician should consider
previous experience, personal patient preference, patients
history of response (both therapeutic and adverse effects)
and pharmacokinetic properties. There is no definite evidence that any one depot drug is superior to another in
terms of efficacy, though they may differ in side effect
profile.
2.3.3. How much?
Table 1 provides an overview of recommended doses
and dose ranges. It should be emphasized that we do not
have good data on dosage equivalence across drugs.
Forsman and Ohman (1977) suggested a monthly injection
of haloperidol decanoate utilizing a dosage 20 times that
which was being administered on a daily basis orally,
would produce a similar blood level. It is unclear whether
the blood level associated with depot treatment needs to be
the same as it was with oral medication.
It is important for clinicians to use their judgment and
not routinely utilize a particular formula. For example, if a
patient is receiving a high dose of oral medication,
consideration should be given to the possibilities that such
a high dose was never proven to be truly helpful or the
patient was not adequately absorbing or was rapidly
metabolizing the oral drug. Given the fact that the depot
drug can overcome these problems the dosage required
will be substantially less than would result from employing
a simple conversion formula.
The strategy for initiating a depot drug will vary
depending upon factors such as prior drug dosage and
clinical state. Some recommend initial loading dose, some
a more gradual titration. (See Table 1 for recommended
starting doses). The need to supplement with oral medication should be considered particularly in the first 4 to 6
weeks. Abrupt discontinuation of oral medication should
be avoided whenever possible. Depot medication dosage
should be increased over a 23 month period. In addition
clinicians should provide careful surveillance during this
transition.
Blood levels are not routinely recommended however

they may be useful in patients who are not responding

adequately or develop significant side effects or are
receiving concomitant drugs which may affect blood
levels.

3. Conclusions
The prevention of relapse in psychotic illness and
particularly schizophrenia is an important public health
goal. The use of depot drugs has important advantages in
facilitating relapse prevention. Certainly pharmacotherapy
must be combined with other treatment modalities as
needed, but the consistent administration of the former is
often what enables the latter.

References
Adamson, L., Curry, S.H., Bridges, P.K., Firestone, A.F., Lavin, N.I.,
Lewis, D.M., Watson, R.D., Xavier, C.M., Anderson, J.A., 1973.
Fluphenazine decanoate trial in chronic inpatient schizophrenics
failing to absorb oral chlorpromazine. Dis. Nerv. System 33, 181191.
Altamura, A.C., Colacurcio, F., Mauri, M.C. et al., 1990a. Haloperidol
decanoate in chronic schizophrenia: a follow up study of 12 months
with plasma levels. Prog. Neuropsychopharmacol. Biol. Psychiatry 14,
2535.
Altamura, A.C., Mauri, M.C., Girardi, T., Panetta, B., 1990b. Clinical and
toxicological profile of fluphenazine decanoate in elderly chronic
schizophrenia. Int. J. Clin. Pharmacol. Res. 10 (4), 223228.
Angst, J., Bech, P., Bruinvels, J., Engel, R.R., Ferner, U., Guelfi, J.D.,
Lingjaerde, O., Muller-Oerlinghausen, B., Paes de-Sousa, M., Paykel,
E. et al., 1994. Report on the fifth consensus conference: methodology
of long-term clinical trials in psychiatry. Pharmacopsychiatry 27 (3),
101107.
Awad, A.G., 1992. Quality of life of schizophrenic patients on medications and implications for new drug trials. Hosp. Commun. Psychiatry
43 (3), 262265.
Awad, A.G., Hogan, T., 1994. Subjective response to neuroleptics and the
quality of life: implications for treatment outcome. Acta Psychiatr.
Scand. 380 (suppl), 2732.
Barnes, T.R.E., Curson, D.A., 1994. Long-term depot antipsychotics: A
risk-benefit assessment. Drug Safety 10, 464479.
Belanger, M.C., Chouinard, G., 1982. Technique for injecting long-acting
neuroleptics. Br. J. Psychiatry 141, 316.
Bland, R.C., Parker, J.H., Orn, H., 1976. Prognosis in schizophrenia: A
ten year follow-up of first admissions. Arch. Gen. Psychiatry 33,
949954.
Brown, G.W., Birley, J.L.T., Wing, J.K., 1972. Influence of family life on
the course of schizophrenic disorders: A replication. Br. J. Psychol.
121, 241248.
Carpenter, W.T., Stephen, J.C., Rey, A.C., Hanlon, T.E., Heinrichs, D.W.,
1982. Psychopharmacol. Bull. 18, 2123.
Carpenter, W.T., Heinrichs, D.W., 1983. Early intervention, time-limited,
targeted pharmacotherapy of schizophrenia. Schizophrenia Bull. 9,
533542.
Carpenter, W.T., Heinrichs, D.W., Hanlon, T.E., 1987. A comparative trial
of pharmacologic strategies in schizophrenia. Am. J. Psychiatry 144,
14661470.
Carpenter, W.T., Hanlon, T.E., Heinrichs, D.W., Summerfelt, A.T.,
Kirkpatrick, B., Levine, J., Buchanan, R.W., 1990. Continuous versus
targeted medication in schizophrenic out-patients: outcome results.
Am. J. Psychiatry 147, 11381148.

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

Collins, E.J., Hogan, T.P., Deasi, H., 1991. Measurement of therapeutic
response in schizophrenia: A critical survey. Schizophr. Res. 5 (3),
249253.
Davies, L.M., Drummond, M.F., 1994. Economics and schizophrenia. Br.
J. Psychiatry 165 (suppl. 25), 1821.
Davis, J.M., Barter, J.T., Kane, J.M., 1989. Antipsychotic drugs. In:
Kaplan, H.I., Sadock, B.J. (Eds.), Comprehensive Textbook of Psychiatry. Williams and Wilkins, Baltimore, pp. 15911626.
Davis, J.M., Matalon, L., Watanabe, M.D., Blake, L., 1994. Depot
antipsychotic drugs. Place in therapy. Drugs 47 (5), 741773.
Drake, R., Gates, C., Cotton, P.C., Whitaker, A., 1984. Suicide among
schizophrenics: Who is at risk?. J. Nerv. Ment. Dis. 172, 613617.
Ereshefsky, L., Jann, M.W., Saklad, S.R. et al., 1985. Effects of smoking
on fluphenazine clearance in psychiatric inpatients. Biol. Psychiatry
40, 329332.
Ereshefsky, L., Saklad, S.R., Jann, M.W., Davis, C.M., Richards, A.,
1984. Future of depot neuroleptic therapy: Pharmacokinetic and
pharmacodynamic approaches. J. Clin. Psychiatry 45 (suppl), 5059.
Ereshefsky, L., Saklad, S.R., Tran-Johnson, T., Toney, G., Lyman, R.C.,
1990. Kinetics and clinical evaluation of haloperidol decanoate loading
dose regimen. Psychopharmacol. Bull. 26, 108114.
Forrest, F.M., Forrest, I.S., Mason, A.S., 1961. Review of rapid urine
tests for phenothiazine and related drugs. Am. J. Psychiatry 118,
300307.
Forsman, A., Ohman, R., 1977. Applied pharmacokinetics of haloperidol
in man. Curr. Ther. Res. 21, 245255.
Frankenburg, F.R., Hegarty, J.D., 1996. Cost considerations in the
treatment of schizophrenia. CNS Drugs 5, 7582.

Gaebel, W., Frick, U., Kopcke, W., Linden, M., Muller,

P., Muller-Spahn,
F., Pietzcker, A., Tegeler, J., 1993. Early neuroleptic intervention in
schizophrenia: are prodromal symptoms valid predictors of relapse?.
Br. J. Psychiatry 163 (suppl. 21), 812.
Gaebel, W., Awad, G., 1994. Prediction of Neuroleptic Treatment
Outcome in SchizophreniaConcepts and Methods, Springer, Vienna.
Gilbert, P.L., Harris, M.J., McAdams, L.A., Jeste, D.V., 1995. Neuroleptic
withdrawal in schizophrenic patients: A review of the literature. Arch.
Gen. Psychiatry 52, 173188.
Glazer, W.M., Kane, J.M., 1992. Depot neuroleptic therapy: An underutilized treatment option. J. Clin. Psychiatry 53, 426433.
Haman, G.L., Egan, T.M., Wells, B.G. et al., 1990. Injection site reactions
after intramuscular administration of haloperidol decanoate 100 mg /
ml J. Clin. Psychiatry 51, 502504.
Hale, A.S., Wood, C., 1996. Comparison of direct treatment costs for
schizophrenia using oral and depot neuroleptics: a pharmacoeconomic
analys. Br. J. Med. Econ. 10, 3745.
Herz, M.I., Melville, C.H., 1980. Relapse in schizophrenia. Am. J.
Psychiatry 137, 801805.
Herz, M.I., Szyinanski, H.V., Simon, J.C., 1982. Intermittent medication
for stable schizophrenic outpatients: an alternative to maintenance
medication. Am. J. Psychiatry 139, 918922.
Herz, M.I., Glazer, W.M., Moster, M.A., Sheard, M.A., Szymanski, H.V.,
Hafez, H., Mirza, M., Vana, J., 1991. Intermittent vs maintenance
medication in schizophrenia: Two-year results. Arch. Gen. Psychiatry
48, 333339.
Hogan, T.D., Awad, A.G., Eastwood, R., 1983. A self-report scale
predictive of drug compliance in schizophrenia: Reliability and
discriminative validity. Psychol. Med. 13, 177183.
Hogarty, G.E., Goldberg, S.G., Schooler, N.R., 1974. Drug and
sociotherapy in the aftercare of schizophrenic patients: Ill Adjustments
of nonrelapsed patients. Arch. Gen. Psychiatry 31, 609618.
Hogarty, G.E., McEvoy, P., Munetz, M., DiBarry, A.L., Bartone, P.,
Cather, R., Cooley, S.J., Uirich, R.F., Carter, M., Madonia, M.J., 1988.
Dose of fluphenazine, familial expressed emotions, and outcome in
schizophrenia: results of a two-year controlled study. Arch. Gen.
Psychiatry 45, 797805.
Jeste, D.V., Caligiuri, M.P., 1993. Tardive dyskinesia. Schizophr. Bull. 19,
303315.

65

Johnson, D.A., Ludlow, J.M., Street, K., Taylor, R.D., 1987. Double blind
comparison of half-dose and standard-dose flupenthixol decanoate in
the maintenance treatment of stabilised out-patients with schizophrenia. Br. J. Psychiatry 151, 634638.
Jolley, A.G., Hirsch, S.R., McRink, A., Manchanda, R., 1989. Trial of
brief intermittent neuroleptic prophylaxis for selected schizophrenic
outpatients: clinical outcome at one year. Br. Med. J 298, 985990.
Jolley, A.G., Hirsch, S.R., Morrison, E., McRink, A., Wilson, L., 1990.
Trial of brief intermittent neuroleptic prophylaxis for selected schizophrenic outpatients: clinical outcome at two years. Br. Med. J 301,
837842.
Kane, J.M., Rifkin, A., Woemer, M.G., Reardon, G., Sarantak, S.,
Schiebel, D., Ramosior, J., 1983. Low-dose neuroleptic treatment of
outpatient schizophrenics: I. Preliminary results for relapse rates. Arch.
Gen. Psychiatry 40, 893896.
Kane, J.M., Borenstein, M., 1985. Compliance in the long-term treatment
of schizophrenia. Psychopharmacol. Bull. 21, 2327.
Kane, J.M., 1986. Dosage strategies with long-acting injectable neuroleptics, including haloperidol decanoate. J. Clin. Psychopharmacol. 6,
205235.
Kane, J.M., Jeste, D.V., Bames, T.R.E., et al., 1992. American Psychiatric
Association Task Force Report on Tardive, Dyskinesia, American
Psychiatric Press, Washington, DC.
Kane, J.M., 1995. Tardive dyskinesia. Paper presented at International
Congress on Schizophrenia Research, April 1013, Warm Springs,
Virginia.
Kane, J.M., 1996. Schizophrenia. N. Engl. J. Med. 334 (1), 3441.
Kinon, B.J., Kane, J.M., Chakos, M., Munne, R., 1993. Possible
predictors of neuroleptic-resistant schizophrenic relapse: Influence of
negative symptoms and acute extrapyramidal side effects. Psychopharmacol. Bull. 29 (3), 365369.
Kreisman, D., Blumenthal, R., Borenstein, M., Woerner, M., Kane, J.M.,
Rifkin, A., Reardon, G., 1988. Family attitudes and patient social
adjustment in a longitudinal study of outpatient schizophrenics receiving low-dose neuroleptics: The familys view. Psychiatry 51, 313.
Lauer, G., Stegmuller-Koenemund, U., 1994. Domain-specific subjective
quality of life and illness-related limitations of schizophrenic patients.
Psychiatr. Pract. 21 (2), 7073.
Levine, J., Schooler, N.R., Severs, J. et al., 1980. Discontinuation of oral
and depot fluphenasine in schizophrenic patients after one year of
continuous medication: a controlled study. Adv. Bioch. Psychopharmacol. 24, 483493.
Lieberman, J.A., Saltz, B.L., Johns, C.A., Pollack, S., Borenstein, M.,
Kane, J.M., 1991. The effects of clozapine on tardive dyskinesia. Br. J.
Psychiatry 158, 503510.
Mann, J.J., 1986. How medication compliance affects outcome. Psychiatr.
Ann. 16, 567570.
Marder, S.R., vanPutten, T., Mintz, J., Mckenzie, J., Lebell, M., May,
P.R.A., 1984. Costs and benefits of two doses of fluphenazine. Arch.
Gen. Psychiatry 41, 10251029.
Marder, S.R., vanPutten, T., Mintz, J., Lebell, M., May, P.R.A., 1987.
Low and conventional dose maintenance therapy with fluphenazine
decanoate: Two year outcome. Arch. Gen. Psychiatry 44, 518521.
Marder, S.R., 1994. The role of dosage and plasma levels in neuroleptic
relapse prevention. Acta Psych. Scandinav. 382, 2527.
May, P.R.A., Tuma, A.H., Dixon, W.J., 1981. Schizophrenia: A follow-up
study of five forms of treatment. Arch. Gen. Psychiatry 38, 776784.
McEvoy, J.P., Howe, A.C., Hogarty, G.E., 1984. Differences in the nature
of relapse and subsequent inpatient course between medication-compliant and non-compliant schizophrenic patients. J. Nerv. Ment. Dis.
172 (7), 412416.

Meise, U., Rossler,

W., Fleischhacker, W.W., 1994. Outpatient care in
schizophrenia: treatment considerations. CNS Drugs 1 (4), 254262.
Miles, P., 1977. Conditions predisposing to suicide: A review. J. Nerv.
Ment. Dis. 164, 231246.
Perel, J.M., 1988. Compliance and optimal dosage during maintenance
therapy: Pharmacokinctic and analytic issues. J. Clin. Chem. 34,
881887.

66

J.M. Kane et al. / European Neuropsychopharmacology 8 (1998) 55 66

Pietzcker, A., Gaebel, W., Kopcke,

W., Linden, M., Muller,
P., MullerSpahn, F., Schussler, G., Tegeler, J., 1986. A German multicenter
study on the neuroleptic long-term therapy of schizophrenic patients:
Preliminary report. Pharmacopsychiatry 19, 161166.

Pietzcker, A., Gaebel, W., Kopcke,

W., Linden, M., Muller,
P., MullerSpahn, F., Tegeler, J., 1993. Intermittent versus maintenance neuroleptic long-term treatment in schizophrenia-Two year results of a German
multicenter study. J. Psychiatr. Res. 27, 321339.
Pikney, A.A., Gerber, G.J., Lafave, H.G., 1991. Quality of life after
psychiatric rehabilitation: the clients perspective. Acta. Psychiatr.
Scand. 83 (2), 8691.
Radomsky, E.D., 1995. Measurement and assessment of compliance with
an oral haloperidol regimen. Doctoral Dissertation, Department of
Psychology, University of South Carolina.
Rice, P., Miller, L.S., 1996. The economic burden of schizophrenia;
conceptual and methodological issues, and costs estimates. In: Moscarelli, M., Rupp, A., Sartorius, N., (Eds.), Handbook of Mental
Health Economics and Healthy Policy. vol. 1, Schizophrenia, Wiley
and Sons, Chichester, pp. 321334.
Saltz, B.L., Woemer, M.G., Kane, J.M., Lieberman, J.A., Alvir, J.M.,
Bergmann, K.J., Blank, K., Koblenzer, J., Kahaner, K., 1991. Prospective study of tardive dyskinesia incidence in the elderly. JAMA 266,
24022406.
Schooler, N.R., 1985. Treatment of schizophrenia: Maintenance strategies
and pharmacologic tactics. In: Alpert M., (Ed.), Controversies in
Schizophrenia, Guilford Press, New York, pp. 366-380.
Schooler, N.R., Keith, S.J., Severe, J.B., Matthews, S.M., 1993. Treatment strategies in schizophrenia: effects of dosage reduction and
family management on outcome. Schizophr. Res. 9, 260.
Schooler, N.R., Keith, S.J., Severe, J.B. et al., 1995. Maintenance

treatment of schizophrenia: a review of dose reduction and family

treatment strategies. Psychiatric Quart. 66, 279292.
Schulz, P., Rey, M.J., Dick, P., Tissot, R., 1989. Guidelines for the dosage
of neuroleptics. II: Changing from daily oral to long acting injectable
neuroleptics. Intl. Clin. Psychopharmacol. 4, 105114.
Schulze-Monking, H., Buchkremer, G., 1995. Emotional family atmosphere and relapse: investigations on the role of relapse. definition,
duration of illness and resignation of relatives. Eur. Psychiatry 10,
8591.
Shalev, A., Hermesh, H., Rothberg, J., Munitz, H., 1993. Poor neuroleptic
response in acutely exacerbated schizophrenic patients. Acta Psychiatr.
Scand. 87, 8691.
Sleator, E.K., 1985. Measurement of compliance. Psychopharmacol. Bull.
21 (3), 10891093.
Smith, K., Shah, A., Wright, K., Lewis, G., 1995. The prevalence and
costs of psychiatric disorders and learning disabilities. Br. J. Psychiatry 166, 918.
Steingard, S., Allen, M., Schooler, N.R., 1994. A study of the pharmacologic treatment of medication compliant schizophrenics who relapse. J.
Clin. Psychiatry 55, 470472.
vanPutten, T., May, P.R.A., 1978. Subjective response as a predictor of
outcome in pharmacotherapy: The consumer has a point. Arch. Gen.
Psychiatry 35, 447480.
Weiden, P., Rapkin, B., Mott, T., Zygmunt, A., Goldman, D., HorvitzLennon, M., Frances, A., 1994. Rating of Medication Influences
(ROMI) Scale in Schizophrenia. Schizophr. Bull. 20 (2), 297306.
Young, J.L., Zonana, H.V., Shepler, L., 1986. Medication compliance in
schizophrenia: Codification and update. Bull. Am. Acad. Psychiatry
Law 14 (2), 105.