The Evaluation and Treatment of First-Episode
Psychosis
by Brian B. Sheitman, Heidi Lee, Rael Strauss, and Jeffrey A. Ueberman
Abstract
A first episode of psychosis is a traumatic experience
for patients and families. At the time of initial evalua-
tion, the differential diagnosis should include a broad
range of neurological, general medical, and psychiatric
conditions. Methodological advances in operationally
defining illness onset, "offset," and remission have
allowed more careful studies of treatment response in
first-episode patients. These studies strongly support
the efficacy of antipsychotic medication as both acute
and maintenance treatment for patients with a first
episode of psychosis. The optimal duration of mainte-
nance treatment, however, has not been determined,
and patients at low risk for relapse following medica-
tion withdrawal cannot be identified with specificity.
First-episode psychotic patients typically experience
12 to 24 months of psychosis before receiving treat-
ment, and a long duration of untreated psychosis may
be associated with a poorer treatment response. Early
intervention may improve outcome in first-episode
psychosis, and the use of novel antipsychotics with
improved efficacy and fewer side effects may improve
medication compliance and reduce morbidity associ-
ated with repeated relapses.
Schizophrenia Bulletin, 23(4):653-661,1997.
Compared with the substantial data available from the
study of chronic multiepisode patients, there is a paucity
of research describing the phenomenology and treatment
of first-episode psychotic patients. Keshavan and
Schooler (1992) identified 53 first-episode schizophrenia
or psychosis studies; two-thirds of which focused on bio-
logical phenomena, most often neuroimaging. They
described differences in clinical assessment methods and
inconsistent reporting of key variables that have limited
our ability to generalize from these investigations. Greater
attention to carefully defining patient populations and ill-
653
ness characteristics has consistently been recommended to
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
increase comparability across first-episode studies (Kirch
et al. 1992; Lieberman et al. 1992; Maurer and Hafner
1995). In this article, we review the differential diagnosis
of a first episode of psychosis, discuss selected method-
ological issues that must be considered in order to inter-
pret available empirical studies, and comprehensively
review the results of acute and maintenance treatment
studies of first-episode psychotic, predominantly schizo-
phrenia patients.
Differential Diagnosis
Evaluating a patient with a first episode of psychosis
requires consideration of a broad differential diagnosis.
Psychosis is not pathognomonic for schizophrenia, but
rather a symptom of a wide array of psychiatric, neurolog-
ical, and general medical disorders. The absence of a fam-
ily history of major mental illness, an acute illness onset,
an age at onset beyond the middle 30s, and psychosis in a
patient undergoing treatment in an emergency, general
medical, or intensive care unit will heighten concern that
psychotic symptoms are secondary to a nonpsychiatric ill-
ness. Even if the index of suspicion is low, the prudent
evaluating psychiatrist will consider a first episode of psy-
chosis to be the result of a neurological or medical condi-
tion until completing a thorough evaluation.
Medical and Neurological Conditions. All first-
episode psychotic patients should have a thorough med-
ical evaluation, including a review of systems and a phys-
ical exam that includes a neurological evaluation. In
addition, a complete blood count, electrolytes, serum cre-
atine, blood urea nitrogen, thyroid function tests, venereal
disease tests, urinalysis, and a toxicology screen should
Reprint requests should be sent to Dr. B.B. Sheitman, Dorothea Dix
Hospital, CRU-Edgerton Bldg., 809 Ruggles Dr., Raleigh, NC 27603.
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
all be obtained. If neurological signs or symptoms includ-
ing asymmetry, weakness, or an altered sensorium are
present, a brain magnetic resonance imaging (MRI) or
computerized axial tomography (CAT) scan should be
obtained. Brain imaging or an electroencephalogram
(EEG) should also be considered for patients who are
uncooperative with the neurological exam or who have an
atypical presentation (e.g., a first episode at age 70).
Psychotic symptoms have been described in many
neurological and general medical disorders (Doran et al.
1986). Neurological conditions to be considered in the
differential diagnosis of first-episode psychosis include
head trauma, infections, brain tumors, seizures, multiple
sclerosis, metachromatic leukodystrophy, Huntington's
disease, and Wilson's disease. General medical conditions
in the differential diagnosis include endocrinopathies
(thyroid, adrenal, pancreatic), autoimmune disorders (sys-
temic lupus erythematosus), vitamin deficiencies (B ]2 ),
and hepatic or erythropoietic disorders of metabolism
(acute intermittent porphyria). A relatively common pre-
cipitant of psychotic symptoms is an adverse reaction to
prescribed medication, including steroids, L-dopa, anti-
cholinergics, and H 2 blockers.
Psychiatric Disorders.
Affective illness. When patients with a first episode
of psychosis also demonstrate manic or depressive symp-
toms, differentiating a primary psychotic disorder from a
mood disorder with psychotic symptoms is problematic.
In order to make this determination, many classification
systems require a retrospective assessment of the tempo-
ral relationship between the onset of mood symptoms and
psychosis, which is often difficult to determine. In addi-
tion, many patients have depressive or manic symptoms,
but do not meet the full syndromal criteria for an affective
disorder. Because differentiating the two syndromes may
have treatment and prognostic implications, a careful his-
tory that includes interviews with family members and
other informants should be obtained. In a sample of first-
episode psychotic patients for whom treatment was not
controlled, Tohen et al. (1992) reported that patients diag-
nosed with affective psychotic disorders had shorter hos-
pitalizations and a higher level of functional recovery than
patients diagnosed with schizophrenia spectrum disorders.
In an open treatment trial using a standardized medication
algorithm, Koreen et al. (1993) reported that the depres-
sive symptoms of the vast majority of acutely psychotic
patients diagnosed with schizophrenia and schizoaffective
disorder were resolved when psychotic symptoms had
remitted.
Substance Abuse. A significant proportion of first-
episode psychotic patients report current or past substance
654
B.B. Sheitman et al.
abuse (Bromet et al. 1992). The hallucinogens LSD and
PCP—as well as cocaine, marijuana, and alcohol—may
induce psychotic symptoms. Illicit substance abuse often
presents a diagnostic challenge (e.g., the patient whose
last use of cocaine is more than 3 days past, but who
remains psychotic and agitated and requires antipsychotic
medication). Strict adherence to DSM—IV (American
Psychiatric Association 1994) requires that the psychotic
symptoms persist for at least 1 month to exclude a diagno-
sis of substance-induced psychotic disorder. The effects of
substance abuse on the onset, treatment response, course,
and outcome of psychotic illnesses are not well under-
stood, although most (DeQuardo et al. 1994; Chouljian et
al. 1995; Gupta et al. 1996), but not all (Dixon et al. 1991)
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
evidence suggests a deleterious effect on treatment
response and outcome.
Brief psychosis. Extremely transient psychotic
episodes, lasting minutes to hours, can sometimes occur
in individuals with borderline or schizotypal personality
disorder. In contrast, the diagnosis of brief psychotic dis-
order applies to psychosis that lasts at least 1 day, but less
than 1 month. The episode is often characterized by florid
symptoms, confusion, and emotional turmoil in the con-
text of an environmental stressor (Frances et al. 1995). In
a prospective study of 221 first-admission patients with
psychosis, Susser et al. (1995) noted that 20 (9%) patients
had a brief psychosis (defined as full remission of symp-
toms at 6-month followup). For patients with acute (as
opposed to gradual) onset, brief psychosis appeared to be
a distinctive and stable subgroup. Most patients with brief
psychosis were women, and none of the patients' disor-
ders had evolved into chronic schizophrenia or affective
disorders by 24 months.
Prospective longitudinal observation over a period of
6 months or more may be required to clarify diagnosis in
patients presenting with a first episode of psychosis.
Methodological Considerations
Illness Onset. Definitions of the beginning of a first
psychotic episode must account for the likelihood that
many dimensions of behavioral and cognitive functioning
can be impaired with the time of onset of each symptom
differing. Prodromal symptoms may appear many years
before the onset of a frank psychosis (Carpenter and
Kirkpatrick 1988; Hafner and Nowotny 1995; Yung and
McGorry 1996), and patients often do not present for
admission to a hospital for a first episode of psychosis
until a significant amount of time after the onset of psy-
chotic symptoms, that is, 12-24 months (Gift et al. 1981;
Fenton 1997). Keshavan and Schooler (1992) have listed
Evaluation and Treatment of First-Episode Psychosis
six distinct clinical events that need to be examined in
identifying the onset of a first episode of psychosis:
decline in social functioning, onset of general behavioral
symptoms, onset of positive symptoms, onset of negative
symptoms, first treatment, and first hospital admission.
Response Criteria (Illness "Offset"). In treatment and
outcome studies where the end of the first episode and
number of subsequent episodes are used as measures of
illness course, a clear definition of the end of the episode
or episode "offset" is required. Multiple dimensions of
psychopathology in schizophrenia make the definition of
response complex, with the potential for a wide disparity
in response rates depending on the operational criteria
used. Response rates based on the resolution of positive
symptoms will be considerably higher than those based
primarily on negative symptom resolution. Social and
vocational disability may extend for a considerable period
beyond that required for symptom resolution. Sole
reliance on a percentage decrease in rating scale scores
from a pretreatment baseline may result in inflated
response rates, since most patients—particularly first-
episode patients—would be expected to have at least
some response to antipsychotic medication. Since man-
aged care has supplanted clinicians' judgment, the dura-
tion of initial hospitalization is likely to be of limited
validity as a measure of treatment response.
The characterization of patients who develop non-
schizophrenic symptoms following the remission of psy-
chosis and those who admit to ongoing residual delusions
can affect the response and relapse rates reported. A rela-
tively common example of the former is the appearance
of a postpsychotic depression (McGlashan and Carpenter
1976; Siris 1991). The concept and significance of a resid-
ual delusion (defined as a delusion about a past event that
persists even though the patient no longer believes it is
occurring at the present time and the patient's behavior is
not affected by it) has received little formal attention. An
example is the patient who claims that her phones were
tapped 1 year ago but not at present and is able to main-
tain work and social functioning despite the residual delu-
sion.
Treatment Studies
Because of the difficulty in enumerating cases and obtain-
ing patient consent, there have been relatively few studies
of the efficacy of standardized treatment administered to
first-episode psychotic patients. Treatment studies are to
be differentiated from outcome studies (May et al. 1981),
which assess illness course without controlling treatment.
Acute treatment studies of the first episode of psychosis,
655
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
primarily patients diagnosed with schizophrenia (May et
al. 1976a, 1976b; Scottish Schizophrenia Research Group
1987a, 1987b; Lieberman et al. 1993a, 1993fo; Szymanski
et al. 1994; Kopala et al. 1996), and maintenance studies
of remitted first-episode patients (Kane et al. 1982; Crow
et al. 1986; Johnstone et al. 1986; Scottish Schizophrenia
Research Group 1989; Nuechterlein et al. 1992, 1994) are
reviewed here.
Acute Treatment Studies. May et al. (1976a, 1976b)
conducted the first treatment study of first-episode schizo-
phrenia. In this study, 228 patients were randomly
assigned to one of five treatment arms: individual psy-
chotherapy, trifluoperazine, psychotherapy plus trifluoro-
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
perazine, electroconvulsive therapy (ECT), or milieu ther-
apy alone. Patients with "no significant prior treatment"
were selected on a "triage" basis from consecutive first-
admission schizophrenia patients to a State hospital
between 1959 and 1962. The authors attempted to select
the middle third of schizophrenia patients by excluding
those they believed would be unlikely to be discharged
and those who had remitted during the 18-day evaluation
period. The treatment continued until the patient was
released or 6-12 months of treatment was determined
unsuccessful. The response rates at the time of discharge
were as follows: psychotherapy plus drugs (95%); drugs
alone (96%); ECT (79%); psychotherapy (65%); milieu
therapy alone (58%). Although this study has method-
ological limitations from today's perspective, considering
when it was conducted it stands as a remarkable achieve-
ment in clinical research. The results demonstrated the
indisputable superiority of antipsychotic medication over
psychotherapy in the treatment of patients with a first
episode of schizophrenic psychopathology. At followup 3
to 5 years after treatment (May et al. 1981), patients in the
groups that had received drugs or ECT had the best out-
come, and those treated with psychotherapy alone had the
poorest outcome. Treatment over the followup period,
however, was not controlled.
The Scottish Schizophrenia Research Group (1987a)
described a randomized, double-blind, 5-week treatment
trial of pimozide versus flupenthixol for 49 first-episode
schizophrenia patients. The dose of each neuroleptic
began at 10 mg/day and could be titrated up to 40 mg for
pimozide and 50 mg for flupenthixol. Approximately half
of the eligible patients from four centers consented to par-
ticipate. Criteria for study entry was a clinician's Inter-
national Classification of Diseases-Ninth Revision
(1CD-9; World Health Organization 1978) diagnosis of a
first episode of schizophrenia. The Present State Exam
(PSE; Wing et al. 1974) administered during the first
week of admission was used to confirm clinical diagnosis;
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
patients were also evaluated with Research Diagnostic
Criteria (RDC; Spitzer et al. 1978a) and Feighner criteria
(Feighner et al. 1972). Using the RDC classification sys-
tem, there were 33 definite and 9 probable cases of schiz-
ophrenia; because of the requirement for 6 months of ill-
ness, Feighner criteria diagnosed only 15 cases of definite
or probable schizophrenia. Sixty-three percent were clas-
sified as "responders" to antipsychotic medication, with
similar results for each medication. Nonresponders tended
to be male, and they had more neurological signs, cogni-
tive deficits, and negative symptoms. Twenty percent of
the patient sample was noted to have a history of possible
neurological dysfunction. Response criteria were not
operationally defined.
In a prospective study of the psychobiology of psy-
chosis, Lieberman et al. (1993a, 19936) reported the re-
sults of an open trial of standardized antipsychotic treat-
ment for 70 RDC-diagnosed schizophrenia (n = 54) and
schizoaffective (n - 16) first-episode patients. The inclu-
sion-exclusion criteria were no prior psychotic episodes, a
diagnosis of definite or probable schizophrenia or
schizoaffective disorder (mainly schizophrenia) by RDC,
age 16 to 40, and no history of neurological or general
medical illness that could influence the diagnosis or the
biological variables that were to be assessed. Of 416
patient admissions screened, 219 met inclusion criteria,
but 127 either refused to participate or were excluded for
clinical or administrative reasons. Of the 92 patients who
entered the study, 19 were later withdrawn based on infor-
mation that became available subsequent to the baseline
assessments.
Patients were administered structured assessments—
Schedule for Affective Disorders and Schizophrenia-
Change Version (Spitzer et al. 1978Z>); Scale for the As-
sessment of Negative Symptoms (Andreasen and Olsen
1982); Clinical Global Impressions scale (CGI; Guy
1976); Simpson-Angus Neurological Rating Scale
(Simpson and Angus 1970); Simpson Dyskinesia Scale
(Simpson et al. 1979)—at baseline and biweekly and were
treated within an open standardized algorithm of sequen-
tial trials of fluphenazine (20 mg for 6 weeks, then 40 mg
for 4 weeks), haloperidol (20 mg for 6 weeks, then 40 mg
for 4 weeks), molindone hydrochloride (up to 300 mg/
day), and finally clozapine (up to 900 mg/day). Patients
proceeded through the algorithm until response criteria
were met. Based on positive and negative symptoms and
CGI scores, operational response criteria were used to
define full or partial remission. Illness onset was
described both as the age at which the patient exhibited
behavioral symptoms that, in retrospect, appear to have
been related to and were contiguous to the current illness
and as the age at first psychotic symptoms. The median
656
B.B. Sheitman et al.
durations of general behavioral symptoms and of psy-
chotic symptoms before study entry were 151 ± 176
weeks and 52 ± 82 weeks, respectively.
Using a survival analysis, the proportion of patients
remitting by 1 year was 83 percent, with the median and
mean time to remission 11 and 35.7 weeks, respectively.
Within the treatment algorithm, 62 percent of patients
developed at least one form of extrapyramidal side effects
(EPS) (Chakos et al. 1992). Brain pathomorphology and
abnormal basal growth hormone secretion significantly
predicted time to remission (Lieberman et al. 1993a). The
length of time a patient had been psychotic before study
entry (i.e., duration of untreated psychosis) was associated
with both a longer time to remission and a lower level of
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
remission among these first-episode patients (Loebel et al.
1992). In a followup to the Lieberman et al. (1993a)
study, Szymanski et al. (1994) reported on the efficacy of
open clozapine treatment among 10 first-episode patients
who failed to respond to typical neuroleptic treatment in
the standardized treatment protocol. The patients had been
psychotic before clozapine treatment for a median time of
131 weeks. Brief Psychiatric Rating Scale (BPRS; Overall
and Gorham 1962) scores were obtained after a 2-week
drug washout and biweekly for 12 weeks after beginning
clozapine treatment. Clozapine was titrated to 500 mg
over 2 weeks and then adjusted as "clinically indicated."
The mean maximum dose of clozapine was 687.5 mg/day
(standard deviation [SD] = 214.8) with a range of
300-900 mg/day. Response criteria were a 20-percent
reduction from a summed baseline BPRS score and a CGI
severity of illness score less than or equal to 3 (mildly ill).
Three of 10 patients met these criteria (2 at 6 weeks and 1
at 12 weeks), although none fully remitted. This 30-per-
cent response rate was comparable to the rate reported by
Kane et al. (1988) in treatment-resistant schizophrenia,
suggesting that patients who are refractory to treatment in
their first episode of illness may be among the most
severely ill patients with schizophrenia.
Kopala et al. (1996) described the efficacy of open
risperidone treatment among 22 (17 male) consecutively
admitted, neuroleptic-naive patients hospitalized for the
first time who met DSM-IV criteria for schizophrenia.
Substance abuse, previous head injury, other "organic
pathology," uncertain diagnosis, or inadequate collateral
history were exclusion criteria; 11 patients were excluded
based on these criteria. Patients had a mean duration of
"illness" before treatment of 245.7 weeks (SD = 217.8).
Psychopathology was assessed using the Positive and
Negative Syndrome Scale (PANSS; Kay et al. 1987), with
the authors reporting interrater reliability scores of 0.90.
Response was defined as a 20 percent or greater reduction
in the total PANSS score. Risperidone was initiated at
Evaluation and Treatment of First-Episode Psychosis
1 mg twice daily and increased to 2—8 mg/day; mean dose
was 4.7 mg/day (SD = 1.5). The mean duration of treat-
ment was 7.1 weeks (SD = 3.2; range, 1.8-14.1). The
authors reported that 59 percent of patients met response
criteria.
Maintenance Treatment Studies. Kane et al. (1982)
were the first to report on the efficacy of prophylactic
antipsychotic medication in a study of 28 (14 female)
patients screened from referrals to an aftercare clinic who
had achieved a stable remission following treatment for a
first episode of schizophrenia. Inclusion criteria were one
schizophrenic episode, no important psychopathology or
mental health contact for a period of 3 months before hos-
pitalization, a stable level of remission for at least 4
weeks (maximum, 1 year) following hospital admission,
no evidence of substance abuse or important medical ill-
nesses, and willingness to provide informed consent.
When rediagnosed by RDC, only 19 of these 28 patients
were diagnosed with schizophrenia. Subjects received
oral fluphenazine (5—20 mg/day), fluphenazine decanoate
(12.5-50 mg every 2 weeks), or placebo. Only subjects
judged likely to be noncompliant were randomized to
fluphenazine or placebo injections, and randomization
was changed to favor placebo 2:1. Relapse was defined as
substantial clinical deterioration, with a potential for
marked social impairment. Of 28 patients enrolled, 7
relapsed and 8 others completed the 1 year of treatment.
Of the RDC schizophrenia patients who received
fluphenazine (n - 6), none relapsed, 4 left the study before
1 year, and 2 subjects completed; of the 13 RDC schizo-
phrenia patients treated with placebo, 6 relapsed, 6
dropped out, and only 1 completed. Overall, 7 of the 17
patients who received a placebo, but none of the 11 drug-
treated patients relapsed over 1 year. Despite methodolog-
ical limitations, this study provided evidence of the effi-
cacy of maintenance antipsychotic medication in reducing
the risk of relapse among first-episode patients.
Crow et al. (1986) described a randomized, placebo-
controlled trial of maintenance neuroleptics for 120
patients discharged following a first episode of schizo-
phrenia. As part of the Northwick Park Study of first-
episode schizophrenia, patients were recruited from sev-
eral medical centers within 35 miles of Harrow, England.
Inclusion criteria were age 15-70 years; a first psychotic
episode (no past psychosis, possible psychosis, or previ-
ous inpatient care exceeding 3 days); admission to inpa-
tient or day-patient care for at least 1 week; clinical diag-
nosis of schizophrenia, categories C, O, or P on the PSE
(Wing et al. 1974); and absence of organic disease of defi-
nite or probable etiological significance. Of the 462 cases
referred, 209 were excluded and 253 agreed to participate
657
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
in the initial assessments. Of these, 120 (74 male) agreed
to participate in the maintenance medication trial. Onset
of illness was defined as the beginning of psychotic symp-
toms; response, as discharge from the hospital for 30 days
or more; and relapse, as readmission to psychiatric care
for any reason. Participating patients had been psychotic
for a median of 2.8 months (range, 1-101 months) before
index admission. Patients were assigned to one of five
antipsychotics or to placebo. The minimum medication
dosages were flupenthixol intramuscular 40 mg/month,
chlorpromazine 200 mg, haloperidol 3 mg, pimozide 4
mg, and trifluoperazine 5 mg per day. Fifty-four patients
(35 male) who received medication and 66 (37 male) who
received placebo were treated up to 2 years or until study
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
termination. Among 107 patients who completed the
study, 62 percent of those on placebo and 46 percent of
those on active medication relapsed. The duration of ill-
ness before beginning neuroleptic medication was the
most important determinant of relapse, suggesting either
that extended duration of untreated symptoms is more
likely to be present in illnesses with a poor prognosis or
that susceptibility to relapse is reduced by early treatment.
The Scottish Schizophrenia Research Group (1989)
reported the results of a trial of maintenance treatment in
a subgroup of patients (« = 15) from the Scottish First-
Episode Schizophrenia Study who responded to drug
treatment during their index episode and were entered into
a double-blind study of either once-weekly pimozide or
intramuscular flupenthixol decanoate for 1 year. After pro-
viding consent, these patients, none of whom relapsed
during the first year, were then entered into a double-blind
study of active medication (once-weekly pimozide or
intramuscular flupenthixol decanoate) or placebo. Relapse
was defined as a deterioration in schizophrenic symptoms
or behavior sufficient to warrant the patient's withdrawal
from the study. None of the eight patients who received
active medication, but four of seven who received
placebo, were readmitted in the second year of treatment.
As part of a longitudinal study of first-episode psy-
chosis, Nuechterlein et al. (1994) described the relation-
ship between life events and illness exacerbation during a
first year of standardized outpatient antipsychotic treat-
ment and a subsequent 24-week double-blind, drug-
placebo crossover period. Patients (n = 106, 87 male)
were recruited from admissions to four public hospitals
and the outpatient service of a university medical center.
Inclusion criteria were recent onset of a psychotic disor-
der with symptoms lasting at least 2 weeks and a first psy-
chotic episode starting not more than 2 years before proj-
ect entry; age 18 to 45 years; an RDC diagnosis of
schizophrenia or schizoaffective disorder (mainly schizo-
phrenia); and Anglo-American, Native American, or
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
acculturated Hispanic or Asian background. Exclusion
criteria were a known neurological disorder; recent (less
than 6 months) significant and habitual substance abuse;
mental retardation (IQ < 70); and African-American
descent (these patients were excluded due to differences
in electrodermal conductivity from other groups, a biolog-
ical variable that was also being studied). Two-thirds of
the patients were neuroleptic naive before enrolling in the
study. Assessments included BPRS and Psychiatric
Assessment Scale (PAS; Krawieka et al. 1977), admin-
stered weekly; life-event interviews, monthly; and
Strauss/Carpenter social and work outcome scale (Strauss
and Carpenter 1972) and the UCLA social attainment
scale (Goldstein 1978), quarterly. Operational criteria
were used to define psychotic exacerbation, psychotic
relapse, and nonpsychotic relapse. Phase 1 of the study
was undertaken after 2 to 3 months of clinical stabiliza-
tion following inpatient discharge. Patients then received
12.5 mg of fluphenazine decanoate every 2 weeks for 12
months. During this period, 11 patients needed their dose
lowered due to side effects, and 6 patients were prescribed
antidepressants. Phase 2 of the study recruited those
patients who showed sufficient remission of psychotic
symptoms over the first year. These patients received
fluphenazine or placebo for 3 months and were then
crossed over to the other treatment. Patients maintaining
stability during this crossover period were invited to par-
ticipate in an open, drug-free treatment period. During
phase 1, independent life events and caregiver expressed
emotion were associated with an elevated risk of relapse.
Although the results from phase 2 are preliminary at this
time, life events appeared to play a lesser role in the pre-
diction of relapse during a medication-free period.
Conclusions
The limited number of available treatment studies of first-
episode psychotic patients, most of whom were diagnosed
with schizophrenia, strongly support the efficacy of
antipsychotic medication, both in the acute phase of the
illness and in relapse prevention in the first 2 years fol-
lowing stabilization. Compared to patients with long-
established illnesses, a greater percentage of first-episode
patients appear to respond to pharmacological treatment
(Lieberman et al. 1996), even though 10 to 20 percent of
patients may be resistant to conventional antipsychotic
medication at the time of first treatment. Consistent with
enhanced efficacy in reducing psychotic symptoms, first-
episode patients also appear more vulnerable to side
effects than chronic multiepisode patients.
Many investigators note that patients typically expe-
rience active psychosis for 12 to 24 months before obtain-
658
B.B. Sheitman et al.
ing treatment (Fenton 1997). Although data are not defini-
tive, the association between the duration of untreated
psychosis and time to and quality of remission (Loebel et
al. 1992) suggests that an active morbid process may be
operative in the early stages of schizophrenia (Wyatt
1991, 1995). In a prospective study of treatment response
in first-episode and chronic patients, Szymanski et al.
(1996) found that a longer duration of untreated illness
was associated with diminished reduction in positive
symptoms over 6-month and 2-year followup in both
patient groups. In the Nottingham study of predictors of
long-term outcome among first-hospitalized patients
(Harrison et al. 1996), longer duration of untreated illness
predicted poorer outcome at 13 years in the domains of
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
disability, psychopathology, and social functioning.
Waddington et al. (1995) found that muteness was the pri-
mary clinical correlate of initial duration of untreated psy-
chosis in a group of older chronically ill schizophrenia
patients. This symptom was viewed as the end stage in
increasing severity of the negative symptom of poverty of
speech; patients with a longer duration of untreated psy-
chosis were more likely to progress to this end stage. It is
possible that early identification and treatment of patients
at or before a first episode of psychosis may limit the
accrual of morbidity or alter the natural history of the ill-
ness (Wyatt 1995; McGlashan and Johannessen 1996).
Although antipsychotic treatment is clearly indicated
for an acute psychosis and reduces the vulnerability to
relapse over the year or two following a first episode, the
important question of how long to continue prophylactic
maintenance treatment is unclear. Many patients with
schizophrenia will ask to stop their medication as soon as
possible after they have recovered from an initial acute
episode. Despite a reduction in relapse rates with mainte-
nance treatment, data from maintenance studies that have
followed patients for up to 2 years indicate that a signifi-
cant (38% in the Crow et al. 1986 study) proportion of
patients will not relapse during this period. A reasonable
clinical approach must balance the possibility of the
accrual of morbidity through relapses with the desire of
most patients to discontinue medication. Maintenance
treatment for 1 to 2 years, followed by a very slow med-
ication taper (20% per month) in conjunction with psy-
chosocial treatment that emphasizes psychoeducation for
the patient and family, is likely the optimal recommenda-
tion (Frances et al. 1996). During medication reduction,
both the patient and his or her family should be encour-
aged to contact the psychiatrist about even a remote suspi-
cion that the patient may be experiencing such prodromal
symptoms of relapse as difficulty sleeping, withdrawal,
suspiciousness, or ideas of reference. In addition, the clin-
icians should see the patients with sufficient frequency to
Evaluation and Treatment of First-Episode Psychosis
detect signs of relapse, especially those patients who are
unable to identify such symptoms themselves. The risks
and benefits of medication continuation and withdrawal
should be fully discussed with patients and family mem-
bers before initiating medication withdrawal, and
informed consent should be obtained.
In our opinion, the novel antipsychotics currently
available (risperidone and olanzapine) and others soon to
be marketed (sertindole, quetiapine, and ziprasidone)
should be seriously considered as a first-line therapy for a
first-episode of psychosis. Enhanced efficacy in the treat-
ment of negative symptoms and a more favorable side-
effect profile, particularly less EPS, have been demon-
strated in chronic patient samples (Marder and Meibach
1994; Beasley et al. 1996) and might be expected to
improve medication compliance, allowing longer periods
of maintenance treatment in patients at the beginning of
their illness (Lieberman 1993a).
Unfortunately, clozapine's potential for hematologic
toxicity, along with the need for indefinite white blood
cell count monitoring, make this medication a less likely
first-line agent.
Further research is needed to clarify a variety of clini-
cal questions concerning first-episode psychosis for which
insufficient data are available. Among these are the fol-
lowing:
1. What is the natural history of disease before the
usual time of presentation to the hospital? That is, when
and how does the illness actually begin, and how can pro-
dromal symptoms be reliably distinguished from normal
behavioral vicissitudes?
2. Do psychotic symptoms develop gradually, inter-
mittently, or persistently?
3. What medication, at what dosage, over what dura-
tion of treatment is the best intervention in first-episode
psychosis?
4. Can reliable clinical or biological indicators of out-
come following drug discontinuation be identified?
5. What effect does early treatment have on long-term
course and outcome?
Carefully designed early-intervention and first-episode
treatment studies should inform these clinical questions
and provide a broadened empirical basis for reducing the
morbidity associated with psychotic illnesses.
References
American Psychiatric Association. DSM-FV: Diagnostic
and Statistical Manual of Mental Disorders. 4th ed.
Washington, DC: The Association, 1994.
659
Schizophrenia Bulletin, Vol. 23, No. 4,1997
Andreasen, N.C., and Olsen, S. Negative v positive schiz-
ophrenia: Definition and validation. Archives of General
Psychiatry, 39:789-794, 1982.
Beasley, C M . ; Tollefson, G.; Tran, P.; Satterlee, W.;
Sanger, T.; Hamilton, S.; and the Olanzapine HGAD Study
Group. Olanzapine versus placebo and haloperidol: Acute
phase results of the North American double-blind olanza-
pine trial. Neuropsychopharmacology, 14:111-123,1996.
Bromet, E.J.; Schwartz, J.E.; Fenning, S.; Geller, L.;
Jandorf, L.; Kovasznay, B.; Lavelle, J.; Miller, A.; Pato,
C; Ram, R.; and Rich, C. The epidemiology of psychosis:
The Suffolk County mental health project. Schizophrenia
Bulletin, 18(2):243-255, 1992.
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
Carpenter, W.T., Jr., and Kirkpatrick, B. The heterogeneity
of the long-term course of schizophrenia. Schizophrenia
Bulletin, 14{4):645-652, 1988.
Chakos, M.H.; Mayerhoff, D.I.; Loebel, A.D.; Alvir, J.M.;
and Lieberman, J.A. Incidence and correlates of acute
extrapyramidal symptoms in first episode of schizophre-
nia. Psychopharmacology Bulletin, 28(l):81-86, 1992.
Chouljian, T.L.; Shumway, M.; Balancio, E.; Dwyer, E.V.;
Surber, R.; and Jacobs, M. Substance use among schizo-
phrenic outpatients: Prevalence, course, and relation to
functional status. Annals of Clinical Psychiatry, 7(1):19-
24, 1995.
Crow, T.J.; MacMillan, J.F.; Johnson, A.L.; and
Johnstone, E.C. The Northwick Park study of first
episodes of schizophrenia: II. A randomized controlled
trial of prophylactic neuroleptic treatment. British Journal
of Psychiatry, 148:120-127, 1986.
DeQuardo, J.R.; Carpenter, C.F.; and Tandon, R. Patterns
of substance abuse in schizophrenia: Nature and signifi-
cance. Journal of Psychiatric Research, 28(3):267—275,
1994.
Dixon, L.; Haas, G.; Weiden, P.J.; Sweeney, J.; and
Frances, A.J. Drug abuse in schizophrenic patients:
Clinical correlates and reasons for use. American Journal
of Psychiatry, 148:224-230,1991.
Doran, A.R.; Breier, A.; and Roy A. Differential diagnosis
and diagnostic systems in schizophrenia. Psychiatric
Clinics of North America, 9(1): 17-33,1986.
Feighner, J.P.; Robins, E.; Guze, S.B.; Woodruff, R.A.;
Winokur, G.; and Munoz, R. Diagnostic criteria for use in
psychiatric research. Archives of General Psychiatry,
26:57-63,1972.
Fenton, W.S. Course and outcome in schizophrenia.
Current Opinion in Psychiatry, 10:40-44, 1997.
Frances, A.; Docherty, J.P.; and Kahn, D.A. Treatment of
schizophrenia. The Expert Consensus Guidelines Series.
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
Journal of Clinical Psychiatry, 57(Suppl. 12B):l-57,
1996.
Frances, A.; First, M.B.; and Pincus, H.A. DSM-1V
Guidebook. Washington DC: American Psychiatric Press,
1995.
Gift, T.E.; Strauss, J.S.; Harder, D.W.; Kokes, R.E.; and
Ritzier, B.A. Established chronicity of psychotic symp-
toms in first-admission schizophrenic patients. American
Journal of Psychiatry, 138(6):779-784, 1981.
Goldstein, M.J. Further data concerning the relationship
between premorbid adjustment and paranoid symptoma-
tology. Schizophrenia Bulletin, 4(2):236-243,1978.
Gupta, S.; Hendricks, S.; Kenkel, A.M.; Bhatia, S.C.; and
Haffke, E.A. Relapse in schizophrenia: Is there a relation-
ship to substance abuse? Schizophrenia Research,
20:153-156, 1996.
Guy, W., ed. ECDEU Assessment Manual for Psycho-
pharmacology, revised. Rockville, MD: U.S. Department
of Health, Education, and Welfare, Publication No.
(ADM) 76-338, 1976.
Hafner, H., and Nowotny, B. Epidemiology of early-onset
schizophrenia. European Archives of Psychiatry and
Clinical Neurosciences, 245:80-92, 1995.
Harrison, G.; Croudace, T.; Mason, P.; Glazebrook, C ;
and Medley, I. Predicting the long-term outcome of schiz-
ophrenia. Psychological Medicine, 26(4):697-705, 1996.
Johnstone, E.C.; Crow, T.J.; Johnson, A.L.; and
MacMillan, J.F. The Northwick Park Study of first
episodes of schizophrenia: I. Presentation of the illness
and problems relating to admission. British Journal of
Psychiatry, 148:115-120, 1986.
Kane, J.M.; Honigfeld, G.; Singer, J.; Meltzer, H.; and the
Clozaril Collaborative Study Group. Clozapine for the
treatment-resistant schizophrenic: A double-blind compar-
ison with chlorpromazine. Archives of General
Psychiatry, 45:789-796, 1988.
Kane, J.M.; Rifkin, A.; Quitkin, F.; Nayak, D.; and
Ramos-Lorenzi, J. Fluphenazine vs. placebo in patients
with remitted, acute first-episode schizophrenia. Archives
of General Psychiatry, 39:70-73,1982.
Kay, S.R.; Fiszbein, A.; and Opler, L.A. The Positive and
Negative Syndrome Scale (PANSS) for schizophrenia.
Schizophrenia Bulletin, 13(2):261-275, 1987.
Keshavan, M.S., and Schooler, N.R. First-episode studies
in schizophrenia: Criteria and characterization.
Schizophrenia Bulletin, 18(3):491-513,1992.
Kirch, D.G.; Keith, S.J.; and Matthews, S.M. Research on
first-episode psychosis: Report on a National Institute of
Mental Health workshop. Schizophrenia Bulletin,
18(2): 179-184, 1992.
660
B.B. Sheitman et al.
Kopala, L.C.; Fredrikson, D.; Good, K.P.; and Honer,
W.G. Symptoms in neuroleptic-naive, first episode schizo-
phrenia: Response to risperidone. Biological Psychiatry,
39:296-298, 1996.
Koreen, A.R.; Siris, S.G.; Chakos, M.; Alvir, J.;
Mayerhoff, D.; and Lieberman, J.A. Depression in first-
episode schizophrenia. American Journal of Psychiatry,
150(ll):1643-1648, 1993.
Krawieka, M.; Goldberg, D.; and Vaughn, M. A standard-
ized psychiatric assessment scale for rating chronic psy-
chotic patients. Acta Psychiatrica Scandinavica, 55:299-
308, 1977.
Lieberman, J.A.; Jody, D.; Alvir, J.; Ashtari, M.; Levy,
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
D.L.; Bogerts, B.; Degreef, G.; Mayerhoff, D.I.; and
Cooper, T. Brain morphology, dopamine, and eye-tracking
in first-episode schizophrenia. Archives of General
Psychiatry, 50:357-367, 1993a.
Lieberman, J.A.; Jody, D.; Geisler, S.; Alvir, J.; Loebel,
A.; Szymanski, S.; Woerner, M.; and Borenstein, M. Time
course and biologic correlates of treatment response in
first-episode schizophrenia. Archives of General
Psychiatry, 50:369-376, 19936.
Lieberman, J.A.; Koreen, A.R.; Chakos, M.; Sheitman,
B.; Woerner, M.; Alvir, J.; and Bilder, R. Factors influenc-
ing treatment response and outcome of first-episode schiz-
ophrenia: Implications for understanding the pathophysi-
ology of schizophrenia. Journal of Clinical Psychiatry,
9:5-9, 1996.
Lieberman, J.A.; Matthews, S.M.; and Kirch, D.G. First-
episode psychosis: Part II. Editor's introduction.
Schizophrenia Bulletin, 18(3):349, 1992.
Loebel, A.D.; Lieberman, J.A.; Alvir, J.; Mayerhoff, D.I.;
Geisler, S.H.; and Szymanski, S.R. Duration of psychosis
and outcome in first episode schizophrenia. American
Journal of Psychiatry, 149(9):1183-1188, 1992.
Marder, S.R., and Meibach, R.C. Risperidone in the treat-
ment of schizophrenia. American Journal of Psychiatry,
151(6):825-835, 1994.
Maurer, K., and Hafner, H. Methodological aspects of
onset assessment in schizophrenia. Schizophrenia
Research, 15(3):265-276, 1995.
May, P.; Tuma, A.H.; and Dixon, W.J. Schizophrenia—A
follow-up study of results of treatment: I. Design and
other problems. Archives of General Psychiatry,
33:474-478, 1976a.
May, P.; Tuma, A.H.; and Dixon, W.J. Schizophrenia: A
follow-up study of the results of five forms of treatment.
Archives of General Psychiatry, 38:776-784,1981.
May, P.; Tuma, A.H.; Yale, C ; Potepan, P.; and Dixon,
W.J. Schizophrenia—A follow-up study of results of treat-
Evaluation and Treatment of First-Episode Psychosis
merit: II. Hospital stay over two to five years. Archives of
General Psychiatry, 33:481^86, 1976b.
McGlashan, T.H., and Carpenter, W.T., Jr. Postpsychotic
depression in schizophrenia. Archives of General
Psychiatry, 33:231-239, 1976.
McGlashan, T.H., and Johannessen, J.O. Early detection
and intervention with schizophrenia: Rationale. Schizo-
phrenia Bulletin, 22(2):201-222, 1996.
Nuechterlein, K.H.; Dawson, M.E.; Gitlin, M.; Ventura,
J.; Goldstein, M.J.; Snyder, K.S.; Yee, CM.; and Mintz, J.
Developmental processes in schizophrenic disorders:
Longitudinal studies of vulnerability and stress.
Schizophrenia Bulletin, 18(3):387^t25, 1992.
Nuechterlein, K.H.; Dawson, M.E.; Ventura, J.; Gitlin,
M.; Subotnik, K.L.; Snyder, K.S.; Mintz, J.; and
Bartzokis, G. The vulnerability/stress model of schizo-
phrenic relapse: A longitudinal study. Ada Psychiatrica
Scandinavica, 89(Suppl. 382):58-64, 1994.
Overall, J.E., and Gorham, D.R. The Brief Psychiatric
Rating Scale. Psychological Reports, 10:799-812, 1962.
Scottish Schizophrenia Research Group. The Scottish first
episode schizophrenia study: I. Patient identification and
categorisation. British Journal of Psychiatry, 150:331-
333, 1987a.
Scottish Schizophrenia Research Group. The Scottish first
episode schizophrenia study: II. Treatment—Pimozide
versus flupenthixol. British Journal of Psychiatry,
150:334-338,1987b.
Scottish Schizophrenia Research Group. The Scottish first
episode schizophrenia study: VII. Two-year follow-up.
Acta Psychiatrica Scandinavica, 80:597-602, 1989.
Simpson, G.M., and Angus, J.W.S. A rating scale for
extrapyramidal side effects. Acta Psychiatrica
Scandinavica, 212(Suppl.):ll-19, 1970.
Simpson, G.M.; Lee, J.H.; Zoubok, B.; and Gardos, G. A
rating scale for tardive dyskinesia. Psychopharmacology,
4:171-179,1979.
Siris, S.G. Diagnosis of secondary depression in schizo-
phrenia: Implications for DSM-1V. Schizophrenia
Bulletin, 17(l):75-97,1991.
Spitzer, R.L.; Endicott, J.; and Robins, E. Research
Diagnostic Criteria: Rationale and reliability. Archives of
General Psychiatry, 35(6):773-782, 1978a.
Spitzer, R.L.; Endicott, J.; and Robins, E. The Schedule
for Affective Disorders and Schizophrenia—Change
Version (SADS-C). New York: Biometrics Research Unit,
New York State Psychiatric Institute, 1978b.
Strauss, J.S., and Carpenter, W.T., Jr. The prediction of
661
Schizophrenia Bulletin, Vol. 23, No. 4, 1997
outcome in schizophrenia: I. Characteristics of outcome.
Archives of General Psychiatry, 27:739-746, 1972.
Susser, E.; Fennig, S.; Jandorf, L.; Amador, X.; and
Bromet, E. Epidemiology, diagnosis and course of brief
psychosis. American Journal of Psychiatry, 152:1743-
1748, 1995.
Szymanski, S.; Cannon, T.; Gallacher, R; Erwin, R.J.; and
Gur, R.E. Course and treatment response in first episode
and chronic schizophrenia. American Journal of
Psychiatry, 153:519-525,1996.
Szymanski, S.; Masiar, S.; Mayerhoff, D.; Loebel, A.;
Geisler, S.; Pollack, S.; Kane, J.; and Lieberman, J.
Clozapine response in treatment-refractory first-episode
schizophrenia. Biological Psychiatry, 35:278-280,1994.
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at NERL on May 30, 2015
Tohen, M.; Stoll, A.L.; Strakowski, S.M.; Faedda, G.L.;
Mayer, P.V.; Goodwin, D.C.; Kilbrener, M.L.; and
Madigan, A.M. The McLean first-episode psychosis proj-
ect: Six-month recovery and recurrence outcome.
Schizophrenia Bulletin, 18(2):273-282, 1992.
Waddington, J.L.; Youssef, H.A.; and Kinsella, A.
Sequential cross-sectional and 10-year prospective study
of severe negative symptoms in relation to duration of ini-
tially untreated psychosis in chronic schizophrenia.
Psychological Medicine, 25:849-857, 1995.
Wing, J.K.; Cooper, J.E.; and Sartorius, N. The
Description and Classification of Psychiatric Symptoms:
An Instruction Manual for the PSE and Catego Systems.
Cambridge, MA: Cambridge University Press, 1974.
World Health Organization. Mental Disorders: Glossary
and Guide to Their Classification in Accordance With the
Ninth Revision of the International Classification of
Diseases. Geneva, Switzerland: The Organization, 1978.
Wyatt, R.J. Neuroleptics and the natural course of schizo-
phrenia. Schizophrenia Bulletin, 17(2):325-351, 1991.
Wyatt, R.J. Early intervention for schizophrenia: Can the
course of illness be altered? Biological Psychiatry, 38:1—
3,1995.
Yung, A.R., and McGorry, P.D. The prodromal phase of
first-episode psychosis. Schizophrenia Bulletin,
22(2):353-370,1996.
The Authors
Brian B. Sheitman, M.D., is Assistant Professor of
Psychiatry, and Jeffrey A. Lieberman, M.D., is Professor
of Psychiatry, University of North Carolina, Chapel Hill,
NC. Heidi Lee, M.D., and Rael Strauss, M.D., are
Research Fellows, Hillside Hospital, Long Island Jewish
Medical Center, Glen Oaks, NY.