S216 Vol. 5 (3C) n March 2005

ABSTRACT

In recent years, several guidelines and algo-rithms have appeared to help clinicians address theissues of best practice in the management of psy-chiatric disorders. As the standard of care hasmoved from the use of first-generation antipsychot-ic medications to the use of atypical antipsychoticsin the treatment of schizophrenia and bipolar dis-order, clinicians are faced with the challenge ofwhen to change medications in patients and whichmedications to choose. And, with the widespreaduse of atypical antipsychotics has come anincreased concern with the evaluation, treatment,and possible prevention of metabolic syndromeand other complications associated with antipsy-chotic pharmacotherapy.

What is the standard of care in treating schizo-phrenia and bipolar disorders? How can clini-cians optimize outcomes in treating seriousmental illness across the lifetime of the patient?What should guide the clinician in choosing atherapeutic agent? And, if complications arise,how can the clinician move confidently from oneagent to another? This article presents a summa-ry of the recent literature on treatment responseand recommended interventions in individualswho present with psychotic illness, paying partic-ular attention to strategies that optimize treatmentand outcomes in the stable phase. It addressesthe clinical issues that bear on the selection

of antipsychotic agents and examines the process ofswitching from one antipsychotic to another. Withthe introduction of second-generation antipsychoticagents has come a change in the concept of stabili-ty and a redefinition of the goals for the stable phaseof illness. This article reviews the process of assess-ment of ongoing treatment, considering what goalsare appropriate, and when to revise treatmentgoals. Finally, this article examines new assessmenttools and collaborative treatment models.(Adv Stud Med. 2005;5(3C):S216-S229)

Schizophrenia and bipolar disorders are life-long illnesses that can have devastatingeffects on the social, occupational, and inter-personal lives of patients and families.1,2

These disorders are associated with high rates of sui-cide, serious medical problems, and enormous eco-nomic burden to patients, families, and to society as awhole.1,2 Treatment across the life cycle involves phar-macotherapy and a coordinated program of psycho-logical interventions, family education and support,and social and vocational rehabilitation.1,2

In recent years, several guidelines have been devel-oped for the treatment of schizophrenia and bipolardisorder to provide clinicians with recommendationson treatment based on empirical data and on expertknowledge. Among these guidelines are the AmericanPsychiatric Association’s (APA) practice guidelines forthe treatment of patients with schizophrenia1 andbipolar disorder2; the Expert Consensus GuidelineSeries3,4; the Schizophrenia Patient Outcomes ResearchTeam treatment recommendations5; the TexasMedication Algorithm Project6,7; and other guide-lines.8,9 The guidelines differ in scientific rigor, com-prehensiveness, and clinical applicability, but taken

REVIEW

THE STANDARD OF CARE IN TREATING PSYCHIATRIC DISORDERS

—

Peter J. Weiden, MD,* and David G. Daniel, MD†

*Professor of Psychiatry, Department of Psychiatry,SUNY Downstate Medical Center, Brooklyn, New York.

†President, Bioniche Development, Inc., ClinicalProfessor, Psychiatry and Behavioral Sciences, GeorgeWashington University, Washington, DC.

Address correspondence to: Peter Weiden, MD,Professor of Psychiatry, Department of Psychiatry, SUNYDownstate Medical Center, Box 1203, 450 ClarksonAvenue, Brooklyn, NY 11203. E-mail: pjweiden@msn.com.

Advanced Studies in Medicine n S217

REVIEW

together, they provide guidance to the clinician basedon available evidence and expert opinion in those areaswhere scientific evidence is limited.10

TREATMENT MODELS

The guidelines for the treatment of disorders suchas schizophrenia and bipolar disorder address thecourse of illness, models of vulnerability and coping,and dimensions of outcome. Although there are lessdata available on the treatment of bipolar disorders,many of the same principles apply. Guidelines empha-size the following goals of treatment: to reduce oreliminate symptoms; to maximize the quality of lifeand adaptive functioning of the patient; and to pro-mote and maintain recovery from the long-termeffects of illness as much as possible. To achieve thesegoals, the psychiatrist must create a supportive thera-peutic alliance with the patient to learn not only aboutthe patient’s symptoms and concerns but also abouthis or her goals and hopes, thus developing a collabo-rative treatment plan, and to maximize the likelihoodof adherence to medication.1 For an optimal outcome,the psychiatrist must identify targets of treatment, useoutcome measures to clarify progress, and be preparedto reassess the treatment plan as the patient’s situationchanges or as new therapeutic tools become available.

MODEL OF ILLNESS AND OUTCOME MEASURES

A useful conceptualization of maintenance treat-ment of schizophrenia is based on a model of vulnera-bility, stress, and coping,11-13 which posits amultifactorial predisposition leading to an increasedrisk of symptomatology. Interaction between neu-rocognitive processes (such as deficits in informationprocessing or executive functioning) and environmen-tal stressors (such as high expressed emotion in thefamily or untoward events at key developmental mile-stones) can lead the patient from a subsyndromal statetoward manifestation of illness. Biological, psycholog-ical, and social protective factors, such as antipsychot-ic medication, good coping strategies, and a calm andstructured environment, may moderate the emergenceof clinical symptoms. This biopsychosocial model sup-ports an approach to treatment that integrates phar-macologic treatment with evidence-based psychosocialinterventions (such as family psychoeducation, socialskills training, and cognitive behavior therapy).

The biopsychosocial model also points toward sev-eral areas that should be considered in assessments oftreatment efficacy and outcome.14 These areas includeclinical dimensions, such as positive and negativesymptoms, depression and anxiety, and side effects ofmedication; rehabilitative dimensions, such as social,vocational, and interpersonal functioning; “humani-tarian” issues, such as quality of life and satisfactionwith treatment; and issues relating to the public wel-fare, such as safety and cost of treatment.

COURSE OF ILLNESS

The treatment plan will vary according to the needsof the patient at a particular time within the longitudi-nal course of illness. Over a lifetime, psychotic illnesswaxes and wanes and generally follows an unsteady andoften unpredictable pattern.15 The course of an episodehas been divided into the acute phase, the stabilizationphase, and the stable phase.1 Treatment planning reflectsthese divisions. Although there is evidence that earlytreatment may reduce morbidity and allow a better qual-ity of life for patients and families, many patients do notreceive adequate psychiatric help until 12 to 24 monthsafter their symptoms first appear.16-18 Therefore, the sub-syndromal or prodromal stage has become the focus ofearly intervention programs that are studying the impactof psychoeducation, cognitive behavior therapy, andantipsychotic medication.

THE ACUTE PHASE

In the acute phase, the goals of treatment are “toprevent harm, control disturbed behavior, reduce theseverity of psychosis and associated symptoms (eg, agi-tation, aggression, negative symptoms, and affectivesymptoms), determine and address the factors that ledto the occurrence of the acute episode, effect a rapidreturn to the best level of functioning, develop analliance with the patient and family, formulate short-and long-term treatment plans, and connect thepatient with appropriate aftercare in the community.”1

Treatment with antipsychotic medication is funda-mental to achieving these aims. Atypical antipsy-chotics available in the United States at this timeinclude aripiprazole, clozapine, olanzapine, quetia-pine, risperidone, and ziprasidone. Because of con-cerns about agranulocytosis, clozapine is notrecommended as a first-line therapy and is treated sep-arately from the other medications in all guidelinesand treatment recommendations. The practice guide-

S218 Vol. 5 (3C) n March 2005

REVIEW

lines of the APA1 divide antipsychotic medication into4 groups:

• Group 1, first-generation agents: chlorpro-mazine, fluphenazine, and haldoperidol;

• Group 2: risperidone, olanzapine, quetiapine,ziprasidone, aripiprazole;

• Group 3: clozapine; and• Group 4: long-acting injectable antipsychotic

agents.

The APA guidelines suggest that group 2 is the“default” treatment for most patients needing treatmentfor an acute psychotic episode and support the choice ofclozapine for persistent suicidal ideation or behavior, per-sistent hostility and aggressive behavior, or for patientswith tardive dyskinesia. Although some meta-analysessuggest that the choice of atypical over first-generationantipsychotics is unsettled,19-21 there is growing consensusabout the use of atypical antipsychotics as first-line drugsin the treatment of first-episode psychosis,1,3 in additionto in the treatment for recurrent psychosis.8 The BritishNational Institute for Clinical Excellence recommendedto the English and Welsh National Health Services thatatypical antipsychotics should be considered alongsidefirst-generation antipsychotics as one of the first-choiceoptions to treat people with newly diagnosed schizo-phrenia.9 The Texas Medication Algorithm Project rec-ommends using atypical antipsychotics for first-episodepatients or those patients never before treated with atyp-ical antipsychotics, and recently revised their recommen-dations to include aripiprazole and ziprasidone amongthose first-line agents.6,7

TREATMENT IN THE STABILIZATION AND STABLE PHASES

The APA defines these goals of treatment in the sta-bilization phase, between acute illness and stability: “toreduce stress on the patient and provide support to min-imize the likelihood of relapse, enhance the patient’sadaptation to life in the community, facilitate continuedreduction in symptoms and consolidation of remission,and promote the process of recovery.”1 The primary aimin the stable phase is to prevent relapse. Relapse preven-tion, once achieved, permits the patient to focus on main-taining or improving his or her level of functioning andquality of life and ensure that increases in symptoms orrelapses are effectively treated and clinical monitoring foradverse treatment effects continues.1 These goals includeestablishing effective control of positive symptoms, suchas delusions and hallucinations, self-injury and suicidal or

homicidal ideation, and aggressive or violent behavior.Additional concerns include the mitigation of negativesymptoms, depression, anxiety, and disorganizedthoughts and the attenuation of social withdrawal. It isessential to identify and address complications, such aspoor medication adherence or comorbid substance abuse,both of which are common barriers to recovery. Otherpsychosocial issues, such as homelessness, social andoccupational functioning, and involvement with thecriminal justice system, must be addressed.

THE WHOLE PATIENT AND THE RECOVERY MODEL

Stability is often defined as the absence of relapse.Used as such, the concept of stability does not tell usanything about most of the other essential features ofoutcome. In particular, although a patient may notshow an acute exacerbation of the illness, they maycontinue to experience symptom breakthrough,impairment in relationships, and side effects frommedications. This phase generally marks a period ofongoing engagement in pharmacologic treatment andpsychosocial programs that support activities of dailyliving. In the past, these were reasonable maintenancegoals in the treatment of chronic psychiatric illness.However, this view is a long way from recovery. Therecovery model has emerged in the context of second-generation antipsychotic medications as an organizingprinciple in thinking about aims of treatment.Recovery should not be misunderstood as cure becausecure is not a realistic expectation at this time. Instead,the recovery model aims not only to prevent relapseand control symptoms but to look at the patient’shopes and goals and to attempt to reintegrate thepatient into as full a life as possible. Recovery is notlinear nor is it a function of medication alone. It is aprocess involving medication evaluations and re-evalu-ation, psychosocial interventions, rehabilitation, andpsychological therapy to address patients’ needs asthose needs change. We do not know how to predictwhich patients will recover, or how much they willrecover, over how long a time. However, we do knowthat if expectations are low, there will be little incen-tive to try innovative therapies that may offer possiblebenefits. Therefore, the assessment of the “stable”patient should address a variety of domains:

• Are there persistent positive symptoms? How dothey impact the patient’s level of functioning? Dothey help (eg, voices may be welcome compan-

Advanced Studies in Medicine n S219

ions) or do they hurt (as when paranoia causessocial withdrawal)?

• Are there cognitive problems? Might they be causedby positive symptoms, sedation, anticholinergiceffects, or drug-drug interactions in polypharmacy?

• Are there persistent negative symptoms? Might theybe caused by positivesymptoms, extrapyrami-dal symptoms (EPS),such as akinesia, seda-tion, or social withdraw-al, in response to stigma?

• Are there persistentaffective symptoms? Arethey caused by depres-sion, EPS, or demoral-ization?

• Are there persistent problems with side effects ofmedication? What are the consequences of thoseside effects for the patient’s health and self-image?

• How do these symptoms and side effects affectthe patient’s quality of life? Does the patient haveclose relationships? How does the patient spendhis or her time? What goals does the patient careabout? What concerns do the patient and familyhave regarding the risks and benefits of changesin medication?22

The psychiatrist who treats a patient with thesequestions in mind will regularly review the currenttreatment plan not only in terms of recurrence ofsymptoms but also in terms of quality of life forpatients and families. In seeking to optimize patientoutcomes, the psychiatrist will consider the patient’slevel of function across many domains.

THE STABLE PHASE: PHARMACOLOGIC TREATMENT

Importantly, the prevention of relapse almostalways requires that the patient accept long-term phar-macologic treatment.23 Unfortunately, in practice,medication nonadherence is the biggest limitation,with postdischarge nonadherence rates as high as 90%in the first year,24-27 which in turn results in higher ratesof psychiatric and medical hospitalization and greateroverall cost.28,29 Although improved compliance withatypical antipsychotics has yet to be convincinglydemonstrated,30-31 if adherence is to increase, the med-ications used must be effective, safe, and free of sig-

nificant side effects.32 Traditional antipsychotic drugs areassociated with high rates of extrapyramidal side effectsand tardive dyskinesia, which occurs in approximately5% of patients after 1 year of antipsychotic treatment.33

Further, traditional antipsychotic drugs do not addressnegative symptoms or ameliorate the depression that

often accompanies chronicpsychotic illness. Atypicalantipsychotics offer the possi-bility of medication effectivein more domains and withmore tolerable side effects. Donew-generation antipsychoticsshow greater efficacy in theprevention of relapse?

PREVENTION OF RELAPSE: THE PRIMARY GOAL

Thus far, there are only a small number of long-termstudies of the effectiveness of atypical antipsychotics onrelapse in the literature. A combined analysis of largecontrolled trials found that risperidone was superior tohaloperidol for the control of positive symptoms, such ashallucinations and delusions.34 Based on this finding,Csernansky et al compared risperidone with haloperidolin a 1-year trial studying relapse rates in patients withschizophrenia and schizoaffective disorder.35 In a double-blind study at 40 sites, they randomly assigned 397 sta-ble adult outpatients with chronic schizophrenia orschizoaffective disorder to receive treatment with flexibledoses of risperidone or haloperidol for a minimum of 1year. Patients in the risperidone group were treated for amedian of 364 days at a mean dose of 4.9 mg per day;those patients in the haloperidol group were treated for amedian of 238 days at a mean dose of 11.7 mg per day.At the end of the study, 25.4% of patients in the risperi-done group had relapsed, significantly lower than the39.9% of relapse in the group treated with haloperidol.Survival curve estimates of the risk of relapse showed aneven more robust difference between the risperidonegroup and the haloperidol group (34% vs 60%; P<.001). In addition, significant differences betweenrisperidone- and haloperidol-treated subjects were seenin total scores on the Positive and Negative SyndromeScale (PANSS) and in 4 of 5 factor scores at the last studyrating. In the risperidone group, improvements frombaseline to 1 year or to last study rating were seen in totalscores and in positive symptoms, negative symptoms,disorganized thoughts, and in anxiety-depression; in thehaloperidol group, symptoms were not improved over

REVIEW

In seeking to optimize patient outcomes,the psychiatrist will consider the patient’s level of function across many domains.

“

”

S220 Vol. 5 (3C) n March 2005

baseline. In both groups, a large number of patients dis-continued treatment for reasons other than relapse,44.1% in the risperidone group and 52.7% in thehaloperidol group. Adverse events were experienced byalmost 90% of patients in both groups. Approximately20% of patients treated with haloperidol experiencedsomnolence, agitation, and hyperkinesia, whereasrisperidone-treated patients experienced a mean increasein body weight of 2.3 kg or 5.0 lbs.

In 2003, Leucht et al published a meta-analysis ofstudies of atypical antipsychotics in relapse prevention inschizophrenia.36 Restricting their analysis to trials with aminimum duration of 6 months, they looked at 6 ran-domized controlled trials comparing new-generationantipsychotic drugs with placebo and at 11 trials com-paring atypical antipsychotics with conventional antipsy-chotic agents. They examined relapse, overall treatmentfailure, and drop-out rates caused by adverse events instudies involving more than 3 000 patients using theantipsychotics amisulpride, clozapine, olanzapine,risperidone, sertindole, ziprasidone, and zotepine.Considered as a group, the new antipsychotics were sta-tistically significantly superior to placebo in relapse ratesand treatment failure. No significant difference wasfound with respect to adverse events. However, the drop-out rate was high across all studies: 43% of patients treat-ed with new antipsychotics compared to 72% of patientstreated with placebo left the studies early because ofrelapse, inefficacy of treatment, adverse events, or loss tofollow-up.

In the comparison of first- versus second-generationantipsychotics, 10 of 11 studies used haloperidol as thecomparator. Considered as a group, raw relapse rates forthe new antipsychotics were statistically significantlysuperior to conventional antipsychotics (risk difference,0.08), particularly when relapse rates estimated from sur-vival curves were analyzed. However, taken individually,only risperidone showed a significantly lower raw relapserate compared to conventional antipsychotics, and olan-zapine showed a lower relapse rate calculated from sur-vival curves. Treatment failure was significantly less forrisperidone and olanzapine individually compared toconventional antipsychotics and, when the results acrossall drugs were pooled, the atypical antipsychotics weresignificantly superior. Approximately 49% of 1314patients treated with new antipsychotics compared to66% of 669 patients treated with conventional antipsy-chotics left the studies early because of an undesirableoutcome. In individual and pooled results, there was no

significant superiority of new antipsychotics comparedto conventional antipsychotics in the number ofdropouts because of adverse events.

The authors point out that there are limitationsto a meta-analysis, which cannot distinguish amongdifferences in study design, and that in this studythe results for individual new drugs were pooled inan exploratory way because of the small number ofrelevant trials. In the studies using placebo as com-parator, only a few of the newer antipsychotics weretested, and patients were for the most part in thecontinuation rather than maintenance phase oftreatment. In studies comparing atypical versusconventional drugs in the prevention of relapse, themagnitude of the advantage for atypical drugs wasmodest. Additionally, most studies used doses ofhaloperidol greater than 5 mg/day, exceeding thedose used by many clinicians for control of psy-chosis and high enough to lead to extrapyramidalside effects in most patients. Clearly more long-term studies of relapse need to be done, involvingeach of the atypical antipsychotic drugs.Nonetheless, these studies thus far lend support tothe superiority of atypical antipsychotic medicationin relapse prevention, in addition to in tolerabilityand symptom suppression.

OTHER GOALS OF TREATMENT

In addition to the prevention of relapse, othermajor goals of treatment in the stable phase of psy-chosis include the improvement of cognitive impair-ment, the prevention or amelioration of negativesymptoms, and the reintegration of the patient into asfull a life as possible. It is in these domains that theatypical antipsychotics offer greater hope than tradi-tional antipsychotics. We will examine each of thesedomains individually.

NEUROCOGNITIVE FUNCTION

Cognitive impairment is an important determinantof functional outcome in schizophrenia and bipolar dis-order, and research is beginning to examine the effects ofnewer antipsychotic drugs on cognitive measures. Keefeet al, in a meta-analysis of 15 studies published between1990 and April 1998 (of which all but one concernedclozapine or risperidone), concluded that newer antipsy-chotics are significantly more effective than convention-al antipsychotics at improving cognitive function.37 Inparticular, verbal fluency, digit symbol substitution, fine

REVIEW

motor function, and executive function showed mostimprovement, whereas there was modest improvementin attention subprocesses and little improvement inlearning and memory. Subsequent reviews have con-firmed an improvement in verbal fluency and psy-chomotor speed on clozapine,38 an improvement inworking memory on risperidone,38,39 and general cogni-tive improvements in patients taking olanzapine andrisperidone compared with haloperidol.40 These earlystudies were characterized by methodological problemsand varying outcome measures that make generalizationsdifficult. Nonetheless, in a review of antipsychotic drugtreatment in 2001, Barnes and Joyce concluded that thenew antipsychotics, as a group, are superior to conven-tional drugs with respect to cognitive function and sug-gested that each of the atypical antipsychotics may havea distinct effect on cognitive function, which furtherresearch could illuminate.31

Subsequent studies have added support to the possi-ble benefits of the atypical antipsychotics on cognitivefunction. An open-label test in which patients wereswitched to olanzapine showed a statistically significantimprovement in 3 cognitive domains: immediate recall,category fluency, and time on Trail A and B makingtasks.41 Harvey et al showed improvement in global cog-nition in 6-week switch studies to ziprasidone from con-ventional antipsychotics olanzapine and risperidone.42

Although no study has shown that the atypical antipsy-chotics actually normalize cognitive function, futureresearch may better define and understand the areas ofcognitive function affected by specific atypical antipsy-chotic medications. At that point, it may be possible totailor drug regimens best suited to individual patients’needs. It is also important to recognize that 2 commonside effects of antipsychotic medications—anticholiner-gic side effects43,44 and sedation45—can also cause or exac-erbate cognitive dysfunction in schizophrenia.

NEGATIVE SYMPTOMS

Negative symptoms may represent a core feature ofschizophrenia and also may be secondary to paranoia,medication side effects, depression, anxiety, or environ-mental deprivation.1 Negative symptoms are already seenin some prodromal states, and their prevalence in first-episode schizophrenia is reportedly between 4% and10%.1 They are more prominent in males than females,and are more prominent with duration of illness.Treatment depends first on evaluating and addressingthose secondary causes that may contribute to the

appearance of negative symptoms. Although there areas yet no treatments of proven efficacy for primarynegative symptoms, the results of short-term studiesindicate that atypical antipsychotic drugs are moreeffective than haloperidol for the treatment of emo-tional blunting and social withdrawal.46-49 Amisulpride,quetiapine, olanzapine, risperidone,1 ziprasidone,50,51

and aripiprazole52 have all been reported to have effi-cacy against negative symptoms. Expert consensusguidelines consistently recommend the use of atypicalantipsychotics for patients with predominantly nega-tive symptoms or for those patients with a combina-tion of positive and negative symptoms.3,4 In additionto pharmacologic therapies, several studies have showna large aggregated effect of cognitive behavior therapyover supportive therapy for reducing negative symp-toms,53 and one study of family psychoeducationreported an improvement in negative symptoms.54

QUALITY OF LIFE

Research indicates that positive symptoms showa low correspondence with functional impairmentsamong patients with schizophrenia1; instead, it isthe cognitive impairment and negative symptomsthat correlate with functional impairment and con-sequent quality of life. In one recent study, switch-ing to an atypical antipsychotic resulted insignificant improvement for most patients in posi-tive symptoms, general psychopathology, and quali-ty of life.55 Clozapine has been associated withsignificant improvement in social and occupationalfunctioning.1 Because thus far there is limited evi-dence to support the efficacy of medications inimproving functional status, the major interven-tions remain psychosocial and rehabilitative.Supported employment, including individualizedjob development, rapid placement in competitiveemployment, ongoing job supports, and integratedvocational and mental health services, has beenfound effective in helping patients achieve compet-itive employment, although studies have found thatthese patients experience difficulty in retainingthese jobs because of cognitive impairments andother illness-related factors.1 Social skills traininghas also been found helpful in addressing difficul-ties in socialization and daily living skills.1 Self-helpgroups and advocacy organizations have also beenfound to be effective in increasing patients’ socialnetworks and quality of life.1

Advanced Studies in Medicine n S221

REVIEW

S222 Vol. 5 (3C) n March 2005

SWITCHING ANTIPSYCHOTICS: INDICATIONS AND RATIONALE

Given the potential benefits to patients in tolera-bility, symptom control, cognitive functioning, andquality of life, there are compelling reasons to consid-er switching patients from a traditional antipsychoticto an atypical, and to consider switching among atyp-icals to obtain the best possible outcome. The decisionto switch a patient is always an individualized assess-ment of potential benefit and risk. Relatively com-pelling factors to consider in switching fromtraditional to atypical antipsychotics include persistentpositive symptoms, persistent negative symptoms,relapse despite compliance, persistent EPS and/or tar-dive dyskinesia, and symptoms of hyperprolactinemia,such as galactorrhea and amenorrhea, in women andgynecomastia in men.56 Relatively strong tolerabilityfactors to consider for switching among atypicalantipsychotics include problems, such as lack of ade-quate response, and side effects from some of thenewer medications, such as obesity, dyslipidemia,residual EPS, or the development of metabolic syn-drome.50 In addition, switching may be indicated forissues related to the patient’s quality of life. Patientswho once had hopes for close relationships and worksatisfaction may have narrowed their social and occu-pational spheres enormously. Although it is importantnot to raise expectations unrealistically, it is equallyimportant to consider whether newer drugs may offera patient a greater chance at a fuller and more satisfy-ing life. Quality-of-life issues should be approachedrespectfully and cautiously with modest expectations,but they are important considerations to raise withpatients and families, along with discussions about thecontrol of symptoms. Patients and families may haveanxieties about any change in medication, fearing thatrelapse could occur or that any change, good or bad,will be destabilizing. And, indeed, change is destabiliz-ing. Potential risks and benefits of switching should bediscussed with patients, families, and with all membersof the treatment team, such as case managers and clin-ic staff, thus all involved can develop realistic expecta-tions and be alert to the changes that may ensue.Patients and families need reassurance that there willbe adequate supervision by psychiatrists and othersinvolved in the patients’ care, not only to managesymptom changes but also to address the patient’s feel-ings about the changes that occur.

How should clinicians approach the decision toswitch? Weiden et al has written extensively on thistopic and recommends the following general rules:

• First, allow an appropriate time frame for assessingsymptom response. For an acute episode, 3 to 6weeks is probably adequate; for chronic but stableoutpatients who are being switched for electivereasons, allow at least 3 months.

• Second, be aware of the complexity of assessing posi-tive symptoms. Effects of monotherapy can onlybe confidently assessed after 4 to 6 weeks becauseuntil that time the prior antipsychotic may exertsome activity, even if blood levels are negligible.An early improvement may be followed by wors-ening at 4 to 6 weeks, leading to disappointmentof patients and families. Alternatively, patientsmay appear to worsen but actually improve, ifthey begin to disclose symptoms that they hadhidden in the past, thus it is important to ask howlong the symptoms have been present.

• Third, be cautious in expecting change in negativesymptoms. Improvements may be partial and lim-ited and too subtle to be detected on standardassessment scales. Families and patients shouldhave modest expectations, and any change shouldbe considered a success.

• Fourth, be aware of changes in affective symptoms.Although there is good reason to think that newerantipsychotics will reduce the burden of depressivesymptoms in schizophrenia, postpsychotic depres-sion is still a risk, particularly if the patient has apast history of depression combined with a recenthistory of positive symptom improvements. In suchcases, aggressive management is called for, includ-ing increasing the level of psychosocial support andmonitoring, and adding antidepressant medication.

• Fifth, pay attention to changes in side effects.Patients switching from a typical to atypical med-ication can be expected to experience a reversal ofpersistent tremor, rigidity, akinesia, or akathisiaand may no longer need an anticholinergic agent.Because amenorrhea and galactorrhea will nor-malize over 3 to 6 months as prolactin normal-izes, it is important to prepare patients for theneed to use birth-control measures if they havenot been using them. Changes in tardive dyski-nesia cannot be predicted, but must be assessed at6 months after switching and yearly thereafter.Atypical antipsychotics, although less likely to

REVIEW

cause tardive dyskinesia than traditional drugs,are not entirely without risk, thus it is still neces-sary to obtain informed consent and monitor fortardive dyskinesia on a regular basis.57

CONTRAINDICATIONS TO SWITCHING

Important contraindications to switching relateprimarily to the risk of relapse, but they also includeother concerns. Elective switching in patients in themaintenance phase is con-traindicated if an exacerba-tion of psychotic symptomsin the near future would rep-resent an unacceptable risk ofdanger to the patient or oth-ers; if the patient is within 3to 6 months of recovery froman acute episode and remainson the medication successful-ly used to treat that episode; and if noncompliancewith oral medication is a problem, but compliance isgood with depot medications.56 Furthermore, if targetsymptoms can be altered more simply by changing thedose of current medication, that should be tried first.Expert consensus guidelines recommend maximizingthe dose of one atypical antipsychotic before switchingto another, except when switching from a typicalantipsychotic because of concerns about side effectsand tardive dyskinesia at higher doses.4 Finally, switch-ing should be attempted when the patient’s psychoso-cial situation is as stable as possible. If there is activesubstance abuse, psychosocial instability, or significantlife stressors such as a change in housing or job, it willbe difficult to evaluate the effects of a new drug.

Switching medications usually requires more fre-quent monitoring by psychiatrists and clinic staff, addi-tional costs of medication and increased visits, andpotentially increased demands on rehabilitation services.However, cost concerns may be offset by decreased hos-pitalization days and ultimately greater stability.58,59

SWITCHING ANTIPSYCHOTICS: HOW TO DO IT

Several studies have examined the process of switch-ing from one antipsychotic to another.50,52,60-63 The opti-mal switching strategy would minimize relapse ofpsychotic symptoms because of inadequate doses ofantipsychotic medication, avoid drug discontinuationsyndromes, and limit the potential for drug-drug inter-

actions characterized by sedation, confusion, and othercognitive problems, and difficulties with motor coordi-nation that may occur when one drug is being tapered atthe same time another is being added.62

SWITCHING STUDIES

In a study of patients switching from clozapine toolanzapine, Tollefson et al abruptly discontinued clozap-ine (mean dose 324 mg/day) in 106 subjects, randomly

assigning them to receive dou-ble-blind placebo or olanzap-ine 10 mg for 3 to 5 days, andthen continuing all subjects inan open-label olanzapine trialfor an additional 9 weeks.61

Statistically significantly moreplacebo-treated than olanzap-ine-treated patients (24.5% vs7.5%) experienced a discon-

tinuation-associated increase of at least one psychoticsign or symptom, most commonly delusions, hallucina-tions, hostility, agitation, and/or paranoia. These symp-toms were reflected in worsening PANSS subscale scoresin general psychopathology (most affected), negativesymptoms, and positive symptoms (least affected).Furthermore, the Montgomery-Asberg DepressionRating Scale indicated a substantial disturbance in mood(P <.001) in those patients assigned to the placebo group.Patients who experienced relapse or clinically significantdiscontinuation symptoms were entered into a crossoverarm of the trial, thus they received clozapine and olanza-pine for a period up to 28 days. Given the symptoms thatemerged with abrupt clozapine discontinuation, theauthors conclude that it is safest to gradually taper cloza-pine before discontinuation and to substitute an alterna-tive antipsychotic agent simultaneously, with a period ofoverlap, even though problems associated with immedi-ate olanzapine monotherapy were few.

Kinon et al compared 4 different methods of switch-ing partially remitted, clinically stable outpatients diag-nosed with schizophrenia or schizoaffective disorder whowere being treated with a conventional antipsychoticdrug or risperidone.62 More than 200 patients wereopenly randomly assigned to abrupt or gradual discon-tinuation of their prior antipsychotic drug, and furtherrandomly assigned in a double-blind fashion to immedi-ate versus stepwise treatment with olanzapine. Therewere 4 treatment paradigms: abrupt discontinuation,immediate olanzapine initiation, 10 mg/day for 3 weeks;

Advanced Studies in Medicine n S223

REVIEW

However, cost concerns may be offset by decreased hospitalization days and

ultimately greater stability.

“

”

S224 Vol. 5 (3C) n March 2005

abrupt discontinuation, stepwise initiation with week 1on placebo, week 2 on olanzapine 5 mg/day, and week 3on olanzapine 10 mg/day; gradual discontinuation,immediate olanzapine initiation, 10 mg/day for 3 weeks;and gradual discontinuation, stepwise initiation withweek 1 on placebo, week 2 on olanzapine 5 mg/day, andweek 3 on olanzapine 10 mg/day.

Patients were assessed using the Clinical GlobalImpressions (CGI) Improvement scale, Patient’s GlobalImpressions (PGI) Improvement scale, and PANSS. Nosignificant difference was seen in discontinuation ratesbetween the switching paradigms. The paradigm ofgradual antipsychotic drug discontinuation combinedwith an initial full dose of olanzapine 10 mg/day demon-strated at week 1 a statistically significant greatest meanchange in CGI and PGI improvement compared to theother paradigms. However, by week 3, there was no sig-nificant difference in improvement measures, and morethan 90% of completing patients on all 4 switching par-adigms were improved or clinically unchanged. No clin-ically significant differences between switchingparadigms were seen in laboratory values or vital signs.Patients in the abrupt discontinuation and immediatetreatment with 10 mg/day of olanzapine paradigmshowed the most sedation, whereas those patients whoexperienced not only abrupt discontinuation but 1 weekof placebo showed the most difficulty with sleep (P <.05for both groups). However, overall, the study demon-strated that switching could be achieved withoutincreased vulnerability to relapse or to occurrence of clin-ically burdensome antipsychotic drug withdrawal symp-toms in the majority of patients.

Weiden et al examined the safety and tolerability ofswitching to ziprasidone using 3 distinct switchingstrategies, looking at 203 patients on 3 different med-ications before the switch.50 Stable outpatients with per-sistent symptoms or troublesome side effects onconventional antipsychotic, olanzapine, or risperidonetherapy were switched to an open-label 6-week flexible-dose trial of ziprasidone. In the crossover to ziprasidone,all groups were started on open-label oral ziprasidone at40 mg twice daily for 2 days, followed by flexible dos-ing between 40 and 160 mg/day administered in divid-ed twice-daily doses. Patients were randomly assigned to1 of 3 schedules to discontinue their prior medicationduring the first week of ziprasidone therapy:

• Complete discontinuation of the previousantipsychotic the day before starting ziprasidone;

• Immediate dose reduction, with a 50% reduction

in the dosage of the previous antipsychotic for thefirst week of ziprasidone, followed by discontinu-ation starting week 2; or

• Delayed dose reduction, with a 50% reduction inthe dosage of the previous antipsychotic on day 4of ziprasidone, followed by discontinuation start-ing week 2.

All patients were on ziprasidone monotherapy afterthe first week of ziprasidone crossover. Baseline andoutcome assessments included PANSS and CGI sever-ity ratings. All 3 switching strategies were well tolerat-ed for all 3 patient groups. There was no statisticaldifference across study groups in completion rates or indiscontinuations because of inadequate response oradverse events. Among all patients in each of the 3 druggroups entering the study, there were no significant pair-wise differences between any 2 switching strategies inoutcome assessment of symptoms. The specific crossovertechnique used during the first week of ziprasidone ther-apy did not influence the outcome at 6 weeks in any ofthe medication groups. After 6 weeks on ziprasidonetherapy, significant (P <.05) improvements wereobserved on all major symptom measures and almost allsubscales for all switched subgroups.

Positive symptoms improved significantly forpatients switched from conventional antipsychotics orolanzapine (P <.05), and negative symptoms improvedsignificantly (P <.005) in all 3 switch groups.Tolerability of ziprasidone was good, with few patientsdiscontinuing treatment. For patients switched fromolanzapine, a significant reduction in mean bodyweight of -1.8 kg or -3.9 lb (P <.001) was observedafter 6 weeks on ziprasidone. A lesser but still signifi-cant decrease in weight was seen in patients switchedfrom risperidone (-0.9 kg or -1.9 lb; P <.05), but therewas no significant weight change in patients switchedfrom conventional antipsychotics. The authors con-clude that stable but symptomatic outpatients beingtreated with a conventional antipsychotic, olanzapine,or risperidone may experience further improvement byswitching to ziprasidone, but that there appears to beno difference in efficacy or tolerability between any ofthe 3 switching strategies. They caution that the 1-week overlap in this study is much shorter than iscommonly used in clinical practice, and that consen-sus guidelines on switching indicate that the preferredswitching method is to overlap the old and the newantipsychotic for several weeks.3,4

REVIEW

Casey et al investigated the efficacy, safety, and tol-erability of 3 dosing strategies for switching chronic,stable patients with schizophrenia from current oralantipsychotic monotherapy to once-daily oral ari-piprazole monotherapy.52 Approximately 247 patientsin this 8-week, open-label, out-patient study were ran-domized to 1 of the following strategies:

1) Immediate initiation of 30 mg/day aripiprazolewith simultaneous immediate discontinuationof current antipsychotic;

2) Immediate initiation of 30 mg/day aripiprazole,while tapering off current antipsychotic over 2weeks; or

3) Up-titrating aripiprazole to 30 mg/day over 2weeks, while simultaneously tapering off currentantipsychotic medication.

Patients entered the study on a stable dose ofhaloperidol, thioridazine, risperidone, or olanzapinefor at least 1 month before study entry. All had ade-quate clinical reason to consider a trial of an antipsy-chotic other than their current therapy. For patients ingroup 2 or 3, current antipsychotic medication wasdecreased by 50% during week 1, then decreased byanother 50% during week 2, and totally discontinued atthe start of week 3. Efficacy, using PANSS, CGI severi-ty, and CGI improvement scores, was maintained witharipiprazole during the study in all 3 groups. The over-all low incidence of adverse events was comparableacross all groups, as were discontinuations caused byadverse events. Body weight was reduced in all groupsbetween 1.3 and 1.7 kg over the 8-week study period,and plasma prolactin levels were reduced across allgroups after the switch to aripiprazole. The authors con-clude that any of the 3 strategies can be used safely forswitching patients to aripiprazole from antipsychoticmonotherapy, and that patients’ symptoms may contin-ue to improve after switching to aripiprazole.

SUMMARY

In summary, studies thus far have shown no distinctadvantage for any particular strategy in switching fromone antipsychotic to another. Clozapine represents a spe-cific exception to this generalization because its discon-tinuation syndrome can be managed best by a gradualtaper rather than abrupt cessation of the drug. Crossovertimes in the studies cited earlier in this section are short-er than those times normally used in clinical practice,and a more conservative approach of 1 to 2 weeks or

longer is recommended by experts and by clinical expe-rience4,64,65 to minimize abrupt transitions.

In switching for efficacy, it is difficult to predict whichpatients will do well with which new drug. Clozapineremains the gold standard, but has many problems asso-ciated with its use. Although there is no consensus onhow many first-line atypicals to try before clozapine, sev-eral guidelines recommend a trial of at least 2 first-lineatypicals. Of the first-line atypical antipsychotics, all areroughly equal in efficacy for positive symptoms. Datashowing improvement from one medication to anotherin switch studies should be viewed with caution. Switchstudies are open label, use patients who have not donewell on prior medications, and provide increased contactand intervention for study subjects. Therefore, the bene-fits cannot be interpreted to show that the postswitchmedication is better than the preswitch medication.Rather, the studies indicate that closely monitoredswitching represents a relatively low risk, particularly forpatients whose level of function is suboptimal or whosecurrent symptoms are persistent, and that there is reasonfor optimism that newer medications may offer patientsa better outcome in control of positive symptoms, nega-tive symptoms, or other outcome domains.

SWITCHING TO MANAGE SIDE EFFECTS

It is much easier to predict the outcome of switchingfor side-effect problems, as the side-effect profiles of theatypical antipsychotics are well established. In the past,most side-effect switches were because of EPS. However,most side-effect switches between atypical antipsychoticsnow occur because of weight gain and sedation, in addi-tion to more subtle residual EPS and a variety of less fre-quent concerns. In a review of side effects in 99 patients,akinesia, weight gain, anticholinergic problems, and sex-ual problems were each reported to cause moderate-to-severe distress by 30% to 40% of patients.66 Musclerigidity and akathisia were each reported by approxi-mately 20% of patients. The degree of benefit thatswitching affords will depend on the side-effect profile ofthe original medication compared to the side-effect pro-file of the new medication.

EXTRAPYRAMIDAL SYMPTOMS

DeNayer et al looked at EPS reduction in morethan 300 patients who were switched to quetiapinefrom olanzapine, risperidone, conventional antipsy-chotic, and haloperidol (dose <10 mg/day) monother-

Advanced Studies in Medicine n S225

REVIEW

S226 Vol. 5 (3C) n March 2005

apy.63 The degree of EPS reduction was statistically sig-nificant for all groups, but it was greatest in thehaloperidol group, less in the conventional antipsy-chotic group, still less in the risperidone group, andleast in the olanzapine group from baseline to week 12of the study. A more recent study suggests that theremay be continuing improvements in EPS with queti-apine over time, extending well beyond the improve-ments seen in the first 6 to 12 months.22,67

WEIGHT GAIN AND ASSOCIATED PROBLEMS

Weight gain has become a major focus of concernin patients taking atypical antipsychotics. It con-tributes to poor self-image and to numerous medicalproblems, including elevated cholesterol and triglyc-erides, both independent risk factors for cardiovasculardisease. Of greatest concern is the development ofmetabolic syndrome, a cluster of coronary heart dis-ease risk factors with common insulin resistance,defined by abdominal obesity, hypertriglyceridemia,low high-density lipoprotein cholesterol, high bloodpressure, and high fasting glucose.68 A comprehensivemeta-analysis of estimated antipsychotic-inducedweight change at 10 weeks showed that olanzapine andclozapine were associated with the greatest increases inweight (4.15 kg and 4.45 kg, respectively).69

Quetiapine and risperidone are associated with lessweight gain, and aripiprazole and ziprasidone are weightneutral. In fact, aripiprazole and ziprasidone have beenassociated with weight loss when patients are switchedfrom olanzapine and other antipsychotics. Following aswitch to aripiprazole from olanzapine, patients showeda mean weight loss of 2 kg (n = 169; P <.001), whereaspatients switching to aripiprazole from risperidoneshowed a mean weight loss of 0.6 kg (n = 106; P <.077).70 Patients switching from olanzapine to ziprasi-done showed a mean weight loss of 1.5 kg after 6 weeks(n = 104; P <.0001), whereas those switching fromrisperidone to ziprasidone showed a weight change ofapproximately 0.5 kg (n = 58; P <.05).50 In a study of185 patients followed over 58 weeks after switching toziprasidone from olanzapine, risperidone, and conven-tional antipsychotics, substantial reductions in bodyweight and body mass index (BMI) observed 6 weeksafter switching from olanzapine and risperidone wereprogressive and sustained over 1 year.71 Substantialreductions in serum cholesterol and trigylceridesobserved 6 weeks after switching to ziprasidone from all3 preswitch antipsychotics were sustained over 1 year.

It is clear that the atypical antipsychotics have verydifferent effects on weight, BMI, and lipid status, andthat these metabolic parameters require monitoringover the course of treatment. Current recommenda-tions72 (American Diabetes Association [ADA]/APAConsensus Statement, 2004) include obtaining base-line screening measures before the initiation of anyantipsychotic medication, including:

• A personal and family history of obesity, dia-betes, dyslipidemia, hypertension, or cardiovas-cular disease;

• Weight and height, thus a BMI can be calculated;• Waist circumference;• Blood pressure;• Fasting plasma glucose; and • Fasting lipid profile.

These assessments can determine if the patient isoverweight (BMI 25–29.9) or obese (BMI >30), hasprediabetes (ie, fasting plasma glucose of 100–125) ordiabetes (fasting plasma glucose of >126 mg/dL), hashypertension (>140/90), or dyslipidemia. If any ofthese conditions are identified, appropriate treatmentor referral should be initiated. Close collaborationwith primary care physicians can expedite the screen-ing and treatment process.

The patient’s weight should be reassessed at 4, 8, and12 weeks after initiating or changing antipsychotic thera-py, and quarterly thereafter. If a patient gains more than5% of his or her initial weight at any time during thera-py, clinicians should consider changing antipsychotics.Fasting plasma glucose, lipid levels, and blood pressureshould also be assessed 3 months after starting an antipsy-chotic medication. After that, blood pressure and plasmaglucose should be obtained annually or more frequently inthose patients who have a higher baseline risk. For thosepatients with a normal lipid profile, repeat testing shouldbe performed at 5-year intervals.

OTHER SIDE EFFECTS OF CONCERN

Decreases in serum prolactin have been reported inswitches to aripiprazole from haloperidol, risperidone,and (to a much smaller degree) olanzapine.70 Sedation,a common complaint, is seen most often with clozap-ine, olanzapine, and quetiapine, and less with arip-iprazole and ziprasidone. Other less common sideeffects that patients may experience with 1 atypicalantipsychotic may be reduced with switching toanother atypical.

REVIEW

FUTURE DIRECTIONS IN TREATMENT

EARLY INTERVENTION

There is increased interest in many countries in earlyrecognition and intervention in the initial prodromalphase of illness to decrease the“duration of untreated psy-chosis.”73 Although the conse-quence of delayed treatmentof psychosis is disruption insocial, educational, and occu-pational development, theissue of to what degree earlyintervention programs candelay or prevent symptomexpression is still an openquestion. In a recent study,ultra high-risk individuals who were administered a 6-month preventive intervention of low-dose risperidoneand cognitive behavior therapy had a significantly lowertransition rate to a first episode (9.7%) compared tothose patients receiving a needs-based intervention(35.7%); however, this difference was not significant at12 months.74 Several international research programs areattempting to define who is at risk, identify what pre-dicts transition to psychosis, evaluate the timing andchoice of interventions, and examine risk-benefit issuesin the early treatment of psychosis.13

MEDICAL CORMORBIDITIES

Recent epidemiologic and population studies haveshown increased morbidity and mortality from medicalillnesses in patients with serious and persistent mental ill-ness.75,76 Individuals with major psychiatric disorders,such as schizophrenia or bipolar disorder, face an elevat-ed mortality rate mostly because of higher rates of med-ical comorbidities, including obesity, diabetes mellitus,cardiovascular and pulmonary diseases, HIV infection,and cancer.77-79 The degree to which psychiatric factorsmay elevate medical comorbidity rates in these patientshas not been well studied or described. The recent meta-bolic recommendations of the ADA/APA have sparkedrenewed interest in behavioral, biological, and psychoso-cial factors associated with schizophrenia and bipolar dis-order that may contribute to comorbid medical illnesses.

PSYCHOSOCIAL TREATMENTS

Psychosocial treatments remain essential in the

treatment of chronic psychosis and its comorbid con-ditions, such as homelessness, joblessness, social isola-tion, substance use, and involvement in the criminaljustice system. However, in spite of evidence to sup-port their efficacy, many patients have no access to

these treatments. Economicarguments have been advancedto re-evaluate cost58,59,75 com-paring the costs of inade-quate treatment, includingincreased days of hospitaliza-tion, increased medical andpsychiatric care needs, andother psychosocial conse-quences of poorly controlledpsychosis, to the costs oftreatment with newer

antipsychotics and access to more robust psychosocialservices. However, funding for mental healthcare andhow to allocate limited resources is likely to remain anongoing policy issue of great relevance to clinicians,patients, and their families.

BEST PRACTICE

In spite of the development of evidence-basedguidelines and consensus recommendations, surveyshave shown that psychiatrists’ adherence to guideline-based outpatient treatment is only approximately50%.5 In community treatment centers, use of currentguidelines was found to be limited by deficiencies inthe medical record, nonstandardized progress notes,lack of computerized prescribing programs, and thestructure of the guidelines.76 In inpatient care, deci-sions about which patients are switched to atypicalantipsychotics depend strongly on factors such as insti-tutional practices, in addition to previous diseasecourse and healthcare use.77 If state-of-the-art treat-ment is to be widely used, clinicians must have accessto treatment plans with outcome measures that areeasy to use and facilitate comparisons among sequen-tial trials of medications.

REFERENCES

1. American Psychiatric Association. Practice guideline for thetreatment of patients with schizophrenia, second edition.Am J Psychiatry. 2004;161:1-56.

2. American Psychiatric Association. Practice guideline for thetreatment of patients with bipolar disorder (Revision). Am JPsychiatry. 2002;159:1-50.

Advanced Studies in Medicine n S227

REVIEW

The degree to which psychiatric factors may elevate medical comorbidity rates

in these patients has not been well studied or described.

“

”

S228 Vol. 5 (3C) n March 2005

3. McEvoy J, Scheifler P, Frances A. Expert consensus guide-line series: treatment of schizophrenia. J Clin Psychiatry.1999;60(suppl 11):3-80.

4. Expert consensus guideline series: optimizing pharmacolog-ic treatment of psychotic disorders. J Clin Psychiatry.2003;64(suppl 12):5-99.

5. Lehman AF, Steinwachs DM. The schizophrenia PatientOutcomes Research Team (PORT) treatment recommenda-tions. Schizophr Bull. 1998;24:1-10.

6. Miller AL, Chiles JA, Chiles JK, et al. The Texas MedicationAlgorithm Project (TMAP) schizophrenia algorithms. J ClinPsychiatry. 1999;60:649-657.

7. Miller AL, Hall CS, Buchanan RW, et al. The TexasMedication Algorithm Project antipsychotic algorithm forschizophrenia: 2003 update. J Clin Psychiatry.2004;65:500-508.

8. Marder SR, Essock SM, Miller AL, et al. The Mount Sinaiconference on the pharmacotherapy of schizophrenia.Schizophr Bull. 2002;28:5-16.

9. National Institute for Clinical Excellence (NICE): Guidance onthe use of newer (atypical) antipsychotic drugs for the treatmentof schizophrenia. Technol Appraisal Guidance. 2002:43.

10. Milner KK, Valenstein M. A comparison of guidelines forthe treatment of schizophrenia. Psychiatr Serv. 2002;53:888-890.

11. Nuechterlein KH, Dawson ME. A heuristic vulnerability/stress model of schizophrenic episodes. Schizophr Bull.1984;10:300-312.

12. Zubin J, Spring B. Vulnerability: A new view of schizophre-nia. J Abnorm Psychol. 1977;86:103-126.

13. Gaebel W, Janssen B, Riesbeck M. Modern treatment con-cepts in schizophrenia. Pharmacopsychiatry. 2003;36(suppl 3):S168-S175.

14. Lehman AF. Evaluating outcomes of treatments for personswith psychotic disorders. J Clin Psychiatry. 1996;57(suppl11):61-67.

15. Breier A, Schreiber JL, Dyer J, Pickar D. Course of illnessand predictors of outcome in chronic schizophrenia;Implications for pathophysiology. Br J Psychiatry.1992:161(suppl 18):38-43.

16. Loebel AD, Lieberman JA, Alvir JM, et al. Duration of psy-chosis and outcome in first-episode schizophrenia. Am JPsychiatry. 1992;149:1183-1188.

17. McGlashan TH, Johannessen JO. Early detection and inter-vention with schizophrenia: rationale. Schizophr Bull.1996;22:201-222.

18. Larsen TK, Johannessen JO, Opjordsmoen S. First-episodeschizophrenia with long duration of untreated psychosis:pathways to care. Br J Psychiatry Suppl. 1998;172:40-45.

19. Geddes J, Freemantle N, Harrison P, Bebbington P. Atypicalantipsychotics in the treatment of schizophrenia: systematicoverview and meta-regression analysis. BMJ. 2000;2:1371-1376.

20. Leucht S, Pitschel-Walz G, Abraham D, Kissling W. Efficacyand extrapyramidal side-effects of the new antipsychoticsolanzapine, quetiapine, risperidone, and sertindole com-pared to conventional antipsychotics and placebo. A meta-analysis of randomized controlled trials. Schizophr Res.1999;4:51-68.

21. Leucht S, Wahlbeck K, Hamann J, Kissling W. New gener-ation antipsychotics versus low-potency conventionalantipsychotics: a systematic review and meta-analysis.Lancet. 2003;361:1581-1589.

22. Weiden P, Aquila R, Standard J. Atypical antipsychoticdrugs and long-term outcome in schizophrenia. J ClinPsychiatry. 1996;57(suppl 1):53-60.

23. Schooler NR. Maintenance medication for schizophreniastrategies for dose reduction. Schizophr Bull. 1991;17:311-324.

24. Luchins D, Malan R, Hanrahan P, et al. Continuity of med-ication in a mental health system. Paper presented at:American Psychiatric Association Annual Meeting; 1997;San Diego, Calif.

25. Mahmoud RA, Engelhart L M, Oster G, et al. Risperidonevs. conventional antipsychotics: a prospective randomizednaturalistic effectiveness trial of outcomes in chronic schizo-phrenia. Paper presented at: American College ofNeuropsychopharmacology; 1997; Kamuela, Hawaii.

26. Mojtabai R, Lavelle J, Gibson PJ, et al. Gaps in use ofantipsychotics after discharge by first-admission patientswith schizophrenia, 1989-1996. Psychiatr Serv.2002;53:337-339.

27. Nageotte C, Sullivan G, Duan N, Camp PL. Medicationcompliance among the seriously mentally ill in a publicmental health system. Soc Psychiatry Psychiatr Epidemiol.1997;32:49-56.

28. Weiden P, Olfson M. Cost of relapse in schizophrenia.Schizophr Bull. 1995;21:419-428.

29. Gilmer TP, Dolder CR, Lacro JP, et al. Adherence to treat-ment with antipsychotic medication and health care costsamong Medicaid beneficiaries with schizophrenia. Am JPsychiatry. 2004;161:692-699.

30. Valenstein M, Blow FC, Copeland LA, et al. Poor antipsy-chotic adherence among patients with schizophrenia: med-ication and patient factors. Schizophr Bull. 2004;30:255-264.

31. Barnes TR, Joyce EM. Antipsychotic drug treatment: recentadvances. Curr Opin Psychiatry. 2001;14:25-37.

32. Geddes J. Prevention of relapse in schizophrenia. N Engl JMed. 2002;346:56-58.

33.Chakos MH, Alvir JM, Woerner MG, et al. Incidence andcorrelates of tardive dyskinesia in first episode of schizo-phrenia. Arch Gen Psychiatry. 1996;53:313-319.

34. Marder SR, Davis JM, Chouinard G. The effects of risperi-done on the five dimensions of schizophrenia derived byfactor analysis: combined results of the North American tri-als. J Clin Psychiatry. 1997;58:538-546.

35. Csernansky JG, Mahmoud R, Brenner R. A comparison ofrisperidone and haloperidol for the prevention of relapse inpatients with schizophrenia. N Engl J Med. 2002;346:16-22.

36. Leucht S, Barnes TR, Kissling W, et al. Relapse preventionin schizophrenia with new-generation antipsychotics: a sys-tematic review and exploratory meta-analysis of random-ized, controlled trials. Am J Psychiatry. 2003;160:1209-1222.

37. Keefe RS, Silva SG, Perkins DO, Lieberman JA. The effectsof atypical antipsychotic drugs in neurocognitive impairmentin schizophrenia: a review and meta-analysis. SchizophrBull. 1999;25:201-222.

38. Meltzer H, McGurk S. The effect of clozapine, risperidoneand olanzapine in cognitive function in schizophrenia.Schizophr Bull. 1999;25:233-256.

39. Green M, Marshall B, Wirshing W, et al. Does risperidoneimprove verbal working memory in treatment-resistant schiz-ophrenia? Am J Psychiatry. 1997;154:799-804.

40. Purdon SE, Jones BD, Stip E, et al. Neuropsychologicalchange in early phase schizophrenia during 12 months oftreatment with olanzapine, risperidone or haloperidol. ArchGen Psychiatry. 2000;57:553-559.

41. Stip E, Remington GJ, Dursun SM, et al. A Canadian multi-center trial assessing memory and executive functions inpatients with schizophrenia spectrum disorders treated with

REVIEW

olanzapine. J Clin Psychopharmacol. 2003;23:400-404. 42.Harvey PD, Meltzer H, Simson GM, et al. Improvement in

cognitive function following a switch to ziprasidone fromconventional antipsychotics, olanzapine, or risperidone inoutpatients with schizophrenia. Schizophr Res. 2004;66:101-113.

43. Minzenberg MJ, Poole JH, Benton C, Vinogradov S.Association of anticholinergic load with impairment of com-plex attention and memory in schizophrenia. Am JPsychiatry. 2004;161:116-124.

44. Velligan DI, Newcomer JW, Pultz J, et al. Does cognitivefunction improve with quetiapine in comparison to haloperi-dol? Schizophr Res. 2002;53:239-248.

45. Lauriello J. Sedation from clozapine. Schizophr Res. 1997.46. Kane J, Honigfeld G, Singer J, Meltzer H. Clozapine for

the treatment-resistant schizophrenic: a double-blind compar-ison with chlorpromazine. Arch Gen Psychiatry. 1988;45:789-796.

47. Marder SR, Meibach RC. Risperidone in the treatment ofschizophrenia. Am J Psychiatry. 1994;151:825-835.

48. Beasley CM Jr, Tollefson G, Tran P, et al. Olanzapine versusplacebo and haloperidol: acute phase results of the NorthAmerican double-blind olanzapine trial. Neuropsycho-pharmacology. 1996;14:111-123.

49. Arvanitis LA, Miller BG. Multiple fixed doses of “Seroquel”(quetiapine) in patients with acute exacerbation of schizo-phrenia: a comparison with haloperidol and placebo. BiolPsychiatry. 1997;42:233-246.

50. Weiden PJ, Simpson GM, Potkin SG, O’Sullivan RL.Effectiveness of switching to ziprasidone for stable butsymptomatic outpatients with schizophrenia. J ClinPsychiatry. 2003;64:580-588.

51. Simpson GM, Glick ID, Weiden PJ, et al. Randomized,controlled, double-blind multicenter comparison of the effica-cy and tolerability of ziprasidone and olanzapine in acutelyill inpatients with schizophrenia or schizoaffective disorder.Am J Psychiatry. 2004;161:1837-1847.

52. Casey DE, Carson WH, Saha AR, et al. Switching patientsto aripiprazole from other antipsychotic agents: a multicen-ter randomized study. Psychopharmacology (Berl).2003;166:391-399.

53. Rector NA, Beck AT. Cognitive behavioral therapy for schiz-ophrenia; an empirical review. J Nerv Ment Dis. 2001;189:278-287.

54. Dyck DG, Short RA, Hendryx MS, et al. Management ofnegative symptoms among patients with schizophreniaattending multiple-family groups. Psychiatr Serv.2000;51:513-519.

55.Cook PE, Goldberg JO, Van Lieshout RJ. Benefits of switch-ing from typical to atypical antipsychotic medications: alongitudinal study in a community-based setting. Can JPsychiatry. 2002;47:870-874.

56. Weiden PJ, Aquila R, Dalheim L, Standard JM. Switchingantipsychotic medications. J Clin Psychiatry. 1997;58(suppl10):63-72.

57. Weiden PJ, Aquila R, Emanuel M, Zygmunt A. Long-termconsiderations after switching antipsychotics. J ClinPsychiatry. 1998;59(suppl 19):36-49.

58. Rey JA. Antipsychotic therapy: a pharmacoeconomic perspec-tive. Am J Health Syst Pharm. 2002;59(suppl 8):S5-S9.

59. Nasrallah HA. Pharmacoeconomic implications of adverseeffects during antipsychotic drug therapy. Am J Health SystPharm. 2002;59(suppl 8):S16-S21.

60. Kinon BJ, Kane JM, Johns C, et al. Treatment of neuroleptic-resistant schizophrenic relapse. Psychopharmacol Bull.1993;29:309-314.

61. Tollefson GD, Dellva MA, Mattler CA, et al. Controlled,double-blind investigation of the clozapine discontinuationsymptoms with conversion to either olanzapine or placebo.J Clin Psychopharmacol. 1999;19:435-443.

62. Kinon BJ, Basson BR, Gilmore JA, et al. Strategies forswitching from conventional antipsychotic drugs or risperi-done to olanzapine. J Clin Psychiatry. 2000;61:833-840.

63. DeNayer A, Martin Jones A, Whiteford JL, et al. Switchingto quetiapine improves CGI and PANSS scores in patientswith schizophrenia. Poster presented at: CINP XXIIICongress; June 23-27, 2002; Montreal, Canada.

64. McGrath J, Emmerson WB. Fortnightly review: treatment ofschizophrenia. BMJ. 1999;319:1045-1048.

65. Ganguli R. Rationale and strategies for switching antipsy-chotics. Am J Health Syst Pharm. 2002;59(suppl 8):S22-S26.

66. Weiden PJ, Miller AL. Which side effects really matter:Screening for common and distressing side effects ofantipsychotic medications. J Psychiatr Pract. 2001;7:41-47.

67. Buckley PF. Efficacy of quetiapine for the treatment of schiz-ophrenia: a combined analysis of three placebo-controlledtrials. Curr Med Res Opin. 2004;20:1357-1363.

68. Ford ES, Giles WH, Dietz WH. Prevalence of the metabol-ic syndrome among US adults: findings from the thirdNational Health and Nutrition Examination Survey. JAMA.2002;287:356-359.

69. Allison DB, Mentore JL, Heo M, et al. Antipsychotic-inducedweight gain: a comprehensive research synthesis. Am JPsychiatry. 1999;156:1686-1689.

70. Casey D, Kujawa M, Carson WH, et al. Switching fromother antipsychotics to aripiprazole [abstract]. Int JNeuropsychopharmacol. 2002;5:S187.

71. Weiden PJ, Loebel A, Yang R, et al. Course of weight andmetabolic benefits one year after switching to ziprasidone.Program and Abstracts of the American Psychiatric AssociationInstitute on Psychiatric Service. Boston, Mass; 2004.

72. American Diabetes Association, American PsychiatricAssociation, American Association of ClinicalEndocrinologists, North American Association for the Studyof Obesity: Consensus development conference on antipsy-chotic drugs and obesity and diabetes. Diabetes Care.2004;27:596-601.

73. Larsen TK, Friis S, Haahr U, et al. Early detection and inter-vention in first-episode schizophrenia: a critical review. ActaPsychiatrica Scandinavica. 2001;103:323-334.

74. McGorry PD, Yung AR, Phillips LJ, et al. Randomized con-trolled trial of interventions designed to reduce the risk ofprogression to first episode psychosis in a clinical samplewith sub-threshold symptoms. Arch Gen Psychiatry.2002;59:921-928.

75. Buda M, Tsuang MT, Fleming JA. Causes of death in DSM-III schizophrenics and other psychotics (atypical group).Comparison with the general population. Arch GenPsychiatry. 1988;45:283-285.

76. Osby U, Brandt L, Correia N, et al. Excess mortality inbipolar and unipolar disorder in Sweden. Arch GenPsychiatry. 2001;58:844-850.

77. Kilbourne AM, Cornelius JR, Han X, et al. Burden of gener-al medical conditions among individuals with bipolar disor-der. Bipolar Disord. 2004;6:368-373.

78. Meyer JM, Nasrallah HA. Medical Illness andSchizophrenia. Washington, DC: American PsychiatricPress, Inc; 2003.

79. Green AI, Canuso CM, Brenner MJ, Wojcik JD. Detectionand management of comorbidity in patients with schizo-phrenia. Psychiatr Clin North Am. 2003;26:115-139.

Advanced Studies in Medicine n S229

REVIEW