Guideline Statements and Implementation

Benefits

In an individual with a possible psychotic disorder, a detailed assessment is important in establishing a diagnosis, recognizing co-occurring conditions (including substance use disorders, other psychiatric disorders, and other physical health disorders), identifying psychosocial issues, and developing a plan of treatment that can reduce associated symptoms, morbidity, and mortality.

Harms*

Some individuals may become anxious, suspicious, or annoyed if asked multiple questions during the evaluation. This could interfere with the therapeutic relationship between the patient and the clinician. Another potential consequence is that time used to focus on a detailed assessment (as outlined in the Practice Guidelines for the Psychiatric Evaluation of Adults, 3rd edition; American Psychiatric Association 2016a) could reduce time available to address other issues of importance to the patient or of relevance to diagnosis and treatment planning.

Patient Preferences

Although there is no specific evidence on patient preferences related to assessment in individuals with a possible psychotic disorder, clinical experience suggests that the majority of patients are cooperative with and accepting of these types of questions as part of an initial assessment.

Balancing of Benefits and Harms

The potential benefits of this guideline statement were viewed as far outweighing the potential harms. This recommendation is also consistent with the APA Practice Guidelines for the Psychiatric Evaluation of Adults, 3rd edition (American Psychiatric Association 2016a). The level of research evidence is rated as low because there is minimal research on the benefits and harms of assessing these aspects of history and examination as part of an initial assessment. Nevertheless, expert opinion suggests that conducting such assessments as part of the initial psychiatric evaluation improves the diagnosis and treatment planning in individuals with a psychiatric disorder. For additional details, see the Practice Guidelines for the Psychiatric Evaluation of Adults. For additional discussion of the research evidence, see Appendix C, Statement 1.

Differences of Opinion Among Writing Group Members

There were no differences of opinion. The writing group voted unanimously in favor of this recommendation.

Benefits

Clinical decision-making, including but not limited to diagnosis and treatment planning, requires a careful and systematic assessment of the type, frequency, and magnitude of psychiatric symptoms as well as an assessment of the impact of those symptoms on the patient’s day-to-day functioning and quality of life. Intuitively, and by analogy with other medical specialties in which treatment is guided by standardized measurement (e.g., of physiological signs or laboratory tests), the use of a systematic and quantifiable approach to assessment would seemingly produce better patient outcomes and greater standardization of care across patients. As electronic health records become more commonly used, electronic capture of quantitative measures can facilitate use of computerized decision-support systems in guiding evidence-based treatment, catalyzing additional improvements in outcomes and quality of care.

Use of a quantitative measure as part of the initial evaluation can establish baseline information on the patient’s symptoms and level of functioning and can help determine specific targets of treatment in the context of shared decision-making. When administered through paper-based or electronic self-report and as compared with a clinical interview, a quantitative measure may help the clinician to conduct a more consistent and comprehensive review of the multiplicity of symptoms that the patient may be experiencing. Using systematic measures may also increase the efficiency of asking routine questions and allow more time for clinicians to focus on symptoms of greatest severity or issues of most concern to the patient. Such measures may also facilitate collection of information from the patient’s family or other collateral informants on factors such as symptoms or functioning. When used on a longitudinal basis, quantitative measures can help determine whether nonpharmacological and pharmacological treatments are having their intended effect or whether a shift in the treatment plan is needed to address symptoms, treatment-related side effects, level of distress, functioning impairments, or potential for harm to the patient or others.

Without the use of a consistent quantitative measure, recall biases may confound the ability of patients and clinicians to compare past and current levels or patterns of symptoms and functioning. When patients have had substantial improvements in symptoms and functioning, it can be easy to focus on the improvements and overlook residual symptoms or side effects of treatment that are contributing to ongoing impairment or reduced quality of life. Thus, ongoing use of quantitative assessments may foster identification of residual symptoms or impairments and early detection of illness recurrence. Systematic use of quantitative measures can also facilitate communication among treating clinicians and can serve as a basis for enhanced management of populations of patients as well as individual patients. Although mobile apps may be capable of assisting with quantitative measurement, there is no current evidence on which to base recommendations about use of mobile apps in the treatment of schizophrenia.

Harms

The harms of using a quantitative measure include the time required for administration and review. The amount of time available for an initial psychiatric evaluation is typically constrained by clinician availability, cost, and other factors. Under such circumstances, time that is used to obtain quantitative measures could introduce harms by reducing time available to address other issues of importance to the patient or of relevance to clinical decision-making. Overreliance on quantitative measures may also cause other aspects of the patient’s symptoms and clinical presentation to be overlooked.

Some patients may view quantitative measures as impersonal or may feel annoyed by having to complete detailed scales, particularly if done frequently. If a patient feels negatively about quantitative measures, this could alter the therapeutic alliance. In addition, some patients may have difficulty completing self-report scales or may interpret questions incorrectly. Patients may also provide inaccurate information about their symptoms, and relying on inaccurate information can have a negative impact on clinical decision-making, including recommendations for treatment.

Systematic use of measures may require changes in workflow or staffing to distribute scales, increase time needed to review results with the patient, or lead to unreimbursed costs (e.g., to integrate measures into electronic health record systems, to pay to use copyrighted versions of scales).

Patient Preferences

Clinical experience suggests that the majority of patients are cooperative with and accepting of quantitative measures as part of an initial or subsequent assessment. Most patients will be able to appreciate the ways in which the use of quantitative measures will be of benefit to them. For example, in the testing of the DSM-5 Cross-Cutting Symptom Measure as part of the DSM-5 field trials, quantitative measures were found to be acceptable to patients (Clarke et al. 2014; Mościcki et al. 2013), and only a small fraction of individuals felt that measurement of symptoms would not be helpful to their treating clinician (Mościcki et al. 2013).

The fact that the clinician is using a systematic approach to address the patients’ symptoms and functioning sends a positive message that could improve the therapeutic relationship. Especially in developed countries, patients are used to and expect digital, computerized information exchange, including for health-related monitoring and communication. For these patients, the use of quantitative measures within the context of an electronic health record, mobile app, or other computerized technology may be more convenient.

Balancing of Benefits and Harms

The potential benefits of this guideline statement were viewed as far outweighing the potential harms. This recommendation is also consistent with Guideline VII, “Quantitative Assessment,” in the APA’s Practice Guidelines for the Psychiatric Evaluation of Adults, 3rd edition (American Psychiatric Association 2016a). Although quantitative measures have been used for reporting purposes as well as research, the level of research evidence for this recommendation is rated as low because it remains unclear whether routine use of these scales in clinical practice improves overall outcomes. Nonetheless, expert opinion suggests that use of quantitative measures will enhance clinical decision-making and improve treatment outcomes. For additional discussion of the research evidence, see Appendix C, Statement 2.

There is minimal research on the harms of using quantitative measures as part of the psychiatric evaluation as compared with assessment as usual. However, expert opinion suggests that harms of assessment are minimal compared with the benefits of such assessments in improving identification and assessment of psychiatric symptoms. For additional details, see the APA’s Practice Guidelines for the Psychiatric Evaluation of Adults, 3rd edition (American Psychiatric Association 2016a).

Differences of Opinion Among Writing Group Members

There were no differences of opinion. The writing group voted unanimously in favor of this recommendation.

Aims of Treatment Planning

The overarching aims of treatment planning are 1) to promote and maintain recovery, 2) to maximize quality of life and adaptive functioning, and 3) to reduce or eliminate symptoms. To achieve these aims, it is crucial to identify the patient’s aspirations, goals for treatment, and treatment-related preferences. Depending on prevailing state laws, psychiatric advance directives are one approach to encouraging patients to contemplate and state their preferences about treatment choices (Easter et al. 2017; Kemp et al. 2015; Shields et al. 2014; Wilder et al. 2010). For patients who have completed a psychiatric advance directive, wellness recovery action plan (Copeland 2000), or individualized crisis prevention or safety plan (Stanley and Brown 2012, 2019; Stanley et al. 2018), these documents will be important to review with the patient when crafting a person-centered approach to care. Discussions with the patient, other treating health professionals, family members, and others involved in the patient’s life can each be vital in developing a full picture of the patient and formulating a person-centered treatment plan, using shared decision-making whenever possible. The patient and others may express opinions about specific treatment approaches or identify practical barriers to the patient’s ability to participate in treatment, such as lack of insight, cognitive impairments, disorganization, or inadequate social resources.

Elements of the Treatment Plan

Depending on the clinical circumstances and input from the patient and others, a comprehensive and person-centered treatment plan will typically delineate treatments aimed at improving functioning, reducing positive and negative symptoms, and addressing co-occurring psychiatric symptoms or disorders. In each of these respects, it is essential to consider both nonpharmacological and pharmacological treatment approaches and recognize that a combination of nonpharmacological and pharmacological treatments will likely be needed to optimize outcomes. Other elements of the treatment plan may include the following:

• determining the most appropriate treatment setting

• delineating plans for addressing risks of harm to self or others (if present)

• addressing barriers to adherence

• engaging family members and others involved in the patient’s life

• providing information to patients, family members, and others involved in the patient’s life about treatment options, early symptoms of relapse, need for ongoing monitoring, coping strategies, case management services, and community resources, including peer-support programs

• incorporating goals of treatment related to

  • social support networks

  • interpersonal, family, or intimate relationships

  • parenting status

  • living situation

  • past trauma or victimization

  • school or employment

  • financial considerations, including disability income support, when indicated

  • insurance status

  • legal system involvement

• identifying additional needs for

  • history or mental status examination

  • physical examination (either by the evaluating clinician or by another health professional)

  • laboratory testing, imaging, electrocardiography (ECG), or other clinical studies (if indicated on the basis of the history, examination, and planned treatments)

• collaborating with other treating clinicians (including provision of integrated care) to avoid fragmentation of treatment efforts and assure that co-occurring substance use disorders and physical health conditions are managed

Engagement of Family Members and Other Persons of Support

Discussions with the patient, family members, and others will typically occur as part of the initial assessment (see Statement 1), and additional input will be needed as treatment proceeds and the treatment plan is updated. Family members and others involved in the patient’s life may also express specific concerns about the individual’s symptoms or behaviors, which, if present, should be documented and addressed. Most individuals welcome involvement of family members and other persons of support (Cohen et al. 2013; Drapalski et al. 2018; Mueser et al. 2015), and family members can be an important part of the care team. Family members can also be provided with educational materials or directed to organizations that offer education to family members and other persons of support (Mental Health America 2019; National Alliance on Mental Illness 2019a).

Strategies to Promote Adherence

Strategies to promote adherence are always important to consider when developing a patient-centered treatment plan (Ferrando et al. 2014). Maintaining adherence to treatment is often challenging (Acosta et al. 2012; Shafrin et al. 2016; Valenstein et al. 2006), and poor adherence is associated with poor outcomes, including increased risks of relapse, rehospitalization, suicidal and aggressive behaviors, and mortality (Bowtell et al. 2018; Cassidy et al. 2018; Goff et al. 2017; Hawton et al. 2005; Hui et al. 2018; Kishi et al. 2019; Leucht et al. 2012; Thompson et al. 2018; Tiihonen et al. 2018; Vermeulen et al. 2017; Witt et al. 2013). Treatment planning to address adherence will depend on the specific contributing factors and whether reduced adherence is related to medication use, missed appointments, or other aspects of treatment. Issues that may influence adherence include, but are not limited to, lack of awareness of illness, forgetting to take doses, difficulties managing complex regimens (e.g., due to cognitive impairment, frequency of doses, or number of medications), side effects that are of particular importance to the patient (e.g., weight gain, akathisia, sexual dysfunction, effects on cognition), financial barriers (e.g., cost, insurance coverage), perceived risks and benefits of treatment, insufficient understanding of medication benefits for symptoms that are important to the patient, ambivalence or suspiciousness of medications or treatment in general, lack of a perceived need for treatment (e.g., due to feeling good or not viewing self as ill; due to personal, religious, or cultural beliefs), co-occurring conditions (e.g., depression; alcohol, cannabis, or other substance use disorder), high levels of hostility, persecutory delusions, prior difficulties with adherence, prior experiences with treatment (e.g., effectiveness, side effects), limited geographic availability or accessibility of services, financial or insurance constraints on medications or visits, difficulties in the therapeutic relationship, lack of support from significant others for treatment, cultural or family beliefs about illness or treatment, and perceptions of stigma about having an illness and taking medication (Acosta et al. 2012; Ascher-Svanum et al. 2006; Czobor et al. 2015; Foglia et al. 2017; García et al. 2016; Haddad et al. 2014; Hartung et al. 2017; Hatch et al. 2017; Higashi et al. 2013; Kane et al. 2013; MacEwan et al. 2016a; Mueser et al. 2015; Pyne et al. 2014; Shafrin et al. 2016; Velligan et al. 2017; Volavka et al. 2016; Wade et al. 2017). Adherence with appointments can also be influenced by financial barriers, difficulties scheduling visits around work or school schedules, or issues with transportation or with childcare. Addressing these barriers as part of the treatment plan will require active collaboration and problem-solving between the clinician and patient, often with input from the patient’s family and others involved in the patient’s’ life (Mueser et al. 2015).

When assessing adherence, it is important to take a patient-centered approach in inquiring in a nonjudgmental way whether the individual has experienced difficulties with taking medication (Haddad et al. 2014). Obtaining information from patient diaries, patient-completed rating scales, pharmacy records, family members, or other collateral sources of information can be useful supplements to subjective patient reporting (Acosta et al. 2012; Haddad et al. 2014; Hatch et al. 2017; Kane et al. 2013). Tablet counts, monitoring using electronic pill bottle caps, and drug formulations with implanted sensors have also been used to assess adherence with antipsychotic medications (Acosta et al. 2012; Haddad et al. 2014). It can also be useful to obtain medication blood levels. Levels of clozapine have been best studied, but blood levels of other antipsychotic medications are also available. Although the utility of routine therapeutic monitoring is unclear for antipsychotic medications other than clozapine, blood levels may help in establishing whether a patient is taking the medication (Hiemke et al. 2018; Horvitz-Lennon et al. 2017; Lopez and Kane 2013; Lopez et al. 2017; McCutcheon et al. 2018; Predmore et al. 2018). Urine levels of antipsychotic medications can also be used to assess for adherence (Velligan et al. 2006).

In terms of enhancing adherence, a wide range of approaches have been tried. However, evidence on the most effective techniques remains limited (Hartung et al. 2017), and different approaches will likely be needed for different patients. A checklist that includes barriers, facilitators, and motivators for adherence has been developed and may be helpful in promoting discussion and identifying adherence-related factors in individual patients (Pyne et al. 2014). In addition to conducting ongoing monitoring of adherence as treatment proceeds, it can be helpful to focus on optimizing treatment efficacy, addressing side effects and concerns about treatment, adjusting dosing to minimize side effects while maintaining efficacy, providing information about the illness and its treatments, engaging in shared decision-making, fostering a strong therapeutic alliance, and engaging family members and other community and social supports, as appropriate (Acosta et al. 2012; Haddad et al. 2014; Hamann and Heres 2019; Kane et al. 2013; Rezansoff et al. 2017).

For some patients, the formulation of the antipsychotic medication may influence adherence (see Statement 4, Table 3). For example, rapidly dissolving tablets, oral concentrates, or LAI formulations may be preferable for patients who have difficulty swallowing pills or who are ambivalent about medications and inconsistent in swallowing them. For individuals who have difficulty remembering to take medication, LAI formulations of medications can be used, oral medication regimens can be simplified to reduce the number of pills or daily doses, watches or cell phone alarms can be used as reminders to take medications, pillboxes may be filled with the week’s medication, and family or significant others may be enlisted to assist with medication if cognitive impairments are present. Another approach that can be used to improve adherence is behavioral tailoring, which involves cuing oneself to take medications by incorporating adherence into one’s daily routine (Kreyenbuhl et al. 2016; Mueser et al. 2002). If financial issues with medications are affecting adherence, reassessment of the treatment regimen may be needed, or patients’ assistance programs may be pursued (e.g., through pharmaceutical company programs or a discount program such as GoodRx, www.goodrx.com). When a patient does not appear for appointments or is nonadherent in other ways, assertive outreach such as telephone calls or secure messages may be helpful in reengaging the patient in treatment.

Addressing Risks for Suicidal and Aggressive Behavior

Identifying risk factors and estimating risks for suicidal and aggressive behaviors are essential parts of psychiatric evaluation (American Psychiatric Association 2016a and as described in detail in Statement 1, subsection “Implementation”). Despite identification of these risk factors, it is not possible to predict whether an individual patient will engage in aggressive behaviors or attempt or die by suicide. However, when an increased risk for such behaviors is present, it is important that the treatment plan reevaluates the setting of care and implements approaches to target and reduce modifiable risk factors. Although demographic and historical risk factors are static, potentially modifiable risk factors may include poor adherence, core symptoms of schizophrenia (e.g., hallucinations, delusions), co-occurring symptoms (e.g., depression, hopelessness, hostility, impulsivity), and co-occurring diagnoses (e.g., depression, alcohol use disorder, other substance use disorders). Additional elements of the treatment plan can address periods of increased risk (e.g., shortly after diagnosis, during incarceration, subsequent to hospital discharge).

Addressing Tobacco Use and Other Substance Use Disorders

Individuals with schizophrenia have high rates of nicotine dependence (Centers for Disease Control and Prevention 2020; Cook et al. 2014; de Leon and Diaz 2005; Dickerson et al. 2018; Hartz et al. 2014; Myles et al. 2012; Smith et al. 2014; Wium-Andersen et al. 2015), cannabis use (Brunette et al. 2018; Hartz et al. 2014; Hunt et al. 2018; Koskinen et al. 2010; Nesvåg et al. 2015; Toftdahl et al. 2016), and use of alcohol and other substances (Brunette et al. 2018; Hartz et al. 2014; Hunt et al. 2018; Nesvåg et al. 2015; Toftdahl et al. 2016). Smoking is a major contributor to increased mortality in individuals with serious mental illness (Reynolds et al. 2018; Tam et al. 2016), and the adverse health consequences of smoking are well documented (Lariscy et al. 2018; U.S. Department of Health and Human Services 2014; Van Schayck et al. 2017). Thus, smoking cessation is recommended for any individual who smokes. Some studies have assessed smoking cessation approaches targeted to individuals with mental illnesses, but specific evidence in patients with schizophrenia is still limited (Sharma et al. 2017). Thus, smoking cessation approaches will typically follow guidelines for the general population (National Cancer Institute 2019; SAMHSA-HRSA Center of Excellence for Integrated Health Solutions 2018; Siu et al. 2015; Van Schayck et al. 2017; Verbiest et al. 2017). Although quit rates may be lower in individuals with schizophrenia than in the general population (Lum et al. 2018; Streck et al. 2020), health education and motivational interviewing approaches can be helpful in those who are ambivalent about stopping cigarette use (Levounis et al. 2017) or who have had prior unsuccessful attempts at smoking cessation.

Rates of cannabis use and other substance use are also increased among individuals with schizophrenia (Hartz et al. 2014; Hunt et al. 2018; Swartz et al. 2006). Cannabis use has been associated with an increased incidence (Nielsen et al. 2017; Vaucher et al. 2018) and earlier onset (Donoghue et al. 2014; Helle et al. 2016; Kelley et al. 2016; Large et al. 2011) of schizophrenia, and it may also contribute to a higher burden of symptoms (Carney et al. 2017; Oluwoye et al. 2018). Other substance use disorders are associated with a poorer prognosis in individuals with schizophrenia (Brunette et al. 2018; Conus et al. 2017; Weibell et al. 2017) and, as noted previously, can contribute to risk of suicide or aggressive behavior. Thus, it is important for the treatment plan to address substance use disorders when they are present. Screening, Brief Intervention, and Referral to Treatment (SBIRT) can be integrated into a range of clinical settings (Agerwala and McCance-Katz 2012; SAMHSA-HRSA Center of Excellence for Integrated Health Solutions 2019). Often, a comprehensive integrated treatment model is suggested in which the same clinicians or team of clinicians provide treatment for schizophrenia as well as treatment of substance use disorders. However, if an integrated treatment is unavailable, the treatment plan should address both disorders with communication and collaboration among treating clinicians. For patients who do not recognize the need for treatment of a substance use disorder, a stagewise motivational approach can be pursued (Catley et al. 2016; Levounis et al. 2017).

Addressing Other Concomitant Psychiatric Symptoms and Diagnoses

Depressive symptoms are common in individuals with schizophrenia and should be addressed as part of treatment planning. The approach to treating depression will be grounded in a careful differential diagnosis that considers the possible contributions of demoralization, negative symptoms of schizophrenia, side effects of antipsychotic medications, substance intoxication or withdrawal, physical health condition, or a co-occurring major depressive episode. Depressive symptoms that occur during an acute episode of psychosis often improve as psychotic symptoms respond to treatment.

Evidence on the use of antidepressants to treat depression in individuals with schizophrenia comes from multiple trials, many of which have small sample sizes or factors that increase the risk of bias in the findings (Dondé et al. 2018; Gregory et al. 2017; Helfer et al. 2016). Nevertheless, a meta-analysis suggests that the addition of antidepressant medications results in small beneficial effects on symptoms of depression, quality of life, and response rates as well as on positive symptoms, negative symptoms, and overall symptoms (Helfer et al. 2016). These effects were more prominent in patients with more severe depressive symptoms. Furthermore, antidepressant treatment did not appear to be associated with exacerbation of psychosis or significant differences in adverse effects (Helfer et al. 2016). Nonpharmacological treatments for depression in schizophrenia have been less well studied but could also be incorporated into treatment planning (Dondé et al. 2018; Opoka and Lincoln 2017).

Treatments for posttraumatic stress disorder (Brand et al. 2018; Sin et al. 2017b) and anxiety (Howells et al. 2017) in individuals with schizophrenia have been less well studied. Nevertheless, many individuals with schizophrenia will have experienced violent victimization (de Vries et al. 2019; Morgan et al. 2016; Roy et al. 2014) or childhood adversity (Bonoldi et al. 2013; Schalinski et al. 2019; Trotta et al. 2015; Varese et al. 2012), and the impact of these experiences needs to be considered as part of a patient-centered treatment plan (Center for Substance Abuse Treatment 2014). When anxiety symptoms are present in individuals with schizophrenia, the possible contributions of psychotic symptoms, medication side effects, substance intoxication or withdrawal, or co-occurring anxiety disorders may suggest an approach to treatment. Given the relative safety of adjunctive antidepressant medications in individuals with schizophrenia and depression, these medications may be considered if otherwise indicated to treat posttraumatic stress disorder or an anxiety disorder. On the other hand, studies on the use of benzodiazepines in schizophrenia are limited (Dold et al. 2012), and long-term use of benzodiazepines may be associated with increased risk of poorer outcomes, including side effects (Dold et al. 2013; Fond et al. 2018; Fontanella et al. 2016; Tiihonen et al. 2016) or misuse (Maust et al. 2019). Nonpharmacological treatments for posttraumatic stress disorder in individuals with schizophrenia have been less well studied but may have modest benefits and do not appear to have significant adverse effects as compared with usual care (Brand et al. 2018; Sin et al. 2017b).

In terms of the use of stimulants to treat preexisting attention-deficit/hyperactivity disorder in individuals with schizophrenia, available evidence is also very limited but suggests a potential for worsening of psychotic symptoms as well as potential for development of a stimulant use disorder (Sara et al. 2014; Solmi et al. 2019). Thus, if stimulant medications are used, monitoring for these possible adverse effects is warranted as part of the treatment plan.

Addressing Other Concomitant Health Conditions

As described in Statement 1, other health conditions are more frequent in individuals with serious mental illness in general (Firth et al. 2019; Janssen et al. 2015; McGinty et al. 2016) and schizophrenia in particular (Henderson et al. 2015). Such disorders include but are not limited to poor oral health (Kisely et al. 2015), hepatitis C infection (Chasser et al. 2017; Hauser and Kern 2015; Hughes et al. 2016), HIV infection (Hobkirk et al. 2015; Hughes et al. 2016), cancer (Olfson et al. 2015), sleep apnea (Myles et al. 2016; Stubbs et al. 2016b), obesity (Janssen et al. 2015), diabetes mellitus (Vancampfort et al. 2016a), metabolic syndrome (Vancampfort et al. 2015), and cardiovascular disease (Correll et al. 2017c). These disorders, if present, can contribute to mortality or reduced quality of life, and some may be induced or exacerbated by psychiatric medications. Impairments in renal and hepatic function, if present, can influence treatment recommendations.

Table 2 (see Statement 1) provides a discussion of suggested physical and laboratory assessments for patients with schizophrenia as part of initial evaluation and follow-up assessments. Such assessments are important for prevention, early recognition, and treatment of abnormalities such as glucose dysregulation, hyperlipidemia, and metabolic syndrome. It is important that patients have access to primary care clinicians who can work with the psychiatrist to diagnose and treat concurrent physical health conditions (American Psychiatric Association 2016a), but the psychiatrist may also provide ongoing monitoring and treatment of common medical conditions in conjunction with primary care clinicians (Druss et al. 2018).

Pregnancy and Postpartum Period

Women with childbearing potential and at risk for pregnancy should be assisted in obtaining effective contraception if pregnancy is not desired. For women who are planning to become pregnant or who are pregnant or in the postpartum period, it is essential to collaborate with the patient, her obstetrician-gynecologist or other obstetric practitioner, and, if involved, her partner or other persons of support. For women who are breastfeeding, collaboration with the infant’s pediatrician is similarly important. The overall goal is to develop a plan of care aimed at optimizing outcomes for both the patient and her infant. In addition, during pregnancy and postpartum, frequent reassessment will be needed to determine whether any modifications to the treatment plan are indicated.

As with any decisions related to the use of psychiatric medications prior to conception, during pregnancy, or while breastfeeding, it is essential to consider the potential benefits of treatment as well as the potential harms of untreated illness and the potential for negative fetal or neonatal effects. Untreated or inadequately treated maternal psychiatric illness can result in poor adherence to prenatal care, inadequate nutrition, increased alcohol or tobacco use, and disruptions to the family environment and mother-infant bonding (ACOG Committee on Practice Bulletins—Obstetrics 2008; American Academy of Pediatrics and the American College of Obstetricians and Gynecologists 2017; Tosato et al. 2017). For women with childbearing potential, decisions about medications and advice about contraceptive practices should consider the potential effects if pregnancy were to occur. Some medications are best avoided in women with childbearing potential; for example, valproic acid should be avoided because of its teratogenic effects (Briggs et al. 2017) and association with maternal metabolic syndrome.

All psychotropic medications studied to date cross the placenta, are present in amniotic fluid, and enter human breast milk (American Academy of Pediatrics and the American College of Obstetricians and Gynecologists 2017). In addition, the period from the third through the eighth week of gestation is associated with greatest risk for teratogenesis (American Academy of Pediatrics and the American College of Obstetricians and Gynecologists 2017). If a woman becomes pregnant while taking an antipsychotic medication, consideration should be given to consulting an obstetrician-gynecologist or maternal/fetal medicine subspecialist in addition to discussion with the prescribing clinician before indicated psychotropic medications are stopped to determine whether the risks of stopping the medication outweigh any possible fetal risks (American Academy of Pediatrics and the American College of Obstetricians and Gynecologists 2017; U.S. Food and Drug Administration 2011a). For many women, the eighth week of gestation will already have passed before obstetric care begins, and stopping medication will not avoid or reduce teratogenic risk. Thus, for a woman with schizophrenia, the benefits of continued treatment with antipsychotic medications in minimizing relapse will generally outweigh the potential for fetal risk (Briggs et al. 2017).

Knowledge of the effects of antipsychotic medications is limited to observational and registry-based studies. Although limited information is known about newer second-generation antipsychotic medications (SGAs), first-generation antipsychotic medications (FGAs) have been in wide use for more than 40 years, and older SGAs have been available for several decades. The available data suggest that these medications have minimal risk in terms of teratogenic or toxic effects on the fetus (ACOG Committee on Practice Bulletins—Obstetrics 2008; Briggs et al. 2017; Chisolm and Payne 2016). There does appear to be a risk of withdrawal symptoms or neurological effects of antipsychotic medications in the newborn if an antipsychotic medication is used in the third trimester (Briggs et al. 2017; U.S. Food and Drug Administration 2011a). Symptoms may include agitation, abnormally increased or decreased muscle tone, tremor, sleepiness, severe difficulty breathing, and difficulty in feeding (Briggs et al. 2017; U.S. Food and Drug Administration 2011a). Nevertheless, tapering of antipsychotic medication late in pregnancy is not advisable because of the associated risk of relapse. In some newborns, the symptoms subside within hours or days and do not require specific treatment; other newborns may require longer hospital stays (U.S. Food and Drug Administration 2011a). The possibility of these effects signals the importance of close monitoring of the newborn in conjunction with the infant’s pediatrician. As noted above, however, the benefits of treatment for the mother and the longer-term benefits of treatment for the infant (e.g., enhanced mother-infant bonding, better adherence to prenatal care and nutrition, lesser rates of prenatal alcohol or tobacco use) will generally favor continuing and not tapering antipsychotic treatment.

A number of other considerations are relevant when treating women with an antipsychotic medication during pregnancy. In general, symptoms should be managed with the lowest effective dose, although it is preferable to maintain efficacy using a single medication at a higher dose rather than using multiple medications at lower doses (American Academy of Pediatrics and the American College of Obstetricians and Gynecologists 2017). If a patient’s symptoms are well controlled on a specific medication, it is usually not advisable to switch to a different antipsychotic medication, even if more safety information is available for a different drug (Chisolm and Payne 2016). Changing medications exposes the fetus to two different medications and also increases possibilities for symptom relapse in the patient.

As with all women who are pregnant, regular prenatal care is essential to ensuring optimal maternal-fetal outcomes (American Academy of Pediatrics and the American College of Obstetricians and Gynecologists 2017; American College of Obstetricians and Gynecologists 2018). Close monitoring for symptom recurrence and for side effects is important during pregnancy and in the postpartum period because the physiological alterations of pregnancy may affect the absorption, distribution, metabolism, and elimination of medications and may necessitate adjustments in medication doses (ACOG Committee on Practice Bulletins—Obstetrics 2008; Chisolm and Payne 2016). Women who are taking antipsychotic medications are also at increased risk of obesity and hyperglycemia, and folate supplementation to reduce risks of neural tube defects and assessment for diabetes during pregnancy are important elements of prenatal care (Briggs et al. 2017).

In terms of breastfeeding, limited information is available, but infants may be exposed to clinically significant levels of medication in breast milk, and the long-term effects of such exposure are not known (Sachs et al. 2013). Accordingly, mothers who wish to breastfeed their infants should review the potential benefits of breastfeeding as well as potential risks in the context of shared decision-making (American College of Obstetricians and Gynecologists’ Committee on Obstetric Practice and the Breastfeeding Expert Work Group 2016; Sachs et al. 2013), with associated monitoring of growth and development by the infant’s pediatrician (Sachs et al. 2013).

Additional information related to the use of antipsychotic medications during pregnancy and while breastfeeding can be found at the websites of the U.S. Food and Drug Administration (FDA; www.fda.gov), REPROTOX (www.reprotox.org), the Teratogen Information System (TERIS; www.depts.washington.edu/terisweb/teris), and the U.S. National Library of Medicine’s LactMed (www.ncbi.nlm.nih.gov/books/NBK501922). For women who have been treated with an SGA during pregnancy, enrolling in the National Pregnancy Registry for Atypical Antipsychotics is suggested (MGH Center for Women’s Mental Health 2019).

Determining a Treatment Setting

In determining a treatment setting, considerations for individuals with schizophrenia are similar to those for individuals with other diagnoses. Thus, in general, patients should be cared for in the least restrictive setting that is likely to be safe and to allow for effective treatment. If inpatient care is deemed essential, efforts should be made to hospitalize patients voluntarily. However, if hospitalization is deemed essential but is not accepted voluntarily by the patient, state or jurisdictional requirements for involuntary hospitalization should be followed. Indications for hospitalization usually include the patient posing a serious threat of harm to self or others or being unable to care for himself or herself and needing constant supervision or support as a result. Other possible indications for hospitalization include psychiatric or other medical problems that make outpatient treatment unsafe or ineffective or new onset of psychosis that warrants initial inpatient stabilization to promote reduction of acute symptoms and permit engagement in treatment.

For individuals with schizophrenia and other significant health issues, determination of a treatment setting will require weighing the pluses and minuses of possible settings to identify the optimal location for care. For example, individuals who require significant medical or surgical interventions or monitoring that are not typically available on a psychiatric inpatient service will likely be better served on a general hospital unit or in an intensive care setting with input from consultation-liaison psychiatrists. Considerable efforts may be needed to help staff who are unfamiliar with psychotic disorders engage with the patient (Freudenreich et al. 2019). In other circumstances, management of the patient on an inpatient psychiatric service in collaboration with consultants of other medical specialties will be optimal.

Less restrictive settings may be indicated when a patient does not meet criteria for inpatient treatment but requires more monitoring or assistance than is available in routine outpatient care. Such settings and programs may include ACT (Substance Abuse and Mental Health Services Administration 2008), assisted outpatient treatment, intensive outpatient treatment, partial hospitalization, and day hospitalization.

Involuntary Treatment Considerations

Under some circumstances, individuals may not wish to participate in treatment or take medications, even if they have severe symptoms. In states where psychiatric advance directives are available, patients may be able to state their preferences about treatment choices while they have capacity in the event of future decompensation and an inability to participate in decision-making. Even in the absence of a psychiatric advance directive, patients can often be helped to accept pharmacological treatment over time and with psychotherapeutic interactions that are aimed toward identifying subjectively distressing symptoms that have previously responded to treatment. Family members and other persons of support can also be helpful in encouraging the patient to engage in treatment.

Prevailing state laws will determine other steps to take if an individual lacks capacity but requires treatment. Some states have processes by which pharmacological treatment may be administered involuntarily, whereas in other states a judicial hearing may be needed to obtain permission to treat a patient who lacks capacity.

For a small subgroup of patients with repeated relapses, rehospitalizations, or even reincarcerations associated with nonadherence or impairments in insight, involuntary outpatient commitment may warrant inclusion in the treatment plan to improve adherence, prevent psychiatric deterioration, enhance outcomes, and promote recovery (American Psychiatric Association 2015; Gaynes et al. 2015; Harris et al. 2019; Segal et al. 2017a, 2017b; Swartz et al. 2017). Involuntary outpatient commitment (which also may be referred to as assisted outpatient treatment, mandated community treatment, outpatient court-ordered treatment, or a community treatment order) is increasingly available but varies among countries (Burns et al. 2013; Harris et al. 2019) and jurisdictions within the United States (Meldrum et al. 2016) in its criteria and implementation. Effective implementation requires adequate resources and individualized treatment planning (American Psychiatric Association 2015) if psychiatric (Gaynes et al. 2015; Harris et al. 2019; Segal et al. 2017a; Swartz et al. 2017) and physical health (Segal et al. 2017b) benefits are to be realized. As with any form of involuntary treatment, decisions about involuntary outpatient commitment require balancing ethical considerations related to patient autonomy and self-determination with considerations about the individual’s best interest (American Psychiatric Association 2015).

Addressing Needs of Patients With Schizophrenia in Correctional Settings

Rates of serious mental illness, including schizophrenia, are higher in correctional settings (e.g., prisons, jails, police lockups, detention facilities) than in the general population (Al-Rousan et al. 2017; Bebbington et al. 2017; Bradley-Engen et al. 2010; Hall et al. 2019; Steadman et al. 2009). Careful assessment and treatment planning are essential when individuals with schizophrenia are in correctional settings. Although some aspects of treatment may need to be adjusted to conform with unique aspects of correctional settings (Tamburello et al. 2018), many individuals experience gaps in care during incarceration (Fries et al. 2013; Reingle Gonzalez and Connell 2014; Wilper et al. 2009), and access should be preserved to essential elements of treatment, including antipsychotic medications (American Psychiatric Association 2009b) and treatment for concomitant substance use disorders (American Psychiatric Association 2007).

While in the correctional system, individuals with schizophrenia may be withdrawn or disorganized or behave in a disruptive manner. These behaviors may result in disciplinary infractions, which may lead the individual with schizophrenia to be placed in a locked-down setting. Such units are often called administrative segregation, disciplinary segregation, or restricted housing units (Krelstein 2002; Semenza and Grosholz 2019) and have been conceptualized as having three main characteristics: social isolation, sensory deprivation, and confinement (Zubek et al. 1969). Each of these elements can vary significantly, but inmates typically spend an average of 23 hours per day in a cell, have limited human interaction and minimal or no access to programs, and are maintained in an environment that is designed to exert maximum control over the person, which has raised broader ethical considerations about the long-term use of such settings (Ahalt and Williams 2016; Ahalt et al. 2017; American Psychiatric Association 2017, 2018; American Public Health Association 2013; Cloud et al. 2015; National Commission on Correctional Health Care 2016).

Inmates’ responses to the segregation experience differ, and relevant scientific literature is sparse (Kapoor and Trestman 2016; O’Keefe et al. 2013). In addition, mental health clinicians working in such facilities frequently report that inmates without preexisting serious mental disorders develop irritability, anxiety, and other dysphoric symptoms when housed in these units for long periods of time (Metzner 2002). Difficulties in providing appropriate and adequate access to mental health care and treatment are especially problematic in any segregation environment and are related to logistical issues that frequently include inadequate office space and limited access to inmates because of security issues (Metzner 2003; Metzner and Fellner 2010). In addition, because of their inherently punitive structure, such units typically provide very little support or access to relevant treatment modalities or a therapeutic milieu. Furthermore, rates of self-injury and suicide appear to be higher in such settings than elsewhere in the correctional system (Baillargeon et al. 2009b; Glowa-Kollisch et al. 2016; Kaba et al. 2014; Way et al. 2005). Consequently, persons with schizophrenia should generally not be placed in a 23-hour/day lockdown for behaviors that are directly related to schizophrenia because such an intervention is likely to exacerbate rather than reduce psychotic symptoms, as well as increase rather than reduce disruptive behaviors (American College of Correctional Physicians 2013; American Psychiatric Association 2016b, 2017; American Public Health Association 2013; National Commission on Correctional Health Care 2016).

Individuals with schizophrenia, like other individuals with serious mental illness, are at increased risk for symptom relapse and gaps in treatment on release from a correctional setting. Services are often needed to reduce the likelihood of recidivism and maintain continuity of care for treatment of schizophrenia and concomitant disorders (e.g., substance use disorders, other medical conditions). Thus, discharge planning is a crucial aspect of care for inmates with schizophrenia, particularly for those who have been incarcerated for significant periods of time. Often, inmates with schizophrenia have been alienated from systems of care and psychosocial supports prior to arrest, and this estrangement is compounded by incarceration. As a result, inmates will likely need assistance around the time of discharge, which can encompass various domains, including housing, treatment needs, financial support, and obtaining supplemental security income/social security disability and related Medicaid benefits (American Psychiatric Association 2009c; Angell et al. 2014; Baillargeon et al. 2009a, 2010; Draine et al. 2010; Gertner et al. 2019; Morrissey et al. 2016; Wenzlow et al. 2011).

Benefits

Development and documentation of a comprehensive, person-centered treatment plan assures that the clinician has considered the available nonpharmacological and pharmacological options for treatment and has identified those treatments that are best suited to the needs of the individual patient, with a goal of improving overall outcome. It may also assist in forming a therapeutic relationship, eliciting patient preferences, permitting education about possible treatments, setting expectations for treatment, and establishing a framework for shared decision-making. Documentation of a treatment plan promotes accurate communication among all those caring for the patient and can serve as a reminder of prior discussions about treatment.

Harms

The only identifiable harm from this recommendation relates to the time spent in discussion and documentation that may reduce the opportunity to focus on other aspects of the evaluation.

Patient Preferences

Clinical experience suggests that patients are cooperative with and accepting of efforts to establish treatment plans.

Balancing of Benefits and Harms

The potential benefits of this guideline statement were viewed as far outweighing the potential harms. The level of research evidence is rated as low because no information is available on the harms of such an approach. There is also minimal research on whether developing and documenting a specific treatment plan improves outcomes as compared with assessment and documentation as usual. However, indirect evidence, including expert opinion, supports the benefits of comprehensive treatment planning. For additional discussion of the research evidence, see Appendix C, Statement 3.

Differences of Opinion Among Writing Group Members

There were no differences of opinion. The writing group voted unanimously in favor of this recommendation.

Selection of an Antipsychotic Medication

Initiation of Treatment With an Antipsychotic Medication

The initial goal of acute treatment with an antipsychotic medication is to reduce acute symptoms, with the aim of returning the individual to his or her baseline level of functioning. Later, maintenance treatment will aim to prevent recurrence of symptoms and maximize functioning and quality of life.

The initial dose of medication will depend on such factors as the medication formulation, the characteristics of the patient, and whether a prior trial of antipsychotic medication has occurred. With the exception of clozapine, the dose of most antipsychotic medications can be increased relatively quickly to a typical therapeutic dose once an initial dose has been tolerated. For patients who have previously been treated with an oral or LAI antipsychotic medication, more rapid resumption of an effective medication dose is often appropriate. Although as-needed or emergency administration of antipsychotic medication may, at times, be useful in individuals with acute agitation, it can also reduce tolerability and contribute to a perception that premature dose increases are needed.

Younger individuals who are experiencing a first episode of psychosis may be more likely to gain weight or develop adverse metabolic effects of antipsychotic medications (Correll et al. 2014; Jensen et al. 2019). This can influence selection of an initial medication. In such individuals, a lower initial medication dose may help in minimizing acute side effects of antipsychotic medication and may improve a patient’s willingness to continue with treatment (Czobor et al. 2015; Gaebel et al. 2010). Because a first episode of psychosis may respond more rapidly and may require a lower medication dose than later episodes do, use of a lower initial medication dose is also reasonable (Takeuchi et al. 2019). In older individuals, particularly those with concomitant physical health issues who are receiving multiple medications, recommended starting doses of medication are one-quarter to one-half of the usual adult starting dose on the basis of pharmacokinetic considerations (Howard et al. 2000).

Determining the optimal dose of antipsychotic medication during acute treatment is complicated by the fact that there is usually a delay between initiation of treatment and full therapeutic response. Patients may take between 2 and 4 weeks to show an initial response and longer periods of time to show full or optimal response. Once a therapeutic dose of the antipsychotic medication is reached, overly rapid or premature escalation of medication doses can affect tolerability. Premature dose increases can also create the false impression of enhanced efficacy due to a higher dose when the observed response is actually related to elapsed time at a steady state level of medication. Available evidence suggests that patients who have not exhibited at least a 20% reduction in symptoms (or minimal improvement) by about 2 weeks on a therapeutic dose are unlikely to be much improved at 4–6 weeks as reflected by at least a 50% reduction in symptoms (Samara et al. 2015). Consequently, monitoring of the patient’s clinical status for 2–4 weeks is warranted on a therapeutic dose unless the patient is having uncomfortable side effects.

Initiation of treatment with clozapine is a notable exception to this general approach because it requires a slow dose titration to minimize the risks of seizure, orthostatic hypotension, and excessive sedation (Clozaril 2019). Large, rapid increases in clozapine dosage have led to cardiovascular collapse and death, particularly in patients taking respiratory depressant medications such as benzodiazepines. From a starting dose of 12.5 mg once or twice daily, the daily clozapine dosage can be increased by, at most, 25–50 mg/day to a target dose of 300–450 mg/day (Clozaril 2019). Subsequent dose increases, if needed, should be of 100 mg or less, once or twice weekly. A slower rate of titration may be needed for patients with an initial episode of schizophrenia and in those who are older, severely debilitated, or sensitive to side effects. Those with a preexisting central nervous system condition, including individuals with 22q11.2 deletion syndrome, also warrant a slower rate of titration and may have an increased risk of seizures at usual doses. Use of divided doses can be helpful in reducing side effects during initial dose titration, although many patients are ultimately treated with a single dose at bedtime to minimize daytime sedation and facilitate adherence (Takeuchi et al. 2016). Although efficacy is often seen at a dosage of 300–450 mg/day, some individuals may need higher dosages of clozapine, to a maximum daily dose of 900 mg, for full response. Blood levels of clozapine can be helpful to obtain if making adjustments to clozapine doses (see Statement 7).

With clozapine, safety monitoring during treatment is important in order to minimize the risk of adverse events. The U.S. Clozapine Risk Evaluation and Mitigation Strategy (REMS) Program (www.clozapinerems.com)1 includes required training that must be completed by prescribers (Clozapine REMS Program 2019a), resource materials (Clozapine REMS Program 2019b), and a shared patient registry for all clozapine manufacturers’ products that permits tracking of absolute neutrophil counts (ANCs) and documentation of decisions about continued treatment. The Clozapine REMS site provides instructions about threshold values for ANCs in hematologically normal individuals and in those with benign ethnic neutropenia, which is most common in individuals of African descent and is associated with normal ANCs that are lower than standard reference ranges (Clozapine REMS Program 2014). The site also describes the required frequencies for ANC monitoring, which vary with ANC values. Because the highest risk of severe neutropenia (ANC < 500/μL) occurs within the initial 6 months of clozapine treatment (Alvir et al. 1993; Clozapine REMS Program 2019c; Myles et al. 2018), the frequency of ANC monitoring is also reduced with longer treatment duration. In patients who have stopped or interrupted treatment with clozapine for 30 days or more, the initial dose titration for clozapine and the monitoring frequency for treatment initiation should be followed.

With clozapine as well as with other antipsychotic medications, some common early side effects such as sedation, postural hypotension, or nausea may improve or resolve after the first several days or weeks of treatment, and patients can be encouraged to tolerate or temporarily manage these short-term effects. Other side effects, notably parkinsonism and akathisia, are likely to persist with long-term treatment, and additional approaches to management may be needed (see Statements 12 and 13).

If treatment with an LAI antipsychotic medication is planned, a trial of the oral formulation of the same medication is usually given to assure tolerability. The conversion from an oral dose of medication to a corresponding dose of an LAI antipsychotic depends on the specific medication (Meyer 2017), and product labeling for each medication describes approximate conversion ratios and whether a period of concomitant oral and LAI medication is needed.

Strategies to Address Initial Nonresponse or Partial Response to Antipsychotic Treatment

If a patient is showing response within several weeks of treatment initiation, continuing with the same medication and monitoring for continued improvement is appropriate. However, if there is no significant improvement after several weeks of treatment (e.g., < 20% improvement in symptoms) or if improvement plateaus before substantial improvement is achieved (e.g., > 50% improvement in symptoms, minimal impairment in functioning), it is important to consider whether factors are present that are influencing treatment response. Such factors may include concomitant substance use, rapid medication metabolism, poor medication absorption, interactions with other medications, and other effects on drug metabolism (e.g., smoking) that could affect blood levels of medication.

Difficulties with adherence are a common contributor to reduced response. (For a detailed discussion of adherence, see Statement 3.) When adherence is poor or uncertain, use of an LAI formulation of an antipsychotic may improve adherence as well as response. Determination of the blood concentration of the drug may also be helpful if the patient is being treated with a medication (e.g., clozapine) for which blood level has some correlation with clinical response. For other antipsychotic medications, a blood level can help to determine if poor adherence or subtherapeutic levels may be contributing to poor response (Bishara et al. 2013; de Oliveira et al. 1996; Hiemke et al. 2004; Lopez and Kane 2013; McCutcheon et al. 2018; Melkote et al. 2018; Sparshatt et al. 2010; Uchida et al. 2011b; Van Putten et al. 1991). Depending on the patient’s symptoms, the possibility of another concomitant disorder should be considered. For example, in a patient with negative symptoms, an untreated major depressive disorder may also be present.

If no factors that would affect treatment response have been identified, raising the dose for a finite period, such as 2–4 weeks, can be tried. Although the incremental efficacy of higher doses has not been established (Samara et al. 2018), some patients may show benefit if they are able to tolerate a higher dose of antipsychotic medication without significant side effects. If dose adjustment does not result in an adequate response, a different antipsychotic medication should be considered. Tables 5 and 6 in Statement 4 can be consulted to identify antipsychotic medications with other receptor binding profiles or different side effects. Because each patient responds differently to antipsychotic medications in terms of therapeutic effects and side effects, adequate trials of multiple antipsychotic medications may be needed before antipsychotic treatment is optimized, and it can be helpful to advise patients of this possibility.

If a patient has had minimal or no response to two trials of antipsychotic medication of 2–4 weeks’ duration at an adequate dose (Howes et al. 2017; Samara et al. 2015), a trial of clozapine is recommended (see Statement 7). A trial of clozapine is also recommended for a patient with a persistent risk of suicide that has not responded to other treatments and is suggested for a patient with a persistent risk of aggressive behavior that has not responded to other treatments (see Statements 8 and 9). A trial of clozapine may also be appropriate in individuals who show a response to treatment (i.e., have more than a 20% reduction in symptoms) yet still have significant symptoms or impairments in functioning (Howes et al. 2017). In fact, clozapine is often underused (Carruthers et al. 2016; Latimer et al. 2013; Olfson et al. 2016; Stroup et al. 2014; Tang et al. 2017), and many patients would benefit from earlier consideration of clozapine initiation.

For individuals with treatment-resistant schizophrenia who are unable to tolerate clozapine or are not interested in pursuing a trial of clozapine, the limited available evidence suggests no benefit from high doses of antipsychotic medication, and treatment-related side effects are likely to be increased (Dold et al. 2015). However, a trial of a different antipsychotic medication may be helpful, particularly if there is no response or only a partial response to the most recently used medication.

Augmentation treatment can also be considered, although a trial of clozapine should not be delayed by multiple attempts at augmentation therapy. Particularly for patients with negative symptoms or depression, augmentation of antipsychotic therapy with an antidepressant medication may also be helpful (Helfer et al. 2016; Stroup et al. 2019). Use of a benzodiazepine, such as lorazepam, is also suggested in patients who exhibit catatonia (Bush et al. 1996b; Fink 2013; Pelzer et al. 2018; Unal et al. 2017). Other augmentation approaches (e.g., antipsychotics, anticonvulsants, benzodiazepines, lithium) have also been studied, although evidence is mixed and primarily from small, short-term open-label studies (Correll et al. 2017b; Galling et al. 2017; Ortiz-Orendain et al. 2017). For combination therapy with two antipsychotic medications, data from a large nationwide cohort study suggest that emergency visits and rehospitalization rates may be reduced in individuals receiving polypharmacy as compared with monotherapy (Tiihonen et al. 2019). In addition, there is no evidence that combining drugs is any more harmful than using a single medication, beyond the common side effects from each drug. Nevertheless, if multiple drugs are used, monitoring for benefits and side effects is important, and it is preferable to limit changes in dose to one drug at a time. In addition, if a patient experiences an exacerbation of symptoms while on a stable dose of medication, a reconsideration of the treatment plan is warranted rather than simply adding medications to the existing regimen.

Monitoring During Treatment With an Antipsychotic Medication

During treatment with an antipsychotic medication, it is important to monitor medication adherence, therapeutic benefits of treatment, and treatment-related side effects. The patient’s clinical status can also be affected by changes in physical health, adjustments to other psychotropic and nonpsychotropic medications, and other factors, such as cessation or resumption of smoking.

Adherence to antipsychotic treatment is a common problem that affects treatment outcomes. There are many barriers to treatment adherence as well as facilitators and motivators of adherence, each of which will differ for an individual patient (Hatch et al. 2017; Kane et al. 2013; Pyne et al. 2014). Thus, it is important to take a patient-centered approach in inquiring in a nonjudgmental way whether the individual has experienced difficulties with taking medication since the last visit. (For a detailed discussion of factors related to adherence, see Statement 3.)

Monitoring of treatment response is also essential for identifying whether there are reductions in the severity of functional impairments or target symptoms, including positive symptoms, negative symptoms, and other symptoms that are a focus of treatment. Use of a quantitative measure (see Statement 2) can assist in determining whether the antipsychotic medication is producing therapeutic benefits, including reductions in symptom severity and improvements in functioning. If a lack of response or a partial response is noted, additional assessment will be needed to identify and address possible contributors as described in the subsection “Strategies to Address Initial Nonresponse or Partial Response to Antipsychotic Treatment.” If an antipsychotic medication dose is being decreased, monitoring can help detect a return of symptoms prior to a more serious relapse.

Monitoring for the presence of side effects is also important throughout the course of antipsychotic treatment. Some side effects (e.g., sedation, nausea) are prominent with treatment initiation but dissipate, at least to some extent, with continued treatment. Other side effects (e.g., hypotension, akathisia) may be present initially and increase in severity with titration of the medication dose. Still other side effects emerge only after longer periods of treatment (e.g., tardive dyskinesia) or become more noticeable to patients as their acute symptoms are better controlled (e.g., sexual dysfunction). Table 2 in Statement 1 gives suggestions for baseline assessments and monitoring frequencies for some side effects, clinical measurements, and laboratory studies. Specific attention may need to be given to clinical workflow to assure that indicated monitoring is conducted because rates of follow-up testing and screening for metabolic side effects of treatment appear to be low (Morrato et al. 2009). Patients should also be asked about other common side effects of antipsychotic medications, which may vary with the specific medication that is prescribed (see Statement 4, Table 6).

Use of rating scales can help assure that patients are asked about side effects in a systematic fashion. Although the clinician-rated Udvalg for Kliniske Undersogelser (UKU) Side Effect Rating Scale (Lingjaerde et al. 1987) is often used to assess side effects of antipsychotic medications in clinical trials (van Strien et al. 2015), it can be time-consuming to administer. However, a self-rated version of the UKU Side Effect Rating Scale (Lindström et al. 2001) is also available (http://scnp.org/fileadmin/SCNP/SCNP/UKU/UKU-Pat-%20English%20.pdf). Another self-rating scale, the Glasgow Antipsychotic Side-effect Scale, has two versions: one for use in patients treated with clozapine (www.sussexpartnership.nhs.uk/sites/default/files/documents/gass_for_clozapine .pdf) (Hynes et al. 2015) and one for patients treated with other antipsychotic medications (www.southernhealth.nhs.uk/_resources/assets/inline/full/0/38222.pdf) (Waddell and Taylor 2008). Other rating scales are aimed at identifying and assessing the severity of a specific type of side effect. For example, the clinician-administered Abnormal Involuntary Movement Scale (AIMS; www.aacap.org/App_Themes/AACAP/docs/member_resources/toolbox_for_clinical_practice_ and_outcomes/monitoring/AIMS.pdf) (Guy 1976) or the Dyskinesia Identification System: Condensed User Scale (DISCUS; https://portal.ct.gov/-/media/DDS/Health/hcsma20002b.pdf) (Kalachnik and Sprague 1993) can complement clinical assessment in identifying and monitoring tardive dyskinesia and other abnormal movements. Another example, the self-rated Changes in Sexual Functioning Questionnaire (www.dbsalliance.org/wp-content/uploads/2019/02/Restoring_Intimacy_CSFQ_ Handout.pdf) (Clayton et al. 1997a, 1997b; Depression and Bipolar Support Alliance 2019; Keller et al. 2006), can help to identify sexual side effects of antipsychotic treatment, which is an issue that patients may find difficult to discuss yet can lead them to discontinue treatment.

Treatment-Emergent Side Effects of Antipsychotic Medications

As with most medications, antipsychotic medications have been associated with a number of side effects that can develop as treatment proceeds. Table 6 in Statement 4 shows the relative tendencies for antipsychotic medications to be associated with specific side effects. In addition, each of these side effects is described in further detail below.

Early in the course of treatment, common side effects include sedation; orthostatic changes in blood pressure; and anticholinergic side effects such as dry mouth, constipation, and difficulty with urination.

Of the side effects related to dopamine D2 receptor antagonist effects of antipsychotics, acute dystonia also appears early in treatment. It is particularly common with high-potency antipsychotic medications (e.g., haloperidol, fluphenazine) and can be life-threatening if associated with laryngospasm. Neuroleptic malignant syndrome (NMS) can also be life-threatening because of associated hyperthermia and autonomic instability. It typically occurs within the first month of antipsychotic treatment, with resumption of treatment, or with an increase in the dose of antipsychotic medication. Akathisia and medication-induced parkinsonism can also occur in the initial weeks of treatment or after increases in medication doses. Hyperprolactinemia, related to D2 receptor antagonism in the hypothalamic-pituitary axis, can lead to breast enlargement, galactorrhea, sexual dysfunction, and, in women, menstrual disturbances. These elevations in prolactin also occur in the initial weeks to months of treatment. On the other hand, tardive syndromes, including tardive dyskinesia, develop later, often months or even years after treatment initiation.

Side effects related to metabolic syndrome are common. Generally, they are observed in the initial months of treatment, but they can also occur later in treatment. These side effects include weight gain; hyperlipidemia; and glucose dysregulation, including development of diabetes mellitus.

Clozapine treatment is associated with a number of side effects that are less commonly seen with other antipsychotic medications. Severe neutropenia is most often seen early in treatment and is potentially life-threatening; however, with current regulatory requirements for monitoring ANC levels during treatment, it is rare. When seizures occur with clozapine, it is typically with very high doses or blood levels of clozapine, rapid increases in clozapine dose, or shifts in medication levels (related to drug-drug interactions or effects of smoking on drug metabolism). Myocarditis is infrequent and generally occurs early in treatment. Cardiomyopathy is rare and generally occurs later in the treatment course. Gastrointestinal effects of clozapine can also be significant, and in some patients it is associated with fecal impaction or paralytic ileus. Sialorrhea and tachycardia are each commonly observed during treatment with clozapine but can generally be managed conservatively.

Benefits

Use of an antipsychotic medication in the treatment of schizophrenia can improve positive and negative symptoms of psychosis (high strength of research evidence) and can also lead to reductions in depression and improvements in quality of life and functioning (moderate strength of research evidence). A meta-analysis of double-blind, randomized, placebo-controlled trials showed a medium effect size for overall efficacy (Leucht et al. 2017), with the greatest effect on positive symptoms. The rates of achieving any response or a good response were also significantly greater in patients who received an antipsychotic medication. In addition, the proportion of individuals who dropped out of treatment for any reason and for lack of efficacy was significantly less in those who were treated with an antipsychotic medication. Research evidence from head-to-head comparison studies and network meta-analysis (McDonagh et al. 2017) showed no consistent evidence that favored a specific antipsychotic medication, with the possible exception of clozapine.

Harms

The harms of using an antipsychotic medication in the treatment of schizophrenia include sedation, side effects mediated through dopamine receptor blockade (e.g., acute dystonia, akathisia, parkinsonism, tardive syndromes, NMS, hyperprolactinemia), disturbances in sexual function, anticholinergic effects, weight gain, glucose abnormalities, hyperlipidemia, orthostatic hypotension, tachycardia, and QTc prolongation. Clozapine has additional harms associated with its use, including sialorrhea, seizures, neutropenia (which can be severe and life-threatening), myocarditis, and cardiomyopathy. Among the antipsychotic medications, there is variability in the rates at which each of these effects occurs, and no specific medication appears to be devoid of possible side effects.

Patient Preferences

Clinical experience suggests that many patients are cooperative with and accepting of antipsychotic medications as part of a treatment plan. A survey of patient preferences reported that patients viewed an ability to think more clearly and an ability to stop hallucinations or paranoia as important efficacy-related reasons to take an antipsychotic medication (Achtyes et al. 2018). However, patients also reported concerns about side effects, particularly weight gain, sedation, and restlessness, as reasons that they might not wish to take antipsychotic medications. Some patients might also choose not to take an antipsychotic medication when they are feeling well or if they do not view themselves as having a condition that requires treatment. Some patients may also prefer one medication over another medication on the basis of prior treatment experiences or other factors.

Balancing of Benefits and Harms

The potential benefits of this guideline statement were viewed as far outweighing the potential harms. Although harms of antipsychotic medications can be significant, the impact of schizophrenia on patients’ lives is also substantial, and consistent benefits of antipsychotic treatment were found. Harms of treatment can be mitigated by selecting medications on the basis of individual characteristics and preferences of patients as well as by choosing a medication on the basis of its side-effect profile, pharmacological characteristics, and other factors. For clozapine, the additional benefits of treatment were viewed as outweighing the additional rare but serious harms and the need for ANC monitoring to reduce the likelihood of severe neutropenia. For additional discussion of the research evidence, see Appendix C, Statement 4.

Differences of Opinion Among Writing Group Members

There were no differences of opinion. The writing group voted unanimously in favor of this recommendation.

Benefits

Use of an antipsychotic medication that has already been associated with symptom response can maintain improvements in symptoms as well as promote enhanced functioning and quality of life (high strength of research evidence). Long-term treatment with an antipsychotic medication has also been associated with a reduction in mortality as compared with no antipsychotic treatment in individuals with schizophrenia. In contrast, discontinuation of antipsychotic treatment can be associated with increases in symptoms and risk of hospitalization and poorer long-term outcomes, including greater mortality in the long term (low strength of research evidence).

Harms

The harms of continuing use of an antipsychotic medication can vary depending on whether the patient is experiencing any significant side effects from the medication that would have long-term untoward effects. For patients whose medications are well tolerated, long-term risks include tardive syndromes from antipsychotic medications. For other patients, long-term risks will vary according to the specific side effect, with metabolic effects of antipsychotic medication serving as a possible contributor to long-term health risks. Some studies have raised concerns about changes in brain region volumes with antipsychotic treatment, but these findings are heterogeneous and inconsistent.

Patient Preferences

Clinical experience suggests that many patients are cooperative with and accepting of antipsychotic medications as part of a treatment plan. This is particularly true when the medication has been associated with a response in symptoms. Indeed, a survey of patient preferences reported that patients viewed an ability to think more clearly and an ability to stop hallucinations or paranoia as important efficacy-related reasons to take an antipsychotic medication (Achtyes et al. 2018). Patients are also likely to value the long-term benefits that have been shown with continued antipsychotic treatment, including reductions in relapses, hospitalizations, and mortality. However, patients also report concerns about side effects, particularly weight gain, sedation, and restlessness, that can make them reluctant to take antipsychotic medications on a long-term basis. In addition, some patients may choose not to take an antipsychotic medication when they are feeling well or if they do not view themselves as having a condition that requires treatment.

Balancing of Benefits and Harms

The potential benefits of this guideline statement were viewed as far outweighing the potential harms. Although harms of antipsychotic medications can be significant and the long-term effects of antipsychotic medications are not well studied, the impact of schizophrenia on patients’ lives is also substantial, and consistent benefits of continued antipsychotic treatment were found. Overall, rates of mortality appear to be reduced by ongoing treatment with an antipsychotic medication as compared with no treatment. In addition, harms of treatment can be mitigated by using the lowest effective dose; by selecting medications on the basis of individual characteristics and preferences of patients; and by choosing a medication on the basis of its side-effect profile, pharmacological characteristics, and other factors. For additional discussion of the research evidence, see Appendix C, Statement 5.

Differences of Opinion Among Writing Group Members

There were no differences of opinion. The writing group voted unanimously in favor of this recommendation.

Benefits

Use of an antipsychotic medication that has already been associated with symptom response can maintain improvements in symptoms as well as promote enhanced functioning and quality of life. In contrast, changes in antipsychotic treatment can be associated with early treatment discontinuation, increases in symptoms, clinical destabilization, and worsening of treatment tolerability.

Harms

The harms of continuing use of the same antipsychotic medication can vary depending on whether the patient is experiencing any significant side effects from the medication that would have long-term untoward effects. Continuing the same medication could lead to greater long-term risks such as metabolic effects or tardive syndromes from antipsychotic medications, but this would depend on the side-effect profile of the medication. In some instances, changing to a different medication could worsen long-term side-effect risk rather than reduce such risks.

Patient Preferences

Clinical experience suggests that most patients prefer to continue to take an antipsychotic medication that has led to a response in symptoms. Once they have found a medication that is effective and well tolerated, many individuals experience anxiety if they are unable to continue with that medication because of realistic concerns about a possible return of symptoms, reductions in functioning, risk of hospitalization, and other potential consequences of medication changes. However, other patients may not wish to remain on a given antipsychotic medication because of concerns about side effects or other factors that make continued treatment difficult.

Balancing of Benefits and Harms

The potential benefits of this guideline statement were viewed as likely to outweigh the potential harms. Although most patients prefer to continue taking the same medication once their symptoms have responded, there are reasons that a change in medication may be indicated, and factors such as medication side-effect profiles, medication availability, and patient preferences for specific medications also may play a role in decisions to continue with the same medication. For additional discussion of the research evidence, see Appendix C, Statement 6.

Differences of Opinion Among Writing Group Members

There were no differences of opinion. The writing group voted unanimously in favor of this suggestion.

Identification of Treatment-Resistant Schizophrenia

Clozapine is recommended for individuals with treatment-resistant schizophrenia, but there is considerable variation in definitions of treatment-resistant schizophrenia in clinical trials and in practice (Howes et al. 2017). For the purpose of future research trials, the Treatment Response and Resistance in Psychosis (TRRIP) Working Group conducted a detailed systematic review of clinical trials in treatment-resistant schizophrenia and used a consensus-based approach to establish minimum and optimum criteria for identifying treatment-resistant schizophrenia (Howes et al. 2017). In addition to a diagnosis of schizophrenia, the identification of treatment-resistant schizophrenia rests on the persistence of significant symptoms despite adequate pharmacological treatment (Howes et al. 2017). More specifically, the TRRIP Working Group recommended that symptoms be of at least 12 weeks’ duration in total, be of at least moderate severity, and be associated with at least moderate functional impairment as determined by validated rating scales (e.g., PANSS, BPRS, or SANS and SAPS for symptoms and a score < 60 on the SOFAS as a measure of functioning). If a prospective medication trial of at least 6 weeks at adequate dose has not led to symptom reduction of more than 20%, this provides additional evidence of treatment resistance. It is helpful to note whether the persistent symptoms include positive, negative, or cognitive symptoms because responses to these symptom domains may differ.

In terms of treatment adequacy, the TRRIP Working Group recommended that at least two antipsychotic trials should be conducted with different antipsychotic medications with at least 6 weeks at a therapeutic dosage of medication for each and with adherence of at least 80% of prescribed dosages (Howes et al. 2017). A therapeutic dosage of medication was defined as the midpoint of the target range for acute treatment of schizophrenia according to the manufacturer’s product labeling or the equivalent of at least 600 mg of chlorpromazine per day (Howes et al. 2017). (For tables of dose equivalents, see College of Psychiatric and Neurologic Pharmacists 2019; Leucht et al. 2014, 2015; Rothe et al. 2018.) The consensus criteria include at least one antipsychotic blood level to assess adherence, obtaining information on adherence from at least two sources (e.g., pill counts, dispensing chart reviews, patient/carer reports), and obtaining information on past treatment response from patient/carer reports and other sources.

For clinical purposes, a common definition is that a patient’s symptoms have shown no response or partial and suboptimal response to two antipsychotic medication trials of at least 6 weeks each at an adequate dosage of medication, and some definitions specify using medications from different classes (e.g., SGA vs. FGA). However, if there is no significant improvement after several weeks of treatment (e.g., < 20% improvement in symptoms), the likelihood of substantial improvement (e.g., > 50% improvement in symptoms) is small (Howes et al. 2017; Samara et al. 2015), and a longer trial of the medication may not be warranted. It should also be noted that a medication trial cannot be viewed as adequate if truncated in terms of duration or dosage because of poor tolerability or if limited by poor adherence. Accordingly, some experts suggest a trial of an LAI antipsychotic medication before deciding that a patient’s symptoms are treatment-resistant.

Initiation of Treatment With Clozapine

After a patient is identified as having treatment-resistant schizophrenia, the clinician should engage the patient in discussion about clozapine treatment. A trial of clozapine may also be appropriate in individuals who show a response to treatment (i.e., have at least a 20% reduction in symptoms) yet still have significant symptoms or impairments in functioning (Howes et al. 2017). In fact, clozapine is often underused (Carruthers et al. 2016; Latimer et al. 2013; Olfson et al. 2016; Stroup et al. 2014; Tang et al. 2017), and many patients would benefit from earlier consideration of clozapine initiation.

Discussion of clozapine should emphasize principles of shared decision-making; including family members or other persons of support in such discussions is often beneficial. In terms of a patient’s recovery goals, most individuals value an ability to think more clearly and stop hallucinations or delusions when deciding about medication changes (Achtyes et al. 2018). In addition, most patients who receive clozapine view it positively. For example, one large survey of individuals with schizophrenia or schizoaffective disorder who were taking an antipsychotic medication found that the vast majority of those taking clozapine adhered to treatment and found it helpful, whereas only approximately 5% found it not helpful (Siskind et al. 2017a). In contrast, most other antipsychotic medications were viewed less positively (Siskind et al. 2017a). Nevertheless, it is important to identify patient concerns about clozapine and address them insofar as is possible. For example, patients may express concerns about the burdens of required blood work and may encounter logistical barriers such as access to transportation (Farooq et al. 2019; Gee et al. 2017; Verdoux et al. 2018). However, they may be willing to consider clozapine if logistical barriers can be overcome or if given the information that blood monitoring requirements become less frequent over time. Concerns about other side effects, such as weight gain or somnolence, may also contribute to a reluctance to switch to clozapine (Achtyes et al. 2018). It can be helpful to have an open discussion of these side effects and a well-defined plan for monitoring as treatment proceeds. Peer-run support groups that directly address living with side effects can help patients develop strategies for coping with side effects.

Clinicians may also have concerns about clozapine that can serve as a barrier to treatment. For example, many clinicians have limited experience in using clozapine and sometimes express concerns about paperwork burdens, patient adherence with monitoring, and side effects (Daod et al. 2019; Farooq et al. 2019; Gee et al. 2017; Kelly and Love 2019; Kelly et al. 2018; Leung et al. 2019; Verdoux et al. 2018; Warnez and Alessi-Severini 2014). Many clinicians overestimate the likelihood of severe neutropenia and are reluctant to begin clozapine on an outpatient basis (Farooq et al. 2019). Education about the use of clozapine and its side effects can be useful in addressing clinician-related prescribing barriers.

When initiating treatment with clozapine, a slow dose titration is essential to minimize the risks of seizure, orthostatic hypotension, and excessive sedation (Clozaril 2019). Large, rapid increases in clozapine dosage have led to cardiovascular collapse and death, particularly in patients taking respiratory depressant medications such as benzodiazepines. From a starting dosage of 12.5 mg once or twice daily, the dosage of clozapine can be increased by, at most, 25–50 mg/day to a target dosage of 300–450 mg/day (Clozaril 2019). Subsequent dose increases, if needed, should be of 100 mg or less, once or twice weekly. Although efficacy is often seen at a dosage of 300–450 mg/day, some individuals may need higher dosages of clozapine, to a maximum daily dose of 900 mg, for full response. A slower rate of titration may be needed for patients with an initial episode of schizophrenia and in those who are older, severely debilitated, or sensitive to side effects. Those with a preexisting central nervous system condition, including individuals with 22q11.2 deletion syndrome, also warrant a slower rate of titration and may have an increased risk of seizures at usual doses. Use of divided doses can be helpful in reducing side effects during initial dose titration, although many patients are ultimately treated with a single dose at bedtime to minimize daytime sedation and facilitate adherence (Takeuchi et al. 2016).

Monitoring for therapeutic benefits and side effects of clozapine should occur throughout the dose titration phase (see Statement 4). Because titration of clozapine proceeds slowly, the therapeutic benefits may not be noticed immediately, and side effects may be more prominent than benefits. Thus, it can be helpful to provide patients with education and reassurance about the expected timetable of therapeutic effects of clozapine.

If clozapine is being resumed after a gap in treatment of 48 hours or more, it should be restarted at 12.5 mg once or twice daily. If that dosage is well tolerated, the dose may be increased to a therapeutic range more quickly than recommended for initial treatment. If a decision is made to stop clozapine, it is best to taper the dose unless the medication is being stopped for medically urgent reasons (e.g., severe neutropenia, myocarditis, NMS).

Use of Clozapine Levels During Treatment With Clozapine

While the dose of clozapine is being titrated, it is useful to obtain blood levels of clozapine and its major active metabolite, norclozapine (N-desmethylclozapine) (Couchman et al. 2010). Blood levels can also be helpful if there are questions about medication adherence, less efficacy or more side effects than expected, potential medication interactions, or other factors that may be influencing clozapine levels. Although there is substantial variation between individuals, clozapine levels on a specific dosage will generally be greater in nonsmokers than in smokers, in heavy caffeine users than in nonusers, in women than in men, and in older individuals than in younger individuals (Carrillo et al. 1998; Ismail et al. 2012). In addition, changing between different generic forms of clozapine can lead to a 5%–10% difference in blood levels. Levels of clozapine should be drawn at steady state (3 days or more after a dose change) and at a trough in medication levels (about 12 hours after the last dose). Typically, patients will receive a bedtime dose of clozapine and then have a level drawn the following morning before receiving an additional dose. There is not an absolute level of clozapine that is associated with either efficacy or toxicity (Remington et al. 2013; Spina et al. 2000; Stark and Scott 2012; Suzuki et al. 2011; VanderZwaag et al. 1996). In most patients, efficacy will be highest at levels greater than 350 ng/mL of clozapine, but some patients will show response or prevention of relapse at levels as low as 200 ng/mL. The risk of developing seizures increases with the blood level of clozapine.

As with the results of any laboratory test, interpretation of clozapine levels should consider the clinical context. For example, if a clozapine level is much higher than expected, assess for dose-related side effects and clinical evidence of toxicity. If the patient’s clinical status does not suggest signs of clozapine toxicity, then determine the timing of the level (e.g., peak vs. trough) and identify any potential for drug interactions, changes in smoking status, or incorrect specimen labeling. If levels are much lower than expected, such factors as poor adherence, rapid metabolism, drug interactions, or changes in smoking status may also be relevant.

Typically, the level of norclozapine will be reported along with the blood level of clozapine. Norclozapine is the major active metabolite of clozapine and appears to differ from clozapine in efficacy and side effects. Nevertheless, the value of norclozapine levels in guiding clinical decisions is unclear. Because the half-life of norclozapine is greater than the half-life of clozapine, a clozapine:norclozapine ratio less than 0.5 can suggest poor adherence over the previous day or rapid metabolism of clozapine (e.g., via CYP1A2 induction). A clozapine:norclozapine ratio greater than 3.0 could suggest that metabolic pathways are saturated or inhibited by a concomitant medication. Shifts in the ratio of clozapine to norclozapine can also result from other drug-drug interactions or if nontrough levels are obtained (Couchman et al. 2010; Ellison and Dufresne 2015).

Monitoring for Side Effects During Treatment With Clozapine

With clozapine, safety monitoring during treatment is important to minimize the risk of adverse events. The Clozapine REMS Program (www.clozapinerems.com)1 is required for prescribing of clozapine in the United States. The REMS program includes required training that must be completed by prescribers (Clozapine REMS Program 2019a), resource materials (Clozapine REMS Program 2019b), and a shared patient registry for all clozapine manufacturers’ products that permits tracking of ANCs and documentation of decisions about continued treatment. The Clozapine REMS site provides instructions about threshold values for ANCs in hematologically normal individuals and in those with benign ethnic neutropenia, which is most common in individuals of African descent and is associated with ANCs that are lower than standard reference ranges (Clozapine REMS Program 2014). It also describes the required frequencies for ANC monitoring, which vary with ANC values. In patients who have stopped or interrupted treatment with clozapine for 30 days or more, the initial dose titration for clozapine and the monitoring frequency for treatment initiation should be followed.

Because the highest risk of severe neutropenia (ANC < 500/μL) occurs within the initial 6 months of clozapine treatment (Alvir et al. 1993; Clozapine REMS Program 2019c; Myles et al. 2018), ANC monitoring is more frequent early in treatment and is required less often with longer treatment duration. The need for ANC monitoring can be a common practical issue for patients because of the time and transportation needed to obtain blood tests at a laboratory. The availability of point-of-care testing for white blood cell counts may mitigate these barriers for patients and facilitate treatment with clozapine.

In addition to neutropenia, clozapine treatment can be associated with several other important side effects. Potentially serious cardiac complications of clozapine treatment include myocarditis and cardiomyopathy. Myocarditis is infrequent and generally occurs during the first month of treatment, whereas cardiomyopathy is rare and generally occurs later in the treatment course (Bellissima et al. 2018; Ronaldson et al. 2015). Myocarditis usually is heralded by shortness of breath, tachycardia, and fever, but diagnosis can be challenging because of the nonspecific nature of other symptoms, which can include fatigue, chest pain, palpitations, peripheral edema, and hypereosinophilia. In patients who do develop myocarditis or cardiomyopathy in conjunction with clozapine treatment, clozapine is typically discontinued. Subsequent decisions about resuming clozapine are individualized and are based on the benefits and risks of treatment as compared with other therapeutic alternatives.

In patients who are treated with clozapine, a brief self-limiting fever (> 38°C) may also occur during the first few weeks of clozapine treatment; this fever responds to supportive measures (Bruno et al. 2015; Lowe et al. 2007; Pui-yin Chung et al. 2008). However, in a febrile patient, it is essential to assess for the presence of potentially life-threatening complications, including NMS, severe neutropenia, infection, and myocarditis.

Other potentially serious side effects of clozapine treatment include seizures, orthostatic hypotension, and gastrointestinal effects. When seizures occur with clozapine, they typically occur with very high doses of clozapine, rapid increases in clozapine dose, or shifts in medication levels (related to drug-drug interactions or effects of smoking on drug metabolism) (Devinsky et al. 1991). Therefore, a slow initial titration of clozapine dose is essential, and patients should be cautioned not to drive or engage in other potentially hazardous activities while clozapine is being titrated. If a seizure does occur with clozapine, dose adjustment may be needed, or adjunctive anticonvulsant medication (e.g., valproate) may be considered in conjunction with neurological consultation (Alldredge 1999; Wong and Delva 2007).

Orthostatic hypotension can also occur with clozapine and is most common with treatment initiation. Older patients and patients with peripheral vascular disease or a compromised cardiovascular status may be at particular risk. When severe, orthostatic hypotension can cause syncope, dizziness, or falls. Patients who experience orthostatic hypotension must be cautioned to sit on the edge of the bed for a minute before standing up, move slowly when going from lying down or sitting to standing, and seek assistance when needed. Management strategies for orthostatic hypotension include supportive measures (e.g., use of support stockings, increased dietary salt and fluid intake), reducing the speed of clozapine dose titration, and decreasing or dividing doses of clozapine. As a last resort, administration of the salt/fluid-retaining corticosteroid fludrocortisone can be considered to increase intravascular volume, but it is important to be mindful of the potential for immunosuppressive effects and development of diabetes with this medication (Mar and Raj 2018; Shen et al. 2017). For patients who are receiving concomitant antihypertensive treatment, adjustments to the dose of these medications may be needed.

Gastrointestinal effects of clozapine can also be significant and in some patients are associated with fecal impaction or paralytic ileus (Every-Palmer and Ellis 2017; Leung et al. 2017). Thus, the patient should obtain urgent medical care if experiencing constipation that is severe or does not resolve. To prevent development of constipation, it is useful to minimize the doses and number of contributory medications such as other anticholinergic medications and opioids. Activity and exercise should be encouraged to stimulate motility. A stool softener (e.g., docusate [Colace]) can be started for patients at increased risk (e.g., older patients). If constipation does develop, initial treatment can include stool softeners (e.g., docusate [Colace]) or osmotic laxatives (e.g., lactulose [Enulose], polyethylene glycol [Miralax], bisacody, [Dulcolax]). Second-line treatments include stimulant laxatives (e.g., senna [Senokot], senna tea, cascara, sodium picosulfate). If constipation persists, an enema (e.g., Fleet) should be considered. A combination of treatments may be needed to treat constipation and then to prevent its recurrence.

Side effects related to metabolic syndrome are common and generally are observed in the initial months of treatment but can also occur later in treatment. These side effects include weight gain (Alvarez-Jimenez et al. 2008; Leucht et al. 2013; Zhang et al. 2013); hyperlipidemia (Buhagiar and Jabbar 2019; Bushe and Paton 2005; Meyer and Koro 2004; Mitchell et al. 2013a); and glucose dysregulation, including development of diabetes mellitus (Hirsch et al. 2017; Ward and Druss 2015; Whicher et al. 2018; Zhang et al. 2017). Monitoring of body mass index, hemoglobin A1c, and lipid levels is important during clozapine treatment, as outlined in Statement 1, Table 2. If diabetes or hyperlipidemia is identified, it should be treated, typically by the patient’s primary care clinician. When weight gain occurs, it is usually progressive over the first 6 months of treatment, although some patients continue to gain weight indefinitely (Alvarez-Jimenez et al. 2008). Prevention of weight gain should, thus, be a high priority because weight loss is difficult for most patients. Efforts should be made to intervene proactively with weight gain of 5–10 pounds, and other medications that can cause weight gain (e.g., valproate) should be tapered and discontinued when possible. Dietary interventions, such as specialized behavioral health interventions, in-person community interventions (e.g., Weight Watchers), services that include meal delivery (e.g., Jenny Craig), or Internet-based interventions (e.g., Omada Health), should be suggested (Bonfioli et al. 2012).

In collaboration with the patient’s primary care clinician, metformin or nonstimulant medications for weight loss can be considered. Metformin has been shown to be safe in individuals without hyperglycemia and can reduce body weight and reverse metabolic abnormalities in patients with obesity or other metabolic problems (Das et al. 2012; de Silva et al. 2016; McGinty et al. 2016; Mizuno et al. 2014; Siskind et al. 2016a; Vancampfort et al. 2019; Zheng et al. 2015; Zhuo et al. 2018). Fewer studies have been done with glucagon-like peptide-1 agonist medications, but the available data suggest that body weight and metabolic risk factors are reduced by these medications as compared with placebo (Siskind et al. 2019). The benefits of exercise appear to be small in terms of weight loss in individuals with schizophrenia (Firth et al. 2015; Pearsall et al. 2014; Vancampfort et al. 2017, 2019). Nevertheless, many individuals with schizophrenia do not engage in physical activity (Stubbs et al. 2016a; Vancampfort et al. 2016b), and exercise can be suggested for its benefits for overall health, improved cardiorespiratory fitness, and other aspects of functioning (Dauwan et al. 2016; Firth et al. 2015, 2017; Vancampfort et al. 2017, 2019).

Sedation, sialorrhea, and tachycardia are each commonly observed during treatment with clozapine but can generally be managed conservatively. Sedation is most pronounced in the initial phases of treatment with clozapine, and many patients develop some tolerance to the sedating effects with continued administration. However, persistent sedation, including daytime drowsiness and increased sleep time, can interfere with social, recreational, and vocational function. Lowering of the daily dose, consolidating divided doses into one evening dose, or changing to a less sedating antipsychotic medication may be effective in reducing the severity of sedation. Coffee or other forms of caffeine can be helpful in the morning but can also interact with medications (e.g., contribute to tachycardia; raise blood levels of medications, including clozapine). Adding a stimulant medication is not typically helpful and can lead to additional side effects. If sedation or the risk of sedation is significant (e.g., during initial clozapine titration), patients should be cautioned not to drive or engage in potentially hazardous activities.

Sialorrhea (or hypersalivation) is also a frequent side effect of clozapine that can contribute to reductions in quality of life and complications such as aspiration pneumonia (Dzahini et al. 2018; Kaplan et al. 2018; Stoecker et al. 2017). Because sialorrhea may be more bothersome at night, patients may be advised to place a towel on their pillow and change to a clean towel in the middle of the night to minimize discomfort. During the day, patients can be encouraged to chew sugarless gum, which stimulates the swallowing reflex. Pharmacological approaches to address sialorrhea come from small studies and case reports and include use of low-dose or topical anticholinergic medications, such as glycopyrrolate or sublingual ophthalmic atropine 1% drops (Bird et al. 2011; Liang et al. 2010; Man et al. 2017). Diphenhydramine has also been studied (Chen et al. 2019); however, because clozapine and other antipsychotics can have significant anticholinergic properties themselves and anticholinergics have small effects on sialorrhea, the use of agents with added anticholinergic effects should be approached cautiously. Terazosin and, in severe refractory cases, botulinum toxin have also been used (Bird et al. 2011; Liang et al. 2010; Man et al. 2017).

Healthy patients can usually tolerate some increase in resting pulse rate, although this may not be the case for patients with preexisting heart disease. In patients with significant tachycardia (heart rates above 110–120 bpm), an ECG is warranted, as is assessment for other potential causes of tachycardia (e.g., fever, anemia, smoking, hyperthyroidism, respiratory disease, cardiovascular disorders, caffeine, other stimulants, and side effects of other medications). Early in treatment with clozapine, the possibility of myocarditis should be considered. If tachycardia is accompanied by pain, shortness of breath, fever, or signs of a myocardial infarction or heart rhythm problem, emergency assessment is essential. Management strategies for tachycardia with any antipsychotic medication include reducing the dose of medication; discontinuing medications with anticholinergic or stimulant properties; and addressing orthostatic hypotension, if present. Case reports have discussed the use of medications such as β-blocking agents for persistent and significant tachycardia with clozapine. Nevertheless, treatment is not indicated unless the patient is symptomatic or the patient’s heart rate is substantially greater than 120 bpm because data from more rigorous studies are not available and these medications can contribute to other side effects such as orthostatic hypotension (Lally et al. 2016a).

Side effects related to dopamine D2 receptor antagonism (e.g., acute dystonia, akathisia, medication-induced parkinsonism, NMS, tardive syndromes, hyperprolactinemia) can occur but are less frequent with clozapine than with many other antipsychotic medications. For additional information on the recognition and management of these side effects, see Statement 4, subsection “Treatment-Emergent Side Effects of Antipsychotic Medications.”

Other Approaches to Treatment-Resistant Schizophrenia

For all patients with treatment-resistant schizophrenia, it is important to conduct a review of the treatment plan at periodic intervals (see Statement 3). In addition to a review of prior medication trials, it is essential to review the psychosocial treatments that a patient has received and whether addition of one or more psychosocial interventions would be of benefit. For example, some patients may not have received cognitive-behavioral therapy for psychosis (CBTp) as recommended in Statement 16 or initial benefits of CBTp may have faded if CBTp was stopped. Under such circumstances, treatment with CBTp may be warranted (Burns et al. 2014; Morrison et al. 2018). Engaging families or persons of support in CBTp can also be helpful (Turkington and Spencer 2019). A similar review of potential additions to the treatment plan can occur with other psychosocial treatments.