International Clinical Psychopharmacology:
The social and economic burden of treatment-resistant schizophrenia: a systematic literature review
Kennedy, James L.a; Altar, C. Anthonyb; Taylor, Danielle L.a; Degtiar, Irinac; Hornberger, John C.c,d
aDepartment of Psychiatry, University of Toronto, Toronto, Ontario, Canada
bAssureRx Health Inc., Mason, Ohio
cCedar Associates LLC, Menlo Park
dDepartment of Internal Medicine, Stanford University School of Medicine, Stanford, California, USA
Correspondence to John C. Hornberger, MD, MS, Cedar Associates LLC, 3715 Haven Ave., Suite 100, Menlo Park, CA 94025, USA Tel: +1 650 257 3315; fax: +1 650 257 3328; e- mail: firstname.lastname@example.org
Received December 24, 2012
Accepted July 04, 2013
Patients with schizophrenia often fail to respond to an initial course of therapy. This study systematically reviewed the societal and economic burden of treatment-resistant schizophrenia (TRS). Studies that described patients with TRS published 1996–2012 were included if they collected primary data on clinical, social, or economic outcomes. All studies were independently reviewed and extracted by at least two investigators. Sixty-five studies were identified. Almost 60% (SD 18%) of patients failed to achieve response after 23 weeks on antipsychotic drug therapy. Patients with TRS had high rates of smoking (56%), alcohol abuse (51%), substance abuse (51%), and suicide ideation (44%). The incidence of severe adverse events to treatment was 4% (SD 7%). Mean quality of life for patients who were unresponsive or intolerant to treatment was ∼20% lower than that of patients in remission. Annual costs for patients with schizophrenia are $15 500–$22 300 and are 3–11-fold higher for patients with TRS. TRS remains common and costly, despite availability of many treatment options, and contributes to a significant loss in patient quality of life. Although estimates in the literature vary greatly, TRS conservatively adds more than $34 billion in annual direct medical costs in the USA.
Almost 50% of the US population reports at least one incident of psychiatric disorder symptoms in their lifetime (Mental Disorders and Illicit Drug Use Expert Group, 2007). According to the National Institute of Mental Health (NIMH), the 12-month prevalence rate of schizophrenia in the US adult population is 1.1% (National Institute of Mental Health, 2011). The WHO ranks schizophrenia as one of the top 10 causes of disability among people in developed countries (Barbato, 1998). In addition, the high annual total costs associated with schizophrenia in the USA, estimated at more than $75 billion in 2012 dollars (Wyatt et al., 1995; Wu et al., 2005; Eaton, 2008), underscore the large economic impact of the disease. Schizophrenia is associated with high levels of social burden as well as immeasurable pain and suffering to afflicted individuals and those who care for them (Barbato, 1998; Eaton, 2008). Mortality rates among people with schizophrenia are more than twice those in the general population, and life expectancy is 12–15 years shorter than that of the general population, often as a result of personal neglect, poverty, and suicide (Barbato, 1998; Mcgrath et al., 2008; Van Os and Kapur, 2009; Wahlbeck et al., 2011).
Schizophrenia is a treatable condition, with many therapeutic alternatives, yet more than half of those with schizophrenia do not receive appropriate treatment (Kane et al., 1988; Barbato, 1998). In addition, 20–30% fail to respond to treatment, and a similar fraction of treatment-adherent patients relapse despite sustained maintenance therapy (Kane et al., 1988, 2007; Conley and Kelly, 2001). One of the most widely cited definitions of ‘treatment-resistant schizophrenia’ (TRS) was first proposed by Kane et al. (1988), and is described by a failure to respond to at least three periods of treatment with neuroleptic agents (from at least two different chemical classes) when administered at appropriate dosages for a period of 6 weeks. However, subsequent studies of treatment resistance in the literature define TRS more conservatively as the presence of pervasive positive symptoms despite one or more antipsychotic trials of adequate length and dosage (Breier and Hamilton, 1999; Lindenmayer et al., 2002; Kane et al., 2011; Suzuki et al., 2012), in which case failure to respond to at least one antipsychotic agent more thoroughly captures those patients for whom some resistance to disease management is evident.
Several studies have explored the various consequences of TRS to patients and society, including rate of symptom response, decreases in labor force participation, and greater need for medical resources. Most studies document a subset of outcomes, but none provide a comprehensive, systematic, and rigorous review of known consequences and the subsequent overall burden of TRS. The aim of this systematic literature review was to comprehensively assess the characteristics and burden of TRS in the USA by compiling available data from published studies on clinical, social, and economic outcomes. Factors that most often affect the societal burden of TRS and potential intervention improvements also were investigated.
MEDLINE was searched using PubMed (National Center for Biotechnology Information, U.S. National Library of Medicine) for reports published between January 1996, the year DSM-IV criteria were published, and May 2011 (see Appendix 1 for search terms, http://cedarecon.com/BurdenOfTRSchizophrenia.html). The search was limited to published studies that had primary data collection (e.g. ‘journal article’); case reports, reviews, editorials, and letters were omitted. Other requirements included English language studies and those including humans. Studies from any region worldwide meeting these criteria were included. The search was augmented by scanning article titles in the reference section of relevant publications and by searching the Tufts Cost-Effectiveness Analyses Registry (Center for the Evaluation of Value and Risk in Health, 2011), which is an online repository of cost-effectiveness studies that provides data on standardized measures of health status (utility weights).
Each abstract of the retrieved references was read by two researchers and screened for the presence of relevant study characteristics and methods (Fig. 1). References were excluded if the abstract failed to refer to primary data collection as a part of the study (e.g. meta-analysis, reviews, etc.), the endpoints were not pertinent to clinical, social, or economic outcomes, the study primarily assessed pharmacokinetic or in-vitro endpoints, treatments were not FDA approved, or the study dealt with an adolescent population (<18 years of age). The entire text of each selected study was then reviewed, and relevant data were extracted and recorded independently by two researchers. These two separate extractions were subsequently reconciled into the final list of data for analyses.
For a comprehensive review, treatment resistance was broadly defined as having failed to respond to one or more adequate trials of an antipsychotic therapy administered for 4 or more weeks and at appropriate doses (Breier and Hamilton, 1999; Lindenmayer et al., 2002; Kane et al., 2011; Suzuki et al., 2012). Treatment response was defined by a 20% or greater decrease in the Brief Psychiatric Rating Scale (BPRS) or total Positive and Negative Syndrome Scale (PANSS), 30% or greater decrease in Scale for the Assessment of Positive/Negative Symptoms (SAPS/SANSS), or a Clinical Global Impression Scale for Severity (CGI-S) of 3 or less. Study participants who were described as ‘resistant’ or ‘refractory’ to treatment were also accepted as ‘treatment resistant’. Studies reporting on a mix of patients with TRS and TR schizoaffective (SA) disorder were also included, as patients with SA have features of both schizophrenia and mood disorders (Brannon, 2012).
Data extraction and assessment of study quality
Extracted data included, were available: authors, year of publication, journal, study design (e.g. randomized control trial, observational), quality of evidence (evidence grade), sample size, mean age, sex distribution, ethnic background, duration of illness at study entry, number of prior psychotic episodes, number of study arms and patients per arm, study length, and mean or median patient survival.
Data were extracted on medical resource use and response rates, duration of use, and costs of antipsychotic drugs, as well as the incidence and cost of medical and psychiatric hospitalizations, emergency room visits, and physician visits. All costs are reported in 2012 US dollars using the medical component of the Consumer Price Index to adjust for inflation (U.S. Bureau of Labor Statistics, 2012). Rates of labor force participation, absenteeism, presenteeism (attending work while not healthy), homelessness, crime, and use of social services were documented when available. We extracted data on incidence, symptom severity, deaths, and costs related to comorbidities and adverse events occurring during the study period. Mortality and suicide rates were recorded, including number and prevalence of previous suicide attempts, suicidal ideation, nonsuicidal self-injury, and suicide attempts resulting in death. Mean baseline and changes in quality-of-life scales, utility scales, quality-adjusted life years (QALYs), and disability-adjusted life years (DALYs) were obtained for patients and caregivers, if reported. Mean baseline and/or changes in symptom severity reported by standardized psychosis assessment scales were recorded. Available data on the proportion of patients with TRS who had a treatment response were also extracted.
The quality of evidence for each study was assessed using the Quality Index developed by the Mental Disorders and Illicit Drug Use Expert Group (2007). The Quality Index includes items pertaining to the completeness of case ascertainment reporting, measurement instruments, diagnostic criteria, outcomes, and follow-up. The index was developed to quantify and assess the representativeness of the studies and ensure transparency of the studies’ quality. The lowest (worst) and highest (best) achievable scores on the Quality Index are zero and 19, respectively.
Evidence tables were developed to facilitate data entry using MS Excel worksheets. Analyses were performed using STATA, version 9.2 (Stata Corp, College Station, Texas, USA). Frequency distributions were generated for categorical variables (e.g. sex). Distributions of continuous variables (e.g. sample size) were summarized using mean, median, SD, and minimum and maximum values. All summary statistics were weighted by sample size. Separate analyses were conducted for studies with a pure TRS population versus those with a mixed TRS/TR SA population. Because of the heterogeneity of the study populations, study designs, and treatments, a pooling of the published data into a formal statistical meta-analysis was not performed.
Literature search and study characteristics
Four hundred and forty articles were identified in the electronic search (Fig. 1). After the title and abstract review, 155 were included for full-text screening. Data from a final set of 65 articles were used for analysis, of which 13 had a mixed TRS/TR SA population, two reported separate data for patients with TRS and TR SA, and eight reported data on health utilities for patients with schizophrenia (Appendix 2, http://cedarecon.com/BurdenOfTRSchizophrenia.html). Of the 44 TRS articles representing 41 studies, most reported findings from nonrandomized, single-arm trials or comparative randomized trials, with 52% of studies having two or more arms (Table 1). The patient sample size per study was 85 on average (SD 107), and ranged from 12 to 518 patients per study. The weighted average study length was 50 months (SD 25). The Quality Index score varied between 4 and 18, for a mean score of 11 (SD 4).
Baseline patient characteristics
The patients with TRS in the included studies had a mean age of 39 years (SD 4), a mean age of onset at 22 years (SD 2), and mean duration of illness of 15 years (SD 4) (Table 2). Men represented 66% (SD 14%) of the study populations. Ethnicity varied widely across studies, with averages of 57% Caucasian (SD 31%), 39% African-American (SD 29%), 6% Asian/Pacific-Islander (SD 23%), and 16% other (SD 12%). The sum of ethnicity exceeds 100% because of variability in reporting.
Mixed patients with TRS/TR SA had similar characteristics to those with TRS [mean age of 40 years (SD 2), mean age of onset of 22 years (SD 2), mean duration of illness of 17 years (SD 5), 65% male (SD 9%), 77% Caucasian (SD 18%), 14% African-American (SD 13%), 4% Asian/Pacific-Islander (1 study), and 8% other (SD 8%)].
Patients with TRS had failed an average of 1.8 drugs at the start of the studies (SD 0.7, range 1–4.5, nine studies) and three drug classes (SD 0, two studies) (Table 3). Two studies defined TR as failure of one antipsychotic trial. An additional 14 studies had failure of two or more antipsychotic trials as eligibility criteria, and five studies required failure of three or more antipsychotic trials. Eleven studies required failure of two or more drug classes.
Symptom severity was most commonly assessed by the CGI-S, BPRS, and PANSS, among the 19 different scales used in the studies (Table 3). Average scores at baseline for patients with TRS were 5.3 on the CGI-S, 52.3 on the BPRS, 21.7 on the PANSS positive scale, 24.6 on the PANSS negative scale, and 92.7 on the PANSS total scale. Consistent with the treatment resistance of these patients, the average scores at the end of the studies were 4.5 on the CGI-S, 39.1 on the BPRS, 16.8 on the PANSS positive scale, 19.6 on the PANSS negative scale, and 75.2 on the PANSS total scale (Fig. 2a). Decreases in scores averaged only 14–25% for the different severity scales, again consistent with outcomes in a treatment-resistant population. Studies including a mix of patients with TRS/TR SA reported similar, 13–29%, average severity score decreases and similar severity scores at baseline to the TRS-only group (Fig. 2b).
The prevalence of health risk behaviors was higher in TRS than in the general population, including smoking (56 vs. 21%), abuse of alcohol (51 vs. 5–15%), and abuse of other substances (51 vs. 15%) (Table 2) (Center for Behavioral Health Statistics and Quality, 2011; Centers for Disease Control and Prevention, 2011; National Center for Health Statistics, 2011). By comparison, general patients with schizophrenia were estimated to have a 55–65% smoking rate (Barnes et al., 2006; Bobes et al., 2010; Ferreira et al., 2010), 27–35% alcohol abuse and 28–35% substance abuse rates (Blanchard et al., 2000; Kavanagh et al., 2002; Barnes et al., 2006). These rates fall between those reported for the TRS population and the general population. Other common TRS comorbidities included aggressive behavior (79%), and the iatrogenic side effects of akathisia (36%), tardive dyskinesia (30%), and Parkinson’s disease (27%) (Table 2). Fifty-four percent of patients had suicidal ideation or attempted suicide.
Treatments consisted of a variety of drugs, including the second generation antipsychotics clozapine, olanzapine, risperidone, the first generation antipsychotic haloperidol, cognitive behavioral therapy, and electroconvulsive therapy. Response rates varied widely after an average treatment duration of 23 weeks. Patients with TRS on pharmacological antipsychotic therapies had an average response of 41% (SD 18%) and a range from 0 to 74% (for patients with TRS on chlorpromazine to ziprasidone, respectively) (Table 4) (Conley et al., 1998; Loebel et al., 2007). Consistent with its greater prophylactic effect in resistant patients, electroconvulsive therapy resulted in an 88% response rate after 3 weeks (Chanpattana and Kramer, 2003). Four studies explicitly mentioned nonresponse rates (patients who did not respond or partially responded), for an average of 44% (SD 7). Mixed patients with TRS/TR SA had a similar average response rate, 43% (SD 16%, range 9–90%), and a 31% nonresponse rate (one study) over a weighted average of 26 weeks of treatment (SD 55).
The incidence of adverse events in TRS was reported in 22 studies (Table 5). Sacchetti et al. (2009) found that 75% of patients experienced adverse events, and Kane et al. (2011) found that 6% of patients had serious adverse events. We found that mild/moderate adverse events, and severe adverse events, had average prevalence rates of 14 and 4%, as reported by 14 and 13 studies, respectively (Table 5). The most frequent severe adverse events were parkinsonism (28%), weight gain resulting in obesity (18%), tremors (17%), and apragmatism (17%). The most common mild or moderate adverse events were drowsiness (39%), headache (35%), sedation (33%), orthostatic hypotension (27%), lethargy (25%), and increased appetite (25%). An average weight gain of 2.5 kg (SD 1.5) was observed over the course of the studies that reported on weight gain (mean trial length of 14.3 weeks; SD 3.5 weeks). For mixed patients with TRS/TR SA, the most frequent severe adverse events were hypersalivation (40%), confusion (30%), and fatigue (30%). The most common mild or moderate adverse events were irritability (40%), decreased concentration (30%), impaired memory (30%), and parkinsonism (27%).
Costs and quality of life
No QALYs or DALYs were reported in the literature for patients with TRS. A Tufts CEA registry search gave health utility weights for patients with schizophrenia (Fig. 3). On a scale between 0 (death) and 1 (perfect health), studies reported an average of 0.56 for severe schizophrenia (SD 0.11), 0.62 for moderate schizophrenia (SD 0.22), and 0.79 for mild schizophrenia (SD 0.09) (Center for the Evaluation of Value and Risk in Health, 2011). See Fig. 3 for a list of terms included in the mild/moderate/severe classifications. Davies et al. (2008) found a 0.61 utility for patients with schizophrenia unresponsive or intolerant to current treatment, whereas Mcintyre et al. (2010) found that patients in remission have a utility of 0.75, an increase of 0.14. Reported adverse event utilities varied from a maximum of −0.37 for agranulocytosis to a minimum of −0.06 for akathisia.
Blieden et al. (1998) and Percudani et al. (1999)) provided information on inpatient and outpatient direct medical costs for TRS, whereas Buckman and Malan (1999) reported on TRS hospitalization costs in the USA, and Luchins et al. (1998) reported these costs for patients with TRS/TR SA. Comparative costs for patients with schizophrenia were obtained from Peng et al. (2011a).
Antipsychotic drug costs for patients with TRS were half the average per-person costs for all drugs in the USA, and for all patients with TRS/TR SA, they were the same as the average per-person costs for all drugs in the USA (Fig. 4) (Organisation for Economic Co-operation and Development, 2010). Drug costs reported for patients with general schizophrenia are six-fold higher than those reported for patients with TRS, and mixed patients with TRS/TR SA had antipsychotic drug costs two-fold higher than those for patients with TRS.
In contrast to drug costs, which are relatively low and a small proportion of total costs (Fig. 4), TRS hospitalization costs and total health resource utilizations were 60- and 19-fold higher, respectively, than for the general US population. Both are 10-fold higher than those for general schizophrenia, and four- and five-fold higher, respectively, than for mixed TRS/TR SA. Contributing to the high hospitalization costs, patients with TRS in these studies had, on average, seven hospitalizations over their lifetime, one hospitalization per year, and 75 hospitalized days per year. In contrast, patients with general schizophrenia have 0.31 inpatient psychiatric admissions and 5.3 hospitalized days in the 6 months after treatment (Peng et al., 2011a). Hospitalization costs comprised 45% of total health resource use for patients with TRS, 55% for the mixture of patients with TRS/TR SA, 46% of general schizophrenia costs, and 14% for the general US population. Blieden et al. (1998), which summarizes comprehensive information for TRS costs within the USA, reported $170 812 for hospitalization costs, $5443 for outpatient costs, and $196 896 for total health resource use in 2012 US dollars.
Studies reported total healthcare costs between $18 182 and $49 846 per patient-year for other schizophrenia subgroups that did not meet this review’s TRS criteria but that appeared to be indicative of the TR population (Table 6) (Wyatt et al., 1995; Blieden et al., 1998; Percudani et al., 1999; Mccombs et al., 2000; Fuller et al., 2002; Wu et al., 2005; Jones et al., 2006; Ascher-Svanum et al., 2008; Ascher-Svanum et al., 2010; Nicholl et al., 2010; Peng et al., 2011a, 2011b). Annual per-patient costs for general schizophrenia, in comparison, are reported between $15 483 and an excess of $22 313 over a nonschizophrenic population (Wyatt et al., 1995; Fuller et al., 2002; Wu et al., 2005; Peng et al., 2011a). Patients with early TRS (those who did not respond after 2 weeks of treatment) had healthcare costs that were from 1.2 to 2.1 times those of early responders (Ascher-Svanum et al., 2008; Peng et al., 2011b), whereas patients with schizophrenia who had a prior relapse, a sign of treatment resistance, had 2.8 times higher healthcare costs than those with no prior relapse (Ascher-Svanum et al., 2010). Untreated patients had costs 1.2 times higher ($8000 more) than treated patients (Mccombs et al., 2000).
No studies directly assessed the loss of productivity for TRS in the USA, but unemployment rates provide a related measure. Ciapparelli et al. (2003) reported that 65% of patients with TRS in an Italian population were unemployed, compared with a 10.4% unemployment rate in the general Italian population in 2000 (Index Mundi, 2011). Unemployment rates among all patients with schizophrenia in the USA are reported to exceed 70% (Marwaha and Johnson, 2004; Rosenheck et al., 2006; Marwaha et al., 2007; Salkever et al., 2007).
No reports were identified on the association between TRS or TR SA disorder and mortality rates, crime rates, or utilization of social services. The cost to nonmedical personnel who serve as informal caregivers and effects on their quality of life also have not been reported in the published literature.
Schizophrenia affects ∼2.6 million adults in the USA, of which approximately one-fifth or 500 000 are treatment resistant. The burden of TRS on patient well being is evidenced in this review by decreases in quality of life, presence of disease-associated and treatment-associated adverse events, increasing medical costs, increasing rates of serious comorbidities, and increasing suicide risk. Symptom severities of patients with TRS differ little at the end of drug trials from those of treatment-naive schizophrenia patients at the beginning of drug trials (Olivares et al., 2009; Macfadden et al., 2011). The small therapeutic responses of TRS patients show that they remain difficult to treat even after multiple treatment trials. Applying a stricter definition of TRS than used in this review would most likely define even lower response rates and higher burdens of illness. The multiple, unsuccessful treatment prolongs the burden of illness for TRS patients and increases their susceptibility to and the occurrence of tardive dyskinesia, akathesia, pseudo-Parkinson’s disease, antipsychotic-induced weight gain, and metabolic syndrome. Poor diet and less attention to personal hygiene and health are well-known and common manifestations of TRS that contribute to more frequent comorbidities (Peet, 2004; Mayo Clinic, 2012).
The studies summarized in this review contained 57% Caucasian patients, thus slightly under-representing the 72% of the US population that is Caucasian (Humes et al., 2011). Antipsychotic drug doses have largely been based on studies of Caucasians, and thus drug doses may be inappropriate for some ethnic groups because of ethnic differences in drug metabolism. Indeed, the few studies that have evaluated allele variants among Caucasians and non-Caucasian populations show population differences in drug metabolism and response that are not simply because of under-representation of certain genotypes in ethnic groups (Burroughs et al., 2002; Subramaniam et al., 2007; Fijal et al., 2009; Bigos et al., 2011). However, antipsychotic prescribing is practiced similarly between ethnicities (Connolly et al., 2011), which may need to be reconsidered to improve treatment responsiveness in ethnic populations.
The quality of life decrement associated with TRS is similar in magnitude to that for patients with stroke or end-stage renal disease who are undergoing maintenance dialysis (Center for the Evaluation of Value and Risk in Health, 2011). In comparison with other mental disorders, TRS has a less severe utility weight (i.e. quality of life, 0.61) than major depressive disorder patients who do not respond to therapy (0.3–0.55) (Nuijten, 2001; Sava et al., 2009; Benedict et al., 2010). As expected, patients with schizophrenia in remission have a better utility weight (0.75) than patients with TRS, and slightly better than those with major depression who respond to therapy (0.64–0.7) (Nuijten, 2001; Benedict et al., 2010). The utility losses of TRS are extrapolated to result in 145 000 QALYs lost annually in the USA (based on 237 million US adults, a 1.1% prevalence of schizophrenia, which includes 20% TRS prevalence, 0.89 average utility for the general US population, and 0.61 utility for patients with TRS) (Kane et al., 1988, 2007; Conley and Kelly, 2001; Noyes et al., 2007). Utility losses for responsive, but more prevalent, forms of schizophrenia are expected to result in an additional 293 000 QALYs lost in the US population (based on 2.6 million patients with schizophrenia, 80% response rate, 0.89 average utility for the general US population, and 0.75 utility for schizophrenia patients in remission).
Ranging from $66 360 to $163 795, total health resource utilization cost estimates for patients with TRS were 3–11-fold higher than those of the general schizophrenia population, which range from $15 500 to $22 300 in different studies. However, as reported in the results, drug costs showed the opposite trend, being six-fold higher for patients with general schizophrenia. This may be because of greater compliance with medication prescriptions among responsive patients. The TRS costs identified by Blieden et al. (1998) far exceeded other costs identified for ‘TRS-like’ populations. The patient population in Blieden et al. (1998) consisted of TRS inpatients supported by public funds; therefore, these patients had a more severe form of the disease [their QLS score was far lower than the average found in this review, indicating poorer functioning (23.1 vs. 42.8)], and their BPRS score was higher, indicating more severe psychopathology (58.7 vs. 52.3)]. Also, patients in other studies did not have TRS as defined by this review’s criteria.
A study on TRS costs by Meltzer et al. (1993) published before this review’s 1996–2011 inclusion criteria found that patients with TRS incurred $104 434 in hospitalization and outpatient costs. This estimate is closer to Blieden et al. (1998), suggesting that TRS costs are indeed far higher than those of other schizophrenia subgroups. Using the Percudani et al. (1999) conservative estimate, TRS adds more than $34 billion in annual direct medical costs within the USA in 2012 US dollars. The total direct medical cost of schizophrenia is estimated to add between $43 and $58 billion annually in the USA (Wyatt et al., 1995; Fuller et al., 2002). These data suggest that even with a conservative estimate for the costs of TRS, the 20% of the schizophrenia population who are treatment resistant constitutes about 60–80% of schizophrenia’s cost burden.
Davies and Drummond (1994) found similar results: care for the 10% of patients with schizophrenia who have a long-term disabling course accounts for 80% of total lifetime direct costs for schizophrenia within Great Britain; 75% of these are because of inpatient or residential care, whereas less than 5% are because of drug costs. The present review found that hospitalization costs were almost half of TRS’s total health resource costs, which is likely an underestimation because of variability in reporting. Examining each study individually, hospitalization costs were 87 and 69% of total health resource usage in Blieden et al. (1998) and Percudani et al. (1999), respectively. The large impact of not receiving effective therapy necessitates more research to clarify the magnitude of TRS costs and to reduce treatment resistance in schizophrenia to lessen its personal and societal burden.
SA disorder has a lifetime prevalence of ∼0.3% (Perala et al., 2007). The TR SA population experiences a slightly less severe burden relative to TRS as measured by common schizophrenia severity scales, the percentage of substance abuse in the population, and by other measures summarized in the results (Fig. 4 and Tables 2–6). TR SA is likewise less resource intensive, with lower overall health utilization costs, although drug costs are reported to be higher, possibly because of treating the affective components of TRS which may in turn increase compliance for medicines.
Our summary of the burden of TRS may be underestimated by the limited published research on costs, crime rates, use of social services, and quality-of-life burden to family members and caregivers. The burden of TRS is likely to have been further underestimated by the exclusion of adolescents from this review. As commonly observed for developmental brain disorders, the earlier appearance of schizophrenia in children and adolescents is frequently more severe and more likely to resist treatment than is adult-onset schizophrenia (Werry et al., 1991). Although 70–80% of adult patients with TRS respond to treatment (Kane et al., 1988, 2007; Conley and Kelly, 2001), only 17% of adolescent patients with schizophrenia were well at follow-up, indicating its lasting effects (Werry et al., 1991). More research is needed to understand the full burden and more effective treatments for child-onset and adolescent-onset schizophrenia (Young and Findling, 2004).
Other limitations to our review include the relatively short average length and small size of some studies, and heterogeneity of the study populations, treatments, and designs. The availability of only one study for some clinical measures, costs, hospitalization rates, comorbidities, and adverse events for patients with TRS, and different payer systems examined, further necessitates caution in generalizing the results.
Patient noncompliance almost certainly contributes to treatment nonresponse in schizophrenia, as better drug adherence is associated with a better prognosis and outcomes (Fenton et al., 1997; Jaeger et al., 2012). Compliance rates are as low as 50% after 6 months of therapy initiation, and patients take less than 60% of prescribed antipsychotic doses (Llorca, 2008). Compliance failure makes it difficult to assess treatment response. Patient education, long-lasting injections, and identification of treatments that are better suited to individual patients may help improve compliance and hence treatment response.
Innovations in personalized medicine may help improve treatment outcomes for patients with TRS. One emerging field is pharmacogenomics, in which genetic factors are used to predict and improve drug response. Pharmacogenomic tests are anticipated to help guide medication selection and dosing, hasten response rates, increase compliance, and improve quality of life for patients, and thereby reduce the escalating rates in annual health expenditures (Wilke and Dolan, 2011). For example, genetic polymorphisms for hepatic metabolizing enzymes could be used to help adjust doses or allow for selection of alternative antipsychotics for those who are poor or ultrarapid metabolizers, thereby increasing efficacy and minimizing treatment resistance (Kang et al., 2009; Suzuki et al., 2011). Variations in genes that produce aberrant hepatic enzyme activities and suboptimal drug effector proteins can predict greater healthcare utilization and costs (Winner et al., 2013) as well as poorer antidepressant efficacy (Hall-Flavin et al., 2012). In addition, a report that integrates these gene variants has, in a clinical setting, improved the efficacy of antidepressant medications (Hall-Flavin et al., 2012).
Genetic testing can also be used to identify patients prone to serious side effects. For instance, 5% of patients carry a genetic variant that puts them at increased risk for clozapine-induced agranulocytosis (CIA) (Athanasiou et al., 2011), and variants in genes for the serotonin 5-HT2C receptor (Sicard et al., 2010) and the melanocortin MCR4 receptor (Malhotra et al., 2012) predict risk for weight gain induced by clozapine and other antipsychotic agents. Identifying patients at greater risk for CIA could lessen the need for regular monitoring and potentially allow a large proportion of the 2.5 million patients at low risk for CIA to switch to clozapine. Patients with TRS may especially benefit from tests for CIA and weight gain, as among all antipsychotic drugs, clozapine produces the greatest clinical improvement in TRS (Kane et al., 1988; Meltzer et al., 1993; Jones et al., 2006). A 1993 study estimated a cost savings of $28 485 per patient-year (2012 US dollars equivalent) for patients with TRS who continued clozapine treatment for at least 2 years, compared with patients who discontinued clozapine therapy (Meltzer et al., 1993).
Such innovations are necessary given the substantial additional toll that resistance to treatment exacts on schizophrenia patients’ quality of life, productivity, and medical costs. Basic research is needed to better identify the mechanisms by which schizophrenia afflicts patients, and to understand genetic and gene–environmental factors that may underlie or modulate these mechanisms. Clinical research has continued to improve upon treatments, lowering the burden of schizophrenia, and lessening the side effects that accompany treatment (Jibson, 2011). New models for effectiveness research, early diagnosis, and improved education for clinicians and patients, when combined with more consistent processes for regulatory requirements and reimbursements of new therapeutics and diagnostics, will help to reduce the burden of TRS in the USA (Mrazek and Lerman, 2011; Wilke and Dolan, 2011).
The authors thank Amal Al-Saadi for her assistance with manuscript preparation and data abstraction and Hialy Gutierrez, Tiffany Yu, Kelly Lui, Austin Grier, and Varun Ektare for their assistance with data extraction.
Conflicts of interest
AssureRx Health (Mason, Ohio, USA) sponsored the study. C.A.A. is an employee of the sponsor and has been granted stock options in AssureRx Health. J.C.H. and I.D. are employees of Cedar Associates, LLC, which received funding to be the research coordinating center for this study. For the remaining authors there are no conflicts of interest.
Ascher-Svanum H, Nyhuis AW, Faries DE, Kinon BJ, Baker RW, Shekhar A .Clinical, functional, and economic ramifications of early nonresponse to antipsychotics in the naturalistic treatment of schizophrenia.Schizophr Bull. 2008; 34:1163–1171.
Ascher-Svanum H, Zhu B, Faries DE, Salkever D, Slade EP, Peng X, Conley RR .The cost of relapse and the predictors of relapse in the treatment of schizophrenia.BMC Psychiatry. 2010; 10:2
Athanasiou MC, Dettling M, Cascorbi I, Mosyagin I, Salisbury BA, Pierz KA, et al .Candidate gene analysis identifies a polymorphism in HLA-DQB1 associated with clozapine-induced agranulocytosis.J Clin Psychiatry. 2011; 72:458–463.
Barnes TR, Mutsatsa SH, Hutton SB, Watt HC, Joyce EM .Comorbid substance use and age at onset of schizophrenia.Br J Psychiatry. 2006; 188:237–242.
Benedict A, Arellano J, De Cock E, Baird J .Economic evaluation of duloxetine versus serotonin selective reuptake inhibitors and venlafaxine XR in treating major depressive disorder in Scotland.J Affect Disord. 2010; 120:94–104.
Bigos KL, Bies RR, Pollock BG, Lowy JJ, Zhang F, Weinberger DR .Genetic variation in CYP3A43 explains racial difference in olanzapine clearance.Mol Psychiatry. 2011; 16:620–625.
Blanchard JJ, Brown SA, Horan WP, Sherwood AR .Substance use disorders in schizophrenia: review, integration, and a proposed model.Clin Psychol Rev. 2000; 20:207–234.
Blieden N, Flinders S, Hawkins K, Reid M, Alphs LD, Arfken CL .Health status and health care costs for publicly funded patients with schizophrenia started on clozapine.Psychiatr Serv. 1998; 49:1590–1593.
Bobes J, Arango C, Garcia-Garcia M, Rejas J .Healthy lifestyle habits and 10-year cardiovascular risk in schizophrenia spectrum disorders: an analysis of the impact of smoking tobacco in the CLAMORS schizophrenia cohort.Schizophr Res. 2010; 119:101–109.
Bourin M, Guitton B, Dailly E, Hery P, Jolliet P .A follow-up study of a population of schizophrenic patients treated with clozapine.Prog Neuropsychopharmacol Biol Psychiatry. 2001; 25:1481–1495.
Breier A, Hamilton SH .Comparative efficacy of olanzapine and haloperidol for patients with treatment-resistant schizophrenia.Biol Psychiatry. 1999; 45:403–411.
Buckman RW, Malan RD .Clozapine for refractory schizophrenia: the Illinois experience.J Clin Psychiatry. 1999; 60:Suppl 1 18–22
Burroughs VJ, Maxey RW, Levy RA .Racial and ethnic differences in response to medicines: towards individualized pharmaceutical treatment.J Natl Med Assoc. 2002; 94:1–26.
Chanpattana W, Kramer BA .Acute and maintenance ECT with flupenthixol in refractory schizophrenia: sustained improvements in psychopathology, quality of life, and social outcomes.Schizophr Res. 2003; 63:189–193.
Chiu NY, Yang YK, Chen PS, Chang CC, Lee IH, Lee JR .Olanzapine in Chinese treatment-resistant patients with schizophrenia: an open-label, prospective trial.Psychiatry Clin Neurosci. 2003; 57:478–484.
Ciapparelli A, Dell'osso L, Pini S, Chiavacci MC, Fenzi M, Cassano GB .Clozapine for treatment-refractory schizophrenia, schizoaffective disorder, and psychotic bipolar disorder: a 24-month naturalistic study.J Clin Psychiatry. 2000; 61:329–334.
Ciapparelli A, Dell’osso L, Bandettini Di Poggio A, Carmassi C, Cecconi D, Fenzi M, et al .Clozapine in treatment-resistant patients with schizophrenia, schizoaffective disorder, or psychotic bipolar disorder: a naturalistic 48-month follow-up study.J Clin Psychiatry. 2003; 64:451–458.
Conley RR, Tamminga CA, Bartko JJ, Richardson C, Peszke M, Lingle J, et al .Olanzapine compared with chlorpromazine in treatment-resistant schizophrenia.Am J Psychiatry. 1998; 155:914–920.
Conley RR, Kelly DL .Management of treatment resistance in schizophrenia.Biol Psychiatry. 2001; 50:898–911.
Connolly A, Taylor D, Sparshatt A, Cornelius V .Antipsychotic prescribing in Black and White hospitalised patients.J Psychopharmacol. 2011; 25:704–709.
Davies LM, Drummond MF .Economics and schizophrenia: the real cost.Br J Psychiatry Suppl. 1994; 165:18–21.
Davies LM, Barnes TR, Jones PB, Lewis S, Gaughran F, Hayhurst K, et al .A randomized controlled trial of the cost-utility of second-generation antipsychotics in people with psychosis and eligible for clozapine.Value Health. 2008; 11:549–562.
Eaton W .The burden of mental disorders.Epidmiol Rev. 2008; 30:1–14.
Fenton WS, Blyler CR, Heinssen RK .Determinants of medication compliance in schizophrenia: empirical and clinical findings.Schizophr Bull. 1997; 23:637–651.
Ferreira L, Belo A, Abreu-Lima C .A case–control study of cardiovascular risk factors and cardiovascular risk among patients with schizophrenia in a country in the low cardiovascular risk region of Europe.Rev Port Cardiol. 2010; 29:1481–1493.
Fijal BA, Kinon BJ, Kapur S, Stauffer VL, Conley RR, Jamal HH, et al .Candidate–gene association analysis of response to risperidone in African-American and white patients with schizophrenia.Pharmacogenomics J. 2009; 9:311–318.
Flynn SW, Macewan GW, Altman S, Kopala LC, Fredrikson DH, Smith GN, Honer WG .An open comparison of clozapine and risperidone in treatment-resistant schizophrenia.Pharmacopsychiatry. 1998; 31:25–29.
Fuller MA, Shermock KM, Secic M, Laich JS, Durkin MB .Service use and costs among VA patients with schizophrenia taking risperidone or olanzapine.Psychiatr Serv. 2002; 53:855–860.
Hall-Flavin DK, Winner JG, Allen JD, Jordan JJ, Nesheim RS, Snyder KA, et al .Using a pharmacogenomic algorithm to guide the treatment of depression.Transl Psychiatry. 2012; 2:e172
Henna Neto J, Elkis H .Clinical aspects of super-refractory schizophrenia: a 6-month cohort observational study.Rev Bras Psiquiatr. 2007; 29:228–232.
Humes KR, Jones NA, Ramirez RR .Overview of Race and Hispanic Origin: 2012. 2010 Census Briefs. 2011; .Washington, DC:US Census Bureau.
Jaeger S, Pfiffner C, Weiser P, Kilian R, Becker T, Langle G, et al .Adherence styles of schizophrenia patients identified by a latent class analysis of the Medication Adherence Rating Scale (MARS): a six-month follow-up study.Psychiatry Res. 2012; 200:83–88.
Jones PB, Barnes TR, Davies L, Dunn G, Lloyd H, Hayhurst KP, et al .Randomized controlled trial of the effect on Quality of Life of second- vs first-generation antipsychotic drugs in schizophrenia: Cost Utility of the Latest Antipsychotic Drugs in Schizophrenia Study (CUtLASS 1).Arch Gen Psychiatry. 2006; 63:1079–1087.
Kane J, Honigfeld G, Singer J, Meltzer H .Clozapine for the treatment-resistant schizophrenic. A double-blind comparison with chlorpromazine.Arch Gen Psychiatry. 1988; 45:789–796.
Kane JM, Meltzer HY, Carson WH Jr, Mcquade RD, Marcus RN, Sanchez R .Aripiprazole for treatment-resistant schizophrenia: results of a multicenter, randomized, double-blind, comparison study versus perphenazine.J Clin Psychiatry. 2007; 68:213–223.
Kane JM, Potkin SG, Daniel DG, Buckley PF .A double-blind, randomized study comparing the efficacy and safety of sertindole and risperidone in patients with treatment-resistant schizophrenia.J Clin Psychiatry. 2011; 72:194–204.
Kang RH, Jung SM, Kim KA, Lee DK, Cho HK, Jung BJ, et al .Effects of CYP2D6 and CYP3A5 genotypes on the plasma concentrations of risperidone and 9-hydroxyrisperidone in Korean schizophrenic patients.J Clin Psychopharmacol. 2009; 29:272–277.
Kavanagh DJ, Mcgrath J, Saunders JB, Dore G, Clark D .Substance misuse in patients with schizophrenia: epidemiology and management.Drugs. 2002; 62:743–755.
Kelly DL, Feldman S, Boggs DL, Gale E, Conley RR .Nonresponse to clozapine and premorbid functioning in treatment of refractory schizophrenia.Compr Psychiatry. 2010; 51:298–302.
Lindenmayer JP, Czobor P, Volavka J, Lieberman JA, Citrome L, Sheitman B, et al .Olanzapine in refractory schizophrenia after failure of typical or atypical antipsychotic treatment: an open-label switch study.J Clin Psychiatry. 2002; 63:931–935.
Llorca PM .Partial compliance in schizophrenia and the impact on patient outcomes.Psychiatry Res. 2008; 161:235–247.
Loebel AD, Khanna S, Rajadhyaksha S, Siu CO, Giller E, Potkin SG .Ziprasidone in treatment-resistant schizophrenia: a 52-week, open-label continuation study.J Clin Psychiatry. 2007; 68:1333–1338.
Luchins DJ, Hanrahan P, Shinderman M, Lagios L, Fichtner CG .Initiating clozapine treatment in the outpatient clinic: service utilization and cost trends.Psychiatr Serv. 1998; 49:1034–1038.
Macfadden W, Desouza C, Crivera C, Kozma CM, Dirani RD, Mao L, Rodriguez SC .Assessment of effectiveness measures in patients with schizophrenia initiated on risperidone long-acting therapy: the SOURCE study results.BMC Psychiatry. 2011; 11:167
Malhotra AK, Correll CU, Chowdhury NI, Muller DJ, Gregersen PK, Lee AT, et al .Association between common variants near the melanocortin 4 receptor gene and severe antipsychotic drug-induced weight gain.Arch Gen Psychiatry. 2012; 69:904–912.
Martin J, Gomez JC, Garcia-Bernardo E, Cuesta M, Alvarez E, Gurpegui M .Olanzapine in treatment-refractory schizophrenia: results of an open-label study. The Spanish Group for the Study of Olanzapine in Treatment-Refractory Schizophrenia.J Clin Psychiatry. 1997; 58:479–483.
Marwaha S, Johnson S .Schizophrenia and employment – a review.Soc Psychiatry Psychiatr Epidemiol. 2004; 39:337–349.
Marwaha S, Johnson S, Bebbington P, Stafford M, Angermeyer MC, Brugha T, et al .Rates and correlates of employment in people with schizophrenia in the UK, France and Germany.Br J Psychiatry. 2007; 191:30–37.
Mccombs JS, Luo M, Johnstone BM, Shi L .The use of conventional antipsychotic medications for patients with schizophrenia in a medicaid population: therapeutic and cost outcomes over 2 years.Value Health. 2000; 3:222–231.
McEvoy JP, Freudenreich O, Wilson WH .Smoking and therapeutic response to clozapine in patients with schizophrenia.Biol Psychiatry. 1999; 46:125–129.
Mcgrath J, Saha S, Chant D, Welham J .Schizophrenia: a concise overview of incidence, prevalence, and mortality.Epidemiol Rev. 2008; 30:67–76.
Mcintyre RS, Cragin L, Sorensen S, Naci H, Baker T, Roussy JP .Comparison of the metabolic and economic consequences of long-term treatment of schizophrenia using ziprasidone, olanzapine, quetiapine and risperidone in Canada: a cost-effectiveness analysis.J Eval Clin Pract. 2010; 16:744–755.
Meltzer HY, Cola P, Way L, Thompson PA, Bastani B, Davies MA, Snitz B .Cost effectiveness of clozapine in neuroleptic-resistant schizophrenia.Am J Psychiatry. 1993; 150:1630–1638.
Mrazek D, Lerman C .Facilitating clinical implementation of pharmacogenomics.JAMA. 2011; 306:304–305.
Nair CJ, Josiassen RC, Abraham G, Stanilla JK, Tracy JI, Simpson GM .Does akathisia influence psychopathology in psychotic patients treated with clozapine? Biol Psychiatry. 1999; 45:1376–1383.
. National Center for Health Statistics .Health, United States, 2010: with special feature on death and dying. 2011; .Hyattsville, MD:National Center for Health Statistics.
Nicholl D, Akhras KS, Diels J, Schadrack J .Burden of schizophrenia in recently diagnosed patients: healthcare utilisation and cost perspective.Curr Med Res Opin. 2010; 26:943–955.
Noyes K, Dick AW, Holloway RG .The implications of using US-specific EQ-5D preference weights for cost-effectiveness evaluation.Med Decis Making. 2007; 27:327–334.
Nuijten MJ .Assessment of clinical guidelines for continuation treatment in major depression.Value Health. 2001; 4:281–294.
Olivares JM, Peuskens J, Pecenak J, Resseler S, Jacobs A, Akhras KS .Clinical and resource-use outcomes of risperidone long-acting injection in recent and long-term diagnosed schizophrenia patients: results from a multinational electronic registry.Curr Med Res Opin. 2009; 25:2197–2206.
. Organisation for Economic Co-operation and Development .OECD Health Data. OECD Health Statistics. 2010; .Paris, France:Organisation for Economic Co-operation and Development.
Peet M .Diet, diabetes and schizophrenia: review and hypothesis.Br J Psychiatry Suppl. 2004; 47:S102–S105.
Peng X, Ascher-Svanum H, Faries D, Conley RR, Schuh KJ .Decline in hospitalization risk and health care cost after initiation of depot antipsychotics in the treatment of schizophrenia.Clinicoecon Outcomes Res. 2011a; 3:9–14.
Peng X, Ascher-Svanum H, Faries DE, Stauffer VL, Kollack-Walker S, Kinon BJ, Kane JM .Cost-effectiveness of early responders versus early nonresponders to atypical antipsychotic therapy.Clinicoecon Outcomes Res. 2011b; 3:79–87.
Perala J, Suvisaari J, Saarni SI, Kuoppasalmi K, Isometsa E, Pirkola S, et al .Lifetime prevalence of psychotic and bipolar I disorders in a general population.Arch Gen Psychiatry. 2007; 64:19–28.
Percudani M, Fattore G, Galletta J, Mita PL, Contini A, Altamura AC .Health care costs of therapy-refractory schizophrenic patients treated with clozapine: a study in a community psychiatric service in Italy.Acta Psychiatr Scand. 1999; 99:274–280.
Rodriguez VM, Catalina ML, Garcia-Noblejas JA, Cuesta P .Schizophrenic syndromes and clozapine response in treatment-resistant schizophrenia.Psychiatry Res. 1998; 77:21–28.
Rosenheck R, Leslie D, Keefe R, Mcevoy J, Swartz M, Perkins D, et al .Barriers to employment for people with schizophrenia.Am J Psychiatry. 2006; 163:411–417.
Sacchetti E, Galluzzo A, Valsecchi P, Romeo F, Gorini B, Warrington L .Ziprasidone vs clozapine in schizophrenia patients refractory to multiple antipsychotic treatments: the MOZART study.Schizophr Res. 2009; 113:112–121.
Salkever DS, Karakus MC, Slade EP, Harding CM, Hough RL, Rosenheck RA, et al .Measures and predictors of community-based employment and earnings of persons with schizophrenia in a multisite study.Psychiatr Serv. 2007; 58:315–324.
Sava FA, Yates BT, Lupu V, Szentagotai A, David D .Cost-effectiveness and cost-utility of cognitive therapy, rational emotive behavioral therapy, and fluoxetine (Prozac) in treating depression: a randomized clinical trial.J Clin Psychol. 2009; 65:36–52.
Sicard MN, Zai CC, Tiwari AK, Souza RP, Meltzer HY, Lieberman JA, et al .Polymorphisms of the HTR2C gene and antipsychotic-induced weight gain: an update and meta-analysis.Pharmacogenomics. 2010; 11:1561–1571.
Subramaniam M, Ng C, Chong SA, Mahendran R, Lambert T, Pek E, Huak CY .Metabolic differences between Asian and Caucasian patients on clozapine treatment.Hum Psychopharmacol. 2007; 22:217–222.
Suzuki T, Uchida H, Watanabe K, Nakajima S, Nomura K, Takeuchi H, et al .Effectiveness of antipsychotic polypharmacy for patients with treatment refractory schizophrenia: an open-label trial of olanzapine plus risperidone for those who failed to respond to a sequential treatment with olanzapine, quetiapine and risperidone.Hum Psychopharmacol. 2008; 23:455–463.
Suzuki T, Mihara K, Nakamura A, Nagai G, Kagawa S, Nemoto K, et al .Effects of the CYP2D6*10 allele on the steady-state plasma concentrations of aripiprazole and its active metabolite, dehydroaripiprazole, in Japanese patients with schizophrenia.Ther Drug Monit. 2011; 33:21–24.
Suzuki T, Remington G, Mulsant BH, Uchida H, Rajji TK, Graff-Guerrero A, et al .Defining treatment-resistant schizophrenia and response to antipsychotics: a review and recommendation.Psychiatry Res. 2012; 197:1–6.
Tollefson GD, Birkett MA, Kiesler GM, Wood AJ .Double-blind comparison of olanzapine versus clozapine in schizophrenic patients clinically eligible for treatment with clozapine.Biol Psychiatry. 2001; 49:52–63.
Van Os J, Kapur S .Schizophrenia.Lancet. 2009; 374:635–645.
Wahlbeck K, Westman J, Nordentoft M, Gissler M, Laursen TM .Outcomes of Nordic mental health systems: life expectancy of patients with mental disorders.Br J Psychiatry. 2011; 199:453–458.
Werry JS, Mcclellan JM, Chard L .Childhood and adolescent schizophrenic, bipolar, and schizoaffective disorders: a clinical and outcome study.J Am Acad Child Adolesc Psychiatry. 1991; 30:457–465.
Wilke R, Dolan M .Genetics and variable drug response.JAMA. 2011; 306:306–307.
Winner J, Allen JD, Altar CA, Spahic-Mihajlovic A .Psychiatric pharmacogenomics predicts health resource utilization of outpatients with anxiety and depression in a staff model health maintenance organization.Transl Psychiatry. 2013; 3:e242
Wirshing DA, Marshall BD Jr, Green MF, Mintz J, Marder SR, Wirshing WC .Risperidone in treatment-refractory schizophrenia.Am J Psychiatry. 1999; 156:1374–1379.
Wu EQ, Birnbaum HG, Shi L, Ball DE, Kessler RC, Moulis M, Aggarwal J .The economic burden of schizophrenia in the United States in 2002.J Clin Psychiatry. 2005; 66:1122–1129.
Wyatt RJ, Henter I, Leary MC, Taylor E .An economic evaluation of schizophrenia – 1991.Soc Psychiatry Psychiatr Epidemiol. 1995; 30:196–205.
Young CM, Findling RL .Pharmacologic treatment of adolescent and child schizophrenia.Expert Rev Neurother. 2004; 4:53–60.
antipsychotic; burden of illness; costs; review; schizoaffective disorder; schizophrenia; treatment resistance
© 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
Highlight selected keywords in the article text.