Real-World Effectiveness, Economic, and Humanistic Outcomes of Selected Oral Antipsychotics in Patients with Schizophrenia: A Systematic Review Evaluating Global Evidence

Introduction

Globally, schizophrenia was ranked 20th amongst the top 25 leading causes of years lived with disability across all ages and 9th amongst the 25–49-year age group.¹ Notwithstanding its prevalence of 23.6 million and age-standardized prevalence of 287.4 per 100,000 people,¹ this severe and heterogeneous mental illness with no current cure is a significant driver of social, economic (with both direct and indirect costs to healthcare), and societal burden for the patient, caregivers, and healthcare systems.^2–5 Patients with schizophrenia often require lifelong treatment⁶ and experience a reduced life expectancy of up to 10–20 years than the general population.⁷ Numerous treatment choices, including those available generically, show considerable heterogeneity across patterns of effectiveness, medication adherence, product switching, and patient treatment burden, making it essential to identify novel management strategies for cost-effective and sustained outcomes in those affected by schizophrenia.

The burden of schizophrenia becomes evident during the initial disease stages;³ hence, treating patients during the first psychotic episode augments chances of disease improvement with optimal recovery and lowers relapse risk.^8–11 While oral antipsychotics remain the mainstay treatment for schizophrenia, with 65 oral antipsychotic formulations available globally,¹² the risk of relapses due to medication non-adherence is high.^13,14 Using long-acting injectables (LAIs) may increase medication adherence and improve patient outcomes.¹³ However, considerable variation remains in clinical practice patterns, patient treatment response, and patients’ perceptions and attitudes, considering the diverse efficacy outcomes of the disease and adverse effect profiles.¹² Besides, there exists variability in response to antipsychotics across countries, suggesting that different healthcare systems and cultural factors could have a global impact on medication patterns.¹⁵ Optimizing patient outcomes for an informed decision on appropriate antipsychotic treatment and best practice guidance is essential for enhanced patient experience. Significant work is needed to bridge this impending gap in identifying and achieving optimal therapeutic opportunities for schizophrenia.

Efforts are being made to identify a balanced benefit-risk profile of currently available antipsychotics to customize treatment to patient’s preferences and treatment requirements. To further understand the benefits of the therapeutic landscape, we conducted a systematic literature review to identify the real-world effectiveness, economic, and humanistic evidence that is available to decipher the value associated with ten globally available select oral antipsychotic treatments (OATs), namely asenapine, brexpiprazole, cariprazine, iloperidone, lumateperone, lurasidone, olanzapine/samidorphan, paliperidone, and quetiapine.

Methods

A systematic literature review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards.¹⁶

Literature Search Strategy

Bibliographic databases of PubMed, the EBSCOhost version of the American Psychological Association (APA) PsycINFO, and the Cumulative Index of Nursing and Allied Health Literature (CINAHL) were searched for relevant publications reporting real-world effectiveness, economic, and humanistic outcomes associated with OATs for schizophrenia from January 2010 to March 2022. Additionally, official websites and Health Technology Assessment (HTA) databases, including the National Institute for Health and Care Excellence (NICE), National Institute of Mental Health (NIMH), Database Commons, Cost Effectiveness Analysis (CEA) Registry, and the HTA Database (INAHTA) were searched for relevant studies published within the timeframe as mentioned above.

A primary search was conducted across the selected databases using controlled search terminology combining OATs (brand and generic names) for schizophrenia and terms related to “effectiveness”, “economic”, “humanistic”, “behavioral”, “adherence”, and “product switching”. A detailed list of keywords used is provided in Supplementary Table 1. A backward citation screening was conducted for the included studies.

Inclusion and Exclusion Criteria

All titles and abstracts obtained through literature searches were assessed against the eligibility criteria, including OATs, adults with schizophrenia, English language publications, and outcomes such as real-world effectiveness, economic outcomes, humanistic outcomes, behavioral outcomes, adherence/persistence, and product switching using a study screening hierarchy. Furthermore, the bibliographies of selected articles were reviewed to identify further articles of relevance. Studies in adults with schizophrenia analyzing real-world effectiveness, economic, or humanistic outcomes of the following OATs approved for schizophrenia (since 2006), namely, asenapine, brexpiprazole, cariprazine, iloperidone, lumateperone, lurasidone, olanzapine, olanzapine/samidorphan, paliperidone, and quetiapine were included.

Eligible studies were full-text articles published in the English language that mentioned at least one recently approved and/or available oral antipsychotic drug, at least one outcome concerning effectiveness endpoints in real-world studies, direct and indirect costs in pharmacoeconomic studies, including work productivity in patients or their caregivers; humanistic outcomes including quality of life (QoL) activities of daily living, and patient satisfaction, among others; behavioral outcomes; adherence/persistence; description of product switching; and drug characteristics, and article types including observational, randomized controlled trials (RCTs), non-randomized experimental, cohort-based, cross-sectional, case-control, and economic evaluation, having the measure of each outcome specified for inclusion. Ineligible studies were those reporting efficacy and safety outcomes (except for HTA databases), non-pharmacological treatment, and/or treatment with different routes of administration (such as a transdermal patch) in pediatric or non-human participants conducted in the United States (US), and type of articles such as conference abstracts, reviews, dissertations, commentaries, editorials, and summary reports.

Data Extraction and Analysis

A faculty investigator (KMK) and a librarian (DAN) supervised and guided the development of the search terms. To expand the search capabilities of each database, exclusive search strings were added, including the Medical Subject Headings terminology for PubMed, APA Thesaurus Terms for APA PsycINFO, and subject headings for the CINAHL. Covidence (Veritas Health Innovation, Melbourne, Australia), a web-based collaboration platform for producing systematic and other literature reviews, was used to screen and extract articles. Searches were executed on each database, and the results were imported to Covidence to remove duplications manually and through the program function. Two reviewers (KJ and KA) checked each electronic database for prospective records under a senior team member’s (KMK) supervision. The title/abstract of articles were identified and screened by two reviewers (KJ and KA) based on the disease, study population, study design, type and name of antipsychotics, and the outcome of interest. Full texts of the selected articles were then obtained, and reviewers conducted a final screening. Any discordance between the reviewers during article selection was arbitrated through discussion with a senior member (KMK).

Covidence automatically uploaded some full-text articles, and for the ones missing, the reports were manually uploaded by the two reviewers (KJ and KA), utilizing the university’s library databases. An extraction log of these publications on Covidence included pertinent information such as the study design and objectives, year of publication, study medications, the country where it was conducted, and the target population. The reviewers (KJ and KA) cleaned and verified the retrieved data. This procedure ensured that included studies satisfied all requirements and that accurate data was collected.

Ethics

Our analysis included previously published evidence and did not involve studies on animals or humans; therefore, ethics approval was not required.

Results

A total of 24,190 records were identified from different databases and HTA websites. Before the initial screening, 14,005 duplicate records were removed, and 10,185 were eligible for initial screening at the title/abstract level. Of the 633 relevant studies assessed for eligibility, most were excluded based on incorrect study design (n=107), those reporting safety and efficacy outcomes (n=112), and ones conducted only in the US (n=27). The US-based study findings have been presented earlier.¹⁷ After excluding studies that did not meet the specified criteria, 48 were selected for final review. In Figure 1, the PRISMA flow diagram summarizes the inclusion and exclusion of studies.

Figure 1 PRISMA flowchart for the selection of studies. Abbreviations: APA, American Psychological Association; CINAHL, Cumulative Index to Nursing and Allied Health Literature; US, United States.

Study Overview

The analysis included 48 studies, including 5 RCTs,^18–22 4 non-randomized experimental studies,^23–26 14 cohort studies,^27–40 10 cross-sectional studies,^41–50, and 15 economic evaluation studies.^51–65 All 48 studies were conducted in countries outside of the United States. The characteristics of these studies are presented in Supplementary Table 2.

The effectiveness, economic, humanistic, and behavioral changes, medication adherence, and product switching evidence for the respective OATs are summarized in Supplementary Table 3.

Diverse Treatment Characteristics and Study Populations

This systematic literature review included all studies that analyzed treatment effects or outcomes of OATs, including asenapine (Saphris^®, Sycrest^®), brexpiprazole (Rexulti^®, Rxulti^®), cariprazine (Reagila^®, Vraylar^®), iloperidone (Fanapt^®, Fanaptum^®), lumateperone (Caplyta^®), lurasidone (Latuda^®), olanzapine, olanzapine samidorphan (Lybalvi^®), paliperidone (Invega^®), and quetiapine fumarate (Seroquel^® slow release [XR]). Studies had diverse study populations and ranged from a limited number of participants (n=20) to over 200,000.^46,54 Similarly, diverse subsets of schizophrenia patients were included in the selected studies. Two studies had a population with psychosis associated with schizophrenia (one cohort study³⁶ and one that evaluated economic outcomes).⁵⁴ In comparison, two studies included patients with schizoaffective disorders (one non-randomized experimental study²⁵ and one that evaluated economic outcomes).⁶²

Effectiveness of Oral Antipsychotics

Effectiveness was measured in 21 studies, including two RCTs,^21,22 three non-randomized experimental studies,^23,24,26 nine cohort studies,^{27,29–31,35–37,39,40} and seven cross-sectional studies.^{41,42,44,46–48,50} A real-world observational study that compared the effectiveness of different antipsychotic drugs in patients with schizophrenia and schizoaffective disorders reported that clozapine (adjusted hazard ratio [HR]: 0.36, P<0.0001) was more effective when compared to first-generation “conventional” drugs (adjusted HR: 1.22, P<0.0001) and LAI second-generation oral antipsychotics (SGAs; adjusted HR: 0.56, P<0.0001) while considering hospital discontinuations for mental disorders.²⁹ In a large cohort study of 64,442 patients, clozapine demonstrated worse survival rates for psychiatric admissions when compared to olanzapine (HR: 0.615).³⁵ Joo et al reported that patients administered LAI antipsychotics had the lowest risk for treatment discontinuation and psychiatric hospitalization compared to patients treated with typical antipsychotic drugs.³⁷ A cross-sectional study with 200 patients reported that risperidone was more effective in treating first-episode schizophrenia than aripiprazole and olanzapine.⁵⁰ Additional details are provided in Table 1.

Table 1 Effectiveness Outcomes Associated with Oral Antipsychotic Treatments

Economic Outcomes of Oral Antipsychotics

Sixteen studies measured the economic outcomes of OATs, including one cross-sectional study (Table 2).^45,51–65 A Brazil-based economic evaluation study involving 174,310 schizophrenia patients showed that atypical antipsychotics (clozapine, olanzapine, quetiapine, risperidone, and ziprasidone) were accountable for most treatment costs and psychiatric hospitalization costs and contributed significantly to direct medical expenses (≈80%).⁵¹ The mean annual costs per patient for olanzapine (US$ 2085.28+485.62) was the highest among atypical antipsychotics, while clozapine had the highest mean annual cost per patient for outpatient psychiatric care (US$ 1105.39+236.70) and psychiatric hospitalization (US$ 3509.34+854.21).⁵¹ Olanzapine was found to be less costly than LAIs for the maintenance treatment of schizophrenia in France.⁵² Another study conducted in Spain showed that paliperidone extended-release (ER) had higher quality-adjusted-life years (QALYs) gained per patient (0.7573) and was less costly than risperidone (€3194), haloperidol (€3322), olanzapine (€3893), amisulpride (€4247), and aripiprazole (€4712).⁵³ A United Kingdom-based economic evaluation study reported that amisulpride was the most cost-effective drug (0.39), followed by risperidone (0.30) and olanzapine (0.17), assuming a willingness to pay threshold of £20,000 (equivalent to $25,552) per QALY gained.⁵⁴ In a multi-country study, it was observed that lurasidone had the lowest lifetime cost for acute treatment compared with all other atypical antipsychotics considered.⁵⁵ Asenapine was dominant over olanzapine from a societal perspective in a Canada-based cost-effectiveness study.⁵⁷ An economic evaluation of OATs in 461 patients reported that cariprazine produced better health outcomes than risperidone, resulting in an estimated QALY gain of 0.029 per patient following one year of treatment.⁶⁰ Phanthunane et al reported that risperidone was the cheapest drug, and treating severe schizophrenic patients with clozapine was associated with an increased cost-effectiveness ratio of 320,000 Baht per disability-adjusted life years (DALYs).⁶¹ A German-based economic evaluation study reported that the number of bed days was comparatively lower in patients treated with typical antipsychotics than those with atypical antipsychotics with a mean predicted treatment cost of €6442 (atypicals) and €4443 (typical) (P<0.0001).⁶⁴

Table 2 Economic Outcomes Associated with Oral Antipsychotic Treatments

Humanistic and Behavioral Outcomes of Oral Antipsychotics

Nine studies measured humanistic outcomes for oral antipsychotics.^{18–20,23,28,32,38,43,49} A QoL study on 1133 patients reported that patients receiving olanzapine and quetiapine showed significant improvements in the role-psychological score of the SF-36 and the goal attainment scale score compared to those receiving chlorpromazine.²³ Gründer et al reported promising improvement in QoL from SGAs compared to First Generation Antipsychotics.³² Taipale et al reported that second-generation LAIs reduced the risk of parental psychiatric healthcare use, while OATs were associated with an increased risk of parental psychiatric healthcare use.³⁸ Patients treated with olanzapine and risperidone reported impaired QoL.⁴³ Patient adherence to medications and perceived general health were significantly higher in patients treated with paliperidone palmitate than with SGAs.⁴⁹ Additional details are presented in Table 3.

Table 3 Humanistic and Behavioral Outcomes Associated with Oral Antipsychotic Treatments

Only one study investigated the effect of three SGAs (olanzapine, aripiprazole, ziprasidone) and their association with different clusters of symptoms and insight (ability to recognize the nature of symptoms and the importance of medication adherence).²⁸ Bianchini et al reported that the three SGAs were significantly more effective than haloperidol (overall P<0.001) and showed better efficacy than haloperidol in insight improvement.²⁸ However, no difference was reported among the different SGAs. An association between insight improvement and psychopathology was also detected (Table 3).

Adherence to Oral Antipsychotics

Two studies assessed medication adherence with SGAs for schizophrenia (Table 4).^27,49 Barbosa et al evaluated the effectiveness of olanzapine and risperidone in patients with schizophrenia; olanzapine had a significantly longer time until treatment discontinuation (P=0.021), and risperidone had a higher discontinuation rate (P=0.021).²⁷ Sağlam Aykut et al identified that medication adherence and general health scores of patients treated with paliperidone palmitate were significantly higher (P<0.001) than those with SGAs, and the medication adverse effects on the patient’s daily performance were substantially lower.⁴⁹

Table 4 Adherence and Product Switching Associated with Oral Antipsychotic Treatments

Impact of Product Switching of Oral Antipsychotics

Five studies identified the effect of product switching with OATs.^{18,19,25,33,34} Awad et al reported improvements in adherence-related attitude and psychosocial functioning in patients who switched to lurasidone from quetiapine, risperidone, aripiprazole, or ziprasidone.¹⁹ An experimental study that evaluated the efficacy, tolerability, and safety of switching from oral olanzapine to risperidone LAI reported significant changes from baseline to study end in the positive and negative syndrome scale (PANSS), the Clinical Global Impression-Severity scale, and Global Assessment of Functioning endpoint efficacy outcomes among 96 patients switching from oral olanzapine to risperidone LAI (P<0.0001).²⁵ Hatta et al identified that augmentation with olanzapine would be superior to switching to olanzapine among early nonresponders to risperidone, and augmentation with risperidone would be superior to switching to risperidone among early nonresponders to olanzapine.³³ Hong et al found that patients switching to olanzapine were less likely to experience relapse (HR: 3.43, 95% CI: 1.43 to 8.26) and extrapyramidal symptoms (Odds ratio [OR]: 4.02, 95% CI: 1.49 to 10.89) and amenorrhea/galactorrhea (OR: 8.99, 95% CI: 2.30 to 35.13).³⁴ Additional details on product switching can be found in Table 4.

Evidence from HTA Databases

Additionally, we reviewed HTA reports on the selected OATs (Supplementary Table 4). Of the 10 HTA articles identified,^66–75 seven studied adverse events,^{66,68–71,74,75} six investigated efficacy,^{66,68–71,74} four examined cost-effectiveness,^69,70,72,75 and one each studied effectiveness⁶⁷ and humanistic outcomes of OATs,⁷³ respectively.

Adverse events were measured and included treatment-emergent adverse events, side effects, metabolic effects, and extrapyramidal symptoms. The most significant side effects include asthenia/ lassitude/fatigue, somnolence/sedation, paresthesia, change in visual accommodation, increased salivation, diarrhea, and weight loss.^{66,68–71,74,75} Efficacy was measured with regards to an improvement in PANSS and related scores,^{66,68–71,74} while cost-effectiveness was measured in cost per QALY,^69,70,72,75 and humanistic outcomes included improvement in QoL.⁷³ Lurasidone significantly improved PANSS, Young Mania Rating Scale, and Montgomery-Asberg Depression Rating Scale scores^66,71 and considerably enhanced cost savings.⁷² The assessments suggested that amisulpride was associated with a more significant side effect burden, including cardiac side effects.⁷⁰ Moreover, economic analyses indicated that in the shorter term, amisulpride augmentation has the potential to be cost-effective.^69,70 Overall, the findings showed that atypical antipsychotic drugs have a favorable efficacy profile, and amisulpride was more effective than other options for treating schizophrenia.^67,69,70

Discussion

Understanding the evolving treatment landscape for schizophrenia is critical for optimal patient management. Our systematic review synthesizes this landscape by scrutinizing several important outcome measures globally, including drug effectiveness, economic considerations, humanistic and behavioral outcomes, medication adherence, and switching of OATs. Overall, the evidence from our systematic review of the ex-US OAT studies was consistent with those reported recently in the US.¹⁷ In addition to identifying the economic impact of OATs on schizophrenia, our review also highlighted a gap in studies assessing humanistic and behavioral outcomes associated with OATs. This finding underscores the need for assessing the humanistic burden related to schizophrenia not only on the patients but also on their family members or other caregivers. Humanistic outcomes may help explain the high prevalence of treatment discontinuation or medication non-adherence frequently observed across studies. Therefore, an opportunity exists to generate evidence-based data to assist physicians, researchers, and payers in directing appropriate treatment for schizophrenia for effective disease management.

Our review observed that the real-world effectiveness of therapeutic options for schizophrenia is a crucial parameter that determines the design of treatment regimens and choices. Any potential treatment/drug is appraised by its ability to diminish the clinical, economic, and humanistic burdens of the disease. According to the 2018 International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Special Task Force report, several value-based factors, including disease severity, insurance value, and value of hope, are being overlooked in drug value assessments, causing inaccurate resource utilization appraisals.⁷⁶ The report has listed 12 value elements to consider when reporting drug value assessments.⁷⁶ Many of these elements are directly associated with a treatment modality’s economic and humanistic outcomes. Our review could identify a few reports concerning economic and humanistic results and minimal regarding behavioral outcomes. Given the burden of schizophrenia on patients and caregivers in terms of productivity loss, most of these economic studies have not explicitly provided the impact of treatments on indirect costs. Another challenge with economic evaluations is to account for switching among OATs, especially with treatment differences in adverse events, adherence, and persistence. Furthermore, only one study reported the association of three OATs with symptom clusters and insight.²⁸ Some patients with schizophrenia exhibit problematic behaviors such as violence or suicidal behavior, and there is a need to evaluate behavioral outcomes in these patients, which will help improve intervention strategies and reduce recurrence. Exploring these outcome measures in value assessments open a window of opportunity to demonstrate greater incremental value for OATs and/or any new drug.

It is well known that in patients with schizophrenia, inadequate medication adherence is a significant barrier to optimal symptom control despite adequate drug efficacy.⁷⁷ An interruption of therapy by the patient can also trigger relapses and subsequent hospitalizations.^78–80 Our review identified seven observational studies reporting treatment adherence (n=2) and product switching (n=5). Olanzapine was reported to have higher treatment adherence with a higher continuation rate (50.8% vs 27.3% for asenapine)⁴⁸ and longer time until treatment discontinuation (p=0.021 vs risperidone),³⁹ while risperidone exhibited an increased risk of discontinuation (p=0.021 vs olanzapine).³⁹ In another study, asenapine was reported to have high discontinuation due to bitter taste and dosing method burden.⁴⁸ Other behavioral and/or humanistic outcomes could contribute to low adherence and high discontinuation of OAT; however, these measures remain primarily unexplored for various OATs, forming an evidence gap in this landscape. Treatment adherence has been associated with reduced relapse event rate, healthcare utilization, and costs, leading to significant savings for commercial payers.⁸¹ A meta-analysis reported that LAIs significantly lower hospitalization rates than OATs (risk/reward ratio: 0.85, 95% CI: 0.78 to 0.93, P<0.001).⁸² Although there is evidence of poor adherence to oral antipsychotics, many clinicians decline the use of LAI antipsychotics in the early stages of schizophrenia, assuming patients would favor oral antipsychotics,⁸³ and that LAIs are generally administered to very sick and difficult-to-handle patients.⁸¹ Our study however, did not explore the underlying parameters leading to OAT medication adherence/non-adherence; this should be a potential area of research for clinicians in the future.

Limitations

Our systematic review should be interpreted relative to certain limitations. First, the studies included in this systematic review were considerably varied regarding study medications, patient populations, and study outcomes. Hence, the degree of coverage of individual themes and measures and extrapolation of one finding to another should be interpreted cautiously. Second, considering this heterogeneity of available evidence for OATs, a formal quality assessment utilizing a standardized tool like the Cochrane Collaboration’s tool⁸⁴ for assessing bias risk could not be performed, but only a qualitative appraisal was conducted. Third, grey literature was not considered for our review. Even though grey literature does not fall into the inclusion criteria to maintain quality standards, first-person descriptions regarding humanistic, caregiver, and familial burdens of schizophrenia are being overlooked. This can increase the risk of reporting bias in the study. Fourth, reports specifying drug-related mortality, lifelong disability, and severe adverse events were not considered in our review. These determinants could influence the economic burden of the disease. Fifth, the effect of OATs on schizophrenia was not studied on the elderly or geriatric population, and this forms a significant gap in the present-day research (indicated by the bibliometric analysis), especially in a scenario where current evidence points out the fact that there is a high prevalence of schizophrenia at the age of 60 years.⁸⁵

Conclusions

Our analysis identified that a single antipsychotic medication or dosage is not best for all patients. Therefore, making decisions about changing a patient’s treatment, including the choice of antipsychotic medication and dose, requires careful consideration and ongoing, shared, collaborative decision-making by the clinician-patient dyad. With scarce evidence on humanistic and behavioral outcomes, developing relevant research questions and implementing well-designed studies that address such impending gaps is the hour’s need. Developing economic and budget models congruent with the ISPOR value flower elements could lead to better global resource allocations and strategies for effective disease management but requires greater evidence, including humanistic outcomes. A comprehensive approach to current treatment options utilizing outcome measures encompassing clinical, economic, behavioral, and humanistic factors is needed to fill the much-needed gap of disease awareness, impact on patient QoL, and comparative treatment value.