A systematic review of the real-world effectiveness and economic and humanistic outcomes of selected oral antipsychotics among patients with schizophrenia in the United States: Updating the evidence and gaps

Abstract

BACKGROUND:

Schizophrenia is a chronic, relapsing, and burdensome psychiatric disorder affecting approximately 0.25%-0.6% of the US population. Oral antipsychotic treatment (OAT) remains the cornerstone for managing schizophrenia. However, nonadherence and high treatment failure lead to increased disease burden and medical spending. Cost-effective management of schizophrenia requires understanding the value of current therapies to facilitate better planning of management policies while addressing unmet needs.

OBJECTIVE:

To review existing evidence and gaps regarding real-world effectiveness and economic and humanistic outcomes of OATs, including asenapine, brexpiprazole, cariprazine, iloperidone, lumateperone, lurasidone, olanzapine/samidorphan, paliperidone, and quetiapine.

METHODS:

We conducted a literature search using PubMed, American Psychological Association PsycINFO (EBSCOhost), and the Cumulative Index of Nursing and Allied Health Literature from January 2010 to March 2022 as per Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. English-language articles describing adults with schizophrenia receiving at least 1 of the selected OATs and reporting real-world effectiveness, direct or indirect costs, humanistic outcomes, behavioral outcomes, adherence/persistence patterns, or product switching were identified.

RESULTS:

We identified 25 studies from a total of 24,190 articles. Real-world effectiveness, cost, and adherence/persistence outcomes were reported for most OATs that were selected. Humanistic outcomes and product switching were reported only for lurasidone. Behavioral outcomes (eg, interpersonal relations and suicide ideation) were not reported for any OAT. The key economic outcomes across studies were incremental cost-effectiveness ratios, cost per quality-adjusted life-years, and health care costs. In studies that compared long-acting injectables (LAIs) with OATs, LAIs had a higher pharmacy and lower medical costs, while total health care cost was similar between LAIs and OATs. Indirect costs associated with presenteeism, absenteeism, or work productivity were not reported for any of the selected OATs. Overall, patients had poor adherence to OATs, ranging between 20% and 61% across studies. Product switching did not impact the all-cause health care costs before and after treatment.

CONCLUSIONS:

Our findings showed considerable gaps exist for evidence on behavioral outcomes, humanistic outcomes, medication switching, and adherence/persistence across OATs. Our findings also suggest an unmet need regarding treatment nonadherence and lack of persistence among patients receiving OATs. We identified a need for research addressing OATs’ behavioral and humanistic outcomes and evaluating the impact of product switching in adults with schizophrenia in the United States, which could assist clinicians in promoting patient-centered care and help payers understand the total value of new antipsychotic drugs.

Plain language summary

Schizophrenia is a mental condition and is responsible for high burden of care for patients, the health care system, and society. The treatment includes oral antipsychotic medications, which help manage the signs and symptoms of the disease. As patients require long-term treatment, it is important to recognize the reasons for treatment failure. Our study will summarize the existing evidence, which will be useful in assessing the value of new oral antipsychotic drugs for schizophrenia.

Implications for managed care pharmacy

Frequent treatment failure in schizophrenia increases health care costs. Our systematic review indicated that treatment with oral antipsychotic treatments (OATs) for schizophrenia is impacted by nonadherence and lack of persistence. The findings illustrate a knowledge gap in humanistic and behavioral outcomes of OATs. By understanding the humanistic, economic, and clinical value of OATs, managed care decision-makers can help facilitate better planning of management policies while addressing unmet needs of patients.

Schizophrenia is a debilitating, chronic, and complex relapsing psychiatric disorder that poses a substantial clinical and economic burden on global health care systems. The global prevalence of schizophrenia in 2022, per the World Health Organization estimates, has been reported to be 0.32%, translating to approximately 24 million people living with schizophrenia.¹ In the United States, the prevalence of schizophrenia has been shown to range between 0.25% and 0.64%.² Both globally and in the United States, schizophrenia has been associated with a considerable humanistic and economic burden that results from impaired patient functioning, decreased quality of life (QoL), the burden to the caregivers and family members, and increased health care resource use and costs.^3-5 In the United States, schizophrenia is responsible for an estimated $62.3 billion in direct health care and up to $251.9 billion in indirect costs related to premature mortality, work productivity, and caregiving.⁵ In 2020, the cumulative societal burden of schizophrenia in the United States was more than $281.6 billion, making this condition an opportunity for appropriate management.⁶

A comprehensive patient-centered treatment plan, including pharmacological treatment, a psychosocial support system, and infrastructure, is considered an effective management strategy for schizophrenia.^7-11 Treatment goals in schizophrenia include the control of acute psychotic episodes, prevention of relapses, and the long-term stabilization of patients.¹² Current treatment approaches per national and international guidelines for patients with schizophrenia emphasize using antipsychotic medications.¹³ The American Psychiatric Association Practice Guideline for the Treatment of Patients with Schizophrenia recommends that patients be treated with antipsychotic medication and monitored for effectiveness and adverse effects.¹³ Globally, 65 antipsychotic medications are used (18 in the United States), and all are available in oral formulations.¹⁴

Although antipsychotics remain the cornerstone of treatment for schizophrenia, many patients remain uncontrolled or discontinue treatment. Understanding the benefits of these treatments beyond clinical efficacy and safety will enhance treatment decision-making for clinicians and assist payers in assessing the value of treatment options, while evaluating evidence gaps related to the value of antipsychotics may help identify unmet needs that impact optimal patient care.^15-18 This systematic literature review will categorize existing real-world effectiveness, economic, and humanistic evidence and identify knowledge gaps to guide future research on the value of OATs for the management of patients with schizophrenia. The review focuses on eligible studies published in the last 12 years on the following second-generation OATs: asenapine, brexpiprazole, cariprazine, iloperidone, lumateperone, lurasidone, olanzapine/samidorphan, paliperidone, and quetiapine.

Methods

SEARCH STRATEGY AND SELECTION CRITERIA

This systematic review was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to identify all eligible studies published between January 2010 and March 2022, with information on real-world effectiveness and economic and humanistic outcomes of OATs in patients with schizophrenia. The search strategy involved a combination of the Medical Subject Headings and keywords (Supplementary Table 1 ^{(245.4KB, pdf)} , available in online article). Searches were performed using PubMed, the EBSCOhost version of the American Psychological Association (APA) PsycINFO database, and the Cumulative Index of Nursing and Allied Health Literature. The search queries involved multiple terminologies to include all target criteria, such as the population, medications, and study outcomes. The key search terms included schizophrenia, oral antipsychotics (brand and generic names), adherence, behavioral outcomes, humanistic outcomes, economic outcomes, and product switching. A primary search was also conducted on health technology assessment reports; however, we did not include this in this manuscript per our focus on the US population.

Study Selection. Following a study screening hierarchy for inclusion and exclusion, all titles and abstracts identified through the literature searches were screened against the eligibility criteria that included OATs, adult patients with schizophrenia, publications in English language, and reported outcomes such as real-world effectiveness, economic outcomes, humanistic outcomes, adherence/persistence, product switching, and behavioral outcomes. Furthermore, bibliographies of identified articles were screened to identify additional studies of relevance. Title and abstracts were independently screened by 2 reviewers (K.J. and K.A.) and results were compared. The review of full-text articles was also done independently by the 2 reviewers (K.J. and K.A.), and in case of disagreement, a third reviewer (K.M.K.) resolved the discrepancies.

We included studies in adult patients with schizophrenia, analyzing real-world effectiveness, economic outcomes, or humanistic outcomes of the following US Food and Drug Administration–approved OATs for schizophrenia: asenapine (approved 2009); brexpiprazole (2015); cariprazine (2015); iloperidone (2009); lumateperone (2019); lurasidone (2010); olanzapine/samidorphan (2021); paliperidone (2006); and quetiapine (2007). Eligible studies were required to have been conducted in the United States, with full-text articles published in the English language and reporting one of the following outcomes: effectiveness endpoints in real-world studies; direct and indirect costs in pharmacoeconomic studies including work productivity studies in patients or their caregivers; humanistic outcomes including QoL, activities of daily living, and patient satisfaction, among others; adherence/persistence; behavioral outcomes; and description of product switching. Furthermore, we excluded articles reporting clinical (efficacy and safety) outcomes with youth and adolescent populations or nonhuman subjects, ones that evaluated nonpharmacological treatments, and article types such as review articles, poster abstracts, meta-analyses, theses/dissertations, commentaries, editorials, or summary reports.

Data Collection and Analysis. The search terms were built under the supervision and guidance of the faculty investigator (K.M.K.) and the librarian (D.A.N.). Exclusive search strings were added to each database to expand the search: Medical Subject Headings terminology for PubMed, APA Thesaurus Terms for APA PsycINFO, and subject headings for the Cumulative Index of Nursing and Allied Health Literature. Covidence, a web-based collaboration software, was used to streamline the production of this review.¹⁹ Searches were executed on each database, and the results were imported to Covidence¹⁹ to remove duplicate studies manually and through the program function. As a result of the initial literature search, titles and abstracts were identified and screened by 2 reviewers (K.J. and K.A.) based on the disease, study population, study design, type and name of antipsychotics, and the outcome of interest (Table 1). Questionable abstracts were included for a full-text review for a final decision. Covidence automatically uploaded available full-text articles, and for those missing, the articles were manually uploaded by the 2 reviewers, using West Virginia University library databases. Data, including study design and objectives, year/time frame, study medications, sample size, and key findings, were extracted from the articles using an extraction form on Covidence.¹⁹ The faculty investigator (K.M.K.) and the librarian (D.A.N.) reviewed the final extraction for confirmation. All disagreement during the review was resolved through consultation with the faculty investigator (K.M.K.).

TABLE 1.

List of Study Inclusion and Exclusion Criteria

Inclusion criteria	Exclusion criteria
Oral antipsychotic medications	Studies presenting efficacy and/or safety measures with nonpharmacological treatment and/or treatment with other routes of administration such as a transdermal patch
Studied in adult patients	Studies with other patient populations (pediatrics only or nonhuman studies)
Studies must be in the English language	Studies conducted outside the United States
Studies should include schizophrenia as a significant disease and should have an outcome that falls under one of the categories below: Real-world effectiveness Economic outcomes Humanistic outcomes Work productivity Adherence/persistent Product switching Behavioral outcomes	Study designs were: Study reviews Theses/dissertations Conference posters or abstracts Commentaries/editorials Summary reports

Results

EVIDENCE FROM PEER-REVIEWED ARTICLES

Inclusion of Studies. Figure 1 represents a flowchart of the procedure for the selection of studies. The initial literature search identified 24,190 articles, of which 14,005 were removed at de-duplication. The remaining 10,185 articles were reviewed for relevancy at the title and abstract stage, and of these, 9,463 were deemed irrelevant. The full text of the remaining 63 articles was examined for final inclusion, and studies were excluded if they were missing disease (ie, schizophrenia), oral medication, or outcome of interest (eg, economic and adherence). They were also excluded if they reported only efficacy or safety data, had the wrong intervention (nonpharmacologic therapy), patient population (pediatrics), route of administration (intramuscular), or study design (case reports). Finally, 25 articles were deemed eligible for inclusion in the study.

FIGURE 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses Flowchart for the Selection of Studies

Characteristics of Studies Evaluating Real-World Outcomes of Oral Antipsychotics. In addition to assessing effectiveness, studies obtained in this literature review evaluated other outcomes such as adherence, economic outcomes, humanistic outcomes, behavioral outcomes, and product switching. As shown in Table 2, 25 studies assessed such outcomes. Effectiveness and economic outcomes related to direct costs were reported for all drugs except lumateperone (Table 3). Adherence was evaluated for all drugs except brexpiprazole and cariprazine. Humanistic outcomes were reported only for lurasidone (Table 3). Indirect costs related to presenteeism, absenteeism, work productivity, and behavioral outcomes were not reported for any OAT (Table 3). Numerous studies examined adherence, effectiveness, economic outcomes, the impact of product switching, and/or humanistic outcomes of OATs as a class,^4,20-28 whereas others assessed the effect of an individual drug such as lurasidone,^29-34 brexpiprazole,^20,38 iloperidone,²⁴ olanzapine,^29,35,39,40 paliperidone,³¹ risperidone,^29,33 and quetiapine.²² Additionally, 15 studies were active-controlled, involving other antipsychotics^{20,28,30,37,38,40} and LAIs (Table 4).^{21,25,31,34-36,41-43}

TABLE 2.

Characteristics of Studies Included in the Systematic Review

Author and year	Study design	Drug name	Sample size	Outcomes
Aigbogun 201820	Economic evaluation Cost-effectiveness Decision analytic framework	Brexpiprazole vs cariprazine, lurasidone	1,000	Economic outcomes Effectiveness
Baser 201521	Economic evaluation Health care cost Resource utilization RWE	Paliperidone LAIOAT as a group	670	Economic outcomes Effectiveness
Berger 201222	Retrospective cohort study Admission and claims data	Aripiprazole, quetiapine, ziprasidone	127 (SCZ = 43; BD = 84)	Adherence/persistence
Citrome 201423	RCT	Lurasidone	198	Product switching
Cutler 201324	RCT	Iloperidone vs placebo, ziprasidone	173	Product switching
El Khoury 202025	Economic evaluation Cost analysis Decision tree model Medicaid payer’s perspective	OAT as a whole group included aripiprazole, asenapine maleate, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine, risperidone, ziprasidone	1 million	Economic outcomes Effectiveness
Kadakia 20224	Retrospective cohort study Cost analysis Medicaid database	OAT as a whole group included asenapine, brexpiprazole, cariprazine, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine fumarate, aripiprazole, clozapine, risperidone, ziprasidone	11,642	Economic outcomes Effectiveness
Lafeuille 201826	Retrospective cohort study Cost analysis Medicaid database	OAT as a whole group included asenapine, iloperidone, lurasidone, olanzapine, paliperidone, aripiprazole, risperidone, ziprasidone	8,667	Adherence/persistence Economic outcomes Effectiveness
Lefebvre 201727	Economic evaluation Cost analysis VA data	OAT as a whole group included aripiprazole, asenapine maleate, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine fumarate, risperidone, ziprasidone	6,872	Economic outcomes Adherence
Newcomer 201828	Retrospective cohort study Medicaid multistate database Cost analysis	Lurasidone vs quetiapine	673	Economic outcomes Effectiveness
Noordsy 201729	RCT	Olanzapine vs risperidone	107	Adherence
O’Day 201330	Economic evaluation 5-year Markov cohort cost-effectiveness model US payer perspective	Lurasidone, olanzapine, risperidone, ziprasidone, aripiprazole, quetiapine ER	NA	Economic outcomes
Patel 202031	Retrospective cohort Economic evaluation Cost analysis VA database	Paliperidone, risperidone	1,169	Economic outcomes Adherence Effectiveness
Patel 2021a32	Retrospective cohort study Cost analysis Medicaid database	OAT as a whole group	832	Economic outcomes Effectiveness Adherence
Patel 2021b33	RCT Post hoc analysis	Urasidone vs risperidone	590	Product switching
Pesa 201534	Retrospective cohort study Economic evaluation Cost analysis Medicaid data	OAT as a whole group included asenapine, iloperidone, lurasidone, olanzapine, paliperidone, aripiprazole, quetiapine, risperidone, ziprasidone	Not available	Economic outcomes Effectiveness
Pesa 201735	Retrospective cohort study Economic evaluation Cost analysis Medicaid data	Olanzapine vs aripiprazole, haloperidole, fluphenazine, risperidone	1,444	Economic outcomes Effectiveness Adherence
Pilon 201736	Retrospective cohort study Economic evaluation	Asenapine maleate, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine fumarate, aripiprazole, risperidone, ziprasidone	24,656	Economic outcomes Effectiveness Adherence
Rajagopalan 201737	Cross sectional study Medicaid data	Lurasidone vs olanzapine, aripiprazole, quetiapine, risperidone, ziprasidone	1,413	Adherence/persistence
Song 201938	Cohort study	Asenapine, iloperidone, lurasidone, olanzapine, quetiapine, aripiprazole, clozapine, olanzapine, fluoxetine, risperidone, ziprasidone	7,029	Adherence
Stahl 201339	RCT	Lurasidone vs olanzapine, placebo	254	Product switching
Weiser 202140	Retrospective observational study VA database	Lurasidone, olanzapine, olanzapine samidorphan, clozapine, quetiapine, aripiprazole, ziprasidone, risperidone	37,368	Effectiveness
Xiao 201641	Economic evaluation	OAT as a whole group included aripiprazole, asenapine maleate, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine fumarate, risperidone, ziprasidone	11,654	Economic outcomes
Yan 202042	Retrospective economic evaluation RWE Medicare supplemental database Medicaid database OPTUM Clinformatics data	Brexpiprazole vs lurasidone, olanzapine, paliperidone, ziprasidone, aripiprazole, quetiapine, risperidone	6,254	Economic outcomes
Young-Xu 201643	Economic evaluation Cost analysis Retrospective longitudinal study VA database	OAT as a whole group included aripiprazole, asenapine maleate, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine fumarate, risperidone, ziprasidone	10,290	Economic outcomes

BD = bipolar disorder; ER = extended release; LAI = long-acting injectable; NA = not applicable; OAT = oral antipsychotic treatment; RCT = randomized controlled trial; RWE = real-world evidence; SCZ = schizophrenia; VA = Veterans Affairs.

TABLE 3.

Outcomes Reported for Each Oral Antipsychotic Drug

OAT	Adherence	Behavioral	Economic	Effectiveness	Humanistic	Product switching
Asenapine	*26,27,36,38	NR	*4,25-27,34,36,41,43	*4,25-27,34,36	NR	NR
Brexpiprazole	NR	NR	√20,42	√20	NR	NR
Cariprazine	NR	NR	√20	√20	NR	NR
Iloperidone	*26,27,36,38	NR	*4,25-27,34,36,41,43	*4,25-27,34,36	NR	√24
Lumateperone	NR	NR	NR	NR	NR	NR
Lurasidone	√37	NR	√20,28,30,42	√20,28,40	NR	√23,33,39
Olanzapine	√29,37	NR	√30,42	√39,40	NR	NR
Paliperidone	√31	NR	√31	√31	NR	NR
Quetiapine	√22,37	NR	√28,30,42	√28,40	NR	NR

Asterisk indicates that the medication was only included in studies that considered OAT as a whole class.

NR = not reported; OAT = oral antipsychotic treatment.

TABLE 4.

Effectiveness, Health Economic, and Humanistic Outcomes Associated With Oral Antipsychotic Treatments

Author and year	Drug name	Study objective	Measures	Main findings
Aigbogun 201820	Brexpiprazole; cariprazine lurasidone	Assess the cost-effectiveness of brexpiprazole vs comparator branded therapies for reducing relapses and hospitalizations among adults with schizophrenia from a US payer perspective	Avoided relapses Avoided hospitalization QALYs Annual per-patient health care ICER	Brexpiprazole was the least costly and most effective treatment strategy for all outcomes—the lowest annual per-patient health care cost, the highest QALYs, and dominant ICER.
Baser 201521	OAT as a group; LAI paliperidone	Compare real-world health care costs and resource utilization between patients with schizophrenia treated with PP long-acting injection and OAT	Health care cost	PP-LAI had lower inpatient costs and admission rates than a matched cohort of OAT patients. Total health care costs were not significantly different.
Berger 201222	Aripiprazole; quetiapine; ziprasidone	Examine patterns of pharmacotherapy among patients who were hospitalized for schizophrenia or bipolar disorder and received aripiprazole, quetiapine, or ziprasidone at discharge	MPR	Medication compliance is poor in patients with SCZ or BD who initiate treatment with aripiprazole, quetiapine, or ziprasidone at hospital discharge. Approximately one-quarter of patients switched to another agent over this period.
Citrome 201423	Lurasidone	Evaluate the treatment failure in patients with schizophrenia and schizoaffective disorder who switched to lurasidone	Product switching (treatment discontinuation)	Lurasidone showed significant efficacy and continued to be generally well tolerated in patients with schizophrenia or schizoaffective disorder who had switched to lurasidone from a broad range of antipsychotic agents.
Cutler 201324	Iloperidone; placebo; ziprasidone	Evaluate the safety and tolerability of flexibly dosed iloperidone for treating schizophrenia over the extension phase of 25 weeks	Impact of treatment switching (CGI-S)	Switching to iloperidone showed improvement in CGI-S and CGI-C score.
El Khoury 202025	OAT as a group	Project the long-term economic impact when a proportion of nonadherent patients with a recent relapse switch from OAAs to PP1M	Number of relapses Pharmacy cost Relapse cost	Pharmacy costs associated with switching nonadherent OAA patients with a recent relapse to PP1M were offset by reduced relapse rates and health care costs at years 1, 2, and 3, with earlier use of PP1M resulting in increased cost savings in all years.
Kadakia 20224	OAT as a whole group	(1) Assess the 1-year incidence of EPS among patients with schizophrenia after initiating a second-generation antipsychotic (2) Assess the annual health care burden for patients who developed EPS compared with patients who did not	1-year incidence of EPS Health care resource utilization Health care cost	More than one-fifth of study patients initiating treatment with OAT developed EPS. EPS increases the risk of hospitalization and higher health care cost. Treatment with minimal risk of EPS may reduce the economic burden for patients with SCZ.
Lafeuille 201826	OAT as a whole group	Compare persistence, costs, and health care resource utilization in patients with schizophrenia and cardiometabolic comorbidities treated with PP1M or an OAA	Health care cost Health care utilization Adherence	Similar 12-month total health care costs between PPM1 and OAA. Lower health care utilization overall in the PP1M group, except for an outpatient visit. Higher adherence rate in PP1M.
Lefebvre 201727	OAT as a whole group	Compare all-cause and substance abuse–related health care resource utilization and costs in veterans with schizophrenia and co-occurring substance abuse who were treated with PP vs OAA	PDC Health care cost Health care utilization	Higher adherence, lower rates of substance abuse–related health care resource utilization, and significant total medical cost savings resulting from fewer hospitalizations in PP compared with OAA.
Newcomer 201828	Lurasidone; quetiapine	(1) Assess hospitalization rates among patients with schizophrenia who switched to antipsychotic monotherapy with lurasidone or quetiapine from other atypical antipsychotics in real-world settings (2) Describe treatment duration and total inpatient admission costs for Medicaid and commercial payers	Rate of hospitalization Health care cost	Patients with SCZ who switched to lurasidone had significantly fewer all-cause and mental health–related hospitalizations and similar rates of schizophrenia-related hospitalization, as well as significantly longer treatment duration rates compared with those switched to quetiapine.
Noordsy 201729	Olanzapine; risperidone	Compare the efficacy and tolerability of olanzapine vs risperidone among patients with schizophrenia who are established in outpatient psychiatric care and entering supported employment	Persistence	Persistence was higher in the olanzapine group compared with the risperidone group.
O’Day 201330	Lurasidone; olanzapine; risperidone; ziprasidone; aripiprazole; quetiapine ER	Evaluate the long-term cost-effectiveness (including hospitalizations and cardiometabolic consequences) of atypical antipsychotics among adults with schizophrenia	ICER	Risperidone was the least costly. Lurasidone was more costly but more effective than risperidone. Cost-effective at WTP thresholds of greater than $25,844 per hospitalization avoided. The favorable cost-effectiveness of lurasidone is driven by its clinical benefits (eg, efficacy in preventing hospitalizations in patients with schizophrenia) and its minimal cardiometabolic adverse effect profile.
Patel 202031	Paliperidone; risperidone	Compare treatment patterns, health care resource utilization, and medical costs before and after a switch from oral antipsychotic drug therapy to the long-acting injectable PP1M in patients with schizophrenia	Health care cost Health care resource utilization PDC	The decrease in total medical costs fully offset an increase in pharmacy costs, resulting in similar total costs post-PP1M switch, indicating the potential economic benefits of switching from oral RIS/PALI to PP1M in patients with schizophrenia.
Patel 2021a32	OAT as a group	Compare adherence, rates of subsequent schizophrenia-related relapses, health care resource utilization, and health care costs among Medicaid beneficiaries with schizophrenia who initiated PP1M vs a new OAT following a recent schizophrenia-related relapse	PDC Health care cost Health care utilization	PP1M was associated with a lower risk of subsequent relapse while remaining a cost-neutral therapeutic option compared with OAT.
Patel 2021b33	Lurasidone; risperidone	To evaluate the impact of switching from risperidone to lurasidone	Product switching PANSS CGI-S	No significant difference in PANSS and CGI-S improvement (lurasidone vs risperidone). Patients switching from risperidone experienced improvements in metabolic parameters and prolactin levels.
Pesa 201534	OAT as a group	Compare health care costs and resource utilization between PP1M and OAT in a population of Medicaid beneficiaries with schizophrenia	Health care cost Health care utilization	LAI may reduce inpatient and outpatient health care services utilization and associated costs. It may also lower the risk of hospitalization compared with OAT.
Pesa 201735	Olanzapine; aripiprazole; haloperidole; fluphenazine; risperidone	Compare all-cause health care utilization and costs between patients with schizophrenia treated with PP1M or OAT	Treatment discontinuation PDC Persistence Health care utilization Health care cost	PP1M had significantly fewer inpatient hospitalizations and associated costs, with no significant difference in the total costs between the 2 cohorts.
Pilon 201736	Asenapine maleate; iloperidone; lurasidone; olanzapine; paliperidone; quetiapine fumarate; aripiprazole; risperidone; ziprasidone; LAI	Compare treatment patterns, HCRU, and Medicaid spending in patients with schizophrenia initiated on LAIs (overall and according to the agent) vs OAT	PDC Persistence Health care cost Health care utilization	LAI had higher odds of being adherent and persistent in therapy. LAI had fewer long-term care visits and home services. PP-LAI was associated with significant medical cost savings, but other LAI costs were similar to OAT. LAI had a higher cost of 1-day mental health institute visits.
Rajagopalan 201737	Lurasidone; olanzapine; aripiprazole; quetiapine;	Compare adherence with lurasidone with other oral atypical antipsychotics among Medicaid and commercially insured patients with schizophrenia	MPR Discontinuation rate Mean time to discontinuation	Lurasidone demonstrated greater adherence compared with other atypical antipsychotics in both the Medicaid and commercially insured
	risperidone; ziprasidone			cohorts. No significance was observed in the mean time to discontinuation (days).
Song 201938	Asenapine; iloperidone; lurasidone; olanzapine; quetiapine; aripiprazole; clozapine; olanzapine/ fluoxetine; risperidone; ziprasidone; LAI	Determine whether LAIs are associated with better persistency compared with OAT among Medicaid recipients with schizophrenia	Persistence	Significantly longer persistence was observed in the LAI than in the OAT.
Stahl 201339	Lurasidone; olanzapine; placebo	Evaluate the impact of switching from olanzapine to lurasidone on body weight	Impact of product switching	Lurasidone showed low propensity for weight gain and minimal effects on metabolic parameters or prolactin and was associated with modest rates of EPS.
Weiser 202140	Lurasidone; olanzapine; olanzapine samidorphan; clozapine; quetiapine; aripiprazole; ziprasidone; risperidone	Compare the clinical effectiveness between all oral and LAI antipsychotic medications used in treating schizophrenia in the US Department of Veterans Affairs health care system	Hazard ratio of treatment discontinuation	Those who initiated antipsychotic treatment with clozapine, second-generation LAI, and antipsychotic polypharmacy experienced longer episodes of continuous therapy and lower treatment discontinuation rates—but not psychiatric hospitalization—than those who initiated oral olanzapine.
Xiao 201641	OAT as a group	Evaluate the impact of PP1M vs OAAs on health care resource utilization, Medicaid spending, and hospital readmission among patients with SAD	Health care cost	Total health care costs associated with the use of PPM1 vs OAAs appeared comparable. Higher pharmacy costs for PPM1 were offset by lower medical costs related to fewer and shorter inpatient visits.
Yan 202042	Brexpiprazole; lurasidone; olanzapine; paliperidone; ziprasidone; aripiprazole; quetiapine; risperidone	Compare all-cause and psychiatric inpatient hospitalization and medical costs in adult patients with SCZ newly treated with brexpiprazole vs other US Food and Drug Administration–approved OAAs in a real-world setting	Health care utilization Health care cost	Patients treated with brexpiprazole had fewer psychiatric hospitalizations and lower psychiatric costs than those treated with paliperidone. Differences in the number of all-cause hospitalizations and medical expenses among treatments were not statistically significant. All-cause total costs and all-cause medical costs did not differ across groups.
Young-Xu 201643	OAT as a group	Compare health care resource utilization and costs in veterans with schizophrenia treated with PP vs OAAs	Health care cost	PP is associated with significantly lower total health care costs attributable to reduced inpatient admissions compared with OAAs. Higher mental health–intensive case management participation among PP users may have contributed to the differences observed.

BD = bipolar disorder; CGI-C = Clinical Global Impression – Change; CGI-S = Clinical Global Impression – Score; EPS = extrapyramidal symptoms; ER = extended release; HCRU = health care utilization; ICER = incremental cost-effectiveness ratio; LAI = long-acting injectable; MPR = medication possession ratio; OAA = oral antipsychotic agent; OAT = oral antipsychotic treatment; PANSS = Positive and Negative Syndrome Scale; PDC = proportion of days covered; PP = paliperidone palmitate; PP1M = once-monthly paliperidone palmitate; RIS = risperidone; PALI = paliperidone; QALY = quality-adjusted life-year; SAD = schizoaffective disorder; SCZ = schizophrenia; WTP = willingness-to-pay.

Measures to Assess Economic Outcomes. Three primary economic outcomes were reported across studies: incremental cost-effectiveness ratio,^21,30 cost/quality-adjusted life-years,²¹ and health care cost.^{4,21,25-28,30-32,34,36,38,41,43} Fourteen studies presented cost analysis,^{4,21,25-28,31,32,34-36,38,41,43} and 2 reported cost-effectiveness.^20,30 Among studies that looked into health care costs, 9 examined pharmacy costs,^{21,25,26,30,32,34-36,41} 7 evaluated overall medical costs,^{5,26-28,32,38,41} 5 assessed inpatient/hospitalization costs,^{21,30,31,34,35} 2 examined outpatient visits,^31,34 and 1 presented the cost of relapse (Table 4).²⁵

The most common measurement of health care utilization (HCRU) was the average number of inpatient days reported in 5 studies.^{21,26,31,35,41} Other HCRU measures included avoidance of hospitalization,^20,38 avoidance of relapse,^20,25 frequency of hospitalization,^21,28 risk of admission,^4,32,41 incidence rate of outpatient visit,^26,27 inpatient admission,^26,27,34 long-term care admission,^26,27 emergency department visits,³⁴ and risk of readmission (Table 4).³⁵

A total of 8 studies compared the direct health care costs associated with LAIs and OATs among patients with schizophrenia^{21,25,31,34-36,41,43} and 2 compared the cost-effectiveness of different OATs (Table 4).^20,30

Costs Associated With Oral Antipsychotics. Among the studies comparing LAIs with OATs, the results indicated that LAIs were associated with significantly higher pharmacy costs vs OATs.^{21,25,26,31,32,34-36,41} However, the expenses related to inpatient and outpatient visits/stays for LAIs were reported to be considerably lower vs OATs (Table 4).

In studies that examined cost-effectiveness, brexpiprazole was more cost-effective than cariprazine and lurasidone,²⁰ and lurasidone had a higher chance of being cost-effective than olanzapine and risperidone.³⁰

HCRU was used to evaluate the effectiveness of OATs in the studies. OATs had fewer schizophrenia-related outpatient visits and fewer intensive care management visits vs LAIs.^26,27 A study by Pilon et al. revealed that OATs had fewer mental health institute admissions and lower cost of 1-day health institute visits vs LAIs.³⁶ Brexpiprazole showed better effectiveness compared with cariprazine, lurasidone, paliperidone, and quetiapine in terms of avoided relapse and hospitalization (Table 4).^20,36,38

Measures of Adherence and Adherence Patterns of Oral Antipsychotics. As shown in Table 4, 7 studies addressed the medication-taking behavior of OATs.^{26,27,31,32,35,36,42} The most common measurements of adherence to the OAT used were proportion of days covered (PDC) across 4 studies^31,32,35,36; medication possession ratio (MPR) in 3 studies^22,31,37; and persistence in 6 studies.^{26,27,29,35,36,42} The follow-up period for both PDC and MPR in the studies was either 6 months or 12 months. Overall, patients with schizophrenia had poor adherence to OATs (Table 4). In studies that used PDC as a measure of adherence, the average adherence rate ranged from 35% to 60%, significantly lower than what is considered adherent (ie, PDC ≥80%).^32,35,36 In a retrospective cohort study by Berger et al., adherence to aripiprazole, quetiapine, or ziprasidone was evaluated after hospital discharge. The overall adherence was low, with mean MPR at 6 months reported as 55.1%.²² Similarly, in other studies that used MPR and persistence as measures, adherence was 41% to 61% (for MPR) and 20% to 61% (for persistence).^{22,26,27,29,32,35,36,42}

As reported in Table 4, in studies comparing OATs, no “within group” differences were noted in adherence patterns for OATs.³⁵ However, in a real-world study by Rajagopalan et al., the adherence patterns of lurasidone and other oral atypical antipsychotics were evaluated using administrative claims data from Medicaid and commercially insured patients.³⁷ Lurasidone was compared with olanzapine, aripiprazole, quetiapine, risperidone, and ziprasidone. A higher MPR was found for patients treated with lurasidone in the Medicaid population vs those treated with other OATs (all P < 0.05). Discontinuation rates were also lower for lurasidone than all other antipsychotics (all P < 0.05). Patients treated with lurasidone had higher MPRs in the commercial population (0.61) relative to patients receiving quetiapine (0.44) and ziprasidone (0.43; both P < 0.05).³⁷

Lafeuille et al. reported that once-monthly paliperidone palmitate was associated with higher persistence over 12 months vs OATs as a group (40% vs 33%; P = 0.006).²⁶ In a retrospective cohort study, Patel et al. reported that once-monthly paliperidone palmitate was associated with higher PDC than OATs (41.2% vs 34.7%; P = 0.008).³² Similar findings of LAI treatment associated with greater adherence have been made by other studies included in this review.^{27,32,35,36,42}

Impact of Drug Switching. As shown in Table 4, Newcomer et al. evaluated the impact of drug switching to lurasidone or quetiapine from other atypical antipsychotics,²⁸ and Patel et al. examined the effect of switching from oral risperidone/paliperidone to once-monthly paliperidone palmitate.³¹ Citrome et al. studied the effectiveness of switching to lurasidone from other antipsychotics,²³ and Cutler et al. assessed the outcome of switching to iloperidone from placebo and ziprasidone.²⁴ Newcomer et al. noted that switching to quetiapine was associated with higher all-cause (P = 0.03) and mental health–related (P = 0.02) hospitalizations relative to lurasidone.²⁸ Patel et al. reported that switching from oral risperidone/paliperidone to once-monthly paliperidone palmitate reduced claims with a diagnosis of mental health–related comorbidities.³¹ After switching, mean PDC was higher, and a reduction in all-cause inpatient stays (odds ratio = 0.39; P < 0.001) and emergency department visits(odds ratio = 0.51; P < 0.001) was observed. Results showed that all-cause total health care costs were similar before and after treatment switching (Table 4).³¹ Detailed findings from reviewed studies of OATs are reported in Supplementary Table 2 ^{(245.4KB, pdf)} .

Discussion

Our findings from the systematic review summarized the evidence on the real-world effectiveness, adherence, and humanistic and/or economic outcomes of OATs for treating schizophrenia in the past 12 years in a real-world setting. To our knowledge, our systematic review provides the most comprehensive and current analysis of available evidence for different OATs in managing schizophrenia in the United States. This evidence highlights the gaps in research of established and novel OATs, which is essential for informing treatment decisions and governing cost-effective treatment choices. Overall, the findings suggested that OATs are associated with high treatment nonadherence and discontinuation. Among the studies that examined OATs as a group, the adherence rate ranged between 20% and 53%.^27,32,36 Among the studies that looked at individual drugs, the lowest adherence rate was observed with oral risperidone/paliperidone (11.0%),³¹ and the highest was lurasidone (61%).³⁷ Compared with the Medicaid population, the commercial population showed a higher adherence rate and lower discontinuation rate among lurasidone, olanzapine, aripiprazole, quetiapine, and risperidone, but the opposite was observed among quetiapine.³⁷ Evaluations of costs and HCRU showed that the total costs of OAT and LAI were not significantly different. LAIs helped reduce total medical costs monthly (−$168 to −$383), annually (−$12,945), and per patient per year (−$6,273).^{21,26,32,36,41} However, this was offset by the increase in pharmacy cost: monthly ($270 to $279), annually ($5,869), per patient per year ($4,770).^{21,26,32,36,41} The differences in total health care costs were not found to be significant in all measures. Studies evaluating the impact of OATs on indirect cost savings, such as presenteeism, absenteeism, and work productivity, were limited, offering an opportunity for future research. Furthermore, the lack of behavioral and humanistic evidence across OATs signifies an unmet need for evidence generation to assist clinicians and payer decision-making.

Many patients with schizophrenia continue to experience acute psychotic episodes and relapses, and the real-world effectiveness of treatment regimens has been reported to be suboptimal despite extensive research.^44,45 In a 3-year clinical study by Emsley et al., recurrence rates were 79% at 12 months, 94% at 24 months, and 97% at 36 months.⁴⁵ In another study, up to 82% of the patients reported a relapse within 9 months of antipsychotic treatment.⁴⁴ Additionally, the long-term benefits of treatment on employment and social relationships also remain unclear.⁴⁶ Treatment nonadherence among patients with schizophrenia is high, ranging between 31.7%⁴⁷ and approximately 70%.⁴⁸ Poor adherence to treatment has been reported as a major detriment to good prognosis among patients with schizophrenia and is also related to increased costs.^49,50 In this regard, adverse events and patients’ perspectives or experiences toward treatment have been identified as major drivers of adherence.¹⁵ Lambert et al. reported that nonadherence was significantly higher among patients with past or present antipsychotic side effects, and Lacro et al. illustrated that poor perception of or negative attitude toward treatment also led to high nonadherence, highlighting the need for education and treatment options for adults with schizophrenia.^48,49

Poor treatment adherence was a driver of inadequate treatment outcomes in patients with schizophrenia. Our findings highlight that OATs’ low adherence and high discontinuation may be related to other humanistic outcomes. However, a knowledge gap exists in our understanding of the impact of OATs on humanistic and behavioral outcomes.

In addition to clinical trial findings, real-world data demonstrating additional product value elements can effectively facilitate managed care formulary decisions and patient care treatment decisions. It has been recommended that the overall value estimation of a novel treatment should include an assessment of the drug’s ability to mitigate economic and humanistic burdens associated with a disorder. A 2018 International Society of Pharmacoeconomics and Outcomes Research special task force recognized that conventional cost-effective analysis presents a narrow view of drug value, often leading to suboptimal resource utilizations and misleading indicators to innovators.⁵¹ In the absence of comprehensive data on humanistic and behavioral outcomes, including the burden on caregivers and employers, a well-informed value-based decision cannot be made. Because much of the burden of caring for patients with schizophrenia falls on caregivers, capturing caregiver-reported outcomes, both in clinical trials and in the real world, is necessary to understand the societal burden of schizophrenia and the value of new drugs. Our systematic review showcased a significant gap in behavioral and humanistic evidence parameters associated with schizophrenia.

LIMITATIONS

Our systematic review has several limitations. Given the scope of this study, there is a potential for selection bias. Disease severity and comorbidities are factors whose impact on outcomes are not always reported in studies. Some of the medications used for schizophrenia are the basis of LAIs, and as LAIs were excluded from this review, we need to exercise caution when interpreting the results. Another concern of selection bias is the lack of details in studies that focused on switching of treatments. It was not clear whether patients switching OATs was typical of patients with schizophrenia. This systematic review also did not evaluate the overall direct and indirect cost of schizophrenia; therefore, the impact of the cost reductions associated with treatment may have been underrepresented. Furthermore, our systematic review did not capture data related to mortality. Mortality and premature mortality are significant contributors to indirect treatment costs. Thus, the impact of treatment on the burden of mortality is likely to exemplify the longer-term value of OATs. Finally, some studies reported patients who may have been uncontrolled on medications. We understand that these may be linked to genetic influences and not necessarily the effectiveness of medications.

Conclusions

Overall, the findings from our systematic review revealed that treatment nonadherence and discontinuation of treatment for OATs are high. Moreover, limited data exist on the effect of schizophrenia treatment with OATs on indirect costs related to presenteeism, absenteeism, and work productivity for both patients and caregivers, highlighting areas for future research. Our study highlights considerable knowledge gaps in the evidence on the humanistic and behavioral outcomes of using OATs in patients with schizophrenia. These findings should be a call to drug developers to incorporate greater patient experience data into clinical trials and real-world outcomes studies, to ensure the endpoints that are measured are meaningful to patients and caregivers. In doing so, the greater evidence will assist clinicians in promoting patient-centered care and help payers understand the total value of new antipsychotic drugs.

Funding Statement

Dr Kamal has received research funding from Cerevel Therapeutics and EMD Serono, served as a consultant for Pfizer/Cytel, and received honorarium from Pharmacy Times Continuing Education. Dr Zacker is employed by Cerevel Therapeutics. Ms Adhikari and Ms Jeun have received funding support from Cerevel Therapeutics. Mr Ashraf and Mr Nolfi have served as consultants and received funding from West Virginia University School of Pharmacy. Cerevel Therapeutics funded the study and was involved with the study design, data collection, analysis, interpretation, and writing.