Research JAMA Psychiatry | Original Investigation Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Who Are Ineligible for Randomized Clinical Trials Heidi Taipale, PhD; Johannes Schneider-Thoma, MD; Justo Pinzón-Espinosa, MD, MSc; Joaquim Radua, MD, PhD; Orestis Efthimiou, PhD; Christiaan H. Vinkers, MD, PhD; Ellenor Mittendorfer-Rutz, PhD; Narcís Cardoner, MD, PhD; Luis Pintor, MD, PhD; Antti Tanskanen, PhD; Anneka Tomlinson, MD, PhD; Paolo Fusar-Poli, MD, PhD; Andrea Cipriani, MD, PhD; Eduard Vieta, MD, PhD; Stefan Leucht, MD; Jari Tiihonen, MD, PhD; Jurjen J. Luykx, MD, PhD Supplemental content IMPORTANCE Most evidence about efficacy and safety of antipsychotics in schizophrenia spectrum disorders relies on randomized clinical trials (RCTs). However, owing to their strict eligibility criteria, RCTs represent only a part of the real-world population (ie, unselected patients seen in everyday clinical practice), which may result in an efficacy-effectiveness gap. OBJECTIVE To quantify the proportion of real-world individuals with schizophrenia spectrum disorders who would be ineligible for participation in RCTs, and to explore whether clinical outcomes differ between eligible and ineligible individuals. DESIGN, SETTING, AND PARTICIPANTS This study applied eligibility criteria typically used in RCTs for relapse prevention in schizophrenia spectrum disorders to real-world populations. Individuals with diagnoses of schizophrenia spectrum disorders recorded in national patient registries in Finland and Sweden were identified. Individuals who had used antipsychotics continuously for 12 weeks in outpatient care were selected. Individuals were followed up for up to 1 year while they were receiving maintenance treatment with any second-generation antipsychotic (excluding clozapine). Follow-up was censored at treatment discontinuation, initiation of add-on antipsychotics, death, and end of database linkage. MAIN OUTCOMES AND MEASURES Proportions of RCT-ineligible individuals with schizophrenia spectrum disorders owing to any and specific RCT exclusion criteria. The risk of hospitalization due to psychosis within 1-year follow-up in ineligible vs eligible persons were compared using hazard ratios (HR) and corresponding 95% CIs. RESULTS The mean (SD) age in the Finnish cohort (n = 17 801) was 47.5 (13.8) years and 8972 (50.4%) were women; the mean (SD) age in the Swedish cohort (n = 7458) was 44.8 (12.5) years and 3344 (44.8%) were women. A total of 20 060 individuals (79%) with schizophrenia spectrum disorders would be ineligible for RCTs (Finnish cohort: 14 221 of 17 801 [79.9%]; Swedish cohort: 5839 of 7458 [78.3%]). Most frequent reasons for ineligibility were serious somatic comorbidities and concomitant antidepressant/mood stabilizer use. Risks of hospitalization due to psychosis was higher among ineligible than eligible individuals (Finnish cohort: 18.4% vs 17.2%; HR, 1.14 [95% CI, 1.04-1.24]; Swedish cohort: 20.1% vs 14.8%; HR, 1.47 [95% CI, 1.28-1.92]). The largest risks of hospitalization due to psychosis were observed in individuals ineligible owing to treatment resistance, tardive dyskinesia, and history of suicide attempts. Finally, with more ineligibility criteria met, larger risks of hospitalization due to psychosis were observed in both countries. CONCLUSIONS AND RELEVANCE RCTs may represent only about a fifth of real-world individuals with schizophrenia spectrum disorders. Underrepresented (ineligible) patients with Author Affiliations: Author schizophrenia spectrum disorders have moderately higher risks of admission due to psychosis affiliations are listed at the end of this while receiving maintenance treatment than RCT-eligible patients. These findings set the article. stage for future studies targeting real-world populations currently not represented by RCTs. Corresponding Author: Jurjen J. Luykx, MD, PhD, Department of Psychiatry and Neuropsychology, School for Mental Health and Neuroscience, Maastricht University Medical Centre, Maastricht, the Netherlands, P. Debyelaan 25, JAMA Psychiatry. 2022;79(3):210-218. doi:10.1001/jamapsychiatry.2021.3990 6229 HX Maastricht, the Netherlands Published online January 26, 2022. (j.luykx@maastrichtuniversity.nl). 210 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 09/15/2023 Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Original Investigation Research M ost evidence about efficacy and safety of medical treatments is based on randomized clinical trials Key Points (RCTs), which are highly standardized, systematic Question What percentage of patients with schizophrenia in the studies. RCT outcomes (efficacy) may differ from the utility real world are represented in randomized clinical trials (RCTs) and of interventions in routine clinical practice (effectiveness), in do their outcomes differ from those not represented in RCTs? what has been termed the efficacy-effectiveness gap. Efficacy- Findings In this study of 25 259 real-world individuals with effectiveness gaps have been identified in several health care diagnoses of schizophrenia spectrum disorders recorded in areas, including pneumology,1,2 oncology,3 infectology,4 and national patient registries in Finland and Sweden, about a fifth internal medicine5; nonpharmacological interventions in were represented in RCTs and their outcomes were better than psychology6; and antidepressants.7 of those individuals with schizophrenia not meeting RCT A possible efficacy-effectiveness gap in effectiveness and inclusion criteria. safety of antipsychotics in individuals with schizophrenia, Meaning Future research should consider the heterogeneity which, to our knowledge, has not been investigated so far, may of individuals with schizophrenia and the patient groups typically stem from the strict exclusion criteria applied in typical RCTs ineligible for participation; RCTs may become more inclusive by aiming at marketing approval. Therefore, a broad and diverse representing a broader spectrum of individuals with schizophrenia set of individuals is excluded from these trials, such as those and by targeting specific currently underrepresented groups. experiencing suicidal ideations, substance use disorders, or somatic and psychiatric comorbidities. Such excluded people zoaffective disorder (referred to from here on as schizophre- may have different courses of illness and possibly also differ- nia). Pseudonymized data were originally extracted by the reg- ent treatment outcomes. ister maintainers via personal identity codes, which enable data Here, we aimed to quantify the real-world population linkage between registries of both countries. Personal iden- (ie, unselected patients seen in everyday clinical practice) tity codes were replaced with research identity codes before not directly represented in RCTs, ie, those who are ineligible data were shared with the researchers. We chose 2 registries owing to any RCT exclusion criteria, as well as the real-world to assess similarity in findings across countries and reduce the populations ineligible owing to specific exclusion criteria. likelihood of chance findings. Moreover, we assessed whether there are differences in key In both registries, we first focused on individuals hospi- outcomes between individuals who are potentially eligible talized at least once owing to schizophrenia and who used sec- and those who are not (overall and for specific RCT exclusion ond-generation antipsychotics at the start of follow-up be- criteria). To answer these research questions, we analyzed data cause those are the typical interventions in modern RCTs10 and from real-world populations in 2 nationwide registries. also the most used antipsychotics in Finland and Sweden nowadays.11 We did not consider individuals using clozapine or first-generation antipsychotics because the former is not a first-line treatment but reserved for treatment resistance (here Methods defined as clozapine or electroconvulsive therapy treatment, In this analysis, we simulated the application of typical inclusion reported ever before follow-up), and the latter are only rarely and exclusion criteria of RCTs conducted in individuals with used in real-world clinical practice in Finland and Sweden.11 schizophrenia (eAppendix 1 in the Supplement) to the national Continuous medication use was derived using the PRE2DUP patient registries of Finland and Sweden. The protocol for our method from dispensed prescriptions.12 analysis was registered on the Open-Science Framework prior The duration of follow-up was 12 months as this is a typi- to analysis on September 15, 2020,8 and we complied with the cal duration of relapse-prevention RCTs,10 with time zero Reporting of Studies Conducted Using Observational Routinely- defined as when the inclusion criteria were fulfilled, ie, after Collected Data (RECORD) reporting guideline (eAppendix 2 in the 12 weeks of continuous antipsychotic use in monotherapy as Supplement).9 The Regional Ethics Board of Stockholm approved an outpatient. We chose this 12-week criterion to ensure clini- this research project (decision 2007/762–31). Permissions were cal stability of schizophrenia in maintenance treatment with also granted by pertinent institutional authorities at the Finnish antipsychotics, which is a starting point for relapse preven- National Institute for Health and Welfare (permission THL/847/ tion trials. We censored follow-up at discontinuation of anti- 5.05.00/2015), the Social Insurance Institution of Finland (65/ psychotics, hospitalization (other types than analyzed as out- 522/2015), and Statistics Finland (TK53-1042-15). The study was come event), death, after the defined follow-up time, and end registry based, and no contact was made with the participants of data linkage. For details about patient involvement and data of the study; therefore, according to legislation in both countries, sharing options, see the eMethods in the Supplement. obtaining informed consent from participants was not required. Outcomes and Statistical Analyses Real-World Databases Used and Follow-up By applying to these databases the standard RCT inclusion and We had access to the data extracted from the national patient exclusion criteria mentioned in the eMethods (eAppendix 1A) registries in Finland (January 2005-December 2017) and Swe- in the Supplement, we defined populations consisting of: den (January 2006-December 2016) (eAppendix 1A in the 1. Individuals potentially eligible for standard RCTs about Supplement includes details about the cohorts), which here relapse prevention with antipsychotics (ie, meeting all in- represent real-world individuals with schizophrenia and schi- clusion criteria but having none of the exclusion criteria); jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2022 Volume 79, Number 3 211 Downloaded From: https://jamanetwork.com/ on 09/15/2023 Research Original Investigation Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Table 1. Characteristics of Individuals Included (Individuals Without Any Exclusion Criteria) vs Excluded (After Application Of All Exclusion Criteria) for Randomized Clinical Trials No. (%) Finnish cohort (n = 17 801) Swedish cohort (n = 7458) Eligible Ineligible Eligible Ineligible Characteristic (n = 3580) (n = 14 221) (n = 1619) (n = 5839) Male 1837 (51.3) 6992 (49.2) 962 (59.4) 3152 (54.0) Female 1743 (48.7) 7229 (50.8) 657 (40.6) 2687 (46.0) Age, y <18 0 26 (0.2) 0 9 (0.2) 18-30 676 (18.9) 2470 (17.4) 205 (12.7) 770 (13.2) 31-45 1016 (28.4) 3624 (25.5) 627 (38.7) 1892 (32.4) 46-65 1888 (52.7) 6332 (44.5) 787 (48.6) 2774 (47.5) >65 0 1769 (12.4) 0 394 (6.8) Age, mean (SD), y 45.6 (12.3) 47.9 (14.1) 44.9 (11.2) 46.5 (12.8) Schizoaffective disorder 385 (10.8) 3459 (24.3) 232 (14.3) 2045 (35.0) Disability pension NA NA 1320 (81.5) 4985 (85.4) Abbreviation: NA, not applicable. 2. Individuals ineligible for such an RCT for any reason primary outcome, we additionally compared individuals with (ie, meeting all inclusion criteria but having ≥1 exclusion a given exclusion criterion with eligible individuals. We also criteria); compared the primary outcome between individuals who met 3. Individuals ineligible for such an RCT owing to each spe- 1, 2, or 3 or more exclusion criteria with those of eligible indi- cific exclusion criterion (ie, forming subpopulations of viduals. We conducted all analyses using SAS statistical soft- ineligible individuals owing to age, substance use, risk ware version 9.4 (SAS Institute) between November 2020 and of suicide, treatment resistance, serious somatic disease, May 2021. We calculated 95% CIs to provide estimates of the mood stabilizer or antidepressant use, intellectual disabil- accuracy of our population parameters. P values are only re- ity, tardive dyskinesia, or pregnancy/breastfeeding). ported in Tables as additional measures of the magnitude and We summarized relevant baseline characteristics and precision of the differences, but we do not characterize re- report the distribution of the prescribed antipsychotics (for sults as statistically significant or according to some arbitrary the most commonly used drugs)11 of the eligible and ineli- P value threshold. gible populations. Based on previous knowledge,11 we catego- rized most commonly prescribed antipsychotics in these cohorts as olanzapine, quetiapine, risperidone, and aripipra- zole, while the rest were grouped as either any long-acting Results injectable (LAI) antipsychotic or other oral antipsychotics. Proportions, Descriptive Statistics, and Antipsychotic Use The primary outcome was hospitalization due to psycho- in Eligible and Ineligible Populations sis (International Statistical Classification of Diseases and The mean (SD) age in the Finnish cohort (n = 17 801) was 47.5 Related Health Problems, Tenth Revision codes F20-F29). Sec- (13.8) years, and 8972 (50.4%) were women; the mean (SD) ondary outcomes were hospitalization due to any psychiatric age in the Swedish cohort (n = 7458) was 44.8 (12.5) years, and reason (International Statistical Classification of Diseases 3344 (44.8%) were women. In the Finnish cohort, 3580 indi- and Related Health Problems, Tenth Revision codes F00-F99), viduals (20.1%) were eligible for RCT participation; 14 221 all-cause hospitalization, the need for add-on antipsychot- (79.9%) met at least 1 exclusion criterion and were thus ineli- ics, and all-cause discontinuation of antipsychotic use. To verify gible. Similarly, in the Swedish cohort, 1619 individuals (21.7%) the robustness of our results for different time points, in were eligible for RCTs, and 5839 (78.3%) were ineligible. addition to the main analyses at 12 months, we conducted There were no major differences in the distribution of age analyses for the primary outcome at 6 months and 9 months and sex between eligible and ineligible individuals (Table 1). of follow-up. Additionally, as sensitivity analyses, we applied Individuals who were ineligible for RCTs were more likely the same primary outcome analyses to separate cohorts of to use oral quetiapine (Finland: 3735 [26.3%] vs 612 [17.1%]; (1) clozapine users and (2) individuals only treated in outpa- Sweden: 809 [13.9%] vs 110 [6.8%]; eFigure 1 in the Supple- tient care (eAppendix 1A in the Supplement). ment). LAI antipsychotics were prescribed less frequently to To compare potential differences in the risk of these out- ineligible than to eligible individuals (Finland: 1767 [12.4%] vs comes between eligible and ineligible individuals, we calcu- 753 [21.0%]; Sweden: 1075 [18.4%] vs 390 [24.1%]; eFigure 1 lated hazard ratios (HRs) and their 95% CIs using a Cox regres- in the Supplement). In the Swedish data set where informa- sion model with eligible individuals as reference. Proportional tion on disability pension was available, ineligible individu- hazards assumption was tested and complied with by plot- als were somewhat more likely (4985 [85.4%]) to receive dis- ting Kaplan-Meier curves and via Schoenfeld residuals. To shed ability pension than eligible ones (1320 [81.5%]), indicating light on the associations of specific exclusion criteria with the more severe decline in occupational function. 212 JAMA Psychiatry March 2022 Volume 79, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 09/15/2023 Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Original Investigation Research Figure. Distribution of the Number of Exclusion Criteria Met and Prevalence of Specific Conditions Among Persons Ineligible for Randomized Clinical Trials A Percentage of individuals meeting exclusion B Prevalence of conditions among excluded individuals by cohort criteria by cohort 40 60 35 Finland 50 Sweden 30 40 Individuals, % Prevalence, % 25 20 30 15 20 10 10 5 0 0 0 1 2 ≥3 Age Age Substance Suicide Treatment Somatic Somatic MS/AD Intellectual Tardive Pregnancy Exclusion criteria met, No. <18 y >65 y use attempt resistance disease, disease, use disability dyskinesia broad narrow Exclusion criteria A, Percentage of individuals in the cohorts who fulfilled none, 1, 2, or ⱖ3 exclusion among individuals ineligible for randomized clinical trials in the Finnish and criteria in the Finnish and Swedish cohorts. B, Prevalence (%) of specific conditions Swedish cohorts. AD indicates antidepressant; MS, mood stabilizer. Ineligibility Reasons and Subpopulations in the Finnish cohort. Risperidone was equally commonly pre- In the Finnish and Swedish cohorts, 5875 (33.0%) and 2514 scribed among pregnant or breastfeeding individuals as olanza- (33.7%), respectively, fulfilled only 1 exclusion criterion, while pine in the Swedish cohort. In Finland, LAI antipsychotics were 3271 (18.4%) and 1338 (17.9%), respectively, met 3 or more cri- used in about 10% of all individuals in most subpopulations teria (Figure, A). The most frequent reasons for ineligibility were (except substance use), while in Sweden the use of LAI anti- serious somatic comorbidities (broad definition: 7202 [51%] and psychotics was more frequent: almost 20% in all groups (ex- 2866 [49%]; narrow: 5287 [36%] and 1747 [30%] in Finland cept mood stabilizers/antidepressants users). and Sweden, respectively) and concomitant use of mood sta- bilizers or antidepressants (7983 [56%] and 3281 [56%]), fol- Primary and Secondary Outcomes in Eligible lowed by a history of substance use (3808 [27%] and 1828 [31%]) and Ineligible Populations and suicide risk (1690 [12%] and 1032 [18%]) (Figure, B). During the 12 months of follow-up, individuals who would be Subpopulations ineligible owing to specific exclusion crite- ineligible for RCTs were more likely to be hospitalized owing ria had variation in age and sex distributions (eTable 1A and B to psychosis, compared with eligible individuals (Finnish co- in the Supplement). The proportion of men was highest in those hort: 2609 [18.4%] vs 615 [17.2%]; HR, 1.14 [95% CI, 1.04- excluded owing to substance use (2510 men [65.9%] in the Finn- 1.24]; Swedish cohort: 1174 [20.1%] vs 240 [14.8%]; HR, 1.47 ish cohort and 1230 [67.3%] in the Swedish cohort), whereas those [95% CI, 1.28-1.92]). Similar risk estimates were observed for excluded owing to age (almost entirely owing to age >65 years) 6- and 9-month follow-up times (Table 2). were mainly women (687 men [38.3%] in the subgroup of the Compared with eligible individuals, individuals who were Finnish cohort excluded owing to age and 143 [36.3%] in the ineligible for RCTs had increased risks of any psychiatric hos- Swedish cohort). There was significant overlap between specific pitalization (HR, 1.34 [95% CI, 1.23-1.45]) in the Finnish co- exclusion criteria with each other. Besides obvious overlap hort; HR, 1.74 [95% CI, 1.52-1.99] in the Swedish cohort) and between broad and narrow definitions of serious somatic comor- for all-cause hospitalization (HR, 1.55 [95% CI, 1.43-1.68] in the bidities, those with a history of suicide attempt often also had Finnish cohort; HR, 1.77 [95% CI, 1.55-2.03] in the Swedish co- substance use (Finnish cohort: 882 [52%]; Swedish cohort: 533 hort; Table 3). In the Swedish cohort, ineligible persons had [52%]) and mood stabilizers/antidepressants use (1009 [58%] a higher risk for needing an additional antipsychotic than in the Finnish and 603 [60%] in the Swedish cohorts) and those eligible persons (HR, 1.31 [95% CI, 1.15-1.48]), which was not with tardive dyskinesia had serious somatic conditions (24 [60%] observed in the Finnish cohort (HR, 1.06 [95% CI, 0.96-1.17]). and 47 [63%] in the Finnish and Swedish cohorts, respectively) The risk of all-cause antipsychotic discontinuation did not and mood stabilizers/antidepressants use (24 [60%] and 36 [48%] differ between ineligible and eligible individuals (Table 3). in the Finnish and Swedish cohorts, respectively). Olanzapine was the most frequently prescribed antipsy- Primary Outcome in Subpopulations Ineligible chotic across specific subpopulations (eFigures 2 and 3 in the for Specific Reasons Supplement). Quetiapine replaced olanzapine as the most fre- The largest risks of hospitalization due to psychosis were ob- quently prescribed antipsychotic among those with a history served in individuals ineligible owing to treatment resis- of suicide attempts and among those with tardive dyskinesia tance, tardive dyskinesia, and history of suicide attempts jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2022 Volume 79, Number 3 213 Downloaded From: https://jamanetwork.com/ on 09/15/2023 Research Original Investigation Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Table 2. Risk of Hospitalization Due to Psychosis in Individuals Ineligible (After Application of ≥1 Exclusion Criteria) vs Eligible (Persons Without Any Exclusion Criterion) Finnish cohort Swedish cohort Time to Time to event/ event/ No. of No. (%) censoring, P No. of No. (%) censoring, P Eligibility individuals with event mean (SD), d HR (95% CI)a value individuals with event mean (SD), d HR (95% CI)a value Main outcome analyses: hospitalization due to psychosis, 12 mo Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.14 (1.04-1.24) .005 1.47 (1.28-1.92) <.001 Ineligible 14 221 2609 (18.4) 257 (137) 5839 1174 (20.1) 248 (137) Alternative time period 1: hospitalization due to psychosis, 6 mo Eligible 3580 443 (12.4) 154 (56) 1619 166 (10.3) 156 (53) 1.15 (1.03-1.27) .01 1.49 (1.26-1.75) <.001 Ineligible 14 221 1939 (13.6) 147 (60) 5839 835 (14.3) 145 (60) Alternative time period 2: hospitalization due to psychosis, 9 mo Eligible 3580 537 (15.0) 218 (93) 1619 212 (13.1) 218 (89) 1.15 (1.05-1.26) .004 1.45 (1.25-1.69) <.001 Ineligible 14 221 2327 (16.4) 205 (98) 5839 1033 (17.7) 200 (99) Abbreviation: HR, hazard ratio. a An HR >1 means higher risk in the ineligible group. Primary analyses were with 12-month follow-up and sensitivity analyses with 6 and 9 months. Table 3. Risk of Secondary Outcomes in Individuals Ineligible (After Application of ≥1 Exclusion Criteria) vs Eligible (Individuals Without Any Exclusion Criteria) at 12-Month Follow-up Finnish cohort Swedish cohort Time to Time to event/ event/ No. (%) censoring, No. (%) censoring, Eligibility with event mean (SD) HR (95% CI)a P value with event mean (SD) HR (95% CI)a P value Hospitalization due to any psychiatric reason Eligible 643 (18.0) 278 (131) 251 (15.5) 273 (127) 1.34 (1.23-1.45) <.001 1.74 (1.52-1.99) <.001 Ineligible 3202 (22.5) 256 (138) 1440 (24.7) 246 (138) All-cause hospitalization Eligible 694 (19.4) 280 (130) 250 (15.4) 273 (127) 1.55 (1.43-1.68) <.001 1.77 (1.55-2.03) <.001 Ineligible 4011 (28.2) 259 (136) 1478 (25.31) 246 (138) Need for additional antipsychotic Eligible 454 (12.7) 254 (138) 286 (17.7) 238 (137) 1.06 (0.96-1.17) .28 1.31 (1.15-1.48) <.001 Ineligible 1783 (12.5) 233 (141) 1221 (20.9) 211 (141) All-cause discontinuation of antipsychotic use Eligible 570 (15.9) 280 (130) 408 (25.2) 275 (126) 1.03 (0.94-1.13) .59 1.01 (0.90-1.12) .91 Ineligible 2191 (15.4) 259 (137) 1370 (23.5) 251 (136) Abbreviation: HR, hazard ratio. a An HR >1 means higher risk in the ineligible group. (Table 4). Finally, with more ineligibility criteria met, larger risks schizophrenia in Finnish and Swedish national registries. We of hospitalization due to psychosis were observed in both coun- found that almost 80% of individuals with schizophrenia tries (eResults and eTables 2 and 3 in the Supplement). would be ineligible to participate in typical RCTs and are there- Sensitivity analysis in the cohort of clozapine users un- fore not represented in them. The most frequent reasons for covered similar proportions of individuals being ineligible ineligibility observed in the 2 cohorts were serious somatic co- (5806 [81.6%] in the Finnish cohort and 1346 [80.2%] in the morbidities and concomitant use of mood stabilizers or anti- Swedish cohort) for RCTs as in the main analyses (eTable 4 in depressants, followed by history of substance use and risk of the Supplement). Results from the second sensitivity analy- suicide. Furthermore, we found that RCT-ineligible real- sis were also similar to the primary analysis: of 4727 individu- world individuals had, on average, a moderately higher risk als treated in outpatient care only, 3508 (74.2%) were ineli- for rehospitalization due to psychosis while receiving main- gible for RCT participation (eTable 5 in the Supplement). tenance treatment with antipsychotics. This increased risk was observed in several subpopulations, ie, individuals ineligible for specific reasons such as substance use, risk of suicide, treat- ment resistance, or tardive dyskinesia. Moreover, ineligible in- Discussion dividuals appeared to have a higher burden of psychiatric and In this study, we applied typical inclusion and exclusion cri- somatic comorbidities, as indicated by increased psychiatric teria of RCTs to the real-world populations of individuals with and any-reason hospitalization rates. 214 JAMA Psychiatry March 2022 Volume 79, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 09/15/2023 Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Original Investigation Research Table 4. Risk of Hospitalization Due to Psychosis Within 12 Months of Follow-up for the Population of Individuals Remaining After Applying Each Specific Exclusion Criterion Separately Compared With Individuals Who Did Not Meet Any Exclusion Criteria (Eligible Group) Finnish cohort Swedish cohort Time to Time to event/ event/ censoring, censoring, No. of Relapsed, mean (SD), P No. of Relapsed, mean (SD), P Eligibility individuals No. (%) d HR (95% CI)a value individuals No. (%) d HR (95% CI)a value Age <18 and >65 y Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 0.71 (0.61-0.83) <.001 1.04 (0.78-1.38) .80 Ineligible 1795 212 (11.8) 267 (133) 403 60 (14.9) 264 (131) Substance use Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.43 (1.29-1.59) <.001 1.88 (1.61-2.21) <.001 Ineligible 3808 801 (21.0) 228 (143) 1828 430 (23.5) 224 (140) Suicide attempt Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.61 (1.42-1.83) <.001 2.13 (1.79-2.54) <.001 Ineligible 1690 395 (23.4) 225 (144) 1032 270 (26.2) 219 (141) Treatment resistance Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.71 (1.52-1.93) <.001 2.31 (1.87-2.85) <.001 Ineligible 1805 476 (26.4) 242 (143) 450 134 (29.8) 233 (141) Serious somatic disease, broader definition Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.09 (0.99-1.20) .09 1.53 (1.31-1.77) <.001 Ineligible 7202 1247 (17.3) 252 (138) 2866 586 (20.5) 243 (139) Serious somatic disease, narrower definition Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.10 (0.99-1.22) .08 1.58 (1.34-1.86) <.001 Ineligible 5087 877 (17.2) 247 (139) 1747 361 (20.7) 236 (138) Mood stabilizer/antidepressant concomitant use Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.10 (1.01-1.22) .03 1.51 (1.30-1.75) <.001 Ineligible 7983 1456 (18.2) 262 (135) 3281 682 (20.8) 251 (136) Intellectual disability Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 0.98 (0.79-1.21) .83 1.19 (0.87-1.62) .28 Ineligible 622 102 (16.4) 269 (133) 282 47 (16.7) 257 (133) Tardive dyskinesia Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 1.77 (0.95-3.31) .07 2.13 (1.36-3.32) <.001 Ineligible 40 10 (25.0) 211 (137) 75 21 (28.0) 235 (129) Pregnant or breastfeeding women Eligible 3580 615 (17.2) 278 (130) 1619 240 (14.8) 273 (127) 0.87 (0.55-1.37) .55 1.29 (0.53-3.13) .57 Ineligible 143 19 (13.3) 240 (145) 30 5 (16.7) 237 (148) Abbreviation: HR, hazard ratio. a An HR >1 means higher risk in the ineligible group. We envision the following implications of our findings. latter could exploit the potential offered by electronic health Because we showed that the majority of individuals with schizo- records for screening and recruiting trial participants and there- phrenia are not represented by typical RCTs and that clinical fore enrich their real-world representativeness. It also under- outcomes can differ between eligible and ineligible individu- lines the importance of aftermarket/postapproval studies (ie, als, targeted RCTs, subgroup analyses of RCTs with broader in- phase 4 studies) requested by regulators and conducted by clusion criteria, and observational cohorts focusing on under- pharmaceutical companies to particularly investigate the safety represented subpopulations are warranted. To date, only a few of new treatments in broader populations. Furthermore, we RCTs have been conducted in specific patient groups.13-15 Ad- found that choices of antipsychotics were somewhat differ- ditionally, because approximately 50% of ineligible individu- ent between ineligible vs eligible individuals. In previous real- als met somatic comorbidities exclusion criteria in our study, world studies using within-individual designs minimizing se- risks of adverse effects and their potential serious conse- lection bias, LAI antipsychotics were associated with lowered quences as well as the risk of clinically significant pharmaco- risk of rehospitalization whereas quetiapine often was not, com- logical interactions could be higher in real-world populations pared with no use of antipsychotics.17,18 Ineligible individuals than in RCTs. This may require clinical attention and further were less likely to use LAI antipsychotics and more likely to use research after pivotal RCTs and drug market approval.16 The quetiapine than eligible individuals. Reasons for these differ- jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2022 Volume 79, Number 3 215 Downloaded From: https://jamanetwork.com/ on 09/15/2023 Research Original Investigation Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice ences are not fully clear from our data; however, it is possible may not be directly generalizable to other types of RCTs or to that LAI antipsychotics are avoided as those are slower to countries with different health care systems or resources. Al- taper, eg, in persons with high risk of extrapyramidal symp- though the main results represent only individuals previ- toms, or active substance use, which increases the risk of ously treated in inpatient care due to schizophrenia spectrum interactive effects leading possibly to respiratory depression disorders, additional analyses in the Swedish outpatient co- or seizures. Quetiapine may be prescribed more often for sub- hort showed similar results, allaying concerns about bias re- groups presenting more affective symptoms (eg, with suicidal sulting from cohort and statistical method selections. (3) Fur- ideation) or for tardive dyskinesia.19 This also describes a fun- thermore, inclusion and exclusion criteria vary between RCTs damental difference between real-world studies (such as the (eAppendix 1B in the Supplement). Some RCTs apply more present study) and RCTs: in the real world, treatments are cho- relaxed criteria than the ones we used, eg, by allowing the sen by clinicians by their best judgment and following clinical participation of individuals with psychiatric or somatic comor- care guidelines, while in RCTs the treatment is preset by the bidities, with stable concomitant antidepressant or mood- design. Therefore, to further elucidate antipsychotic use and stabilizing medications, history of suicide attempts (without effectiveness in practice, in addition to typical RCTs, which active suicidal thoughts or behaviors), or substance use (when are important to examine whether a drug works in principle in inactive at the time or when criteria for dependence are not selection bias–free conditions (efficacy), pragmatic trials (such met). Of note, the data from real-world cohorts do not allow as STAR*D20 and CATIE21) and observational studies (such as one to apply all eligibility criteria exactly as in RCTs because the SOHO study22,23) may in future be of benefit.24 These are only diagnoses and not clinical ratings are available, and some performed on less selected populations and resemble clinical symptoms are often underreported in diagnostic data (eg, practice more closely than typical RCTs. Finally, our results suicidality and substance use). However, previous research provide estimates for risks of rehospitalization for schizophre- has shown that register-based inpatient diagnoses of schizo- nia in different patient populations, which could be used to in- phrenia are valid.25 Consequently, while our estimates refer form individuals and clinicians about the expected outcome to standard RCT with rather strict criteria, possibly other RCTs on antipsychotics within 1 year. Of note, individuals with a pre- represent more than the 20% of real-world patients. Nonethe- vious history of substance use, suicide attempt, or clozapine less, our results highlight that there is considerable heteroge- use (as a proxy for treatment resistance) had only a moder- neity in real-world individuals, which is not addressed by most ately higher risk of rehospitalization for acute psychosis. In this standard RCTs. context, it needs to be considered that these estimates only ap- ply to individuals with schizophrenia already stable taking medication for 12 weeks before the start of the 1-year obser- vation period. For some individuals in these subpopulations, Conclusions it might be difficult to reach this level of stability. Neverthe- In conclusion, based on comprehensive main and sensitivity less, the observed differences in rehospitalization rates be- analyses leveraging sizeable nationwide cohorts and in line with tween subpopulations call for more specific epidemiological hypotheses put forward before but backed with less solid studies on expected absolute risks and predictors of relapse evidence,26-31 only a minority (about one-fifth) of real-world in- and rehospitalization. dividuals with schizophrenia may be eligible for typical RCTs and their clinical outcomes were estimated to differ from ineli- Limitations gible individuals. However, because we did not investigate rela- Our analysis is somewhat limited because our selection of in- tive treatment effects (eAppendix 1A in the Supplement) and dividuals eligible for RCTs matches the population in actual because the observed differences in absolute risks for clinical RCTs only to a certain extent for different reasons. (1) Partici- outcomes were not extreme, we emphasize that there are no pation in a trial requires participants not only to meet eligibil- major indications from our research that overall RCT results on ity criteria but also to be willing to participate in a trial; the efficacy and safety of antipsychotics would not apply to ineli- latter might be an important driver of outcomes, which we can- gible individuals. Nevertheless, our results indicate that spe- not disentangle in the real-world population. This, in addition cific subgroups among the majority of real-world individuals in- to other factors such as dropouts, might explain the differ- eligible for RCTs can have a different course of illness, which ence in rehospitalization rates on antipsychotic maintenance also means they might experience differential treatment therapy between RCTs (4% at 7-12 [median, 9] months)10 and benefits. Therefore, in line with previous literature,28,30,32-36 the real world (here 14% at 9 months in eligible individuals). future studies focusing on specific subpopulations, pragmatic (2) The selection criteria used for our analysis are typical for trials to investigate treatment strategies, and well-designed a specific, relatively common type of RCTs (ie, relapse preven- observational studies are needed to investigate and improve tion of schizophrenia with antipsychotics) and specific real- the outcomes of the many individuals afflicted by schizophre- world samples (Finland and Sweden). Therefore, our results nia and currently underrepresented in research settings. ARTICLE INFORMATION Published Online: January 26, 2022. Open Access: This is an open access article Accepted for Publication: November 20, 2021. doi:10.1001/jamapsychiatry.2021.3990 distributed under the terms of the CC-BY License. © 2022 Taipale H et al. JAMA Psychiatry. 216 JAMA Psychiatry March 2022 Volume 79, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 09/15/2023 Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice Original Investigation Research Author Affiliations: Department of Forensic Schneider-Thoma, Pinzón-Espinosa, Efthimiou, by an NIHR Research Professorship (grant Psychiatry, Niuvanniemi Hospital, University of Cardoner, Tomlinson, Vieta, Luykx. RP-2017-08-ST2-006), by the NIHR Oxford and Eastern Finland, Kuopio, Finland (Taipale, Critical revision of the manuscript for important Thames Valley Applied Research Collaboration, and Tanskanen, Tiihonen); Divisions of Insurance intellectual content: Taipale, Pinzón-Espinosa, by the NIHR Oxford Health Biomedical Research Medicine, Department of Clinical Neuroscience, Radua, Efthimiou, Vinkers, Mittendorfer-Rutz, Centre (grant BRC-1215-20005). Dr Taipale was Karolinska Institutet, Stockholm, Sweden (Taipale, Cardoner, Pintor-Perez, Tanskanen, Fusar-Poli, funded by the Academy of Finland (grants 315969 Radua, Mittendorfer-Rutz, Tanskanen, Tiihonen); Cipriani, Vieta, Leucht, Tiihonen, Luykx. and 320107). Dr Radua was funded by the Spanish School of Pharmacy, University of Eastern Finland, Statistical analysis: Taipale, Radua, Efthimiou, Ministry of Science, Innovation and Universities/ Kuopio, Finland (Taipale); Department of Psychiatry Cipriani, Luykx. Economy and Competitiveness/Instituto de Salud and Psychotherapy, School of Medicine, Technical Obtained funding: Luykx. Carlos III (grants CPII19/00009 abd PI19/00394), University of Munich, Munich, Germany Administrative, technical, or material support: co-financed by ERDF Funds from the European (Schneider-Thoma, Leucht); Department of Mental Mittendorfer-Rutz, Tanskanen, Tomlinson, Leucht, Commission (A Way of Making Europe). Dr Luykx Health, Parc Tauli University Hospital, Sabadell, Luykx. was supported by a personal Rudolf Magnus Talent Barcelona, Spain (Pinzón-Espinosa, Cardoner); Supervision: Radua, Cardoner, Pintor-Perez, Fellowship from the University Medical Center Department of Medicine, University of Barcelona Fusar-Poli, Cipriani, Vieta, Leucht, Luykx. Utrecht (grant H150). School of Medicine, Barcelona, Spain Conflict of Interest Disclosures: Dr Taipale Role of the Funder/Sponsor: The funders had no (Pinzón-Espinosa); Department of Clinical reported grants and personal fee from Janssen role in the design and conduct of the study; Psychiatry, University of Panama School of Pharmaceuticals, grants from Eli Lilly and Company, collection, management, analysis, and Medicine, Panama City, Panama (Pinzón-Espinosa); and personal fees from Otsuka Pharmaceutical interpretation of the data; preparation, review, or Institut d’Investigacions Biomèdiques August outside the submitted work. Dr Pinzón-Espinosa approval of the manuscript; and decision to submit Pi i Sunyer (IDIBAPS), CIBERSAM, Barcelona, reported personal fees and nonfinancial support the manuscript for publication. Spain (Radua, Pintor, Vieta); Early Psychosis: from Lundbeck, Angelini, Neuraxpharm, and Interventions and Clinical-detection (EPIC) Lab, Disclaimer: The views expressed are those of Pfizer and nonfinancial support from Janssen the authors and not necessarily those of the Department of Psychosis Studies, Institute of Pharmaceuticals and Casen Recordati outside the Psychiatry, Psychology, and Neuroscience, King’s UK National Health Service, the Conflict of Interest submitted work. Dr Mittendorfer-Rutz reported Disclosures: National Institute for Health Research, College London, London, United Kingdom (Radua, grants from Janssen Pharmaceuticals outside the Fusar-Poli); Institute of Social and Preventive or the UK Department of Health. submitted work. Dr Cardoner has received grants Medicine, University of Bern, Bern, Switzerland and acted as a CME consultant, advisor, or Additional Contributions: We are indebted to (Efthimiou); Department of Psychiatry, University speaker for the following entities in the last experienced expert Russel Cummins (Altrecht of Oxford, Oxford, United Kingdom (Efthimiou, 5 years (all unrelated to the present work): Mental Health), who took the effort of critically Tomlinson, Cipriani); Department of Psychiatry, Angelini, Esteve, Exeltis, Janssen Pharmaceuticals, reading our manuscript and providing his take on Amsterdam Neuroscience, Amsterdam UMC, Lundbeck, Novartis, Pfizer, and Servier. the current unmet needs in psychosis studies and Vrije Universiteit Amsterdam, Amsterdam, Dr Tanskanen reported other support from the potential value of our findings (eAppendix 1 in the Netherlands (Vinkers); Institut d’Investigació I Janssen and Eli Lilly and Company outside the the Supplement). Compensation was not received. Innovació Parc Tauli (I3PT), CIBERSAM, Sabadell, submitted work. Dr Tomlinson reported personal Barcelona, Spain (Cardoner); Department of fees from INCiPiT (Italian Network for Paediatric REFERENCES Psychiatry and Forensic Medicine, Universitat Trials) and Angelini Pharma outside the submitted 1. Cramer-van der Welle CM, Peters BJM, Autònoma de Barcelona, Barcelona, Spain work. Dr Fusar-Poli reported personal fees from Schramel FMNH, Klungel OH, Groen HJM, (Cardoner); Department of Psychiatry and Lundbeck, Angelini Pharma, and Menarini outside van de Garde EMW; Santeon NSCLC Study Group; Psychology, Hospital Clinic, University of Barcelona, the submitted work. Dr Cipriani reported personal Santeon NSCLC study group. Systematic evaluation Barcelona, Spain (Pintor, Vieta); Oxford Health NHS fees from INCiPiT (Italian Network for Paediatric of the efficacy-effectiveness gap of systemic Foundation Trust, Warneford Hospital, Oxford, Trials) and CARIPLO Foundation and personal fees treatments in metastatic nonsmall cell lung cancer. United Kingdom (Tomlinson, Cipriani); Department and grants from Angelini Pharma outside the Eur Respir J. 2018;52(6):1801100. doi:10.1183/ of Brain and Behavioral Sciences, University of submitted work. Dr Vieta reported grants from 13993003.01100-2018 Pavia, Pavia, Italy (Fusar-Poli); OASIS Service, AB-Biotics, Ferrer, GlaxoSmithKline, Janssen South London and Maudsley NHS Foundation Trust, 2. Buhl R, Criée C, Kardos P, et al. Patients in clinical Pharmaceuticals, and Sanofi-Aventis; personal fees trials on COPD triple therapy compared to real London, United Kingdom (Fusar-Poli); Center for from AbbVie, Aimentia, Angelini Pharma, Celon Psychiatry Research, Stockholm City Council, world populations. Am J Respir Crit Care Med. 2019; Pharma, Sumitomo Dainippon Pharma, Gedeon 199:A1117. doi:10.1164/ajrccm-conference.2019.199.1_ Stockholm, Sweden (Tiihonen); Department of Richter, Janssen Pharmaceuticals, Organon, Psychiatry, UMC Utrecht Brain Center, University MeetingAbstracts.A1117 Otsuka Pharmaceutical, Sunovion, and Takeda; and Medical Center Utrecht, Utrecht University, received grants and served as consultant, advisor, 3. Phillips CM, Parmar A, Guo H, et al. Assessing the Utrecht, the Netherlands (Luykx); Department of or CME speaker from GH Research, Lundbeck, and efficacy-effectiveness gap for cancer therapies: Psychiatry and Neuropsychology, School for Mental Sage outside the submitted work. Dr Leucht a comparison of overall survival and toxicity Health and Neuroscience, Maastricht University reported personal fees from Alkermes, Angelini between clinical trial and population-based, Medical Centre, Maastricht, the Netherlands Pharma, Eisai, Gedeon Richter, Janssen real-world data for contemporary parenteral cancer (Luykx); Outpatient second opinion clinic, GGNet Pharmaceuticals, Johnson & Johnson, Lundbeck, therapeutics. Cancer. 2020;126(8):1717-1726. Mental Health, Warnsveld, the Netherlands (Luykx) Medichem, Merck Sharp & Dohme, Otsuka doi:10.1002/cncr.32697 . Pharmaceutical, Casen Recordati, Rovi, Sandoz, 4. Bor J, Thirumurthy H. Bridging the Author Contributions: Drs Taipale and Tanskanen Sanofi, Sunovion, Teva Pharmaceuticals, efficacy-effectiveness gap in HIV programs: had full access to all of the data in the study and Boehringer Ingelheim, and LTS Lohmann outside lessons from economics. J Acquir Immune Defic Syndr. take responsibility for the integrity of the data and the submitted work. Dr Tiihonen reported personal 2019;82(suppl 3):S183-S191. doi:10.1097/QAI. the accuracy of the data analysis. Drs Taipale, fees from Finnish Medicines Agency, European 0000000000002201 Schneider-Thoma, and Pinzón-Espinosa are Medicines Agency, Eli Lilly and Company, 5. Ankarfeldt MZ, Adalsteinsson E, Groenwold RHH, co–first authors. Janssen-Cilag, Otsuka Pharmaceutical, and Ali MS, Klungel OH. A systematic literature review Concept and design: Taipale, Schneider-Thoma, Lundbeck and grants from the Stanley Foundation on the efficacy-effectiveness gap: comparison of Pinzón-Espinosa, Radua, Cardoner, Cipriani, Vieta, and the Sigrid Jusélius Foundation outside the randomized controlled trials and observational Leucht, Tiihonen, Luykx. submitted work. No other disclosures were studies of glucose-lowering drugs. Clin Epidemiol. Acquisition, analysis, or interpretation of data: reported. 2017;9:41-51. doi:10.2147/CLEP.S121991 Taipale, Schneider-Thoma, Pinzón-Espinosa, Radua, Funding/Support: Dr Efthimiou was supported by Efthimiou, Vinkers, Mittendorfer-Rutz, 6. Halford WK, Pepping CA, Petch J. The gap the Swiss National Science Foundation (Ambizione between couple therapy research efficacy and Pintor-Perez, Tanskanen, Tomlinson, Fusar-Poli, grant 180083). Dr Cipriani is supported by the Cipriani, Leucht, Tiihonen, Luykx. practice effectiveness. J Marital Fam Ther. 2016;42 National Institute for Health Research (NIHR) (1):32-44. doi:10.1111/jmft.12120 Drafting of the manuscript: Taipale, Oxford Cognitive Health Clinical Research Facility, jamapsychiatry.com (Reprinted) JAMA Psychiatry March 2022 Volume 79, Number 3 217 Downloaded From: https://jamanetwork.com/ on 09/15/2023 Research Original Investigation Representation and Outcomes of Individuals With Schizophrenia Seen in Everyday Practice 7. Schneider C, Breilmann J, Reuter B, Becker T, and devices after approval. Lancet. 2020;395 2005;255(2):143-148. doi:10.1007/s00406-004- Kösters M. Systematic evaluation of the (10228):998-1010. doi:10.1016/S0140-6736(19) 0547-5 'efficacy-effectiveness gap' in the treatment 33177-0 27. Freudenthal R, Marston L, Stansfeld JL, of depression with venlafaxine and duloxetine. 17. Tiihonen J, Mittendorfer-Rutz E, Majak M, et al. Priebe S, Moncrieff J. How do participants in Acta Psychiatr Scand. 2021;144(2):113-124: Real-world effectiveness of antipsychotic clinical trials compare with other patients with doi:10.1111/acps.13293 treatments in a nationwide cohort of 29 823 schizophrenia? Contemp Clin Trials Commun. 2021; 8. Luykx JJ, Pinzón-Espinosa J, Schneider-Thoma J, patients with schizophrenia. JAMA Psychiatry. 2017; 22:100803. doi:10.1016/j.conctc.2021.100803 et al What is the Impact of Randomized Controlled 74(7):686-693. doi:10.1001/jamapsychiatry.2017.1322 28. Kline E, Hendel V, Friedman-Yakoobian M, et al. Trials Inclusion and Exclusion Criteria on Outcomes 18. Taipale H, Mehtälä J, Tanskanen A, Tiihonen J. A comparison of neurocognition and functioning in People using Antipsychotics in the Real World? Comparative effectiveness of antipsychotic drugs in first episode psychosis populations: do research A Systematic Analysis in National Registries. for rehospitalization in schizophrenia: a nationwide samples reflect the real world? Soc Psychiatry OSF Regist; 2020. doi:10.17605/OSF.IO/ZRBW7 study with 20-year follow-up. Schizophr Bull. 2018; Psychiatr Epidemiol. 2019;54(3):291-301. 9. Benchimol EI, Smeeth L, Guttmann A, et al; 44(6):1381-1387. doi:10.1093/schbul/sbx176 doi:10.1007/s00127-018-1631-x RECORD Working Committee. The REporting 19. Ricciardi L, Pringsheim T, Barnes TRE, et al. 29. Lally J, Watkins R, Nash S, et al. of studies Conducted using Observational Treatment recommendations for tardive dyskinesia. The Representativeness of participants with severe Routinely-collected health Data (RECORD) Can J Psychiatry. 2019;64(6):388-399. doi:10.1177/ mental illness in a psychosocial clinical trial. Front statement. PLoS Med. 2015;12(10):e1001885. 0706743719828968 Psychiatry. 2018;9:654. doi:10.3389/fpsyt.2018. doi:10.1371/journal.pmed.1001885 00654 20. Rush AJ, Trivedi MH, Wisniewski SR, et al. 10. Ceraso A, Lin JJ, Schneider-Thoma J, et al. Acute and longer-term outcomes in depressed 30. Kennedy-Martin T, Curtis S, Faries D, Maintenance treatment with antipsychotic drugs outpatients requiring one or several treatment Robinson S, Johnston J. A literature review on the for schizophrenia. Cochrane Database Syst Rev. steps: a STAR*D report. Am J Psychiatry. 2006;163 representativeness of randomized controlled trial 2020;8:CD008016. doi:10.1002/14651858. (11):1905-1917. doi:10.1176/ajp.2006.163.11.1905 samples and implications for the external validity CD008016.pub3 of trial results. Trials. 2015;16:495. doi:10.1186/ 21. Stroup TS, McEvoy JP, Swartz MS, et al. 11. Taipale H, Puranen A, Mittendorfer-Rutz E, et al. The National Institute of Mental Health Clinical s13063-015-1023-4 Antipsychotic use among persons with Antipsychotic Trials of Intervention Effectiveness 31. Hofer A, Hummer M, Huber R, Kurz M, Walch T, schizophrenia in Sweden and Finland, trends and (CATIE) project: schizophrenia trial design and Fleischhacker WW. Selection bias in clinical trials differences. Nord J Psychiatry. 2021;75(5):315-322. protocol development. Schizophr Bull. 2003;29(1): with antipsychotics. J Clin Psychopharmacol. 2000; doi:10.1080/08039488.2020.1854853 15-31. doi:10.1093/oxfordjournals.schbul.a006986 20(6):699-702. doi:10.1097/00004714- 12. Tanskanen A, Taipale H, Koponen M, et al. 22. Haro JM, Suarez D, Novick D, Brown J, 200012000-00019 From prescription drug purchases to drug use Usall J, Naber D; SOHO Study Group. Three-year 32. Shah JL, Peters MI. Early intervention in periods: a second generation method (PRE2DUP). antipsychotic effectiveness in the outpatient care psychiatry: scotomas, representativeness, and the BMC Med Inform Decis Mak. 2015;15:21. of schizophrenia: observational versus randomized lens of clinical populations. Soc Psychiatry Psychiatr doi:10.1186/s12911-015-0140-z studies results. Eur Neuropsychopharmacol. 2007; Epidemiol. 2019;54(9):1019-1021. doi:10.1007/ 13. Krause M, Huhn M, Schneider-Thoma J, 17(4):235-244. doi:10.1016/j.euroneuro.2006.09.005 s00127-019-01686-x Bighelli I, Gutsmiedl K, Leucht S. Efficacy, 23. Haro JM, Edgell ET, Jones PB, et al; 33. Taipale H, Tiihonen J. Registry-based studies: acceptability and tolerability of antipsychotics SOHO Study Group. The European Schizophrenia what they can tell us, and what they cannot. Eur in patients with schizophrenia and comorbid Outpatient Health Outcomes (SOHO) study: Neuropsychopharmacol. 2021;45:35-37. doi:10.1016/ substance use: a systematic review and rationale, methods and recruitment. Acta Psychiatr j.euroneuro.2021.03.005 meta-analysis. Eur Neuropsychopharmacol. 2019; Scand. 2003;107(3):222-232. doi:10.1034/j.1600- 29(1):32-45. doi:10.1016/j.euroneuro.2018.11.1105 34. Correll CU, Kishimoto T, Kane JM. Randomized 0447.2003.00064.x controlled trials in schizophrenia: opportunities, 14. Krause M, Huhn M, Schneider-Thoma J, 24. Vieta E. Observational, pragmatic, and clinical limitations, and trial design alternatives. Dialogues Rothe P, Smith RC, Leucht S. Antipsychotic drugs trials in bipolar disorder. J Clin Psychiatry. 2008;69 Clin Neurosci. 2011;13(2):155-172. doi:10.31887/ for elderly patients with schizophrenia: (9):e27. doi:10.4088/JCP.0908e27 DCNS.2011.13.2/ccorrell a systematic review and meta-analysis. Eur Neuropsychopharmacol. 2018;28(12):1360-1370. 25. Isohanni M, Mäkikyrö T, Moring J, et al. 35. Stroup TS, Geddes JR. Randomized controlled doi:10.1016/j.euroneuro.2018.09.007 A comparison of clinical and research DSM-III-R trials for schizophrenia: study designs targeted to diagnoses of schizophrenia in a Finnish national distinct goals. Schizophr Bull. 2008;34(2):266-274. 15. Krause M, Zhu Y, Huhn M, et al. Efficacy, birth cohort: clinical and research diagnoses of doi:10.1093/schbul/sbm156 acceptability, and tolerability of antipsychotics schizophrenia. Soc Psychiatry Psychiatr Epidemiol. in children and adolescents with schizophrenia: 36. Alphs L, Benson C, Cheshire-Kinney K, et al. 1997;32(5):303-308. doi:10.1007/BF00789044 Real-world outcomes of paliperidone palmitate a network meta-analysis. Eur Neuropsychopharmacol. 2018;28(6):659-674. doi:10.1016/j.euroneuro.2018. 26. Riedel M, Strassnig M, Müller N, Zwack P, compared to daily oral antipsychotic therapy in 03.008 Möller HJ. How representative of everyday clinical schizophrenia: a randomized, open-label, review populations are schizophrenia patients enrolled board-blinded 15-month study. J Clin Psychiatry. 16. Cipriani A, Ioannidis JPA, Rothwell PM, et al. in clinical trials? Eur Arch Psychiatry Clin Neurosci. 2015;76(5):554-561. doi:10.4088/JCP.14m09584 Generating comparative evidence on new drugs 218 JAMA Psychiatry March 2022 Volume 79, Number 3 (Reprinted) jamapsychiatry.com Downloaded From: https://jamanetwork.com/ on 09/15/2023