array(41) {
["project_status"]=>
string(7) "ongoing"
["project_assoc_trials"]=>
array(9) {
[0]=>
object(WP_Post)#5552 (24) {
["ID"]=>
int(1166)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2014-09-22 14:26:00"
["post_date_gmt"]=>
string(19) "2014-09-22 14:26:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(216) "NCT00334126 - A Randomized, Double-blind, Placebo-controlled, Parallel Group Study to Evaluate the Efficacy and Safety of Paliperidone ER Compared to Quetiapine in Subjects With an Acute Exacerbation of Schizophrenia"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(194) "nct00334126-a-randomized-double-blind-placebo-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-paliperidone-er-compared-to-quetiapine-in-subjects-with-an-acute-exacerbation"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-05-14 16:51:16"
["post_modified_gmt"]=>
string(19) "2025-05-14 20:51:16"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(243) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00334126-a-randomized-double-blind-placebo-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-paliperidone-er-compared-to-quetiapine-in-subjects-with-an-acute-exacerbation/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[1]=>
object(WP_Post)#5551 (24) {
["ID"]=>
int(1227)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2014-10-20 14:30:00"
["post_date_gmt"]=>
string(19) "2014-10-20 14:30:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(211) "NCT00397033 - A Randomized, Double-blind, Placebo-controlled, Parallel-group Study to Evaluate the Efficacy and Safety of Two Dosages of Paliperidone ER in the Treatment of Patients With Schizoaffective Disorder"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(195) "nct00397033-a-randomized-double-blind-placebo-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-two-dosages-of-paliperidone-er-in-the-treatment-of-patients-with-schizoaffecti"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-05-14 16:53:55"
["post_modified_gmt"]=>
string(19) "2025-05-14 20:53:55"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(244) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00397033-a-randomized-double-blind-placebo-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-two-dosages-of-paliperidone-er-in-the-treatment-of-patients-with-schizoaffecti/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[2]=>
object(WP_Post)#5550 (24) {
["ID"]=>
int(1230)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2014-10-20 14:32:00"
["post_date_gmt"]=>
string(19) "2014-10-20 14:32:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(211) "NCT00412373 - A Randomized, Double-blind, Placebo-controlled, Parallel- Group Study to Evaluate the Efficacy and Safety of Flexible-dose Paliperidone ER in the Treatment of Patients With Schizoaffective Disorder"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(194) "nct00412373-a-randomized-double-blind-placebo-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-flexible-dose-paliperidone-er-in-the-treatment-of-patients-with-schizoaffecti"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-05-14 16:54:48"
["post_modified_gmt"]=>
string(19) "2025-05-14 20:54:48"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(243) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00412373-a-randomized-double-blind-placebo-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-flexible-dose-paliperidone-er-in-the-treatment-of-patients-with-schizoaffecti/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[3]=>
object(WP_Post)#5548 (24) {
["ID"]=>
int(1395)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2015-10-20 11:15:00"
["post_date_gmt"]=>
string(19) "2015-10-20 11:15:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(301) "NCT00083668 - A Randomized, Double-blind, Placebo- and Active-controlled, Parallel-group, Dose-response Study to Evaluate the Efficacy and Safety of 3 Fixed Dosages of Paliperidone Extended Release (ER) Tablets and Olanzapine, With Open-label Extension, in the Treatment of Patients With Schizophrenia"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(192) "nct00083668-a-randomized-double-blind-placebo-and-active-controlled-parallel-group-dose-response-study-to-evaluate-the-efficacy-and-safety-of-3-fixed-dosages-of-paliperidone-extended-release-e"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-22 12:20:47"
["post_modified_gmt"]=>
string(19) "2025-10-22 16:20:47"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(241) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00083668-a-randomized-double-blind-placebo-and-active-controlled-parallel-group-dose-response-study-to-evaluate-the-efficacy-and-safety-of-3-fixed-dosages-of-paliperidone-extended-release-e/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[4]=>
object(WP_Post)#5547 (24) {
["ID"]=>
int(1434)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2016-01-22 08:48:00"
["post_date_gmt"]=>
string(19) "2016-01-22 08:48:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(268) "NCT00085748 - A Randomized, 6-Week Double-Blind, Placebo-Controlled Study With an Optional 24-Week Open-Label Extension to Evaluate the Safety and Tolerability of Flexible Doses of Paliperidone Extended Release in the Treatment of Geriatric Patients With Schizophrenia"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(196) "nct00085748-a-randomized-6-week-double-blind-placebo-controlled-study-with-an-optional-24-week-open-label-extension-to-evaluate-the-safety-and-tolerability-of-flexible-doses-of-paliperidone-extend"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-10-22 12:18:30"
["post_modified_gmt"]=>
string(19) "2025-10-22 16:18:30"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(245) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00085748-a-randomized-6-week-double-blind-placebo-controlled-study-with-an-optional-24-week-open-label-extension-to-evaluate-the-safety-and-tolerability-of-flexible-doses-of-paliperidone-extend/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[5]=>
object(WP_Post)#5544 (24) {
["ID"]=>
int(1499)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2016-03-31 12:54:00"
["post_date_gmt"]=>
string(19) "2016-03-31 12:54:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(248) "NCT00524043 - A Randomized, Double-Blind, Placebo- and Active-Controlled, Parallel-Group Study to Evaluate the Efficacy and Safety of a Fixed Dosage of 1.5 mg/Day of Paliperidone Extended Release (ER) in the Treatment of Subjects With Schizophrenia"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(192) "nct00524043-a-randomized-double-blind-placebo-and-active-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-a-fixed-dosage-of-1-5-mg-day-of-paliperidone-extended-release-er"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-12-09 13:28:52"
["post_modified_gmt"]=>
string(19) "2025-12-09 18:28:52"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(241) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00524043-a-randomized-double-blind-placebo-and-active-controlled-parallel-group-study-to-evaluate-the-efficacy-and-safety-of-a-fixed-dosage-of-1-5-mg-day-of-paliperidone-extended-release-er/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[6]=>
object(WP_Post)#5549 (24) {
["ID"]=>
int(1356)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2015-07-16 10:24:00"
["post_date_gmt"]=>
string(19) "2015-07-16 10:24:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(158) "NCT00249132 - A Canadian multicenter placebo-controlled study of fixed doses of risperidone and haloperidol in the treatment of chronic schizophrenic patients"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(156) "nct00249132-a-canadian-multicenter-placebo-controlled-study-of-fixed-doses-of-risperidone-and-haloperidol-in-the-treatment-of-chronic-schizophrenic-patients"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-06-24 16:49:18"
["post_modified_gmt"]=>
string(19) "2025-06-24 20:49:18"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(205) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00249132-a-canadian-multicenter-placebo-controlled-study-of-fixed-doses-of-risperidone-and-haloperidol-in-the-treatment-of-chronic-schizophrenic-patients/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[7]=>
object(WP_Post)#5546 (24) {
["ID"]=>
int(1461)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2016-03-31 12:30:00"
["post_date_gmt"]=>
string(19) "2016-03-31 12:30:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(167) "NCT00088075 - A Randomized, Double-Blind, Placebo-Controlled Clinical Study of the Efficacy and Safety of Risperidone for the Treatment of Schizophrenia in Adolescents"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(163) "nct00088075-a-randomized-double-blind-placebo-controlled-clinical-study-of-the-efficacy-and-safety-of-risperidone-for-the-treatment-of-schizophrenia-in-adolescents"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2025-12-18 11:16:38"
["post_modified_gmt"]=>
string(19) "2025-12-18 16:16:38"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(212) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00088075-a-randomized-double-blind-placebo-controlled-clinical-study-of-the-efficacy-and-safety-of-risperidone-for-the-treatment-of-schizophrenia-in-adolescents/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
[8]=>
object(WP_Post)#5545 (24) {
["ID"]=>
int(1946)
["post_author"]=>
string(4) "1363"
["post_date"]=>
string(19) "2016-03-31 12:36:00"
["post_date_gmt"]=>
string(19) "2016-03-31 12:36:00"
["post_content"]=>
string(0) ""
["post_title"]=>
string(106) "NCT00253136 - Risperidone Depot (Microspheres) vs. Placebo in the Treatment of Subjects With Schizophrenia"
["post_excerpt"]=>
string(0) ""
["post_status"]=>
string(7) "publish"
["comment_status"]=>
string(6) "closed"
["ping_status"]=>
string(6) "closed"
["post_password"]=>
string(0) ""
["post_name"]=>
string(103) "nct00253136-risperidone-depot-microspheres-vs-placebo-in-the-treatment-of-subjects-with-schizophrenia-2"
["to_ping"]=>
string(0) ""
["pinged"]=>
string(0) ""
["post_modified"]=>
string(19) "2026-01-21 13:16:27"
["post_modified_gmt"]=>
string(19) "2026-01-21 18:16:27"
["post_content_filtered"]=>
string(0) ""
["post_parent"]=>
int(0)
["guid"]=>
string(152) "https://dev-yoda.pantheonsite.io/clinical-trial/nct00253136-risperidone-depot-microspheres-vs-placebo-in-the-treatment-of-subjects-with-schizophrenia-2/"
["menu_order"]=>
int(0)
["post_type"]=>
string(14) "clinical_trial"
["post_mime_type"]=>
string(0) ""
["comment_count"]=>
string(1) "0"
["filter"]=>
string(3) "raw"
}
}
["project_title"]=>
string(67) "Individual prediction of placebo response in acute psychosis trials"
["project_narrative_summary"]=>
string(480) "Clinical trials to test new antipsychotics may be jeopardized by a large placebo response which may make difficult to detect true treatment effects. Identifying placebo responders may allow to interpret the data more accurately and even to design and conduct clinical trials in a more efficient manner. Here, we are using individual characteristics from before treatment onset to predict the individuals who will respond to treatment despite being randomly assigned to a placebo. "
["project_learn_source"]=>
string(9) "colleague"
["principal_investigator"]=>
array(7) {
["first_name"]=>
string(4) "Jose"
["last_name"]=>
string(5) "Rubio"
["degree"]=>
string(7) "MD, PhD"
["primary_affiliation"]=>
string(59) "Northwell Health Feinstein Institutes for Medical Research "
["email"]=>
string(22) "JRubio13@northwell.edu"
["state_or_province"]=>
string(2) "NY"
["country"]=>
string(13) "United States"
}
["project_key_personnel"]=>
array(2) {
[0]=>
array(6) {
["p_pers_f_name"]=>
string(7) "Mehrdad"
["p_pers_l_name"]=>
string(5) "Sadri"
["p_pers_degree"]=>
string(2) "MD"
["p_pers_pr_affil"]=>
string(59) "Northwell Health Feinstein Institutes for Medical Research "
["p_pers_scop_id"]=>
string(0) ""
["requires_data_access"]=>
string(3) "yes"
}
[1]=>
array(6) {
["p_pers_f_name"]=>
string(4) "Joey"
["p_pers_l_name"]=>
string(5) "Zhang"
["p_pers_degree"]=>
string(21) "Bachelors of Science "
["p_pers_pr_affil"]=>
string(58) "Northwell Health Feinstein Institutes for Medical Research"
["p_pers_scop_id"]=>
string(0) ""
["requires_data_access"]=>
string(3) "yes"
}
}
["project_ext_grants"]=>
array(2) {
["value"]=>
string(2) "no"
["label"]=>
string(68) "No external grants or funds are being used to support this research."
}
["project_date_type"]=>
string(18) "full_crs_supp_docs"
["property_scientific_abstract"]=>
string(1581) "Background: Increasing placebo responses in schizophrenia trials impede true drug effect detection, leading to costly late-stage failures. Predicting placebo response from baseline characteristics can enable flexible trial designs for true drug effects. Objective: (1) Develop individual predictive models for symptom remission in placebo-treated acute psychosis patients; (2) Develop secondary models for treatment response (20%, 30%, and 50% total PANSS reductions); (3) Identify key baseline characteristics. Study Design: Extract baseline features from placebo-randomized acute psychosis trial participants. Define symptom remission (PANSS Positive items <4, CGI 20%, 30%, and 50% total PANSS reductions) with individual data. A leave-one-site-out cross-validation random forest classifier will use these features to predict outcomes. Participants: Acute schizophrenia or schizoaffective disorder patients in psychosis trials, randomized to placebo. Outcome Measures: Primary: Symptom remission: PANSS psychosis score ≤3 and CGI ≤3 at any post-baseline visit. Secondary: Treatment response: >20%, 30%, and 50% PANSS Total reduction at week 8, using a linear mixed model for repeated measures. Statistical Analysis: Data preprocessing per Chekroud et al, Science 2024 will extract baseline features. A random forest classifier will predict using these features and outcomes. Hyperparameter tuning with grid-search cross-validation will optimize the model. Performance will be assessed by ROC AUC. Feature importance will use standardized coefficients and SHAP values. "
["project_brief_bg"]=>
string(628) "Placebo response in antipsychotic trials has undermined detection of medication effects and has been implicated in expensive failures of drug development [1,2]. Acute psychosis trials are particularly sensitive to placebo and early non-pharmacologic improvements [3]. By modeling placebo trajectories and identifying predictors of placebo response at the patient level, we can (a) improve interpretation of existing trial outcomes; (b) inform trial design (sample enrichment, duration, endpoints); and (c) provide clinicians and investigators with individualized estimates of expected short-term change under placebo-level care."
["project_specific_aims"]=>
string(354) "(1) Develop individual predictive models of symptom remission in individuals with acute psychosis randomly allocated to placebo
(2) Develop secondary models for treatment response under these circumstances (defined as 20%, 30%, and 50% reductions in total PANSS)
(3) Identify the baseline characteristics that drive such predictions
"
["project_study_design"]=>
array(2) {
["value"]=>
string(5) "other"
["label"]=>
string(5) "Other"
}
["project_purposes"]=>
array(1) {
[0]=>
array(2) {
["value"]=>
string(50) "research_on_clinical_prediction_or_risk_prediction"
["label"]=>
string(50) "Research on clinical prediction or risk prediction"
}
}
["project_research_methods"]=>
string(656) "- Data Source: Patient-level data from placebo controlled schizophrenia/schizoaffective trials available via the YODA Project:
NCT00334126
NCT00397033
NCT00412373
NCT00083668
NCT00085748
NCT00524043
NCT00249132
NCT00088075
NCT00253136
- Inclusion Criteria:
- Trial duration >=6 weeks
- Diagnosis of Schizophrenia or schizoaffective
- Acutely ill participants (i.e., study defined or equivalent to PANSS > 70)
- Random allocation to placebo in at least one group
- Exclusion Criteria: None beyond failing inclusion requirements.
"
["project_main_outcome_measure"]=>
string(312) "Primary:
- Symptomatic remission defined as PANSS psychosis items score ≤3 and CGI ≤3 in at least one visit after baseline
Secondary:
- Treatment response defined as >20%, 30%, and 50% PANSS Total improvement at week 8 estimated from a linear mixed model for repeated measures"
["project_main_predictor_indep"]=>
string(243) "Predictors:
- All baseline features; All available information (demographic, medical and psychiatric illness, adverse events, psychopathology) which will be used as the input features of the Random Forest Model.
"
["project_other_variables_interest"]=>
string(3) "N/A"
["project_stat_analysis_plan"]=>
string(3172) "1. Preprocessing
Data preprocessing per Chekroud et al. 2024 will extract baseline features [4]. We will harmonize all placebo-controlled acute schizophrenia and schizoaffective disorder trials by aligning observations on a common time axis (days since baseline).
Missing data will be imputed using k-nearest neighbor (KNNs).
If measures of the same construct were conducted using comparable yet different rating scales, we will z score values, otherwise raw scores will be used.
Trial and site indicators will be included to control for heterogeneity across studies.
For the placebo-response prediction model, we will generate a baseline-only feature set including demographic, clinical, illness history, baseline PANSS subscale items, baseline psychosis factor scores, and baseline functioning.
All preprocessing will occur within each trial to avoid data leakage.
2. Modeling Approaches
a) Linear Mixed-Effects Models (Primary: Symptom Trajectory Modeling)
We will use linear mixed-effects (LME) models to estimate individual-specific symptom trajectories under placebo. We will estimate for each individual the predicted PANSS Total score at week 8
• Outcome: continuous PANSS total score across visits.
• Fixed effects: time, baseline PANSS, baseline covariates, and trial/site indicators.
• Random effects: random intercepts and random slopes for time at the participant level to capture individualized symptom trajectory under placebo.
• The random slopes extracted from these models will quantify each individual's rate of change in symptoms.
b) Machine Learning Model (Random Forest) using baseline features to predict primary and secondary outcome as described above.
• Features: all baseline demographic, clinical, laboratory, PANSS item-level scores, illness severity indices, duration of illness, acute exacerbation characteristics, comorbidity, side effects, site, and trial indicators.
• Rationale: This model estimates whether baseline characteristics can predict who naturally improves in acute psychosis without antipsychotic medication, which is crucial for interpreting drug effects.
Hyperparameters will be tuned via nested cross-validation within the training folds.
3. Validation
Generalizability will be assessed by cross-trial validation:
• Leave-One-Trial-Out:
o Train models on all but one trial
o Test on the held-out trial
o Rotate through all trials
All preprocessing and imputation will be done within each trial, and hyperparameter tuning occur only within the training portion of each split.
For the Random Forest model:
• Balanced class weights will be used to address imbalanced placebo response rates.
• Performance metrics: AUC value of the ROC analysis
4. Feature Importance will be explored based on SHAP values to understand variable contributions to the Random Forest model performance.
These analyses will identify baseline predictors most strongly associated with placebo improvement or remission.
"
["project_software_used"]=>
array(2) {
[0]=>
array(2) {
["value"]=>
string(6) "python"
["label"]=>
string(6) "Python"
}
[1]=>
array(2) {
["value"]=>
string(7) "rstudio"
["label"]=>
string(7) "RStudio"
}
}
["project_timeline"]=>
string(189) "We plan to start as soon as we are granted access to the data. We anticipate 3 months for data preprocessing, 6 months for analyses, and 3 months for manuscript preparation and submission. "
["project_dissemination_plan"]=>
string(382) "Findings will be submitted to high-impact psychiatry and psychopharmacology journals and presented at national conferences. Code and analytic pipelines will be shared in line with YODA's open-science principles. We will target the Lancet Psychiatry as preferred journal, otherwise we will submit to JAMA Psychiatry and the American Journal of Psychiatry, depending on the findings. "
["project_bibliography"]=>
string(1104) "1. Agid, O., Siu, C. O., Potkin, S. G., Kapur, S., Watsky, E., Vanderburg, D., … & Remington, G. (2013). Meta-regression analysis of placebo response in antipsychotic trials, 1970–2010. American Journal of Psychiatry, 170(11), 1335-1344.
2. Rutherford, B. R., Pott, E., Tandler, J. M., Wall, M. M., Roose, S. P., & Lieberman, J. A. (2014). Placebo response in antipsychotic clinical trials: a meta-analysis. JAMA psychiatry, 71(12), 1409-1421.
3. Leucht, S., Leucht, C., Huhn, M., Chaimani, A., Mavridis, D., Helfer, B., … & Davis, J. M. (2017). Sixty years of placebo-controlled antipsychotic drug trials in acute schizophrenia: systematic review, Bayesian meta-analysis, and meta-regression of efficacy predictors. American Journal of Psychiatry, 174(10), 927-942.
4. Chekroud, A. M., Hawrilenko, M., Loho, H., Bondar, J., Gueorguieva, R., Hasan, A., … & Paulus, M. (2024). Illusory generalizability of clinical prediction models. Science, 383(6679), 164-167.
"
["project_suppl_material"]=>
bool(false)
["project_coi"]=>
array(3) {
[0]=>
array(1) {
["file_coi"]=>
array(21) {
["ID"]=>
int(18491)
["id"]=>
int(18491)
["title"]=>
string(21) "COI_Mehrdad-Sadri.pdf"
["filename"]=>
string(21) "COI_Mehrdad-Sadri.pdf"
["filesize"]=>
int(17831)
["url"]=>
string(70) "https://yoda.yale.edu/wp-content/uploads/2025/12/COI_Mehrdad-Sadri.pdf"
["link"]=>
string(67) "https://yoda.yale.edu/data-request/2025-0900/coi_mehrdad-sadri-pdf/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "2274"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(21) "coi_mehrdad-sadri-pdf"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(18456)
["date"]=>
string(19) "2025-12-26 18:09:00"
["modified"]=>
string(19) "2025-12-26 18:09:03"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
}
[1]=>
array(1) {
["file_coi"]=>
array(21) {
["ID"]=>
int(18493)
["id"]=>
int(18493)
["title"]=>
string(18) "COI_Jose_Rubio.pdf"
["filename"]=>
string(18) "COI_Jose_Rubio.pdf"
["filesize"]=>
int(28125)
["url"]=>
string(67) "https://yoda.yale.edu/wp-content/uploads/2025/12/COI_Jose_Rubio.pdf"
["link"]=>
string(64) "https://yoda.yale.edu/data-request/2025-0900/coi_jose_rubio-pdf/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "2274"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(18) "coi_jose_rubio-pdf"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(18456)
["date"]=>
string(19) "2025-12-29 17:22:37"
["modified"]=>
string(19) "2025-12-29 17:22:38"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
}
[2]=>
array(1) {
["file_coi"]=>
array(21) {
["ID"]=>
int(18492)
["id"]=>
int(18492)
["title"]=>
string(18) "COI_Joey_Zhang.pdf"
["filename"]=>
string(18) "COI_Joey_Zhang.pdf"
["filesize"]=>
int(18486)
["url"]=>
string(67) "https://yoda.yale.edu/wp-content/uploads/2025/12/COI_Joey_Zhang.pdf"
["link"]=>
string(64) "https://yoda.yale.edu/data-request/2025-0900/coi_joey_zhang-pdf/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "2274"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(18) "coi_joey_zhang-pdf"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(18456)
["date"]=>
string(19) "2025-12-29 17:21:56"
["modified"]=>
string(19) "2025-12-29 17:21:58"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
}
}
["data_use_agreement_training"]=>
bool(true)
["human_research_protection_training"]=>
bool(true)
["certification"]=>
bool(true)
["request_data_partner"]=>
string(15) "johnson-johnson"
["project_study_design_exp"]=>
string(65) "Leave one site out cross validated individual predictive modeling"
["search_order"]=>
string(1) "0"
["project_send_email_updates"]=>
bool(false)
["project_publ_available"]=>
bool(true)
["project_year_access"]=>
string(4) "2026"
["project_rep_publ"]=>
bool(false)
["project_assoc_data"]=>
array(0) {
}
["project_due_dil_assessment"]=>
array(21) {
["ID"]=>
int(18778)
["id"]=>
int(18778)
["title"]=>
string(47) "YODA Project Due Diligence Assessment 2025-0900"
["filename"]=>
string(51) "YODA-Project-Due-Diligence-Assessment-2025-0900.pdf"
["filesize"]=>
int(132525)
["url"]=>
string(100) "https://yoda.yale.edu/wp-content/uploads/2025/12/YODA-Project-Due-Diligence-Assessment-2025-0900.pdf"
["link"]=>
string(93) "https://yoda.yale.edu/data-request/2025-0900/yoda-project-due-diligence-assessment-2025-0900/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1885"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(47) "yoda-project-due-diligence-assessment-2025-0900"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(18456)
["date"]=>
string(19) "2026-02-17 15:27:51"
["modified"]=>
string(19) "2026-02-17 15:27:51"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
["project_title_link"]=>
array(21) {
["ID"]=>
int(18779)
["id"]=>
int(18779)
["title"]=>
string(46) "YODA Project Protocol - 2025-0900 - 2025-12-29"
["filename"]=>
string(46) "YODA-Project-Protocol-2025-0900-2025-12-29.pdf"
["filesize"]=>
int(185832)
["url"]=>
string(95) "https://yoda.yale.edu/wp-content/uploads/2025/12/YODA-Project-Protocol-2025-0900-2025-12-29.pdf"
["link"]=>
string(88) "https://yoda.yale.edu/data-request/2025-0900/yoda-project-protocol-2025-0900-2025-12-29/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1885"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(42) "yoda-project-protocol-2025-0900-2025-12-29"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(18456)
["date"]=>
string(19) "2026-02-17 15:28:11"
["modified"]=>
string(19) "2026-02-17 15:28:11"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
["project_review_link"]=>
array(21) {
["ID"]=>
int(18780)
["id"]=>
int(18780)
["title"]=>
string(36) "YODA Project Review - 2025-0900_site"
["filename"]=>
string(38) "YODA-Project-Review-2025-0900_site.pdf"
["filesize"]=>
int(1331843)
["url"]=>
string(87) "https://yoda.yale.edu/wp-content/uploads/2025/12/YODA-Project-Review-2025-0900_site.pdf"
["link"]=>
string(80) "https://yoda.yale.edu/data-request/2025-0900/yoda-project-review-2025-0900_site/"
["alt"]=>
string(0) ""
["author"]=>
string(4) "1885"
["description"]=>
string(0) ""
["caption"]=>
string(0) ""
["name"]=>
string(34) "yoda-project-review-2025-0900_site"
["status"]=>
string(7) "inherit"
["uploaded_to"]=>
int(18456)
["date"]=>
string(19) "2026-02-17 15:28:27"
["modified"]=>
string(19) "2026-02-17 15:28:27"
["menu_order"]=>
int(0)
["mime_type"]=>
string(15) "application/pdf"
["type"]=>
string(11) "application"
["subtype"]=>
string(3) "pdf"
["icon"]=>
string(62) "https://yoda.yale.edu/wp/wp-includes/images/media/document.png"
}
["project_highlight_button"]=>
string(0) ""
["request_overridden_res"]=>
string(1) "3"
}
data partner
array(1) {
[0]=>
string(15) "johnson-johnson"
}
pi country
array(0) {
}
pi affil
array(0) {
}
products
array(2) {
[0]=>
string(6) "invega"
[1]=>
string(9) "risperdal"
}
num of trials
array(1) {
[0]=>
string(1) "9"
}
res
array(1) {
[0]=>
string(1) "3"
}
General Information
How did you learn about the YODA Project?:
Colleague
Conflict of Interest
Request Clinical Trials
Associated Trial(s):
- NCT00334126 - A Randomized, Double-blind, Placebo-controlled, Parallel Group Study to Evaluate the Efficacy and Safety of Paliperidone ER Compared to Quetiapine in Subjects With an Acute Exacerbation of Schizophrenia
- NCT00397033 - A Randomized, Double-blind, Placebo-controlled, Parallel-group Study to Evaluate the Efficacy and Safety of Two Dosages of Paliperidone ER in the Treatment of Patients With Schizoaffective Disorder
- NCT00412373 - A Randomized, Double-blind, Placebo-controlled, Parallel- Group Study to Evaluate the Efficacy and Safety of Flexible-dose Paliperidone ER in the Treatment of Patients With Schizoaffective Disorder
- NCT00083668 - A Randomized, Double-blind, Placebo- and Active-controlled, Parallel-group, Dose-response Study to Evaluate the Efficacy and Safety of 3 Fixed Dosages of Paliperidone Extended Release (ER) Tablets and Olanzapine, With Open-label Extension, in the Treatment of Patients With Schizophrenia
- NCT00085748 - A Randomized, 6-Week Double-Blind, Placebo-Controlled Study With an Optional 24-Week Open-Label Extension to Evaluate the Safety and Tolerability of Flexible Doses of Paliperidone Extended Release in the Treatment of Geriatric Patients With Schizophrenia
- NCT00524043 - A Randomized, Double-Blind, Placebo- and Active-Controlled, Parallel-Group Study to Evaluate the Efficacy and Safety of a Fixed Dosage of 1.5 mg/Day of Paliperidone Extended Release (ER) in the Treatment of Subjects With Schizophrenia
- NCT00249132 - A Canadian multicenter placebo-controlled study of fixed doses of risperidone and haloperidol in the treatment of chronic schizophrenic patients
- NCT00088075 - A Randomized, Double-Blind, Placebo-Controlled Clinical Study of the Efficacy and Safety of Risperidone for the Treatment of Schizophrenia in Adolescents
- NCT00253136 - Risperidone Depot (Microspheres) vs. Placebo in the Treatment of Subjects With Schizophrenia
What type of data are you looking for?:
Individual Participant-Level Data, which includes Full CSR and all supporting documentation
Request Clinical Trials
Data Request Status
Status:
Ongoing
Research Proposal
Project Title:
Individual prediction of placebo response in acute psychosis trials
Scientific Abstract:
Background: Increasing placebo responses in schizophrenia trials impede true drug effect detection, leading to costly late-stage failures. Predicting placebo response from baseline characteristics can enable flexible trial designs for true drug effects. Objective: (1) Develop individual predictive models for symptom remission in placebo-treated acute psychosis patients; (2) Develop secondary models for treatment response (20%, 30%, and 50% total PANSS reductions); (3) Identify key baseline characteristics. Study Design: Extract baseline features from placebo-randomized acute psychosis trial participants. Define symptom remission (PANSS Positive items <4, CGI 20%, 30%, and 50% total PANSS reductions) with individual data. A leave-one-site-out cross-validation random forest classifier will use these features to predict outcomes. Participants: Acute schizophrenia or schizoaffective disorder patients in psychosis trials, randomized to placebo. Outcome Measures: Primary: Symptom remission: PANSS psychosis score <=3 and CGI <=3 at any post-baseline visit. Secondary: Treatment response: >20%, 30%, and 50% PANSS Total reduction at week 8, using a linear mixed model for repeated measures. Statistical Analysis: Data preprocessing per Chekroud et al, Science 2024 will extract baseline features. A random forest classifier will predict using these features and outcomes. Hyperparameter tuning with grid-search cross-validation will optimize the model. Performance will be assessed by ROC AUC. Feature importance will use standardized coefficients and SHAP values.
Brief Project Background and Statement of Project Significance:
Placebo response in antipsychotic trials has undermined detection of medication effects and has been implicated in expensive failures of drug development [1,2]. Acute psychosis trials are particularly sensitive to placebo and early non-pharmacologic improvements [3]. By modeling placebo trajectories and identifying predictors of placebo response at the patient level, we can (a) improve interpretation of existing trial outcomes; (b) inform trial design (sample enrichment, duration, endpoints); and (c) provide clinicians and investigators with individualized estimates of expected short-term change under placebo-level care.
Specific Aims of the Project:
(1) Develop individual predictive models of symptom remission in individuals with acute psychosis randomly allocated to placebo
(2) Develop secondary models for treatment response under these circumstances (defined as 20%, 30%, and 50% reductions in total PANSS)
(3) Identify the baseline characteristics that drive such predictions
Study Design:
Other
Explain:
Leave one site out cross validated individual predictive modeling
What is the purpose of the analysis being proposed? Please select all that apply.:
Research on clinical prediction or risk prediction
Software Used:
Python, RStudio
Data Source and Inclusion/Exclusion Criteria to be used to define the patient sample for your study:
- Data Source: Patient-level data from placebo controlled schizophrenia/schizoaffective trials available via the YODA Project:
NCT00334126
NCT00397033
NCT00412373
NCT00083668
NCT00085748
NCT00524043
NCT00249132
NCT00088075
NCT00253136
- Inclusion Criteria:
- Trial duration >=6 weeks
- Diagnosis of Schizophrenia or schizoaffective
- Acutely ill participants (i.e., study defined or equivalent to PANSS > 70)
- Random allocation to placebo in at least one group
- Exclusion Criteria: None beyond failing inclusion requirements.
Primary and Secondary Outcome Measure(s) and how they will be categorized/defined for your study:
Primary:
- Symptomatic remission defined as PANSS psychosis items score <=3 and CGI <=3 in at least one visit after baseline
Secondary:
- Treatment response defined as >20%, 30%, and 50% PANSS Total improvement at week 8 estimated from a linear mixed model for repeated measures
Main Predictor/Independent Variable and how it will be categorized/defined for your study:
Predictors:
- All baseline features; All available information (demographic, medical and psychiatric illness, adverse events, psychopathology) which will be used as the input features of the Random Forest Model.
Other Variables of Interest that will be used in your analysis and how they will be categorized/defined for your study:
N/A
Statistical Analysis Plan:
1. Preprocessing
Data preprocessing per Chekroud et al. 2024 will extract baseline features [4]. We will harmonize all placebo-controlled acute schizophrenia and schizoaffective disorder trials by aligning observations on a common time axis (days since baseline).
Missing data will be imputed using k-nearest neighbor (KNNs).
If measures of the same construct were conducted using comparable yet different rating scales, we will z score values, otherwise raw scores will be used.
Trial and site indicators will be included to control for heterogeneity across studies.
For the placebo-response prediction model, we will generate a baseline-only feature set including demographic, clinical, illness history, baseline PANSS subscale items, baseline psychosis factor scores, and baseline functioning.
All preprocessing will occur within each trial to avoid data leakage.
2. Modeling Approaches
a) Linear Mixed-Effects Models (Primary: Symptom Trajectory Modeling)
We will use linear mixed-effects (LME) models to estimate individual-specific symptom trajectories under placebo. We will estimate for each individual the predicted PANSS Total score at week 8
- Outcome: continuous PANSS total score across visits.
- Fixed effects: time, baseline PANSS, baseline covariates, and trial/site indicators.
- Random effects: random intercepts and random slopes for time at the participant level to capture individualized symptom trajectory under placebo.
- The random slopes extracted from these models will quantify each individual's rate of change in symptoms.
b) Machine Learning Model (Random Forest) using baseline features to predict primary and secondary outcome as described above.
- Features: all baseline demographic, clinical, laboratory, PANSS item-level scores, illness severity indices, duration of illness, acute exacerbation characteristics, comorbidity, side effects, site, and trial indicators.
- Rationale: This model estimates whether baseline characteristics can predict who naturally improves in acute psychosis without antipsychotic medication, which is crucial for interpreting drug effects.
Hyperparameters will be tuned via nested cross-validation within the training folds.
3. Validation
Generalizability will be assessed by cross-trial validation:
- Leave-One-Trial-Out:
o Train models on all but one trial
o Test on the held-out trial
o Rotate through all trials
All preprocessing and imputation will be done within each trial, and hyperparameter tuning occur only within the training portion of each split.
For the Random Forest model:
- Balanced class weights will be used to address imbalanced placebo response rates.
- Performance metrics: AUC value of the ROC analysis
4. Feature Importance will be explored based on SHAP values to understand variable contributions to the Random Forest model performance.
These analyses will identify baseline predictors most strongly associated with placebo improvement or remission.
Narrative Summary:
Clinical trials to test new antipsychotics may be jeopardized by a large placebo response which may make difficult to detect true treatment effects. Identifying placebo responders may allow to interpret the data more accurately and even to design and conduct clinical trials in a more efficient manner. Here, we are using individual characteristics from before treatment onset to predict the individuals who will respond to treatment despite being randomly assigned to a placebo.
Project Timeline:
We plan to start as soon as we are granted access to the data. We anticipate 3 months for data preprocessing, 6 months for analyses, and 3 months for manuscript preparation and submission.
Dissemination Plan:
Findings will be submitted to high-impact psychiatry and psychopharmacology journals and presented at national conferences. Code and analytic pipelines will be shared in line with YODA's open-science principles. We will target the Lancet Psychiatry as preferred journal, otherwise we will submit to JAMA Psychiatry and the American Journal of Psychiatry, depending on the findings.
Bibliography:
1. Agid, O., Siu, C. O., Potkin, S. G., Kapur, S., Watsky, E., Vanderburg, D., … & Remington, G. (2013). Meta-regression analysis of placebo response in antipsychotic trials, 1970--2010. American Journal of Psychiatry, 170(11), 1335-1344.
2. Rutherford, B. R., Pott, E., Tandler, J. M., Wall, M. M., Roose, S. P., & Lieberman, J. A. (2014). Placebo response in antipsychotic clinical trials: a meta-analysis. JAMA psychiatry, 71(12), 1409-1421.
3. Leucht, S., Leucht, C., Huhn, M., Chaimani, A., Mavridis, D., Helfer, B., … & Davis, J. M. (2017). Sixty years of placebo-controlled antipsychotic drug trials in acute schizophrenia: systematic review, Bayesian meta-analysis, and meta-regression of efficacy predictors. American Journal of Psychiatry, 174(10), 927-942.
4. Chekroud, A. M., Hawrilenko, M., Loho, H., Bondar, J., Gueorguieva, R., Hasan, A., … & Paulus, M. (2024). Illusory generalizability of clinical prediction models. Science, 383(6679), 164-167.