The catchment area for this observational cohort study is East Azerbaijan, a province in the North of Iran with more than 3,700,000 inhabitants, with the majority having an Azeri ethnic background. Iranian Azeris are the largest minority in the country with a Caucasian ethnic background, and are the world's largest Azeri population who are a Turkic ethnic group. The dominant religion in this region is Islam.
The study target population includes all inhabitants of East Azerbaijan province who are referred with (signs of) a first psychotic episode or evaluated for a recurrent psychotic episode that was previously undiagnosed (< 2 years). The cohort will include patients with a diagnosis of schizophrenia, schizophreniform disorder, delusional disorder, brief psychotic disorder, schizotypal personality disorder, schizoaffective disorder, or substance-induced psychotic disorder based on the Diagnostic and Statistical Manual of Mental Disorders (DSM)-5.(1)
The ARAS project was approved by the ethical committee of the National Institute for Medical research Development (NIMAD) in Iran in 2017 (record number: IR.NIMAD.REC.1396.101). All study procedures will be carried out according to the Declaration of Helsinki. Written informed consent will be obtained from all participants and their caregivers or legal guardians. Patients can leave the study at any time for any reason with no influence on their quality of care. A code will be given to each patient and all data will be recorded anonymously. The participants will be informed that all study findings will be stored and handled in strict confidence according to national guidelines. The principal investigator and the main investigators have access to the materials.
This project started in January 2018. For the first baseline assessment (referred to as T0, Figure 1), clinical interviews and assessments will take place at the treating clinics, starting at the time of admission or as soon as possible if forced admission is required. Psychological testing will be performed within a one-month period after inclusion. Diagnostic interviews, questionnaires, and other instruments will be conducted in Farsi, the official language of the country.
Upcoming assessment as routine outcome monitoring will be scheduled for all patients at the one year follow up (T1), which will include the same assessments used in T0, including outcome measurements and assessment for direct costs. The second follow-up (T3) is scheduled at 3 years and the final follow-up assessment at 5 years (T5) after inclusion. Treatment strategy and visit intervals will be defined by the treating psychiatrist between inclusion and follow-up periods.
Evaluation of patients will be performed in separate meetings depending on their availability (Figure 1). Clinical assessments will be performed by test psychologists/psychiatrists who are trained by repeated workshops.
Clinical diagnosis and biography
Clinical diagnoses will be made by referring psychiatrists from in- and out-patient clinics. The clinical diagnosis within the SSD will be confirmed by the Structured Clinical Interview for DSM-5 (SCID),(30) or for patients < 18 years by the Kiddie-Schedule for Affective Disorders and Schizophrenia Present and Lifetime Versions (K-SADS-PL).(31) Farsi versions of these two questionnaires have been used by several research teams inside Iran and contain few cultural adaptations.(32)
Dimensions of Psychosis will be rated using the symptom severity dimension tool of DSM-5.(33) Severity of symptoms will be rated using the Positive and Negative Syndrome Scale (PANSS), which is a semi-structured interview on the symptom severity including three subscales of positive and negative symptoms and general psychopathology.(34) The severity of depressive symptoms will be measured using the Calgary Depression Scale in Schizophrenia (CDSS).(35) This instrument differentiates between depression and the negative and positive symptoms of schizophrenia.(36)
Side-effects of antipsychotics and the patients’ adherence to their medication will be assessed using the Subject’s Response to Antipsychotics (SRA-34).(37) Evaluation for hyperkinesia, parkinsonism, akathisia, and dystonia will be performed using the St. Hans Rating Scale (SHRS).(38)
Medical history will be obtained from the patient and family members (mostly parents). The nature, severity, and consequences of symptoms for daily functioning will be recorded, and, besides their medication-history, nicotine, illicit drug and alcohol use, and history of illness in the family will be recorded. The biography focuses on signs of an early developmental disorder, current living situation, occupation and daytime activities, and social interactions.
A physical health status examination, including a physical examination and measurement of blood pressure, heart rate, length, weight, and waist circumference will be carried out. Laboratory tests will be conducted to test for general health conditions, including complete blood cell count, lipid profile, fasting blood sugar and HbA1c, liver function, kidney, and pituitary function, and urine screening for drugs (morphine, amphetamine, cannabis, tramadol, methadone if indicated). If necessary, additional tests will be performed based on clinical indications, e.g., brain imaging, electroencephalography, or electrocardiography. The requirement for external consultation for any general medical condition will be decided by the psychiatrist and results will be recorded.
Risk and resilience factors
The evaluation process will continue with semi-structured interviews and self-report questionnaires in different domains. History of experienced adversities during early life and peer interaction in primary school will be evaluated using the Illinois Bully Scale,(39) Olweus Bully/Victim Questionnaire,(40) and the Retrospective bullying experience.(41) The Illinois Bully Scale includes items that address how often a youth is engaged in social aggression, physical and verbal types of victimization, and physical fighting with peers.(35) The Olweus Bully/Victim Questionnaire includes the initiation of an act of bullying against the participant, as well as the expression of bullying behavior against others. The questions include several aspects of bully/victim problems including physical, verbal, indirect, racial, and sexual bullying annoyance. The questionnaire includes pro-bullying and pro-victim attitudes and the reaction of teachers, peers, and parents to the bullying behavior. (36) The Retrospective bullying experience addresses the adult population. It measures the frequency, significance, and extent of bully-victimization. Important aspects like bully-related psychological trauma, suicidal ideation if bullied, and bullying in college and the workplace are also addressed. (37)
Life events will be recorded using the modified Holmes-Rahe stress scale.(42) This inventory includes 50 stressful events adapted to the cultural and social context of the Iranian population that are evaluated by Yes/No questions. The Internalized Stigma of Mental Illness scale (ISMI), a self-report questionnaire measuring self-stigma among persons with psychiatric disorders,(43) will also be used. The Multidimensional Scale of Perceived Social Support (44) will evaluate the extent of support from family, friends, or significant people in patients’ lives.
Religiosity will be measured using a scale based on Stark and Glock's dimensions of religiosity. This scale is adapted to the Islamic religion measuring dimensions of religious beliefs, practice, experience, and consequences with 26 questions.(45)
Assessment of functioning will include i) WHO Disability Assessment Schedule (WHODAS 2.0),(46) a standard tool applicable to clinical and general populations that covers six dimensions of cognition, mobility, self-care, getting along, life activities, and participation;(42) ii) Global Assessment of Functioning (GAF) Scale form DSM (1) and iii) the three item Functional Recovery Tool including daily living and self-care, work, study, and housekeeping, and social contacts.(47, 48) The Manchester Short Assessment of Quality of Life (MANSA) (49) will be used to assess satisfaction with life as a whole and with several domains like leisure, relationships, and mental health.(43)
A comprehensive neurocognitive battery has been designed in accordance with the recommendations of the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) initiative,(50) including eight cognitive domains (see Table 1).
Working memory will be tested with the forward and backward digit span and letter-number sequencing tasks.(51) Divided attention, which is closely tied to executive functioning,(52) will be assessed using the comprehensive (part B) test of the Trail making task. Verbal memory will be tested using The Rey Auditory Verbal Learning Test, measuring memory span, new learning, retention, recognition, and delayed recall.(53) Visual learning will be measured using the Rey-Osterrieth Complex Figure. Executive functioning will be evaluated using the symbol digit modality task, letter digit modality task, backward digit span, and letter-number sequencing task. Speed of information processing will be measured with the average score of both the basic (part A) tests of Stroop (54) and Trail making, (55) measured as time to complete in seconds, as well as category fluency by naming items from three different categories. The Stroop test will also give an estimation about inhibition and flexibility. Intelligence will be tested using the Wechsler Intelligence Scale IV short version,(56) validated for the Iranian population. Regarding social cognition, emotional recognition will be evaluated using the Reading the mind in the eyes test.(57) The Benton Facial Recognition Test will be used to evaluate performance in face discrimination. Theory of mind will be tested using a Sally-Ann story in the Farsi language.
A trained nurse will collect 10 mL venous blood samples in EDTA (ethylenediaminetetraacetic acid) tubes by means of a standard venipuncture at the local hospital laboratory. An individual specific barcode for each patient will be automatically assigned to each tube. This scannable code is unique for every study subject and does not contain elements of their demographics. The code will identify study subjects throughout documentation and evaluation and will be traceable only by the principal investigators. The biobank, located at the study coordinating center, will be monitored by the principal investigator.
Separate samples of whole blood, red blood cells, buffy coat, plasma, and serum will be coded and stored in freezers at -80°C available at the sampling site until further analysis.
All of the described assessments will be repeated during follow-up visits at 1, 3 and 5 years. Relevant outcomes will be measured on three different domains of i) clinical outcome (i.e., symptoms, treatment efficacy, tolerability and cost, diagnostic stability), ii) cognition, and iii) social functioning and recovery (Table 1).
Patients will be invited and interviewed again to assess diagnostic stability and all symptoms will be recorded. Other assessments will include physical examination and laboratory test mentioned earlier, any change in living situation and functioning of the patient, performance on neuro- and social cognition tests, function and disability, perceived stigma, components of metabolic disturbances, and all-cause mortality. Number of relapses, referral to the mental health care system, and re-hospitalization will be recorded.
Primary outcomes will be measured in two categories, symptom remission, including cognitive performance, and functional remission. Analysis will be performed to relate risk/resilience factors to the outcome measures and evaluate the course of outcome. We considered five types of outcomes: (1) dimension of psychosis,(58) (2) cognitive functioning,(58) (3) symptom remission,(59) (4) diagnostic stability,(60) and (5) functional remission.(61) Based on rates reported by previous studies, a minimum number was calculated to be 374 patients. Allowing for dropouts during follow-up, we estimated the total sample size to be 500.
At the level of measurement, inter-rater concordance will be examined for observer rated tools. Raw data will be checked for invalid or nonsense entries. Some raw data will be recoded to composite measures and data for investigators. For example, each component of the cognitive battery might have several tasks and repetitions. These multiple results will be combined to give a summary score based on the structure of the battery. A similar process is also applicable to the other measures.
Descriptive statistics using tabulations and graphical methods will be performed to achieve objective I, after double-checking data entry, data cleaning, and checking for consistency between different types of measurements. Neuropsychological test scores will be converted to domain z scores for correlational analysis to determine predictors of decline or reserve. For other questionnaires, there are predefined values for interpretation. After identifying the distribution of the data, kappa coefficients, correlations, multilevel regression, multidimensional analysis, cluster analysis, and longitudinal trajectory analysis will be used.
The magnitude of the difference between risk and resilience factors will be evaluated by comparing standardized mean differences to achieve objective II. The relationship between risk and resilience factors will be assessed by logistic regression (for binary outcomes) and multivariable linear regressions (for continuous normally distributed outcomes). Appropriate longitudinal data methods will be used for the assessment of disease progression from baseline first episode status to disease at follow-up in years 1, 3, and 5. Multivariable models will determine risk factors that predict cognitive, personal, and clinical outcomes, when appropriate. Multilevel analysis to compare disease progression will also be considered.
We will perform a cluster analysis to achieve objective III. The hierarchical cluster method will be applied to identify the number of clusters for each outcome and seed points for a k-means cluster analysis. As an example, neurocognitive subtypes will be compared on baseline clinical and socio-demographic variables by conducting Chi-square analyses or Kruskal Wallis tests on gender, age, social support, educational level, substance use, schizophrenia spectrum disorder diagnosis, symptom severity, dosage of antipsychotic medication, and adverse life events.
We will define predictive models for objective IV to assess the predictive value of (using the previous example) cognition on symptomatic and functional outcomes at follow-up. Separate clinically and statistically meaningful trajectories will be identified using censored normal group-based trajectory modeling (62) and evaluated for model classification accuracy. Predicted trajectories will be evaluated using a random-effects ordinal regression analysis.