Development of an individualized risk calculator of treatment resistance in patients with first-episode psychosis (TRipCal) using automated machine learning: a 12-year follow-up study with clozapine prescription as a proxy indicator.

About 15-40% of patients with schizophrenia are treatment resistance (TR) and require clozapine. Identifying individuals who have higher risk of development of TR early in the course of illness is important to provide personalized intervention. A total of 1400 patients with FEP enrolled in the early intervention for psychosis service or receiving the standard psychiatric service between July 1, 1998, and June 30, 2003, for the first time were included. Clozapine prescriptions until June 2015, as a proxy of TR, were obtained. Premorbid information, baseline characteristics, and monthly clinical information were retrieved systematically from the electronic clinical management system (CMS). Training and testing samples were established with random subsampling. An automated machine learning (autoML) approach was used to optimize the ML algorithm and hyperparameters selection to establish four probabilistic classification models (baseline, 12-month, 24-month, and 36-month information) of TR development. This study found 191 FEP patients (13.7%) who had ever been prescribed clozapine over the follow-up periods. The ML pipelines identified with autoML had an area under the receiver operating characteristic curve ranging from 0.676 (baseline information) to 0.774 (36-month information) in predicting future TR. Features of baseline information, including schizophrenia diagnosis and age of onset, and longitudinal clinical information including symptoms variability, relapse, and use of antipsychotics and anticholinergic medications were important predictors and were included in the risk calculator. The risk calculator for future TR development in FEP patients (TRipCal) developed in this study could support the continuous development of data-driven clinical tools to assist personalized interventions to prevent or postpone TR development in the early course of illness and reduce delay in clozapine initiation.

Subjective-Objective Sleep Discrepancy in Schizophrenia.

OBJECTIVE/BACKGROUND: Subjective methods are often employed for sleep assessment due to their ease of use, but the results may not concur with objective findings. This discrepancy may be present in schizophrenia; however, limited data are available. We performed a secondary analysis to evaluate the agreement between 1-week actigraphy and sleep diary-derived parameters and factors that contribute to subjective-objective sleep discrepancy. PARTICIPANTS: 66 outpatients with schizophrenia (mean age = 44.08 years; 45.45% males). METHODS: Agreement between subjective-objective parameters was assessed using two-way repeated measures ANOVA, Pearson's correlation, and Bland-Altman plot. The magnitude of discrepancy was quantified using Cohen's d. Pearson's correlation was used to determine the significant factors of subjective-objective sleep discrepancy. Benjamini-Hochberg adjustment was performed to account for multiple testing. RESULTS: On average, sleep diaries overestimated sleep onset latency by 20.45 min, total sleep time by 37.63 min, and sleep efficiency by 4.29%, but underestimated wake after sleep onset by 33.28 min. Cohen's d ranged between 0.61 and 1.41. Subjective-objective discrepancies were significantly associated with marital and employment status, self-reported sleep disturbance, delayed sleep-wake phase disorder, chronotype, and psychosocial functioning (r = 0.32-0.44; Benjamini-Hochberg corrected p < .05). CONCLUSIONS: Our findings show that differences between subjective and objective measurements of sleep are present in schizophrenia. Although actigraphy is not a standard procedure for sleep disturbance in schizophrenia, clinical judgment should be used if patients are suspected to have overestimated their sleep difficulties. Further studies should examine whether feedback based on actigraphy can benefit patients with schizophrenia and comorbid sleep disturbances.

Correlates of sleep irregularity in schizophrenia.

Circadian rhythm disruption, manifested as circadian misalignment, difficulty initiating sleep, and sleep-wake irregularity, is often left unrecognized in patients with schizophrenia. Studies have shown multiple adverse consequences of sleep irregularity, but limited data is available on schizophrenia. This is a secondary analysis of a case-control study of 66 schizophrenia patients with delayed sleep-wake phase (with or without meeting criteria for the disorder) and normal sleep-wake phase (mean age = 44.08 years; 45.45% males). Potential correlates included sleep quality, daytime sleepiness, social rhythms, chronotype, psychiatric symptoms, psychosocial functioning, metabolic index, cognitive function, and sociodemographic, lifestyle and pharmacological factors. Square successive difference (SSD) scores, derived from 1-week sleep diary and actigraphy, were indexes of sleep irregularity. Multilevel modeling analysis, with SSD scores as level-1 measures and the hypothesized correlates as level-2 measures, was performed. Statistical significance was Bonferroni-adjusted. Higher SSD scores of sleep diary and actigraphy variables were significantly associated with positive and depressive symptoms, poor sleep quality, daytime sleepiness, irregular social rhythm, evening chronotype, delayed sleep-wake phase disorder, later caffeine use and dinner time, greater cigarette use, and lower dosages of hypnotics and antipsychotics. Our findings highlight the implications of sleep irregularity in schizophrenia and the need for a multimodal intervention.

Risk factors for suicide in the immediate post-discharge period.

BACKGROUND: Suicide risk is highest in the first few months following psychiatric in-patient care. Most data on post-discharge suicides have come from Western countries. Many studies collected cases of suicide over a long post-discharge period and did not focus on this high-risk period. This study aims to describe the characteristics and examine the risk factors of suicides occurring in the immediate post-discharge period in Hong Kong. METHODS: A case-control study based on discharged patients from all psychiatric hospitals/units in Hong Kong in 1997-1999. Suicides occurring within 60 days of discharge from psychiatric hospitals (N = 97) were ascertained by record linkage with Coroner's court data. Controls were matched for age, gender, diagnoses, discharge hospitals, and dates of discharge. Possible risk factors were extracted from in- and out-patient records, and were identified by conditional logistic regression. RESULTS: The commonest diagnosis and suicide method were schizophrenia and falling from a height, respectively. There were no significant case-control differences in the drug treatment received. Risk factors for suicides were: previous deliberate self-harm (OR = 2.3, 95% CI = 1.07-5.05), admission for deliberate self-harm (OR = 3.2, 95% CI = 1.3-7.8), compulsory admission (OR = 3.1, 95% CI = 1.1-8.7), living alone (OR = 5.8, 95% CI = 1.4-23), work stresses (OR = 5.4, 95% CI = 1.5-18) and being out of contact (OR = 7.9, 95% CI = 1.87-33). The overall number of risk factors had greater screening efficacy for suicide than any single factor. CONCLUSIONS: Vulnerable (previous suicidality) and uncooperative (compulsory admission and out of contact) patients who live alone and are exposed to work stresses are prone to immediate post-discharge suicide. Thorough treatment of the circumstances leading to the index admissions, management of work stresses, improved engagement in follow-up care and systematic assessment of suicide risk are indicated.