Isabelle Chéreau-Boudet

Background: Missing data in patient-reported outcome (PRO) databases is a pervasive challenge, particularly in psychiatry and psychosocial rehabilitation. Incomplete data may reduce the generalizability of findings and introduce selection bias. It may also signal loss of access to care, potentially hindering recovery and rehabilitation efforts. Proactively anticipating missingness can help mitigate this issue by identifying individuals at risk of missing values, enabling services to take timely and targeted actions in response. However, the predictability of incomplete PRO data remains underexplored. Methods: We used data from the French multicentric psychosocial rehabilitation database REHABase, focusing our analysis on patients with schizophrenia. We developed an ensemble machine learning model to predict missing data occurrence across six PROs, incorporating treatment center affiliation, sociodemographic features and clinical predictors. To ensure interpretability, we applied the concept of Shapley values to quantify individual predictor contributions to missing data patterns. Results: Our sample comprised N = 2,363 participants. Averaged areas under the receiving operating curve (AUC) measured on the holdout testing observations ranged from 0.73 to 0.78 across the six PRO scales, demonstrating good predictive performance of our ensemble model. Treatment center affiliation emerged as a critical predictor of missing data. The ten most influential patient-level predictors were: being a disabled worker beneficiary; educational attainment; housing status; duration of illness; antipsychotic medication; origin of the referrer; number of suicide attempts; addictions comorbidity; having a forensic history; and sex. We also identified directional contributions distinguishing positive (increased likelihood of missing values) and negative effects (decreased likelihood). Conclusions: To our knowledge, this work represents the first predictive analytics framework for the occurrence of missing PRO data in psychosocial rehabilitation. Our ensemble algorithm holds dual potential: improving data collection strategies and informing targeted interventions to enhance patient engagement and retention. By proactively identifying at-risk individuals and refining study designs, our model could also indirectly support better functional recovery outcomes for schizophrenia patients.

Background: There is a lack of knowledge regarding the actionable key predictive factors of homelessness in psychiatric populations. Therefore, we used a machine learning model to explore the REHABase database (for rehabilitation database—n = 3416), which is a cohort of users referred to French psychosocial rehabilitation centers in France. Methods: First, we analyzed whether the different risk factors previously associated with homelessness in mental health were also significant risk factors in the REHABase. In the second step, we used unbiased classification and regression trees to determine the key predictors of homelessness. Post hoc analyses were performed to examine the importance of the predictors and to explore the impact of cognitive factors among the participants. Results: First, risk factors that were previously found to be associated with homelessness were also significant risk factors in the REHABase. Among all the variables studied with a machine learning approach, the most robust variable in terms of predictive value was the nature of the psychotropic medication (sex/sex relative mean predictor importance: 22.8, σ = 3.4). Post hoc analyses revealed that first-generation antipsychotics (15.61%; p < 0.05 FDR corrected), loxapine (16.57%; p < 0.05 FWER corrected) and hypnotics (17.56%; p < 0.05 FWER corrected) were significantly associated with homelessness. Antidepressant medication was associated with a protective effect against housing deprivation (9.21%; p < 0.05 FWER corrected). Conclusions: Psychotropic medication was found to be an important predictor of homelessness in our REHABase cohort, particularly loxapine and hypnotics. On the other hand, the putative protective effect of antidepressants confirms the need for systematic screening of depression and anxiety in the homeless population.

Schizophrenia patients are at high risk of developing severe COVID-19 outcomes but recent evidence suggests that they are under-vaccinated. This study explored the role of potential attitudinal barriers by comparing schizophrenia patients with participants from the general population regarding COVID-19 vaccination rates, general attitudes towards vaccines, and willingness to take a COVID-19 vaccine. We conducted a cross-sectional study between April 2021 and October 2021. A total of 100 people with schizophrenia and 72 nonclinical controls were recruited. In our study, individuals with schizophrenia were under-vaccinated, despite similar general attitudes towards vaccination and higher willingness to be vaccinated against COVID-19 compared to nonclinical participants. In patients, negative attitudes toward vaccines were related to higher levels of negative psychotic symptoms and higher levels of paranoid ideation. As a whole, participants with more negative attitudes towards vaccines were less likely to be vaccinated against COVID-19 and had lower levels of trust in institutions. Vaccine hesitancy does not appear to be a major barrier for COVID-19 vaccine uptake amongst people with schizophrenia. This study suggests that disparities in COVID-19 vaccination rates in schizophrenia do not seem related to attitudinal but rather structural barriers.

Background and Aims: Few studies have been conducted on the long-term evolution of gambling disorder (GD). The aim of this study was to identify factors that could predict GD relapse.Methods: Data were part of a dataset from a large 5-year cohort of gamblers who were assessed at inclusion and each year thereafter. Participants were recruited from an outpatient addiction treatment center, from various gambling places and through the press. For this specific study, inclusion criteria included (i) transitioning from GD to recovery at a follow-up time and (ii) undergoing at least one follow-up visit afterwards. Participants were evaluated using a structured clinical interview and self-report questionnaires assessing sociodemographic, gambling and clinical characteristics. “Relapse” was defined as the presence of GD (according to the DSM-5) at the N+1^th visit following the absence of GD at the N^th visit. A Markov model-based approach was employed to examine predictive factors associated with relapse at a subsequent follow-up visit.Results: The sample consisted of 87 participants, aged 47.6 years (sd = 12.6), who were predominantly male (65%). Among the participants, 49 remained in recovery, whereas 38 relapsed. Participants who reported not having experienced at least one month of abstinence and those with a low level of self-directedness at the previous follow-up visit were more likely to relapse.Conclusions: Our findings suggest the existence of factors that are predictive of relapse in individuals with GD who had previously achieved recovery. These results can inspire the development of measures to promote long-term recovery.

Background: Up to half of the patients with schizophrenia attempt suicide during their lifetime. Better insight is associated with better functioning but also with increased suicidality. The direction of the relationship between insight and suicidality is not clear, hence we aimed to provide new elements using structural equation modeling. Methods: Insight, quality of life (QoL), depression, and suicidality were measured at baseline and at 12 months in individuals with schizophrenia spectrum disorders. The relationships between these variables were investigated by latent difference score models, controlling for chlorpromazine doses, positive and negative symptoms, and general psychopathology. Results: 738 patients were included, and 370 completed the study. Baseline levels of insight predicted changes in suicidality, whereas baseline levels of suicidality did not predict changes in insight, suggesting that better insight underlies suicidality and predicts its worsening. Our results suggest this temporal sequence: better insight → worse QoL → increased depression → increased suicidality, while insight also affects the three variables in parallel. Conclusion: Better insight predicts a worsening of QoL, depression and suicidality. These findings contribute to our global understanding of the longitudinal influence of insight on suicidality. We advocate that insight-targeted interventions should not be proposed without the monitoring of depression and suicide prevention.

There is abundant literature on how to distinguish problem gambling (PG) from social gambling, but there are very few studies of the long-term evolution of gambling practice. As a consequence, the correlates of key state changes in the gambling trajectory are still unknown. The objective of the JEU cohort study is to identify the determinants of key state changes in the gambling practice, such as the emergence of a gambling problem, natural recovery from a gambling problem, resolution of a gambling problem with intermediate care intervention, relapses or care recourse. The present study was designed to overcome the limitations of previous cohort study on PG. Indeed, this longitudinal case–control cohort is the first which plans to recruit enough participants from different initial gambling severity levels to observe these rare changes. In particular, we plan to recruit three groups of gamblers: non-problem gamblers, problem gamblers without treatment and problem gamblers seeking treatment. Recruitment takes place in various gambling places, through the press and in care centers. Cohort participants are gamblers of both sexes who reported gambling on at least one occasion in the previous year and who were aged between 18 and 65. They were assessed through a structured clinical interview and self-assessment questionnaires at baseline and then once a year for five years. Data collection comprises sociodemographic characteristics, gambling habits (including gambling trajectory), the PG section of the DSM-IV, the South Oaks Gambling Screen, the Gambling Attitudes and Beliefs Survey – 23, the Mini International Neuropsychiatric Interview, the Wender-Utah Rating Scale-Child, the Adult ADHD Self-report Scale, somatic comorbidities (especially current treatment and Parkinson disease) and the Temperament and Character Inventory – 125. The JEU cohort study is the first study which proposes to identify the predictive factors of key state changes in gambling practice. This is the first case–control cohort on gambling which mixes non-problem gamblers, problem gamblers without treatment and problem gamblers seeking treatment in almost equal proportions. This work may help providing a fresh perspective on the etiology of pathological gambling, which may provide support for future research, care and preventive actions. (ClinicalTrials.gov): NCT01207674 .