Autistic symptoms represent a frequent feature in schizophrenia spectrum disorders (SSD). However, the prevalence and the cognitive and functional correlates of autistic symptoms in unaffected first-degree relatives of people with SSD remain to be assessed.

A total of 342 unaffected first-degree relatives related to 247 outpatients with schizophrenia were recruited as part of the multicenter study of the Italian Network for Research on Psychoses (NIRP). Autistic features were measured with the PANSS Autism Severity Scale. Three groups of participants, defined on the presence and severity of autistic symptoms, were compared on a wide array of cognitive and functional measures.

Of the total sample, 44.9% presented autistic symptoms; 22.8% showed moderate levels of autistic symptoms, which can be observed in the majority of people with SSD. Participants with higher levels of autistic symptoms showed worse performance on Working Memory (p = 0.014) and Social Cognition (p = 0.025) domains and in the Global Cognition composite score (p = 0.008), as well as worse on functional capacity (p = 0.001), global psychosocial functioning (p < 0.001), real-world interpersonal relationships (p < 0.001), participation in community activities (p = 0.017), and work skills (p = 0.006).

A high prevalence of autistic symptoms was observed in first-degree relatives of people with SSD. Autistic symptoms severity showed a negative correlation with cognitive performance and functional outcomes also in this population and may represent a diagnostic and treatment target of considerable scientific and clinical interest in both patients and their first-degree relatives.

Deficits in social cognition (SC) are significantly related to community functioning in schizophrenia (SZ). Few studies investigated longitudinal changes in SC and its impact on recovery. In the present study, we aimed: (a) to estimate the magnitude and clinical significance of SC change in outpatients with stable SZ who were assessed at baseline and after 4 years, (b) to identify predictors of reliable and clinically significant change (RCSC), and (c) to determine whether changes in SC over 4 years predicted patient recovery at follow-up.

The reliable change index was used to estimate the proportion of true change in SC, not attributable to measurement error. Stepwise multiple logistic regression models were used to identify the predictors of RCSC in a SC domain (The Awareness of Social Inference Test [TASIT]) and the effect of change in TASIT on recovery at follow-up.

In 548 participants, statistically significant improvements were found for the simple and paradoxical sarcasm of TASIT scale, and for the total score of section 2. The reliable change index was 9.8. A cut-off of 45 identified patients showing clinically significant change. Reliable change was achieved by 12.6% and RCSC by 8% of participants. Lower baseline TASIT sect. 2 score predicted reliable improvement on TASIT sect. 2. Improvement in TASIT sect. 2 scores predicted functional recovery, with a 10-point change predicting 40% increase in the probability of recovery.

The RCSC index provides a conservative way to assess the improvement in the ability to grasp sarcasm in SZ, and is associated with recovery.

Resilience is defined as the ability to modify thoughts to cope with stressful events. Patients with schizophrenia (SCZ) having higher resilience (HR) levels show less severe symptoms and better real-life functioning. However, the clinical factors contributing to determine resilience levels in patients remain unclear. Thus, based on psychological, historical, clinical and environmental variables, we built a supervised machine learning algorithm to classify patients with HR or lower resilience (LR).

SCZ from the Italian Network for Research on Psychoses (N = 598 in the Discovery sample, N = 298 in the Validation sample) underwent historical, clinical, psychological, environmental and resilience assessments. A Support Vector Machine algorithm (based on 85 variables extracted from the above-mentioned assessments) was built in the Discovery sample, and replicated in the Validation sample, to classify between HR and LR patients, within a nested, Leave-Site-Out Cross-Validation framework. We then investigated whether algorithm decision scores were associated with the cognitive and clinical characteristics of patients.

The algorithm classified patients as HR or LR with a Balanced Accuracy of 74.5% (p < 0.0001) in the Discovery sample, and 80.2% in the Validation sample. Higher self-esteem, larger social network and use of adaptive coping strategies were the variables most frequently chosen by the algorithm to generate decisions. Correlations between algorithm decision scores, socio-cognitive abilities, and symptom severity were significant (pFDR < 0.05).

We identified an accurate, meaningful and generalizable clinical-psychological signature associated with resilience in SCZ. This study delivers relevant information regarding psychological and clinical factors that non-pharmacological interventions could target in schizophrenia.

Many forms of mental disorders, especially psychotic disorders are characterized also by a worsening of sexual functioning. Sexual dysfunction has been shown to significantly correlate with a longer duration of untreated psychosis and with heavier psychotic symptomatology.

The aim of this study is to validate the Italian version of the Arizona Sexual Experience (ASEX), a very handy and reliable tool to assess sexual dysfunction, in a population of people suffering from psychotic spectrum disorders.

Seventy-three psychiatric patients were recruited and assessed for mental illness and sexual functioning. We administered the Italian version of ASEX, adequately translated by two expert bilinguals. After 15 days we administered once again the test for test-retest reliability.

Validation of ASEX revealed Cronbach’s coefficients >0.70 in both single items as in the total score. In addition, the test-retest reliability revealed Pearson’s coefficients >0.50 in the various domains. Confirmatory factor analysis revealed good fit indexes for the two factors model of ASEX (SRMR=0.54; CFI=0.974; RMSEA=0.135).

This study represents the first validation in the Italian psychiatric context of a very useful specific tool for the sexual assessment in people suffering from mental illness. Our analysis revealed good psychometric characteristics in terms of confirmatory factor analysis, internal consistency, and test-retest reliability.

No significant relationships.

Several studies highlighted how COVID-19-related isolation and quarantine deeply weighed on the mental health of both the general and psychiatric population. There has been limited investigation about self-harm and impulsivity during the COVID-19 pandemic.

The aim of this study is to evaluate how COVID-19-related lockdown affected self-harm rates in an Italian hospital.

Data on 59 patients were retrospectively collected from the Emercency deparment of the Policlinico Tor Vergata, Rome, from March 11 to May 4, 2020 (Italian mass quarantine) and the same periods of 2019 and 2021. Demographics, psychiatric history, substance use/abuse, types of self-harm and admission in psychiatric acute unit (PAU) rates were recorded.

No statistical difference was reported in self-harm rates [9.8%(26/266) in 2019 vs 13.2%(10/76) in 2020 vs 10.7%(23/215) in 2021;p>0.05]. In 2020 subjects were younger (31.9±12.1 vs 39.2±14.4,p=0.22;vs 38.1±14.4;p=0.15) and had higher incidence of psychiatric history [90%(9/10) vs 73.1%(19/26), p=0.42;vs 65.2% (15/23),p=0.29],than 2019 and 2021 respectively. Substance use/abuse rates were significantly lower in 2020 compared to 2019 and 2021 [10%(1/10) vs 53.8%(14/26),p=0.04;vs 60.9% (14/23), p=0.02]. In 2020, subjects committing self-harms were more frequently admitted to PAU compared to 2019 and 2021 [60%(6/10)vs19.2%(5/26),p=0.04; vs 17.4% (4/23), p=0.04).

Consistent with the literature, lockdown-related measures negatively impacted on younger people, with higher rates of self-harm between March and May 2020. This, together with a higher rate of admissions to PAU, should warn the mental health system to target with specific programs to support adolescents and youngers.

No significant relationships.

A few studies have analyzed the impact of COVID-19 pandemic on psychiatric Emergency Department (ED) accesses. The pandemic may indeed have influenced the phase of day accesses for patients with psychiatric disorders.

Aim of this cross-sectional study is to analyze how COVID-19 weighed on psychiatric patients daily accesses over the course of three years.

Data on 219 patients were retrospectively collected from the ED in the Policlinico Tor Vergata, Rome. According to the stage of the day, accesses were divided into 4 groups: between 00:00 and 6:00; between 6:00 a.m. and 12:00 a.m.; between 12:00 a.m. and 18:00 p.m.; between 18:00 p.m. and 00:00 p.m.

Performing a regression analysis, a relation was found between psychiatric symptoms, stage of the day admission and year. In 2019 the admissions seem to be homogeneously distributed, however during 2021 and 2020 the admissions rates have a delayed evening trend.

Despite the low number of accesses considered, the Covid-19 pandemic appears to exert an effect that still lasts in terms of both accesses and worsening or new onset of psychiatric symptoms. Measures taken to prevent the spread of infections may have affected access in the ED of patients in various ways. However, the trend of increasing evening accesses could be related to a saturation of territorial psychiatric services that work mainly until the afternoon. Thus, an enhancement of territorial psychiatric services seems highly necessary to cope with what could be an increase in psychopathology in patients without previous diagnosis.

No significant relationships.

People with severe mental illnesses (SMI) have a mortality rate two times higher compared to the general population, with a decade of years of life lost. In this randomized controlled trial (RCT), we assessed in a sample of people with bipolar disorder, major depressive disorder, and schizophrenia spectrum disorder, the efficacy of an innovative psychosocial group intervention compared to a brief psychoeducational group intervention on patients’ body mass index (BMI), body weight, waist circumference, Framingham and HOMA-IR indexes.

This is a multicentric RCT with blinded outcome assessments carried out in six Italian university centers. After recruitment patients were randomized to receive a 6-month psychosocial intervention to improve patients’ physical health or a brief psychoeducational intervention. All recruited patients were assessed with standardized assessment instruments at baseline and after 6 months. Anthropometric parameters and blood samples have also been collected.

Four-hundred and two patients with a diagnosis of bipolar disorder (43.3%), schizophrenia or other psychotic disorder (29.9%), or major depression (26.9%) were randomly allocated to the experimental (N = 206) or the control group (N = 195). After 6 months, patients from the experimental group reported a significant reduction in BMI (odds ratio [OR]: 1.93, 95% confidence intervals [CI]: 1.31–2.84; p < 0.001), body weight (OR = 4.78, 95% CI: 0.80–28.27, p < 0.05), and waist circumference (OR = 5.43, 95% CI: 1.45–20.30, p < 0.05). Participants with impaired cognitive and psychosocial functioning had a worse response to the intervention.

The experimental group intervention was effective in improving the physical health in SMI patients. Further studies are needed to evaluate the feasibility of this intervention in real-world settings.

International Classification of Diseases 11th Revision (ICD-11) has inserted complex post-traumatic stress disorder (cPTSD) as a clinically distinct disorder, different from PTSD. The diagnosis of cPTSD has the same requirements for the one of PTSD, in addition to disturbances of self-organization (DSO – e.g., disturbances in relationships, affect dysregulation, and negative self-concept).

This study aimed to explore suicidality in PTSD and cPTSD. We examined also the association between clinical dimensions of hopelessness (feelings, loss of motivation, future expectations) and other symptomatologic variables.

The sample, recruited at the Fondazione Policlinico Tor Vergata, Rome, Italy, consisted of 189 subjects, 132 diagnosed with PTSD, and 57 with cPTSD, according to the ICD-11 criteria. Participants underwent the following clinical assessments: Clinician-Administered PTSD Scale (CAPS), Impact of Event Scale-Revised (IES), Beck Depression Inventory (BDI), Symptom Checklist-90-Revised (SCL-90), Dissociative Experience Scale (DES), Beck Hopelessness Scale (BHS).

cPTSD showed significantly higher BHS-total (p = 0.01) and BHS-loss of motivation subscale (p <0.001) scores than PTSD. Besides, cPTSD showed significantly higher scores in all clinical variables except for the IES-intrusive subscale. By controlling for the confounding factor “depression”, suicidality in cPTSD (and in particular the BHS-total) appears to be correlated with IES-total score (p = 0.042) and with DES-Absorption (p = 0.02). Differently, no such correlations are found in PTSD.

Our study shows significant symptomatologic differences between PTSD and cPTSD, including suicidality. Indeed, suicidality in cPTSD appears to be correlated with the “loss of motivation” dimension, which fits well within the ICD-11 criteria of DSO.

No significant relationships.

Cognitive function is impaired in depressive disorders. Among several factors implicated in regulation of the cognitive function, metabolic syndrome has been showed have a pivotal role cognitive functioning in healthy controls. However, the role of metabolic syndrome in regulating the cognitive functioning of subjects affected by depressive disorders is little studied.

To investigate the effect of metabolic syndrome in regulation of cognition in unipolar and bipolar depression.

One-hundred-sixty-five people affected by a depressive disorder (unipolar depression, UP; bipolar depression, BP) were enrolled at the Psychiatric and Clinic Psychology Unit of the University of Rome Tor Vergata, Rome, Italy. A group of healthy controls (HC) matched for agender and age was enrolled. The cognitive functions were evaluated with a computerized tool, THINC-it.

UP and BP had lower performances in THINC-it cognitive domains than HC. Metabolic syndrome is a negative, independent predictor of low performance in the THINC-it cognitive domains of people with depressive disorders.

Our findings confirm that metabolic syndrome has a prominent role in determining the cognitive efficiency in depressive disorders, independently by the presence of a unipolar or bipolar depressive disorder. Metabolic syndrome has to be considered a major factor that should be considered in the treatment strategies of cognitive functioning improvement of people affected by depressive disorders.

No significant relationships.

Amongst different subtypes of Conversion Disorder (CD), DSM-V lists the Psychogenic Non-epileptic seizures (PNES). PNES are defined as episodes that visually resemble epileptic seizures but, etiologically, they are not due to electrical discharges in the brain.

Our study aims to explore the differences between PNES and other CDs. In particular, we studied the suicidality and its correlations with dissociation and alexithymia.

Patients, recruited from the Psychiatry and Clinical Psychology Unit of the Fondazione Policlinico Tor Vergata, Rome, Italy, were diagnosed with PNES (n=22) and CD (n=16) using the DSM-5 criteria. Patients underwent the following clinical assessments: HAM-D, BDI, DES, BHS, TAS, CTQ.

PNES showed significantly higher scores than CD in all assessments, except for BDI-somatic (p=0.39), BHS-feeling (p=0.86), and the presence of childhood trauma. PNES also showed significantly higher suicidality (p = 0.003). By controlling for the confounding factor “depression”, in PNES suicidality (and in particular the BHS-loss of motivation) appears to be correlated with DES-total score (p = 0.008), DES-amnesia (p = 0.002) and DES -derealization-depersonalization (p = 0.003). On the other hand, in CDs, the BHS-total score shows a correlation with the TAS-total score (p = 0.03) and BHS-Feelings with TAS-Externally-Oriented Thinking (p = 0.035), while only the BHS-Loss of motivation appears correlated with DES-Absorption (p = 0.011).

Our study shows significant differences between PNES and CD, in several symptomatologic dimensions, including suicidality. Indeed, in PNES suicidality appears to be related to dissociation, while in CDs it appears mainly to be correlated with alexithymia.

The deficiency of polyunsaturated fatty acids (PUFAs) and an alteration between the ratio of omega-6 and omega-3 PUFAs may contribute to the pathogenesis of depressive disorders.

To investigate the levels of omega-3 and omega-6 in red cell membranes (mPUFAs) and plasma (pPUFAs) of patients with treatment-resistant (TRD) and non-treatment resistant depression (non-TRD).

TRD and non-TRD consisted of 75 patients enrolled at the Psychiatric and Clinic Psychology Unit of the University of Rome Tor Vergata, Rome, Italy, and met the DSM-IV criteria for major depressive disorder (MDD). A group of healthy controls (HC) matched for agender and age was enrolled. All blood samples were performed in conditions of an empty stomach between 07:00 am and 09:00 am. For each subject were obtained 5 ml of whole blood with the use of tubes for plasma with EDTA as an anticoagulant. Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) for omega-3 and arachidonic acid (AA) for omega-6 were measured.

Levels of pPUFAs did not differ between the three groups. The mPUFAs were altered in the MDD. TRD and non-TRD had lower EPA and AA values respect to the HC. DHA in red cell membranes was lower in TRD than non-TRD and HC.

Changes in levels of PUFAs in red cell membranes, but not in plasma, may be an important factor to evaluate the resistance to the pharmacological treatment.

Previous researches highlighted among patients with schizophrenia spectrum disorders (SSD) a significant presence of autistic traits, which seem to influence clinical and functional outcomes. The aim of this study was to further deepen the investigation, evaluating how patients with SSD with or without autistic traits may differ with respect to levels of functioning, self-esteem, resilience, and coping profiles.

As part of the add-on autism spectrum study of the Italian Network for Research on Psychoses, 164 outpatients with schizophrenia (SCZ) were recruited at eight Italian University psychiatric clinics. Subjects were grouped depending on the presence of significant autistic traits according to the Adult Autism Subthreshold Spectrum (AdAS Spectrum) instrument (“AT group” vs “No AT group”). Other instruments employed were: Autism Spectrum Quotient (AQ), Specific Levels of Functioning (SLOF), Self-Esteem Rating scale (SERS), Resilience Scale for Adults (RSA), and brief-COPE.

The “AT group” reported significantly higher scores than the “No AT group” on SLOF activities of community living but significantly lower scores on work skills subscale. The same group scored significantly lower also on SERS total score and RSA perception of the self subscale. Higher scores were reported on COPE self-blame, use of emotional support and humor domains in the AT group. Several correlations were found between specific dimensions of the instruments.

Our findings suggest the presence of specific patterns of functioning, resilience, and coping abilities among SSD patients with autistic traits.

Autism spectrum disorders (ASDs) and schizophrenia spectrum disorders (SSDs), although conceptualized as separate entities, may share some clinical and neurobiological features. ASD symptoms may have a relevant role in determining a more severe clinical presentation of schizophrenic disorder but their relationships with cognitive aspects and functional outcomes of the disease remain to be addressed in large samples of individuals.

To investigate the clinical, cognitive, and functional correlates of ASD symptoms in a large sample of people diagnosed with schizophrenia.

The severity of ASD symptoms was measured with the PANSS Autism Severity Scale (PAUSS) in 921 individuals recruited for the Italian Network for Research on Psychoses multicenter study. Based on the PAUSS scores, three groups of subjects were compared on a wide array of cognitive and functional measures.

Subjects with more severe ASD symptoms showed a poorer performance in the processing speed (p = 0.010), attention (p = 0.011), verbal memory (p = 0.035), and social cognition (p = 0.001) domains, and an overall lower global cognitive composite score (p = 0.010). Subjects with more severe ASD symptoms also showed poorer functional capacity (p = 0.004), real-world interpersonal relationships (p < 0.001), and participation in community-living activities (p < 0.001).

These findings strengthen the notion that ASD symptoms may have a relevant impact on different aspects of the disease, crucial to the life of people with schizophrenia. Prominent ASD symptoms may characterize a specific subpopulation of individuals with SSD.

One in six adolescents suffers from mental health problems. Despite the presence of general information on Italian adolescents' mental health, researches conducted with standardized assessment tools are scarce in the literature. We evaluated the prevalence of self-reported behavioral and emotional problems in a group of Italian adolescents and examined their relation to socio-demographical variables.

This population-based sampling survey was conducted on high school students aged 14–18 from urban areas of Rome and Latina. Participants completed Youth Self-Report (YSR) and a socio-demographic schedule to collect information on age, gender, type of school attended, socio-economic status, urbanicity.

Final sample consisted of 1400 adolescents (38.61% male, mean age 16 years, s.d. 1.42). Prevalence of Internalizing Problems, Externalizing Problems and Total Problems was 29.55%, 18.34% and 24.13%. In our multivariable model, Internalizing Symptoms were not explained by sociodemographic variables while Externalizing Symptoms were explained by Male Gender [OR = 1.53 (1.14–2.06)], older age [OR = 2.06 (1.52–2.79)] and attending a Technical and Professional Institute [OR = 2.15 (1.53–3.02)], with an adjusted R2 = 4.32%. Total Problems were explained by School Type [Technical and Professional Institutes and Art and Humanities v. Grammar and Science School; OR respectively 1.93 (1.40–2.67) and 1.64 (1.08–2.47)], adjusted R2 = 1.94.

The study provides, for the first time, evidence of a great prevalence of self-reported behavioral and emotional problems in a large sample of Italian adolescents, highlighting the role of different socio-demographic variables as risk factors for externalizing behaviors. Our results emphasize the urgent need for implementing prevention programs on mental health in adolescence.

Greater levels of insight may be linked with depressive symptoms among patients with schizophrenia, however, it would be useful to characterize this association at symptom-level, in order to inform research on interventions.

Data on depressive symptoms (Calgary Depression Scale for Schizophrenia) and insight (G12 item from the Positive and Negative Syndrome Scale) were obtained from 921 community-dwelling, clinically-stable individuals with a DSM-IV diagnosis of schizophrenia, recruited in a nationwide multicenter study. Network analysis was used to explore the most relevant connections between insight and depressive symptoms, including potential confounders in the model (neurocognitive and social-cognitive functioning, positive, negative and disorganization symptoms, extrapyramidal symptoms, hostility, internalized stigma, and perceived discrimination). Bayesian network analysis was used to estimate a directed acyclic graph (DAG) while investigating the most likely direction of the putative causal association between insight and depression.

After adjusting for confounders, better levels of insight were associated with greater self-depreciation, pathological guilt, morning depression and suicidal ideation. No difference in global network structure was detected for socioeconomic status, service engagement or illness severity. The DAG confirmed the presence of an association between greater insight and self-depreciation, suggesting the more probable causal direction was from insight to depressive symptoms.

In schizophrenia, better levels of insight may cause self-depreciation and, possibly, other depressive symptoms. Person-centered and narrative psychotherapeutic approaches may be particularly fit to improve patient insight without dampening self-esteem.

Multiple-complex Developmental Disorder (McDD) is a developmental disorder characterized by peculiar clinical features: affective dysregulation (anxiety, panic and aggressivity), impairment of social behaviour and hypersensitivity, impaired cognitive processing. McDD is usually included within the Pervasive Developmental Disorders not otherwise specified (PDD NOS) (Cohen et al. 1986; Towbin et al. 1993). Aripiprazole is a new atypical antipsychotic drug. Efficacy of aripiprazole in children and adolescents is supported by some studies (Findling et al. 2007, Wink at al 2010; Kim et al 2010).

The aim of our study is to describe the efficacy and safety of aripiprazole in 4 children with Multiple-complex Developmental Disorder.

Mean dosage of aripiprazole was 5 mg/day. Response was evaluated by clinical assessment and by Clinical Global Impressions Scale-Severity, Clinical Global Impressions Scale- Improvement, Children's Global Assessment Scale (CGAS) and by Brief Psychiatric Rating Scale. This assessment was administered at baseline, and at weeks 4, 12 and 24.

A significant reduction of social impairments, thought disorders and affective dysregulation was observed. Drug therapy was well tolerated, even though children complained of mild and transient nausea and somnolence.

To date there are only a few reports on the use of aripiprazole in McDD patients. The present case series suggest that aripiprazole may be effective and well tolerated in McDD patients.

To explore tolerability, safety and treatment response of flexible doses of oral paliperidone ER in patients with schizophrenia suffering from an acute episode.

Interim analysis of a 6-week prospective, open-label, international study. Endpoints were the rate of responders defined as a ≥30% improvement in the Positive and Negative Syndrome Scale (PANSS) from baseline to endpoint, the Clinical Global Impression-Severity Scale (CGI-S), weight change and adverse events (AEs).

100 patients were analyzed (51% male, mean age 39.0±11.6 years). 82% of patients completed the study. Most frequent reasons for early discontinuation were subject choice (10%) and lack of efficacy (7%). the mean dose of paliperidone ER was 5.9 mg/day at baseline and 7.9 mg/day at endpoint. an improvement of ≥30% in total PANSS was observed in 68% of patients (95% confidence interval [CI]58%;77%], with a decrease in mean total PANSS scores from 98.2±16.2 at baseline to 71.1±20.3 at endpoint (mean change -27.1±19.9; 95%CI -31.1;-23.2, p< 0.0001) and onset of efficacy as of day 2. the percentage of patients rated as at least markedly ill in CGI-S decreased from 69% to 20.3%. AEs reported in ≥5% were insomnia (14%), tachycardia (10%), akathisia (6%), extrapyramidal disorder (6%), headache (5%) and schizophrenia (5%). Median weight gain was 0.7 kg (95% CI 0.19;1.96) from baseline to endpoint.

This analysis supports data from recent controlled studies that flexibly dosed paliperidone ER is safe, well tolerated and associated with a clinically meaningful treatment response in patients with an acute schizophrenic episode.

Treatment Resistant-Depression (TRD), known as the failure to respond to at least two different adequate trials of antidepressant treatments (ADT) in the current episode, is a relatively frequent clinical condition, associated to a high number of relapses, hospitalizations, and an elevated use of multiple pharmacological treatments. To date, however, the association between clinical variables and non-response in TRD remains unclear.

To identify predictors of non-response in inpatients with Major Depressive Disorder (MDD) and TRD.

To investigate clinical variables as potential predictors of non-response in TRD.

Two hundred fifty-three inpatients with MDD and TRD were divided into two groups: responders and non-responders to drug therapies, according to a decrease of 50% or more of the severity of depression (measured with HAM-D 17 items) at the end of forth week of hospitalization. A general model of Cox regression (with backward stepwise method) was used to identify independent predictors of non-response to treatment.

One hundred fifty-four TRD inpatients were responders and 99 non-responders. Cox regression identified three independent clinical predictors independently associated with the group of non-responders: (1) the presence of 5 or more depressive episodes in the medical history (OR = 2.27); (2) a current comorbid anxiety disorder (OR = 1.85); (3) a history of early life adversities (ELAs) (OR = 1.60).

The findings of this study suggest that the phenomenon of non-pharmacological response in the TRD is associated with different clinical variables, which might act through separate mechanisms in determining the persistence of depressive symptomatology.