Attention Deficit / Hyperactivity Disorder (ADHD) is a highly prevalent neurodevelopmental condition characterized by inattention, motor hyperactivity and impulsivity. ADHD cognitive and behavioral presentation is characterized by a high heterogeneity (APA, 2013). Indeed, a complex diagnostic process, that considers several validated tools, is, to date, necessary.

The main aim is to develop supervised machine learning (ML) algorithms that could be used to support the diagnostic process for ADHD, by identifying the most relevant features in discriminating between the presence or absence of the ADHD diagnosis in children.

We analyzed data from 342 children (Mean age: 8y 8m ± 1y; 61 F) referred for possible ADHD symptomatology. Assessments were performed by an expert clinician and through questionnaires: Social Responsiveness Scale (SRS), Child Behavior Checklist (CBCL), Conners Rating Scale for Parents (CPRS) and for Teachers (CTRS). Data were analyzed using a decision tree classifier and random forest algorithms.

The decision tree model performed an accuracy of 0.71. The random forest model that was identified as the best tested, performed an accuracy of 0.77 (Figure 1) and it identified as most informative parent- and teacher-rated DSM-oriented ADHD symptoms (Figure 2).

Figure 1: Random forest confusion matrix and statistics.

Figure 2: Ranking of variables importance.

A random forest classifier could represent an effective algorithm to support the identification of ADHD children and to simplify the diagnostic process as an initial step. The use of supervised machine learning algorithms could be useful in helping the diagnostic process, highlighting the importance of a personalized medicine approach.

No significant relationships.

Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder characterized by lack of self-regulation and deficits in organizing behaviors in response to emotional stimuli. Methylphenidate (MPH) is one of the most effective psychostimulant drugs for ADHD, however, a possible predictive utility of brain hemodynamic data related to MPH administration and its relation to clinical symptomatology is still not clear. To address these questions, we used Near Infrared Spectroscopy (NIRS) technology, a non-invasive optical technique that allows to investigate the effect of psychopharmacological treatment on cortical hemodynamics.

Twenty children with ADHD underwent a three-waves study and 25 healthy controls were recruited at W1. At W2 children with ADHD received first MPH administration and at W3 they reached the titration phase. At each phase children performed - during NIRS recording - an emotional continuous performance task with visual stimuli of different emotional content. Clinical data were also collected at W1 and W3. We investigated the relationship among the difference between NIRS activation at W2 and W1 (Delta1) and W3 and W2 (Delta2), for each subject, task condition and brain region. Lastly, we investigated correlations between the Delta1 and clinical symptomatology indexes at W1 and between Delta2 and clinical data at W3.

Our study results suggest that hemodynamic changes in right prefrontal region probably induced by first MPH administration could predict hemodynamic changes related to MPH titration phase. These biological indexes could be associated to clinical evidences related not only to core ADHD symptoms but also to affective correlates.

No significant relationships.

Epigenetics hypothesizes a crucial link between postnatal risk factors, individual response to stress, DNA methylation and psychiatric symptomatology changes during life.

We analyzed methylation within two gene exons: NR3C1 and SLC6A4, which are involved in responses to environmental stressors. We investigated the relationship between methylation, postnatal risk factors and psychopathology assessed by Child Behavior Checklist (CBCL) in our help-seeking sample evaluated in infancy (W1), preadolescence (W2) and adult life (W3).

Postnatal risk factors data were collected at W1 in 205 clinical subjects (156 M, 49 F; age=9,13±1,95). The CBCL scores were collected at W1 and W2 (W2 age=14,52±2,12). Data regarding methylation were collected at W2. At W3 we are also collecting clinical scores. A Spearman correlation coefficient was calculated between methylation percentage and clinical data at W2. The externalizing and internalizing trajectories were evaluated through repeated measure ANOVA with postnatal risk factors (presence/absence) as between-groups factor.

Significant associations were found between methylation and internalizing and total clinical scores (Table 1). The rm-ANOVA results showed a significant interaction between the CBCL internalizing score and presence/absence of postnatal risk, with higher internalizing problems in subjects that were exposed to postnatal risk factors. This effect was significant at W2 but not at W1 (Figure 1).

Psychopathological symptoms trajectories could depend on epigenetics and early environmental risk factors. Further analyses will address a Linear Discriminant Analysis to proceed to a machine learning oriented approach.

No significant relationships.

Our 15-years follow-up ReMIND project aims to re-assess an epidemiological and a clinical sample of adults (Wave 3), who were assessed in preadolescence (Wave 1) and adolescence (Wave 2), to evaluate symptoms trajectories and their relationship with genetic/epigenetic data, environmental risk factors and neuroimaging measures.

Here, we depict preliminary results regarding the epidemiological sample.

We assessed internalizing and externalizing symptoms in 40 italian subjects (25 F) from general population at three waves (W1 mean age: 12±0,82; W2 mean age: 17±0,88, W3 mean age: 28±1), through the Child Behavior Checklist (W1 and W2) or the Adult Self Report (W3), and perinatal risk factors through a socio-anamnestic questionnaire, by a new online platform (MedicalBit). We analyzed symptoms trajectories and their relation with perinatal risk factors through a repeated measures multivariate analysis of variance (rm-MANOVA).

rm-MANOVA results show that high number of perinatal risks was significantly associated with higher internalizing symptomatology in preadolescence but not in adolescence and adult life. The mean difference was 8 T-points. The same trend is evident in adolescence but not in adult age (Graph 1). Perinatal risk factors did not have a significant effect on externalizing symptoms at any time point, despite a non-significant trend is evident (Graph 2).

Our preliminary results suggest a trend of increased internalizing symptoms from childhood to adulthood and a significant role of perinatal risk factors in pre-adolescence. Further investigations are necessary to better understand symptoms trajectories and the role of biological and environmental factors.

No significant relationships.

Literature underlines that the Edinburgh Postnatal Depression Scale (EPDS) was the most common measure to assess pregnancy and postpartum depression worldwide and suggests that the rate of false positive cases is high. Furthermore, the EPDS measures, but not distinguish between depression and anxiety. The aim of this study was to value the ability of the Mood Spectrum Self Report Last Month (MOOD SR LM) to detect the rate of false positive obtained with EPDS

We recruited 81 pregnant women at the third month of pregnancy. 32 women were affected by Major Depression (MD) assessed by Structured Clinical Interview for DSM IV Axis I Diagnosis and 49 women had EPDS score ≥ 13 but didn’t reach the diagnosis of MD by SCID I(False posivite group). Women were administered with the Postpartum Depression Predictors Inventory-Revised (PDPI-R), MOOD SR-LM, Work and Social Adjustment Scale(WSAS) at the third month of pregnancy.

The two groups wasn't different concerning the average scores using PDPI and EPDS. The WSAS average scores were higher in the depressed women group than in the false positive group. Tree factors of MOOD SR LR (“Depressive Mood”, “Psychomotor Retardation and “Drug/Illness related Depression”) had higher average scores in the depressed women group than in the false positive group.

To our results, MOOD SR LM seem to be able to decrease the false positive cases. Further studies concerning this use of MOOD SR LM are necessary.

Perinatal depression is a particolar challenge to clinicians, and its prevalence estimates are difficult to compare across studies. Furthermore,there are no studies that systematically assessed the incidence of perinatal depression. The aim of this study is to estimate prevalence, incidence, recurrence and new onset of DSM IV minor and major depression (mMD) in an unselected population of pregnant women.

1066 pregnant women were recruited at third month of pregnancy (T0), and minor/major depression (mMD) was assessed by the Structured Clinical Interview for DSM IV disorders (SCID I). The SCID I was administered at baseline evaluation (T0), the Edimburgh Postnatal Depression Scale (EPDS) was administered at third, 6th (T1), 8th (T2) month of pregnancy, and the SCID I Mood module was administered to confirme an eventual DSM-IV minor or major depression diagnosis when the EPDS score was≥13.

The pregnancy period prevalence of mMD was 12,4%.The point prevalence of mMD decreased from 8,6% at the 3rd month of pregnancy to 1,7% at the 8th month of pregnancy.The cumulative incidence of mMD was 2,2%. The weighted incidence of new onsets during pregnancy was 1,6%. The weighted percentage of recurrences during pregnancy was 3,7%.

The decline in the point prevalence during the second and third trimester of pregnancy found in our study may be attributed to psychological counselling and/or pharmacological treatment.Further studies about new onsets of depression during pregnancy are highly important in order to improve clinical prediction of risk in any individual woman.

Second generation antipsycotics (SGAs) represent an essential treatment for Psychotic Disorders. However SGAs cause weight gain and metabolic disorders with an unknown mechanism.

The aim of this study is to investigate whether weight gain may be caused by a reduction in resting energy expenditure (REE).

We recruited 34 pazients who received a clinical diagnosis of Bipolar Disorder I, Schizoaffective Disorder or Schizophrenia. Patients were in antipsychotic monotherapy (clozapine, olanzapine or aripiprazole). We measured REE by indirect calorimetry and we calculated theorical REE utilizing LARN and Harris-Benedict (H-B) equation.

The difference between the average value of REE measured in patients (1295Kcal/die) and the average value of REE estimated by LARN and H-B equation (1719Kcal/die and 1741Kcal/die) was respectively 423Kcal/die and 445Kcal/die: both the differences resulted statistically significative (p< 0.01). This reduction of REE results in a weight gain of 1,7 kg in a month.

It is necessary to inform patients about the side effects of SGAs. Lifestyle counseling and pharmacological assistance, or switching to another antipsychotic are possible strategies.

Depression during pregnancy is associated to physical symptoms that can impair the functioning of women; furthermore some of the depression somatic symptoms (i.e., sleep disturbance, fatigue, weight change and appetite) are also features of pregnancy. The overlap of symptomatology can interfere with the identification and the diagnosis of the mood episode. Aim of this study is to compare the the depressive phenomenology and the presence of Axis I comorbidity between pregnant and non pregnant depressed women.

We diagnosed Major Depression (MD) using the Structured Clinical Interview for Axis I Diagnosis DSM IV (SCID I) in 32 pregnant women at third month of pregnancy and 87 non pregnant women and we compared the depressive phenomenology in the two groups. Then we administered the Mood Spectrum Self Repost Last Month (MOOD SR-LM) in the two group in order to study the mood spectrum symptomatology.

Pregnant depressed women have higher psychomotor retardation, higher levels of concentration and lower agitation than non pregnant depressed women.

The severity of depression symptoms was similar in the two depressed groups.

Our results agree with current litterature about the presence of psychomotor retardation in depressed pregnant women.

The higher level of concentration in pregnant women could be explained by the high comorbidity with Generalized Anxiety Disorder (GAD). In the pregnant depressed women the Obsessive-Compulsive Disorder (OCD) and Panic Disorder (PD) comorbidity are more rappresentated.

Cognitive impairment in patients with eating disorders was reported by the majority of studies addressing this issue. However, heterogeneous patterns of cognitive dysfunctions were observed and, in a minority of studies, no impairment was found. The present study was aimed to define the pattern of neurocognitive impairment in a large sample of bulimia nervosa (BN) patients and to demonstrate that neuroendocrine, personality and clinical characteristics influence neurocognitive performance in BN.

Attention, executive control, conditional and incidental learning were evaluated in 83 untreated female patients with BN and 77 healthy controls. Cortisol and 17β-estradiol plasma levels were assessed. Cloninger's Temperament and Character Inventory-Revised (TCI-R), the Eating Disorder Inventory-2 and the Montgomery-Asberg Depression Rating Scale were administered.

No impairment of cognitive performance was found in subjects with BN vs. healthy controls. The higher the cortisol level and “Self-directedness” scores the better the performance on conditional learning, while 17β-estradiol levels showed an opposite pattern of association; “Reward dependence” scores were associated to a worse performance on incidental learning; depressive symptomatology negatively influenced the performance on the WCST.

No cognitive impairment was found in untreated patients with BN in the explored cognitive domains. An influence of neuroendocrine, personality and clinical variables on neurocognitive functioning was found, which might explain discrepancies in literature findings.

To examine the predictors of aggressive behaviours occurring before acute hospitalisation.

We analysed 350 acute admissions to a psychiatric ward during a 12-month period. The diagnoses were formulated according to the DSM IV axis I and II criteria. Aggressive behaviours occurring in the week before admission were retrospectively assessed using the modified overt aggression scale. The patients' clinical and sociodemographic variables, concurrent drug or alcohol abuse, and admission status were recorded at the time of admission.

Aggressive and violent behaviours were highly prevalent, respectively, in 45% and 33% of the cases. Violence before admission was independently associated with drug abuse, involuntary admission status, and severe psychopathology. A diagnosis of a psychotic disorder did not increase the risk of aggression or violence, compared to the other psychiatric diagnoses. Personality disorders were significantly more associated to aggressive behaviours than psychotic disorders.

The diagnosis of psychotic disorder is a poor predictor of aggression in a sample of psychiatric patients. Other clinical and non-clinical variables are associated to aggression before hospitalisation: they include drug abuse, involuntary admission status, general severity of symptoms, and diagnosis of personality disorder.

To investigate the predictive role of any specific (DSM-IV) Anxiety Disorders (AD) on depressive symptoms and Major or Minor Depressive Disorder (MDD, mDD) during early postpartum period.

Women (at the 12th-15th gestational week, N=1066) were recruited in the framework of the Program ‘Perinatal Depression - Research & Screening Unit (PND-ReScU)’. Depressive symptoms were assessed by the Edinburgh Postnatal Depression Scale (EPDS), and Axis-I disorders (AD, MDD, mDD) were diagnosed with the Structured Clinical Interview for Axis-I Disorders (SCID-I).

Any current AD at baseline (3rd month of pregnancy) was detected in 231 (21.7%). Having at least one current AD, was associated with a greater likelihood of having MDD or mDD during the early postpartum period, even after the adjustment for the confounding factor of having a lifetime history of MDD (RR=3.86 95%CI 1.58-9.42).

In particular, women affected by Obsessive Compulsive Disorder (N=17; 1.6%) or Panic Disorder (N=43; 4%) had at higher risk to develop depressive symptoms (EPDS≥13) during the postpartum period than women without these AD (RR=6.9 and 6.7 respectively). As for the risk of developing PPD, the strongest association was found for women with Panic Disorder (RR=7.6 95% CI 2.62-22.0).

AD are associated with a greater likelihood to develop depressive symptoms and MDD or mDD during the early postpartum period. Women with current PD have the strongest risk to develop both MDD or mDD and depressive symptoms during early postpartum period compared to other anxiety disorders.

Bipolar disorder and substance use disorder have well documented genetic component. Among Axis I disorder bipolar subjects have highest comorbidity for alcohol and substance use disorder, ranging from 17 to 60%. Some Authors find a familial association between bipolar disorder and substance use disorder, suggesting the question if these disorders share a common genetic liability.

To investigate familial loading in bipolar disorder (with and without alcohol/substance use disorder) and in patients with substance use disorder.

Sampe (62 patients with bipolar disorder and alcohol/substance use disorder-DD, 23 patients with bipolar disorder-BD, 22 patients with substance use disorder without mood disorder-SUD) was recruited in Psichiatric clinic of Pisa and Dependance Department of Pisa and Bolzano. Instruments: SCID-IV-I/P for Axis I diagnosis and Family History Screen (Weissman, 2000).

Bipolar pts (DD and BD) have higher familial diathesis for manic and depresive episodes respect DUS probands (p≤0.001). Moreover DD show higher familial loading for alcohol/substance use respect BD (p=0.000). Abuser subjects (DD and SUD) show higher familiarity for conduct disorder respect BD (p=0.008)

DD pts show a double familial loading, both for mood disorder (respect SUD), and for alcohol/substance use disorder (respect BD). Probably the familial loading for substance use disorder and mood disorder are distinct and they are both present in dual diagnosis patients.

To investigate the predictive role of any specific (DSM-IV) Anxiety Disorders (AD) on depressive symptoms and Major or Minor Depressive Disorder (MDD, mDD) during early postpartum period.

Women (at the 12th-15th gestational week, N=1066) were recruited in the framework of the Program ‘Perinatal Depression - Research & Screening Unit (PND-ReScU)". Depressive symptoms were assessed by the Edinburgh Postnatal Depression Scale (EPDS), and Axis-I disorders (AD, MDD, mDD) were diagnosed with the Structured Clinical Interview for Axis-I Disorders (SCID-I).

Any current AD at baseline (3rd month of pregnancy) was detected in 231 (21.7%). Having at least one current AD, was associated with a greater likelihood of having MDD or mDD during the early postpartum period, even after the adjustment for the confounding factor of having a lifetime history of MDD (RR=3.86 95%CI 1.58-9.42).

In particular, women affected by Obsessive Compulsive Disorder (N=17; 1.6%) or Panic Disorder (N=43; 4%) had at higher risk to develop depressive symptoms (EPDS≥13) during the postpartum period than women without these AD (RR=6.9 and 6.7 respectively). As for the risk of developing PPD, the strongest association was found for women with Panic Disorder (RR=7.6 95% CI 2.62-22.0).

AD are associated with a greater likelihood to develop depressive symptoms and MDD or mDD during the early postpartum period. Women with current PD have the strongest risk to develop both MDD or mDD and depressive symptoms during early postpartum period compared to other anxiety disorders.

Mental disorders occurring in pregnancy need attention and specifically targeted treatments; if untreated, they may have severe consequences, for the woman, for her family, and mostly for the newborn. The woman is at higher risk for suicide; she may increase tobacco, alcohol and substances use, and may frequently develop postpartum depression (Halbreich, 2004).

We aimed to describe the prevalence of Axis I disorders (DSM IV-TR criteria) at the 3rd month of pregnancy in a large non-clinical sample of women.

Women between the 12th and the 15th gestational week were enrolled. A total of 1066 subjects (49.9%) of those eligible (N=2138) signed an informed consent and completed the baseline interview. Standard demographic information were collected and Axis I diagnoses were performed using the SCID-I (First et al, 1995).

Mean age was 32.3 years (±3.9); 280 women (26.3%) had a current Axis I Disorder. Mood and anxiety disorders were the most frequent current diagnoses: Major Depressive Episode (N=34;3.2%); Minor Depression (N=45;4.3%); Specific Phobia (N=114; 10.7%); Panic Disorder (N=43; 4%); Social Phobia (N=41; 3.8%); Obsessive-Compulsive Disorder (N=17; 1.6%); Anxiety Disorder NOS (N=29; 2.7%); Generalized Anxiety Disorder (N=20; 1.9%). Current comorbidity between depressive and anxiety disorders was present in 38 women (3.6%).

The sample size and the administration of the SCID, that provides more complete and reliable information about lifetime and current psychiatric history, than instruments used in previous similar studies (Andersson et al., 2003; Spitzer et al., 2000), represent major strengths of the study.

After the development of the Postpartum Depression Predictors Inventory-Revised (PDPI-R) only one study was conducted to determine the predictive validity of the Prenatal and Full Versions of the instrument. However this study did not succeed in identifying the cut-off for the Full Version.

We aimed to determine the predictive validity of the PDPI-R as a screening instrument for post-partum depression (PPD).

Women completed the PDPI-R at the 3rd month of pregnancy and at the 1st month after childbirth. PPD symptoms were assessed using the Edinburgh Postnatal Depression Scale (EPDS) at multiple time points during pregnancy and during the post-partum. When the EPDS score was ≥13, a Structured Clinical Interview for DSM-IV Disorders was conducted to determine whether criteria for depression were met.

The Prenatal and Full Versions of the PDPI-R predicted accurately 80.3% and 88.2% of PPD. The Prenatal PDPI-R yielded a sensitivity of .72 and a specificity of .74 at a cut-off score of 4.5, while the Full version yielded a sensitivity of .83 and a specificity Conclusions

The PDPI-R is a useful and valid screener for PPD.

To assess the dosing patterns for risperidone long-acting injectable (RLAI) in patients with schizophrenia, participating in the 6-month, open-label Switch to Risperidone Microspheres (StoRMi) trial.

Treatment was initiated at RLAI 25 mg intramuscularly every 2 weeks, although higher (starting) doses were permitted if clinically necessary. Efficacy was evaluated using the Positive and Negative Syndrome Scale (PANSS), Clinical Global Impression (CGI) and Global assessment of functioning (GAF). Treatment-emergent adverse events (AEs) were monitored.

A total of 1,849 patients were included. The mode dose was 25 mg for 52.9% of patients, the remainder evenly distributed among 37.5 and 50 mg doses. At baseline, patients treated with lower RLAI doses were more likely to be female, have shorter disease duration, milder symptoms, and be using less polypharmacy. The strongest predictors that a patient would remain on 25 mg RLAI were baseline PANSS hallucinatory behaviour item (OR = 0.78), baseline CGI (OR = 0.69), gender (OR = 1.56) and country (P< 0.001 for all). Efficacy measures improved for all dosage groups, with the greatest improvement in patients treated with lower doses. AEs were more frequent in patients treated with 50 mg RLAI (68% vs. 57% with lower doses, P<0.0001). Most AEs were mild to moderate in severity.

In this large, European sample, most patients were treated with 25 mg RLAI. Patients treated with lower doses tended to have milder baseline symptoms. Dosing patterns varied among different countries. RLAI was effective and well tolerated over the full range of allowed doses.