T30. TIPPING POINTS – PREDICTING TRANSITIONS TO PSYCHOSIS IN AT-RISK YOUNG PEOPLE

Abstract Background In traditional psychosis prediction research, the assumption is that a single “snapshot” of clinical disturbance at time point one (i.e. baseline) can reliably predict the future emergence of psychosis over time (i.e., follow-up). This is a linear, static approach to psychosis prediction. However, the field increasingly recognizes that mental health behaves as a non-linear, dynamic system, common to other complex structures such as ecosystems, financial markets or the climate. Methods Increasing evidence points toward the existence of generic “tipping points” in these complex dynamic systems. A tipping point refers to a critical threshold whereby a system shifts from one state into another. Evidence suggests there are universal early warning signals/resilience indicators (such as a phenomenon called ‘critical slowing down’), which predict close proximity to a critical tipping point. Results There is growing evidence for the presence of these early warning signals in psychopathology. This presentation will introduce theoretical concepts of tipping points and resilience indicators in the context of transitioning from at-risk mental state to frank psychosis. Discussion This new framework may represent a paradigm shift from static prediction approaches to dynamic, individualized models of psychosis prediction and may inform the development of new clinical identification tools and early and individualized interventions to prevent such transitions.


T29. ELECTRORETINOGRAPHIC RESPONSE IN YOUTHS AT GENETIC RISK OF SCHIZOPHRENIA AND BIPOLAR DISORDER AND IN NORMAL CONTROLS: TRANSVERSAL AND LONGITUDINAL DIFFERENCES AND IMPLICATIONS FOR THE RISK TRAJECTORY
Anne-Marie Gagné 1 , Thomas Paccalet* ,1 , Valérie Jomphe 1 , Daphné Lussier 1 , Michel Maziade 1 1 Centre de Recherche CIUSSS-CN Background: Visual defects have been widely reported in major psychosis. This includes altered eye tracking, retinal thinning and electrophysiological anomalies. 1 One of the most replicated alterations is decreased electroretinographic (ERG) responses that are observed in both bipolar disorder and schizophrenia. Our previous study showed a diminished rod b-wave amplitude in a small sample of children born to an affected parent. 2 The fact that an anomaly found in patients would also be observed in children at genetic risk suggests a neurodevelopmental origin and may represent a vulnerability marker. Little data exists on the stability of ERG measures in childhood and adolescence. We wanted to evaluate rod and cone ERG response in larger samples of young offspring of an affected parent (HR) and age and gender balanced controls. By comparing a subsample of 33 offspring to controls, we were able to evaluate the stability and change of ERG over time.
Methods: ERGs of 71 offspring (mean age of 19 y.o.) and 224 healthy controls (mean age of 20 y.o.) was recorded. From this sample, 33 HR and 26 healthy controls had ERG recordings at 2 different moments (mean interval of 4 years). We then compared the amplitudes obtained at Time 1 and Time 2 in order to assess whether the ERG amplitudes remained stable or varied over time. Results: Congruent with our 2010 report, this larger HR sample showed a reduced rod b-wave amplitude (p<.05). Probably due to higher statistical power, two other differences were found: an increased cone b-wave latency (p<.05) as well as a diminished mixed rod/cone ERG amplitude (p <.05). None of the ERG amplitudes of the healthy controls changed over time. In contrast, 12 out of 33 HR participant showed a variation of more than one standard deviation (either increase or decrease) on the rod b-wave amplitude which was significantly more frequent than in healthy controls (2/26; p<.05). Change in offspring occurred in both directions: some of them had an increased ERG amplitude response that was sufficient to end up in the confidence interval of the controls whereas others experienced a decreased of their rod amplitude over time. Discussion: These young high-risk offspring displayed three ERG anomalies that we have already reported in adult patients. 2 Our finding bolstered the evidence that ERG anomalies observed in patients may have neurodevelopmental or childhood roots. We observed only little variation in the ERG of the healthy controls over time in that early age range and this appears concordant with existing literature. Of particular interest is the finding that rod b-wave amplitudes were diminished in patients and in offspring. The offspring also showed increased variability over time in comparison to controls. Future studies will seek to understand the relationship between the transversal or longitudinal patterns of rod b-wave amplitudes, as an indicator of risk, and the risk endophenotypes previously reported in the children born to an affected parent.

Interdisciplinary Center Psychopathology and Emotion regulation
Background: In traditional psychosis prediction research, the assumption is that a single "snapshot" of clinical disturbance at time point one (i.e. baseline) can reliably predict the future emergence of psychosis over time (i.e., follow-up). This is a linear, static approach to psychosis prediction. However, the field increasingly recognizes that mental health behaves as a non-linear, dynamic system, common to other complex structures such as ecosystems, financial markets or the climate. Methods: Increasing evidence points toward the existence of generic "tipping points" in these complex dynamic systems. A tipping point refers to a critical threshold whereby a system shifts from one state into another. Evidence suggests there are universal early warning signals/resilience indicators (such as a phenomenon called 'critical slowing down'), which predict close proximity to a critical tipping point. Results: There is growing evidence for the presence of these early warning signals in psychopathology. This presentation will introduce theoretical concepts of tipping points and resilience indicators in the context of transitioning from at-risk mental state to frank psychosis. Background: Early onset psychosis (onset prior to age 18) may be associated with poorer long-term outcomes than when onset takes place in adulthood. Studies so far prospectively following youth with early onset psychosis have suggested that only 20% of patients experienced a "good" outcome, and that nearly 50% experience a "poor" outcome. However, the major body of evidence so far has focused on cohorts recruited over 20 years ago, while a recent review (Clemmensen 2012) has noted that outcomes may be more favourable in more recent studies. We set out to update the evidence by prospectively assessing youth with a first episode of early onset psychosis after a 10 year follow-up period with clinical and functional measures. Methods: Patients were recruited from a child and adolescent psychiatry unit at a university hospital in Barcelona, Spain between 2003 and 2008. Inclusion criteria were: age 7 to 17 years and presence of positive psychotic symptoms of less than 6 months duration. Exclusion criteria: presence of a concomitant Axis I disorder that could account for the psychotic symptoms (e.g., substance abuse, autistic spectrum disorders, post-traumatic stress disorder, or acute stress disorder); learning disability according to DSM-IV criteria; neurological disorders. Occasional substance use was not an exclusion criterion. A sample of age matched healthy controls was recruited from the same geographical area. Controls had the same exclusion criteria in addition to no family history of psychotic disorders in first degree relatives. All participants were assessed with clinical (K-SADS, PANSS, GAF, CGI) measures by a psychologist and psychiatrist with experience with child and adolescent population at baseline and 10 year follow-up. Results: Sixty-nine patients and thirty-one controls were assessed at baseline; 36 patients (52%) and 24 controls (77%) were re-assessed at ten year follow-up. There were no differences in age (M=26.4 SD=1.4 vs. M=26.0 SD=2.0; t=.84; p=.41) or gender (47% vs. 46% females, Chi square=.011; p=.92). We were unable to locate 9 patients (13%) and 4 controls (12.5%). Two patients were deceased at follow-up (one committed suicide; one from metastasic cancer), one patient was in prison, and the rest declined participation. Patients who attended follow-up had trend-level baseline poorer functioning (17.85 vs. 22.4, p=.065), and significantly greater baseline clinical severity 6.44 vs 6.0, p=.017) than those who did not. Diagnoses at baseline were as follows: n=16 (23.2%) schizophrenia; n=20 (29%) affective disorders with psychotic symptoms; n=29 (42%) psychosis not otherwise specified. At follow-up: n=18 (50%) schizophrenia, n=11 (30.6%) affective disorders, n=1 (2.8%) personality disorder; n=1 (2.8%) eating disorder and n=5 (13.9%) had no psychiatric diagnosis. Twenty patients (58.8%) had been hospitalized during the follow-up period and thirty (88.2%) were currently receiving at least one antipsychotic drug. Seven patients (19.4%) were categorised as "poor outcome" (GAF < 50), 18 patients (50%) as "moderate" outcome (GAF 51-70) and 9 patients (26.5%) as "good outcome" (GAF >70). Seven patients (19.4%) were in full time employment and five (13.9%) were in full time education. The rest were either unemployed or working part time or in an assisted setting. Discussion: Early onset psychosis is associated with poor long term outcomes in a portion of patients, although current functional outcomes in these patients may be comparatively better in relation to data from historical reports (Clemmensen 2012). Factors related to improved healthcare services, such as reduction in duration of untreated psychosis and new treatment modalities may potentially underlie these differences.

T32. USING PSYCHOSOCIAL INTERVENTION TO ENHANCE KNOWLEDGE OF ATTENUATED PSYCHOSIS SYMPTOMS AND HELP SEEKING BEHAVIORS AMONG AFRICAN AMERICAN YOUNG ADULTS
Huijun Li* ,1 , Tanisha Pelham 2 1 Florida A&M University; 2 Florida State Hospital Background: The lack in knowledge of mental illnesses is of primary concern with regard to help-seeking and treatment outcomes, especially when faced with chronic and severe illnesses such as psychotic disorders. Where mental health knowledge lacks, so does the ability to recognize the signs and symptoms, risk factors, and causes of mental disorders; as well as the appropriate routes of care for these illnesses. Psychotic disorders and attenuated/subclinical psychotic symptoms are often the target of stigma due to the distinctive symptoms, disruptive behavior and perceived dangerousness of both. Furthermore, the social stigma and discrimination historically faced by African Americans in the United States magnifies the disparity in treatment outcomes among this population. The enrollment of minority college students has increased from 15 percent to 33 percent over the past three decades; cases of students with mental illnesses have also increased. It is becoming more important to explore psychosocial intervention strategies geared to promote knowledge of attenuated psychotic symptoms and helpseeking behavior among African Americans young adults. Methods: The sample consists of 177 students from a Historically Black College and University (HBCU) in the Southeast region of the United States. The participants ranged in age from 18-25. A within group testretest design was used to conduct the study. The participants received a pretest, participated in a psychosocial training on attenuated psychosis syndrome, and a posttest.

Results:
The results suggest that the training was effective in enhancing the participants' knowledge of early warning signs of psychosis and improving their help-seeking behavior. However, stigma unexpectedly increased after the training. Discussion: Discussion: Enhancing mental health literacy has implications for influencing the effects of stigma and discrimination. Colleges and universities are optimal settings for improving mental health literacy because of the high-risk age groups served at these institutions. Mental health literacy is an important life skill that should be taught before the need arises. In order to increase the likelihood that African American college students seek the appropriate help for mental health problems and understand the effects of stigma on help seeking behaviors, more cultural specific interventions are necessary among this population. Interventions should include strategies to cope with stigma and discrimination in order to reduce the effects of both. Future research in this area should also consider how one's ethnic identity correlates with stigma and help seeking behavior.