percent of participants indicated they enjoyed using the monitor tool. Monitor metrics were assessed for repeated assessment stability by calculating an interclass correlation coefficient (ICC). The most reliable measure produced an ICC of 0.90, indicating high test-retest reliability. Additional metrics from distinct components of the monitor’s tasks showed similar high levels of reliability (ICC > 0.70). For assessment, monitor metrics were analyzed for associated signals that were related to diagnostic group. The most sensitive metric was Reaction Time Standard Deviation (RTSD). An ANOVA showed a main effect of the diagnostic group of scores on RTSD [F(3,55) ¼ 11.08, P < 0.00001]. In post hoc analyses, significant differences between the ASD and all other groups were found (P < 0.01). In addition, single and multivariate models were examined for association with baseline TOVA Attention Perfor- mance Index (API) scores. A three-factor multivariate model was found to offer the most explanatory power with the fewest number of factors. The three- factor model was strongly associated with baseline TOVA API [F(4,19) ¼ 9.615, P ¼ 0.0002]. Conclusions: This pilot study provides initial support for the feasibility of Akili Interactive’s novel video game-like repeat-assessment monitor. In addition, the findings suggest that it is highly reliable, is sensitive to attention dysfunction, and may have diagnostic specificity for specific disorders such as ASD, potentially filling a gap in practice for cognitive assessment and monitoring tools that are tolerable, accessible, and affordable. RI, COG, SAC http://dx.doi.org/10.1016/j.jaac.2017.09.190 3.43 THE RELATION OF PARENT-CHILD INTERACTION THERAPY (PCIT) IN WELL-BEING OF YOUNG CHILDREN WITH ATTENTION-DEFICIT/HYPERACTIVITY DISORDER (ADHD) AND OPPOSITIONAL DEFIANT DISORDER Naser Ahmadi, MD, PhD, University of California, Los Angeles, drnaserahmadi@gmail.com; Shahzad Chaudhry, MFT, University of California, Los Angeles Kern Medical, SChaudhry@co.kern.ca.us; Garth Olango, MD, University of California, Los Angeles Kern Medical, olangog@ kernmedctr.com; Mohammed Molla, MD, University of California, Los Angeles Kern Medical, shahriarmolla@ gmail.com Objectives: Previous studies have demonstrated effectiveness of Parent-Child Interaction Therapy (PCIT) in improving children’s disruptive behavioral problems. This study investigates the effect of PCIT on the well-being of young children with ADHD and ODD. Methods: Eleven consecutive children (ages 5–8 years, 7 males) who scored  120 on the Eyberg Child Behavior Inventory (ECBI) received PCIT. The child and families were assessed at baseline and at four months after beginning treatment. Parenting skills were measured using the Dyadic Parent-Child Interaction Coding System (DPICS), and child behavior problems were measured using the ECBI and the Child Behavior Checklist (CBCL). Well-being was measured using multidimensional PERMA model (positive emotions, engagement, positive relationships, meaning, and accomplishment). Well- being was defined as low (PERMA score  3), average (PERMA score 4–5), and high (PERMA score  6). Results: At follow up, after PCIT intervention, a significant improvement in children’s behavior (ECBI baseline: 147  22 vs. follow-up: 80  15, P ¼ 0.001) and parents’ do and don’t skills, measured by observer rating on DPICS (d ¼ 2.9, d ¼ 1.6, respectively, P ¼ 0.01), were noted. The total PERMA score, as well as scores in each dimension, were significantly improved at follow-up in both children and parents compared with baseline (P ¼ 0.01). The most robust improvement was noted in positive relationships, engagements, and accom- plishments (P < 0.05). Conclusions: The current findings reveal that PCIT is associated with an in- crease in well-being of both children and parents, in addition to significant reduction of children’s disruptive behavior. This highlights the importance of simultaneous intervention to address vulnerable symptoms, as well as enhancing well-being in youth with ADHD and ODD. ADHD, ODD, PAT http://dx.doi.org/10.1016/j.jaac.2017.09.191 3.44 A COMPARISON OF MOTHER-CHILD INTERACTIONS DURING FEEDING AND PLAY IN THE ASSESSMENT OF FEEDING DISORDERS Irene Chatoor, MD, Children’s National Medical Center, ichatoor@cnmc.org; Cristina Sechi, PhD, University of Cagliari, Italy, cristina.sechi79@gmail.com; Loredana Lucarelli, PsyD, University of Cagliari, Italy, llucarelli@unica.it Objectives: The assessment of the quality of the parent-infant/toddler rela- tionship in children with feeding disorders is crucial, not only to observe parent-child interactions in settings where the child and his/her mother are distressed but also in settings where activities can be pleasurable. Most observational tools for assessing parent-child interactions suffered from a lack of empirical research, and the psychometric properties of these tools should be considered more carefully. Therefore, this work explores the factor struc- ture and validity of the Parent-Child Play Scale. It is a complement to the Feeding Scale to evaluate mother-child interactions in children with feeding disturbances and disorders. The Play Scale can be used with children ranging in age from one month to three years and provides ratings of mother-child interactions during 10 minutes of videotaped free-play observations. The scale consists of 32 mother and child interactive behaviors, which are rated by trained observers. Methods: This study examines the factor structure and the ability of the Play Scale to differentiate between healthy children and children with three sub- types of developmental feeding disorders; the relationship between the mother-infant/toddler interactions measured on the Feeding Scale versus the Play Scale. A sample group of 124 mother-child dyads (74 children with feeding disorders; 50 children who were typically developing) participated in the study. Children ranged in age from one to 36 months. Results: The factor analysis extracted four factors, which we labeled: dyadic reciprocity, maternal unresponsiveness to infant/toddler’s cues, dyadic con- flict, and maternal intrusiveness. All four factors together explain 57.7 percent of the total variance. The Play Scale discriminates between children with and without feeding disorders, as well as between children with different subtypes of feeding disorders. Significant correlations (P < 0.05; P < 0.01) between the subscales of the Feeding Scale and the Play Scale were found for mothers and children who had feeding disorders. Conclusions: The Play Scale can be used for research or clinical practice to enhance observational skills, to assess the pervasiveness of mother-infant/ toddler difficulties, and to monitor changes following therapy. R, EC, EA http://dx.doi.org/10.1016/j.jaac.2017.09.192 3.45 CHARACTERISTICS AND POTENTIAL SUBTYPES OF CHILD AND ADOLESCENT PSYCHIATRIC PATIENTS WITH AVOIDANT RESTRICTIVE FOOD INTAKE DISORDER THAT PRESENTED TO A TERTIARY CARE PSYCHIATRIC FACILITY Mili Khandheria, MD, University of Texas Southwestern Medical Center, milik617@gmail.com; Kinjal Patel, DO, University of Texas Southwestern Medical Center, kinjal6891@ gmail.com; Briana Sacco, MD, University of Texas Southwestern Medical Center, briana.sacco@gmail.com; Graham J. Emslie, MD, University of Texas Southwestern Medical Center, graham.emslie@UTsouthwestern.edu Objectives: Few studies have been done on avoidant-restrictive food intake disorder, which is a relatively new diagnosis in DSM-5. This retrospective chart JOURNAL OF THE AMERICAN ACADEMY OF CHILD &ADOLESCENT PSYCHIATRY S218 www.jaacap.org VOLUME 56 NUMBER 10S OCTOBER 2017 NEW RESEARCH POSTERS 3.43 — 3.45