Prevalence of autism spectrum disorders has increased over recent years, however, little is known about the identification and management of autism spectrum disorder in Africa. This report summarizes a workshop on autism spectrum disorder in Africa under the auspices of the International Child Neurology Association and the African Child Neurology Association through guided presentations and working group reports, focusing on identification, diagnosis, management, and community support. A total of 47 delegates participated from 14 African countries. Although there was a huge variability in services across the countries represented, numbers of specialists assessing and managing autism spectrum disorder was small relative to populations served. Strategies were proposed to improve identification, diagnosis, management and support delivery for individuals with autism spectrum disorder across Africa in these culturally diverse, low-resource settings. Emphasis on raising public awareness through community engagement and improving access to information and training in autism spectrum disorder. Special considerations for the cultural, linguistic, and socioeconomic factors within Africa are discussed.

Substantial evidence indicates that parents of autistic individuals often display milder forms of autistic traits referred to as the broader autism phenotype (BAP). To determine if discrete endophenotypes of autism can be identified, we reviewed the literature to assess the evidence of behavioral, cognitive, and psychiatric profiles of the BAP. A systematic review was conducted using EMBASE, MEDLINE, PsycINFO, PsycEXTRA, and Global Health. Sixty papers met our inclusion criteria and results are discussed according to the proportion of studies that yield significant deficits per domain. The behavioral, cognitive, and psychiatric endophenotypes in parents of autistic probands are still not clarified; however, evidence suggests mild social/communication deficits, rigid/aloof personality traits, and pragmatic language difficulties as the most useful sociobehavioral candidate endophenotype traits. The existence of deficits in the cognitive domain does suggest familial vulnerability for autism. Furthermore, increased depressed mood and anxiety can also be useful markers; however, findings should be interpreted with caution because of the small number of studies in such heterogeneously broad domains and several methodological limitations.

IntroductionThe Paediatric Early Warning Trigger (PEWT) Tool is used at our hospital to aid early identifi cation of clinical deterioration and promote effective escalation of care. Our hospital is a teaching hospital, with inpatient general, respiratory, haematology, surgery, neurosurgery, hepatology and gastroenterology beds and a paediatric critical care unit (CCU). Method The use of PEWT was prospectively audited during July 2010 for all Paediatric admissions, excluding critical care or NICU admissions. All observation charts included a PEWT scoring system that assessed 5 physiological parameters. The data were collected twice daily using a standard proforma including demographics, documentation of triggering, frequency of triggering and admission to CCU. ResultsData was collected on 331 children (age [range] 6.3 [0.01-18.6] years), 62 (19%) children triggered. Seven children (all of whom triggered) were admitted to CCU. The children admitted to CCU were more likely to have triggered in more than one time period (median [range] number of triggers 3 [1-7] v 0 [0-16], p<0.0001). Overall, identifi cation of triggering on PEWT detected CCU admission with a sensitivity of 100% and a specifi city of 83% with a positive predictive value of 0.13 and a negative predictive value of 1.00. Area under receiver operated curve (AUC) analysis showed that triggering PEWT was highly predictive of CCU admission (AUC [95% CI] 0.91 [0.87-0.96]) as was the number of triggering epochs (AUC [95% CI] 0.92 [0.88-0.97]).Conclusion Triggering PEWT is strongly associated with CCU admission. However, given its low positive predictive value, PEWT has poor effi cacy for establishing the need for critical care based on triggering scores alone. Therefore, CCU

years), 62 (19%) children triggered. Seven children (all of whom triggered) were admitted to CCU. These seven were more likely to have triggered in more than one time epoch (median [range] number of triggers 3 [1-7] v 0 [0-16], p< 0.0001). Triggering PEWT detected CCU admission with a sensitivity of 100% and a specificity of 11% with a positive predictive value of 100% and a negative predictive value of 83%. Area under receiver operated curve (AUC) analysis showed that triggering PEWT was highly predictive of CCU admission (AUC 0.91) as was the number of triggering epochs (AUC 0.92). Conclusion: Triggering PEWT is strongly associated with CCU admission. However, given its low specificity, PEWT is poorly predictive for establishing the need for critical care based on triggering scores alone. Therefore, CCU admission may not be the best measure for evaluating the effectiveness of PEWT, instead effectiveness of interventions would be more appropriate but more difficult to measure.