Behaviour disorders in urban primary school children in Dhaka, Bangladesh MG Rabbani 1 and MM Hossain 2 * 1 Department of Psychiatry, Sir Salimullah Medical College and Mitford Hospital, Dhaka, Bangladesh; and 2 Department of Community Medicine, Faculty of Medicine and Health Sciences, United Arab Emirates University, PO Box 17666, Al Ain, United Arab Emirates Data on the prevalence and correlates of behaviour disorders in children are scanty or absent in many countries, includingBangladesh.AsampleofprimaryschoolchildreninDhakacity,Bangladesh,wasscreenedduringFebruaryto April 1994 with the Rutter B2 Scale to estimate the prevalences of different types of behaviour disorders and to assess whether and how these prevalences were associated with the children's age and gender, and gender segregation at school. Of the 1288 children in grades 1 to 5 screened, 13.4% had some type of behaviour disorder, with males more than twice as commonly affected as females (20.4 vs 9.9%). Emotional, conduct, and undifferentiated disorders were detected in 3.2, 8.9 and 1.2%, respectively, of the children. All three disorders were more prevalent in males than in females. In multivariate logistic regression analyses, signi®cant positive correlates were male gender (odds ratio (OR):3.1)andhighergrades(reference:grades1±2;ORforgrade3:9.1;ORforgrade5:4.2)forconductdisorder, and male gender (OR:3.1) and higher grade (OR for grade 3:6.2) for all disorders combined. None of the three independent variables examined was signi®cantly associated with emotional disorder. Separate analyses could not be done for undifferentiated disorder because of the small numbers involved. The high prevalence of behaviour disorders in Dhaka city primary school children suggest that a mental health care programme for these children comprising screening and, if indicated, full evaluation and expert care would be worthwhile. Keywords: Bangladesh; behaviour disorder; primary school children; Rutter B2 Scale Introduction During the past three decades, the importance of primary prevention, early detection, and proper management of behaviour disorders in children has been increasingly recognised world-wide. Published estimates of the pre- valenceofbehaviourdisordersinchildrenindifferentparts oftheworldvarywidelyfrom3to30%. 1 However,dataon the prevalence and correlates of behaviour disorders in children are scanty or absent in many countries, including Bangladesh. Community-based random sample surveys would pro- vide the best estimates of the true prevalence of behaviour disorders in children. But full psychiatric evaluation, in the communitysetting,oflargenumbersofchildrenwouldbea very dif®cultÐif not impossibleÐtask anywhere. In contrast, studies of behaviour disorders in children would beeasiertodoinpsychiatricclinics,butsuchstudieswould underestimate the true prevalence by missing those who do not seek care from such clinics. To avoid these problems, school-based studies have been done to estimate the prevalence of behaviour disorders in children by using screening questionnaires completed by the teachers. 2 We did this study of primary school children in Dhaka city, Bangladesh, to estimate the prevalences of different typesofbehaviourdisordersandtoassesswhetherandhow these prevalences were associated with the children's age and gender, and gender segregation at school. The relation betweenbehaviourdisordersandgendersegregationwasof particular interest to us because some Dhaka city schools admitchildrenofonlyonegenderevenattheprimarylevel. Methods Area and population ThisstudywasdoneinDhakacity(estimatedpopulation10 million), the capital of Bangladesh. The target population of this study was all primary school children in the city. Dhaka city schools offering primary education differ by various characteristics including management (public or private), gender segregation (boys' only, girls' only, or coeducational),instructionallanguage(BengaliorEnglish), and courses offered (only primary or both primary and secondary). The duration of primary education in Bangla- desh is 5 years (grades 1 to 5). Subjects Logistic and resource constraints and the absence of an up- to-date listing of all schools in Dhaka city prevented us from adopting a random sampling strategy. We, therefore, selected a convenience sample of 11 schools from different parts of the city taking special care to ensure representa- tiveness according to the schools' characteristics. Our goal was to obtain information about all children enrolled in grades 1 to 5 in each selected school. Thisstudywasapprovedbytheethicalcommitteeofthe Bangladesh Medical Research Council and by the head teachers of each participating school. Data collection Data for this study were collected during February to April 1994. We used the Rutter B2 Scale for completion by teachers to screen the study children for behaviour disorders. This scale, with documented validity and reliability in English and several other languages, has been *Correspondence: Dr M M Hossain, Department of Community Medicine, Faculty of Medicine and Health Sciences, United Arab Emirates University, PO Box 17666, Al Ain, United Arab Emirates. Accepted 8 March 1999 Public Health (1999) 113, 233±236 ß R.I.P.H.H. 1999 http://www.stockton-press.co.uk/ph