The determination ot t~e prevel ence of tuberculosis infection withinrnage groups and the relation of lnfection to factors unde r study, i,e.,rnprevious BeG vaccination, raw milk consumption, and intimate exposure torncattle were evaluated in a cross-sectlcnal community-based tuberculin skinrntest survey, and case-control analysis of the relation of infection to thernfactors under study 1n Mendeyo Awraja, Bale region (south-east Ethiopia),rnin chi ldren 5 months to 15 years ot age. A ~urvey cons i st i ng of anrninterviev/, examination of BeG scar on the right shoulder, and tuberculinrnskin testing and reading was conductea on 1892 children, 1dentified byrnsystematic household survey from ten randomly selected peasant Vl I lages inrnMendeyo Awraja.rnOut of the total 1892 children surveyed 1002 (53%) were males andrn890 (47%) were females. 1028 (54.3%) had a BeG scar and 864 (45.7%) hadrnno BeG scar; the prevalence of tuberculosis infection in these groups wasrn11 . 8% and 15.3% respect i ve 1 y . The overall .prevalence of tubercu los isrninfection was 13.4%. The mean age of the surveyed children was 6 years.rnPrevalence of infection was lew in infants and children under 5 years ofrnage and increases with age, being high in 12 to 15 years old. There was arnsignificant association between infection and previous non-BeGrnvaccination, and intimate exposure to cattle, as determined by chi squarernstatistics (p < . 05 ) . There was no slgnificant association betweenrninfection and ra'~ CO';J'S milk consumption (p > .05).The avera!(e alm":1-l risK of illt'ection as calcuLlted from the r esults of' thernstudy in non-vaccinated ch~laren "'as 2.73%. InCidence of smear-positivernpulmonary tuberculosis cases Has calculated to be about 337 cases perrn290) 000 population (the set tIed awraja) s population). The study hasrnenabled us to determine the annual risk of infection and incidence ofrnsmear-positive pulmonary tuberculosis cases in the community (awraja).