In an attempt to evaluate the effectiveness ofrnneonatal BeG vaccination policy, a tuberculin survey wasrnconducted among different age groups of children inrnLideta awraja, Addis ababa who had received BeGrnvaccination shortly after birth . The survey consisted ofrnassessing the immunization records and nutritional statusrnand conducting clinical histories and physicalrnexaminations. Thereafter, each study subject's site ofrnBeG vaccination on the right shoulder was assessed andrnthe size of the scar graded. Tuberculin PPD tests werernalso performed simultaneously, and tuberculin reaction ___ _rnwere read within 96- 120 hours. A standard datarncollection form was used to record address, age, sex,rnbody weight, BeG Scar and tuberculin response .rnIn addition, 60 children were revaccinated at thernage of 10 years (within the study period) and tuberculinrntests were performed 2 months after the secondrnvaccination .rnFurthermore , 70 bacter iologically provenrntuberculosis patients were also tuberculin tested at thernbeginning of the study, in order to check the potency ofrntuberculin PPD test and to estimate the diagnostic va luernof the test.rnA total of 895 children were studied between Octoberrn1992 and February 1993. Of these 563 (63 %) had a definiternBeG scar while the rest 331 (37%) were without a Scar.rnThe sex distribution in each group was similar. Thernpercentage of tuberculin non reactors in different agerngroups was 39.5% at 2 months, 34 . 5% at 18 months, 39 .1%rnat 5-7 years, and 46% at 9-11 years. The number ofrntuberculin non reactors initially showed a slight declinernfrom age 2 months and then an increase from age 18 monthsrnonward; the overall pattern of in durationstatistically significant different between the agerngroups (P< 0.001). Some 55% of the children without arnBCG scar showed no response to the tuberculin test, whilern67% of those with a scar showed a positive tuberculinrnresponse. A direct correlation was observed betweenrntuberculin reactivity and the size of BCG scar (P6 mm) wasrnobserved in 11% of c hildren despite the absence of thernBCG scar. Children without the BCG scar have 2.53 timesrnthe risk of a negative tuberculinrnthose with a BCG scar (P 4 rom) . The tuberculosis patients showedrntuberculin reactions according to an unimodalrndistributi on. Of these 12% showed 10 -15rom induration.rnA reaction of 10- 15 rom were also observed in 3.3% ofrnchildren without a scar,ll% of children with a large scarrnsize( >5rom) and in 10% Of revaccinated children with outrntuberculosis infection . This revealed that the lack ofrnspecificity of the test .rnOverall, this finding emphasized the poor tuberculinrnresponse in infants at the age of 2 months compared to 18rnmonths after vaccination at birth and the waning effectrnof tuberculin response at the age five years compared torn18 months. It indicates that the BCG induced a llergyrntended to disappear in this age group and the need forrnrevaccination. Thus we suggest a further study inrndeciding the age of first vaccination and revaccinationrnat the age of 5 years without tuberculin test .rnIn estimating coverage of vacc ination, the presencernof BCG scar used as an indicator may overestimate byrnalmost a factor of two. Hence for accurate estimates wernrecommend determining the rate of tuberculin reactors inrnpopulations who have a BCG s car. size was In assessing the diagnostic value of tuberculin PPDrntest, it was found less specific to determine the problemrnof tuberculosis in the community at a cut-off point ofrn10mm. Therefore for a better estimate, a cut- off pointrnof 15mm and above should be considered.