RESUMO
A cross sectional study was conducted to determine the variables which have an individual predictive power on nonutilization of immunisation services using logistic regression model. Children between 12 and 23 months of age were assessed and enrolled using a pretested interview schedule. Statistical analysis was done using descriptive statistics, univariate analysis using Chi Square tests and multivariate logistic regression analysis. Of the 200 children studied 56% were fully immunised. Only illiterate mothers (p < 0.05) and poor access (> 3 km) to health facility (p < 0.001) were found to have an independent predictive power on the nonutilization of immunisation services. Sex of the child, parity levels, socioeconomic status did not independently influence the utilisation pattern. It is possible to identify these still existing small pockets of nonusers of immunisation services through the predictive variables and target them through special efforts.