ABSTRACT
Background: Many Asian countries have persistently high infant mortality because of various biological and social determinants. In particular, infant mortality has increased in Asia in recent decades. The geographic distribution of people having health problems and its relationship to potential risk factors can be of great value for planning cost effective interventions
Objectives: To determine and map the spatial nature of infant mortality in Punjab, at district level and model fit spatial autocorrelation
Materials and Methods: This paper identified and mapped high risk clusters of infant mortality by mapping the infant mortality rate, as well as examines the impact of range of determinants on infant mortality by regressing 15 socio-economic covariates on dependent variable i.e. infant mortality rate by using Multiple Regression model and Conditional Autoregressive Model to quantify spatial risk of infant death
Results: The results showed that the Conditional Autoregressive model yield more reliable results as compared to the multiple regression mode] having lower value of Akaike Information Criterion i.e. 257.8 as compared to multiple regression model i.e. 283.9. The analysis showed that household size, literacy of women, houses having place of hand wash, women having comprehensive knowledge of HIV transmission, households getting benefits from government schemes of social protection and the ownership of assets etc. had low infant mortality rate while adolescent birth rate, underweight live births, unwilling pregnancy, early marriage, infant suspected pneumonia in the last two weeks, people diagnosed with Hepatitis during last one year had high infant mortality rate
Conclusion: Spatial analysis was useful in accessing the determinants of infant mortality rate