Socio-Economic Factors of Misconception About HIV/Aids Among Ever-Married Women in Punjab: A Comparison of Non- Spatial And Spatial Hierarchical Bayesian Poisson Model

摘要

Combating HIV/Aids is the sixth goal of Millennium Development Goals and has become an increasing health concern in Pakistan. Pakistan is among the 11 countries in the Asia-Pacific which houses a majority of people infected with HIV. The only way to battle HIV is to provide accurate knowledge about HIV / Aids transmission among general public especially women of child bearing age i.e. 15-49 years. Prevention of HIV and misconception about its transmission are associated to each other. Therefore, the present study aimed to identify the spatial distribution of the three types of misconception factors of HIV transmission (i.e. transmitted by mosquito bite, supernatural means and sharing food with HIV positive person). This study also provides the core socio-economic factors to stop the misconception about HIV/ Aids transmission and helped in reducing its epidemic in Pakistan. Spatial and Non-Spatial Bayesian Hierarchical model were applied to the data and results from them revealed that the Conditional Autoregressive Bayesian Hierarchical Models (Spatial Model) were more appropriate. The results showed that Conditional Autoregressive Bayesian Hierarchical models at level 2 are best fit to the data. The results identified that average size of household, mean age of ever-married women in each tehsil, ownership of assets, percentage of TB patients, adolescent birth rate and intrauterine device used as contraceptive measures were the negatively associated with the response variables. While, percentage of ever-married women who attended matric level school, percentage of household received benefits from government social protection schemes, percentage of Hepatitis patients, access to mass media by TV, Female sterilization and condom used as a contraceptive measure were the positively significant factors for misconception indicators about HIV/ Aids Transmission.