Socio-demographic and behavioural factors associated with prominent misconceptions of HIV/AIDS transmission among ever married women in Sri Lanka: An application of modelling correlated binary outcomes using a bridge distribution function

Human Immunodeficiency Virus (HIV) case reporting usually represents only the trending pattern of HIV infected people in Sri Lanka. Having a decent level of knowledge about HIV largely contributes to preventing HIV prevalence in the country. People with HIV are not hazardous to the public who live together with them and with whom they have ordinary, non-sexual contact. However, suffering from HIV in a culture like Sri Lanka generates an extreme level of stigmatizing by people living with HIV and many psychological and social effects. This study aims to identify the factors that are associated with possessing knowledge, among ever married women in the reproductive age in Sri Lanka, regarding two prominent misbeliefs on HIV/AIDS transmission. This will be useful to identify various socio-economic, geographic subgroups who are lacking knowledge on HIV/AIDS spread. Secondary data from 2016 Sri Lanka Demographic and Health Survey was used in the analysis. Joint modelling was considered since it was found that two outcome variables are highly associated and hence the ever-married woman who believes one misconception is more likely to believe the second misconception too. To capture the association between two outcomes, we incorporated a shared random effect and assumed the bridge distribution for the random effect. Respondent’s province, highest level of education, access to mass media, religion, age and wealth index were found to have a significant effect on prominent misconceptions on HIV/AIDS transmission.