Association between community knowledge and stigmatising attitudes towards tuberculosis across five provinces in Zimbabwe

A Research Report Submitted to the Faculty of Health Sciences, University of the Witwatersrand in partial fulfilment of the requirements for the Degree of Masters in Epidemiology in the field Epidemiology and Biostatistics. School of Public Health Johannesburg November 2017.Background: People with tuberculosis (TB) often get stigmatised in the communities they live. High stigma contributes to delays in seeking care, poor adherence, and adverse treatment outcomes. Most public health interventions that address stigma focus only on increasing community knowledge and awareness on TB. There is however limited local evidence to support this practice. This study was carried out to determine the association between community knowledge and stigmatising attitudes towards TB in five provinces in Zimbabwe. I also wanted to determine the factors associated with good knowledge of TB and stigma towards it. Methods: I conducted a cross-sectional study using secondary data collected in 2016 from a Knowledge Attitudes and Practices (KAP) survey. The primary study was carried out in five provinces in Zimbabwe by The International Union against Tuberculosis and Lung Disease (The Union). A total of 634 adults from randomly selected households were interviewed using a pre-validated structured questionnaire. The participants’ knowledge was determined based on their total score on simple questions around TB. Logistic regression was used to identify factors associated with community knowledge of TB. An ordinal logistic regression was fitted to determine the association between knowledge of TB and stigmatising attitudes. Also, confirmatory factor analysis (CFA) using generalised structural equation model (GSEM) was done to demonstrate the factors directly and indirectly associated with stigmatising attitudes. viii Results: More than half of my participants had sound knowledge of TB. Participants who had attained a tertiary level of education (AOR 6.86; 95% CI 1.82 – 25.9) and those who knew of someone who has/had TB (AOR 2.67; 95% CI 1.71 – 4.17) were more likely to have good knowledge of TB. The majority of my participants (54%) had some form of stigma towards TB. Surprisingly, there was no association between community knowledge of TB and stigmatising attitudes towards it (AOR 1.03; 95% CI 0.75 – 1.43). However, participants from other smaller religious groups (AOR 2.01; 95% CI 1.03 – 3.91) had higher levels of stigma. The likelihood of having higher levels of stigma decreased with advancing age (AOR 0.99; 95% CI 0.97 – 1.00). Religious groups (traditional churches and apostolic sects), age and socioeconomic status were all directly and indirectly associated with stigmatising attitudes. Their total effect was to increase the likelihood of having higher levels of stigmatising attitudes. Conclusion: Most individuals in the five provinces are aware of TB and have received information on TB however high levels of stigma exist. There is no association between community knowledge and stigma towards TB. There is need to take into consideration the demographic, religious and socioeconomic factors in the design and implementation of public health interventions aimed at improving community knowledge and reducing stigma on TB Keywords: Tuberculosis, stigma, “community knowledge”, generalised structural equation model (GSEM), principal component analysis (PCA)LG201