Tuberculosis in the era of infection with the human immunodeficiency virus: assessment and comparison of community knowledge of both infections in rural Uganda

UNLABELLED BACKGROUND In Uganda, despite a significant public health burden of tuberculosis (TB) in the context of high human immunodeficiency virus (HIV) prevalence, little is known about community knowledge of TB. The purpose of this study was to assess and compare knowledge about TB and HIV in the general population of western Uganda and to examine common knowledge gaps and misconceptions. METHODS We implemented a multi-stage survey design to randomly survey 360 participants from one district in western Uganda. Weighted summary knowledge scores for TB and HIV were calculated and multiple linear regression (with knowledge score as the dependant variable) was used to determine significant predictors. Six focus group discussions were conducted to supplement survey findings. RESULTS Mean (SD) HIV knowledge score was 58 (12) and TB knowledge score was 33 (15), both scores out of 100. The TB knowledge score was statistically significantly (p < 0.001) lower. Multivariate regression models included age, sex, marital status, education, residence, and having a friend with HIV/TB as independent variables. TB knowledge was predicted by rural residence (coefficient = -6.27, 95% CI: -11.7 to -0.8), and age ≥45 years (coefficient = 7.45, 95% CI: 0.3-14.6). HIV knowledge was only predicted by higher education (coefficient = 0.94, 95%CI: 0.3-1.6). Focus group participants mentioned various beliefs in the aetiology of TB including sharing cups, alcohol consumption, smoking, air pollution, and HIV. Some respondents believed that TB was not curable. CONCLUSION TB knowledge is low and many misconceptions about TB exist: these should be targeted through health education programs. Both TB and HIV-infection knowledge gaps could be better addressed through an integrated health education program on both infections, whereby TB program managers include HIV information and vice versa.