Prevalence of Tuberculosis in Nigeria

: Studies on the prevalence and risk of tuberculosis (TB) among diagnosed human immunodeficiency virus (HIV)-infected patients in sub-Saharan Africa are alarming especially Nigeria ranking second in Africa and seventh in the world. In other to ascertain the relationship between new smear positive cure and new smear died, new smear complete, failed and defaulted, and variation also to establish if it’s a good fit. In other to get our coefficient of determination at 5% level of significance, on the regression line is 0.963 explaining a 96.3% variation in the patients who died and 0.741 explaining 74.1% variation in patients that where cured. This shows that the models have an unexplained error as 3.7% and 25.9% respectively. There’s no first order linear auto-correlation in the multiple linear regression data for Durbin Watson for patient that died. In regressions, this implies an under estimated level of statistical significance for patients cured. determination, variance inflation factor (VIF) which mirrors the interpretation of multiple determination, indicates that multicollinearity in new smear positive complete, and failed is not enough to worry about but for new smear positive defaulted, its value is greater than 4 for both patients cured and patients that died. F-test explains a significant variance of patients that where cured and those that died within the study period. The coefficient of determination ((cid:1844) (cid:2870) ) indicates a 96.3% and 74.1% variation in both patients that where cured and those that died within the study period. Durbin Watson shows a zero-autocorrelation in the residuals for patients cured and a non-zero autocorrelation for patients that died