Bio-statistics Essay Example | Topics and Well Written Essays - 1000 words

Bio-statistics - Essay Example It can be either additive or subtractive. In case of additive procedure, the significant predictive variables are included in the regression model one after another. In case of subtractive approach all the expected variables are included in the model and then those which do not show statistically significant association with the predicted variable are dropped from the equation. In univariate analyses each variable is analysed separately for its statistics like mean, median, mode, standard deviation, range, skewness etc. Association on the other hand comes by looking at trend between two variables. If values of two variables show a tandem movement, then there exists significant association between the two variables. Though it looks very similar to correlation, it is completely different as in case of association, there may not be any causal relationship between two variables, which is there in case of correlation. What has been stated in the quoted sentence is that univariate analysis resulted in significant association between ‘allogeneic transfusion’ and ‘older age’; ‘allogeneic transfusion’ and ‘female sex’; ‘allogeneic transfusion’ and ‘hip procedure’ and so on. It means that instances of allogeneic transfusion were more in older people, females and so on. But it does not necessarily mean that this relationship was causal as well, it may be or may not be. g) ‘Revision hip’ was associated with the highest probability of transfusion as ‘Risk ratio’ is highest for this procedure. This means that those going for â€Å"Revision hip’ are at the greatest risk for allogeneic transfusion than those going for other procedures. h) The older person (aged 77) is at greater risk of requiring a transfusion after operation than the younger one. This comes from the fact that the multivariate regression model throws up risk ratio of 1.77 for those aged from 70 – 80 as compared to that of those aged below 70.