ANALYSING FACTORS WHICH INFLUENCE TREATMENT OUTCOMES BY MULTIVARIABLE LINEAR REGRESSION
Abstract
Multivariable linear regression is a procedure of building mathematical model of straight line with several independent variables which are added one to another and to a constant to give value of dependent variable, i.e. of outcome. Dependent variable has to be continuous and numeric, while independent variables may be of both numeric and categorical type. Multivariable linear regression is very useful method for testing influence of multiple independent variables on a treatment outcome (dependent variable). With this method it is possible not only to estimate whether certain independent variable may influence the outcome significantly, but also to quantify this influence and compare strength of influence between the variables. Multivariable linear regression is especially useful method for analyzing data of observational clinical studies, particularly “cross-sectional” ones.
References
Marill KA. Advanced statistics: linear regression, part I: simple linear regression. Academic emergency medicine. 2004;11(1):87-93.
12.4 - Detecting Multicollinearity Using Variance Infla-tion Factors | STAT 501 [Интернет]. [цитирано 06. Април 2018.]. Available at: https://onlinecourses.science.psu.edu/stat501/node/347.
Marill KA. Advanced statistics: linear regression, part II: multiple linear regression. Academic emergency medicine. 2004; 11(1): 94-102.
Schneider A, Hommel G, Blettner M. Linear regression analysis: part 14 of a series on evaluation of scientific publications. Deutsches Ärzteblatt International. 2010; 107(44): 776-82.
Chan YH. Biostatistics 201: Linear Regression Analysis. Singapore Med J 2004; 45(2): 55-61.
Kaya Uyanik G, Guler N. A study on multiple linear regression analysis. Procedia - Social and Behavioral Sciences 2013; 106: 234 – 240.
Dallal GE. How to Read the Output From Multiple Linear Regression Analyses [Интернет]. [цитирано 05. Април 2018.]. Available at: http://www.jerrydallal.com/lhsp/regout.htm.