Author

Rama Hiola

Subject

- Kesehatan

Abstract

Observational studies are the basis of epidemiological research to draw the conclusions of the effects or a response treatment. In general, a randomized trial is required in order to meet the assumption of independence to minimize the bias effects. However in an observational study, particularly in medical field, randomization not able to implement because conduces in doubtful treatment effects estimation. Propensity score is the conditional probability to get certain treatments involving the observed covariates. This method is used to reduce bias in the estimation of the impact of treatment on observational data for their confounding factors. If treatment is binary, then the logistic regression model is one estimated of propensity score because of easiness in terms of estimation and interpretation. In the analysis of observational studies, propensity score stratification (PSS) has proven to be one of methods to adjust the unbalanced covariate for the purposes of causal inference. The data used in this study is the medical records of patients DM in X hospital about the factors that influence the type of diabetes mellitus. In this study PSS used in diabetes mellitus cases to reduce bias due to confounding factors, so that can be known the factors affect the type of diabetes mellitus with obesity as confounding factors. The results of PSS analysis is known that the variables directly influence the type of DM are obesity, age, gender and variable does not directly influence the type of DM are genetic variable, sport activities and dietary habit of patients DM.

Publisher

Faculty of Mathematics and Science Yogyakarta State University

Contributor

-

Publish

2016

Material Type

PROSIDING

Right

-