Adjusted risk estimates of confounding factors determining heart disease among elderly: LASI data analysis
Author(s): CP Prakasam
Abstract: In this paper attempt is made to estimate adjusted relative risk, risk difference by collecting data from the Longitudinal ageing study in India (LASI). Data were collected from 65380 elderly population above 45 years of age (? 45) who have reported “ever diagnosed Chronic heart disease (HEART)”, “Ever diagnosed diabetes (DI)”, “Ever diagnosed hypertension (HT)”, SEX and “Current age (AGE)” of elderly. Hyper tension, Age of elderly and sex factors considered as confounding variables and odds ratio, risk ratio and adjusted risk difference has been estimated by applying Mantel-Haenszel and inverse-variance stratified methods for categorical data and generalized linear regression with a log link and binomial distribution for continuous data set. To overcome residual error due to confounding, age of elderly considered as continuous variable and generalized regression model has been applied and STATA 16/1 software used for analysis. Odds ratio was around 3.3 and the magnitude of confounding due to age and hypertension factor influencing chronic heart disease through diabetes was 85% and 1.04 (4%) times at higher risk than those who are not reported diabetes with confound factors viz age group and hypertension. Risk estimated by using generalized linear model by considering age as continuous variable shows that 6.5 percent increase in chronic heart disease with diabetes among elderly who had hypertension as confounding variable. The estimated risk rates are low with Age of elderly and sex as confounding factors influencing heart disease reveals that those factors may not be potential confounders.
CP Prakasam. Adjusted risk estimates of confounding factors determining heart disease among elderly: LASI data analysis. Int J Stat Appl Math 2025;10(3):42-46. DOI: 10.22271/maths.2025.v10.i3a.2006
