Page 93 - ISMCON souvenir 2021

ISMSCON - 2021

OS71: Impact of gender on the Lipid Profile of Patients with
Coronary Artery Disease: A Bayesian Analytical Approach

Vivek Verma , Ashwani Kumar Mishra , Anita Verma , Hafiz T.A. Khan , Dilip C. Nath ,
1
4
5
2
3
Rajiv Narang 6
1 Department of Statistics, Assam University, Assam, 788011, India; Department of Neurology, All India Institute
of Medical Sciences (AIIMS), New Delhi. 110029, India
Email: viv_verma456@yahoo.com
2 National Drug Dependence Treatment Centre (NDDTC), All India Institute of Medical
Sciences (AIIMS), New Delhi, 110029, India Email: ashwaniiop@gmail.com
3 Department of Neurology, All India Institute of Medical Sciences( AIIMS), New Delhi.
110029, India Email: anii121514@gmail.com
4 Health Promotion and Public Health, College of Nursing, Midwifery and Healthcare, University of West London,
Paragon House, Boston Manor Road, Brentford TW8 9GB
United Kingdom, Email: hafiz.khan@uwl.ac.uk. Also an associate professorial fellow at the Oxford Institute of
Population Ageing, University of Oxford, UK.
5 Assam University, Silchar, Assam, 788011, India Email: dilipc.nath@gmail.com
6 Department of Cardiology, All India Institute of Medical Sciences (AIIMS), New Delhi.
110029, India Email: r_narang@yahoo.com

Abstract

Background: The variation of the lipid profile by sex in patients with coronary artery disease (CAD) is
well known. However, when the sample size is small, it is difficult to establish this association.

Objectives: The present study uses the Bayesian paradigm to understand the association between
sex and lipid profile in patients with coronary artery disease and compares the results with classical
approaches.
Methods: The study is based on a secondary analysis of CAD patients’ data (n = 1045) from NHANES
(2015-2016), older than 50 years for which lipid profile measurements were available. The clinical
diagnosis of CAD was positive in 91 individuals and negative in 945. The comparison of differences
in lipid profiles by sex was performed under the classical paradigm (independent sample t-test and
Wilcoxon Rank Sum Test) and Bayesian.

Results: CAD-positive men were younger (54-80 years) than women (57-80 years). Lipid parameters
(total cholesterol, LDL, direct HDL, non-HDL) differed significantly according to sex, under both classical
and Bayesian paradigms; except triglycerides and two proportions (TC: HDL, LDL: HDL). However, the
Bayesian paradigm has suggested differences even for triglycerides and two sex ratios.

Conclusions: The present study demonstrates the application of Bayesian t test in the case of a small
sample size. This clearly suggests that even when the sample size is small, the Bayesian paradigm
closely approximates our prior knowledge of the lipid profile as a risk factor for the development of CAD.
The Bayesian paradigm revealed the importance of clinical parameters (triglycerides, TC: HDL and LDL:
HDL), which remained hidden under the classical t-test and the non-parametric Wilcoxon Rank Sum Test.

Key Words: Bayes factor; posterior distribution; Triglyceride; Ratio of TC:HDL; Ratio of LDL:HDL.

CONFERENCE SOUVENIR 91