# Understanding 30-Day Mortality After First STEMI Through DAGs: Unravelling Epidemiological Cause-Effect Links

## TL;DR

This study uses DAGs to identify factors influencing 30-day mortality after first STEMI, revealing sex as a key factor with direct and indirect effects.

## Contribution

The study introduces DAGs with NOTEARS and permutation testing to uncover causal relationships in STEMI mortality.

## Key findings

- 246 out of 2,946 patients died within 30 days post-STEMI, with higher mortality in older patients.
- Sex was a significant factor affecting mortality directly and indirectly through age, stress, and hypertension.
- Variables like HbA1c, triglycerides, and socioeconomic status showed significant associations with mortality.

## Abstract

Background and aim: Traditional statistical tests have limitations in analyzing cause-and-effect relationships. Directed acyclic graphs (DAGs) offer a structured representation of causality. This study aimed to utilize DAGs to explore the causal impact of epidemiological factors on 30-day mortality among patients following their first acute ST-elevation myocardial infarction (STEMI).

Method: The study employs data from the North India (NORIN)-STEMI study registry, comprising 3,192 first-time STEMI patients collected prospectively from two tertiary care hospitals in Delhi, India. Continuous optimization structure learning using the Non-combinatorial Optimization via Trace Exponential and Augmented Lagrangian for Structure Learning (NOTEARS) method is applied to learn the DAG. Additionally, a permutation testing framework is proposed for the statistical validation of the links of the DAG.

Results: Among 2,946 first-time STEMI patients, 246 (7.7%) experienced mortality during the study period. A t-test revealed that age was significantly different between the survival and mortality groups within 30 days post-STEMI (p<0.0001). Patients who died within 30 days had a higher mean age (59.90±13.89 years). Furthermore, the study identified a statistically significant association between mortality and HbA1c, triglycerides, smoking, sex, education, occupation, socioeconomic status, physical activity, overall stress, and hypertension.

Conclusion: Our DAG reveals causal relationships and identifies confounding variables affecting mortality after STEMI. Sex is identified as a significant factor influencing mortality both directly and indirectly. This influence occurs through its effects on age, alcohol consumption, stress, hypertension, and socioeconomic status. Additionally, sex is recognized as a confounding factor whose impact on mortality is modified by other factors.