What statistical test would be best for determining the effect of a patient's triglyceride levels on the incidence of myocardial infarctions (heart attacks), if you have a longitudinal cohort study and the baseline laboratory values for triglycerides? Data is at the patient level, coming from EHR data for the patient's hospital admission for MI (identified as such via ICD codes). Hint: EHR data is all time-stamped.

Break the situation down into the components for choosing a statistical test.

• dependent variable?
• dependent variable?
• independent variable(s)?
• type(s) of independent variable(s)?
• What are other things that could affect the statistical analysis?
• which of the following statistical analysis would you use:

One-sample t-test

Paired t-test

Two-sample t-test

ANOVA

Chi-square test

Simple linear regression

General linear model

Logistic regression

Wilcoxon rank-sum test

Kruskal-Wallis test

Spearman correlation

Fisher's exact test

Log-rank test

Cox proportional hazards model