Relationship between alcohol consumption and myocardial infarction among ageing men using a marginal structural model

Area of Science:

• Cardiovascular Epidemiology
• Public Health
• Biostatistics

Background:

• Traditional studies on alcohol consumption and myocardial infarction (MI) rely on baseline data.
• This approach may not accurately reflect the dynamic relationship between alcohol intake and cardiovascular risk.

Purpose of the Study:

• To investigate the association between alcohol consumption and MI.
• To account for time-varying alcohol intake and confounders in the analysis.

Main Methods:

• Utilized data from 1030 Finnish males in the Kuopio Ischaemic Heart Disease Risk Factor Study.
• Employed conventional discrete-time hazard and marginal structural models (MSMs).
• Incorporated time-varying covariates including prior alcohol consumption and smoking status.

Main Results:

• Conventional models with baseline data showed an insignificant increase in MI risk for heaviest drinkers.
• Models incorporating time-varying alcohol consumption revealed a significantly increased MI risk (RR=1.71).
• Marginal structural models indicated a trend towards increased risk among heaviest consumers (RR=1.59).

Conclusions:

• Baseline data analysis may lead to biased estimates of the alcohol-MI association.
• Time-varying analyses are crucial for accurately assessing cardiovascular risk related to alcohol consumption.