Abstract
BACKGROUND
Both all-cause and alcohol mortality follow a social gradient. Studies have provided evidence of interaction between socioeconomic status (SES) and alcohol use in predicting alcohol-attributable mortality. It is unclear whether this holds true for all-cause mortality. This study examines interaction effects and investigates the role of biomarkers in all-cause mortality.
METHODS
We included data from 4307 participants of the baseline survey (1997-2001) of SHIP (Study of Health in Pomerania)-START and a mortality follow-up from baseline until January 2023. All-cause mortality was the outcome; exposure variables were income or education as SES proxies, alcohol use, and biomarkers related to chronic alcohol consumption. Confounders were age, sex, marital status, smoking, and Body Mass Index (BMI). We used Cox proportional hazards regression (CPHR) in main analyses and CPHR and Aalen additive hazards regression (AAHR) in interaction analyses.
RESULTS
Our final samples (either with income or education) consisted of 3771/3939 individuals of whom 981 (26.0%) and 1047 (28.6%) died during the follow-up period. Adjusting for confounders, individuals with low income (HR = 1.31, CI = 1.11, 1.56; compared to high income) and both current abstainers that formerly consumed alcohol (HR = 1.56, CI = 1.27, 1.92) and individuals with high alcohol intake (HR = 1.31, CI = 1.03, 1.66; compared to individuals with low alcohol intake) showed increased mortality. Biomarkers were positively associated with mortality. There was evidence of statistical interaction effects, for example, increased risks in low-income current abstainers as well as individuals with moderate or high alcohol use.
CONCLUSIONS
We found a social gradient in all-cause mortality that was attenuated, but not fully explained, by risk factors. There was some evidence of statistical interaction effects between alcohol use and SES, but small subgroup counts limit the certainty of related conclusions.