Introduction: To evaluate the broad impact of alcohol on health and healthcare utilization, the dose-response relationship of alcohol intake with all-cause and cause-specific hospitalizations was examined.
Methods: In the Moli-sani study, an Italian population-based cohort, we followed 20,682 initially healthy individuals (48% men, age ≥35 y) free of CVD or cancer. Alcohol intake in the year before enrolment was assessed by the Italian EPIC-FFQ and classified as: abstainers (referent), ex-drinkers, occasional drinkers (<1 gr/day), 1-12, 12.1-24, 24.1-48 and >48 gr/day. We identified hospitalizations by linkage to the regional hospitalization registry. Cause-specific hospitalizations were assigned by the ICD9 code of the primary admission diagnosis. We estimated incidence rate ratios (IRR) for admission by Poisson regression, accounting for repeated hospitalizations.
Results: At baseline, 27% of participants were abstainers, 3% ex-drinkers, 6% occasional drinkers and 64% regular current drinkers. During a median follow-up of 6.3 y, 12,996 hospital admissions occurred. In multivariable analyses, occasional consumption and intake up to 48 gr/day were associated with a lower risk of all-cause hospitalization than was abstention. There was a roughly dose-dependent association with lower risk of hospitalization for vascular disease. Excessive alcohol consumption was associated with a higher risk of hospitalization for alcohol-related diseases (IRR: 1.92, 95% CI: 1.43-2.59) and for cancer (IRR: 1.33, 95% CI: 1.08-1.63). Former drinkers were at higher risk for vascular and alcohol-related hospitalization. No association was observed with hospitalization for trauma.
Conclusions: Heavy alcohol consumption is associated with higher risk of hospitalization for alcohol-attributable conditions and cancer, but intake up to 48 gr/day is associated with lower risk of all-cause and vascular hospitalization. These estimates highlight the different healthcare burden imposed by varying levels of alcohol intake.