Alcohol consumption and hospitalization burden in an adult Italian population: prospective results from the Moli-sani study

Background and aims: Epidemiological evidence on the impact of different alcohol drinking patterns on health-care systems or hospitalizations is sparse. We investigated how the different average volumes of alcohol consumed relate to all-cause and cause-specific hospitalizations. Design: Prospective cohort study (baseline 2005–10) linked to a registry of hospital discharge records to identify hospitalizations at follow-up (December 2013). Setting: Molise region, Italy. Participants: A total of 20 682 individuals (48% men, age ≥ 35 years) who participated in the Moli-sani Study and were free from cardiovascular disease or cancer at baseline. Measurements: The alcohol volume consumed in the year before enrolment was classified as: life-time abstainers, former drinkers, occasional drinkers and current drinkers who drank 1–12 (referent), 12.1–24, 24.1–48 and > 48 g/day of alcohol. Cause-specific hospitalizations were assigned by Italian Diagnosis Related Groups classification or by ICD-9 code of main admission diagnoses. Incidence rate ratios (IRR) of hospitalization were estimated by Poisson regression, taking into account the total number of admissions that occurred during the follow-up per person. Findings: During a median follow-up of 6.3 years, 12 996 multiple hospital admissions occurred. In multivariable analyses, life-time abstainers and former drinkers had higher rates of all-cause [IRR = 1.11, 95% confidence interval (CI) = 1.05–1.17 and IRR = 1.19, 95% CI = 1.02–1.31, respectively] and vascular (IRR = 1.14, 95% CI = 1.02–1.27 and IRR = 1.48, 95% CI = 1.24–1.76, respectively) hospitalizations compared with light alcohol consumers. Alcohol consumption > 48 g/day was associated with a higher rate of hospitalization for both alcohol-related diseases (IRR = 1.74, 95% CI = 1.32–2.29) and cancer (IRR = 1.36, 95% CI = 1.12–1.65). The magnitude of the association between heavier alcohol intake and hospitalization tended to be greater in smokers than non-smokers. No associations were observed with hospitalization for trauma or neurodegenerative diseases. Conclusions: Moderate alcohol consumption appears to have a modest but complex impact on global hospitalization burden. Heavier drinkers have a higher rate of hospitalization for all causes, including alcohol-related diseases and cancer, a risk that appears to be further magnified by concurrent smoking.