Background: Most of the available epidemiological evidence on alcohol and chronic disease agrees on recommending alcohol abstention to young people, but some controversy exists about the most appropriate recommendation for alcohol abstention for people of older ages. A growing body of evidence suggests that the pattern of alcohol consumption is likely to be a strong effect modifier. The Mediterranean Alcohol Drinking Pattern (MADP) represents a score integrating several dimensions of drinking patterns (moderation, preference for red wine, drinking with meals, and avoiding binge drinking). Our aim was to clarify this issue and provide more precise recommendations on alcohol consumption. Methods: We prospectively followed-up 2226 participants (men older than 50 years and women older than 55 years at baseline) in the Seguimiento Universidad de Navarra (SUN) cohort. We classified participants into three categories of adherence to the MADP score (low, moderate, and high), and we added a fourth category for abstainers. Cox regression models estimated multivariable-adjusted hazard ratios (HR) of all-cause death and 95% confidence intervals (CI) using low MADP adherence as the reference category. Results: The strongest reduction in risk of mortality was observed for those with high adherence to the MADP, with an HR of 0.54 (95% CI: 0.37-0.80). The moderate adherence group (HR = 0.65, 95% CI: 0.44-0.96) and the abstention group (HR = 0.60, 95% CI: 0.36-0.98) also exhibited lower risks of mortality than the low MADP adherence group. Conclusions: based on the available evidence, a public health message can be provided to people older than 50 years as follows: among those who drink alcohol, high adherence to the MADP score could substantially reduce their risk of all-cause mortality.