Moderate alcohol intake has been related to lower mortality. However, alcohol use includes other dimensions beyond the amount of alcohol consumed. These aspects have not been sufficiently studied as a comprehensive entity. We aimed to test the relationship between an overall alcohol-drinking pattern and all-cause mortality. In a Mediterranean cohort study, we followed 18 394 Spanish participants up to 12 years. A validated 136-item FFQ was used to assess baseline alcohol intake. We developed a score assessing simultaneously seven aspects of alcohol consumption to capture the conformity to a traditional Mediterranean alcohol-drinking pattern (MADP). It positively scored moderate alcohol intake, alcohol intake spread out over the week, low spirit consumption, wine preference, red wine consumption, wine consumed during meals and avoidance of binge drinking. During the follow-up, 206 deaths were identified. For each 2-point increment in a 0–9 score of adherence to the MADP, we observed a 25 % relative risk reduction in mortality (95 % CI 11, 38 %). Within each category of alcohol intake, a higher adherence to the MADP was associated with lower mortality. Abstainers (excluded from the calculations of the MADP) exhibited higher mortality (hazard ratio 1·82, 95 % CI 1·14, 2·90) than participants highly adherent to the MADP. In conclusion, better adherence to an overall healthy alcohol-drinking pattern was associated with reduced mortality when compared with abstention or departure from this pattern. This reduction goes beyond the inverse association usually observed for moderate alcohol drinking. Even moderate drinkers can benefit from the advice to follow a traditional MADP.

In the present study the relationship between the consumption of different beverage groups and body-weight status in 5 years of study participation in German adolescents was investigated. We used anthropometric and dietary data from 3 d weighed records of 244 subjects between 9 and 18 years of age participating in the Dortmund Nutritional and Anthropometric Longitudinally Designed (DONALD) study. Only subjects with at least four out of six possible weighed dietary records were considered. A repeated-measures regression model (PROC MIXED) was used to analyse the effect of beverage consumption on body-weight status. BMI standard deviation scores (BMI-SDS) and body fat percentage (%BF) were chosen as the dependent variables. In boys, energetic beverage consumption was not associated with BMI-SDS or %BF, neither cross-sectionally nor prospectively. In girls, baseline consumption of energetic beverages did not predict baseline BMI-SDS, baseline %BF, or change in either variable over the study period. However, an increase in energetic beverage consumption over the study period was associated with an increase in BMI-SDS (+0.070 SDS/MJ increase in energetic beverage consumption; P = 0·01). Separate consideration of regular soft drinks and fruit juices revealed that, in girls, BMI-SDS increased with increased fruit juice consumption (+0·096 SDS/MJ increase in fruit juice consumption; P = 0·01), and to a lesser extent with regular soft drink consumption (+0·055 SDS/MJ increase in regular soft drink consumption; P = 0·08). In conclusion, these results suggest that an increase in energetic beverage consumption may result in weight gain, at least in adolescent girls.