The beneficial association of the Mediterranean diet (MedDiet) with longevity has been consistently demonstrated, but the associations of MedDiet components have not been accordingly evaluated. We performed an updated meta-analysis of prospective cohort studies published up to 31 December 2017, to quantify the association of adherence to MedDiet, expressed as an index/score (MDS) and of its components with all-cause mortality. We estimated summary relative risks (SRR) and 95 % CI using random effects models. On the basis of thirty studies (225 600 deaths), SRR for the study-specific highest/lowest and per 1sd MDS increment were 0·79 (95 % CI 0·77, 0·81, Ι2=42 %, P-heterogeneity 0·02) and 0·92 (95 % CI 0·90, 0·94, Ι2 56 %, P-heterogeneity <0·01), respectively. Inversely, statistically significant associations were evident in stratified analyses by country, MDS range and publication year, with some evidence for heterogeneity across countries overall (P-heterogeneity 0·011), as well as across European countries (P=0·018). Regarding MDS components, relatively stronger and statistically significant inverse associations were highlighted for moderate/none-excessive alcohol consumption (0·86, 95 % CI 0·77, 0·97) and for above/below-the-median consumptions of fruit (0·88, 95 % CI 0·83, 0·94) and vegetables (0·94, 95 % CI 0·89, 0·98), whereas a positive association was apparent for above/below-the-median intake of meat (1·07, 95 % CI 1·01, 1·13). Our meta-analyses confirm the inverse association of MedDiet with mortality and highlight the dietary components that influence mostly this association. Our results are important for better understanding the role of MedDiet in health and proposing dietary changes to effectively increase adherence to this healthy dietary pattern.

No abstract

For many diseases it is reasonable to assume that the hazard rate is not constant across time, but also that it changes in different time intervals. To capture this, we work here with a piecewise survival model. One of the major problems in such piecewise models is to determine the time points of change of the hazard rate. From the practical point of view this can provide very important information as it may reflect changes in the progress of a disease. We present piecewise Weibull regression models with covariates. The time points where change occurs are assumed unknown and need to be estimated. The equality of hazard rates across the distinct phases is also examined to verify the exact number of phases. An example based on herpes zoster data has been used to demonstrate the usefulness of the developed methodology.