The mediterranean diet and physical activityinteraction analysis and assessment of a mediterranean lifestyle score for the prevention of chronic diseases and premature mortality

Resumen

This dissertation includes 3 longitudinal analyses conducted among Spanish university graduates from the Seguimiento Universidad de Navarra (SUN) prospective cohort (publications 1, 3, and 4). Research previously conducted in the SUN cohort has focused on the association of risk factors with disease and mortality risk. This has been studied by estimating measures of association, mainly the relative risk either as the HR or OR, and potential interactions have been assessed on the multiplicative scale. A novel aspect that this dissertation has contributed within this cohort has been the study of measures of interaction between the MedDiet and PA on the additive scale. This methodology, which estimated RERI, provided further insights towards the joint association between these risk factors mainly using an additive scale for comparing absolute risks in contrast with the previously reported evidence on relative risks using a multiplicative scale. Furthermore, we also estimated the APs of the joint effect aimed to quantify the effect of the interaction in absolute values, taking a first step towards future analyses on measures of population impact associated with this risk factor combination. Another key contribution of this dissertation includes evidencing the association of the previously validated MEDLIFE index, which comprehensively captures the traditional Mediterranean lifestyle, with mortality and depression. In addition, this dissertation includes 2 cross-sectional studies that used baseline measurements from Feeding Americas Bravest, a two-year randomized trial aimed to test a nutritional intervention in US career firefighters (publications 2 and 5). This study design allowed us to apply our line of research to a non-Mediterranean population. Therefore, our findings have contributed new insights to the field of chronic disease epidemiology with regard to measures of interaction between modifiable lifestyle factors and the Mediterranean lifestyle among a non-Mediterranean population. In the following discussion we summarize the primary results of the five publications, state strengths and limitations of the studies, provide interpretations given the currently available evidence, discuss biological plausibility, generalizability, and the potential application along with future directions of our findings.