UC Irvine

Heart disease has been the leading cause of death in the United States. The influence of long-term dietary behaviors, such as alcohol intake and fast-food consumption, on cardiovascular disease remains a topic of ongoing debate. As the population of older adults (65 and older) continues to grow, it becomes increasingly important to consider dietary strategies that promote optimal body composition, a key factor for disease risk and health and wellness. The goal of this dissertation is to estimate the effect of long-term dietary behaviors and cardiovascular disease risk in populations. The research additionally aims to leverage modern causal inference methodologies that allow for more robust conclusions about causation and impact.The dissertation consists of three separate research studies aiming to estimate the effect of 1) a range of sustained alcohol intake interventions on long-term cardiovascular disease risk, 2) reducing fast-food intake and the incidence and progression of coronary arterial calcium in adults with a history of frequent consumption, and 3) protein intake levels and changes in body composition measures in postmenopausal women. In the first two projects, analyses were conducted using a target trial emulation approach on participants from the Coronary Artery Risk Development in Young Adults (CARDIA) study, focusing on repeated measures of alcohol and fast-food consumption histories. The first study examined the effect of different levels of sustained alcohol consumption in young adults—ranging from abstention to more than 2 drinks a day for women and more than 3 drinks a day for men—and the 30-year cumulative incidence of all-cause cardiovascular disease. The second project examined the same population but focused on participants aged 33 to 45 with a history of frequent fast-food consumption (averaging ≥ 2 times per week) and the 10-year incidence and progression of coronary arterial calcium. The third study also utilized the target trial emulation approach but shifted to using data from the Women’s Health Initiative (WHI), aiming to examine the effect of different levels of protein intake thresholds on body composition, such as minimizing visceral adiposity accumulation and maximizing lean mass, in postmenopausal women. All three studies utilized Robin’s g-formula to estimate mean risks and outcomes. We found that abstaining from alcohol or consuming it in light to moderate amounts does not significantly affect long-term CVD risk in young adults. However, heavy alcohol intake did elevate CVD risk, with the impact ranging from minimal to modest. Additionally, we found that for adults aged 33-45 who have habitually consumed fast food since early adulthood, significantly reducing or eliminating fast-food intake over the next 10 years could lower the risk of developing incident coronary arterial calcium and also slow its progression. And in our third study, we estimated a protein intake of 1.2 g/kg/day or higher over three years would have a favorable effect on body composition in postmenopausal women, including lower visceral adiposity, regional subcutaneous adiposity, body fat percentage, and body weight, along with higher lean mass.