A META-REGRESSION AND BAYESIAN REGRESSION FRAMEWORK FOR COMBINING RESULTS OF SCIENTIFIC RESEARCH AND SURVEYS OF PEOPLE’S LIFESTYLES TO MAKE RECOMMENDATIONS ON WHAT INTERVENTIONS WILL HELP THEM LIVE LONGER AND HEALTHIER

Abstract

There has been increasing research into interventions that people can do in areas of diet, exercise, nutritional supplements, sleep, and stress management to improve their health and even slow down or reverse their aging process. Challenges that people face applying these research findings to their daily lives include the fact that this research is evolving and not yet “settled science,” most research looks at single interventions in order to maintain experimental control (whereas people are interested in maximizing their benefits by adopting multiple interventions), and different interventions or combinations thereof may be differentially effective for different people. The present paper presents a methodology that uses meta-regression and Bayesian regression to combine results from the scientific literature and surveys of people’s lifestyles into an overall predictive model that can make recommendations to people based on their individual characteristics and lifestyles as to what interventions are likely to produces the largest gains in wellness and lifespan extension.