Over the past 60 years, evidence has accumulated on the fundamental role of supportive social relationships in individual health and longevity. This paper first summarizes the results of 23 meta-analyses published between 1994 and 2021, which include 1,187 longitudinal and cross-sectional studies with more than 1,458 million participants. The effect sizes reported in these meta-analyses are highly consistent with regard to the predicted link between social support and reduced disease and mortality; the meta-analyses also highlight various theoretical and methodological issues concerning the multi-dimensionality of the social support concept and its measurements, and the need to control potential confounding and moderator variables. This is followed by an analysis of the experimental evidence from laboratory studies on psychobiological mechanisms that may explain the effect of social support on health and longevity. The stress-buffering hypothesis is examined and extended to incorporate recent findings on the inhibitory effect of social support figures (e.g., the face of loved ones) on fear learning and defensive reactions alongside evidence on the effect of social support on brain networks that down-regulate the autonomic nervous system, HPA axis, and immune system. Finally, the paper discusses the findings in the context of three emerging research areas that are helping to advance and consolidate the relevance of social factors for human health and longevity: (a) convergent evidence on the effects of social support and adversity in other social mammals, (b) longitudinal studies on the impact of social support and adversity across each stage of the human lifespan, and (c) studies that extend the social support framework from individual to community and societal levels, drawing implications for large-scale intervention policies to promote the culture of social support.