Research and theory increasingly suggest that attendance at religious services is associated with better self-rated health, life satisfaction and happiness and functional health. However, prior studies are limited and do not extensively explore potential explanations for these associations. This study estimates the impact of service attendance on mortality in a national probability sample and provides an extensive empirical examination of potential explanations. Data was gathered from the Americans' Changing Lives study, conducted by the Survey Research Center at the University of Michigan. The stratified, multistage, area probability sample included 3,617 noninstitutionalized persons aged 25 and over living in the United States. The analyses revealed a significant and sizable protective effect of service attendance on mortality (mortality hazards reduced by 30-35%), an effect which approximates that of moderate physical activity. To understand fully how and why attendance at religious services has a protective effect on mortality, future studies should assess the collective properties of religion, such as its reinforcement among parishioners of social support, improved self-concept and sense of well-being.