No abstract

BackgroundSocioeconomic disadvantage is a fundamental cause of morbidity and mortality. One of the most important ways that governments buffer the adverse consequences of socioeconomic disadvantage is through the provision of social assistance. We conducted a systematic review of research examining the health impact of social assistance programs in high-income countries.MethodsWe systematically searched Embase, Medline, ProQuest, Scopus, and Web of Science from inception to December 2017 for peer-reviewed studies published in English-language journals. We identified empirical patterns through a qualitative synthesis of the evidence. We also evaluated the empirical rigour of the selected literature.ResultsSeventeen studies met our inclusion criteria. Thirteen descriptive studies rated as weak (n = 7), moderate (n = 4), and strong (n = 2) found that social assistance is associated with adverse health outcomes and that social assistance recipients exhibit worse health outcomes relative to non-recipients. Four experimental and quasi-experimental studies, all rated as strong (n = 4), found that efforts to limit the receipt of social assistance or reduce its generosity (also known as welfare reform) were associated with adverse health trends.ConclusionsEvidence from the existing literature suggests that social assistance programs in high-income countries are failing to maintain the health of socioeconomically disadvantaged populations. These findings may in part reflect the influence of residual confounding due to unobserved characteristics that distinguish recipients from non-recipients. They may also indicate that the scope and generosity of existing programs are insufficient to offset the negative health consequences of severe socioeconomic disadvantage.Electronic supplementary materialThe online version of this article (10.1186/s12889-018-6337-1) contains supplementary material, which is available to authorized users.

Since the turn of the 21st century, during which White mortality has been rising, there has been a sharp increase in only three causes of death, drug use, alcohol use, and suicide. Because all three of these causes conjure notions of anguish and hopelessness, they have been conceptualized as a collective “deaths of despair” phenomenon. Simons and Masters challenge this conceptualization, by asking if these three causes are empirically associated with each other. Their analyses produce small correlations, which lead them to call into question that the three causes are part of a unified phenomenon. We contest their work on several grounds. Their analyses suffer from several technical problems, including the fact that, for any given year and cause of death, 65.8-97.6% of counties have death counts under 10. More fundamentally, it is unclear that we should expect these causes of death to rise and fall together, even if they are connected to a singular phenomenon. Instead, ‘despair’ may manifest differently in different places (i.e. these causes may be substitutes for each other). We argue that the best answer to the authors’ important question comes from assessing whether there is a common, despair-based causal mechanism underlying all three of them.