Primary Submission Category: Mortality
Socioeconomic and psychosocial predictors of all-cause mortality in the MIDUS Study
Authors: David Curtis Ming Wen Sara Grineski Amelia Herder Yehua Dennis Wei Yue Zhang
Presenting Author: David Curtis*
Trends in life expectancy in the US plateaued, even declining in years, during the two decades prior to the COVID-19 pandemic. First reported for White working-age adults, “deaths of despair” was cited as an explanation as job opportunities worsened for less educated Whites. Later research showed rising mortality across racial/ethnic groups and for an expanded set of causes of death that challenged the narrow explanation of despair. This study tests affective, social, and psychological predictors on mortality, and the extent to which they explain the socioeconomic influences on mortality.
Data come from the Midlife in the United States (MIDUS) study. Participants were interviewed in 1995-1996 with follow-up data collection in 2004-2006 and 2013-2014 and mortality tracing through 2019. The sample includes 6,115 participants and 1,531 deaths observed.
Risk of all-cause mortality is modeled using Cox proportional hazards analysis with repeated observations. We find large differences in mortality risk by education and employment. Adults with less than a high school diploma had 29-34% higher hazard relative to adults with a diploma through a 2-year college degree/vocational school and 53-56% higher hazard than individuals with a 4-year degree or graduate degree. Unemployment was associated with more than two times higher mortality hazard relative to part- or full-time employment. Adjusting for psychosocial measures reduced hazard ratios for higher education categories by 4-9% and for unemployment by 35%. Multidimensional composite measures of affect and stress were associated with all-cause mortality, equivalent to 1 SD higher negative affect and stress predicting 17% and 11% greater mortality hazard.
Our results are consistent with vast, interdisciplinary research literatures that identify negative affect, psychosocial stress, and social support as predictors of morbidity and mortality, and demonstrate plausible mechanisms for socioeconomic inequities in mortality.