Monitoring well-being as a warning indicator

The science of well-being is relatively new. It was initially built on the work of the positive psychology movement but then developed as a collaboration between a small group of economists and psychologists three decades ago. It has now grown into a subfield in economics that integrates social science disciplines with medical science. Thanks to the diversity of perspectives, data sources, and methods used in this research, the field is uniquely well-suited to address a wide range of policy questions by bridging disciplinary divides. As a result, many governments around the world have begun including well-being metrics into official statistics and in policy design, monitoring, and evaluation.

In this recently published paper, Emily Dobson, Isabel Shaheen O’Malley, Kasey Vangelov, and I demonstrate the contributions well-being studies can make to policy through a case study of the crisis of deaths of despair in the U.S. While such deaths were initially most prevalent among less-than-college-educated, middle-aged whites, they have become more common among minorities and the young. We build upon a methodological approach that uses trends in subjective well-being to track and create warning indicators of populations and places that are vulnerable to these deaths, focusing on these emerging groups. 

We rely on Centers for Disease Control mortality data and the well-being data from the Behavioral Risk Factor Surveillance System (BRFSS) and apply econometric techniques that are considered best-practice for predictive analysis. In our earlier work we found that despair increases pre-dated the increase in deaths in populations and places where they were common. As before, we matched our data on well-being trends for the specific populations under study with those on despair-related mortality.

Our study identifies a significant relationship between despair and the expansion of mortality to new racial and age cohorts in the U.S. The relationship is more complicated and varied across the different kinds of deaths than in our previous research. The inclusion of a broader range of population cohorts at least in part explains that change.

To address some of these challenges in our econometric analysis, we used panel vector-autoregressions, which allow us to identify a dual causality in the relationship between mental health and deaths from suicide, drugs, and alcohol. This suggests a cyclical pattern, where periods of high despair related deaths are associated with future worsening of mental health and distress and vice-versa. While having various channels of causality may complicate the design of policy interventions, the results suggest that intervening early to address increases in despair in vulnerable places are more effective than trying to address the cascading effects of increases in both despair and deaths at the same time.

It is important to note that while some of the years in our study corresponded with COVID-related declines in youth mental health, driven in part by the increase in social isolation made necessary by the COVID lockdowns, the roots of the youth mental health crisis and the spread of deaths of despair to younger cohorts and races preceded the COVID-19 pandemic. The declines in youth mental health that we document in the BRFSS began in 2011, and the spread of the deaths we documented began in 2018 or earlier. While it may have had an exacerbating effect, it is important to note that the increase in deaths, except for suicides, was still ongoing well after the COVID-19 crisis, as the final year of our analysis was 2023.

The results from our paper, coupled with more expansive research on the topic, could lead to a model that state and local governments could use to translate data on well-being and mental health into projections of deaths of despair in the subsequent years. That in turn would facilitate the use of similar models in other fields where despair plays a role (for example, crime and homicide trends, populist voting, and the spread of misinformation) to try and mitigate the spread of these trends before they occur. The situation in the U.S. today shows how damaging widespread population despair is on many facets of society.