Marital Loss and Well-Being in the Second Half of Life: Sociohistorical Context, Individual Differences, and Methodological Advances

Marital loss, whether through spousal bereavement or divorce, represents a major life
event with significant implications for well-being. However, it is an open question whether
such implications evolve over historical time. Further, it is also less well understood to what
extent the well-being changes reported in earlier studies can be attributed to different forms
of marital loss, beyond other age-related processes affecting well-being. This dissertation
aims to extend prior research by addressing these two research gaps, focusing on well-being
implications of marital loss in midlife and older adulthood. It also explores how
sociodemographic characteristics as well as various personal and social resources moderate
the consequences of marital loss across historical time and nature of loss.
Study I investigates historical shifts in the impact of spousal bereavement on wellbeing over the past four decades. Using data from N = 2,044 individuals (Mage at event = 65.73
years, 70% women) in the German Socio-Economic Panel (1984–2020), within-person
longitudinal changes were quantified with Area Under the Curve (AUC) metrics and latent
basis growth models. The findings indicate that while immediate reactions to losing a spouse
have remained largely stable over the period of study, individuals experiencing spousal loss
in the 2000s and 2010s exhibited less severe well-being declines compared to those in the
1980s and 1990s. These improvements were primarily driven by shallower anticipatory
declines and faster post-loss adaptation rather than changes in the acute reaction phase.
Notably, socio-demographic, health, and social moderators of bereavement effects remained
relatively stable over time. Study II aims to quantify the amount of change in well-being attributable to marital
loss both before and after the loss, and how the loss-related change varies across bereavement
and divorce. This study analyzes data from the Health and Retirement Study (1998–2018; N= 7,234; Mage = 66 years, 56% women, 75% White). Leveraging generalized synthetic control
methods, matched control groups of continuously-married or never-married individuals were
artificially constructed with identical pre-loss well-being trajectories, who served as
counterfactuals (in terms of not experiencing marital loss) for their bereaved and divorced
peers. Results demonstrated that spousal bereavement led to significant well-being declines,
particularly in the two years before and four years after the event. In contrast, no consistent
negative well-being effects emerged for divorce. Furthermore, larger social network, older
age, and non-White racial background appeared to buffer the adverse impacts of
bereavement.Together, these findings enhance our understanding of how marital loss shapes wellbeing in later life, underscoring the role of both sociohistorical contexts as well as individual
differences (e.g., age, social resources) in this process. They also draw important distinctions
between consequences of spousal bereavement and divorce. The implications and limitations
of the methodological innovations used in the studies are discussed. Finally,
recommendations are offered to guide future research toward more mechanism-oriented and
causally informative approaches.