Take a sad song and make it better: Spousal activity limitations, caregiving, and depressive symptoms among couples

Abstract

Objectives: Framed around key concepts of the life course perspective, we examined the linkages between spousal activity limitations, caregiving transitions, and depression among married couples. The key study objectives were 1) to demonstrate how the caregiving-depression link widely reported in earlier research may have been over-stated, and 2) to investigate whether caregiving yields mental health benefits by weakening the link between spousal activity limitations and depressive symptoms.

Methods: We used longitudinal data from the Health and Retirement Study (2004-2016) to examine a national sample of coupled individuals (6,475 couples; 57,844 person-wave observations). A series of longitudinal actor-partner interdependence models were used to estimate within-person associations between spousal activity limitations, caregiving transitions, and depressive symptoms among coupled individuals.

Results: Findings demonstrated that spousal activity limitations function as a confounder for the association between caregiving transitions and depressive symptoms. Results further provided evidence that transitioning into a caregiving role in the context of spousal activity limitations alleviated symptoms of depression for the caregiver.

Conclusion: The findings provide an explanation for the extended longevity benefit reaped by caregivers increasingly reported in recent population studies. Implications for policy, practice, and future research are discussed.

Keywords: Agency; Caregivers; Caregiving system model; Depression; Disability; Health and Retirement Study; Linked lives; Transitions.

Figure 1

A. Conventional framework: Spousal activity limitations are a necessary condition for caregiving behavior (defined in this study as providing help with activity limitations) to take place (path A1). Caregiving is often found to be a risk factor for depression in earlier studies (path A2). Another body of research subsequently identified a direct causal link between spousal activity limitations and depressive symptoms (path A3), often citing caregiving as an explanatory mechanism for the link between spousal activity limitations and depression. However, what becomes evident with path A3 is that spousal activity limitations qualify as a confounder for the effect of caregiving on depression. That is, because spousal activity limitations serve as an antecedent of both caregiving and depression, models estimating the link between caregiving and depression without considering spousal activity limitations would suffer from a significant omitted variable bias, potentially creating a spurious association between caregiving and depression (even when there is no relationship). As such, while the association between spousal activity limitations and depression can be estimated without considering caregiving, the association between caregiving and depression would lead to biased estimates and erroneous results if spousal activity limitations are not considered in the model. B. Current study framework: In contrast to Figure A, in Figure B we conceptualized caregiving as a downstream behavioral outcome of spousal activity limitations that would modify the link between spousal activity limitations and depression—an effect sometimes referred to as “mediated interaction” (VanderWeele, 2014). In the current framework, spousal activity limitations are also a necessary condition for caregiving behavior to take place (B1; marked in a dotted line because the association is not estimated in the analyses). However, caregiving behavior is evaluated as a moderator (B3) that would ameliorate the adverse association between spousal activity limitations and depressive symptoms (B2).