Purpose. Widowhood is an extremely stressful life event in old age. While previous studies have extensively examined the buffering effects of social support in adjustment to widowhood, little research has explored how widowhood affects older adults' social relationships. Understanding the ways social relationships change following spousal loss is critical in order to provide effective assistance to widowed older adults. By knowing precisely in which aspects of social relationships people lose (or gain) support after they become widowed, interventions can be targeted at helping widowed older adults maintain the kinds of social support and network that are optimal for them. Thus, this paper examines the extent to which widowhood affects changes in emotional support that older adults receive from their children, friends, and relatives.

Methods. Analyses are based on data from the Changing Lives of Older Couples (CLOC), a prospective study of 1,532 married individuals aged 65 and older. After baseline interview, widowed persons were interviewed at six, 18, and 48 months after spousal death. Controls matched along the dimensions of age, race, and sex, were interviewed at roughly the same time. The analytic sample for this paper includes 297 individuals (211 widowed persons and 86 married persons) who participated in the baseline and the six-month follow-up interviews. Ordinary least squares regression models are used to examine the effect of widowhood on (1) support from adult children and (2) support from friends and relatives six months after spousal death, controlling for support at baseline. Changes in both positive (e.g., caring, willing to listen) and negative (e.g., demanding, critical) aspects of support are evaluated.

Results. Compared to still-married persons, widowed persons report higher levels of support from their children and friends/relatives six months after spousal death, net of support at baseline. When the effect of widowhood is evaluated separately for positive and negative support, widowhood does not affect positive support, yet it significantly decreases negative support. Pre-widowhood levels of support are significantly associated with the support after the loss for both social relationships. In addition, baseline marital quality, emotional stability, and support from friends are significant predictors of children's support following widowhood. Emotional stability, good spousal health, and children's support at baseline predicted higher levels of support from friends and relatives at six-month follow-up.

Implications. The findings provide useful insights into the ways in which close social ties respond to the distress of the surviving spouse. One of major findings is that widowhood significantly influences negative support but not positive support. Knowing that widowed persons go through a stressful life transition, especially at an earlier stage of bereavement, children and friends may try to refrain from being critical or demanding to the surviving spouse. Another important finding is that support from these relationships during times of bereavement is influenced by levels of support prior to widowhood as well as other important pre-loss characteristics (e.g., marital quality, spousal health). Practitioners may be able to better serve widowed older adults by taking pre-loss contexts of widowhood into account.