Social Factors Associated with the Rate of Disablement Following Loss of a Spouse

Background/Purpose: This study investigates social context and demographic factors that could influence the trajectory of disablement in older adults who lost a spouse. Loss of a spouse is regarded as a major life stressor, capable of altering a survivor’s ongoing health trajectory. Such adverse life events are often analyzed cross-sectionally using retrospective data to assess the influence of stress on health status. A life-course perspective emphasizes longitudinal processes seen within social contexts and that are punctuated by potentially consequential life transitions. Therefore the present study uses longitudinal data to track changes in level of functional limitations over a time period that spans loss of a spouse, in the hope that this approach will better inform risk assessment and intervention planning in social work with older adults.

 

Methods: A nationally representative sample of 3937 individuals aged 50 and older who lost a spouse during up to 20 years of participation in the Health and Retirement Study (HRS) was selected. Separate measures for basic and instrumental activity limitations (ADL, IADL) were taken at two-year intervals. ADL include limitations associated with bathing, eating, dressing, walking across a room, and getting in or out of bed; IADL comprise using a telephone, taking medication, handling money, shopping, and preparing meals. The harmonized HRS data file (RAND-L) includes measures of basic demographic characteristics (age, sex, race, ethnicity, education) and both ADL and IADL scores and household size at each time. The selected cases were observed on at least two occasions prior to losing a spouse, and at least one time afterward. Data were collected through a combination of face-to-face interviews, by phone, and by mail-back questionnaire. Trajectories were analyzed using piecewise latent growth models with Mplus.

 

Results: The average number of observations per individual is 4.3 pre-loss occasions and 3.2 post-loss. The mean age at loss of spouse was 72.1 years, and 82% were deaths. Lower initial level of both ADL and IADL was associated with greater education; age and race emerged as significant predictors of pre- and post-loss rate of increase in ADL, while age alone was associated with rate of IADL increase before spousal loss; post-loss slopes for ADL were more deleterious among those who lost their spouse by separation/divorce; educational attainment was protective against increasing post-loss IADLs, while separation/divorce and larger household size exacerbated the rate of change. A binary variable indicating whether net household size (excluding the late spouse) had increased following the loss was not significant.

 

Conclusions and Implications: Loss due to separation/divorce was surprisingly more deleterious than death to trajectories of ADL and IADL, as was larger household size. The latter might be expected if family moved in with the surviving spouse to help out with worsening health, however the test of net increase did not support the interpretation. Knowledge of the protective effect of education and worsening associated with separation/divorce highlights individuals who may be more vulnerable to suffering limitations after losing a spouse.