Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: The number of children reared by relatives other than their parents is at an all time high and continues to grow most dramatically among families with the lowest incomes and the greatest social service needs. Prior research indicates that kinship caregivers are unaware of many services and supports for which they are eligible and reluctant to make themselves known to social service agencies to ask for resources that may be available. Although prior research reveals elevated levels of caregiving burden and distress among kinship caregivers, little is know about the factors that exacerbate or reduce this stress over time. We address this gap in our knowledge base by examining the effects of child behavioral and emotional functioning, financial/material resources, family functioning, and social support on caregiver stress through structured face to face interviews conducted with 207 informal kinship caregivers. Data were collected in four waves over an 18-month period.

Method: Caregiver stress was measured by the short form of the Parenting Stress Index; children's functioning by the Child Behavior Checklist; financial/material resources by the Family Resource Scale; family functioning by the Beavers Self-report; and social support by the Family Support Scale. Analysis of four-wave interview data was conducted to assess if factors that have been shown to be associated with caregiver stress in cross-sectional data were also associated with the change of caregiver stress. Using SAS PROC MIXED, a series of hierarchical linear models were fitted to the data. First, an unconditional growth (or change) model of time in Level 1 and no predictors in Level 2 was estimated. After comparing the likelihood ratio tests of a model with a quadratic function of time and a linear function time, a linear individual growth model was chosen. Then, factors that might potentially account for the variation in the individual change parameter were included as Level 2 covariates.

Results: Our model showed that the individual change parameter as represented by the linear function of time was not statistically significant, which indicated there was no significant change in caregiver stress. Also, the variance of the change parameter was not statistically significant, which suggested that there was no significant variation in the change of caregiver stress. After controlling for demographic variables, caregiver stress at baseline was significantly associated with children's problematic behavior, family resources, family functioning and social support. Family resources moderated the relationship between family functioning as well as social support and caregiver stress.

Implications: Results suggest that associations between the predictor variables and caregiver stress remain stable over time. These associations suggest that interventions designed to reduce stress of informal kinship caregivers should begin with an assessment of the adequacy of financial and material resources. Family group conferencing, mediation and other family interventions are increasingly common in work with kinship caregiving families. These interventions are designed to strengthen family support and improve family functioning. The results of this study suggest that these interventions may be successful in reducing caregiver stress only after ensuring that financial and material resources are adequate.