Predictors of caregiver burden over a three month period following hospitalization of the patient

The purpose of this secondary analysis was to determine the pattern of caregiver burden over a twelve-week period following the patient\u27s hospitalization, and the ability of baseline variables to predict caregiver burden. Interview data were collected from 125 caregivers and their care recipients in a larger study designed to test the effectiveness of a discharge planning protocol implemented by advanced practice nurses. Data collection points were during hospital admission, and at 2, 6, and 12 weeks after discharge. Caregiver burden was measured with a modified form of the Caregiver Strain Index. All analyses controlled for random group assignment and multiple data collection points. An analysis of variance revealed significant changes in the mean scores for caregiver burden across the episode of illness. Caregiver burden peaked at 2 weeks, and then slowly declined to baseline by 12 weeks after discharge. Setwise mixed model regression analyses indicated that caregiving context, patient health stressors, and perceived social support explained from 9–18 percent of incremental variance over the benchmark model. The final model, including the timemark, accounted for 26.8 percent of the explained variance in caregiver burden scores across an episode of the patient\u27s illness. Perceived social support, the number of prior hospitalizations within six months, and the patient\u27s functional status were found to be significant predictors of caregiver burden in the final model. Social support mediated the relationship between caregiver burden and the caregiver\u27s age and self-rated health. The implications for further research and clinical interventions are discussed