Abstract
INTRODUCTION: Caregiver-oncologist concordance regarding the patient's prognosis is associated with worse caregiver outcomes (e.g., depressive symptoms), but mechanisms underpinning these associations are unclear. We explored whether caregiving esteem mediates these associations. METHODS: At enrollment, caregivers and oncologists used a 5-point ordinal scale to estimate patient survival; identical responses were considered concordant. At 4-6 weeks, caregivers completed an assessment of the extent to which caregiving imparts self-esteem (Caregiver Reaction Assessment self-esteem subscale; range 0-5; higher score indicates greater esteem). They also completed Patient Health Questionnaire-2 (PHQ-2) for depressive symptoms, Distress Thermometer, and 12-Item Short Form Survey for quality of life (QoL). Mediation analysis with bootstrapping (PROCESS macro by Hayes) was used to estimate the extent to which caregiving mediated the effects of prognostic concordance on caregiver outcomes through caregiving esteem. RESULTS: Prognostic concordance occurred in 28% the caregiver-oncologist dyads; 85% of the discordance were due to caregivers estimating a longer patient's survival. At 4-6 weeks, mean caregiving esteem score was 4.4 (range 1.5-5.0). Lower caregiving esteem mediated the associations of concordance with higher PHQ-2 [indirect effect = 0.12; 95% Confidence Interval (CI) 0.03, 0.27], greater distress (indirect effect =0.25; 95% CI 0.08, 0.48), and poorer QoL (indirect effect = -1.50; 95% CI -3.06, -0.41). Caregiving esteem partially mediated 39%, 64%, and 48% of the associations between caregiver-oncologist concordance and PHQ-2, distress, and SF-12, respectively. CONCLUSIONS: Caregiver-oncologist concordance was associated with lower caregiving esteem. Lower caregiving esteem mediated the negative relationship between caregiver-oncologist concordance and caregiver outcomes.