Objective: To examine the relationship between care recipient (person with Alzheimer's disease) ability to perform daily tasks and caregivers' (CG) perceived burden and depression, guided by the caregiver identity theory. We also examine the mediating effect of CG abilities to meet their basic needs. Methods: This study utilizes the baseline data of the REACH II study. Spearman's rho (ρ) was used to test for relationships between burden, reported depression, and each ADLs and IADLs. To further explore the relationship between burden and each ADLs and IADLs, structural equation modeling was conducted using Mplus 8.0. Results: Reported CG total scores indicated increased perceived CG burden with greater number of assisted daily activities. CG depression scores were significantly predicted by reported burden scores and caregiver's ability to pay for basic needs. Importantly, 34.6% of variation in CG reported depressions scores were explained by reported burden scores. A multivariate regression model with reported burden scores, controlling for caregiver's ability to pay for basic needs, explained 36.6% of the variance in CG depression scores. Burden scores and CG ability to pay for basic needs significantly predicted depression scores. Results from the three models indicated that CG burden fully mediated the relationship between daily living skill scores and CG depression. Conclusion: Our study findings suggest the need to more closely examine the link between AD caregiving, financial instability, and mental health and bolster support for policies and programs that offer tangible supports and services to offset the costs of informal AD CG.