Objectives Spousal caregivers of older adults, especially new spousal caregivers, face increased risks of negative health outcomes due to the demands of caregiving and their own health decline. Estimating the impacts of caregiving on health without controlling for caregivers' own aging-related health decline could exaggerate the negative health consequences of caregiving, while focusing solely on caregivers could result in selection bias where healthier individuals enter and/or remain in caregiving. This study aims to estimate the impacts of caregiving on health of new spousal caregivers while controlling for observable confounders. Methods We utilized coarsened exact matching analysis to compare health outcomes between new spousal caregivers and spousal noncaregivers using pooled panel data from 2006 to 2018 in the Health and Retirement Study. We analyzed 242,123 person-wave observations from 42,180 unique individuals, among whom 3,927 were new spousal caregivers. Variables used for matching were classified into 3 categories: care needs, willingness to provide care, and ability to provide care. Two-year outcomes assessed are spouse's self-rated health, depressive symptoms, and cognitive functioning. Results A total of 3,417 (87.01%) new spousal caregivers were matched with 129,798 observations of spousal noncaregivers. Regression analysis indicated being a new spousal caregiver was associated with a 0.18- (standard error = 0.05) unit increase in number of depressive symptoms. No statistically significant results were identified for self-rated health and cognitive functioning. Discussion Our results highlighted the needs to address mental health among new spousal caregivers and emphasized the importance of addressing mental health in long-term care programs and policies. [ABSTRACT FROM AUTHOR]