54% of families report feeling at least somewhat guilty after placing a relative in a nursing home, according to a Care.com survey of 978 US adults (2019), with 14% feeling “extremely guilty.” A peer-reviewed parallel from Gaugler et al. (2021), tracking 83 dementia caregivers through residential long-term care transitions, found that nearly half (46.3%) experienced guilt from the care recipient, other family members, or facility staff. Adult children reported higher guilt rates than spouses — 37.5% versus 26.7% at baseline in the Residential Care Transition Module trial — and the qualitative record is thick with the specific fear that placement happened “too soon.” Action-side guilt is well-documented, replicated across research designs, and begins immediately after the move.
The inaction side lacks a symmetrical direct-regret measure, but the burden data are substantial. The 2020 NAC/AARP nationally representative survey of 1,707 US caregivers found that 23% said caregiving had made their own health worse — up from 17% in 2015. The same report found 28% had stopped saving and 23% had taken on more debt because of caregiving. Schulz and Sherwood’s 2008 synthesis of three decades of caregiver health research concluded that caregiving “creates physical and psychological strain over extended periods of time” and that roughly two-thirds of caregivers experience some negative consequence. The floor of inaction-side harm is real but harder to frame as regret because most home caregivers are embedded in an ongoing situation rather than looking back on a discrete choice.
Under Gilovich and Medvec’s framework, actions generate more regret in the short term, and the placement guilt literature fits this pattern: guilt is highest immediately after the move and tends to attenuate as the care recipient stabilises and the caregiver observes professional care quality. Inaction regret — regret about not having placed sooner — is harder to study because it surfaces mainly when home care has already broken down (hospitalisation, caregiver health crisis, or care recipient injury from inadequate supervision). The 31-point gap between sides should be read cautiously: the two measures are not equivalent constructs, and a direct bilateral survey asking both groups the same retrospective question does not yet exist in the published literature.