Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Session Type
Personal Schedule
Sign In
Access for All
Exhibit Hall
Hotels
WiFi
Search Tips
The global care crisis has intensified demands on informal caregivers, yet a critical puzzle persists: why do caregivers experiencing prevalent burnout refrain from engaging with available support? Using a novel population-representative survey data from 2,284 caregivers of older adults in Hong Kong, a Confucian-influenced, residual-welfare society where families absorb the primary duty of care, we examine patterns of perceived need and utilization across five caregiver services: allowance, respite, emotional support, skills training, and consultation. Latent class analysis identifies six configurations of service need and usage, ranging from non-users without perceived need to comprehensively underserved caregivers and comprehensive service users. Multinomial logistic regression reveals that both need recognition and service utilization are systematically stratified along classed and gendered lines. Those facing “multiple jeopardy”—economically marginalized, unmarried, managing multiple health conditions, and engaged in the most intensive caregiving—are paradoxically less likely to perceive themselves as needing supports, let alone access them. Two processes explain this paradox: drawing on Bourdieusian theory, prolonged structural disadvantage cultivates a classed habitus that normalizes hardship and forecloses recognition of suffering as warranting assistance; additionally, “intensive caregiving” arrangements including co-residence with care recipients and providing over 40 hours of care per week leave little cognitive or emotional space for self-reflection on one’s own needs. We further demonstrate how caregivers “do gender” in support-seeking: as care demands increase, women experience sharper rises in unmet need across most services but a decline for respite, reflecting gendered moral scripts around caregiving and help-seeking. This paper makes three contributions: (1) leveraging rare population-representative data that directly measure carers’ unmet need and service utilization; (2) examining need-usage configurations across five services via latent class analysis, with correlates assessed under Andersen’s behavioral model; (3) extending this model by theorizing need recognition as a socially structured process that reproduces inequality even before service access.