Using cox hazard regression, this study conducted a longitudinal analysis to investigate the longitudinal effects of social cohesion and living status on entry into long-term care. Study sample included older adults dwelling in community at round 1. After removing proxy respondents and dropped cases through round2 and round 6, the final sample size was 3,181 in the present study. Total 396 (12.4%) older adults were institutionalized through 6 rounds and 2,785(87.6%) were censored. Based on a health behavioral model introduced by Andersen (1995), influential factors in entry into long-term care were categorized into personal characteristics (age, gender, education, race), enabling factors (long-term care insurance, Medicaid, social cohesion, living arrangement), and need factors (mental health, chronic conditions, functional disabilities (ADL), self-rated health).
As for need factors, functional disability was strongly associated with entry into long-term care (HR = 1.60, p < .001). Mental health status (phq4) was significantly associated (HR = 1.09, p < .01). Among enabling factors, Medicaid and long-term care insurance were significantly associated with entry into long-term care (HR = 1.89, p < 001, and HR = 1.31, p < .05 respectively). Contrary to expectation, social cohesion was not a protective factor for entry into long-term care. Older adults who perceived higher social cohesion were more likely to enter long-term care (HR = 1.13, p < .001). Living arrangement was identified as a strong predictor for entry into long-term care (HR = .28, p <.001). The interaction between living with others and functional disability was also significant (HR = .74, p <.001).
In this longitudinal study, social cohesion increased the likelihood of older adults’ entry into long-term care. This finding implies that social cohesion might play a role in diffusing health-relevant information and increasing access to long-term care services (Echeverría, Diez-Roux, Shea, Borrell, & Jackson, 2008). Therefore, social cohesion might be regarded as an enabling factor of health behavioral model instead of a protective factor that improves individual health and delays the entry into long-term care. This study also revealed that living arrangement (living with others) is the most significant protective factor decreasing the likelihood of entry of long-term care. In addition, this study identified the moderating effect of living arrangement (living with others) in protecting against facilitative effects of functional disability on entry of long-term care.