Predictors of developing a new need for long-term care of older adults aged ≥70 years: Results from a population-based cohort study in Taiwan.

Abstract

To determine the characteristics associated with developing a new need for long-term care among older Taiwanese adults. We analyzed the 1999 and 2003 datasets of the Taiwan Longitudinal Survey on Aging. A total of 1986 persons who were aged ≥70 years and had no disability in activities of daily living or need for long-term care at the baseline served as the study participants. We first identified the potential variables associated with developing a need for long-term care in demographic, socioeconomic, anthropometric, functional status, mental health, medical and global domains based on the literature and our earlier studies. We then carried out univariable logistic regression analyses to identify the variables most strongly associated with developing a new need for long-term care within each domain. Variables showing the strongest association were further analyzed in a forward stepwise multivariable logistic regression model to determine the significant predictors. The multivariable model identified age, sex, doing heavy housework, walking 200 m, carrying out calculation, appetite status, frequency of hospitalizations during the past 12 months, living arrangement and weight loss as significant (P < 0.05) predictors. The present study identified nine characteristics that predict the subsequent 4-year new need for long-term care of ≥70-year-old Taiwanese adults. Age, sex and functional status are the most powerful predictors. The results are useful for identifying older individuals at risk of developing a new need for long-term care within the next 4 years, enabling implementation of preventive strategies or timely care planning. Geriatr Gerontol Int 2019; 19: 641-646.