Decomposition and comparative analysis of differences in loneliness between urban and rural older adults: evidence from a national survey

Loneliness is a significant public health issue in China, particularly among the older population. This study sought to explore differences in the incidence of loneliness among older adults in urban versus rural regions and to quantify the contribution of relevant influencing factors. This study assessed data from 13,735 older adults derived from the 2018 Chinese Longitudinal Healthy Longevity Survey (CLHLS). The chi-squared test was employed to examine the distribution characteristics of relevant indicators among older adults in the two regions. Logistic regression was utilized to explore the factors affecting loneliness in older adults across different areas. Additionally, the Fairlie decomposition analysis method was applied to quantify the four categories of influencing factors—demographic characteristics, sociological characteristics, personal lifestyle, and health status—that contributed to differences in loneliness and to estimate their respective contributions among urban and rural older adults. A total of 26.79% of older adults reported feeling lonely, with a higher proportion of rural older adults (27.77%) experiencing loneliness than their urban counterparts (24.27%). In rural regions, risk factors for loneliness included a body mass index (BMI) of < 18.5 kg/m², being widowed, and other marital statuses, as well as self-reported local income status as “rich.” Conversely, protective factors included being male, having an educational level of 1–6 years or ≥ 7 years, self-reported local income status as “average,” alcohol consumption, engaging in exercise, and self-reported health (SRH) as “good.” In urban areas, risk factors included being widowed or in other marital statuses, as well as self-reported local income status as “rich.” In contrast, protective factors comprised being ≥ 100 years old, engaging in exercise, having sleep durations of 6.0–7.9 h, 8.0–9.9 h, or ≥ 10 h, and self-rated health as “good.” The Fairlie decomposition analysis revealed that 95.38% of the disparity in loneliness symptoms stemmed from observable factors, whereas the remaining 4.62% was attributed to urban–rural differences. Factors such as gender, educational attainment, marital status, living conditions, self-reported local income level, alcohol consumption, physical exercise, sleep length, and SRH played a significant role in explaining the disparities in loneliness symptoms between the two groups. Older adults residing in rural regions exhibited a greater prevalence of loneliness than their counterparts in urban areas. This difference is mainly caused by gender, educational level, marital status, living status, self-reported local income status, drinking, exercise, sleep duration, and SRH. Addressing these influential factors of loneliness would aid in developing targeted and more precise intervention strategies specifically tailored to enhance the mental wellness of high-risk older adults.