Loneliness and Pain Among Racially Diverse Community-Dwelling Older Adults in the US: Untangling Directionality in the Relationship

Background: Loneliness and Pain are significant public health challenges. Both are associated with multiple negative health outcomes. Although some literature documents a relationship between loneliness and pain, little is known about this relationship among older adults, the directionality and whether the association varies as a function of race/ethnicity. We examine whether loneliness predicts pain and if pain predicts loneliness among Medicare recipients using a nationally representative data set of racially diverse community dwelling older adults in the US.

Methods: Our study is guided by the Minority Stress Framework and Biopsychosocial Model of Chronic Pain. We use data from Waves 2 and 3 of the National Social Life, Health, and Aging Project (NSHAP, which included well validated measures of objective and subjective measures of physical and mental health. Because nativity and insurance factors affect health, we focused our analyses on U.S. born Whites and African American and U.S. and foreign-born Latino Medicare beneficiaries aged 65 and older who participated in wave 2, and were alive at wave 3. We treated missing data using multiple imputation. Specifically, we performed Multiple Imputation using Fully Conditional Specification, Chained Equation methods, with 20 imputations, using Predictive Mean Matching for models for scale variables. We used two items to assess presence and intensity of pain. We defined Clinically Significant Pain (CSP) as pain with moderate to extreme intensity. We used the Felt Loneliness Scale to identify individuals with no loneliness, single wave and chronic loneliness. We used logistic regression models adjusting for key demographic, functional, chronic disease and psychosocial characteristics and interaction to test moderation effect of race/ethnicity.

Results: Among individuals with no pain at wave 2, 32% reported CSP at Wave 3. Pooled models (n=620) indicated that individuals reporting loneliness at both waves were 2.4 (p=0.01) times more likely to report CSP at wave 3 than individuals without loneliness at either wave. However, reporting loneliness at wave 2 only was not associated with CSP at wave 3. Among individuals with no loneliness at wave 2 approximately 10.7% reported loneliness at wave 3. Pooled models (n=1057), demonstrated that pain at wave 2 (acute or chronic) did not significantly predict the onset of loneliness. Race did not moderate effects in either direction.

Implications: Findings suggest that loneliness is a stressor that contributes to the development of CSP. However, the presence of CSP does not appear to contribute to development of loneliness. Particularly among racially diverse older adults, further research is necessary to gain deeper understanding of the subjective experiences of loneliness and pain and how cultural, minority stress and coping strategies influence their relationship. Researchers and practitioners should collaborate on examining if and how addressing loneliness may delay onset of CSP thru adaptation and implementing of culturally appropriate interventions in ‘real-world’ settings.