Correlates of Emotional and Social Loneliness Among Persons with Serious Mental Illness

Background and Purpose: Loneliness is associated with negative physical and mental health outcomes. Persons with serious mental illness (SMI; i.e., major depressive, bipolar, schizophrenia spectrum disorders) may experience greater loneliness compared to the general population (Borge et al, 1999). Multidimensional models of loneliness have been proposed: social loneliness refers to the absence of social network and integration; emotional loneliness refers to the lack of intimate connection with another person (Weiss, 1973; de Jong-Gierveld, 1987). This study examines the extent to which social and emotional loneliness are distinct concepts among older adults with SMI. We hypothesize that emotional response variables—reactivity to fear, anger, and positive emotional cues—will more strongly correlate with emotional loneliness, while social loneliness will be more strongly correlated with perceived social support.

Methods: Cross-sectional data were collected from persons receiving intensive case management services at a community mental health center (N = 136). Participant race, gender, age, and educational attainment were self-reported. Social and emotional loneliness were measured by the 11-item DeJong Gierveld Loneliness Scale (DJGL; de Jong-Gierveld & Kamphuls, 1985). Reactivity to positive emotions, fear, and anger were collected with the Emotional Numbing and Reactivity Scale (Orsillo, et al., 2007). Perceived social support from family, friends, and a significant other was measured by the 12-item Multidimensional Scale of Perceived Social Support (MSPSS; Zimet et. al., 1988). Two multiple linear regression models were conducted to identify the contribution of perceived social support and emotional reactivity to social and emotional loneliness.

Results: Participants’ mean age was 58.31 years (SD = 5.84). The sample was primarily male (54.41%) and African-American (61.76%). Demographic, emotional response, and social support variables explained 32.2% of the variance in emotional loneliness (F(11,124) = 5.35, p < .001). Emotional loneliness was significantly associated with reactivity to fear (B = .193, p < .05) and lower reactivity to positive emotions (B = -.246, p < .01). Reactivity to anger and social support variables did not significantly contribute to the model. The model for social loneliness explained 43.2% of the variance in scores (F(11, 124) = 8.57, p<.001), with friend (B = -.311, p < .01) and family (B = -.233, p < .01) support contributing significantly to the model. Social loneliness was uncorrelated with support from a significant other or emotional response.

Conclusions and Implications: This study provides support for the multidimensionality of loneliness among an SMI sample. Emotional loneliness was associated with decreased reactivity to positive cues and increased reactivity to fear cues. Social loneliness covaried with perceived low support from family and friends. Emotional loneliness was strongly associated with emotional response styles suggestive of negative affect, which could disrupt attempts to affiliate with others. Comprehensive and accurate identification of both dimensions of loneliness may support practitioners in tailoring psychiatric rehabilitation and other therapeutic interventions for older persons with SMI. Future research may explore the differential effects of each type of loneliness on health and mental health outcomes of adults with SMI.