LONELINESS, AGE, CARDIOVASCULAR AND METABOLIC HEALTH IN SCHIZOPHRENIA

Abstract

Introduction Loneliness is common in schizophrenia with prevalence estimates as high as 80% among people with schizophrenia. Recent studies among non-psychiatric samples have indicated deleterious effects of loneliness on physical health and mortality, including increased risk of hypertension, central obesity, metabolic syndrome, coronary heart disease, stroke, and other cardiovascular disorders. The all-cause mortality risk of loneliness is double that from obesity and equivalent to that from smoking ¾ of a pack of cigarettes per day. However, despite the increased risk for loneliness and cardiovascular problems in schizophrenia, these associations have not been examined directly in previous studies of people with schizophrenia. This study aims to explore the relationship of loneliness and age with cardiovascular and metabolic markers of physical health among participants with schizophrenia in comparison to non-psychiatric comparison (NC) participants. Methods The third edition of the University of California, Los Angeles Loneliness Scale (UCLA-3) was administered to 53 participants with schizophrenia, as diagnosed by DSM IV criteria, and 32 NC participants. Sociodemographic and clinical characteristics such as age, education, gender, duration of illness, and antipsychotic use were measured. Participants were also evaluated in terms of severity of psychopathology (positive, negative, depressive, and anxiety symptoms) using standard clinical symptom rating scales. Metabolic and cardiovascular health markers including LDL, HDL, total cholesterol, triglycerides, homeostatic model assessment of insulin resistance (HOMA-IR), systolic and diastolic blood pressure, fasting glucose, as well as waist circumference, body mass index were also collected. The Framingham 10-year Cardiovascular risk score was also calculated. This data was drawn from a larger ongoing study of aging in schizophrenia. We included only those participants with complete data for all of the above measures in the present analyses. Differences between the two groups on continuous variables were evaluated with independent t-tests; differences on categorical variables were evaluated with Pearson's Chi-square. Within each group, we also employed hierarchical multiple regression analyses to evaluate the association of loneliness with each health marker, while adjusting for gender and age. We also examined interactions between age and loneliness. (Given that loneliness is associated with an increased risk of mortality, the presence or absence of the latter could indicate potential survivor biases in the sample.) Results Participants with schizophrenia had a mean age of 51.41 (SD 11.03), with 20(37.7%) females, and 28 (52.8%) non-Caucasians. The NC participants had a mean age of 51.52 (SD 9.98), with 19 (59.4%) females, and 20 (62.5%) non- Caucasians. Relative to the NC group, participants with schizophrenia had worse loneliness, as well as worse positive, negative, depressive and anxiety symptoms. The also had fewer years of education, higher 10- year risk of cardiovascular disease, and were more likely to meet criteria for metabolic syndrome. Among the NC group, multivariate analyses indicated no association of loneliness with of any of the physical health measures, although the association of loneliness with total cholesterol suggested a non-trivial amount of shared variance (R2= 0.85, p= .086). Within the schizophrenia group, loneliness was associated with worse triglyceride levels; there was no age*loneliness interactions for most variables, suggesting the association was similar across age levels. However, there were significant loneliness x age interactions for cholesterol, t = -2.40, p = .021, R2 = .102, and LDL levels, t = -2.67, p = .011, R2 = .129. Specifically, loneliness was associated with worse cholesterol levels among individuals 1 SD below the mean in age (i.e., age 40.39), t= 2.30, p= .020, and better LDL levels among individuals 1 SD above the mean age (i.e., age 62.45), t= -2.037, p= .047. Conclusions Given the small sample size (particularly among NCs), and the exploratory nature of the present analyses, findings must be interpreted with caution until replicated in larger independent samples. However, the overall pattern of findings within the schizophrenia group is intriguing in suggesting a possible link between loneliness and blood lipid levels. There have been some GWAS studies suggesting an overlap in polygenic risk scores for cholesterol and loneliness. The differential effects of loneliness on cholesterol measures among older versus younger patients must be interpreted with particular caution, but if replicated could indicate those people with schizophrenia who survive to older age have a protective factor moderating the association of loneliness and cholesterol levels. The latter warrants further prospective research among older and younger people with schizophrenia. This research was funded by: Support provided by National Institute of Mental Health (NIMH) grant 5R01 MH094151-04 (Jeste), the NIMH T32 Geriatric Mental Health Program MH019934 (PI: Jeste), and UC San Diego's Stein Institute for Research on Aging (Director: Jeste).