[WITHDRAWN] Predictors of Loneliness Among People with Psychosis over a Six-Month Period

Background and Purpose:

Loneliness is a subjective and distressing experience of feeling disconnected or socially isolated from others. Loneliness is understudied among populations who experience psychosis, despite research showing that people with psychosis report experiencing loneliness at significantly higher rates than the general public. Loneliness is typically not assessed at the outset or during traditional treatments for psychosis, yet early research has shown that loneliness is likely related to social and clinical factors frequently experienced by persons with psychosis. The aim of this study was to examine the associations between loneliness, psychosocial, clinical, and social cognitive factors over time among people with psychosis spectrum diagnoses.

Methods:

This remote study collected data at baseline, four-, and six-months. The sample consisted of people with confirmed psychosis spectrum diagnoses who were between the ages of 18-59. Approximately 60% of the sample was white and about 57% identified their sex as female. Baseline measures were completed by 108 participants with 77 people completing measures at all three timepoints.

A drop-out analysis was conducted to assess whether baseline differences were present between those that completed all measurement timepoints and those that did not. Linear mixed-effect models were fit to examine associations between loneliness (ULS), internalized stigma (ISMI-10), defeatist beliefs (DBS), motivation and pleasure (MAPS-SR), depression (BDI-II), emotion recognition (ER-40), positive and negative psychosis symptoms (SAPS/SANS), as well as demographic information (e.g., age, race/ethnicity, sex, etc.). A random-effect at the participant-level was included in these models. Time-varying variables were decomposed to establish within- and between-person values more clearly. Model fits were compared with an ANOVA to test which was the best fit for predicting loneliness.

Results:

The drop-out analysis indicated that participants who completed all measurement timepoints had significantly lower BDI-II scores than those who did not; no other differences were detected.

The model that best predicted loneliness included a significant positive main-effect for within- and between-person stigma, a significant negative main-effect for between-person motivation to engage in activities, and a significant positive main-effect for within-person depression scores. Between-person scores for stigma and motivation to engage in activities were the strongest predictors of loneliness in the model. Approximately 46% of the variance in loneliness was explained by the fixed effects in this model, while about 76% of the variance was explained by the total model.

Implications:

These findings suggest that stable experiences (between-person) of heightened internalized stigma and reduced motivation to engage in activities are strongly associated with higher levels of loneliness. Fluctuating levels (within-person) of stigma and depression were also shown to impact self-ratings of loneliness.

The study suggests that loneliness is not a simple artifact of depression and that interventions to improve the social experiences of people with psychosis may need to address stigma and motivation to engage in activities. Additional research is needed to gain insight into whether loneliness is associated with difficulties anticipating pleasure, which is a common feature of psychosis, as well as how to better address the potentially alienating impact of internalized stigma associated with psychosis spectrum diagnoses.