The research team recruited a convenience sample of 294 patients with Parkinson’s disease from a tertiary care hospital in Henan Province, China, between November 2021 and October 2023. Each participant completed a battery of validated instruments. Objective social isolation was measured using the Lubben Social Network Scale-6, a compact instrument that quantifies the actual size and frequency of a patient’s social network across family and friendship domains. Loneliness was assessed with the Chinese version of the 20-item UCLA Loneliness Scale, version 3, the field’s most widely used subjective measure of perceived social disconnection. Illness acceptance was captured with the Acceptance of Illness Scale, which gauges how well a patient has psychologically integrated the limitations imposed by their condition, and community self-efficacy was measured with the Community Self-efficacy Scale, reflecting a patient’s confidence in their ability to access and mobilize neighborhood and community resources.

With these four variables in hand, the researchers tested a moderated mediation model using Hayes’ PROCESS macro, version 3.4, Model 14. This statistical architecture is specifically designed for scenarios in which a moderator affects only the mediating path, the sequence running from independent variable to mediator to dependent variable, without regulating the direct path from independent variable to outcome. In practical terms, the team wanted to know whether illness acceptance statistically carries the association between objective social isolation and loneliness, and whether community self-efficacy changes the strength of that carried pathway. The answer to both questions was yes, and the pattern of effects carries significant implications for clinical nursing.

The numbers tell a striking story. The total effect of objective social isolation on loneliness was significant, with a standardized coefficient of -0.352 and a p-value of 0.005. The indirect effect flowing through illness acceptance was also significant, at -0.152 with a 95 percent bootstrap confidence interval spanning -0.286 to -0.039 and a p-value of 0.018. Remarkably, the direct effect was non-significant, at -0.200 with a p-value of 0.092. This pattern indicates that illness acceptance fully mediated the statistical association between objective social isolation and loneliness in this sample. In other words, social isolation does not appear to feed loneliness directly; rather, isolation erodes a patient’s acceptance of their illness, and that erosion of acceptance is what statistically translates into heightened loneliness.

This full mediation finding reframes the psychology of loneliness in Parkinson’s disease. Objective social isolation, the measurable shrinkage of a patient’s social network, appears to operate on loneliness through an internal psychological gate. When a patient’s network contracts, whether because motor symptoms make social outings harder, speech changes make conversation more taxing, or friends gradually drift away, the patient’s capacity to accept the illness diminishes. That diminished acceptance, in turn, is statistically associated with greater loneliness. The pathway suggests that simply adding social contacts may not be enough; interventions must also address the internal psychological process of acceptance if they hope to break the chain from isolation to loneliness.

But the study’s second finding may be even more consequential. Community self-efficacy significantly moderated the relationship between illness acceptance and loneliness, with a coefficient of 0.040 and a p-value below 0.001. The conditional indirect effects revealed a fascinating stratification: the mediating role of illness acceptance was significant at low levels of community self-efficacy, one standard deviation below the mean, with a coefficient of -0.135 and a 95 percent confidence interval of -0.280 to -0.029, and at average levels, with a coefficient of -0.083 and a confidence interval of -0.186 to -0.014. At high levels of community self-efficacy, one standard deviation above the mean, the indirect effect disappeared entirely, with a coefficient of -0.032 and a confidence interval of -0.101 to 0.020, which crosses zero.

This conditional pattern means that patients with higher community self-efficacy rely less on illness acceptance to mitigate loneliness. When a patient feels confident in their ability to navigate and draw upon community resources, the psychological burden of accepting the illness appears to matter less for their loneliness. The protective function of acceptance is effectively redundant when community confidence is high. Conversely, for patients with low community self-efficacy, illness acceptance becomes the critical psychological buffer, and its erosion by social isolation is statistically associated with the greatest loneliness. The indirect effect essentially vanishes at high efficacy levels, suggesting that community self-efficacy can substitute for the protective work that acceptance performs in less confident patients.

The authors argue that these findings provide a theory-driven empirical basis for stratified psychosocial nursing for patients with Parkinson’s disease. Rather than applying a one-size-fits-all intervention, clinicians should assess each patient’s community self-efficacy and tailor their approach accordingly. For patients with low community self-efficacy, health education should be prioritized to improve illness acceptance, since this psychological variable is the statistically significant mediator whose erosion by isolation is most strongly linked to loneliness in this group. For patients with high community self-efficacy, interventions should emphasize maintaining social and community networks during disease fluctuations rather than encouraging resource utilization, because these patients already possess the confidence to access resources and may instead need support to preserve the networks themselves as symptoms evolve.

The study’s design carries important caveats. As a cross-sectional investigation, it captures associations at a single point in time and cannot establish causal direction; it is possible that loneliness itself erodes illness acceptance or that both are shaped by unmeasured variables such as disease severity, depression, or cognitive status. The convenience sample of 294 patients from a single tertiary care hospital in Henan Province may not generalize to all patients with Parkinson’s disease, and the statistical mediation tested here is a model of associations rather than a demonstration of mechanism in the experimental sense. The authors themselves describe illness acceptance as functioning as a psychological statistical mediator, a careful phrasing that acknowledges the inferential limits of the design. Longitudinal and experimental studies will be needed to confirm whether boosting illness acceptance or community self-efficacy actually reduces loneliness over time.

Nevertheless, the findings arrive at a moment when the psychological dimensions of Parkinson’s disease are gaining unprecedented attention. Loneliness is highly prevalent among patients with Parkinson’s disease, and prior work had established that objective social isolation is associated with greater loneliness, but the integrated pathway involving social isolation, illness acceptance, and community self-efficacy had never been studied together. By demonstrating that illness acceptance fully mediates the statistical association and that community self-efficacy moderates this mediating pathway, the study adds a crucial piece to the puzzle of why some patients with shrinking networks remain emotionally connected while others spiral into profound loneliness. The research received funding from the Nursing Research Special Fund of the First Affiliated Hospital of Zhengzhou University under grant number HLKY20230001, was performed in accordance with the Declaration of Helsinki with ethical approval from the hospital’s Ethics Review Committee for Life Sciences under approval number 2023-KY-1373-002, and was published open access on 16 October 2026. For the millions of patients worldwide living with Parkinson’s disease, the message is quietly hopeful: the road from isolation to loneliness runs through acceptance, and confidence in one’s community can pave over that road entirely.

Subject of Research: Psychology

Article Title: Mechanisms underlying loneliness in Parkinson’s disease: the mediating role of illness acceptance and the moderating effect of community self-efficacy

Article References: Wang, W., Ma, K., Yang, C., & Guo, Y. (2026). Mechanisms underlying loneliness in Parkinson’s disease: the mediating role of illness acceptance and the moderating effect of community self-efficacy. BMC Psychology. https://doi.org/10.1186/s40359-026-05780-z

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05780-z

Keywords: Mechanisms, underlying, loneliness, Parkinson, disease, mediating, role, illness, acceptance, moderating, effect, community