Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Dec 24;82(8):7942–7951. doi: 10.1111/jan.70452

Social Isolation and Its Influencing Factors in Maintenance Haemodialysis Patients: A Latent Profile Analysis

Luwen Zhang 1, Fangli Liu 1,✉, Jinghui Liu 1, Xia Zheng 2
PMCID: PMC13356399  PMID: 41439303

ABSTRACT

Aims

To explore latent profiles of social isolation in maintenance haemodialysis (MHD) patients and to analyse the factors influencing different latent profiles.

Design

Multicentre cross‐sectional study.

Methods

Between November 2024 to March 2025, 305 MHD patients from the haemodialysis centres of three hospitals in Henan Province, China, were recruited using a convenience sampling method. All participants completed the general information questionnaire, Lubben Social Network Scale 6 (LSNS‐6), UCLA Loneliness Scale‐6 (ULS‐6) and Personal Mastery Scale. Latent Profile Analysis (LPA) was used to classify the participants into potential subgroups with different types of social isolation. The influencing factors of profiles were explored by univariate analysis and multiple logistic regression analysis.

Results

Social isolation of 305 patients can be divided into three profiles: the family‐friend dual isolation group (14.10%), friend isolation‐only group (47.54%), and social network well‐being group (38.36%). Multivariable logistic regression analysis revealed that monthly personal income, living arrangement, social participation, dialysis time, post‐dialysis fatigue, number of comorbidities, loneliness and personal mastery were identified as factors influencing the profiles.

Conclusions

There is heterogeneity in social isolation among MHD patients. It is therefore necessary to implement targeted intervention measures based on the distinct characteristics of each subgroup to facilitate their social reintegration.

Implications for the Profession and Patient Care

Nurses should identify differences in social isolation among MHD patients. It is necessary to establish tripartite connections between families, hospitals and communities, and develop personalised psychosocial interventions to alleviate social isolation.

Implications

The study identified distinct subgroups of social isolation among MHD patients, while emphasising the impact of psychological resources such as loneliness and personal mastery on social isolation. This may offer critical insights for nurses to develop targeted interventions for patients' social health.

Reporting Method

The study followed the STROBE guidelines for cross‐sectional studies.

Patient or Public Contribution

No patient or public involvement.

Keywords: latent profile analysis, loneliness, maintenance haemodialysis, nursing, personal mastery, social isolation, social network

What Does This Paper Contribute to the Wider Global Clinical Community?

  • LPA was used to identify heterogeneous subgroups of social isolation among MHD patients, which is expected to provide a scientific basis for designing stratified and personalised psychosocial clinical interventions.

  • The differential roles of loneliness and personal mastery across subgroups could promote the development of an assessment framework that integrates subjective psychological and objective social functions in MHD nursing practice.

1. Introduction

End‐Stage Renal Disease (ESRD) refers to the final stage of chronic kidney disease, defined by an estimated glomerular filtration rate (eGFR) below 15 mL/min/1.73 m2 (KDIGO Group 2024). According to the 2024 United States Renal Data System (USRDS) report (Johansen et al. 2024), over 815 thousand people in the US had ESRD by the end of 2022, a 7.56% increase from 2017. Patients at this stage suffer from severe impairment of renal function and rely on long‐term renal replacement therapies such as maintenance haemodialysis (MHD) to sustain life (Torreggiani et al. 2023). ESRD patients undergoing MHD require dialysis 2–3 times per week, with each session lasting 4.0–4.5 h. Although MHD significantly prolongs patients' survival, the frequent and long‐term nature of dialysis treatment often prevents them from maintaining normal social interactions and work, thereby limiting their engagement in social activities and increasing the risk of social isolation (Sluiter, Cazzolli, et al. 2024).

Social isolation is defined as a detrimental state characterised by limited social connections or infrequent social interactions, resulting from active or passive detachment from social networks (Nicholson Jr. 2009). Research indicated that social isolation not only diminishes subjective well‐being and induces negative emotions (Li et al. 2024) but may also correlate with increased unhealthy behaviours, such as reduced physical activity (Herbolsheimer et al. 2018), poor dietary patterns (Hanna et al. 2023) and decreased medication adherence (Yu et al. 2024). Furthermore, Zhou et al. (2021) found that high levels of social isolation are associated with a rapid decline in renal function. Therefore, social isolation plays a significant role in health management and disease progression among MHD patients, and it is a public health issue that should be valued. However, existing self‐management and symptom management strategies for MHD patients often neglect the psychosocial dimension that has a significant impact on patients' health (Friedrich et al. 2025). Thus, exploring social isolation and its influencing factors among MHD patients holds great significance for future nursing practices in health management among MHD patients.

2. Background

Currently, studies mainly assess social isolation in MHD patients using total scale scores, treating these patients as a homogeneous group while overlooking individual heterogeneity. Latent profile analysis (LPA), a person‐centred approach, categorises individuals into subgroups based on their distinct patterns across observed variables (Kongsted and Nielsen 2017). This method identifies subgroups with similar latent trait profiles, acknowledges inter‐individual variability, avoids homogenising all patients, and thereby provides a more accurate structural understanding of the data (Naldi and Cazzaniga 2020). Such stratification enables clinical staff to implement precise interventions at the group level.

Loneliness is a negative emotion that arises when an individual lacks intimacy and social contact in life (Bekhet et al. 2008). A qualitative study shows that MHD patients often experience intense feelings of loss and loneliness due to the constraints of dialysis treatment, particularly experiencing extreme loneliness when their family members and friends fail to understand their situation (Jeong et al. 2025). Loneliness reflects unmet needs for social connection, which may trigger withdrawal behaviours (e.g., avoiding interactions, reducing social participation), directly exacerbating social isolation (Beller and Wagner 2018). Therefore, investigating loneliness can provide deeper insights into the subjective experiences and psychological states of patients facing social isolation, grasping its impact mechanism from a subjective level. Although numerous current studies have examined the relationship between social isolation and loneliness, the impact of loneliness on MHD patients with distinct social isolation profiles remains insufficiently explored.

Personal mastery refers to an individual's perception of their ability to influence and control the outcomes of life events or their surrounding environment (Peterson and Stunkard 1989). It reflects patients' sense of control and confidence in their lives. The composite theory of personal control, suggesting that an individual's sense of mastery exists in the interaction between the person and the world, represents a belief about how one engages with the world (Peterson and Stunkard 1989). For MHD patients, they must strictly adhere to fixed dialysis schedules and locations without the flexibility to make autonomous adjustments. This treatment pattern may reduce their interaction with the outside world, make them feel that their life is dominated by dialysis treatment, lead to diminished personal autonomy, and even create a sense of being unable to master their own lives effectively. Studies have shown that higher levels of personal mastery can promote positive changes in health‐related behaviours in individuals (Fateh et al. 2021). Therefore, individuals with high personal mastery may take the initiative to adapt to life changes imposed by treatment (e.g., dialysis schedule conflicts) and maintain social role functioning. Currently, only the study has shown that the long‐term disease burden of peritoneal dialysis patients significantly affects their personal mastery, and that these patients have poor stress‐coping abilities and are prone to social alienation (Diao et al. 2024). The level of personal mastery in MHD patients and its impact on their social isolation remain to be further investigated.

This study is the first to apply LPA to categorise social isolation among MHD patients into distinct groups, breaking through the homogenization limitation of previous assessments that relied solely on total scale scores. Meanwhile, it incorporates loneliness and personal mastery to explore the impact of subjective psychological experiences on objective social functions among MHD patients. The findings are expected to offer critical insights for mitigating social isolation, facilitating social reintegration and designing targeted interventions for this population.

3. The Study

Aims: To explore latent profiles of social isolation among MHD patients and analyse the sociodemographic, disease‐related and psychological factors influencing these profiles.

4. Methods

4.1. Design

A multicentre cross‐sectional study was conducted following the STROBE guidelines.

4.2. Study Setting and Sampling

This study was conducted from November 2024 to March 2025. Using convenience sampling, patients undergoing maintenance haemodialysis were recruited from the haemodialysis centres of three tertiary hospitals in Kaifeng city, Henan Province, China. According to Kendall's sample size estimation method, the sample size should be at least 5 to 10 times the number of independent variables, with an additional 10%–20% allowance for invalid responses. Considering lost or invalid questionnaires, we opted to add a 20% allowance for invalid responses. The study included 13 items of general demographic information, 6 items of LSNS‐6, 6 items of ULS‐6 and 7 items of Personal Mastery Scale, resulting in a total of 32 independent variables. Thus, the required sample size ranged from 192 to 384. With reference to Ferguson et al. (2020), who recommended a minimum sample size of 300 for LPA, we ultimately enrolled 305 MHD patients.

4.3. Inclusion and Exclusion Criteria

Inclusion criteria: (1) Age ≥ 18 years; (2) undergoing regular haemodialysis with a duration of dialysis ≥ 3 months; (3) possess basic verbal response and comprehension abilities for routine communication, which are assessed by the researcher through direct conversation; and (4) have signed an informed consent form and voluntarily participate. Exclusion criteria: (1) Individuals with cognitive or psychiatric disorders; (2) currently in an acute exacerbation phase of disease; (3) comorbid with other severe conditions, for example, critical illnesses, organ transplantation; and (4) presence of hearing or visual impairments.

4.4. Instruments

4.4.1. Demographic and Disease‐Related Characteristics

Based on a literature review, the research team developed a general information questionnaire. It includes 13 items: gender, age, educational level, economic status (monthly income/pension/alimony, etc.), marital status, living arrangement, residential area, occupational status, participation in social activities (e.g., card games, square dancing, volunteering), dialysis time, vascular access type, the number of other comorbid chronic diseases (diabetes mellitus, hypertension, cardiovascular diseases, cerebrovascular diseases, etc.) and post‐dialysis fatigue.

4.4.2. Lubben Social Network Scale 6 (LSNS‐6)

The Lubben Social Network Scale‐6 (LSNS‐6), developed by Lubben et al. (2006), is designed to accurately and rapidly assess social isolation in older adults. The scale comprises two dimensions: family network and friend network, with a total of 6 items. Each item has 6 options scored 0–5 (none, 1, 2, 3–4, 5–8, ≥ 9). The total score ranges from 0 to 30, with a score below 12 indicating social isolation. A family or friend network score under 6 suggests family or friend isolation. The Chinese version of LSNS‐6 has construct validity coefficients ranging from 0.84 to 0.96 and has a Cronbach's α of 0.832.

4.4.3. ULS‐6

Xiao and Du (2023) revised the Chinese version of the 8‐item UCLA Loneliness Scale (ULS‐8) to form the 6‐item ULS‐6. All 6 items are positively scored, using a 4‐point scale from 1 to 4. An item score of ≥ 2 suggests the individual feels some degree of loneliness. The total score ranges from 6 to 24, with higher scores indicating greater loneliness: 6–9 (none), 10–13 (mild), 14–17 (moderate) and 18–24 (severe). The ULS‐6 has a Cronbach's α of 0.891. Additionally, the ULS‐6 exhibits a robust unidimensional structure with good construct validity in measuring loneliness among adults.

4.4.4. Personal Mastery Scale

Developed by Pearlin and Schooler (1978), this scale measures individuals' perceived control over life outcomes. It consists of 7 items rated on a 5‐point Likert scale: 1 (Strongly Disagree) to 5 (Strongly Agree). Items 1–5 are reverse‐scored, while items 6–7 are positively scored. Total scores range from 7 to 35, with higher scores indicating greater personal mastery. The Chinese version has a Cronbach's α of 0.81.

4.5. Data Analysis

Data were imported into Epidata 3.1 after dual‐entry verification. Latent profile analysis (LPA) was conducted utilising Mplus 8.3 software, with scores from the six items of the Lubben Social Network Scale as exogenous variables. Starting from a one‐class model, the number of classes in the model was gradually increased until the best‐fitting model indicators were achieved. Model fitting indicators include the following:

(1) Information criteria: Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and adjusted BIC (aBIC), where lower values denote better fit.

(2) Classification accuracy: Entropy value (range: 0–1), with values closer to 1 indicating higher classification precision. Entropy ≥ 0.8 corresponds to ≥ 90% classification accuracy.

(3) Lo–Mendell–Rubin likelihood ratio test (LMR) and Bootstrap‐based Likelihood Ratio Test (BLRT). p < 0.05 indicated that the kth model is better than the k–1th model.

The optimal model was selected by comprehensively evaluating these indices. Patients were assigned to distinct latent classes, and subsequent statistical analyses were conducted using IBM SPSS 26.0. Multiple logistic regression was employed to identify factors associated with latent class membership. Categorical data were described using frequencies and percentages (%) and the χ 2 test or Fisher's exact probability method were used for comparison between multiple groups. Normally distributed continuous data were expressed as means ± standard deviation and analysed using ANOVA for comparison between multiple groups. Test level was α = 0.05.

4.6. Ethical Considerations

This study was approved by the University Institutional Ethics Committee (Approval No: HUSOM2025‐551). Each patient participated voluntarily and provided signed informed consent forms.

5. Results

5.1. Characteristics of the Sample

A total of 305 MHD patients were included in the study. The ages of the participants ranged from 23 to 88 years (57.66 ± 12.56), with an average dialysis duration of 56 months (approximately 4 years). The cohort included 188 males (61.6%) and 117 females (38.4%), with 114 residing in rural areas (37.4%) and 191 in urban areas (62.6%). A total of 236 patients (77.4%) reported never participating in social activities (e.g., playing chess and cards, square dancing, Tai Chi, volunteer work). Autogenous arteriovenous fistula was the predominant vascular access type (81.31%). 69.5% of patients reported post‐dialysis fatigue. Additionally, 49.18% had two or more comorbid chronic conditions.

Among MHD patients, the social isolation score was 13.27 ± 3.88, with a prevalence of 49.18%. Specifically, the family network score was 8.86 ± 1.73, with a family isolation incidence of 11.80%, while the friend network score was 4.41 ± 3.30, with a friend isolation incidence of 61.64%. The loneliness score was 10.15 ± 3.20, with 39.67% of patients reporting feelings of loneliness. The personal mastery score was 25.47 ± 1.874.

5.2. Results of Latent Profile Analysis

Latent profile analysis was conducted using the six items of the LSNS‐6 as exogenous variables. Starting with a one‐class model, we sequentially fitted models by increasing 1 to 5 latent profiles (Table 1). As the number of profiles increased from 1 to 3, the AIC, BIC and aBIC values decreased, and both the LMR and BLRT were statistically significant (p < 0.05). However, when we increased the number of profiles from 3 to 4, the LMR test was no longer significant (p = 0.601). Based on comprehensive comparisons of model fit indices, the 3‐profile model was selected as the optimal best‐fitting model.

TABLE 1.

Indicators for latent profile of social isolation in MHD patients.

Profiles AIC BIC aBIC Entropy LMR (P) BLRT (P) Proportion (%)
1 4725.715 4770.359 4732.301 — — — —
2 3878.622 3949.308 3889.049 0.980 < 0.001 < 0.001 0.603/0.397
3 3733.252 3829.980 3747.520 0.914 0.004 < 0.001 0.141/0.475/0.384
4 3072.033 3194.804 3090.144 1.000 0.601 < 0.001 0.351/0.259/0.262/0.128
5 3028.497 3177.310 3050.449 0.978 0.0155 < 0.001 0.3508/0.0656/0.1967/0.2590/0.1279
Open in a new tab

Table 2 presents the classification probability matrix. The results show that the average probability of each group belonging to its respective latent profile is above 0.90, and the average probability of being classified into the other two profiles is below 0.10, indicating high reliability of the three‐profile classification.

TABLE 2.

The average probability of participants being assigned to each group in the three latent profiles.

Profile Profile 1 Profile 2 Profile 3
Profile 1 0.933 0.057 0.010
Profile 2 0.050 0.942 0.008
Profile 3 0.000 0.003 0.977
Open in a new tab

The characteristics of the three latent social isolation profiles among MHD patients are illustrated in Figure 1. Profile 1, profile 2 and profile 3 were named based on their LSNS‐6 item score distributions. Profile 1 (n = 43, 14.10%): The mean scores of all LSNS‐6 items were at the lowest level overall, with both family and friend subscale means ≤ 6, indicating overall isolation in both family and friend network. Thus, it was labelled the ‘Family‐friend dual isolation group’. Profile 2 (n = 145, 47.54%): The mean scores of LSNS‐6 items 1–3 were relatively high, indicating a well‐functioning family network. However, the mean scores of LSNS‐6 items 4–5 were the lowest, and the friend dimension mean was < 6, indicating isolation limited to friend only. Thus, it was labelled the ‘Friend isolation‐only group’. Profile 3 (n = 117, 38.36%): The mean scores of all LSNS‐6 items were the highest, with both family and friend subscale means > 6, indicating a well‐functioning social network overall. Thus, it was labelled the ‘Social network well‐being group’.

FIGURE 1.

FIGURE 1

Open in a new tab

Characteristic distribution of three latent profiles of social isolation in MHD patients.

5.3. Univariate Analysis of Latent Profiles

Univariate analysis revealed statistically significant differences among the three social isolation profile groups in age, education level, monthly income, living arrangements, occupational status, social participation, dialysis time, number of comorbidities and post‐dialysis fatigue (p < 0.05) (Table 3). Additionally, the three latent social isolation profiles exhibited significant differences in loneliness scores (13.16 ± 6.07 vs. 10.88 ± 2.85 vs. 8.15 ± 1.66 points, p < 0.05) and personal mastery scores (24.07 ± 2.27 vs. 25.12 ± 1.53 vs. 26.41 ± 1.63 points, p < 0.05) (Table 4).

TABLE 3.

Results of a univariate analysis of latent profiles of social isolation in MHD patients.

Variable Total N (%) (n = 305) Family‐friend dual isolation group (n = 43) Friend isolation‐only group (n = 145) Social network well‐being group (n = 117) χ 2 or F p
Age (years)
18–44 53 (17.38%) 5 (7.69%) 17 (32.69%) 31 (59.62%) 17.808 0.001
45–59 103 (33.77%) 14 (14.42%) 44 (42.31%) 45 (43.27)
≥ 60 149 (48.85%) 24 (16.11%) 84 (56.38%) 41 (27.52%)
Educational level
Primary school or below 68 (22.29%) 9 (13.24%) 45 (66.18%) 14 (20.59%) 23.696 0.001
Junior high school 131 (42.95%) 19 (14.50%) 66 (50.38%) 46 (35.11%)
High school/Technical secondary school 77 (25.25%) 10 (12.99%) 25 (32.47%) 42 (54.54%)
University or above 29 (9.51%) 5 (17.24%) 9 (31.03%) 15 (51.73%)
Personal monthly income (CNY)
< 1000 114 (37.38%) 17 (14.91%) 66 (57.90%) 31 (27.19%) 14.876 0.004
1000–5000 176 (57.70%) 25 (14.21%) 76 (43.18%) 75 (42.61%)
> 5000 15 (4.92%) 1 (6.67%) 3 (20.00%) 11 (73.33%)
Living arrangements
Living with family 278 (91.15%) 29 (10.43%) 137 (49.28%) 112 (40.29) 24.920 < 0.001
Living alone or other 27 (8.85%) 14 (51.85%) 8 (29.63%) 5 (18.52%)
Occupational status
Employed 26 (8.53%) 1 (3.85%) 4 (15.38%) 21 (80.77%) 20.654 < 0.001
Retired 134 (43.93%) 22 (16.42%) 65 (48.51%) 47 (35.07%)
Unemployed 145 (47.54%) 20 (13.79%) 76 (52.41%) 49 (33.79%)
Social participation
Never 236 (77.38%) 42 (17.80%) 131 (55.51%) 63 (26.69%) 60.833 < 0.001
Occasionally/often 69 (22.62%) 1 (1.45%) 14 (20.29%) 54 (78.26)
Dialysis time (year)
< 1 41 (13.44%) 6 (14.64%) 16 (39.02%) 19 (46.34%) 13.822 0.008
1–5 164 (53.77%) 18 (10.98%) 72 (43.90%) 74 (45.12%)
> 5 100 (32.79%) 19 (19.00%) 57 (57.00%) 24 (24.00%)
Number of comorbidities
0–1 155 (50.82%) 12 (7.74%) 62 (40.00%) 81 (52.26%) 28.670 < 0.001
≥ 2 150 (49.18%) 31 (20.67%) 83 (55.33%) 36 (24.00%)
Post‐dialysis fatigue
Yes 212 (69.51%) 37 (17.45%) 117 (55.19%) 58 (27.36%) 36.042 < 0.001
No 93 (30.49%) 6 (6.45%) 28 (30.11%) 59 (63.44%)
Open in a new tab

TABLE 4.

Results of loneliness and personal mastery scores among three profiles of social isolation.

Family‐friend dual isolation group (n = 43) Friend isolation‐only group (n = 145) Social network well‐being group (n = 117) F p
Loneliness 12.93 ± 3.98 10.64 ± 2.63 8.15 ± 1.63 64.36 < 0.001
Personal mastery 24.07 ± 2.27 25.12 ± 1.53 26.41 ± 1.63 31.01 < 0.001
Open in a new tab

5.4. Multiple Logistic Regression Analysis of Latent Profiles

The multicollinearity test showed that the VIF values of all independent variables were < 5, so there were no multicollinearity issues. Using the three latent social isolation profiles of MHD patients as the dependent variable (with the ‘Social network well‐being group’ as the reference), a multinomial logistic regression analysis was performed. Independent variables included the nine factors identified as statistically significant in univariate analysis, along with loneliness and personal mastery scores. In the model fitting information results, the p‐value of the likelihood ratio test was < 0.001, indicating that the model is statistically significant. The R 2 values ranged from 0.37 to 0.61, demonstrating a good fit. The results indicated that personal monthly income, living arrangement, social participation, dialysis time, post‐dialysis fatigue, number of comorbidities, loneliness and personal mastery were significantly associated with social isolation profiles (p < 0.05) (Table 5).

TABLE 5.

Multiple logistic regression analysis of latent profiles of social isolation in MHD patients.

Variable Family‐friend dual isolation group vs. Social network well‐being group Friend isolation‐only group vs. Social network well‐being group
β p OR 95% CI β p OR 95% CI
Age (years)
≥ 60 −1.291 0.144 0.275 (0.049, 1.555) −0.494 0.418 0.610 (0.184, 2.020)
45–59 −0.705 0.379 0.494 (0.103, 2.373) −0.517 0.325 0.596 (0.213, 1.669)
Educational level
Primary school or below 0.238 0.815 1.269 (0.172, 9.361) 0.615 0.360 1.850 (0.496, 6.900)
Junior high school −0.098 0.915 0.906 (0.149, 5.506) −0.070 0.905 0.932 (0.294, 2.954)
High school/Technical secondary school −0.750 0.437 0.472 (0.071, 3.133) −0.677 0.271 0.508 (0.152, 1.696)
Personal monthly income (CNY)
< 1000 1.737 0.229 5.679 (0.334, 96.500) 2.156 0.026 8.635 (1.294, 57.606)
1000–5000 0.787 0.531 2.197 (0.187, 25.796) 1.390 0.091 4.013 (0.800, 20.152)
Occupational status
Employed −0.415 0.780 0.699 (0.036, 13.421) −0.305 0.712 0.737 (0.146, 3.722)
Retired 1.092 0.231 2.982 (0.499, 17.816) 0.637 0.298 1.890 (0.570, 6.274)
Living arrangements
Living with family −2.636 0.001 0.072 (0.015, 0.332) −0.287 0.692 0.750 (0.182, 3.100)
Social participation
Occasionally/Often −3.213 0.003 0.040 (0.005, 0.337) −1.677 < 0.001 0.189 (0.086, 0.413)
Dialysis time (year)
> 5 0.860 0.240 2.364 (0.563, 9.923) 1.488 0.035 3.041 (1.083, 8.538)
1–5 −0.468 0.561 0.665 (0.168, 2.630) 0.245 0.602 1.277 (0.509, 3.206)
Post‐dialysis fatigue
Yes 1.758 0.004 5.803 (1.779, 18.923) 1.420 < 0.001 4.138 (1.992, 8.594)
Number of comorbidities
≥ 2 2.264 < 0.001 9.620 (3.400, 27.221) 1.283 < 0.001 3.609 (1.855, 7.020)
Loneliness 0.731 < 0.001 2.077 (1.681, 2.568) 0.553 < 0.001 1.739 (1.443, 2.096)
Personal mastery −0.551 < 0.001 0.577 (0.444, 0.749) −0.367 < 0.001 0.693 (0.567, 0.846)
Open in a new tab

6. Discussion

6.1. Social Isolation Is Prevalent Among MHD Patients, With Clear Heterogeneity

This study reveals that the prevalence of social isolation among MHD patients is 49.18%. LPA identified three distinct subgroups: the family‐friend dual isolation group, the friend isolation‐only group and the social network well‐being group. The largest proportion of patients was classified into the friend isolation‐only group. These results highlight the significant inter‐individual heterogeneity of social isolation in this population.

Patients in the friend isolation‐only group have high family network scores but low friend network scores. This indicates a stable family network but significant friend isolation, likely due to the impact of disease management on social activities. Dialysis‐related fatigue, discomfort and treatment schedules often limit patients' social participation—activities like outings, gatherings or community events may exceed their physical or time constraints (Zheng et al. 2025). In contrast, home‐based activities offer flexibility and require less energy. As a result, patients' life focus shifts significantly toward the family, while their friend networks gradually narrow. Healthcare providers can strengthen peer support systems, such as establishing hospital‐based patient mutual aid groups. The community health system should create a supportive community environment by providing patients with opportunities for social interaction, such as community interest groups, group health education sessions, volunteer activities and community walking events (Barragan 2021; Zheng et al. 2022).

Patients in the family‐friend dual isolation group have low scores in both family and friend networks. Physical changes in MHD patients, such as invasive catheters, arteriovenous fistula formation and limb edema, can trigger patients' concerns about body image and reduce their self‐esteem, thereby hindering their confidence to initiate and maintain intimate relationships (Sluiter, van Zwieten, et al. 2024). Healthcare providers should focus on these patients. First, family support interventions should be strengthened, such as conducting multi‐family group therapy (Lemmens et al. 2009), to enhance family members' participation. Secondly, it is crucial to establish a psychological foundation for social interaction. Group Cognitive Behavioural Therapy (CBT) can be employed to restructure patients' negative cognitions (Smith et al. 2021), such as the ‘perceived social uselessness’, while teaching them emotional regulation skills. In addition, an online healthcare community (OHC) platform can also be established for MHD patients (Yang et al. 2022), and gradually expand their friend networks.

Patients in the social network well‐being group have high scores in both family and friend networks, indicating little to no social isolation. The focus of intervention for this group of patients is to maintain their well‐functioning social networks. Patients should be guided to actively utilise their robust social networks to improve their health management. Additionally, it is necessary to conduct regular assessments of their social and mental health.

6.2. The Factors Influencing the Social Isolation Profile

6.2.1. Personal Monthly Income, Living Arrangement and Social Participation

The results showed that haemodialysis patients with a personal monthly income below CNY 1000 had a significantly higher probability of belonging to the friend isolation‐only group. In this study, most patients with low incomes were unemployed, rural farmers or relied on subsistence allowances. Although there are many middle‐aged and young patients, most ESRD patients lose their working ability, leading to a sharp drop in family income. They may lack the resources to maintain and expand their friend networks, leading to a higher risk of friend isolation. MHD patients living with family members were likely to be in the social network well‐being group. This may be because family companionship provides emotional support and practical help, such as accompanying patients to medical visits and assisting with dietary management, thereby reducing their sense of isolation and loneliness.

Furthermore, MHD patients who occasionally or regularly engaged in social activities were likely to belong to the social network well‐being group. This aligns with findings from Ejiri et al. (2019). Social participation is critical for building and maintaining social networks (Nishio et al. 2021). In this study, 77.4% of MHD patients rarely engaged in social activities, leading to prolonged isolation in closed environments. Research indicates that long‐term social disengagement deprives patients of emotional support, exacerbating loneliness, anxiety and depression, which further erode their willingness and capacity to participate socially (Liang 2024), amplifying risks of family and friend isolation.

6.2.2. Dialysis Time, Post‐Dialysis Fatigue and Comorbidities

Patients with a dialysis duration exceeding 5 years had a higher probability of belonging to the friend isolation‐only group. Prolonged dialysis exposes patients to persistent physical discomfort, complications and psychological stress, leading to chronic physical and mental exhaustion that may gradually diminish their energy and interest in maintaining friendships. Patients reporting post‐dialysis fatigue and those with two or more other comorbid chronic conditions were more likely to belong to the family‐friend dual isolation group or friend isolation‐only group. In this study, 69.5% of patients reported post‐dialysis fatigue. Research shows that this fatigue worsens after dialysis and can take hours to return to baseline, often lasting into the evening or the next day (Alvarez et al. 2020). This makes patients more likely to rest at home and less able to interact with others. Future clinical interventions should prioritise addressing post‐dialysis fatigue through strategies such as low‐to‐moderate intensity exercise, aromatherapy, acupressure, cognitive‐behavioural interventions and enhanced nutritional management (Natale et al. 2023). Older MHD patients, in particular, often have hypertension or diabetic nephropathy with multi‐system involvement and complications like anaemia, bone/joint pain, fatigue, limb edema and muscle cramps, as well as cardiovascular/cerebrovascular diseases (Zhu et al. 2024).

6.2.3. Loneliness

The results indicate that loneliness is linked to social isolation profiles in MHD patients. Higher loneliness levels increase the likelihood of family and friend isolation, which is consistent with findings of Richardson et al. (2023). The loneliness model of Hawkley and Cacioppo (2010) posits that lonely individuals perceive the social world as a more threatening place, expect more negative social interactions and retain more negative social information, thereby distancing themselves from social partners. Lieberz et al. (2021) found that individuals with high loneliness scores exhibit reduced oxytocinergic and affective responsiveness to positive conversations, along with lower interpersonal trust, thereby decreasing motivation for social interaction and promoting social avoidance behaviours. 39.67% of MHD patients in this survey reported experiencing loneliness. Healthcare providers should facilitate communication between patients and their family members, encourage families to offer patience and emotional support, and implement social network and social support interventions to mitigate loneliness risks.

6.2.4. Personal Mastery

In this study, it specifically denotes patients' perceived capacity to manage their lives and disease despite changes in life and environment resulting from haemodialysis treatment. The multivariate logistic regression analysis revealed that higher levels of personal mastery act as protective factors against family‐friend isolation and friend isolation. This may be because patients with greater personal mastery can adjust more quickly and adapt better to life and environmental changes caused by dialysis treatment (Peterson and Stunkard 1989). They also typically exhibit improved emotional states and stronger self‐efficacy (Gao et al. 2022), enabling them to actively address social pressures and uncertainties, seek solutions and avoid social withdrawal due to illness. The findings indicate that personal mastery among MHD patients in this study needs enhancement. This suggests that healthcare providers, based on considering cultural and individual differences, can implement programs such as group psychotherapy and PERMA model‐based positive psychology interventions (Michaelis et al. 2018) to strengthen the positive psychological resources of MHD patients.

6.3. Limitations

Several limitations exist in this study. First, this study employed a cross‐sectional design, which hinders the determination of causal relationships between variables. Future research could conduct longitudinal studies to explore the dynamic trajectory of social isolation and its interactions with other variables in MHD patients. Second, the sample's representativeness is limited due to the use of convenience sampling, and selection bias may be present, which could affect the generalisability of the results. Future studies should adopt multicentre designs spanning diverse regions to include a broader population of MHD patients. In addition, reliance on self‐report measures may lead to recall or reporting bias. Future research could consider utilising data triangulation to mitigate such potential biases. Finally, other factors that may influence the heterogeneity of social isolation, such as ethnicity, personality traits, quality of social relationships and living environment, should also be considered and warrant further investigation in future studies.

7. Conclusions

This study used LPA to identify three distinct social isolation profiles among MHD patients, revealing significant heterogeneity. The largest proportion of patients were classified into the ‘friend isolation‐only group’. The sociodemographic and disease‐related factors influencing these profiles included personal income, living arrangement, social participation, dialysis duration, post‐dialysis fatigue and number of comorbidities. Additionally, the study emphasised the importance of loneliness and personal mastery in MHD patients as factors affecting social isolation profiles. The findings of this study may contribute to the development of targeted intervention programs to address social isolation. Interventions should focus on establishing collaboration among families, hospitals and communities to optimise effectiveness.

Author Contributions

Fangli Liu conceptualised the study. Fangli Liu and Luwen Zhang designed the study. Luwen Zhang, Jinghui Liu and Xia Zheng collected the data. Luwen Zhang and Fangli Liu performed statistical calculations and drafted the initial manuscript. All authors contributed to data acquisition, result interpretation and critical revision of the manuscript. The authors read and approved the final manuscript.

Funding

This work was supported by Henan Provincial Department of Science and Technology Key project (Grant No. 242102310252; 252102311067), Henan University Youth Cross Fund General (Grant No. S23060Y) and Postgraduate Education Reform and Quality Improvement Project of Henan Province (Grant No. YJS2025XQLH33).

Ethics Statement

This study was approved by the Institutional Ethics Committee of Henan University (Approval No. HUSOM2025‐551).

Consent

All patients voluntarily participated and provided signed informed consent forms.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgements

We extend our sincere appreciation to all members and funding bodies that contributed to this research.

Data Availability Statement

The data generated during the current study are available from the corresponding author on reasonable request.

References

  1. Alvarez, L. , Brown D., Hu D., Chertow G. M., Vassalotti J. A., and Prichard S.. 2020. “Intradialytic Symptoms and Recovery Time in Patients on Thrice‐Weekly In‐Center Hemodialysis: A Cross‐Sectional Online Survey.” Kidney Medicine 2, no. 2: 125–130. 10.1016/j.xkme.2019.10.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barragan, C. 2021. “Social Relationships and the Importance of Community‐Based Fitness Programs (CBFP).” Journal of Women & Aging 33, no. 4: 428–441. 10.1080/08952841.2021.1915685. [DOI] [PubMed] [Google Scholar]
  3. Bekhet, A. K. , Zauszniewski J. A., and Nakhla W. E.. 2008. “Loneliness: A Concept Analysis.” Nursing Forum 43, no. 4: 207–213. 10.1111/j.1744-6198.2008.00114.x. [DOI] [PubMed] [Google Scholar]
  4. Beller, J. , and Wagner A.. 2018. “Loneliness, Social Isolation, Their Synergistic Interaction, and Mortality.” Health Psychology 37, no. 9: 808–813. 10.1037/hea0000605. [DOI] [PubMed] [Google Scholar]
  5. Diao, K. , Wang J., Zhang Y., Huang Y., and Shan Y.. 2024. “The Mediating Effect of Personal Mastery and Perceived Social Support Between Emotional Intelligence and Social Alienation Among Patients Receiving Peritoneal Dialysis.” Frontiers in Public Health 12: 1392224. 10.3389/fpubh.2024.1392224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Ejiri, M. , Kawai H., Fujiwara Y., et al. 2019. “Social Participation Reduces Isolation Among Japanese Older People in Urban Area: A 3‐Year Longitudinal Study.” PLoS One 14, no. 9: e0222887. 10.1371/journal.pone.0222887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Fateh, A. , Mustamil N., and Shahzad F.. 2021. “Role of Authentic Leadership and Personal Mastery in Predicting Employee Creative Behavior: A Self‐Determination Perspective.” Frontiers of Business Research in China 15, no. 1: 3. 10.1186/s11782-021-00100-1. [DOI] [Google Scholar]
  8. Ferguson, S. L. , Moore E. W. G., and Hull D. M.. 2020. “Finding Latent Groups in Observed Data: A Primer on Latent Profile Analysis in Mplus for Applied Researchers.” International Journal of Behavioral Development 44, no. 5: 458–468. 10.1177/0165025419881721. [DOI] [Google Scholar]
  9. Friedrich, N. K. , Schmitt N., Hruby H., and Kugler C.. 2025. “Self‐Management Interventions for Adult Haemodialysis Patients: A Scoping Review of Randomized Controlled Trials.” BMC Nephrology 26, no. 1: 285. 10.1186/s12882-025-04229-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gao, Y. , Shan Y., Wang S., et al. 2022. “Decisional Control Preference Among Chinese Patients With Advanced Kidney Disease: The Role of Mastery, Perceived Social Support and Self‐Efficacy.” Psychology, Health & Medicine 27, no. 10: 2126–2137. 10.1080/13548506.2021.1952282. [DOI] [PubMed] [Google Scholar]
  11. Hanna, K. , Cross J., Nicholls A., and Gallegos D.. 2023. “The Association Between Loneliness or Social Isolation and Food and Eating Behaviours: A Scoping Review.” Appetite 191: 107051. 10.1016/j.appet.2023.107051. [DOI] [PubMed] [Google Scholar]
  12. Hawkley, L. C. , and Cacioppo J. T.. 2010. “Loneliness Matters: A Theoretical and Empirical Review of Consequences and Mechanisms.” Annals of Behavioral Medicine 40, no. 2: 218–227. 10.1007/s12160-010-9210-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Herbolsheimer, F. , Ungar N., and Peter R.. 2018. “Why Is Social Isolation Among Older Adults Associated With Depressive Symptoms? The Mediating Role of Out‐of‐Home Physical Activity.” International Journal of Behavioral Medicine 25, no. 6: 649–657. 10.1007/s12529-018-9752-x. [DOI] [PubMed] [Google Scholar]
  14. Jeong, D. , Choi Y. J., and Sok S.. 2025. “Experience of Loneliness Among Middle‐Aged Hemodialysis Patients: Qualitative Study.” Journal of Nursing Management 2025: 1013725. 10.1155/jonm/1013725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Johansen, K. L. , Gilbertson D. T., Li S., et al. 2024. “US Renal Data System 2023 Annual Data Report: Epidemiology of Kidney Disease in the United States.” American Journal of Kidney Diseases 83: A8–A13. 10.1053/j.ajkd.2024.01.001. [DOI] [PubMed] [Google Scholar]
  16. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group . 2024. “KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease.” Kidney International 105, no. 4s: S117–s314. 10.1016/j.kint.2023.10.018. [DOI] [PubMed] [Google Scholar]
  17. Kongsted, A. , and Nielsen A. M.. 2017. “Latent Class Analysis in Health Research.” Journal of Physiotherapy 63, no. 1: 55–58. 10.1016/j.jphys.2016.05.018. [DOI] [PubMed] [Google Scholar]
  18. Lemmens, G. M. , Eisler I., Buysse A., Heene E., and Demyttenaere K.. 2009. “The Effects on Mood of Adjunctive Single‐Family and Multi‐Family Group Therapy in the Treatment of Hospitalized Patients With Major Depression. A 15‐Month Follow‐Up Study.” Psychotherapy and Psychosomatics 78, no. 2: 98–105. 10.1159/000201935. [DOI] [PubMed] [Google Scholar]
  19. Li, K. , Tang F., Albert S. M., Rauktis M. E., and Ohmer M. L.. 2024. “Social Isolation, Loneliness, and Depressive Symptoms Among Older Adults: The Moderating Effect of Resilience.” Gerontologist 64, no. 7: gnae056. 10.1093/geront/gnae056. [DOI] [PubMed] [Google Scholar]
  20. Liang, L. Y. 2024. “The Impact of Social Participation on the Quality of Life Among Older Adults in China: A Chain Mediation Analysis of Loneliness, Depression, and Anxiety.” Frontiers in Public Health 12: 1473657. 10.3389/fpubh.2024.1473657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lieberz, J. , Shamay‐Tsoory S. G., Saporta N., et al. 2021. “Loneliness and the Social Brain: How Perceived Social Isolation Impairs Human Interactions.” Advanced Science 8, no. 21: e2102076. 10.1002/advs.202102076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lubben, J. , Blozik E., Gillmann G., et al. 2006. “Performance of an Abbreviated Version of the Lubben Social Network Scale Among Three European Community‐Dwelling Older Adult Populations.” Gerontologist 46, no. 4: 503–513. 10.1093/geront/46.4.503. [DOI] [PubMed] [Google Scholar]
  23. Michaelis, R. , Niedermann C., Reuber M., Kuthe M., and Berger B.. 2018. “‘Seizures Have Become a Means of Somehow Learning Things About Myself’—A Qualitative Study of the Development of Self‐Efficacy and Mastery During a Psychotherapeutic Intervention for People With Epilepsy.” Epilepsy & Behavior 84: 152–161. 10.1016/j.yebeh.2018.04.019. [DOI] [PubMed] [Google Scholar]
  24. Naldi, L. , and Cazzaniga S.. 2020. “Research Techniques Made Simple: Latent Class Analysis.” Journal of Investigative Dermatology 140, no. 9: 1676–1680.e1671. 10.1016/j.jid.2020.05.079. [DOI] [PubMed] [Google Scholar]
  25. Natale, P. , Ju A., Strippoli G. F., et al. 2023. “Interventions for Fatigue in People With Kidney Failure Requiring Dialysis.” Cochrane Database of Systematic Reviews 8, no. 8: Cd013074. 10.1002/14651858.CD013074.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nicholson, N. R., Jr. 2009. “Social Isolation in Older Adults: An Evolutionary Concept Analysis.” Journal of Advanced Nursing 65, no. 6: 1342–1352. 10.1111/j.1365-2648.2008.04959.x. [DOI] [PubMed] [Google Scholar]
  27. Nishio, M. , Takagi D., Shinozaki T., and Kondo N.. 2021. “Community Social Networks, Individual Social Participation and Dietary Behavior Among Older Japanese Adults: Examining Mediation Using Nonlinear Structural Equation Models for Three‐Wave Longitudinal Data.” Preventive Medicine 149: 106613. 10.1016/j.ypmed.2021.106613. [DOI] [PubMed] [Google Scholar]
  28. Pearlin, L. I. , and Schooler C.. 1978. “The Structure of Coping.” Journal of Health and Social Behavior 19, no. 1: 2–21. [PubMed] [Google Scholar]
  29. Peterson, C. , and Stunkard A. J.. 1989. “Personal Control and Health Promotion.” Social Science & Medicine 28, no. 8: 819–828. 10.1016/0277-9536(89)90111-1. [DOI] [PubMed] [Google Scholar]
  30. Richardson, A. , König H. H., and Hajek A.. 2023. “Volunteering, Loneliness and Perceived Social Isolation: Evidence From a Representative Sample of Middle‐Aged and Older Adults in Germany.” Aging & Mental Health 27, no. 11: 2289–2294. 10.1080/13607863.2023.2222676. [DOI] [PubMed] [Google Scholar]
  31. Sluiter, A. , Cazzolli R., Jaure A., et al. 2024. “Experiences of Social Isolation and Loneliness in Chronic Kidney Disease: A Secondary Qualitative Analysis.” Clinical Journal of the American Society of Nephrology 19, no. 11: 1405–1416. 10.2215/CJN.0000000000000529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sluiter, A. , van Zwieten A., Shen J. I., and Manera K.. 2024. “Measuring Social Functioning in Chronic Kidney Disease.” Seminars in Nephrology 44, no. 3–4: 151547. 10.1016/j.semnephrol.2024.151547. [DOI] [PubMed] [Google Scholar]
  33. Smith, R. , Wuthrich V., Johnco C., and Belcher J.. 2021. “Effect of Group Cognitive Behavioural Therapy on Loneliness in a Community Sample of Older Adults: A Secondary Analysis of a Randomized Controlled Trial.” Clinical Gerontologist 44, no. 4: 439–449. 10.1080/07317115.2020.1836105. [DOI] [PubMed] [Google Scholar]
  34. Torreggiani, M. , Piccoli G. B., Moio M. R., et al. 2023. “Choice of the Dialysis Modality: Practical Considerations.” Journal of Clinical Medicine 12, no. 9: 3328. 10.3390/jcm12093328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Xiao, R. , and Du J. W.. 2023. “Reliability and Validity of the 6‐Item UCLA Loneliness Scale (ULS‐6) for Application in Adults.” Journal of Southern Medical University 43, no. 6: 900–905. 10.12122/j.issn.1673-4254.2023.06.04. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Yang, J. X. , Luo B. A., and Tsai F. S.. 2022. “Continuance Intention of Online Healthcare Communities: The Mediation Mechanism of Social Interaction Ties.” Journal of Organizational and End User Computing 34, no. 7: 1–25. 10.4018/JOEUC.302892. [DOI] [Google Scholar]
  37. Yu, Y. , Huang Q., Ren Z., and Ou Z.. 2024. “The Association Between Social Isolation and Medication Adherence Among Chinese Older Adults With Chronic Diseases: Serial Mediation of Social Support and Loneliness.” Annals of Behavioral Medicine 58, no. 10: 670–678. 10.1093/abm/kaae049. [DOI] [PubMed] [Google Scholar]
  38. Zheng, X. , Yang Z., Xu L., and Wang A.. 2025. “Insights Into the Complexities of Symptom Management for Hemodialysis Patients: A Systematic Review of Qualitative Studies.” Health Psychology Review 19: 1–44. 10.1080/17437199.2025.2496230. [DOI] [PubMed] [Google Scholar]
  39. Zheng, Z. , Liu W., Lu Y., Sun N., Chu Y., and Chen H.. 2022. “The Influence Mechanism of Community‐Built Environment on the Health of Older Adults: From the Perspective of Low‐Income Groups.” BMC Geriatrics 22, no. 1: 590. 10.1186/s12877-022-03278-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zhou, W. , Li Y., Ning Y., et al. 2021. “Social Isolation Is Associated With Rapid Kidney Function Decline and the Development of Chronic Kidney Diseases in Middle‐Aged and Elderly Adults: Findings From the China Health and Retirement Longitudinal Study (CHARLS).” Frontiers in Medicine 8: 782624. 10.3389/fmed.2021.782624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhu, B. , Wu H., Lv S., and Xu Y.. 2024. “Association Between Illness Perception and Social Alienation Among Maintenance Hemodialysis Patients: The Mediating Role of Fear of Progression.” PLoS One 19, no. 4: e0301666. 10.1371/journal.pone.0301666. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data generated during the current study are available from the corresponding author on reasonable request.

Articles from Journal of Advanced Nursing are provided here courtesy of Wiley

RESOURCES