Introduction

With advances in antiretroviral therapy (ART) and the continuous optimization of treatment programs, AIDS has been transformed from a severe, life-threatening condition into a manageable chronic infectious disease. The life expectancy of people living with HIV (PLWH) has gradually approached that of the general population1,2. However, the extension of survival time does not necessarily correlate with an improvement in quality of life (QOL). PLWH face multiple challenges throughout the long course of managing their condition. In the 21st century, the medical community emphasizes that the ultimate goal of HIV treatment and management should be to enhance the QOL of PLWH globally.

WHO defines QOL as an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards, and concerns3. This broad concept includes physical health, mental state, independence, social relationships, and an individual’s beliefs and relationships with their surroundings. Therefore, UNAIDS has been working to improve the QOL of PLWH in their whole life cycle as one of its priority actions4.

Over the past 40 years, researchers worldwide have actively explored methods to enhance the QOL of PLWH. They have examined and identified various factors associated with QOL, including sociodemographic characteristics5 disease-related factors6 related symptoms7 and behavioral health factors8. Among these, sociodemographic characteristics—such as gender, race, economic status, education, and marital status—play a significant role. The Self-regulatory HIV/AIDS Symptom Management Model (SSMM-HIV) is a classical theory that offers a comprehensive symptom management framework. Sociodemographic characteristics influencing factors, although not a central focus in the SSMM-HIV, shape illness representations, symptom experiences, symptom management, adherence to ART, and clinical outcomes, including QOL9. These factors have a variety of influences on the central concepts in the model. In Ahmed’s study10 female gender, older age (> 50 years), low education, married and widowed patients, and unemployed patients were the factors that could predict poor QOL. Martin and his colleagues present that income and employment impact QOL significantly11. Musumaria12 also reveals that perceived financial and employment status were associated with QOL. The female gender was a predictor of the decline in QOL. In addition, some studies reported that racial discrimination was significantly associated with reduced quality of life13,14.

Despite existing research on sociodemographic factors among PLWH and their association with QOL, particularly concerning marital status15,16 there have been a lack of studies investigating other influential factors, such as household structures among PLWH. As modernization progresses and urbanization increases, marital status may not accurately reflect the social conditions of an individual’s living environment. The household environment encompasses various living conditions that can significantly impact health and well-being. These factors may include physical aspects, such as housing quality, as well as social circumstances17 such as living alone, which can influence the development of intervention strategies for social support systems and provide new insights for future research on social support. Therefore, this study focuses on the household structures in PLWH and aims to: (1) examine the differences in QOL among PLWH living in rural and urban areas; and (2) explore the rural-urban relationships and disparity between the number of inhabitants, types of residential populations, and marital status concerning QOL in PLWH.

Methods

Study design and settings

A cross-sectional descriptive design was employed in this multicenter study. Based on the latest analysis of the spatial patterns and temporal evolution of AIDS18 the research was conducted across nine provinces or municipalities with varying incidence levels. This included three high-incidence areas: Xinjiang, Guangxi, and Sichuan Province; three regions with slightly elevated incidence: Beijing, Hunan, and Guangdong Province; and two moderate-incidence areas: Qinghai Province and Henan Province. Additionally, Shenyang City, located in northeastern Liaoning Province, was classified as a low-incidence area19,20.

Sample collection

The investigation was conducted over a total period of 4 months, from November 2021 to January 2022, and from April 2022 to May 2022. We recruited PLWH from nine designated HIV/AIDS medical institutions through quota sampling based on gender, age, and residential areas. Eligibility for participation required that PLWH meet the following criteria: (1) HIV infection diagnosed according to the Chinese AIDS Diagnosis and Treatment Guidelines (2021 edition), (2) at least 18 years old, (3) inclusion of both new and previously treated patients from the inpatient and outpatient departments of the institutions, and (4) provision of informed consent to participate in this study. Participants diagnosed with severe conditions that would impede their ability to complete the survey were excluded. All participants provided online informed consent before responding to the survey questions.

Measures

Sociodemographic and clinical data

The sociodemographic variables included age, gender, race, education, marital status, employment, income, residential population structure, and residential areas. These data were collected through an online questionnaire. The clinical variables were HIV-positive duration, ART treatment, CD4+ T cell count, disease stage, and comorbidities. The cut-off points of categorical variables were defined based on WHO and China’s guidelines21.

QOL score

We used the WHOQOL-HIV BREF (World Health Organization Quality Of Life Questionnaire For HIV Brief Version) in our study to collect the QOL of PLWH. The WHOQOL-HIV BREF included 31 items and was developed by WHO22. A five-point Likert scale rated all items. Among 31 items, 29 domain-specific items are used to measure individual QOL across six domains. The other two items focus on participants’ perceptions of their general QOL and health status. Every domain score ranged from 4 to 20. Higher scores in each domain indicated higher QOL for that domain. Psychometric properties of a Chinese version among PLWH have been proven to be reliable and valid, with Cronbach’s α of 0.93, and the test-retest reliability revealed a statistically significant intraclass correlation coefficient of 0.72–0.82 (P < 0.001)23. We also used this scale in another study24.

Ethics statement

This study was approved by the Institutional Review Board of Beijing Ditan Hospital, Capital Medical University, where the study was conducted, and the reference number is NO. DTEC-KY2021-015-01. All participants signed written informed consent. All methods were performed following the relevant guidelines and regulations (Declaration of Helsinki).

Statistical analysis

Data analysis was conducted in the software package R (http://www.R-project.org, The R Foundation), and Free Statistics software version 1.7 was used to perform all statistical analyses.

We used descriptive statistics to summarize the characteristics of the study participants. Based on the normality of data distribution, continuous variables were presented as the deviation/median ± interquartile range and compared between groups using the t-test/Wilcoxon rank-sum test. Meanwhile, Categorical variables were presented as percentages and compared using the χ2 test. The Kruskal-Wallis test, one-way analysis of variance, was applied to assess the significance of group differences. The Sociodemographic, clinical, and household structures associated with QOL were examined using multivariable linear regression. A P-value less than 0.05 was considered statistically significant.

Results

Participant characteristics

Among the 711 participants, the mean age was 39.22 ± 11.60 years (range 18–74). Of these, 64.14% resided in urban areas. Over two-thirds (72.15%) were male, and 84.95% identified as Han Chinese. Additionally, 59.22% had attained at least a high school education or its equivalent. Nearly half of the participants (48.80%) were single, and 34.18% reported a monthly income of ¥3,000 or less. Employment status revealed that 38.12% were employed, while 56.26% lived with their families. Almost 90% of participants received ART, and one-third (34.74%) had been living with HIV for over 6 years. Furthermore, 77.78% reported a CD4 + T cell count of fewer than 200 copies/ml, and 74.68% were asymptomatic. Finally, 81.29% indicated that they had no comorbidities (see Table 1).

Table 1 General characteristics of study participants.
Full size table

In this study, the average age of PLWH in urban areas was found to be younger than that of their rural counterparts (P < 0.01). Additionally, the educational level of PLWH in urban settings was generally higher than that of those in rural areas. Furthermore, statistically significant differences were observed between the urban and rural groups in terms of income, marital status, and employment (P < 0.01). Clinical characteristics, such as disease staging, CD4 + T cell count, and comorbidities, also exhibited significant differences (P < 0.01). However, no significant differences were found in the number of inhabitants (χ² = 7.254, P = 0.06), type of residential population (χ² = 6.074, P = 0.05), HIV duration (χ² = 3.435, P = 0.33), or ART treatment (χ² = 0.715, P = 0.4) (see Table 1).

Comparison of differences in QOL among PLWH in different household environments

Overall, the mean QOL among PLWH in China was 84.90 ± 15.93. The mean QOL was comparable between the urban and rural groups (t = 1.405, P = 0.236). No statistically significant differences were observed in the other five domains, except for physical health (PH) (Table 2). The QOL stratified by marital status is presented in Table 2. Additionally, no statistically significant differences were detected in the overall QOL among individuals who were single, married, cohabiting, or in other marital statuses (F = 1.468, P = 0.231). However, five of the six domains exhibited similar levels across different marital statuses, with the exception of social participation (SP) (F = 3.999, P = 0.019), where the married or cohabiting group reported the lowest QOL (12.34 ± 3.70 compared to 12.88 ± 3.92 and 13.56 ± 4.50).

Table 2 Differences in QOLQOL among PLWH in different residential areas and marital status.
Full size table

Table 3; Fig. 1 revealed significant differences in the mean of QOL among the living alone group, living with a family group, and living with friends/others group (F = 4.295, P < 0.01). PLWH living alone reported the highest QOL (87.32 ± 17.82 versus 84.01 ± 15.11 versus 82.41 ± 13.33). In the sub-domains analysis, there was no statistical difference among the three groups in the PH domain (F = 1.285, P = 0.28) and SP domain (F = 0.613, P  = 0.54).

Table 3 Differences in QOL among PLWH in different residential population structure.
Full size table
Fig. 1
figure 1

Differences in QOL among PLWH living with different type of residential population.

Full size image

After stratification according to the number of inhabitants, the QOL of each group was compared and analyzed. A statistically significant difference was examined in groups with several inhabitants (F = 6.926, P < 0.01). PLWH living only oneself compared with other groups reported the highest QOL (89.14 ± 18.42 versus 85.45 ± 15.81 versus 82.62 ± 14.22 versus 82.53 ± 14.20), and similar results were presented in the sub-domains analysis (Table 3; Fig. 2).

Fig. 2
figure 2

Differences in QOL among PLWH living with different number of inhabitants.

Full size image

Following stratification based on the number of inhabitants, the QOL for each group was compared and analyzed. A statistically significant difference was observed among groups with varying numbers of inhabitants (F = 6.926, P < 0.01). Individuals living alone reported the highest QOL compared to other groups, with scores of 89.14 ± 18.42, 85.45 ± 15.81, 82.62 ± 14.22, and 82.53 ± 14.20, respectively. Similar results were noted in sub-domain analysis (see Table 3; Fig. 2).

Association between the number of inhabitants and QOL among PLWH

In full multivariable models, more inhabitants were negatively associated with the QOL among PLWH. In the unadjusted model, participants living with three individuals experienced a decrease in QOL by 6.52 points (β = – 6.52 [95% CI, – 9.78 to – 3.26]), while those living with four or more individuals had a decrease of 6.61 points (β = – 6.61 [95% CI, – 9.90 to – 3.32]) (P < 0.001) (see Table 4). After adjusting for confounding sociodemographic factors presented in Table 4, the coefficients were − 6.83 (three-person group) and − 7.43 (four-person or more group) (P = 0.002 versus 0.001). These results were consistent in Model II, which adjusted for all confounding factors, including sociodemographic characteristics and clinical aspects; the coefficients were − 5.71 (three-person group) and − 6.49 (four-person or more group) (P = 0.009 versus 0.004). No associations were observed between fewer inhabitants (living with only one or two individuals) and QOL.

Table 4 Association between number of inhabitants and QOL in multiple regression model.
Full size table

Table 5 presents stratified analyses of the association between the number of inhabitants and QOL in urban and rural populations. After full adjustment, rural residents exhibited a stronger inverse association between household size and QOL (β = – 3.89, 95% CI: – 6.03 to – 1.75; P = 0.0004) compared to urban counterparts (β = – 1.55, 95% CI: – 2.97 to – 0.14; P = 0.032). Despite this divergence in effect magnitude, the interaction term between residential setting and household size was nonsignificant (P for interaction = 0.178).

Table 5 Subgroup analysis between urban and rural group.
Full size table

Table 6 delineates differential patterns of QOL determinants across urban and rural subgroups. In rural areas, cohabiting with two or three household members was associated with clinically meaningful QOL reductions (two persons: β = – 10.83, P = 0.001; three persons: β = – 9.90, P = 0.001), whereas urban residents showed significant QOL declines only in households with four or more inhabitants (β = – 4.68, P = 0.020). Employment status demonstrated contrasting effects: rural employed individuals reported substantially higher QOL than students (β = 17.20, P = 0.021), exceeding the urban employment benefit (β = 7.42, P = 0.010). Comorbidity burden exerted more substantial detrimental effects in rural settings (β = – 14.02 vs. urban β = – 8.47; both p < 0.001). Stratified analyses revealed that PLWH with monthly household incomes 3,000 ~ 6000 RMB had significantly higher QOL scores than those earning < 3,000 RMB, with a more pronounced effect in rural areas (β = 7.10 for 3,000 ~ 6,000 RMB, P = 0.004) compared to urban settings (β = 4.45, P = 0.009).

Table 6 Association between the number of inhabitants and QOL in urban and rural group.
Full size table

Discussion

Our study investigated the relationship between household structures and the QOL in a sample of PLWH across nine regions in China. Several findings are both exciting and noteworthy. Firstly, significant differences were observed between urban and rural groups in terms of sociodemographic and clinical characteristics. The general education levels and income of the urban group were found to be higher than those of the rural group, aligning with the theoretical framework regarding the construction of the urban-rural dual structure in China25,26,27. Although previous policies limited healthcare access for the rural population, resulting in inadequate HIV care services in rural areas28the initiation of the “Four Free, One Care” policy has enabled an increasing number of rural residents to access medical support by providing free ART for rural populations29.

This study also demonstrated that urban and rural PLWH have equal access to ART, with no statistically significant differences identified. However, PLWH from rural areas appeared to be at a more advanced stage of disease, exhibiting lower CD4 + T cell counts and a higher prevalence of comorbidities than their urban counterparts. This finding suggests that PLWH in rural areas may lack the knowledge and resources necessary for effective self-care, adherence to standardized ART, and management of complications. Researchers such as Wen30 and Yu31 have identified rural areas as a critical gap in HIV/AIDS control, highlighting the need for targeted interventions for PLWH living in rural communities with lower educational levels.

Some previous research reported that PLWH experienced lower QOL than the general population32,33,34. In our study, we found that PLWH had an average or slightly higher QOL score in China than twenty years before, which could be attributed to the advances in ART and the care and support provided by society29. In addition, PLWH living in Urban areas have similar QOL scores to their rural counterparts in this investigation, which was consistent with other studies. Wardojo and colleagues35 presented that living in an urban area was associated with a better QOL by multivariate linear regression analysis. One possible reason is that our study was conducted more than three years after the previous survey, and many efforts have been made in recent years to allocate resources for HIV prevention and control in poor areas under the support of national policies by our government, medical institutions, and social organizations. Therefore, the gap between the urban and the rural was getting smaller. Huang36 investigated 410 HIV-positive individuals and found that rural household registration was negatively associated with physical health summary score (PHS), one of the domains in QOL. Although our study showed no difference in overall QOL between urban and rural groups, PLWH living in rural areas had lower PH scores than urban groups, consistent with Huang’s results.

Previous studies examining the relationship between marital status and QOL among PLWH have yielded inconsistent results. A survey conducted between 2016 and 2017 involving 364 PLWH in Iran indicated that married individuals reported a higher overall QOL compared to their single counterparts16. This discrepancy may be attributed to married individuals having greater opportunities for social interaction with at least one partner, which serves as a source of social support. Conversely, a cohort study involving 763,137 PLWH in the United States15 demonstrated that marital status significantly influenced HIV mortality, revealing that divorced or separated individuals experienced a mortality rate 4.3 times higher than that of married individuals, who tended to live longer and healthier lives. However, Medeiros37 found no evidence that marital status was an independent factor influencing QOL among PLWH, a finding that aligns with the results of the current study. Additionally, Ahmed10 reported that married or cohabiting individuals exhibited a lower QOL, possibly due to feelings of self-blame related to mutual infection and transmission of the virus between partners, as well as fears of potential future loss of a partner. In contrast, single individuals generally reported a better overall QOL.

Due to China’s rapid industrialization over the past two decades, many rural residents have migrated to urban areas, yet they frequently return to visit their home villages38. Consequently, marital status alone cannot adequately explain the domestic family structure or ecological environment in recent years, especially in light of urban modernization, job mobility, and the complexities of marital relationships, particularly for PLWH. As a result, their family structure and marital status may be more intricate than previously understood. Furthermore, as Medeiros observed, many PLWH lead double lives; they may be married yet experience solitude for various reasons, or they may not have a legally recognized marital status while still being in a relationship. Therefore, in contemporary society, marital status does not fully capture the living conditions faced by modern individuals. Considering these changes, we incorporated additional factors related to the household environment into our regression model for analysis, including the type of residential population and the number of inhabitants. Our findings indicate that PLWH living alone reported a better QOL. In contrast, those with more than three cohabitants experienced a significant decline in their QOL scores, providing a foundation for further research.

Our stratified analyses reveal significant urban-rural disparities in how household environments influence the QOL among Chinese PLWH, thereby extending prior population-level findings by identifying context-dependent risk profiles. Three key patterns emerge from these results, each carrying implications for targeted interventions. First, the pronounced inverse association between household size and QOL in rural settings (β = -3.89 vs. urban β = -1.55) suggests that overcrowding may disproportionately burden rural PLWH. This observation aligns with evidence that rural households in China often face compounded stressors, such as limited privacy, multigenerational caregiving obligations, and reduced access to community-based HIV support services28,39. The lack of significant interaction (P = 0.178) despite divergent effect sizes implies that residential context modifies the magnitude rather than the direction of this relationship. This nuance has been overlooked in previous national surveys. Second, the differential socioeconomic impacts underscore rural-urban inequities within HIV care ecosystems. The 2.3-fold greater QOL benefit of employment among rural PLWH (β = 17.20 vs. urban β = 7.42) likely reflects the fragmented social safety nets in rural China40where formal employment may uniquely confer access to health insurance and reduce stigma. Conversely, the stronger comorbidity burden among rural residents (β = -14.02 vs. urban β = -8.47) mirrors documented gaps in the integration of primary care for HIV and chronic disease co-management in rural areas. Third, the threshold effects of income increments reveal environment-mediated economic mechanisms. Rural PLWH achieve clinically meaningful QOL gains at lower income thresholds (3,000 ~ 6,000 RMB: β = 7.10) compared to their urban counterparts, potentially because marginal income improvements in rural areas disproportionately enhance housing quality and medical affordability. This finding challenges the assumption that economic interventions should prioritize urban HIV hotspots and advocates for poverty-alleviation strategies tailored to the cost-of-living realities in rural settings41. Methodologically, our use of interaction testing and stratified beta coefficients advances beyond conventional subgroup comparisons by quantifying effect modification gradients. While the nonsignificant interaction terms caution against overinterpreting urban-rural differences for certain variables (e.g., household size), the clinically meaningful effect size disparities for employment and comorbidities warrant attention despite their statistical non-significance—a critical distinction often neglected in observational HIV research. Applying HIV-specific Minimal Clinically Important Difference thresholds (MCID), our observed physical domain differences (Δ = 5.4 points) represent 75% of the established 7.2-point benchmark. While sub-threshold cross-sectionally, longitudinal environmental exposures may cumulatively exceed clinical relevance—a hypothesis supported by housing intervention trials demonstrating 2.3-point annual QOL gains42.

Limitations

The strengths of our survey include the diversity of the study participants and the multiple centralizations. However, our study has several limitations. The data were collected by self-report, which is subject to recall bias. Another limitation is that almost all participants in this study were recruited in designated HIV/AIDS medical institutions located in cities by convenience sampling. Therefore, the results may be slightly overestimated or underestimated compared to real situations. Further studies are needed to assess the external validity of our results in various settings. In addition, we primarily focused on household structures in the current research, and future studies should explore house quality, biological environment, and psychosocial environment. The absence of granular metrics related to housing quality, such as sanitation facilities and ventilation, may lead to an underestimation of environmental effects [15]. Additionally, unmeasured confounding factors arising from internal migration patterns could significantly influence urban-rural comparisons. Future longitudinal studies should investigate whether urbanization initiatives moderate these associations over time.

Conclusion

In summary, the results demonstrate that the type of residential population and the number of inhabitants can have a negative association with QOL among PLWH in our study in China. On the other hand, PLWH living alone or with only one person have better well-being and QOL. Additionally, our analysis reveals significant urban-rural disparities in how household environments influence the QOL among Chinese PLWH, extending prior population-level findings by identifying context-dependent risk profiles. The differential socioeconomic impacts underscore rural-urban inequities within HIV care ecosystems. Therefore, future studies should explore interpersonal space in PLWH by profound design based on the present research, which can inject new vigor and inspiration into the future social support for PLWH. This finding will encourage healthcare providers, social workers, case managers, and other relevant professionals to focus on assessing household size and other pertinent factors when providing care, mental health, and social support to PLWH, based on the disparity between rural and urban areas.