The impact of social support on pain intensity in older adults with chronic pain: the serial mediating roles of positive and negative affect and pain catastrophizing

Abstract

Background

Chronic pain in older adults represents a major public health concern, with non-pharmacological interventions gaining importance as a key strategy for pain management. Social support, positive and negative affect, and pain catastrophizing are known to be associated with pain intensity in older adults. This study aimed to investigate the effect of social support on pain intensity in older adults with chronic pain and to explore the mediating roles of positive affect, negative affect, and pain catastrophizing.

Methods

This cross-sectional study involved 363 older adults with chronic pain who completed the Social Support Rating Scale (SSRS), Positive and Negative Affect Scale (PANAS), Pain Catastrophizing Scale (PCS), and Numeric Rating Scale (NRS). Serial mediation analysis was conducted to test whether positive and negative affect and pain catastrophizing act sequentially as mediators between social support and pain intensity. Structural equation modeling (SEM) was conducted using Mplus 8.3 to examine whether positive and negative affect and pain catastrophizing serially mediate the association between social support and pain intensity.

Results

Social support was negatively associated with pain intensity, both directly and indirectly, through the separate mediating effects of positive affect (β = -0.129, 95% CI: -0.176, -0.082) and pain catastrophizing (β = -0.199, 95% CI: -0.262, -0.136). The sum of these indirect effects accounted for 49.84% of the total effect. Furthermore, social support was indirectly associated with reduced pain intensity via the serial mediation of negative affect and pain catastrophizing (β = -0.088, 95% CI: -0.131, -0.059) and the serial mediation of positive affect and pain catastrophizing (β = -0.066, 95% CI: -0.099, -0.043). These pathways collectively accounted for 23.40% of the total effect.

Conclusion

Healthcare providers (e.g., nurses, physicians, and therapists) should enhance chronic pain management by assessing patients’ social support levels and implementing interventions focused on emotional regulation and cognitive restructuring. An interdisciplinary approach is recommended to effectively reduce pain intensity in older adults.

Trial registration

Not applicable.

Introduction

Chronic pain, defined as pain that persists or recurs for more than three months, constitutes a complex biopsychosocial health issue that extends beyond a mere somatic symptom [1]. Globally, chronic pain affects approximately 20%–30% of adults, with a higher prevalence among older adults [2–5]. The adverse effects of chronic pain are multifaceted, leading to declines in physical functioning.(e.g., reduced physical activity, mobility impairments, and frailty), an increased risk of falls, sleep disturbances, and significantly elevated rates of depression and anxiety [6, 7]. In China, the rapidly aging population has made chronic pain a pressing public health issue that demands urgent attention.

Pain intensity is a key indicator of functional status in older adults with chronic pain, reflecting the severity of their pain experience [8]. It is typically measured using tools such as the Numerical Rating Scale (NRS) or the Visual Analog Scale (VAS) [9]. Studies show that greater pain intensity correlates with increased disability risk in older adults with chronic pain [10, 11]. Compared to younger populations, older adults are more prone to chronic pain, often accompanied by multiple comorbidities, which complicates pain management [12]. Additionally, older patients may face limitations in the use of analgesic medications (e.g., opioids) due to cognitive decline or increased sensitivity to drugs [13]. Consequently, non-pharmacological interventions, particularly those targeting psychosocial factors, have become a priority in chronic pain research for older adults.

The biopsychosocial model is a widely recognized framework in pain research, positing that pain perception is modulated by psychosocial factors beyond neurophysiological signals [1, 14]. Social support is believed to alleviate pain through multiple mechanisms [15]. It encompasses emotional, informational, and instrumental assistance provided by an individual’s social network [16]. Evidence suggests that higher levels of social support are associated with lower pain intensity and improved pain management outcomes [17, 18]. Social support is thought to influence pain intensity by reducing stress, shaping cognitive appraisals, and enhancing coping strategies [19]. However, most studies on social support and pain-related outcomes in older adults have focused on direct associations, often overlooking the underlying mediating processes [20].

In this study, positive and negative affect refer to individuals’ general tendencies to experience pleasant or unpleasant emotional states, respectively. Positive and negative affect represent independent psychological dimensions: positive affect reflects sensitivity to pleasurable experiences (e.g., interest, vitality), while negative affect represents undue vigilance toward daily stressors (e.g., anxiety, frustration) [21]. In this study, positive affect and negative affect were treated as two distinct dimensions of affect. Growing evidence underscores the unique role of positive affective states in promoting psychological adjustment among patients with chronic pain [22, 23]. Specifically, positive affect is theorized to facilitate adaptive coping by counteracting fear-induced negative effects, promoting neutral reappraisal of pain, and buffering pain-related negative cognitions (i.e., rumination, helplessness, and magnification) [24, 25]. Prior research emphasizes the critical role of positive affect in pain alleviation [26]. For instance, Tang et al. found that adults with higher positive affect exhibited improved pain perception and adaptation [27]. In contrast, negative affect, including emotions such as anxiety, depression, and irritability, has been shown to exacerbate pain experiences in individuals with chronic pain. Elevated negative affect may intensify pain perception through heightened physiological arousal, increased attentional bias toward pain-related stimuli, and maladaptive cognitive appraisal processes [28, 29]. Higher negative affect is associated with increased pain intensity, greater pain-related interference, diminished quality of life, and reduced medication efficacy in this population [30]. From a lifespan perspective, emotion regulation in older adulthood differs from that in earlier life stages. According to the Socioemotional Selectivity Theory (SST), as perceived future time becomes limited, older adults increasingly prioritize emotionally meaningful goals, with greater emphasis on maintaining positive affect and regulating negative emotions [31]. Consistent with this view, the“positivity effect”suggests an age-related preference for positive over negative information in emotional processing [32]. In the context of chronic pain, positive affect may function as a key regulatory resource that supports adaptive pain appraisal and emotional resilience, whereas negative affect may reflect emotional burden under persistent stress and exert its influence more indirectly through maladaptive cognitive processes such as pain catastrophizing. Integrating SST with the biopsychosocial framework thus provides an age-sensitive basis for understanding affective and cognitive pathways linking social support to pain outcomes in older adults.

Pain catastrophizing, defined as an exaggerated, negative cognitive response to pain encompassing rumination, magnification, and helplessness [33], is a key mediator within the biopsychosocial framework of chronic pain [34]. Among older adults, it is associated with higher pain intensity, increased disability, and reduced quality of life [35, 36]. Previous research has demonstrated the important role of pain catastrophizing in shaping pain-related outcomes in older adults with chronic pain. For example, Gong et al. showed that pain catastrophizing mediated the association between pain intensity and pain-related activity patterns, highlighting catastrophizing as a key cognitive factor in chronic pain management [37]. However, the upstream psychosocial factors that contribute to pain catastrophizing in older adults remain insufficiently understood. Experimental and clinical research also highlights a bidirectional relationship between pain catastrophizing and pain intensity, wherein catastrophic thinking exacerbates pain, and pain, in turn, triggers negative cognitions [38]. Social support, by fostering positive emotional regulation (e.g., enhancing positive affect), may disrupt this cycle, as supportive interventions have been shown to decrease catastrophizing, pain severity, and emotional distress [39]. These findings suggest that social support may indirectly influence pain intensity through pathways involving pain catastrophizing and affect — a hypothesis warranting further exploration in Chinese older adults with chronic pain.

Previous research has established that social support, affect, and catastrophizing collectively influence pain intensity in older adults with chronic pain [15]. While the impact of social support on pain intensity is recognized, the mechanisms through which social support influences pain outcomes remain incompletely understood. In particular, most prior studies have examined affect and pain catastrophizing as independent or parallel mediators, without considering their potential sequential relationship. A serial mediation model allows for the examination of ordered pathways in which one mediator influences another, thereby providing a more nuanced understanding of underlying mechanisms. In the context of chronic pain, affective states may shape maladaptive cognitive responses such as pain catastrophizing, which in turn directly intensify pain perception. Accordingly, this study aims to elucidate the relationships among social support, positive and negative affect, pain catastrophizing, and pain intensity in older adults with chronic pain by constructing and testing a serial mediation model. Notably, research examining these relationships in older adults in China is limited, making it crucial to understand how these factors operate within this cultural context. Given the distinct social, cultural, and healthcare dynamics in China, it is important to investigate whether these relationships may be stronger or weaker compared to other populations. This approach seeks to clarify the mechanisms for reducing pain intensity and provide a theoretical and empirical basis for developing targeted psychological interventions in clinical settings tailored to the unique needs of Chinese older adults. The hypothetical model is presented in Fig. 1.We proposed the following hypotheses:

• H1. Social support is directly and negatively associated with pain intensity.

• H2. Positive and negative affect mediate the relationship between social support and pain intensity.

• H3. Pain catastrophizing mediates the relationship between social support and pain intensity.

• H4. Positive and negative affect, as parallel affective components, and together with pain catastrophizing, exert a sequential mediating effect on the relationship between social support and pain intensity.

Fig. 1.

Hypothesized model

Methods

Study design and participants

This cross-sectional study was conducted at a tertiary hospital in Zhuhai, Guangdong Province, China, following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Older adults with chronic pain were recruited via convenience sampling between August 2024 and February 2025. Inclusion criteria were: (1) age ≥ 60 years, and (2) chronic pain persisting or recurring for more than 3 months. Exclusion criteria were: (1) diagnosis of a psychiatric disorder or cognitive impairment, and (2) history of malignant tumor or severe chronic systemic disease. The required sample size was estimated using a Monte Carlo simulation approach for testing indirect effects [40]. Specifically, the sample size was determined using Monte Carlo simulations for a model with two serial mediators, targeting a power of 0.8 with 5000 repetitions. Furthermore, we accounted for the treatment of pain intensity as an ordered categorical variable with three levels, which necessitated an approximately 15% increase in the estimated sample size. Accordingly, the final required sample size was 350. Out of 435 eligible participants invited, 363 provided completed and valid questionnaires, yielding a response rate of 83.4%.

Data collection

Nurses from relevant departments introduced the study purpose and questionnaires to eligible participants. Participants were invited to complete the survey electronically by scanning a QR code with their mobile devices or, upon request, using a paper version. Clinical data were retrieved from the hospital’s electronic health record system.

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the fifth Affiliated Hospital of Sun Yat-sen University [Ethics No. (2024) K124-1]. Prior to participation, all participants were informed about the purpose and procedures of the study. Written informed consent was obtained from all participants. Participants were informed that their participation was voluntary and that they could withdraw from the study at any time without any consequences. Participants also provided consent for the linkage of their self-reported questionnaire data with relevant clinical records for research purposes. All data were anonymized prior to analysis to ensure confidentiality. For participants with limited literacy, the consent form was read aloud by trained research staff in a standardized manner. All study procedures, potential risks, and participants’ rights—including the right to withdraw at any time without consequences—were explained verbally, and participants were encouraged to ask questions. Consent was documented by signature or thumbprint in accordance with ethical guidelines.

Measurement tools

Demographic and clinical information

Data collection

Demographic and clinical data included gender, age, marital status, educational level, occupational status, medical payment method, personal income, residence, dwelling state, and number of pain sites.

Pain intensity

Pain intensity was assessed using the Numeric Rating Scale (NRS), a widely used and validated instrument for evaluating pain intensity in adults with chronic pain [41]. In the present study, the NRS was a priori recorded and analyzed as an ordinal severity variable with three categories: mild (0–3), moderate (4–6), and severe (7–10), based on established cut-off points [42]. This categorical classification was selected at the design stage to facilitate clinical interpretation and to align with commonly used pain severity stratification in chronic pain research.

Social support

Social support was measured by the Social Support Rating Scale (SSRS). Developed by Xiao Shui-Yuan, the SSRS is a validated 10-item instrument widely used in Chinese medical psychology research [43]. It assesses three dimensions: subjective support, objective support, and support utilization, with total scores ranging from 12 to 66 (higher scores indicate greater social support). Items 1–4 and 8–10 are single-response multiple-choice items, where respondents select one option from four response categories, scored from 1 to 4 points; item 5 includes five subitems (A–E), each scored from 0 to 4 points; and items 6–7 are scored based on the number of support sources or 0 for none. The SSRS has good reliability in Chinese populations (Cronbach’s α = 0.81–0.92) [44, 45]. In the present study, the scale demonstrated good internal consistency, with a Cronbach’s α of 0.82. Due to high inter-dimensional correlations (r > 0.80), suggesting limited dimensional distinctiveness and potential multicollinearity, the total SSRS score was used as a continuous variable in Mplus mediation analyses.

Positive and negative affect scale

Positive and Negative Affect were measured using the Positive and Negative Affect Scale (PANAS). Developed by Watson et al. [21] in 1988, the PANAS measures positive and negative affect as distinct constructs. This 20-item self-report scale includes two 10-item subscales: positive affect (e.g., enthusiasm, interest) and negative affect (e.g., anxiety, sadness). Participants evaluate each item using a 5-point Likert scale (1 = “not at all or very slightly” to 5 = “very much”), with higher scores reflecting increased affect levels. The PANAS has high reliability (Cronbach’s α = 0.92 for positive affect; 0.94 for negative affect).

Pain catastrophizing

Pain catastrophizing was measured using the Pain Catastrophizing Scale (PCS). Developed by Sullivan et al. [46] in 1995 and validated in Chinese by Yap et al. [47], the PCS comprises 13 items across three dimensions: rumination (e.g., preoccupation with pain), magnification (e.g., amplifying the threat of pain), and helplessness (e.g., perceived inability to cope with pain). Each item is rated on a 5-point Likert scale (0 = “not at all” to 4 = “all the time”). Total scores range from 0 to 52, with higher values reflecting increased levels of pain catastrophizing. PCS has good reliability (Cronbach’s α = 0.87) [48].

Statistical analysis

Descriptive statistics were performed using SPSS 27.0. There were no missing data for the main study variables. Categorical data were presented as frequencies and percentages, while continuous variables (social support, positive affect, negative affect, pain catastrophizing), which met normality assumptions, were reported as means with standard deviations. Pearson correlation analysis examined associations among continuous variables, while Spearman’s correlation analysis assessed associations between these continuous variables and pain intensity due to its ordinal nature. A serial mediation model was tested using structural equation modeling (SEM) in Mplus 8.3. Specifically, social support was specified as the independent variable, positive affect and negative affect were modeled as parallel first-stage mediators, pain catastrophizing was specified as a second-stage mediator, and pain intensity was specified as the outcome variable. Given the ordinal outcome, the model was estimated using the weighted least squares mean and variance adjusted (WLSMV) estimator. Model fit was evaluated using χ²/df, the Comparative Fit Index (CFI), the Tucker–Lewis Index (TLI), the Root Mean Square Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR). Acceptable model fit was defined according to commonly recommended criteria: χ²/df < 5, CFI > 0.90, TLI > 0.90, RMSEA < 0.08, and SRMR < 0.08 [49, 50].The model was validated using a bootstrap procedure (5000 repetitions) and tested for mediation effects with 95% confidence intervals. A two-tailed P < 0.05 was considered statistically significant.

Results

Participants’ characteristics

The participants (N = 363) had a mean age of 68.31 ± 6.49 years (range: 60–89); 180 (49.59%) were male and 183 (50.41%) were female. Most were married (92%). Nearly 40% were illiterate, and 18% had attended junior high school. Over 90% lived with family members. Detailed demographic characteristics are presented in Table 1.

Table 1.

Participants’ demographic characteristics and clinical data (N = 363)

Variable	N (%)
Total	363
Age (years)
60–74	293 (80.72%)
75–89	70 (19.28%)
Education
Illiterate	145 (39.94%)
Below primary and primary school	151 (41.60%)
Junior high school	40 (11.02%)
High school and above	27 (7.44%)
Gender
Male	180 (49.59%)
Female	183 (50.41%)
Marital status
Married	334 (92.01%)
Unmarried	4 (1.10%)
Widowed	19 (5.23%)
Divorced	6 (1.65%)
Occupational status
Employed	55 (15.15%)
Retired	236 (65.01%)
Unemployed	72 (19.83%)
Medical payment method
Medical insurance for urban residents	128 (35.26%)
Employee medical insurance	168 (46.28%)
New rural cooperative medical insurance system	67 (18.46%)
Dwelling state
Live alone	28 (7.71%)
Live with family	327 (90.08%)
Other	8 (2.20%)
Residence
Urban	274 (75.48%)
Rural	89 (24.52%)
Pain intensity(NRS:0–10)
Mild(0–3)	156 (42.98%)
Moderate(4–6)	131 (36.09%)
Severe(7–10)	76 (20.94%)
Personal income(yuan)
< 1000	92 (25.34%)
1000–3000	94 (25.90%)
3000–5000	92 (25.34%)
> 5000	85 (23.42%)

Descriptive statistics and correlation analysis

Means, standard deviations, and correlation coefficients for all study variables are presented in Table 2. Social support was positively correlated with positive affect (r = 0.535, P < 0.01) and negatively correlated with negative affect (r = -0.473, P < 0.01), pain catastrophizing (r = -0.647, P < 0.01), and pain intensity (r = -0.588, P < 0.01). Positive affect was negatively correlated with pain catastrophizing(r = -0.557, P < 0.01) and pain intensity(r = -0.588, P < 0.01), whereas negative affect was positively correlated with pain catastrophizing(r = 0.586, P < 0.01) and pain intensity(r = 0.480, P < 0.01). And Pain catastrophizing was positively correlated with pain intensity (r = 0.746, P < 0.01).

Table 2.

Descriptive statistics and correlation analysis between social support, positive/negative affect, pain catastrophizing and pain intensity (N = 363)

	Mean ± SD	Social Support	Positive Affect	Negative Affect	Pain Catastrophi-zing	Pain Intensity
Social Support	40.29 ± 7.89	1
Positive Affect	27.22 ± 7.81	0.535**	1
Negative Affect	18.60 ± 6.52	-0.473**	-0.375**	1
Pain Catastrophizing	20.28 ± 14.09	-0.647**	-0.557**	0.586**	1
Pain Intensity	—	-0.588**	-0.588**	0.480**	0.746**	1

Multicollinearity analysis

Prior to the mediation analyses, multicollinearity among the predictors and mediators was assessed. We examined the Variance Inflation Factors (VIFs) and tolerance values, with all variables (social support, positive affect, negative affect, and pain catastrophizing) entered as predictors.The results yielded VIF values ranging from 1.530 to 2.183 and tolerance values ranging from 0.458 to 0.654, indicating that multicollinearity was not a concern.

Mediating effect analysis

The serial mediation model demonstrated excellent fit to the data: χ²/df = 1.592; RMSEA = 0.040; CFI = 0.994; TLI = 0.985; SRMR = 0.011. The path coefficients are illustrated in Fig. 2 and detailed in Table 3. Social support significantly predicted higher positive affect (β = 0.529, p < 0.001) and lower negative affect (β= -0.487, p < 0.001), pain catastrophizing (β= -0.372, p < 0.001), and pain intensity (β= -0.151, p < 0.001). The total, direct, and specific indirect effects are summarized in Table 3. The total effect of social support on pain intensity was significant (β= -0.658, p < 0.001). The direct effect was also significant (β= -0.151, p < 0.001), accounting for 22.95% of the total effect. The total indirect effect was significant (β= -0.507, p < 0.001), accounting for 77.05% of the total effect. Several specific indirect paths were significant:

Fig. 2.

The serial-mediating model of social support, positive and negative affect, pain catastrophizing and pain intensity; ** P < 0.01

Table 3.

The standardized total, direct, and indirect effects of social support on pain intensity with positive and negative affect and pain catastrophizing

Model Pathway	β	SE	95%CI		Percent (%)
			Lower	Upper
Total effect	-0.658	0.091	-0.836	-0.480	100
Direct effect	-0.151	0.056	-0.256	-0.039	22.95
Social Support→ Pain Intensity	-0.151	0.056	-0.256	-0.039	22.95
Indirect effect	-0.507	0.071	-0.646	-0.368	77.05
Social Support→ Positive Affect→ Pain Catastrophizing→ Pain Intensity	-0.066	0.015	-0.099	-0.043	10.03
Social Support→ Negative Affect→ Pain Catastrophizing→ Pain Intensity	-0.088	0.018	-0.131	-0.059	13.37
Social Support→ Positive Affect→ Pain Intensity	-0.129	0.024	-0.176	-0.082	19.60
Social Support→ Negative Affect→ Pain Intensity	-0.025	0.026	-0.011	0.003	/
Social Support→ Pain Catastrophizing→ Pain Intensity	-0.199	0.032	-0.262	-0.136	30.24

These significant indirect paths, as evidenced by bootstrap confidence intervals that did not include zero, support hypotheses H2, H3, and H4.

Discussion

This study validated the proposed hypotheses, demonstrating that social support, positive and negative affect, and pain catastrophizing are significantly associated with pain intensity in older adults with chronic pain. Social support exerted both direct and indirect effects on pain intensity. The indirect pathways, primarily through positive affect and pain catastrophizing, accounted for a substantial portion (over 70%) of the total effect. These results underscore the complex interplay of emotional and cognitive processes within the biopsychosocial framework, shaping the connection between social support and pain outcomes.

The direct effects of social support

Supporting H1, social support was directly associated with lower pain intensity. This finding is consistent with longitudinal studies linking lower perceived social support to higher pain intensity [51], as well as studies showing that socially integrated activities can reduce pain prevalence [52]. Potential mechanisms include emotional support alleviating loneliness and stress, thereby decreasing central nervous system sensitivity to pain [1, 53], and instrumental support promoting treatment adherence [54]. Neuroimaging studies suggest social support can modulate brain activity in regions like the anterior cingulate cortex and prefrontal cortex, reducing pain perception [55]. In the context of Chinese older adults, where social networks are culturally significant, assessing and strengthening social support should be a clinical priority.

The mediating role of positive and negative affect

The findings partially supported H2. Positive affect served as a significant mediator, whereas the simple mediation via negative affect was not significant. Higher social support was linked to increased positive affect, which in turn was associated with lower pain intensity. This is consistent with research showing social support alleviates pain by enhancing positive affect [53] and that positive affect is associated with improved pain coping [56]. Positive affect may counteract pain through physiological pathways, such as stress reduction and endorphin system activation [57, 58]. The absence of a significant direct mediation effect via negative affect may suggest that its influence is primarily conveyed through its subsequent impact on pain catastrophizing (i.e., the serial mediation pathway), or that its effect is buffered by other resources, such as social support itself. From an aging perspective, this pattern is also consistent with theories of emotion regulation in later life, which propose that older adults may be more effective in down-regulating negative emotional states, while the impact of residual negative affect is more likely to manifest through maladaptive cognitive processes. Clinically, fostering positive affect through social and psychological interventions is a promising avenue.

The mediating role of pain catastrophizing

Supporting H3, pain catastrophizing was a significant mediator. Greater social support was associated with lower levels of catastrophizing, which subsequently was associated with lower pain intensity. Pain catastrophizing, with its components of rumination, magnification, and helplessness, amplifies pain perception and distress [36]. Social support likely mitigates catastrophizing by providing emotional reassurance, fostering adaptive coping strategies, and buffering psychological stress [59]. Physiologically, catastrophizing is linked to heightened stress responses (e.g., HPA axis activation) which social support may help regulate [60]. Routine assessment of pain catastrophizing and interventions targeting cognitive restructuring are crucial components of comprehensive pain management.

The serial mediating role of positive and negative affect and pain catastrophizing

The results strongly supported H4, revealing significant serial mediation pathways. Social support influenced pain intensity sequentially through: (1) enhancing positive affect, which was associated with lower catastrophizing, and (2) reducing negative affect, which was associated with lower catastrophizing. The total indirect effect accounted for approximately 73.24% of the relationship between social support and pain intensity, indicating that psychological mechanisms—specifically emotional states and cognitive appraisals—are the primary drivers of this association. Within this framework, the serial mediation pathways contributed 23.40% to the total effect. The remaining indirect effects—mediated independently by positive affect (19.60%) and pain catastrophizing (30.24%)—suggest that these factors also provide parallel, independent benefits. This highlights a dynamic process where social support was associated with beneficial emotional states (high positive affect, low negative affect), which promote less catastrophic appraisals of pain, ultimately relating to lower reported pain intensity [61, 62]. This serial mediation model underscores the interconnected nature of social, emotional, and cognitive factors in shaping the pain experience. It argues for interdisciplinary care where social workers, psychologists, and clinicians collaborate to address these different layers.

Importantly, the present findings should be interpreted in the context of our Chinese clinical sample. Approximately 40% of participants were illiterate, suggesting that written materials and text-based digital health information may be less accessible for many older adults. In this setting, social support networks—particularly family members—may play an indispensable role in conveying health information and supporting pain self-management, for example by helping interpret medical advice, monitor pain fluctuations, and facilitate more adaptive pain-related appraisals through everyday communication. Accordingly, strong social support may partly compensate for limited health literacy and show a more pronounced association with lower pain intensity through enhanced positive affect and reduced catastrophizing. Moreover, 7.79% of participants lived alone, reflecting the continued importance of family co-residence in China. As potential support providers are often physically present, the observed associations may reflect differences in the quality of support rather than its mere availability. Variations in emotional, instrumental, and informational support—and in family members’ ability to provide autonomy-supportive assistance—may meaningfully shape affective and cognitive responses to chronic pain.

Future longitudinal studies are warranted to clarify temporal ordering and potential causal pathways. Further research could adopt more fine-grained assessments of social support quality and examine whether the observed mechanisms differ by living arrangement, literacy level, and support source (e.g., adult children, spouses, peers, or community organizations). In addition, studies focusing on older adults living alone are needed to identify barriers to accessing support and to inform targeted community-based interventions. Finally, research on intergenerational support may explore whether assistance from grandchildren in using low-barrier digital tools (e.g., voice messaging or short videos) can help maintain social connectedness and facilitate access to remote support and telehealth guidance. From a health system perspective, these findings are also relevant to the Chinese healthcare context, where chronic pain management is often fragmented across primary care, specialty services, and mental health care. Although China’s basic medical insurance schemes provide broad population coverage, access to integrated psychosocial pain management remains uneven, particularly for older adults. In this context, family-based support may serve as a critical facilitator of coordinated care by bridging gaps between clinical encounters, supporting treatment adherence, and reinforcing coping strategies outside formal healthcare settings.

Implications for clinical practice

This study underscores the necessity of transitioning from a conventional biomedical approach to a holistic pain management strategy that integrates psychosocial elements. First, clinical practice should move beyond a siloed approach. All healthcare professionals involved in the care of older adults (including nurses, primary care physicians, and geriatricians) should prioritize assessing and strengthening social support networks, particularly in Chinese communities where family and social ties are culturally significant. Interventions such as family involvement, peer support groups, and community programs may enhance emotional and informational support, potentially reducing pain catastrophizing and improving pain-related outcomes.

Second, pain management protocols should be redesigned to integrate emotional and psychological interventions as a standard component. For older adults with limited literacy or limited access to digital resources, interventions may be more feasible when delivered through face-to-face education during clinical visits, family-assisted guidance, community-based group sessions, and materials using pictorial or audio formats. For older adults with adequate literacy and digital access, technology-assisted approaches (e.g., simplified mobile applications with audio guidance or voice-based modules) may complement traditional care and support emotion regulation and self-management. More intensive psychological interventions such as cognitive-behavioral therapy (CBT), delivered by qualified psychologists or trained clinicians, may be considered for individuals with elevated pain catastrophizing or persistent emotional distress.

Finally, psychosocial assessments should be a core component of comprehensive geriatric assessments. This requires systematic interdisciplinary collaboration, where nurses lead in initial screening and ongoing monitoring, social workers address support network gaps (particularly for socially isolated individuals), and psychologists or pain specialists manage complex cognitive and emotional aspects. Overall, tailoring psychosocial interventions to patients’ literacy level, living arrangement, and available support resources may improve feasibility and effectiveness, ultimately enhancing pain outcomes and quality of life for older adults with chronic pain.

Limitations

This study has several limitations. First, its cross-sectional design precludes causal inference. The serial mediation model was theory-driven and aimed to examine potential psychological mechanisms rather than establish causality. While social support is a complex and multifaceted construct, quantitative modeling remains a widely used approach to test hypothesized pathways in chronic pain research, and qualitative or mixed-methods designs may further complement these findings in future studies. Second, the use of convenience sampling from a single tertiary hospital may limit the generalizability of the findings. Specifically, the sample may over represent older adults who actively seek medical care for chronic pain, potentially those with more severe symptoms, longer pain duration, or greater functional impairment. Conversely, older adults who manage chronic pain primarily in community or home settings, those with limited access to tertiary healthcare services, and socially isolated individuals (e.g., older adults living alone) may be underrepresented in the present study. In addition, modeling pain intensity as an ordinal variable may reduce variability and statistical power compared with using the raw 0–10 NRS score; future research using continuous measurements may further validate the findings and explore potential nonlinear effects. Furthermore, the study did not differentiate between types of chronic pain conditions (e.g., neuropathic vs. musculoskeletal), which may involve distinct biopsychosocial pathways. Future studies should examine these mechanisms across different pain etiologies and in more diverse and representative populations.

Conclusions

This study demonstrates that social support significantly was associated with pain intensity in older adults with chronic pain, both directly and indirectly through the serial mediating roles of positive affect, negative affect, and pain catastrophizing. The identified pathways clarify the psychosocial mechanisms through which social resources translate into better pain outcomes. Our findings underscore the necessity of a multidisciplinary, biopsychosocial approach to chronic pain management in older adults. Therefore, effective management strategies should integrate coordinated efforts to enhance social support, foster positive affect, and mitigate pain catastrophizing within a collaborative care framework that brings together clinicians, mental health professionals, and social support systems.

Authors’ contributions

Jiali Zhang contributed to project conceptualization, study design, academic supervision, and funding acquisition. Lirong Zeng led project administration, protocol development, resource coordination, and oversight of study implementation. Xing Zhang, Ruike Zhang and Xiaoha Lin were responsible for data management, manuscript drafting, statistical analyses, visualization and data collection. Jiali Zhang and Lirong Zeng are the co-guarantors of this study. All authors critically reviewed, revised, and approved the final version of the manuscript.

Funding

This study was supported by the Zhuhai Science and Technology Innovation Bureau, Guangdong Province, under Grant No. 2420004000305.

Data availability

The data in this study can be obtained from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

According to the Declaration of Helsinki, this study was approved by the Ethics Committee of the Fifth Affiliated Hospital of Sun Yat-sen University [Ethics No. (2024) K124-1]. All participants provided informed consent for the study and had the right to withdraw at any time.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.