Pain and depression mediate the relationship between multimorbidity and frailty in older adults with diabetes Authors: Yaqing Liu, Xun Liu, Longhan Zhang, Lu Lin, Xiaoyun Li, Sixuan Guo, Xingyu Wei, Yuanhong Sun, Yiting Pan, An Luo, Hao Bai, Li Liao, Jing Yang Published: 2026 ABSTRACT Frailty is increasingly recognized as a significant risk factor for death and disability in older adults with diabetes. Although previous studies have confirmed the correlation between the number of comorbid chronic conditions and frailty in older adults with diabetes, the mechanism underlying this effect remains unclear. Pain and depression are common symptoms accompanying diabetes. Therefore, this study aims to examine the mediation effect of pain and depression on the relationship between multimorbidity and frailty in older adults with diabetes. This study used data from the China Health and Retirement Longitudinal Study conducted by Peking University in 2018 and 2020, selecting 799 older adults with diabetes. Descriptive and correlation analyses were performed using SPSS 27.0, and the mediation analyses were conducted using the PROCESS macro. Pain and depression played mediating roles between multimorbidity and frailty in older adults with diabetes, with direct (0.062) and indirect (0.075) effects accounting for 45.26% and 54.74% of the total effect (0.137), respectively. Further analysis revealed three significant indirect pathways: (1) multimorbidity → pain → frailty (8.03% of the total effect), (2) multimorbidity → depression → frailty (43.07%), and (3) multimorbidity → pain → depression → frailty (3.65%). These findings highlight the importance of early intervention for older adults with diabetes who have chronic diseases, particularly those experiencing pain and depression. However, the generalizability of these findings may be limited to community-dwelling older adults, and further validation in hospitalized or institutionalized populations is warranted.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-51580-1. Introduction Diabetes continues to be a significant public health challenge worldwide. With the world’s population aging, the proportion of people over 60 with diabetes will continue to rise. Evidence shows that frailty is becoming the third type of complication of diabetes in addition to the traditional microvascular and macrovascular complications1. The progressive decline in physiological reserves associated with frailty, driven by age-related dysregulation of multiple systems, significantly increases the risk of functional decline and mortality2. For older adults with diabetes, the added burden of metabolic dysfunction and chronic complications further amplifies this risk, positioning frailty as a key determinant of poor health outcomes in this group3. Due to chronic hyperglycemia, insulin resistance, and other metabolic disturbances4, the prevalence of frailty among older adults with diabetes is 3 to 5 times higher than that among their non-diabetic counterparts3. Frailty significantly increases the risk of adverse outcomes such as falls, fractures, disability, and cognitive impairment in this population5, severely compromising their physical function, quality of life, and overall prognosis. Indeed, frailty is not simply a consequence of aging and is not always progressive6. Instead, the functional deficits linked to frailty in older adults are reversible and dynamic7. Therefore, exploring the factors influencing frailty and its underlying mechanisms in patients with diabetes is imperative for implementing possible early screening and suitable intervention strategies, subsequently reducing the incidence of complications and improving the prognosis. Older patients with diabetes mellitus are a unique and complex group, often burdened with multiple co-existing chronic diseases, such as hypertension and dyslipidemia8. While multimorbidity refers to two or more coexisting conditions in an individual9, frailty represents a distinct geriatric syndrome characterized by decreased physiological reserve and increased vulnerability to stressors10. Multimorbidity may contribute to frailty, but the latter encompasses broader functional and biological dysregulation11. The high prevalence of multimorbidity and frailty in older people with diabetes is attributed to both complications directly linked to diabetes and other conditions associated with diabetes12. Frailty results from the combined effects of the natural aging process and the exacerbating influence of some chronic diseases, which accelerate age-related physiological decline11. High blood sugar levels predict a transition to higher frailty levels, as chronic hyperglycemia contributes to systemic inflammation, oxidative stress, and cellular dysfunction13. Additionally, vascular complications of diabetes mellitus (e.g., peripheral artery disease, stroke) and malnutrition further exacerbate functional decline in frail older adults, creating a vicious cycle of disability and dependency. Frailty is an age-related syndrome resulting from a multidimensional process leading to increased vulnerability, reduced ability to respond to stressors, adverse health outcomes, and also death 10. Coexisting chronic conditions exacerbate chronic hyperglycemia and insulin resistance, significantly increasing the risk of frailty and seriously affecting the quality of life of patients with diabetes. Frailty in older adults with diabetes is understood as an age-related syndrome resulting from a multidimensional process, leading to increased vulnerability, reduced ability to respond to stressors, and ultimately adverse health outcomes and increased mortality. This multimorbidity exacerbates hyperglycemia and insulin resistance, significantly increasing the risk of frailty and seriously affecting the quality of life. It is also known that frailty is not limited to just physical aspects but has psychological and social dimensions as well. However, the mechanisms linking these chronic diseases to frailty are not fully understood. Specifically, identifying modifiable factors that lie on the path between multimorbidity burden and frailty could offer novel targets for intervention. This study aims to investigate the mediating roles of pain and depressive symptoms in this relationship. Pain is a common symptom accompanying different chronic diseases such as cancer, diabetes, neurodegenerative diseases, and liver, renal, and cardiovascular diseases14. These chronic diseases are often accompanied by a chronic inflammatory state, and elevated inflammatory factors such as C-reactive protein (CRP) and interleukin-6 (IL-6) may be associated with the onset of pain15. Studies have shown that patients with more chronic diseases have more severe pain than those with fewer chronic diseases16. Chronic pain could serve as a pivotal stressor in the ageing process, potentially instigating or hastening frailty17. In older adults, persistent pain might lead to a loss of physiological reserves and impaired mobility, thus contributing to the likelihood of frailty18. Depression and pain are highly comorbid and share biological mechanisms. Zis found that both depression and pain might be risk factors for each other19. Pain can have demoralizing consequences, negatively affecting physical condition, overall quality of life, and emotional well-being20. Research suggests that chronic pain may lead to abnormal functioning in areas of the brain associated with emotion regulation, such as the prefrontal lobe and amygdala, which can increase the risk of depression21. Moreover, research indicates that multimorbidity can contribute to depression, with pain possibly being one of the factors22. Through the above literature analysis, it can be seen that pain and depression play an important role in the relationship between multimorbidity and frailty in older adults with diabetes. Gender is a key factor influencing frailty. Research indicates that women are generally more likely to experience frailty compared to men23. This difference is often attributed to biological factors, such as hormonal changes and lower muscle mass, as well as social factors, including caregiving responsibilities and longer life expectancy24. Frailty is strongly associated with age. As individuals grow older, the risk of frailty increases due to the natural decline in physiological reserves, the accumulation of chronic diseases, reduced physical activity and so on25. Higher education levels are often linked to a lower risk of frailty. Education can influence health behaviors, access to healthcare, and socioeconomic status, all of which contribute to better health outcomes and reduced frailty26. Marital status can also affect frailty. Married individuals, particularly those in supportive relationships, tend to have lower rates of frailty compared to those who are single, divorced, or widowed. Social support from a spouse can promote healthier lifestyles, provide emotional stability, and encourage timely medical care, all of which can mitigate the risk of frailty27. In summary, gender, age, education level, and marital status have significant impacts on frailty. Therefore, this study included these variables as covariates to control for confounding factors, thereby enhancing the accuracy and reliability of the analysis. The management of chronic diseases in older adults has traditionally focused on a disease-centered approach, which often fails to address the broader, multifaceted nature of aging28. This approach primarily targets specific diseases like diabetes without considering the interplay of other health conditions, psychological factors, and functional limitations that significantly impact older adults. Frailty, a key geriatric syndrome, embodies this multidimensional aspect of aging, as it involves a complex combination of physical, cognitive, and emotional factors that go beyond the mere presence of chronic diseases29. Given that frailty increases the vulnerability of older adults to adverse health outcomes, it is crucial to move away from a purely disease-centric care model. In this study, we focused on the relationship between comorbid chronic diseases and frailty, as well as the mediating roles of pain and depression. We treated depression as a mediating variable. By examining the chain mediation effect of pain and depression, we seek to highlight the importance of adopting a more holistic approach in geriatric care—one that considers not only the management of diabetes and other chronic conditions but also the psychological and functional dimensions that contribute to frailty and overall well-being. Materials and methods Research hypotheses This study proposes the following hypotheses: H1 Multimorbidity has a significant predictive effect on frailty in older adults with diabetes. H2 Pain plays a mediating role in the relationship between multimorbidity and frailty in older adults with diabetes. H3 Depression is a mediator between multimorbidity and frailty in older adults with diabetes. H4 Pain and depression play a chain mediation effect between multimorbidity and frailty in older adults with diabetes. Study population This is a cross-sectional study. The research data were derived from the China Health and Retirement Longitudinal Study (CHARLS) led by Peking University in 2018 and 2020. In the baseline survey (2011), CHARLS surveyed 17,000 middle-aged and older people aged 45 and above in about 10,000 households30. In terms of survey content, CHARLS questionnaire covered individual and family information of middle-aged and older people, including physical and mental health status represented by depressive symptoms, chronic disease status, and self-rated health, as well as demographic variables such as gender, age, marital status, and education level. CHARLS designs its survey scheme based on the frameworks of a series of international aging studies, such as the Health and Retirement Study (HRS) of the University of Michigan in the United States, to ensure the comparability of research findings. CHARLS data collectors should have relevant bachelor’s or higher degrees, good communication and computer skills, and essential qualities. Training focuses on improving data collection quality, covering theory, skills, and fieldwork. It combines multiple methods and has assessments. Only passers participate, with regular follow-up training. The combined data from 2018 and 2020 were analyzed in this study. The inclusion criteria are as follows: (a) Age ≥ 60 years old; (b) Diagnosed with diabetes; (c) Data missing for frailty, depression, pain, and comorbid chronic disease items in CHARLS < 10%. Exclusion criteria: (a) Abnormal data; (b) Missing covariate values. Finally, 799 valid samples were included. The diagnostic criteria for diabetes mellitus in this study were as follows, in accordance with the Guideline for the Management of Diabetes Mellitus in the Elderly in China (2024 edition)31: Fasting plasma glucose (FPG) ≥ 7.0 mmol/L (confirmed by repeat testing on a separate day for asymptomatic individuals); or classic diabetes symptoms (polyuria, polydipsia, polyphagia, and unexplained weight loss) plus random plasma glucose or venous plasma glucose ≥ 11.1 mmol/L; or 2-h postprandial glucose (2hPG) ≥ 11.1 mmol/L during oral glucose tolerance test (OGTT) (confirmed by repeat testing for asymptomatic individuals); or Glycated hemoglobin (HbA1c) ≥ 6.5%. CHARLS datasets are available for download from CHARLS homepage: http://charls.pku.edu.cn/en. CHARLS was approved by the Biomedical Ethics Committee of Peking University (approval number: IRB00001052-11,015), and informed consent was obtained from all the participants. This study was conducted with reference to the guidelines for analyzing mediation, conditioning and conditioning processes32. The detailed flowchart of participant selection is presented in Fig. 1. The missing data were calculated as the median of all other values due to the non-normal distribution of the data.Fig. 1Flowchart of the selection. Flowchart of the selection. Measurement variables Frailty Frailty was assessed using the validated Fried Frailty Phenotype (FP) criteria, which is the most widely used frailty assessment tool in epidemiological studies. Consistent with previous CHARLS-based research33,34, we operationalized frailty as a binary variable according to the following standardized criteria: (1) Weakness was measured using the self-reported item “having difficulty in lifting or carrying weights over 5 kg”; (2) Slowness was considered present if the participant had difficulty in walking 100 m or climbing several flights of stairs without resting; (3) Exhaustion was present if the participant answered “Most or all of the time” or “Occasionally or a moderate amount of the time”, in response to either of the questions: “I felt everything I did was an effort during last week” or “I could not get going during last week”; (4) Low physical activity was considered to be present if the participant did not undertake a physical activity or walk at least 10 min at a time during a usual week; (5) Weight loss was defined as the unintentional loss of 5 kg or more in the last year, or current body mass index (BMI) ≤ 18.5 kg/m2. Frailty was defined as the presence of three or more of the five components (each component was scored 1 point). Those with a frailty score ≥ 3 were considered frail, while those with a score < 3 were non-frail. The total score was included in the model for mediation analysis. Depression The Center for Epidemiology Studies Depression Scale-10 (CES-D-10) developed by Andresen was used to assess depression in older people in CHARLS35. Studies have shown that the CES-D-10 has good validity and reliability among the middle-aged and older Chinese population36. CES-D-10 includes 10 items: (1) I was bothered by things that don’t usually bother me; (2) I had trouble keeping my mind on what I was doing; (3) I felt depressed; (4) I felt everything I did was an effort; (5) I felt hopeful about the future; (6) I felt fearful; (7) My sleep was restless; (8) I was happy; (9) I felt lonely; (10) I could not get “going”. There are 4 single-choice options for each item: (a) Rarely or none of the time (< 1 d); (b) Some or a little of the time (1 ~ 2 d); (c) Occasionally or a moderate amount of the time (3 ~ 4 d); (d) Most or all of the time (5 ~ 7 (d). These choices were marked as 0, 1, 2, and 3, respectively. Additionally, item 5 and item 8 were scored in reverse. The total score of the CES-D-10 is 30 points, and a participant with 10 points or above was considered depressed 37,38. The total score was entered into the model for mediation analysis. Pain Pain was assessed by asking “On what part of your body do you feel pain? Please list all parts of the body you are currently feeling pain”, in 16 parts of the body including head, shoulder, arm, chest, back, waist and so on. A score of 1 is assigned to each area, and the total score is 16, with higher scores indicating more areas of pain. Pain score represents a count of body areas with pain, making it a discrete numerical variable. Multimorbidity In this study, the number of comorbid chronic diseases was used to represent the degree of multimorbidity in older adults with diabetes. There are 14 categories of chronic diseases in CHARLS, including hypertension, dyslipidemia, emphysema, and so on. It was assessed by asking “Have you been diagnosed with [conditions listed below, read one by one] by a doctor?” Because the study population is people with diabetes, each chronic disease reported was scored as 1 point, with 13 points in total (except diabetes). This count represents the total number of different chronic diseases a participant had in addition to their diabetes diagnosis. Higher scores represent a greater number of comorbid chronic diseases in people with diabetes39. Potential covariates According to prior knowledge, the sociodemographic characteristics factors were also considered in our study. Demographic variables include gender (male, female), age, education level (illiterate, primary school, junior high school and above), and marital status (married, separated/divorced/widowed). Statistical analysis The data were organized and analyzed by SPSS 27.0. Nominal variables were described by constituent ratios. Because numerical variables were not normally distributed, they were expressed as M (P25, P75). Mann–Whitney U test was used for comparison between two groups, and the Kruskal–Wallis H test was used for multiple comparisons. The correlation between variables was verified by Spearman correlation coefficient analysis. The chain mediation effect of pain and depression between multimorbidity and frailty in older people with diabetes was tested by applying Model 6 in PROCESS 4.340, controlling for gender, age, education, and marriage. PROCESS Model 6 was selected for its ability to test sequential mediation, with assumptions verified through residual analysis and collinearity diagnostics41. Bootstrap test42 for mediation effects was used to ensure greater accuracy and stability in estimating confidence intervals43. We collected available data from CHARLS database to the greatest extent possible and conducted a power analysis using G*Power 3.1 software44. The results indicated that our sample size (N = 799) achieved 95% statistical power for detecting a medium effect size (f2 = 0.15, α = 0.05). Results Descriptive statistics Table 1 summarizes the demographic characteristics of the study participants. A total of 799 older adults with diabetes were enrolled in this study, including 350 (43.80%) males and 449 (56.20%) females; age was 66.0(63.0, 72.0) years; education: illiterate 135 (16.90%), primary school 571 (71.46%), junior high school and above 93 (11.64%); marital status: married 659 (82.48%), separated/divorced/widowed 140 (17.52%); duration of diabetes mellitus was 2.0 (1.0, 2.0) years.Table 1Demographic characteristics of study participants (N = 799).CharacteristicM (P25, P75)N (%)Gender Male350(43.80) Female449(56.20) Age(years)66.0(63.0, 72.0)Education Illiterate135(16.90) Primary school571(71.46) Junior high school and above93(11.64)Marital status Married659(82.48) Separated/divorced/widowed140(17.52) Duration of diabetes mellitus (years)2.0(1.0, 2.0) Number of pain area(s)5.0(1.7, 6.0) CES-D-10* score9.0(4.0, 14.0) People with depression313(39.17) Number of comorbid chronic diseases2.0 (1.0, 4.0) Frailty score1.0 (1.0, 2.0) People with frailty153(19.15)*CES-D-10: Center for Epidemiology Studies Depression Scale-10. Table 1Demographic characteristics of study participants (N = 799).CharacteristicM (P25, P75)N (%)Gender Male350(43.80) Female449(56.20) Age(years)66.0(63.0, 72.0)Education Illiterate135(16.90) Primary school571(71.46) Junior high school and above93(11.64)Marital status Married659(82.48) Separated/divorced/widowed140(17.52) Duration of diabetes mellitus (years)2.0(1.0, 2.0) Number of pain area(s)5.0(1.7, 6.0) CES-D-10* score9.0(4.0, 14.0) People with depression313(39.17) Number of comorbid chronic diseases2.0 (1.0, 4.0) Frailty score1.0 (1.0, 2.0) People with frailty153(19.15)*CES-D-10: Center for Epidemiology Studies Depression Scale-10. Demographic characteristics of study participants (N = 799). *CES-D-10: Center for Epidemiology Studies Depression Scale-10. The number of comorbid chronic diseases in older adults with diabetes was 2.0 (1.0, 4.0), frailty score was 1.0 (1.0, 2.0), pain score was 5.0 (1.7, 6.0) and depression score was 9.0 (5.0, 14.0). The prevalence of frailty in older adults with diabetes was 19.15%. The prevalence of depression was 39.17%. See Table 1 for details. Correlations statistics of major study variables Table 2 presents the correlation analysis results, exploring the relationships among multimorbidity, pain, frailty, and depression. Multimorbidity showed significant positive correlations with pain, depression, and frailty in older adults with diabetes (r = 0.151, 0.189, 0.206; P < 0.01). Additionally, pain showed a significant positive correlation with depression and frailty (r = 0.109, 0.171; P < 0.01). Moreover, depression also showed significant positive correlations with frailty (r = 0.510, P < 0.01).Table 2Correlations between multimorbidity, pain, depression and frailty.multimorbidityPainDepressionFrailtyMultimorbidity1.000Pain0.151**1.000Depression0.189**0.109**1.000Frailty0.206**0.171**0.510**1.000**p < 0.01 (two-tailed). Table 2Correlations between multimorbidity, pain, depression and frailty.multimorbidityPainDepressionFrailtyMultimorbidity1.000Pain0.151**1.000Depression0.189**0.109**1.000Frailty0.206**0.171**0.510**1.000**p < 0.01 (two-tailed). Correlations between multimorbidity, pain, depression and frailty. **p < 0.01 (two-tailed). Mediation effects This study conducted a mediation analysis using the SPSS PROCESS macro, version 4.3 (model 6), developed by Preacher and Hayes41. We used 5,000 bootstrap samples to examine the mediating effects of pain and depression on the relationship between multimorbidity and frailty in older adults with diabetes, while controlling for age, gender, education level, and marital status. The statistical significance level was set at p < 0.001 (two-tailed). Multiple regression analysis showed that multimorbidity significantly positively predicted pain (β = 0.386, P < 0.001). When both multimorbidity and pain were included as predictor variables, their effects on depression were significant (β = 0.735, 0.165; P < 0.001). When multimorbidity, pain, and depression were included as predictor variables simultaneously, all three positively predicted frailty (β = 0.062, 0.029, 0.080; P < 0.001). See Table 3 for details.Table 3Regression analysis between variables.Outcome variablePredictor variableGlobal fit indexSignificance of regression coefficientRR2FβtPainMultimorbidity0.2270.0528.4670.3865.896***DepressionMultimorbidity0.2670.0719.9200.7355.389***Pain0.1652.252*FrailtyMultimorbidity0.5540.30748.9770.0623.057**Pain0.0292.680**Depression0.08015.223***Adjusted age, gender, education, and marital status; *p < 0.05 (two-tailed); **p < 0.01 (two-tailed); *** p < 0.001 (two-tailed) Table 3Regression analysis between variables.Outcome variablePredictor variableGlobal fit indexSignificance of regression coefficientRR2FβtPainMultimorbidity0.2270.0528.4670.3865.896***DepressionMultimorbidity0.2670.0719.9200.7355.389***Pain0.1652.252*FrailtyMultimorbidity0.5540.30748.9770.0623.057**Pain0.0292.680**Depression0.08015.223***Adjusted age, gender, education, and marital status; *p < 0.05 (two-tailed); **p < 0.01 (two-tailed); *** p < 0.001 (two-tailed) Regression analysis between variables. Adjusted age, gender, education, and marital status; *p < 0.05 (two-tailed); **p < 0.01 (two-tailed); *** p < 0.001 (two-tailed) As shown in Table 4, analysis of the total indirect effect indicated that pain and depression partially mediate the relationship between multimorbidity and frailty (Effect = 0.075, SE = 0.013, 95%CI 0.050 ~ 0.100), accounting for 54.74% of the total effect. Meanwhile, when tested separately, the mediating paths were significant: (a) the indirect effects of multimorbidity on frailty through pain (Effect = 0.011, SE = 0.005, 95%CI 0.002 ~ 0.022), accounting for 8.03% of the total effect; (b)the indirect effects of multimorbidity on frailty through depression (Effect = 0.059, SE = 0.012, 95%CI 0.036 ~ 0.083), accounting for 43.07% of the total effect; (c) the indirect effects of multimorbidity on frailty through pain to depression(Effect = 0.005, SE = 0.002, 95%CI 0.001 ~ 0.010), accounting for 3.65% of the total effect. The specific paths are presented in Fig. 2. All the indices concerning the model effects are shown in Table 4.Table 4The mediating effect of pain and depression between multimorbidity and frailty in older adults with diabetes.Effect valueBoot SEBoot CI lowerBoot CI upperRelative mediation effect (%)Power#Total effect0.1370.0220.0930.181100.00Total direct effect0.0620.0200.0220.10245.26Total indirect effect0.0750.0130.0500.10054.74Indirect path 1*0.0110.0050.0020.0228.030.93Indirect path 2**0.0590.0120.0360.08343.071.00Indirect path 3***0.0050.0020.0010.0103.650.65*Indirect path 1: the indirect effects of multimorbidity on frailty through pain; **Indirect path 2: the indirect effects of multimorbidity on frailty through depression; ***Indirect path 3: the indirect effects of multimorbidity on frailty through pain and then depression; # The power analyses was conducted using G*Power 3.1 software.Fig. 2Model of mediating roles of pain and depression between multimorbidity and frailty in older adults with diabetes. Notes: **p < 0.01 (two-tailed); *** p < 0.001 (two-tailed). Table 4The mediating effect of pain and depression between multimorbidity and frailty in older adults with diabetes.Effect valueBoot SEBoot CI lowerBoot CI upperRelative mediation effect (%)Power#Total effect0.1370.0220.0930.181100.00Total direct effect0.0620.0200.0220.10245.26Total indirect effect0.0750.0130.0500.10054.74Indirect path 1*0.0110.0050.0020.0228.030.93Indirect path 2**0.0590.0120.0360.08343.071.00Indirect path 3***0.0050.0020.0010.0103.650.65*Indirect path 1: the indirect effects of multimorbidity on frailty through pain; **Indirect path 2: the indirect effects of multimorbidity on frailty through depression; ***Indirect path 3: the indirect effects of multimorbidity on frailty through pain and then depression; # The power analyses was conducted using G*Power 3.1 software. The mediating effect of pain and depression between multimorbidity and frailty in older adults with diabetes. *Indirect path 1: the indirect effects of multimorbidity on frailty through pain; **Indirect path 2: the indirect effects of multimorbidity on frailty through depression; ***Indirect path 3: the indirect effects of multimorbidity on frailty through pain and then depression; # The power analyses was conducted using G*Power 3.1 software. Model of mediating roles of pain and depression between multimorbidity and frailty in older adults with diabetes. Notes: **p < 0.01 (two-tailed); *** p < 0.001 (two-tailed). Discussion To the best of our knowledge, this is the first study to investigate the relationship between multimorbidity, pain, depression, and frailty among older adults with diabetes, as well as the mediating effects of pain and depression. The results showed that multimorbidity affected frailty both directly and indirectly through pain and depression. Specifically, pain and depression partially mediated the effects of multimorbidity on frailty. The number of comorbid chronic diseases was associated with the number of pain sites, which in turn was associated with depression, and depression predicted increased frailty. While it is important to acknowledge that depression can also contribute to pain perception through mechanisms such as altered pain modulation and heightened sensitivity to nociceptive stimuli22, our findings primarily support the pathway where pain precedes and exacerbates depressive symptoms in older adults with diabetes. This aligns with the biopsychosocial model of chronic pain, which posits that persistent pain disrupts emotional regulation and increases vulnerability to depression45. The chain mediation model helps reveal the progression of frailty among older adults with diabetes. These findings help explain how multimorbidity contributes to frailty, highlighting potential targets for managing this condition in older adults with diabetes. In this study, the prevalence of frailty among older Chinese adults with diabetes was 19.15%, significantly higher than that of the general older population in China (9.9%)46. Long-term chronic hyperglycemia, insulin resistance, and related factors contribute to a frailty prevalence that is 3–5 times higher in older adults with diabetes compared to their non-diabetic counterparts3. Therefore, multisite pain—a key mediator in our model—has been linked to frailty progression. A study by Thapa et al. (2022) demonstrated that 49% of those with frailty also had multisite pain47. Frail older individuals are more susceptible to adverse events, including hypoglycemia and mortality. Considering the high prevalence of frailty in older adults with diabetes, clinicians give greater consideration to frailty when planning diabetes treatment for this population. Furthermore, the prevalence of depression in our study population was 39.17%, slightly lower than that of the general older population in China (40.80%)48. Depression is a common negative emotional response in people with diabetes, due to the decline in physical functioning, which can easily occur in the face of the lifestyle changes caused by diabetes and its heavy management burden49. Healthcare professionals should be attentive to the mental status of older patients with diabetes, recognizing depression as early as possible and implementing appropriate interventions. Of note, the findings of this study are consistent with previous research23,24, indicating a higher prevalence of frailty among women (23.4%) than men (13.7%). The construction of a chain mediating model, supported by previous research, effectively explains the relationship between multimorbidity and frailty by elaborating on the mediating mechanisms of pain and depression. Firstly, the data confirmed that pain mediated the relationship between multimorbidity and frailty, validating H2. Pain is associated with poor physical function outcomes, including reduced grip strength, slower gait speed, malnutrition, and lower physical activity, all of which are key components of the physical phenotype of frailty50. The significant positive association between multimorbidity and frailty aligns with the research of Zazzara51. Chronic pain may activate the hypothalamic–pituitary–adrenal (HPA) axis, leading to a sustained elevation of stress hormones such as cortisol, which is associated with the development of frailty18. In addition to discomfort, pain is associated with poorer health outcomes, including depression, sleep disturbances, decreased socialization, and functional impairment, all contributing to the frailty in older people with diabetes52. Meanwhile, the continuous pain signals stimulate the relevant neural pathways in the brain, affecting the balance of neurotransmitters such as serotonin. A decrease in serotonin levels is closely related to the development of depressive mood, and depression further exacerbates the decline of physical function, forming a vicious cycle53. Secondly, the research data also showed that depression played a mediating role between multimorbidity and frailty in older adults with diabetes, validating H3. The finding indicated that depression significantly predicted frailty in a positive manner, which was consistent with the studies of Ning34 and Bu33. Furthermore, a recent bidirectional Mendelian randomization study demonstrated a bidirectional causal relationship between frailty and depression54. Depression could lead to poor nutritional status, sleep disturbance, and emotional disorders, which may seriously affect physical health and cause frailty. Depression activates the hypothalamic–pituitary–adrenal (HPA) axis, leading to persistent hypercortisolemia55. Elevated cortisol levels promote muscle catabolism by inhibiting protein synthesis and increasing proteolysis, resulting in loss of skeletal muscle mass and strength- a key feature of frailty. Additionally, cortisol disrupts glucose metabolism, contributing to insulin resistance and exacerbating diabetes-related metabolic dysregulation, which further impairs physical function. Inflammatory pathways are also dysregulated in depression. Chronic low-grade inflammation, characterized by increased levels of cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α)56, is associated with both depression and frailty. These cytokines interfere with normal energy metabolism, reducing appetite and nutrient absorption, thereby leading to malnutrition. Moreover, IL-6 can directly affect muscle function by promoting muscle atrophy and impairing muscle regeneration. At the neuronal level, depression-induced alterations in neurotransmitter systems, particularly reduced serotonin and dopamine signaling, impact motor control and motivation57. Serotonin modulates muscle tone and movement coordination, while dopamine influences the reward system that drives physical activity. Deficits in these neurotransmitters can lead to psychomotor retardation, decreased physical activity, and accelerated muscle deconditioning. Furthermore, depression is associated with disrupted circadian rhythms, which govern physiological processes including sleep–wake cycles, hormone secretion, and metabolism. Misaligned circadian rhythms contribute to sleep disturbances, metabolic dysfunction, and impaired immune function-all factors that predispose to frailty development58. Although potential biological mechanisms linking depression to frailty were discussed above, the absence of objective biomarkers in the study data limits biological validation of this pathway. Therefore, future research should focus on collecting objective biomarkers to empirically verify these mechanistic links. Finally, in exploring the mechanism between multimorbidity and frailty, the study also identified an important pathway consisting of chain mediation from pain to depression. Research showed that pain may positively predict depression, and the result was consistent with the research conclusions of Zis19 and Viderman20. Although depression can exacerbate pain through mechanisms such as central sensitization and maladaptive coping behaviors, our findings highlight that pain-related disability and distress are more salient contributors to depressive symptoms in older adults with diabetes59. The chronic physical discomfort and functional limitations associated with pain may lead to increased psychological stress, which may induce or exacerbate depression. Overall, the interactions between multimorbidity, pain, depression, and frailty in older people with diabetes are multifaceted and involve multiple biological, psychological and social dimensions. These factors are intertwined to form a complex pathological network. Therefore, in clinical practice, multidimensional intervention strategies, including pharmacological treatment, psychotherapy, lifestyle modification and rehabilitation training, should be adopted to interrupt this pathological cycle and improve the overall health of patients. Limitations and prospects This study had some limitations. First, self-reported measures for pain, depression, and frailty may introduce recall and reporting bias, particularly in older populations. Second, since our data were derived from CHARLS, a cohort of community-dwelling older adults in China, the findings may lack generalizability. Moreover, we cannot rule out the effects of residual confounding due to unmeasured factors, including physical activity, nutritional status, cognitive function, and social support. Additionally, as a cross-sectional study, it cannot establish causal relationships between variables. Notably, using a single frailty measurement method may limit the comprehensiveness of findings, as different measures can affect baseline risk and clinical decisions60. To address these limitations, future research should adopt longitudinal and intervention-based designs to clarify causal relationships. It should also incorporate objective biomarkers, expand sample scope to enhance generalizability, assess potential confounding factors comprehensively, and use diverse frailty measures to improve the robustness of the findings. Conclusion This research demonstrates that pain and depression mediate the association between multimorbidity and frailty in older adults with diabetes. These findings emphasize the need for early interventions targeting pain and depression in older diabetic individuals with multiple chronic conditions. Future research should focus on developing and evaluating targeted interventions that address these pathways to improve health outcomes for this vulnerable population. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Supplementary Material 1.