Document resource
WHAT IS ALREADY KNOWN ON THIS TOPIC The association between type 2 diabetes and depression is well researched, with ORs typically ranging from 1.15 to 2.55. However, studies investigating other psychosocial aspects are scarce, and often impose restrictive inclusion criteria.WHAT THIS STUDY ADDS We investigate whether there is an association between type 2 diabetes and a range of psychosocial burdens among an otherwise hard-to-reach population of adults aged 65 years and above in Germany. Older adults with type 2 diabetes had higher odds of experiencing depressive symptoms, loneliness, and lower levels of energy, life satisfaction, and satisfaction with one’s activity level, even after adjustment for socio-demographic, lifestyle-related and comorbidity variables.HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Findings call for integrated care models that routinely screen for and address psychosocial burdens in older people with type 2 diabetes.Introduction Diabetes mellitus is a chronic metabolic disease characterized by elevated levels of blood glucose. In 2024, 10.3% of adults in Germany reported having diagnosed diabetes, with higher prevalence among men and adults aged 65 years and older. 1 The most common form is type 2 diabetes, which accounts for about 97% of all diabetes cases in the population aged 65 years and older.2 Type 2 diabetes is often accompanied by major cardiovascular diseases and, if not treated adequately, can also lead to different long-term complications, including diabetic kidney disease, eye disease, and diabetic foot syndrome.3 However, its association with psychosocial health is often overlooked, despite the increasing evidence linking it to mental health problems, such as depressive symptoms.3Previous studies have extensively examined the link between type 2 diabetes and depressive symptoms or depression. Prospective and retrospective cohort studies point towards an increased risk of depression in participants with diabetes, with ORs typically ranging from 1.15 to 2.55.4–6 The risk of depression or major depressive disorder was found to be particularly strong among adults aged above 60 years and among women.7 8 However, many inconsistencies still exist in the literature, with some well-powered studies reporting no association between diabetes and depressive symptoms or major depression.9 10 Additionally, some studies found significant associations for depressive symptoms only under certain conditions, such as before adjusting for comorbidities11 or only in participants with prior diagnosis of diabetes (vs previously undetected cases).12 Finally, studies that pay particular attention to the older population focus on community-dwellers and impose restrictive inclusion criteria, limiting the participation of higher age groups and people with a disability.Despite considerable attention to depressive symptoms or depression, studies examining the association between type 2 diabetes and other psychosocial aspects like loneliness and low vitality, life satisfaction, satisfaction with one’s activity level, and perceived social support are scarce. However, some studies found type 2 diabetes or the presence of diabetic complications to be associated with lower health-related quality of life, including lower social functioning and vitality.13 14 Additionally, a study found that participants with diabetes were more likely to experience social exclusion and loneliness,15 although another study found that participants with diabetes reported no greater life dissatisfaction than those without diabetes.16In this study, we address this research gap by investigating the association between type 2 diabetes and a range of psychosocial burdens (depressive symptoms, low energy, low life satisfaction, loneliness, low satisfaction with one’s activity level, and perceived low social support) among older adults in Germany. Through its inclusive research design, the present study engages an otherwise hard-to-reach population, including higher age groups and people with a disability.Research design and methods Study design and participants The present study uses data from the Gesundheit 65+study, which is a longitudinal population-based study between June 2021 and April 2023 among persons 65 years and older permanently residing in Germany conducted by the Robert Koch Institute (RKI). The study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. The study design included a baseline survey between June 2021 and April 2022, as well as follow-ups 4, 8 and 12 months later (wave 0, wave 1, wave 2, wave 3, respectively). In addition, at wave 3, a nested health examination was conducted for a subsample of participants. 17The sampling and recruitment procedure is described in more detail elsewhere.18 In brief, baseline participants were selected using a two-stage cluster sampling procedure. In the first stage, 128 primary sampling units (PSUs) were randomly drawn from all municipalities in Germany, stratified by region and the BIK-10 classification (“Gemeindegrößenklassen nach Fläche, Bevölkerung und Bevölkerungsdichte”, a scheme that groups municipalities into 10 categories based on population size and urban–rural structure). In the second stage, within each PSU, sex-stratified and age-stratified (65–79 and ≥80 years) random samples of residents were drawn from local population registers.17 Sampled individuals were invited to participate in the study using a sequential mixed-mode contact design involving personalized postal letters, telephone calls, and face-to-face contacts via home visits for as long as possible given the pandemic situation. A mixed-mode data collection design was applied, offering a paper or online questionnaire, or an interview on the phone or during a home visit, to facilitate study participation. In the event that participants were unable to answer the survey on their own, proxy participation was allowed. Individuals not able to understand the German language, as well as those who died/moved before the data collection started, were excluded.17Of 12,248 individuals who were invited to participate in the study, 307 were excluded based on the above-mentioned criteria. A total of 3,694 individuals fully or partially answered the baseline survey, representing a response rate of 31% (following the American Association for Public Opinion Research standards). Among them, 530 did not respond to wave 1, 583 to wave 2, and 745 to wave 3. Over all waves, the majority of participants completed paper questionnaires (90%), while a smaller proportion completed online questionnaires (8%), in-person interviews (1%) and telephone interviews (1%). Proxy participation occurred in 8% of all cases. Additionally, 1,493 participants underwent the nested examination survey, which included a physical examination and medication review. Although the nested health examination provides important information on various health indicators (eg, resting blood pressure, handgrip strength), in our study, it was used exclusively to help determine participants’ type 2 diabetes status based on medication use. Overall, 147 individuals with missing data for their type 2 diabetes status were excluded, resulting in a final sample of 3,547 individuals with 12,514 observations across all study waves.Type 2 diabetes Diabetes status was assessed at baseline and wave 3 within a list of self-reported chronic conditions according to the European Health Interview Survey (EHIS). 18 Participants were asked: “Have you had any of the following diseases or health conditions in the last 12 months?” (answer options: yes/no). In addition, the subsample of participants who underwent the examination at wave 3 were asked about their medication use.Diabetes was assumed if it was reported at baseline or wave 3. For participants with examination, diabetes was assumed if diabetes was reported at baseline or wave 3 or diabetes medication use was documented at wave 3 (Anatomical Therapeutic Chemical Classification, ATC code A10). Although the diabetes type was not specifically recorded, we assume it to be type 2 diabetes given the older age of the study participants.2Psychosocial burdens Depressive symptoms in the last 2 weeks were measured at all study waves with the two-item Patient Health Questionnaire (PHQ-2), which assesses two core symptoms of major depressive disorder: anhedonia (loss of interest or pleasure in activities) and low mood (feelings of sadness, melancholy, or hopelessness). Participants were asked how frequently they have experienced these symptoms during the past 2 weeks, with responses scored on a 4-point scale from “not at all” to “nearly every day” (total score range: 0–6).19 A total score of 3 or higher was defined as positive screening for depressive symptoms.Low energy was measured at baseline and wave 3 using a single item drawn from the Short Form 36 Health Survey to assess health-related quality of life.20 Participants were asked, “How much of the time during the past four weeks did you have a lot of energy?”. Those responding “never” or “rarely” were classified as having low energy.General life satisfaction was assessed at all study waves with the question “How satisfied are you currently, all in all, with your life?” (answer categories: 0—completely dissatisfied to 10—completely satisfied).21 A score of 6 or lower was considered low or middle life satisfaction.22Loneliness was assessed at all study waves with the three-item Revised University of California, Los Angeles (UCLA) Loneliness Scale, which included missing company, feeling marginalized, and feeling isolated from others (range of the total score: 3–9).23 A total score of 6 or higher was considered indicative of loneliness.24Satisfaction with activity level was assessed at all study waves using a single item from the World Health Organization Quality of Life - Older Adults (WHOQOL-OLD) questionnaire. Participants were asked, “How satisfied are you with your level of activity?”.25 Participants were grouped into three categories: (1) satisfied/very satisfied, (2) neither satisfied nor unsatisfied, and (3) unsatisfied/very unsatisfied.Perceived social support was measured at all study waves using the three-item Oslo-3 Social Support Scale, which assesses the perception of social network size, the sympathy and interest of others, and the receipt of practical help from neighbors (total score range: 3–14). A total score less than 9 was considered low support, from 9 to 11 moderate support, and from 12 to 14 high support.26Covariates The following covariates were considered based on their inclusion in previous studies and the availability of information in the survey. 17 Age group (65–79, ≥80 years), sex, and educational level defined by the Comparative Analysis of Social Mobility in Industrial Nations classification system (categorized into primary, secondary, and tertiary level) were included as socio-demographic covariates.27 Additionally, the level of physical activity (inactive, low (<2 hours weekly), active (≥2 hours weekly)) and of body mass index (no obesity (<30 kg/m2), obesity (≥30 kg/m2)) were considered as lifestyle-related variables.28 29 Finally, the presence of at least one cardiovascular comorbidity (including ischemic heart disease, stroke, or chronic sequelae thereof, based on the EHIS list described above) and of cancer were included as comorbidity covariates.30 While sex and education were considered only at baseline, all other variables were considered at their respective collection points throughout the study waves.Statistical analysis Despite the longitudinal design of the Gesundheit 65+ survey, our study does not focus on modeling change over time. Rather, clustering was used to account for repeated measurement within participants. The complex sampling design of the Gesundheit 65+ survey was accounted for by using appropriate survey procedures within the statistical software R (V.4.4.1) for all analyses. These procedures accounted for the clustering of the data (ie, PSU and repeated measures within the same individuals) and included a weighting factor correcting for deviations from the population structure with respect to sex, age, region, municipality size, education and loss to follow-up. 17 The proportion of missing data was below 5% for all variables, except for obesity (6%), presence of cancer (11%) and cardiovascular comorbidities (16%). Multivariate imputation by chained equations (MICE) was used to impute all missing variables of the dataset, except for diabetes status.First, bivariate associations between type 2 diabetes and the different psychosocial variables were assessed using Rao-Scott adjusted Pearson’s χ2 tests. Percentages with 95% CIs and p values are reported. P values equal or inferior to 0.05 were considered significant. Second, multivariate associations with type 2 diabetes were examined using binomial and multinomial logistic regressions. Multinomial logistic regressions were chosen due to the violation of the proportional odds assumption, assessed graphically using binary splits. Separate models were fitted for each psychosocial variable. In addition to unadjusted models (M1), multivariable models were adjusted for socio-demographic variables (M2), additionally for lifestyle-related variables (M3), and additionally for comorbidities (M4). ORs with 95% CI are reported, representing the odds of reporting a psychosocial burden at a single assessment occasion (ie, each study wave is considered a separate unit of analysis; the OR is not cumulative across waves).To test the robustness of our findings, sensitivity analyses were performed with complete case data (sensitivity analysis 1, online supplemental table S1) as well as with MICE imputed type 2 diabetes status data (sensitivity analysis 2, online supplemental table S2). Further, analyses were run by excluding observations based on proxy participation (sensitivity analysis 3, online supplemental table S3) and with baseline only data (cross-sectional design; sensitivity analysis 4, online supplemental table S4). Finally, we tested whether the association of psychosocial variables with type 2 diabetes was modified by age, sex, or study wave by adding the respective interaction terms (product term) of diabetes with age group, sex and study wave into the regression models (interaction analysis, online supplemental table S5).SP110.1136/bmjdrc-2026-006056.supp1Supplementary dataData and resource availability statement The authors state that some access restrictions apply to the data on which the results are based. The data set cannot be made publicly available because the informed consent of the study participants does not cover the release of the data. The data set underlying the results is archived at the Research Data Centre of the Robert Koch Institute and can be accessed by researchers on reasonable request (dataset: Gesundheit 65+, V.4, year 2025). The data can be accessed on site in the Secure Data Centre of the Research Data Centre of the RKI. Requests can be sent by email to fdz@rki.de.Results Characteristics of the study population Table 1 summarizes the baseline characteristics of the study population stratified by type 2 diabetes status. In total, 722 (21%) of the 3,547 participants had type 2 diabetes. Participants with type 2 diabetes were more often male and in the higher age group than those without type 2 diabetes. On average, they had less often received tertiary level education, were more often obese and physically inactive, and had a higher prevalence of cardiovascular comorbidities. However, the prevalence of cancer did not differ between the two groups. Additionally, participants with type 2 diabetes were more likely to rely on proxy participation to answer the survey.Table 1Baseline characteristics of the study participants by type 2 diabetes status among 3547 participantsCharacteristicNo type 2 diabetes (N=2825)Type 2 diabetes (N=722)P value†n% (95% CI)*n% (95% CI)*Sex<0.001 Male144242 (40 to 43)42552 (47 to 57) Female138358 (57 to 60)29748 (43 to 53)Age group (years)<0.05 65–79115165 (63 to 67)24760 (56 to 65) 80+167435 (33 to 37)47540 (35 to 44)Education level<0.05 Primary131952 (49 to 55)40058 (52 to 64) Secondary78232 (29 to 35)19632 (27 to 37) Tertiary72416 (14 to 18)12610 (7 to 13)Obesity<0.001 No237981 (79 to 83)50267 (62 to 72) Yes44619 (17 to 21)22033 (28 to 38)Level of physical activity<0.001 Inactive114638 (35 to 41)40855 (49 to 60) Low98735 (33 to 37)21232 (27 to 37) Active69227 (25 to 30)10213 (10 to 17)Cardiovascular comorbidities<0.001 No225482 (80 to 85)48771 (66 to 76) Yes57118 (15 to 20)23529 (24 to 34)Cancer0.6 No237486 (84 to 88)60286 (82 to 89) Yes45114 (12 to 16)12014 (11 to 18)Participant answered the questionnaire in their own name<0.001 Yes260793 (92 to 94)63188 (84 to 91) No (answered by proxy)2187 (6 to 8)9112 (9 to 16)*Weighted prevalence and 95% CI.†Pearson’s χ2: Rao and Scott adjustment.Prevalence of psychosocial burdens in relation to type 2 diabetes Table 2 shows the prevalence of the different psychosocial burdens stratified by type 2 diabetes status. Participants with type 2 diabetes generally reported higher levels of psychosocial burdens than those without type 2 diabetes. Specifically, they reported more frequently depressive symptoms (15 vs 8%), low energy (26 vs 16%) and loneliness (29 vs 21%). Furthermore, they were less likely to be highly satisfied with their lives (62 vs 74%), less often satisfied with their activity level (49 vs 62%), and more likely to report low levels of perceived social support (24 vs 19%).Table 2Prevalence of psychosocial burdens by type 2 diabetes status among 3,547 participants across all study waves (N=12,514 observations)Psychosocial burdenNo type 2 diabetes(N=9,992 observations)Type 2 diabetes(N=2,522 observations)P value†n% (95% CI)*n% (95% CI)*Depressive symptoms<0.001 No926692 (91 to 93)217585 (82 to 88) Yes7268 (7 to 9)34715 (12 to 18)Low energy level<0.001 No835684 (82 to 86)181774 (70 to 77) Yes163616 (14 to 18)70526 (23 to 30)Low or middle life satisfaction<0.001 No744174 (72 to 76)158662 (58 to 66) Yes255126 (24 to 28)93638 (34 to 42)Loneliness<0.001 No789479 (77 to 81)178871 (67 to 75) Yes209821 (19 to 23)73429 (25 to 33)Satisfaction with activity level<0.001 Satisfied/very satisfied626462 (60 to 64)125749 (45 to 54) Neither satisfied nor unsatisfied212421 (20 to 23)69729 (25 to 32) Unsatisfied/very unsatisfied160417 (15 to 19)56822 (19 to 26)Perceived social support<0.005 High251926 (24 to 28)48521 (18 to 24) Moderate567455 (53 to 58)145055 (52 to 59) Low179919 (17 to 21)58724 (20 to 28)*Weighted prevalence and 95% CI.†Pearson’s χ2: Rao and Scott adjustment.Association between diabetes and psychosocial burdens Table 3 shows the results of the regression models, ran separately for each psychosocial burden. For the binary psychological burdens, without any adjustment (M1), type 2 diabetes was associated with higher odds of having depressive symptoms, low energy, loneliness, and low or middle life satisfaction. After adjusting for socio-demographic variables (M2), there was only minimal change in the ORs. The associations were attenuated but remained significant after further adjustment for lifestyle-related variables (M3) and comorbidity variables (M4). In the fully adjusted model (M4), type 2 diabetes was associated with a 60% higher odds of depressive symptoms, 41% higher odds of low or middle life satisfaction, 37% higher odds of low energy, and 31% higher odds of loneliness.Table 3ORs (95% CI) for the association between type 2 diabetes and psychosocial burdens among 3,547 participants (N=12,514 observations)Psychosocial burdenM1M2M3M4OR95% CIOR95% CIOR95% CIOR95% CIDepressive symptoms NoRefRefRefRef Yes2.151.59 to 2.892.051.52 to 2.761.751.30 to 2.361.601.18 to 2.17Low energy level NoRefRefRefRef Yes1.961.52 to 2.531.861.44 to 2.411.521.17 to 1.971.371.06 to 1.77Low or middle life satisfaction NoRefRefRefRef Yes1.751.41 to 2.161.691.38 to 2.081.511.22 to 1.861.411.15 to 1.73Loneliness NoRefRefRefRef Yes1.521.18 to 1.961.501.18 to 1.921.391.09 to 1.781.311.03 to 1.66Satisfaction with activity level Satisfied/very satisfiedRefRefRefRef Neither satisfied nor unsatisfied1.721.38 to 2.141.701.37 to 2.111.461.17 to 1.811.381.11 to 1.72 Unsatisfied/very unsatisfied1.631.26 to 2.111.601.24 to 2.071.391.06 to 1.811.260.98 to 1.63Perceived social support HighRefRefRefRef Moderate1.241.00 to 1.541.180.94 to 1.471.180.94 to 1.481.150.92 to 1.44 Low1.601.20 to 2.131.491.11 to 2.011.371.02 to 1.831.320.99 to 1.76Model 1: unadjusted; Model 2: adjusted for sex, age and education; Model 3: additionally adjusted for obesity and level of physical activity; Model 4: additionally adjusted for cardiovascular comorbidities and cancer.Ref, reference group.For the ordinal psychological burdens, without any adjustment (M1), type 2 diabetes was associated with higher odds of reporting being “neither satisfied nor unsatisfied” (middle category) and “unsatisfied” (low category) with one’s activity level and of reporting low perceived social support, whereas no significant association was found for moderate perceived social support. The associations were attenuated but remained significant for the middle and low category of satisfaction with one’s activity level and low perceived social support after adjustment for the socio-demographic (M2) and lifestyle-related variables (M3), but lost significance except for the middle category of satisfaction with one’s activity level after adjusting for the comorbidity variables (M4). In the fully adjusted model (M4), type 2 diabetes was associated with a 38% higher odds of reporting being neither satisfied nor unsatisfied with one’s activity level.Sensitivity analyses and interaction The sensitivity analyses supported the main findings, with similar results in the analyses with complete case data ( online supplemental table S1), with imputed data of type 2 diabetes status (online supplemental table S2), without data based on proxy participation (online supplemental table S3), and with baseline data only (online supplemental table S4) with one exception. In the sensitivity analysis using baseline data, the associations between type 2 diabetes and loneliness, low levels of energy, life satisfaction, and being neither satisfied nor unsatisfied with one’s activity level were no longer significant in the fully adjusted models (M4) (online supplemental table S4).No significant interaction for sex, age group, or study wave was found for the association between type 2 diabetes and the different psychosocial burdens (online supplemental table S5).Conclusions In the present population-based study among adults aged 65 years and older in Germany, we found significant associations between type 2 diabetes and several psychosocial burdens. In the presence of type 2 diabetes, the odds of depressive symptoms were increased by 60%, the odds of reporting low or middle life satisfaction, low energy, and being neither satisfied nor unsatisfied with one’s activity level were increased by about 40%, and the odds of reporting loneliness were increased by about one-third. In the model with adjustment for socio-demographic, lifestyle-related and comorbidity variables, only the association between type 2 diabetes and perceived social support did not remain significant.Similar to previous studies on depression, participants with type 2 diabetes were about two times as likely to report depressive symptoms.5 6 31 Although the strength of the association decreased after adjusting for the presence of cardiovascular comorbidities and cancer, it remained significant, contrary to some previous studies that suggested the association could be entirely explained by comorbidities.11 The mechanisms underlying the association between type 2 diabetes and depressive symptoms remain unclear; however, the persistent association even after adjustment for comorbidities suggests that metabolic and inflammatory pathways intrinsic to diabetes could be involved in this relationship. Indeed, chronic inflammation and insulin resistance in diabetes have been linked to altered brain function and mood regulation, which could contribute to depressive symptoms.32 Additionally, the daily burden of diabetes self-management (eg, dietary restrictions and insulin administration) could also explain the association.33Our study found that the psychosocial burdens of type 2 diabetes go beyond depressive symptoms. Similar to previous studies reporting loneliness, reduced vitality and social functioning among individuals with diabetes, we found significant associations between diabetes and loneliness, low energy, and being neither satisfied nor unsatisfied with one’s activity level.13–15 34 These associations were not significant in the sensitivity analysis using baseline data only, likely due to lower statistical power. Furthermore, we found that type 2 diabetes was associated with higher odds of reporting being unsatisfied with one’s activity level and of reporting low perceived social support, but only before adjusting for comorbidity variables. The associations remained nearly significant (95% CI lower bound 0.99) in the sensitivity analysis using imputed type 2 diabetes status data, suggesting that the lack of significance in some models may be due to lower statistical power. Finally, type 2 diabetes was associated with higher odds of reporting low or middle life satisfaction, in contrast to findings by Grigg et al, who observed no such association, possibly due to the different study population (aged 17 years and above living in the Bella Coola Valley, Canada) or the smaller number of participants (N=971).16The bidirectionality of the association between type 2 diabetes and depression has been well documented.5 However, similar conclusions have not yet been made for the other psychosocial burdens. While our study demonstrates a significant association between type 2 diabetes and several other psychosocial burdens, longitudinal studies are needed to better understand the directionality of this association.Finally, similar to previous studies reporting none or very little differences in the associations with regard to sex, we observed no significant interactions by sex.5 35 We also observed no significant interactions by age group or study wave. This indicates that the associations between type 2 diabetes and different psychosocial burdens were consistent across sex and age groups, and did not change over the study period.Strengths and limitations Our study has several strengths. Most importantly, the study design was made as inclusive as possible, offering several possibilities for participation (including proxy participation) and printing survey questions in large print to ensure even the higher age groups and people with a disability could be included. The target population was not limited to city-dwellers, and efforts were made to include people living in nursing homes. The only restriction on participation was sufficient German language proficiency. 17 As a result, this study provides valuable insights into the health challenges faced by a population that is typically hard to reach and under-represented in research, and findings are broadly generalizable to the older population in Germany.36 Additionally, unlike previous studies that have primarily focused on depressive symptoms or depression, our survey collected data on a broad range of psychosocial burdens. This allows for a more comprehensive understanding of the psychosocial aspects associated with diabetes.However, our study also has limitations. First, our study design does not allow us to draw conclusions about the causal relationship between type 2 diabetes and psychosocial burdens. While we demonstrate significant associations, the directionality of these relationships and the underlying mechanisms remain unclear. Second, diabetes status was self-reported, and we did not explicitly differentiate between type of diabetes, nor did we collect additional information on diabetes severity (eg, glycemic control), which could influence psychosocial burdens.37 Regarding the type of diabetes, we assumed it to be type 2 diabetes given that, in the population aged 65 years and above, about 97% of all diabetes cases in Germany are type 2 diabetes.2 Further, undiagnosed diabetes could not be accounted for due to the lack of blood samples, possibly resulting in misclassification. Third, although the survey assessed a wide range of psychosocial burdens, many variables relied on single-item indicators or shortened versions of validated scales (eg, PHQ-2), which may not fully capture the complexity of these burdens. However, it is important to note that the PHQ-2 has shown excellent psychometric properties and high reliability, making it a robust tool for screening depressive symptoms. In fact, studies have demonstrated that the PHQ-2 is comparable to the PHQ-9 in identifying major depressive disorder, with the added benefit of being a more time-efficient option for initial screening.38 Furthermore, satisfaction with one’s activity level relied on a single item of the WHOQOL-OLD questionnaire. This item was phrased in an ambiguous manner and did not clarify the specific nature of “activity”. Participants may have interpreted the item differently (ie, social activity or physical activity), leading to measurement error and potentially reducing the ability to detect significant associations. Future studies should distinguish more clearly between social participation and physical activity by using separate, conceptually precise measures. Finally, although our study examines psychosocial burdens as separate outcomes by separate models, they are not entirely independent constructs. For example, loneliness has been shown to be closely associated with depressive symptoms in older adults.39 Consequently, some degree of overlap between psychosocial burdens should be acknowledged when interpreting the results.Public health relevance and implications for future research This study showed that older adults with type 2 diabetes were more prone to psychosocial burdens than those without type 2 diabetes. This highlights the importance of integrated care models that routinely screen for and address psychosocial burdens among persons with diabetes. This is especially important because psychosocial factors, such as depressive symptoms, can reduce adherence to diabetes self-management, increasing the risk of glycemic instability and long-term complications. 40 Integrated care models, such as in the Pathways Study in Washington State, have shown that combining diabetes and depression treatment in primary care settings can improve both depression outcomes and glycemic control.41 Similarly, a psychodynamic therapy program (Psychosomatic Treatment Program for People with Diabetes, Psy-PAD) aiming to improve glycemic control and decrease diabetes-related distress and depressive symptoms yielded promising results in Germany.42While our study demonstrates that type 2 diabetes is associated with psychosocial burdens, the mechanisms underlying this relationship remain unclear. Studies specific to depressive symptoms and depression suggest several possible explanations, including inflammatory processes32 and the burden of living with a chronic illness alongside complex care management demands.33 Understanding the mechanisms behind this association is essential to develop more effective treatment strategies, as current interventions may not address key biological or behavioral factors.Finally, while depressive symptoms or depression have been the primary focus of most studies examining the psychosocial burden of diabetes, other psychosocial aspects remain underexplored. Future research should expand the focus to include a broader range of psychosocial challenges, which would enable the design of more comprehensive care models that address the full spectrum of needs in individuals with type 2 diabetes.