Exploring the correlates of diabetes-related stigma in young and middle-aged adults with type 2 diabetes: A cross-sectional study

Abstract

Diabetes-related stigma is prevalent among young and middle-aged adults with type 2 diabetes mellitus (T2DM), particularly in Chinese cultural contexts where psychological distress is often somatized. This study aimed to investigate the prevalence of diabetes-related stigma and identify modifiable correlates among working-age adults with T2DM to inform targeted clinical interventions. A total of 180 patients with T2DM aged 18 to 59 years were recruited using stratified sampling based on age, education level, and disease duration. Data were collected using the Chinese version of the type 2 diabetes stigma assessment scale and a sociodemographic questionnaire. Statistical analyses included independent t-tests, 1-way analysis of variance, and hierarchical multiple linear regression. Participants demonstrated moderate-to-high levels of diabetes-related stigma, with the highest scores observed in the blame and judgment dimension (mean item score 3.17 ± 0.44). The total stigma score was 56.14 ± 8.05. Hierarchical regression analysis identified 5 independent predictors of stigma: lower educational level (β = −0.386, P < .001), younger age (18–44 years), lower household income (<3000 RMB/mo), shorter disease duration (<5 years), and multiple comorbidities (≥2 conditions). The regression model explained 38% of the variance in stigma levels (R² = 0.38, F = 28.066, P < .001). Diabetes-related stigma among young and middle-aged adults with T2DM is moderate to high, particularly in younger, less educated, and financially constrained patients. Healthcare providers should integrate routine stigma assessment into clinical practice and develop tailored interventions addressing age-specific, socioeconomic, and disease-related factors to improve patient outcomes and quality of care.

1. Introduction

Diabetes mellitus has become a defining global public health challenge in the twenty-first century, with the International Diabetes Federation projecting a worldwide prevalence of 783 million by 2045, including 141 million cases in China.^[1] The fastest growth occurs among young and middle-aged adults (15–59 years), whose role as primary breadwinners amplifies their metabolic and psychosocial burdens. Disease-related stigma, marked by shame, concealment, and anticipated discrimination.^[2] Type 2 diabetes mellitus (T2DM) involves insulin resistance and progressive β-cell dysfunction, and requires lifelong dietary restrictions, continuous glucose monitoring, and prevention of complications. This regimen generates stress, with 47.3% of patients avoiding social situations due to insulin injections or self-monitoring, and 34.1% experiencing self-deprecation linked to weight changes.^[3] The World Health Organization classifies this self-worth erosion as “stigma,” manifesting as fearful concealment of complications, avoidance of public insulin administration, and internalized judgments of “laziness.”^[4] In China, diabetes-related stigma is more prevalent than in Western populations due to cultural tendencies to somatize psychological distress, attributing emotional suffering to bodily dysfunction rather than to social devaluation.^[5,6] Validated assessment tools and culturally attuned interventions are scarce and existing scales inadequately capture the interplay between social exclusion and self-stigma among young and middle-aged adults.^[4,6,7]

2. Materials and methods

This study conducted a survey to investigate the prevalence and identify modifiable predictors of diabetes-related stigma among working-age adults with type 2 diabetes using the diabetes stigma assessment scale (DSAS-2) in a clinical population.^[8]

2.1. Study design

This cross-sectional study was conducted between November 2023 and February 2024. The reporting adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

2.2. Study setting and sampling method

A questionnaire survey was conducted on 180 young and middle-aged adults with type 2 diabetes mellitus at a tertiary hospital in Chongqing, China. Participants were recruited using stratified sampling based on three key characteristics: age group (18–44 vs 45–59 years), educational level (junior high school or below, senior high school/vocational diploma, college, or above), and disease duration (1–5, 6–10, and >10 years). Proportional allocation was applied according to the distribution of these characteristics in the hospital’s electronic health records.

2.3. Inclusion criteria

1. Confirmed diagnosis of T2DM for ≥6 months.

2. Aged 18 to 59 years.

3. Ability to communicate independently and complete the questionnaire.

4. Willingness to provide informed consent.

2.4. Exclusion criteria

1. Severe acute complications (e.g., diabetic ketoacidosis, hyperosmolar hyperglycemic state).

2. Cognitive impairment or psychiatric disorders.

3. Comorbidities with life-threatening conditions (e.g., advanced cancer, end-stage renal disease).

2.5. Data collection tools and procedures

Information was collected using a structured questionnaire developed based on literature review and validated by the research team. The questionnaire included sections on sociodemographic characteristics and the diabetes acceptance and stigma assessment. The surveys were conducted face-to-face by trained interviewers fluent in both Mandarin and the local dialect.

2.6. Data analysis

Statistical analyses were performed with the use of IBM SPSS Statistics for Windows (version 29.0; IBM corp., https://www.ibm.com/spss) software, and data that were missing by more than 10% were excluded. Census data are presented as frequencies and percentages. Measurement data were expressed as mean and standard deviation. t test or analysis of variance was used to compare the evaluation value of stigma in young and middle-aged patients with type 2 diabetes with different characteristics. Multiple linear regression was used to analyze the influencing factors of stigma in young and middle-aged patients with type 2 diabetes.

2.7. Ethical considerations

This study was approved as exempt from ethical review by the hospital’s scientific research management department for being a low-risk observational study. Only anonymous data were collected; no interventions were used. Informed consent was obtained from all participants. Data were encrypted and stored securely, and all procedures complied with the Declaration of Helsinki.

3. Results

3.1. Basic information of young and middle-aged patients with type 2 diabetes mellitus

There were 88 males (48.89%) and 92 females (51.11%), and the age range was 18.0 to 59.0 years. Among them, 102 (56.67%) were 45 to 59 years old, 50 (27.78%) had college education or above, 17 (9.44%) had a per capita monthly family income of 5000 yuan or more, and 107 (59.44%) had a family history of diabetes. Other relevant basic information is detailed in Table 1.

Table 1.

Basic information of young and middle-aged patients with type 2 diabetes (N = 180).

Variables	Groups	Numbers	Percentage
Gender	Male	88	48.89
	Female	92	51.11
Age	18–44 yr old	78	43.33
	45–59 yr old	102	56.67
Education background	Junior high school and below	71	39.44
	Technical secondary school/high school	59	32.78
	College or above	50	27.28
Marital status	Married	138	76.67
	Unmarried	34	18.89
State of work	Get divorced	8	4.44
	On the job	74	41.11
	Go out of retired	64	35.56
	Unempolyed	72	23.33
Monthly income per capita	<3000	65	36.11
	3000–5000	98	54.44
	≥5000	17	9.44
Payment methods of medical treatment expenses	Out of pocket	13	7.22
	Reimbursement of expenses	167	92.78
Family history of diabetes	None	73	40.56
Glucose-lowering REGIMENS	Yes	107	59.44
	No medication used	27	15.00
	Oral hypoglycemic agents	80	44.44
	Insulin	27	15.00
	Oral medication combined with insulin	46	25.56
Comorbidity	None	42	25.56
	1 species	97	53.89
	2 or more	41	22.78
Duration of diabetes	1–5 yr	49	27.22
	6–10 yr	99	55.00
	>10 yr	32	17.78

3.2. Current status of stigma

The highest average score of blame and judgment items was (3. 17 ± 0.44; Table 2). The results of single factor analysis showed that there were statistically significant differences in the level of stigma among young and middle-aged patients with type 2 diabetes mellitus with different ages, education levels, monthly incomes, years of illness, working status, diabetes complications and different ways of medical expenses (P < .05). The stigma scores of young and middle-aged patients with type 2 diabetes with different characteristics are shown in Table 3.

Table 2.

Stigma scores of young and middle-aged patients with type 2 diabetes.

Items	Number of items (n)	scores (x ± s)	Item mean score (x ± s)
Guilt and judgment	7	22.17 ± 3.08	3.17 ± 0.44
Self-stigma	6	17.62 ± 3.75	2.94 ± 0.63
Perceived discrimination	6	16.36 ± 3.40	2.73 ± 0.57
Total	19	56.14 ± 8.05	2.94 ± 0.43

Table 3.

Comparison of the stigma of type 2 diabetes stigma in different characteristics.

Items	Scores (x± s)	Value of statistics (F/t)	P
Age		t = 4.194	.022
18–44	58.90 ± 8.62
45–59	54.04 ± 6.92
Gender		t = −0.958	.677
Male	55.56 ± 7.77
Female	56.71 ± 8.32
Education background		F = 23.607	<.01
Junior high school and below	60.56 ± 7.38
Technical secondary school/high school	54.36 ± 6.25
College or above	51.98 ± 7.92
Marital status		F = 0.403	.669
Married	56.30 ± 8.33
Unmarried	55.18 ± 7.42
Get divorced	57.43 ± 6.57
State of work		F = 94.077	<.01
On the job	56.81 ± 5.40
Go out of retirement	49.63 ± 5.21
Unempolyeed	64.90 ± 6.59
Monthly income per capita (Yuan)		F = 24.232	<.01
<3000	61.08 ± 7.90
3000–5000	53.52 ± 6.50
>5000	52.41 ± 7.92
Payment methods for medical expenses		T = 18.032	<.01
Out of pocket	69.31 ± 4.2
Reimbursement of expenses	55.12 ± 7.34
Family history of diabetes		t = −1.179	.269
Yes	55.56 ± 7.71
None	57.00 ± 8.50
Duration of diabetes
1–5 yr	61.53 ± 8.33	F = 18.779	<.01
5–10 yr	54.57 ± 6.97
>10 yr	52.78 ± 6.90
Comorbidity		F = 4.080	.019
None	54.26 ± 9.57
1 species	55.73 ± 6.95
2 or more	59.05 ± 8.20
Glucose-lowering REGIMENS		F = 1.585	.195
No medication used	58.00 ± 8.90
Oral hypoglycemic agents	56.10 ± 7.34
Insulin	53.44 ± 8.18
Oral medication combined with insulin	56.72 ± 8.47

3.3. Univariate analysis

The total score of stigma in young and middle-aged patients with type 2 diabetes was taken as the dependent variable, and the statistically significant items in univariate analysis were taken as the independent variables for multiple linear regression analysis. The assignment method of independent variables is shown in Table 4, and the results of regression analysis are shown in Table 5. There were statistically significant differences in the level of stigma among patients with different ages, education levels, family per capita monthly income, working status, duration of diabetes and number of complications (P < .05).

Table 4.

Assignment methods of independent variables.

Item	Mode of assignment
Age (years old)	18–44 = 0;45–59 = 1
Education background	Junior high school and below = 1; high school/technical secondary school = 2, college or above = 3
State of work
Monthly household income per capita (Yuan)	<3000 = 1; 3000–5 000 = 2; >5000 = 3
Payment methods for medical expenses	Out of pocket = 1; reimbursement of expenses = 2
Duration of diabetes (years old)	1–5 = 1; 6–10 = 2; >10 = 3
Comorbidity	None = 1; 1 piece = 2; 2 or more = 3

Table 5.

Results of multiple linear regression analyses of factors affecting stigma in young and middle-aged patients with type 2 diabetes mellitus.

Items	B	SE	β	t	P
Term of constant	68.338	2.251	–	30.357	<.01
Education background	−3.822	0.590	−0.386	−6.482	<.01
State of work	2.103	0.615	0.205	3.420	.001
Monthly household income per capita	−3.719	1.053	−0.246	−3.018	.003
Duration of diabetes	−1.740	0.976	−0.144	−1.783	.076

R² = 0.391; △R² = 0.377; F = 28.066; P = .000.

4. Discussion

The stigma of type 2 diabetes among young and middle-aged individuals is in the upper-middle range. Our study found a total stigma score of 56.14 ± 8.05, higher than previous reports, indicating an intermediate to high level of stigma.^[8] This may be due to the unique cultural and demographic context, with participants facing multiple pressures from their work, family, and society. Young adults (18–25 years old) face additional challenges in education, employment, and disease management, leading to higher stigma levels.^[9–11] Factors influencing stigma levels included age, education, family financial burden, diabetes duration, and comorbidities. Younger patients (18–44 years) had higher self-stigma scores (P < .05), likely because of heightened social self-consciousness and lifestyle perceptions.^[12–14] Lower education levels were associated with higher stigma scores (P < .01), highlighting the need for targeted interventions.^[14] Financial strain, particularly among unemployed individuals and those with low monthly incomes (<3000 RMB), exacerbated stigma (P < .01), likely due to increased treatment burden. Patients with a diabetes duration of 1 to 5 years had higher stigma scores (P < .01), possibly because of inadequate psychological adaptation and social image concerns. Comorbidities also increased stigma levels (P < .05), likely due to reduced self-care capacity and increased family burden.^[15–18]

While our multiple linear regression model explained approximately 38% (R² = 0.38, P < .001) of the variance in stigma levels among young and middle-aged adults with T2DM, the substantial unexplained variance suggests additional influencing factors warrant exploration. Social support quality may play a critical role, as inadequate emotional or instrumental support could exacerbate feelings of isolation and stigma, particularly among patients facing socioeconomic challenges. Psychological resilience, which enables adaptive coping strategies, might buffer against stigma by fostering self-efficacy and reducing negative self-appraisal, yet was not measured in this study. Patient-provider communication quality also merits attention, as stigmatizing attitudes or poor information exchange during clinical encounters could intensify perceived discrimination. Cultural factors, such as China’s “face” culture, may further complicate stigma experiences by discouraging disease disclosure and reinforcing internalized shame. Additionally, disease-related variables like glycemic control variability, treatment regimen complexity, and comorbid mental health conditions (e.g., depression) could independently contribute to stigma but were not captured in our model. Future research should employ mixed-methods approaches, combining quantitative assessments of these factors with qualitative interviews to uncover context-specific stigma mechanisms. Longitudinal designs tracking stigma trajectories alongside psychological and social determinants would clarify causal pathways, while culturally tailored interventions addressing communication barriers, resilience-building, and social support networks could reduce stigma more effectively. Acknowledging these gaps strengthens the translational value of our findings and provides a roadmap for developing holistic, patient-centered stigma reduction strategies.

4.1. Limitation

The exclusive recruitment of participants from a single tertiary care academic medical center in southwestern China engenders non-negligible selection bias. Tertiary institutions function as referral hubs for individuals with advanced metabolic decompensation, refractory hyperglycemia, or multiple macro- and microvascular complications. Consequently, the study cohort likely overrepresented patients with longer disease duration, higher HbA1c levels, and more complex therapeutic regimens compared with the broader T2DM population managed in primary or secondary care settings. This selection bias may have inflated the observed DSAS-2 aggregate score (mean = 56.14 ± 8.05), yielding an upward-biased estimate of disease-related stigma among working-age adults. Tertiary hospitals are geographically and financially inaccessible to substantial segments of the population, particularly rural inhabitants, uninsured migrants, and individuals below the poverty threshold. This socioeconomic truncation distorts the income–stigma gradient, attenuates variance at the lower tail of the distribution, and potentially masks the most extreme experiences of financial-medical hardship and concomitant stigma. Within our sample, a high proportion of patients exhibited a long disease duration and complex treatment regimens, which were independently associated with heightened stigma. Thus, the reported associations may reflect tertiary care selection effects rather than true epidemiological relationships, precluding any causal inference regarding the temporal dynamics of stigma accruals. These biases collectively constrain external validity, rendering our prevalence estimates and predictor coefficients non-generalizable to the national T2DM population. Future investigations should employ multi-level, probability-proportional-to-size sampling that integrates tertiary referral centers, district hospitals, and primary-care/community health stations, supplemented by capture–recapture methods to enroll non-attending individuals. Such a design would enable post-stratification weighting to correct for differential selection probabilities and yield unbiased, nationally representative estimates of the stigma burden and its determinants.

5. Conclusion and recommendations

This study highlights moderate-to-high stigma levels among young and middle-aged T2DM patients and identifies the key predictors. These findings provide evidence for healthcare providers to develop targeted interventions and improve patient outcomes. Integrating stigma assessment into routine clinical practice is essential for enhancing the quality of care. Future research should expand these findings to refine intervention strategies.

Targeted interventions were developed based on the 5 independent predictors of diabetes-related stigma identified in this study: lower education, younger age, shorter disease duration, lower household income, and ≥2 comorbidities. For younger adults (18–44 years), a WeChat Mini-program named “T2DM-Youth Hub” offers a 6-week cognitive-behavioral micro-curriculum with gamified incentives and peer-led support groups facilitated by “diabetes ambassadors.” Additionally, precision social media campaigns on platforms such as TikTok and Little Red Book feature insulin injection challenges to engage this demographic. For individuals with lower educational levels (≤junior high school), a visual-narrative health literacy package was designed, including a 1-page pictorial glycemic diary, “photo-novella” group education sessions, and audio messages in local dialects to ensure accessibility.^[18] For those with a short disease duration (<5 years), immediate post-diagnostic stigma screening using a rapid DSAS-2 tool is recommended, followed by stepped psychosocial support involving narrative nursing interviews, WeChat micro-groups, and referral for cognitive-behavioral therapy if needed. For patients with low household income (<3000 RMB/month), dual financial pathways optimize insurance reimbursement and charity enrollment, complemented by monthly hospital-community workshops integrating metabolic education and psychological counseling. For patients with multiple comorbidities (≥2), a multidisciplinary “3C” model (endocrinology–cardiology–clinical psychology) is employed, featuring automated EHR flagging, stigma heat-maps, and quarterly 1-stop combined clinics.^[19–21] All interventions were integrated into the hospital information system with mandatory DSAS-2 reassessment at 3, 6, and 12 months using interrupted time-series analysis to monitor trends and ensure iterative quality improvement.^[22,23]

Author contributions

Conceptualization: Ruiting Yang.

Data curation: Ruiting Yang.

Formal analysis: Ruiting Yang.

Investigation: Jie Yun.

Methodology: Jie Yun, Ruiting Yang.

Project administration: Jie Yun, Ruiting Yang.

Resources: Jie Yun.

Supervision: Jie Yun.

Visualization: Jie Yun.

Writing – original draft: Ruiting Yang.

Writing – review & editing: Ruiting Yang.