Abstract
Diabetes distress (DD) significantly burdens and negatively impacts self-management and health outcomes in patients with type 2 diabetes (T2D). Early detection and management of DD are crucial for effective T2D management. The Diabetes Distress Assessment System (T2-DDAS) is a recently validated tool for measuring DD, but its psychometric properties in non-Western populations, such as India, have not been evaluated. This study aimed to assess the validity and reliability of T2-DDAS in Indian patients with T2D. Data from 408 T2D patients enrolled in a one-year diabetes management program at the Freedom from Diabetes Clinic in Pune, India, from January to March 2024, were analyzed. Sociodemographic characteristics, medical history, and anthropometric parameters were extracted from existing records. T2-DDAS, previously validated in a Western population, was administered. Statistical analyses included descriptive statistics, Principal Component Analysis (PCA), Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and Structural Equation Modeling (SEM), using IBM SPSS, R, and AMOS software. The T2-DDAS demonstrated good internal consistency (Cronbach’s α = 0.91). CFA yielded satisfactory fit indices (χ² = 1036.92, RMSEA = 0.07, TLI = 0.905, CFI = 0.918), supporting the construct validity of T2-DDAS for the Indian population. The T2-DDAS is a valid and reliable tool for assessing DD in Indian patients with T2D.
Keywords: Validation study, Type 2 diabetes, Reliability, Validity, Diabetes Distress Assessment System, India
Subject terms: Diseases, Endocrinology, Health care, Risk factors
Introduction
India faces a staggering burden of type 2 diabetes (T2D), with its prevalence increasing exponentially over the past decades; from 5.0% in 1980 to 7.3% in 2000 to 8.8% in 20141. The International Diabetes Federation (IDF) estimated that 77 million people in India had diabetes in 2019, with projections reaching 134 million by 20452,3. Effective T2D management is urgently needed with the understanding that management extends beyond glycemic control and recognizing that emotional and psychological well-being impacts the disease4,5.
Diabetes distress (DD), the emotional burden associated with managing T2D, is a significant psychological concern affecting the well-being, treatment adherence, and overall outcomes of people living with diabetes6,7. It is significantly linked to poorer glycemic control, reduced medication adherence, unhealthy diet, and lower physical activity levels, underlining its significant clinical impact6,8–10. In India, approximately one-third (33%) of patients with T2D experience DD6. Recognizing DD as a crucial factor in diabetes management, early identification, and intervention can significantly improve treatment adherence, glycemic control, and quality of life9,11. Therefore, effective assessment tools are essential to identify and address this issue.
Standardized tools, such as the Diabetes Distress Scale (DDS-17)12, are widely used to assess DD. However, in recent years, these tools have not fully captured the impact of contemporary factors, such as evolving social norms, medication advancements, and healthcare access challenges13–15. While diabetes management guidelines now emphasize addressing DD, there is need for more up-to-date, comprehensive, and culturally sensitive assessment tools that provide actionable research and clinical practice data.
Cultural variations significantly influence how people perceive illnesses, express emotions, and interact with healthcare systems, which can affect response to standardized assessment tools16,17. The T2-Diabetes Distress Assessment System (T2-DDAS), designed and validated to assess DD in Western populations18, offers a comprehensive evaluation of core emotional distress related to diabetes and the specific factors contributing to it13. While the T2-DDAS has demonstrated good psychometric properties in Western settings, its applicability in culturally diverse settings such as India remains unexplored.
The T2-DDAS, while validated for Western populations, may not fully capture the diabetes-related distress experienced by culturally diverse populations in India. Therefore, we hypothesize that validation and adaptation of the T2-DDAS for Indian cultural context will result in an accurate assessment of DD, addressing the unique sociocultural factors influencing patient experiences with diabetes. Without such validation, misinterpretation of DD levels and challenges in developing culturally sensitive interventions can arise. Therefore, this study aimed to evaluate the psychometric properties of the T2-DDAS in an Indian population with T2D. Our findings will contribute significantly to the field by determining T2-DDAS’ suitability for identifying and addressing DD in the Indian population.
Methodology
Study design and participants
This validation study included 408 patients with T2D enrolled in an online diabetes management program at the Freedom from Diabetes Clinic in Pune, India, between January 2024 and March 2024.
Ethical consideration
This validation study adhered to the ethical principles outlined in the Declaration of Helsinki. This study was approved by the Freedom from Diabetes Research Foundation - Institutional Ethics Committee (approval number: FFDRF/IEC/2023/5; approved on 15-10-2023). Informed consent was obtained.
Data collection and preparation
Patients were administered the T2-DDAS questionnaire via Google Forms. All responses were encoded with a unique ID in numerical format and stored in Excel spreadsheets for further analysis.
T2-diabetes distress assessment system
The T2-DDAS is a tool designed to measure diabetes-related emotional distress in individuals with T2D. It assesses how patients feel about managing their diabetes, including concerns about glycemic control, treatment adherence, long-term health complications, and the psychosocial impact of the disease. It was developed in 2022 to overcome the limitations of previous measures of DD. It comprises two main parts: the Diabetes Distress CORE, which includes eight items, and the elevated SOURCE, which consists of 21 items. The first eight questions in the CORE assess the intensity of core DD reported by the patient, with higher scores indicating greater intensity. The remaining 21 questions in the SOURCE section evaluate the common sources of DD for adults with T2D. These sources are divided into seven subscales, covering various life aspects of a patient with diabetes: Hypoglycemia (three items), Long-term health (three items), Healthcare provider (three items), Interpersonal Issues (three items), Shame/Stigma (three items), Healthcare Access (three items), and Management Demands (three items). Each source pertains to a specific aspect of living with and managing diabetes that can contribute to DD for a particular patient. A higher score on a particular source indicates a greater impact on the patient’s DD. The response to each item is based on a 5-point Likert scale, rated from 1 to 5 (1: Not A Problem, 2: A Little Problem, 3: A Moderate Problem, 4: A Serious Problem, and 5: A Very Serious Problem) concerning diabetes over the past month. Thus, higher values indicate greater distress. A mean item score of ≥ 2 is considered a level of distress for both the elevated DD CORE and elevated SOURCE, while an overall mean score of < 2.0 indicates no distress13,18.
Statistical analysis
The Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test of sphericity were used to assess the sample adequacy and factorability of the datasets. KMO values closer to 1.0 are deemed optimal, whereas values < 0.5 are considered inadequate19. Descriptive statistics, including the mean, standard deviation, skewness, and kurtosis, were analyzed for each item using IBM SPSS (version 21.0).
Reliability and validity assessment
The internal consistency and reliability of each factor were assessed using Cronbach’s α. To evaluate the quality of the developed questionnaire, both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were conducted using R software version 4.0.2. A scree plot was used to determine the optimal number of factors to be included in the final set. Principal component analysis (PCA) with varimax rotation was applied to all items during the EFA. Model fit for CFA was evaluated using various indices including the Goodness of Fit Index (GFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximations (RMSEA), modified χ2 fit statistics, and the Comparative Fit Index (CFI).
Structural equation modeling
Structural equation modeling (SEM) was conducted using AMOS version 22 to assess the relationship between the structural paths and factors within the DDAS questionnaire.
Results
Baseline characteristics
A total of 408 patients with T2D across 81 cities [51 urban areas (cities/metropolitan regions) and 37 rural or semi-urban areas (districts/smaller towns)] in India were included in this study. The mean age was 49.9 ± 10.0 (years), and the duration of diabetes was 8.6 ± 7.2 (years). Females comprised 41.2% of the study population. Most of the participants (88.7%) were married. In terms of education, nearly half (49.3%) had a graduate degree or higher level of education, and 47.1% were salaried employees. The average weight, body mass index (BMI), and glycated hemoglobin (HbA1c) levels of the participants were 77.3 ± 15.0 (kg), 28.0 ± 5.1 (kg/m²), and 8.2 ± 1.6 (%), respectively. Notably, 74.5% of the patients were taking oral hypoglycemic agents (OHAs) for diabetes management, whereas 12% were drug-naive (never started medications for diabetes). Table 1 provides a detailed breakdown of sociodemographic characteristics and clinical conditions of the study population.
Table 1.
Socio-demographic and medical characteristics (N = 408).
| Variables | Total (%) |
|---|---|
| Age | |
| More than 55 years | 116 (28.4) |
| 45–55 years | 167 (40.9) |
| Less than 45 years | 125 (30.6) |
| Gender | |
| Male | 240 (58.8) |
| Female | 168 (41.2) |
| Marital status | |
| Unmarried | 19 (4.7) |
| Divorced and separated | 9 (2.2) |
| Widowed | 18 (4.4) |
| Married | 362 (88.7) |
| Education | |
| Graduate and above | 201 (49.3) |
| Non-graduate | 194 (47.5) |
| Other—do not wish to share | 13 (3.2) |
| Occupation | |
| Salaried | 192 (47.1) |
| Self employed | 97 (23.8) |
| Retired | 27 (6.6) |
| Other—do not wish to share | 24 (5.9) |
| Homemaker | 68 (16.7) |
| Family history of diabetes | |
| Maternal—yes | 106 (26.0) |
| Paternal—yes | 123 (30.1) |
| Both (maternal and paternal)—yes | 74 (18.1) |
| Siblings—yes | 13 (3.2) |
| None | 92 (22.5) |
| Duration of diabetes | |
| < 5 years | 172 (44.3) |
| 6–10 years | 60 (15.5) |
| > 10 years | 156 (40.2) |
| Diabetes medication status | |
| Oral hypoglycemic agents only | 304 (74.5) |
| Insulin only | 3 (0.7) |
| Both oral hypoglycemic agents and insulin | 52 (12.7) |
| Drug naïve | 49 (12.0) |
| Other comorbidities | |
| Hypertension | 157 (38.5) |
| Dyslipidemia | 189 (46.3) |
| Hypothyroidism | 74 (18.1) |
| Body mass index (BMI) | |
| Normal (18.5 kg/m2–22.9 kg/m2) | 57 (14.2) |
| Overweight (23.0 kg/m2–24.9 kg/m2) | 67 (16.7) |
| Obese (≥ 25 kg/m2) | 278 (69.2) |
| Variables | (Mean ± SD) |
| Weight (kg) | 77.3 ± 15.0 |
| Glycated hemoglobin (HbA1c) (%) | 8.2 ± 1.6 |
| Fasting blood glucose (mg/dl) | 144.0 ± 48.4 |
| Postprandial blood glucose (mg/dl) | 182.7 ± 72.0 |
Descriptive statistics
The DD Scale (Table 2), which consists of eight items, was administered to 408 participants. The mean scores for these items ranged from 2.12 to 2.52, indicating DD. The data did not follow a normal distribution; the skewness values were positive, ranging from 0.46 to 0.86, suggesting slightly skewed distributions towards lower distress levels while the kurtosis values ranged from − 0.73 to -0.34, indicating relatively platykurtic distribution.
Table 2.
Descriptive statistics of the T2-DDAS tool.
| Factors | Items | N | Minimum | Maximum | Mean | Standard deviation | Skewness | Kurtosis |
|---|---|---|---|---|---|---|---|---|
| Diabetes distress | DD1 | 408 | 1 | 5 | 2.32 | 1.176 | 0.682 | -0.348 |
| DD2 | 408 | 1 | 5 | 2.52 | 1.143 | 0.46 | -0.616 | |
| DD3 | 408 | 1 | 5 | 2.26 | 1.197 | 0.615 | -0.683 | |
| DD4 | 408 | 1 | 5 | 2.14 | 1.18 | 0.788 | -0.371 | |
| DD5 | 408 | 1 | 5 | 2.23 | 1.231 | 0.755 | -0.468 | |
| DD6 | 408 | 1 | 5 | 2.12 | 1.289 | 0.864 | -0.427 | |
| DD7 | 408 | 1 | 5 | 2.28 | 1.281 | 0.684 | -0.638 | |
| DD8 | 408 | 1 | 5 | 2.34 | 1.23 | 0.594 | -0.731 | |
| Hypoglycemia | HY1 | 408 | 1 | 5 | 1.55 | 0.929 | 1.783 | 2.606 |
| HY2 | 408 | 1 | 5 | 1.54 | 0.919 | 1.755 | 2.254 | |
| HY3 | 408 | 1 | 5 | 1.56 | 0.92 | 1.658 | 2.008 | |
| Long term health | LH1 | 408 | 1 | 5 | 2.69 | 1.292 | 0.307 | -1.02 |
| LH2 | 408 | 1 | 5 | 2.1 | 1.213 | 0.836 | -0.41 | |
| LH3 | 408 | 1 | 5 | 2.26 | 1.219 | 0.679 | -0.58 | |
| Healthcare provider | HP1 | 408 | 1 | 5 | 2.0 | 1.208 | 1.024 | -0.026 |
| HP2 | 408 | 1 | 5 | 1.65 | 1.021 | 1.609 | 1.893 | |
| HP3 | 408 | 1 | 5 | 1.57 | 0.959 | 1.762 | 2.43 | |
| Interpersonal issues | II1 | 408 | 1 | 5 | 1.91 | 1.183 | 1.178 | 0.358 |
| II2 | 408 | 1 | 5 | 1.68 | 1.048 | 1.539 | 1.523 | |
| II3 | 408 | 1 | 5 | 1.84 | 1.154 | 1.25 | 0.543 | |
| Shame/stigma | SS1 | 408 | 1 | 5 | 1.67 | 1.111 | 1.671 | 1.807 |
| SS2 | 408 | 1 | 5 | 1.35 | 0.791 | 2.5 | 6.025 | |
| SS3 | 408 | 1 | 5 | 1.55 | 0.99 | 1.904 | 2.904 | |
| Healthcare access | HA1 | 408 | 1 | 5 | 1.61 | 0.957 | 1.591 | 1.913 |
| HA2 | 408 | 1 | 5 | 1.83 | 1.11 | 1.216 | 0.538 | |
| HA3 | 408 | 1 | 5 | 1.61 | 1.015 | 1.616 | 1.508 | |
| Management demands | MD1 | 408 | 1 | 5 | 2.3 | 1.274 | 0.634 | -0.707 |
| MD2 | 408 | 1 | 5 | 2.31 | 1.26 | 0.63 | -0.77 | |
| MD3 | 408 | 1 | 5 | 2.63 | 1.288 | 0.274 | -1.057 |
DD1 to DD8 for DD - Diabetes Distress; HY1 to HY3 for HY – Hypoglycemia; LH1 to LH3 for LH - Long-term Health; HP1 to HP3 for HP - Health Provider; II1 to II3 for II - Interpersonal Issues; SS1 to SS3 for SS - Shame Stigma; HA1 to HA3 for HA - Health Access; MD1 to MD3 for MD - Management Demands.
Structural validity: exploratory and confirmatory factor analysis
Exploratory factor analysis
The DDAS had eight items (DD1 to DD8) with factor loadings (Table 3) ranging from 0.70 to 0.83 and a high Cronbach’s alpha of 0.91, indicating good internal consistency. The communalities for these items ranged from 0.71 to 0.79. The Hypoglycemia factor had three items (HY1 to HY3) with factor loadings from 0.61 to 0.63 and Cronbach’s alpha of 0.64. Communalities were high, ranging from 0.77 to 0.86. The remaining factors - Long-term Health (3 items), Health Provider (3 items), Interpersonal Issues (3 items), Shame Stigma (3 items), Health Access (3 items), and Management Demands (3 items) - had factor loadings ranging from 0.56 to 0.82 and Cronbach’s alpha values ranged from 0.67 to 0.89, indicating acceptable to good internal consistency. The Kaiser-Meyer-Olkin measure of sampling adequacy was 0.95, and Bartlett’s test of sphericity was significant (p < 0.001), suggesting that the data were suitable for factor analysis.
Table 3.
Exploratory factor analysis.
| Factors | Items | Factor Loading | Cronbach’s Alpha | Communalities |
|---|---|---|---|---|
| Diabetes distress | DD1 | 0.707 | 0.914 | 0.730 |
| DD2 | 0.785 | 0.780 | ||
| DD3 | 0.746 | 0.763 | ||
| DD4 | 0.759 | 0.751 | ||
| DD5 | 0.832 | 0.794 | ||
| DD6 | 0.821 | 0.756 | ||
| DD7 | 0.78 | 0.724 | ||
| DD8 | 0.776 | 0.713 | ||
| Hypoglycemia | HY1 | 0.629 | 0.64 | 0.863 |
| HY2 | 0.637 | 0.853 | ||
| HY3 | 0.613 | 0.771 | ||
| Long term health | LH1 | 0.777 | 0.76 | 0.795 |
| LH2 | 0.821 | 0.756 | ||
| LH3 | 0.739 | 0.800 | ||
| Healthcare provider | HP1 | 0.786 | 0.89 | 0.789 |
| HP2 | 0.669 | 0.637 | ||
| HP3 | 0.641 | 0.640 | ||
| Interpersonal issues | II1 | 0.778 | 0.79 | 0.754 |
| II2 | 0.616 | 0.666 | ||
| II3 | 0.706 | 0.744 | ||
| Shame/stigma | SS1 | 0.588 | 0.87 | 0.841 |
| SS2 | 0.585 | 0.684 | ||
| SS3 | 0.576 | 0.865 | ||
| Healthcare access | HA1 | 0.641 | 0.67 | 0.684 |
| HA2 | 0.575 | 0.624 | ||
| HA3 | 0.629 | 0.566 | ||
| Management demands | MD1 | 0.669 | 0.78 | 0.682 |
| MD2 | 0.643 | 0.813 | ||
| MD3 | 0.564 | 0.684 |
DD1 to DD8 for DD - Diabetes Distress; HY1 to HY3 for HY – Hypoglycemia; LH1 to LH3 for LH - Long-term Health; HP1 to HP3 for HP - Health Provider; II1 to II3 for II - Interpersonal Issues; SS1 to SS3 for SS - Shame Stigma; HA1 to HA3 for HA - Health Access; MD1 to MD3 for MD - Management Demands.
Confirmatory factor analysis
The fit indices for the confirmatory factor analysis model were χ2 = 1036.92, RMSEA = 0.07, TLI = 0.905, CFI = 0.918, and CMIN/DF = 2.97 (p < 0.001). The RMSEA value of 0.07 suggests a reasonable approximation error, while the TLI and CFI values above 0.9 indicate a good incremental fit of the model.
Structural equation modeling
The hypothesized model was visually represented (Fig. 1) through the SEM pathway, with latent variables depicted as ellipses, items as rectangles, and measurement errors as circles. The diagram illustrates the relationships between measured indicators (observed items) and latent variables (unobserved factors) through regression paths and covariances. The numerical values represent the standardized regression weights and measurement error terms.
Fig. 1.
SEM pathway for the T2-DDAS tool. Ellipses represent the latent (unobserved) variables; DD - Diabetes Distress, HY – Hypoglycemia, LH - Long-term Health, HP - Health Provider, II - Interpersonal Issues, SS - Shame Stigma, HA - Health Access, MD - Management Demands; Rectangles depict observed variables; DD1 to DD8 for DD - Diabetes Distress, HY1 to HY3 for HY – Hypoglycemia, LH1 to LH3 for LH - Long-term Health, HP1 to HP3 for HP - Health Provider, II1 to II3 for II - Interpersonal Issues, SS1 to SS3 for SS - Shame Stigma, HA1 to HA3 for HA - Health Access, MD1 to MD3 for MD - Management Demands; Small circle depicts error such as e1, e2, etc., connected to the observed variables, indicating the measurement error associated with each observed variable. Directional Arrows indicate hypothesized causal relationships between the variables. The direction of the arrow signifies the direction of the relationship. Bidirectional Arrows indicate covariances or correlations between variables. Numerical values along the paths (arrows) represent the strength and significance of the relationships between variables.
Interpretation of latent variables and indicators
DD was strongly influenced by its indicators (DD1-DD8), as evidenced by path coefficients ranging from 0.59 to 0.88. DD8 had a coefficient of 0.88, suggesting it was highly indicative of DD; the higher the coefficient, the stronger the relationship20. This indicated a robust relationship between the latent variable and its measures. Hypoglycemia (HY) was well-represented by HY1, HY2, and HY3, with coefficients of 0.84, 0.90, and 0.78 respectively. HY2 had the strongest association (coefficients of 0.90) with the latent variable. Interpersonal Issues (II) were significantly reflected in II1, II2, and II3, with II3 showing the strongest relationship (coefficients of 0.90). Health Access (HA) was moderately indicated by HA1, HA2, and HA3, with coefficients around 0.60. Management Demands (MD) were strongly represented by MD1, MD2, and MD3, particularly MD1 (coefficients of 0.82). Health Provider (HP) was moderately indicated by HP1, HP2, and HP3, with HP1 showing the strongest relationship (coefficients of 0.66). Shame Stigma (SS) was strongly represented by SS1, SS2, and SS3, with SS2 having the strongest association (coefficients of 0.85). Long-term Health (LH) was strongly indicated by LH1, LH2, and LH3, with LH2 showing the strongest relationships (coefficients of 0.85).
Interpretation of relationships among latent variables
The relationships among latent variables underscored the interconnectedness of factors affecting individuals with diabetes. Each coefficient represented the strength and direction of the effects observed. Diabetes Distress (DD) had a strong direct effect on Hypoglycemia (HY) (coefficient of 0.90), suggesting higher diabetes-related stress significantly increased the likelihood of hypoglycemic events. Hypoglycemia (HY), in turn, strongly affected Interpersonal Issues (II) (coefficient of 0.90), suggesting that experiencing hypoglycemia can greatly impact an individual’s relationships and interactions with others. Interpersonal Issues (II) strongly influenced the relationship with Health Providers (HP) (coefficient of 0.83), implying interpersonal difficulties significantly affect patient-provider interactions. The role of Health Providers (HP) was also strongly related to feelings of Shame Stigma (SS) (coefficient of 0.79) indicating that the way healthcare providers interact with patients can profoundly impact the patients’ feelings of shame and stigma. Shame Stigma (SS) had a strong direct effect on Long-term Health (LH) outcomes, with a coefficient of 0.85. This relationship underscores how feelings of shame and stigma can negatively impact the long-term health of individuals with diabetes. Finally, Long-term Health (LH) moderately influenced Diabetes Distress (DD) (coefficient of 0.55) suggesting poorer long-term health outcomes can contribute to increased diabetes-related stress, though the effect is not as strong as the other relationships observed.
Covariances between latent variables
Health Access (HA) and Management Demands (MD) showed a significant correlation (coefficient of 0.65). Health Provider (HP) and Long-term Health (LH) shared a strong correlation with a coefficient of 0.77.
The SEM pathway diagram effectively captured the intricate relationships between various aspects of the DDAS tool in the context of diabetes management. The path coefficients highlight the strength and direction of these relationships, providing valuable insights into the system’s dynamics.
Discussion
This study aimed to assess the psychometric validity of the English version of the T2-DDAS in an Indian T2D population to measure diabetes-related distress among patients. The results of the exploratory factor analysis and confirmatory factor analysis suggest that the English version of the T2-DDAS tool is a reliable and valid instrument for assessing various dimensions of diabetes distress in the Indian Population.
The T2-DDAS tool demonstrated good overall reliability, with the Diabetes Distress factor showing high internal consistency21,22 (Cronbach’s α = 0.91) and satisfactory factor loadings (0.70–0.83), indicating an effective measurement of diabetes distress. Other factors like Long-term Health, Health Provider, Interpersonal Issues, Shame Stigma, Health Access, and Management Demands showed acceptable to strong internal consistency (Cronbach’s α = 0.67–0.89) and factor loadings (0.56–0.82), indicating that they are robust and reliable measures of the respective constructs of DD21–23. The Hypoglycemia factor, while showing strong communalities (0.77–0.86), had lower factor loadings (0.61–0.63) and Cronbach’s α (0.64), suggesting room for improvement, though close to that reported for the original tool (> 0.65). Although values below 0.7 suggest scope for refinement, such values are not uncommon in psychological scales and may reflect the unique characteristics of the construct being measured in this population. Furthermore, some studies have considered Cronbach’s alpha acceptable up to 0.5, while stating that higher values strengthen the internal consistency of the measure23 Future research should aim to refine the Hypoglycemia factor and explore alternative reliability measures to enhance its robustness in the Indian context. The Principal Component Analysis (PCA) was employed as an effective initial method for examining the dimensionality of the T2-DDAS. However, future validation efforts could benefit from using Principal Axis Factoring (PAF), which isolates shared variance and is better suited for uncovering latent constructs. PAF could further distinguish between overlapping dimensions, such as emotional distress and hypoglycemia concerns, improving the clinical precision of the tool. Additionally, PAF could offer deeper insights into cultural and demographic variations in distress patterns, enhancing the applicability of the T2-DDAS across diverse populations.
Confirmatory factor analysis (CFA) showed satisfactory fit indices: χ² (1036.92), RMSEA (0.07), TLI (0.905), and CFI (0.918), supporting the structural validity of T2-DDAS and indicating that the items measure the intended latent constructs24. The RMSEA indicates reasonable approximation error, and TLI and CFI values above 0.9 suggest a good model fit. Structural equation modeling (SEM) further confirmed the tool’s validity, with a strong path (0.90) between Hypoglycemia and Interpersonal Issues, highlighting the impact of frequent hypoglycemia on social interactions and anxiety. Additionally, the moderate effect of Shame Stigma on Long-term Health (0.85) underscores the importance of addressing psychosocial factors, such as stigma, in improving long-term health outcomes. Future research should explore how these relationships evolve over time and develop interventions targeting stigma and patient-provider relationships to enhance diabetes care. Overall, the CFA and SEM analyses provided strong evidence of the reliability and validity of the English T2-DDAS in assessing diabetes distress in the Indian population.
The T2-DDAS shows promise as a reliable tool for measuring DD, although the Hypoglycemia factor requires further refinement for optimal reliability. Furthermore, future research should prioritize linguistic translation to increase the generalizability of this tool, particularly for non-English speakers. The absence of linguistic validation in this study may limit the findings to urban or educated populations with a higher English proficiency. The strength of this study lies in its large, geographically diverse sample from 81 Indian cities [51 urban areas (cities/metropolitan regions) and 37 rural or semi-urban areas (districts/smaller towns)], mitigating the selection bias typical of single-center studies. However, the online diabetes management program’s subscription model may limit generalizability by introducing a self-selection or non-response bias. Respondents with better Internet access, digital literacy, or education may be overrepresented, while rural and lower socioeconomic groups could be underrepresented, possibly underestimating diabetes distress in these populations. To address these limitations, future studies should use mixed methods, combining online and face-to-face surveys, and employ longitudinal approaches to assess the stability and predictive validity of the T2-DDAS over time. Further, while this study establishes the cross-sectional validity of the T2-DDAS, its utility could be significantly enhanced through longitudinal research. By capturing changes in diabetes distress over time, such studies might explain the interplay between distress, treatment adherence, and long-term health outcomes. Longitudinal data would also provide a deeper understanding of how specific interventions—psychological counseling, medication adjustments, or lifestyle modifications—impact distress trajectories. This approach could refine individualized care strategies and inform broader policy frameworks for diabetes management, ensuring that programs are adaptable to the dynamic nature of patient experiences.
In conclusion, the English T2-DDAS is a reliable tool for assessing diabetes distress in Indian patients with T2D. It provides clinicians with a validated method to identify individuals at risk of poor glycemic control, treatment nonadherence, and psychological issues, promoting better-coping mechanisms and adherence to treatment. By integrating the T2-DDAS into regular monitoring protocols and electronic health systems, clinicians can track distress longitudinally, enabling proactive and holistic diabetes management. Future research should prioritize linguistic adaptation and inclusion of broader populations to enhance the T2-DDAS’s generalizability, improve its internal consistency and factor structure, and support the development of patient-centered psychological interventions.
Acknowledgements
The authors would like to express their sincere gratitude to Dr. Maheshkumar Kuppusamy from the Department of Physiology, Government Yoga and Naturopathy Medical College and Hospital, Arumbakkam, Chennai, Tamil Nadu, India, for his invaluable expertise in reviewing the statistical analyses for this study.
Author contributions
Concept and design — PT, NK; acquisition of data— BSh, DT; analysis and interpretation of data — BSh, AV; drafting of the manuscript — BSh, TK, MG; revising the article critically for important intellectual content — NK, MG; final approval of the version to be submitted —PT, NK.
Data availability
All the data supporting our findings have been presented in the manuscript; the datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Anjana, R. M. et al. Prevalence of diabetes and prediabetes in 15 states of India: results from the ICMR–INDIAB population-based cross-sectional study. Lancet Diabetes Endocrinol.5, 585–596 (2017). [DOI] [PubMed] [Google Scholar]
- 2.Sun, H. et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res. Clin. Pract.183, (2022). [DOI] [PMC free article] [PubMed]
- 3.Barman, P., Das, M. & Verma, M. Epidemiology of type 2 diabetes mellitus and treatment utilization patterns among the elderly from the first wave of Longitudinal Aging study in India (2017-18) using a Heckman selection model. BMC Public. Health. 23, 1–12 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sugandh, F. et al. Advances in the management of diabetes mellitus: a focus on personalized medicine. Cureus. 15, (2023). [DOI] [PMC free article] [PubMed]
- 5.Kalra, S., Jena, B. N. & Yeravdekar, R. Emotional and psychological needs of people with diabetes. Indian J. Endocrinol. Metab.22, 696 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sinha, R., Priya, A. & Sinha, A. & Hifz Ur Rahman, M. Prevalence of diabetes distress among type 2 diabetes mellitus patients in India: a systematic review and meta-analysis. Health Psychol. Behav. Med.12, (2024). [DOI] [PMC free article] [PubMed]
- 7.Hiasat, D. A. et al. The prevalence of diabetes distress among patients with type 2 diabetes in Jordan. J. Taibah Univ. Med. Sci.18, 1237–1243 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Adugnew, M. et al. Original research: diabetes-related distress and its associated factors among people with type 2 diabetes in Southeast Ethiopia: a cross-sectional study. BMJ Open.14, 77693 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hoogendoorn, C. J. et al. Emotional distress predicts reduced type 2 diabetes treatment adherence in the Glycemia reduction approaches in diabetes: a comparative effectiveness study (GRADE). Diabetes Care. 47, 629–637 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Parviniannasab, A. M., Faramarzian, Z., Hosseini, S. A., Hamidizadeh, S. & Bijani, M. The effect of social support, diabetes management self-efficacy, and diabetes distress on resilience among patients with type 2 diabetes: a moderated mediation analysis. BMC Public. Health. 24, 1–10 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wojujutari, A. K., Idemudia, E. S. & Ugwu, L. E. Psychological resilience mediates the relationship between diabetes distress and depression among persons with diabetes in a multi-group analysis. Sci. Rep.14(1), 1–8 (2024). [DOI] [PMC free article] [PubMed]
- 12.Polonsky, W. H. et al. Assessing psychosocial distress in diabetes development of the diabetes distress scale. Diabetes Care. 28, 626–631 (2005). [DOI] [PubMed] [Google Scholar]
- 13.Polonsky, W. H. et al. Toward a more comprehensive understanding of the emotional side of type 2 diabetes: a re-envisioning of the assessment of diabetes distress. J. Diabetes Complications. 36, 108103 (2022). [DOI] [PubMed] [Google Scholar]
- 14.Niroomand, M. et al. Distress and depression among patients with diabetes mellitus: prevalence and associated factors: a cross-sectional study. J. Diabetes Metab. Disord. 20, 141–151 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chew, B. H., Vos, R. C., Pouwer, F. & Rutten, G. E. H. M. The associations between diabetes distress and self-efficacy, medication adherence, self-care activities and disease control depend on the way diabetes distress is measured: comparing the DDS-17, DDS-2 and the PAID-5. Diabetes Res. Clin. Pract.142, 74–84 (2018). [DOI] [PubMed] [Google Scholar]
- 16.Taylan, C. & Weber, L. T. Don’t let me be misunderstood: communication with patients from a different cultural background. Pediatr. Nephrol.38, 643–649 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gopalkrishnan, N. Cultural diversity and mental: considerations for policy and practice. Front. Public. Health. 6, 179 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fisher, L. et al. A new perspective on diabetes distress using the type 2 diabetes distress assessment system (T2-DDAS): prevalence and change over time. J. Diabetes Complications. 36, 108256 (2022). [DOI] [PubMed] [Google Scholar]
- 19.Tian, L., Wang, X. A. & Dynamic prediction neural network model of cross-border e-commerce sales for virtual community knowledge sharing. Comput. Intell. Neurosci. 2022, 2529372 (2022). [DOI] [PMC free article] [PubMed]
- 20.Pruttiakaravanich, A. & Songsiri, J. Convex formulation for regularized estimation of structural equation models. Sig. Process.166, 107237 (2020). [Google Scholar]
- 21.Jahrami, H., Saif, Z., Trabelsi, K., Bragazzi, N. L. & Vitiello, M. V. Internal consistency and structural validity of the nomophobia questionnaire (NMP-Q) and its translations: a systematic review with meta-analysis. Heliyon9, e15464 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Xiao, T. et al. Original research: psychometric validation of the perceived stress scale (PSS-10) among family caregivers of people with schizophrenia in China. BMJ Open.13, (2023). [DOI] [PMC free article] [PubMed]
- 23.Taber, K. S. The use of Cronbach’s alpha when developing and reporting research instruments in science education. Res. Sci. Educ.48, 1273–1296 (2018). [Google Scholar]
- 24.Stacciarini, T. S. G. & Pace, A. E. Confirmatory factor analysis of the appraisal of self-care agency scale - revised. Rev. Lat Am. Enfermagem25, (2017). [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All the data supporting our findings have been presented in the manuscript; the datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.