Abstract
Background
Diabetes mellitus (DM) significantly impairs quality of life (QoL) through its chronic course, treatment burden, and complications. Disease-specific QoL tools provide more clinically relevant assessment than generic instruments. Although the Diabetes Quality of Life–Brief Clinical Inventory (DQOL-BCI) has been widely validated internationally, no validated Tamil version exists despite large Tamil-speaking populations affected by type 2 diabetes mellitus (T2DM). This study aimed to translate, culturally adapt, and validate a Tamil version of the DQOL-BCI (DQOL-BCI-T) among Tamil-speaking adults with T2DM in Sri Lanka.
Methods
A cross-sectional validation study of DQOL-BCI-T was conducted among 165 native Tamil-speaking adults with T2DM attending an urban outpatient clinic in Colombo from May to October 2025). Translation followed a standardized forward–backward procedure with expert panel review and cognitive debriefing. Reliability was assessed using Cronbach’s alpha. Construct validity was evaluated through convergent validity with the Tamil WHOQOL-BREF and discriminant validity across clinical variables. Factor structure was examined using principal component analysis with oblimin rotation.
Results
The DQOL-BCI-T demonstrated acceptable internal consistency (Cronbach’s α = 0.682) and stability over two weeks. Significant moderate negative correlations with WHOQOL-BREF total (r = − 0.559, p < 0.001) and domain scores supported convergent validity. Discriminant validity was demonstrated by significantly lower QoL among patients with microvascular complications (p = 0.040). Factor analysis showed satisfactory sampling adequacy (KMO = 0.729) and acceptable item loadings, explaining 72% of total variance.
Conclusions
The DQOL-BCI-T demonstrates satisfactory validity, reliability, and cultural appropriateness for measuring diabetes-specific quality of life among Tamil-speaking adults with T2DM. Its concise format and strong psychometric performance make it a practical tool for routine clinical assessment and research, supporting patient-centred diabetes care. Further validation across diverse clinical populations, socio-cultural contexts, and treatment regimens is warranted to confirm its broader applicability and generalisability.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s12913-026-15638-0.
Keywords: Diabetes, Diabetes Quality-of-Life Brief Clinical Inventory, Tamil translation, Psychometric properties, Validity
Introduction
Quality of life and diabetes mellitus
The World Health Organisation (WHO) defines quality of life (QoL) as an individual’s perception of their position in life within the context of their culture, value systems, goals, expectations, standards, and concerns [1]. Historically, healthcare systems emphasized reducing morbidity and mortality, often overlooking QoL as an essential health outcome. However, in recent decades, patient-perceived outcomes, including functional, emotional, and social well-being have gained prominence, particularly in the management of chronic diseases such as diabetes mellitus (DM) [2]. Diabetes requires continuous self-management and imposes lifestyle and treatment burdens that significantly affect QoL. Consequently, assessing diabetes-related QoL is essential for understanding patient well-being and informing treatment decisions.
Global and regional burden of diabetes
Diabetes mellitus is one of the most common chronic metabolic disorders, contributing significantly to morbidity, mortality, and healthcare expenditure worldwide. Approximately 1 in 11 adults globally has diabetes, with type 2 diabetes mellitus (T2DM) accounting for nearly 90% of cases. By 2050, the global diabetic population is projected to rise to 852 million [3].
Southeast-Asia (SEA) has a prevalence of 9.7%, with 107 million persons affected with DM currently, and is among the regions expected to experience an upsurge in diabetes prevalence, with a projected increase of 73%3. Sri Lanka reports one of the highest prevalences in Asia and is one of the top five countries in SEA, with 1 in 5 adults having diabetes or prediabetes [3]. A nationally representative study in 2018/2019 found a crude adult prevalence of 23.0% and an age-standardized prevalence of 21.8% [4].
Impact of diabetes on quality of life
QoL among individuals with diabetes is influenced by four overarching domains: physical, psychological, social, and spiritual well-being. Both demographic factors, such as age, gender, socioeconomic status, marital status, and, disease-related factors - insulin use, hypoglycaemia, macrovascular and microvascular complications -are strongly associated with QoL impairments [5]. Complications such as neuropathy, retinopathy, nephropathy, cardiovascular disease, and erectile dysfunction too, markedly lower QoL and contribute to anxiety and depression. QoL also influences diabetes management. Complex treatment regimens, including diet modification, frequent glucose monitoring, medication adherence, and lifestyle changes, could reduce QoL and impair effective self-management. Conversely, greater illness understanding and empowerment of patients tend to improve QoL levels.
Quality of life measurement tools
Generic QoL instruments
There are several generic tools that have been developed and validated for assessing health-related QoL, including SF-36 [6] and SF-12 [7] that measure physical, emotional, and social functioning [8]. Sickness Impact Profile (SIP) [9] and EuroQol (EQ-5D) [10]. Although widely used, generic measures may not adequately capture disease-specific QoL concerns and are mostly used in the absence of a disease-specific measure.
Diabetes-specific QoL instruments
Disease-specific tools provide more relevant information for patient care and clinical trials [11]. The Diabetes Quality of Life Brief Clinical Inventory (DQOL-BCI), developed by Burroughs et al. [12], is a 15-item instrument derived from the original 46-item DQOL developed by the Diabetes Control and Complications Trial (DCCT) Research Group in 1988. It assesses the negative impact of diabetes, treatment burden, treatment satisfaction, and overall QoL. Each item is scored from 1 (lowest satisfaction) to 5 (highest satisfaction). Total scores range from 15 to 75. The advantages of this tool include brief administration time (~ 10 min), strong psychometric properties, its suitability for busy clinical settings. This tool is being extensively used across multiple countries, and validated translations exist in Iranian, Polish, Malaysian, Greek, Chinese, Indonesian, Urdu, Turkish, and Mongolian, with all demonstrating strong reliability and validity [13, 14].
Need for a Tamil version
Tamil, a major Dravidian language, is spoken by approximately 90 million individuals worldwide and is an official language in Sri Lanka, the State of Tamil Nadu in India, and Singapore [15]. Despite the significant numbers of native Tamil-speaking populations across Sri Lanka (18%), Southern India (5.8%), Malaysia (3.9%), Singapore (4.4%), Myanmar (2.1%), and Mauritius (0.8%) [16], no validated Tamil version of the DQOL-BCI exists. This study therefore aimed to develop a Tamil translation of the DQOL-BCI (DQoL-BCI-T) and evaluate its psychometric properties—including reliability, validity, and factor structure—among native Tamil-speaking adults with T2DM in Sri Lanka.
Methods
Ethics
Ethics approval was obtained from the Ethics Review Committee, Faculty of Medicine, University of Colombo, Sri Lanka (EC/25/038). Written informed consent was obtained from all participants. Data collection followed the principles of the Helsinki Declaration (1964 and subsequent amendments). Privacy and confidentiality were maintained throughout the study.
Study design and setting and sample size
A cross-sectional validation study was conducted at the Outpatient Department, Colombo Municipal Council Clinic, Slave Island, Colombo, from May 2025 to October 2025. Sample size determination was based on recommendations for psychometric validation studies, which suggest recruiting at least 5–10 participants per questionnaire item and a minimum sample exceeding 150 participants for principal component analysis. For the 15-item DQOL-BCI, a minimum sample of 150 participants was required. Allowing for potential incomplete responses, 165 participants were recruited.
Participants
A total of 165 native Tamil-speaking adults with T2DM participated in the study after providing informed written consent. They were included in the study if they were ≥ 18 years of age, diagnosed as having T2DM for ≥ 6 months, receiving at least one oral hypoglycaemic agent or insulin. Patients who were unable to self-report due to severe illness or had cognitive or psychiatric impairment were excluded from the study. All 165 participants completed the questionnaire fully.
Study instruments
Information on age, sex, marital status, education level, ethnicity, diabetes treatment type, and complications was collected using a demographic questionnaire. The Tamil translated version of the Diabetes Quality of Life–Brief Clinical Inventory (DQOL-BCI - T), the test version, and a validated Tamil version of WHOQOL-BREF [17].
Translation and cultural adaptation process
An email request was sent to the developer of DQOL-BCI, who provided the DLQI-BCI English version and permission via email. The translation of the DQOL-BCI followed the forward-backward procedure that was previously described [18, 19]. Briefly, the English questionnaire was translated independently to Tamil by two bilingual medical professionals well versed in both Tamil and English. Their first language was Tamil. The translations carried out by a person in his or her first language are more likely to accurately reflect the nuances of that language [20]. During the translation process, the translators were asked to pay special attention to ensuring conceptual and normative equivalence, rather than focusing merely on preserving the literal meaning of each word. Technical terms and long sentences were avoided to enhance the understanding of the target groups. Any discrepancies and ambiguities between the translated versions and the original English version were resolved by consensus between the two bilingual translators, and a finalised Tamil-translated version was agreed upon for backward translation. Two independent bilingual medical professionals, who were completely blind to the original English version, translated the Tamil version back into English. Henceforth, the two independent translators, together with the team of translators, compared the English translation with the original tool to determine any discrepancies or conceptual errors. A panel comprising an endocrinologist, social scientist, and linguist evaluated the face, content, and consensual validity and clarity. For the cognitive debriefing, 20 Tamil-speaking, diabetic patients were selected to check the face validity of the finalised questionnaire with respect to item clarity, relevance, and acceptability. Minor wording adjustments were made before arriving at the final version.
Data analysis
The convergent validity was assessed using the correlation between DQoL-BCI-T scores and WHOQOL-BREF Tamil scores. A correlation ≥ 0.30 in expected directions was considered acceptable. To evaluate the dimensional structure of the translated Tamil version of the DQOL-BCI, a Principal Component Analysis (PCA) was conducted. PCA was chosen to align strictly with the analytical methodology used in the development of the original instrument, allowing for a direct comparison of how the items clustered in this new linguistic context. The PCA was performed using an oblique (oblimin) rotation with Kaiser normalization to accommodate expected correlations between quality-of-life dimensions. To assess reliability and internal consistency, Cronbach’s alpha was computed for the total scale and subdomains. The Intraclass Correlation Coefficient (ICC) using the two-way random effects model assessed stability.
Results
Among the 165 native Tamil participants, the mean age of the study participants was 60.8 years, with a standard deviation (SD) of 9.0 (Table 1). Around 58% of the study participants (n = 96) were elders (60 years and above). Approximately half of the participants were female (50.3%). Multiple comorbidities were found among the majority (n = 133, 80.6%), and microvascular complications of DM were detected among 49.7% (n = 82) of patients. The mean and +/- SD of the disease duration were 7.5 and +/- 6.3 years, while nearly 60% of the patients had DM for more than five years.
Table 1.
Basic characteristics of the sample (N = 165)
| Number (%) | |
|---|---|
| Female | 83(50.3) |
| Age category | |
| < 50 years | 20(12.1) |
| 50–64 years | 87(52.7) |
| 65 years and above | 58(35.2) |
| Level of education | |
| Up to grade 5 | 11(6.7) |
| Grade 5 to GCE O/L | 45(27.3) |
| GCE O/L | 71(43.0) |
| GCE A/L | 35(21.2) |
| Higher degree | 3(1.8) |
| Presence of comorbidities | |
| Hypertension | 96(58.2) |
| Dyslipidemia | 114(69.1) |
| Ischemic heart disease | 14(8.5) |
| Treatment modality | |
| OHA | 157(95.2) |
| Insulin | 8(4.8) |
| Duration of the disease (mean+/-SD = 7.5+/- 6.3 years) | |
| Less than 5 years | 65(39.4) |
| 5 years and above | 100(60.6) |
| Presence of microvascular complications | |
| Retinopathy | 53(32.1) |
| Nephropathy | 8(4.8) |
| Neuropathy | 40(24.2) |
Cognitive validation
Target group and expert evaluation confirmed that DQoL-BCI-T is a valid tool to assess quality of life among patients with diabetes mellitus based on its content. The cognitive validation was confirmed based on face validity, content validity, and consensual validity.
Assessment of reliability
The DQoL-BCI-T was pilot tested on 34 respondents (Table 2). At baseline, the 15-item DQoL-BCI-T exhibited a Cronbach’s α of 0.727 with acceptable internal consistency. Then Cronbach’s alpha was retested in the DQoL-BCI- T version to assess the internal consistency of the tool for the sample in the validation study after two weeks and it exhibited a Cronbach’s value of 0.682, confirming acceptable internal consistency, as presented in Table 2. As this value was slightly lower than the conventional threshold of 0.70, an item level analysis was carried out. The removal of individual items did not meaningfully improve Cronbach’s alpha as shown in Table 2.
Table 2.
Reliability values for all items in the tool
| Items in DQoL-BCI-Tamil | Reliability Values at Pilot test | Scale When Item Deleted | For the validation sample (n = 165) |
Scale When Item Deleted |
|---|---|---|---|---|
| How satisfied are you with your current diabetes treatment? | 0.727 | 0.738 | 0.682 | 0.659 |
| How satisfied are you with the amount of time it takes to manage your diabetes? | 0.690 | 0.661 | ||
| How often do you find that you eat something you should not rather than tell someone that you have diabetes? | 0.727 | 0.673 | ||
| How often do you worry about whether you will miss work? | 0.691 | 0.654 | ||
| How satisfied are you with the time it takes to determine your sugar level? | 0.709 | 0.663 | ||
| How satisfied are you with the time you spend exercising? | 0.704 | 0.657 | ||
| How often do you have a bad night’s sleep because of diabetes? | 0.700 | 0.675 | ||
| How satisfied are you with your sex life? | 0.692 | 0.683 | ||
| How often do you feel diabetes limits your career? | 0.727 | 0.668 | ||
| How often do you have pain because of the treatment for your diabetes? | 0.729 | 0.674 | ||
| How satisfied are you with the burden your diabetes is placing on your family? | 0.715 | 0.679 | ||
| How often do you feel physically ill? | 0.689 | 0.648 | ||
| How often do you worry about whether you will pass out? | 0.689 | 0.671 | ||
| How satisfied are you with time spent getting checkups for your diabetes? | 0.746 | 0.674 | ||
| How satisfied are you with your knowledge about your diabetes? | 0.693 | 0.667 |
Test–retest reliability was assessed among 34 participants over a two-week interval using a two-way random-effects model with absolute agreement. The ICC for the total DQOL-BCI-T score was 0.843 (95% CI: 0.684–0.935), indicating good temporal stability. While some individual items showed lower agreement, the overall scale demonstrated good reproducibility over time, supporting the temporal stability of the instrument.
Construct validity
Construct validity was appraised by assessing convergent and discriminant validity.
Convergent validity
The correlation of the DQoL-BCI-T score with scores of the validated Tamil translation of the World Health Organization Quality of Life Brief (WHOQOL-BREF) questionnaire was assessed in the convergent validity assessment. A negative correlation is anticipated between these two constructs; As higher WHOQOL-BREF scores indicate better quality of life and higher DQOL-BCI-T scores indicate poorer diabetes-related quality of life, the observed negative correlations were expected and theoretically consistent, as shown in Table 3.
Table 3.
Convergent validity assessment (N = 165)
| Pearson r value (P value) | |
|---|---|
| WHO QoL Score (Total) | -0.559(< 0.001)* |
| WHO QoL Score (Physical health) | -0.544(< 0.001)* |
| WHO QoL Score (Psychological health) | -0.482(< 0.001)* |
| WHO QoL Score (Social relationship) | -0.386(< 0.001)* |
| WHO QoL Score (Environment) | -0.381(< 0.001)* |
*Statistically significant at 0.001 level
Discriminant validity
In order to establish discriminant validity, the DQoL-BCI-T scores were compared with the duration of the disease, treatment status, presence of comorbidities, and experience of microvascular complications by assuming those factors can discriminate the QoL of the patients as shown in Table 4. However, the only significant difference was observed with experience of the microvascular complications. With regard to microvascular complications, those with retinopathy had a slightly higher mean DQOL-BCI-T score than those without retinopathy (31.79 ± 6.55 vs. 30.56 ± 6.13; p = 0.241). Similarly, participants with neuropathy had higher mean scores than those without neuropathy (32.63 ± 6.08 vs. 30.42 ± 6.27; p = 0.053), while those with nephropathy also showed slightly higher scores than those without nephropathy (32.00 ± 7.43 vs. 30.90 ± 6.23; p = 0.631). None of these differences reached statistical significance. However, it is important to note that the subgroup sizes in this study are rather small for conducting meaningful analysis and should be considered exploratory.
Table 4.
Discriminant validity of the Tamil version of the DQoL-BCI according to clinical characteristics (N = 165)
| Variable | Category | Mean (SD) DQoL-BCI score | p value | Cronbach’s α |
|---|---|---|---|---|
| Duration of disease | Less than 5 years (n = 65) | 31.20(7.37) | 0.851 | 0.779 |
| Five years and above (n = 97) | 31.00(5.39) | 0.559 | ||
| Treatment option | OHA+ Insulin (n = 8) | 31.25 (5.63) | 0.893 | 0.520 |
| OHA (n = 157) | 30.94 (6.32) | 0.691 | ||
| Presence of comorbidities | Yes (n = 133) | 31.11(6.22) | 0.519 | 0.668 |
| No (n = 32) | 30.31(6.56) | 0.748 | ||
| Presence of any microvascular complications | Yes (n = 82) | 32.86(6.25) | 0.040* | 0.640 |
| No(n = 83) | 30.03(6.20) | 0.716 | ||
| Retinopathy | Yes (n = 53) | 31.79 (6.55) | 0.241 | 0.670 |
| No (n = 112) | 30.56 (6.13) | 0.691 | ||
| Neuropathy | Yes (n = 40) | 32.63 (6.08) | 0.053 | 0.619 |
| No (n = 125) | 30.42 (6.27) | 0.694 | ||
| Nephropathy | Yes (n = 8) | 32.00 (7.43) | 0.631 | 0.767 |
| No (n = 157) | 30.90 (6.23) | 0.680 |
*Statistically significant at 0.05 significance level
Principal component analysis
The Kaiser–Meyer–Olkin (KMO) sampling adequacy for the factor analysis was 0.729; hence, was rated meritorious. Also, with χ2 = 358.46 and p < 0.001, Bartlett’s Test of Sphericity revealed adequacy of the data and suitability of CFA. The extracted communalities and loadings factors for the DQoL-BCI-T are presented in Table 5. We used both pattern and structure matrices to avoid probabilities of value suppression because of the factorial relationships. Four factors explaining the total variance of 72% were extracted. The first factor reflected diabetes-related functional impact and physical concerns, including physical illness, treatment burden, hypoglycaemia-related worries, and limitations at work. The second factor represented treatment satisfaction and diabetes self-management, encompassing satisfaction with treatment, glucose monitoring, exercise, clinic visits, and diabetes knowledge. The third factor encompassed psychosocial well-being by capturing the emotional and social consequences of diabetes, including dietary restrictions, sleep disturbance, and satisfaction with sexual life. The fourth factor embodied family burden, reflecting the perceived impact of diabetes on family life. For the retention of items, we used Field’s proposal of communalities > 0.30. As shown in Table 5, loading values for all 15 items were acceptable (> 0.39). All items of the translated DQoL-BCI-T were retained, proving the validity of the translated questionnaire.
Table 5.
Survey items, rotated factor loading, and communalities (n = 165)
| Item | Pattern Matrix Component |
Structure Matrix Component |
Communalities Extraction |
||||||
|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 1 | 2 | 3 | 4 | ||
| 1 | − 0.043 | 0.555 | 0.325 | 0.129 | 0.059 | 0.571 | 0.348 | 0.163 | 0.449 |
| 2 | 0.054 | 0.649 | 0.032 | − 0.296 | 0.157 | 0.642 | 0.065 | − 0.264 | 0.506 |
| 3 | 0.273 | 0.420 | − 0.510 | − 0.022 | 0.280 | 0.432 | − 0.464 | − 0.016 | 0.494 |
| 4 | 0.641 | 0.077 | 0.256 | − 0.025 | 0.677 | 0.174 | 0.322 | − 0.043 | 0.531 |
| 5 | 0.052 | 0.635 | − 0.223 | 0.058 | 0.113 | 0.635 | − 0.188 | 0.087 | 0.456 |
| 6 | 0.104 | 0.556 | 0.126 | − 0.284 | 0.203 | 0.561 | 0.160 | − 0.257 | 0.426 |
| 7 | 0.239 | 0.047 | 0.590 | − 0.060 | 0.306 | 0.104 | 0.615 | − 0.061 | 0.445 |
| 8 | 0.105 | 0.080 | 0.566 | 0.412 | 0.155 | 0.142 | 0.585 | 0.418 | 0.531 |
| 9 | 0.700 | 0.004 | − 0.234 | 0.192 | 0.670 | 0.098 | − 0.162 | 0.163 | 0.539 |
| 10 | 0.712 | − 0.156 | − 0.093 | 0.116 | 0.677 | − 0.057 | − 0.029 | 0.078 | 0.503 |
| 11 | 0.087 | 0.101 | 0.111 | 0.809 | 0.080 | 0.160 | 0.132 | 0.812 | 0.695 |
| 12 | 0.674 | 0.108 | 0.216 | − 0.196 | 0.718 | 0.199 | 0.285 | − 0.214 | 0.609 |
| 13 | 0.594 | − 0.036 | 0.094 | − 0.063 | 0.601 | 0.045 | 0.150 | − 0.087 | 0.375 |
| 14 | − 0.080 | 0.584 | − 0.100 | 0.195 | − 0.019 | 0.578 | − 0.079 | 0.227 | 0.391 |
| 15 | − 0.141 | 0.637 | 0.058 | 0.152 | − 0.055 | 0.628 | 0.075 | 0.192 | 0.441 |
Notes: KMO = 0.729,Bartlett’s Test of Sphericity (p < 0.001). Extraction method: principal component analysis; rotation method: oblimin with Kaiser normalization
Discussion
This study successfully translated and validated the Tamil version of the Diabetes Quality of Life–Brief Clinical Inventory (DQOL-BCI-T), demonstrating it to be a reliable, valid, and culturally appropriate instrument for assessing diabetes-specific quality of life among Tamil-speaking adults with type 2 diabetes mellitus. The translation and cognitive validation processes adhered to international best practices, ensuring conceptual rather than literal equivalence, and were supported by expert panels and target population feedback, which together confirmed strong face and content validity.
Reliability
Test–retest reliability over a two-week interval demonstrated stability (pilot test sample and validation sample), suggesting that the instrument consistently measures the underlying constructs of diabetes-related quality of life in the absence of clinical change. Slight variations in reliability coefficients across time points may reflect the dynamic nature of quality of life, which is influenced by daily experiences, symptoms, and treatment demands, particularly in individuals with long-standing diabetes and multiple comorbidities. Additionally, brief tools tend to show lower Cronbach’s alpha due to fewer items per domain, and the values obtained here are within the expected range for such measures and similar to those reported in the broader QoL literature.
Internal consistency
The internal consistency of the DQOL-BCI-T demonstrated acceptable reliability for a brief, multidimensional clinical tool. Although slightly below the conventional threshold of 0.70, the Cronbach’s alpha value of 0.682 remains acceptable for a brief multidimensional instrument assessing diverse aspects of quality of life. Furthermore, item-level analyses showed that removal of individual items did not meaningfully improve Cronbach’s alpha, supporting retention of the complete instrument to preserve its conceptual integrity and comparability with the original version. The item analysis Furthermore Cronbach’s alpha values for the validation sample were comparable to those reported in other language versions of the DQOL-BCI, including Iranian, Polish, Chinese, and Turkish adaptations, where alpha values have ranged from approximately 0.65 to 0.85. Given the brief nature of the instrument and the heterogeneity of QoL domains assessed, the observed alpha values are considered satisfactory and consistent with the original instrument’s intended clinical utility [21]. These findings support the use of DQOL-BCI-T in clinical and research settings, particularly where brief tools are needed to minimize respondent burden.
Construct validity
Construct validity of the DQOL-BCI-T was supported through both convergent and discriminant validity analyses. The significant moderate to strong negative correlations between DQOL-BCI-T scores and WHOQOL-BREF domain scores were in the expected direction and magnitude. As higher WHOQOL-BREF scores indicate better quality of life, the observed inverse relationship confirms that higher diabetes-specific burden (higher DQOL-BCI-T scores) is associated with lower general quality of life. The strongest correlations were observed with the physical and psychological health domains, consistent with the substantial physical symptom burden and emotional distress associated with diabetes management.
Discriminant validity
Discriminant validity findings demonstrated that the DQOL-BCI-T distinguished between patients with and without microvascular complications, a known determinant of QoL impairment in diabetes (Table 4). Other clinical factors such as disease duration and treatment modality did not significantly discriminate QoL scores in this sample, this pattern aligns with prior research demonstrating the more pronounced QoL impact of complications than duration alone [22].
The absence of significant differences according to disease duration, treatment modality, and comorbidity status may reflect the relatively homogeneous clinical characteristics of the study population. In contrast, microvascular complications directly affect physical functioning and symptom burden and therefore may have demonstrated a stronger association with diabetes-specific QoL. However, the absence of statistical significance among individual microvascular complications, is likely attributable to the limited sample sizes in these subgroups. These findings should not be interpreted as definitive evidence of the absence of discriminant validity due to the consequent risk of occurrence of Type II errors.
Factor structure
The instrument’s structural validity was evidenced through factor analysis, which revealed a multidimensional structure with satisfactory sampling adequacy (KMO > 0.70) and all items loading acceptably on factors. Although the original DQoL-BCI was developed as a brief unidimensional measure without predefined domains, factor analysis of the Tamil version identified four conceptually meaningful factors. Collectively, these factors correspond to the major constructs measured by the original DQoL-BCI treatment: satisfaction, diabetes burden, physical and psychosocial functioning, and social impact while supporting interpretation of the Tamil version as a single overall measure rather than independent subscales. The reliance on Principal Component Analysis (PCA) rather than latent variable modeling techniques such as Exploratory Factor Analysis (EFA) or Confirmatory Factor Analysis (CFA) was identified as a limitation. While PCA was appropriate for replicating the original instrument’s development pipeline, it analyzes total variance and does not isolate unique measurement error. Consequently, while our results demonstrate structural comparability to the original tool, CFA would provide a more robust assessment of construct validity. Future multi-center studies should utilize CFA to cross-validate these findings in a larger Tamil-speaking population. While factor structures may vary across cultural adaptations, a finding documented in other validation studies of the DQOL-BCI, such variability reflects cultural interpretations of QoL dimensions rather than deficiencies in the measure itself. Similar variability was reported in the Malaysian and Turkish validation studies, where optimal factor structures differed from the original conceptual model but still supported overall construct integrity [23].
Clinical and research implications
The brevity and ease of administration of the DQOL-BCI-T make it particularly suitable for busy outpatient clinics and routine care, enabling healthcare providers to systematically incorporate patient-reported outcomes into diabetes management. At a public health level, the tool can support outcome evaluation in clinical trials, community interventions, and health system performance assessments. Thus, the availability of a validated Tamil version of the DQOL-BCI fills an important gap in patient-centred outcomes for a large global population. Tamil-speaking individuals comprise substantial proportions of populations in Sri Lanka, Southern India, Malaysia, Singapore, and beyond, yet most QoL instruments remain untranslated or unvalidated in Tamil, a limitation this study addresses. The availability of a validated Tamil version of the DQOL-BCI has important implications for both clinical practice and research. Tamil-speaking populations constitute a substantial proportion of individuals with diabetes in Sri Lanka and other regions, yet QoL assessment tools in their native language have been limited. Yet, it is well known that a language may exhibit minor linguistic variations across different cultural and geographical settings, particularly when it is spoken in multiple countries or sociocultural contexts [24]. Similarly, it is well established that Tamil exhibits linguistic variation across different geographical and sociocultural settings, with regional and social dialects differing in their spoken forms despite sharing a common linguistic foundation. Consequently, minor variations in wording, semantic interpretation, or conceptual expression may arise when translating or adapting a questionnaire for Tamil-speaking populations from different linguistic and sociocultural backgrounds. Thus, it may be useful to further refine and field optimise this tool to suit different sociocultural settings [25].
Limitations and future research
Several study limitations need to be acknowledged. The study was conducted at a single urban clinic, which may limit generalisability to rural populations or different healthcare settings. The majority of participants in this study were being treated with oral hypoglycaemic agents, while only a small proportion were receiving insulin therapy. Although this distribution reflects the characteristics of the study population, the psychometric properties observed may not be fully generalisable to patients receiving insulin, who may have different treatment-related experiences and QoL concerns. Future studies should include a larger proportion of insulin-treated individuals to further evaluate the performance of the this tool across different treatment modalities. Furthermore, the absence of objective clinical indicators such as HbA1c limits the evaluation of criterion validity. Future studies should examine the association between DQoL-BCI-Tamil scores and glycaemic control and other clinical outcomes to further establish its clinical validity. Another limitation of the present study is that responsiveness to clinical change was not evaluated because of its cross-sectional design. Although the Tamil version of the DQOL-BCI demonstrated satisfactory reliability and validity, its ability to detect clinically meaningful changes over time remains to be established, This requires longitudinal studies to evaluate sensitivity to treatment interventions before the instrument is routinely used to assess outcomes in intervention studies or to monitor changes in QoL following treatment. Although the Tamil version of the DQOL-BCI was developed through a rigorous translation and cultural adaptation process, the present study was conducted in a tertiary care setting, and the findings may not fully represent all Tamil-speaking populations. Differences in cultural practices, educational attainment, health literacy, socioeconomic conditions, and access to healthcare among rural populations may influence responses to QoL items. Therefore, further validation in rural Tamil-speaking populations with diverse socio-economic and cultural settings is recommended to confirm the instrument’s applicability and generalisability across all Tamil-speaking populations.
Conclusion
The Tamil version of the Diabetes Quality of Life–Brief Clinical Inventory (DQOL-BCI-T) is a valid, reliable, and culturally appropriate instrument for assessing diabetes-specific quality of life among Tamil-speaking adults with T2DM. The tool demonstrates acceptable internal consistency, strong construct validity, and stable psychometric performance, while retaining the brevity and clinical utility of the original instrument. The DQOL-BCI-T fills an important gap in patient-reported outcome measurement for Tamil-speaking populations and provides clinicians, researchers, and policymakers with a practical tool to enhance patient-centred diabetes care and outcomes assessment in Sri Lanka and other Tamil-speaking settings. As the structural evaluation was limited to Principal Component Analysis (PCA), subsequent confirmatory validation (CFA) in larger and more diverse cohorts is warranted. Testing the instrument across varied socio-cultural settings is essential to firmly establish its psychometric robustness, particularly when applied to specific clinical contexts such as hypoglycaemic events, macro- and microvascular complications, and specialized treatment profiles.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank Thomas Burroughs, Professor of Health Management & Policy of St.Louis University, Missouri, USA for granting permission to use DQOL-BCI and for providing us with the administration manual.
Author contributions
All authors contributed to the fulfillment of this study. ND obtained funding, and JP coordinated the conduct of the study; ND, and JP conceptualised the study and wrote the study design; ND, IT and AV collected the data and did the analysis; ND and JP interpreted the data analysis; ND and JP wrote the manuscript and all authors reviewed, edited, and approved the final manuscript and agreed on the journal.
Funding
This study was financially supported by a grant received from the Research & Development Centre, Link Natural Products (Pvt.) Ltd, Dompe, Sri Lanka. The funder had no role in the design, data collection, data analysis, and reporting of the study results.
Data availability
The data supporting the study findings are included within the manuscript. The raw datasets used during the current study are available from the corresponding author upon request.
Declarations
Ethical approval
The study was performed in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Review Committee, Faculty of Medicine, University of Colombo, Sri Lanka (EC/25/038). Written informed consent was obtained from all study participants included in the study.
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.World Health Organization. Programme on mental health: WHOQOL user manual, 2012 revision. World Health Organization; 1998. https://iris.who.int/handle/10665/77932.
- 2.Rubin RR, Peyrot M. Quality of life and diabetes. Diabetes Metab Res Rev. 1999;15(3):205–18. 10.1002/(sici)1520-7560(199905/06)15:3<205::aid-dmrr29>3.0.co;2-o. PMID: 10441043. [DOI] [PubMed]
- 3.International Diabetes Federation. IDF Diabetes Atlas. 11th ed. Brussels: IDF; 2025. [Google Scholar]
- 4.Rannan-Eliya RP, Wijemunige N, Perera P, Kapuge Y, Gunawardana N, Sigera C, Jayatissa R, Herath HMM, Gamage A, Weerawardena N, Sivagnanam I, Dalpatadu S, Samarage S, Samarakoon U, Samaranayake N, Pullenayegam C, Perera B, SLHAS Collaborators. Prevalence of diabetes and pre-diabetes in Sri Lanka: a new global hotspot-estimates from the Sri Lanka Health and Ageing Survey 2018/2019. BMJ Open Diabetes Res Care. 2023;11(1):e003160. 10.1136/bmjdrc-2022-003160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ferreira PL, Morais C, Pimenta R, Ribeiro I, Amorim I, Alves SM. Empowerment and Knowledge as Determinants for Quality of Life: A Contribution to a Better Type 2 Diabetes Self-Management. Int J Environ Res Public Health. 2023;20(5):4544. 10.3390/ijerph20054544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30(6):473–83. [PubMed] [Google Scholar]
- 7.Ware J Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220–33. 10.1097/00005650-199603000-00003. [DOI] [PubMed] [Google Scholar]
- 8.Development of the World Health Organization WHOQOL-BREF quality of life assessment. WHOQOL Group Psychol Med. 1998;28(3):551–8. 10.1017/s0033291798006667. [DOI] [PubMed] [Google Scholar]
- 9.Bergner M, Bobbitt RA, Carter WB, Gilson BS. The Sickness Impact Profile: development and final revision of a health status measure. Med Care. 1981;19(8):787–805. 10.1097/00005650-198108000-00001. [DOI] [PubMed] [Google Scholar]
- 10.Devlin NJ, Brooks R. EQ-5D and the EuroQol Group: Past, Present and Future. Appl Health Econ Health Policy. 2017;15(2):127–37. 10.1007/s40258-017-0310-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.El Achhab Y, Nejjari C, Chikri M, Lyoussi B. Disease-specific health-related quality of life instruments among adults diabetic: A systematic review. Diabetes Res Clin Pract. 2008;80(2):171–84. 10.1016/j.diabres.2007.12.020. [DOI] [PubMed] [Google Scholar]
- 12.Burroughs TE, Desikan R, Waterman BM, Gilin D, McGill J. Development and Validation of the Diabetes Quality of Life Brief Clinical Inventory. Diabetes Spectr. 2004;17(1):41–9. 10.2337/diaspect.17.1.41. [DOI] [Google Scholar]
- 13.Udzińska M, Tarach JS, Burroughs TE, Zwolak A, Matuszek B, Smoleń A, Nowakowski A. Validation of the Polish version of Diabetes Quality of Life - Brief Clinical Inventory (DQL-BCI) among patients with type 2 diabetes. Arch Med Sci. 2014;10(5):891–8. 10.5114/aoms.2014.46210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tang Z, Jiang X, Hong L, Feng Z, He Q, Yuan J, Xiang X. Validation of the Simplified Chinese Version of the Brief Diabetes Quality of Life (DQoL) Questionnaire Based on a Cross-Sectional Study. Int J Environ Res Public Health. 2020;17(23):8792. 10.3390/ijerph17238792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pakpour AH, Saffari M, Burri A. Translation and validation of an Iranian version of the Diabetes Quality of Life measure. J Diabetes Investig. 2012;3(5):471–8. 10.1111/j.2040-1124.2012.00217.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tamil - Worldwide Distribution – Worlddata.info. Available from: https://www.worlddata.info/languages/tamil.php#:~:text=With%2082.69%20million%20native%20speakers,Tamil%20as%20their%20mother%20tongue.&text=Unless%20otherwise%20described%20in%20the,the%20total%20number%20of%20speakers. Accessed 7 Dec 2025.
- 17.WHOQOL-BREF: introduction, administration, scoring and generic version of the assessment: field trial version, December 1996. World Health Organisation.
- 18.WHODAS 2.0 translation package (Version 1.0) 2021. Translation and linguistic evaluation protocol and supporting material. World Health Organisation.
- 19.Wild D, Grove A, Martin M, et al. ISPOR Task Force for Translation and Cultural Adaptation. Principles of good practice for the translation and cultural adaptation process for patient-reported outcomes (PRO) measures: report of the ISPOR Task Force for Translation and Cultural Adaptation. Value Health. 2005;8(2):94–104. 10.1111/j.1524-4733.2005.04054.x. [DOI] [PubMed] [Google Scholar]
- 20.Hendricson WD, Russell IJ, Prihoda TJ, Jacobson JM, Rogan A, Bishop GD. An approach to developing a valid Spanish language translation of a health-status questionnaire. Med Care. 1989;27(10):959–66. 10.1097/00005650-198910000-00006. [DOI] [PubMed] [Google Scholar]
- 21.Oluchi SE, Manaf RA, Ismail S, Kadir Shahar H, Mahmud A, Udeani TK. Health Related Quality of Life Measurements for Diabetes: A Systematic Review. Int J Environ Res Public Health. 2021;18(17):9245. 10.3390/ijerph18179245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Holmes-Truscott E, Broadley MM, Søholm U, Cooke DD, Hendrieckx C, Coates EJ, Heller SR, Speight J. Acceptability and psychometric properties of four scales assessing the impact of Type 2 diabetes on quality of life-Results of ‘YourSAY: Quality of Life’. Diabet Med. 2025;42(3):e15461. 10.1111/dme.15461. [DOI] [PubMed] [Google Scholar]
- 23.Lim AA, Ramasamy SM, Mat Nasir K, Han N, Burroughs YW. Linguistic and psychometric validation of the Malaysian version of Diabetes Quality of Life-Brief Clinical Inventory (DQoL-BCI). Res Social Adm Pharm. 2017;13(6):1135–41. 10.1016/j.sapharm.2016.10.017. [DOI] [PubMed] [Google Scholar]
- 24.Botha W, Bernaisch T. World Englishes and sociolinguistic variation. World Engl. 2025;44(1–2):2–11. 10.1111/weng.12695. [DOI]
- 25.Schiffman HF. Standardization or restandardization: the case for Standard Spoken Tamil. Lang Soc. 1998;27(3):359–85. 10.1017/S0047404500020017. [DOI]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the study findings are included within the manuscript. The raw datasets used during the current study are available from the corresponding author upon request.