Detailed study methodology has been previously published and is presented here in brief (19). The study adopts an exploratory qualitative design underpinned by Bronfenbrenner’s ecological system theory (30) and Whittemore et al.’s (31) conceptual framework based on the social-ecological model (19, 32–34). Data collection methods included focus group discussions (FGDs) to enable the exploration of various aspects of facilitators and barriers to T2DM self-management practices. There was a special focus on social context through gathering perspectives on viewpoints, thoughts, expectations, and suggestions (35, 36) from people with T2DM and their caregivers in rural Bangladesh. Furthermore, FGDs investigated preferences, barriers, and facilitators for community-based DSMP intervention programs to meet the unique needs and circumstances of the rural population.

The study has been conducted in Bangladesh, a lower-middle-income country in the South Asian region. Bangladesh is geographically divided into eight administrative regions, known as divisions (the first layer), which are further subdivided into 64 districts (the second layer). Each district has several upazilas (the third layer), and each upazila comprises several union parishads (the lowest layer). Each union Paris had is further divided into approximately ~15 to 20 villages (37). This research was implemented in four districts: Jhenaidah et al. encompassing four Upazilas and eight villages (Figure 1). The selected villages were geographically dispersed to ensure a diverse representation of the population. The sampling process has been detailed previously (19).

Using purposive sampling (38, 39), two community champions, who were local social workers and well-connected with the community, facilitated engagement with potential participants in common settings (i.e., mosques, community centers, and local village markets or tea stalls). They initiated this during religious or community gatherings leveraging their trustable community relationships to explain the study’s purpose and encourage participation.

Participants were recruited in dyads, which included an immediate caregiver to enable the exploration of relationships and roles in T2DM management. Eligibility for participants with T2DM included adults who were (i) aged ≥18 years, (ii) diagnosed with T2DM at least 1 year before recruitment, and (iii) under the active care of a registered physician. Eligibility for caregivers of those with T2DM included (i) a spouse or immediate family member residing with the participant with T2DM, (ii) a provider of support and care to the participant with T2DM at home for at least 1 year prior to data collection (32, 40, 41). Exclusion criteria for all participants included those diagnosed with cognitive or mental impairment or other significant illnesses preventing participation.

Focus group discussion (FGD) interview guide (attached in Appendix 1) included key topic areas of knowledge, motivation, and routines toward DSMP, financial and time constraints, emotional health and well-being, self-monitoring, healthy eating, physical activity, and overall facilitators, barriers, and expectations for continuity of diabetes care. The interview guide also included the key topic areas of expectations for developing a community-based practice approach to implementing DSMP (e.g., the participants’ preferences for the place, mode of intervention delivery, intervention format, duration of intervention, preferred provider, and financial issues). Interview guides were piloted (sample size n = 4, data not presented) to ensure cultural sensitivity and acceptability; they were iteratively revised as required. Widely framed and open-ended questions provided ample opportunities for the study participants to share their personal experiences that facilitate and impede their habits and expectations toward DSMP. Participants were also encouraged to shape their own views and thoughts and contribute any additional information relevant to the topic.

The focus group discussions were conducted between July and October 2023 and consisted of eight to 10 participants to facilitate in-depth and manageable interactions. Three interviewers (HAC, ZI, RR) conducted FGDs in Bangla, the native language of both the interviewers and participants. Each FGD was conducted at a mutually convenient location and time, averaging 45–60 min. Discussions were conducted until data saturation was reached, which was indicated when no further information, dimensions, or ideas emerged from the interviews, as outlined by Guest et al. (42, 43). Interviews were audio-recorded and transcribed verbatim by an independent transcribing service (at the Center for Qualitative Research, Dhaka, Bangladesh) and, then translated into English by the first author (HAC) and reviewed by another author (AK). Participant details were deidentified for anonymity.

This study was guided the Theoretical Domains Framework (TDF), which builds on the Bronfenbrenner’s social-ecological system theory (30) and Whittemore et al.’s conceptual framework (31), rooted in the social-ecological model (SEM), which is widely applied in health services research, provide a robust basis for understanding and addressing diabetes prevention and management (32). This model suggests that long-term behavior change requires targeting multiple levels of influence for a program to succeed. Influence levels are categorized into intrapersonal factors, interpersonal factors, institutional factors, community factors, and public policy levels (30, 31, 33, 34, 44). Intrapersonal factors assess individuals’ ability to change their behaviors to optimize health in diabetes self-management practices, which are influenced by knowledge, attitudes, self-efficacy, and self-confidence (45, 46). A patient’s interpersonal relationships within a societal context, such as family, caregivers, friends, neighbors, colleagues, and healthcare service providers, positively influence an individual’s DSMP (33, 47–49). Social support consists of interpersonal transactions and includes companionship and emotional, tangible, and informational care (37). Institutional influences on individuals include work, school or college, and religious surroundings. Institutional influences provided a structural framework to promote healthy behaviors and prevention activities for diabetes-related comorbidities to a large group, thus forming social supports that further facilitate healthy lifestyle adoption (33). Their prescribed geographical location determines an individual’s community factor. Evidence indicates that the characteristics of the neighborhood and communities affect individuals’ health behavior (31, 46, 50). Further, public policy influences individuals’ health behaviors through the execution of laws and policies by local and national authorities. This model is considered an appropriate framework to explain the dynamic interplay of multiple factors that acknowledge influencing causes as either facilitators, barriers, or expectations to T2DM self-management practices in rural Bangladesh (19).

Transcripts were independently coded by two researchers (HAC and SAD) using the NVivo 12 software (Nvivo [Version 12], QSR International Pty Ltd., Melbourne, Australia [2018]). Data were searched for concepts in relation to research questions using thematic deductive analysis according to the ‘15-point checklist of criteria for efficient thematic analysis’ by Braun and Clark (43). The process involved: (a) establishing a set of a priori codes derived from literature and the collective expertise of the research team, (b) line-by-line coding based on the predefined set of codes, (c) initiating coding using both a priori and posteriori codes, incorporating insights from the transcripts, (d) reviewing and organizing all coded data to create a thematic matrix, (e) integrating a priori and posteriori codes into final themes and sub-themes, and (f) presenting the findings under these identified themes. Themes and subthemes were derived by organizing and clustering related codes into coherent groups, followed by iterative discussions and refinement based on emerging patterns and insights from the FGDs and participant feedback. Any disagreements were discussed until a consensus was reached (51). The research findings were presented in accordance with the ‘Consolidated Criteria for Reporting Qualitative Studies: 32-Item Checklist’ (36). All quantitative data were presented as mean (± standard deviation [SD]) unless otherwise stated.

Four subthemes were identified that underpin the theme of DSMP implementation including motivation, implementation barriers, cultural and religious values, and financial constraints.

Participants with T2DM and their caregivers stated that motivation toward a healthy lifestyle, including regular exercise, selecting healthier dietary options, regular glucose monitoring, and medication compliance, was critical for sustained DSMP adherence.

Participants with T2DM also cited their family as a significant source of extrinsic motivation to ensure optimum health and avoid financial burdens on family members.

While participants often demonstrated knowledge of the general aspects of a healthy diet or healthier choices and motivation facilitated ongoing behavior change, cultural traditions related to meal preparation frequently presented as a primary barrier to implementing changes. Participants often perceived traditional Bangladeshi diets as difficult to modify, particularly substituting or omitting foods high in complex carbohydrates (e.g., rice) that influence blood glucose levels. This, in part, may be owing to financial constraints related to purchasing alternatives at a higher price point.

Further, many participants reported confusion in changing their diet to improve glycemic control, leading to increased reliance on medication to manage blood glucose levels.

In addition to dietary challenges, participants faced physical health limitations that affected their ability to engage in essential DSMP activities. Poor eyesight made it difficult to read medication labels and monitor blood sugar levels accurately, while foot soreness restricted mobility and reduced opportunities for physical exercise. Time constraints and competing responsibilities were also noted as significant barriers, particularly for female participants. Many women expressed being overburdened with household tasks and caregiving responsibilities, leaving them too fatigued to incorporate additional activities, such as walking or exercising, into their daily routines. These challenges highlight the interplay of individual, physical, and socio-cultural factors in shaping DSMP behaviors.

Several participants perceived an interconnection between cultural beliefs and DSMP, noting that beliefs in alternative medicines influenced their initiation and adherence to prescribed medications.

Further, most participants reported that their religious beliefs were deeply entrenched and personal, positively influencing their DSMP, with prayer and faith providing a sense of control over their health and improving glycemic control. This also encouraged regular physical activity by attending mosques and promoting discipline and social interaction in rural communities.

Financial constraints presented a barrier to DSMP, contributing to reduced access to resources and the inability to meet healthcare costs related to medical consultations, diagnostic tests, buying blood glucose monitoring devices, medications, and ongoing consumables. As a result, some participants reported refraining from using health services.

The theme of community spirit and interconnectedness is underpinned by two subthemes including social support from close-knit relationships (e.g., spouse, children, and in-laws) and broader social networks (e.g., neighbors, friends, and community members).

The presence of social support from family members (e.g., spouse, son/daughter, and daughters-in-law/sons-in-law), appeared to have an inverse relationship with the ability to engage in and maintain, DSMP effectively. This was particularly pronounced in the case of female participants, who were more likely to report receiving less support, challenging their ability to engage in healthy self-management practices. In contrast, participants who reported receiving ongoing support were likelier to have increased motivation and ability to implement DSMP. In addition, caregivers enhance well-being by facilitating self-management. Their support fosters an environment that facilitates effective DSMP by addressing medical, emotional, and logistical needs, ultimately leading to better health outcomes and an improved quality of life for the people with T2DM.

The supportive roles of neighbors, peers, or community members in rural settings were essential for providing emotional support, advice, and lived experience of diabetes management strategies and practical support, including help buying medications.

This theme includes two sub-themes: availability and accessibility of community infrastructure and resources, and community influences/social stigma.

Most participants reported that barriers to DSMP included a lack of healthy food options in markets, lack of exercise facilities, busy roads or narrow village paths, and related to safety issues, such as crime or traffic hazards that hinder walking in the local community setting.

Participants reported that social stigma negatively impacts people with T2DM. Most of the participants shared that the community perceives T2DM as a curse, often blaming those affected and fostering in them feelings of personal failure or guilt regarding their health conditions. Moreover, many female participants described receiving verbal abuse due to societal norms and perceptions during physical exercise.

This theme includes two sub-themes: doctor-patient relationships and lack of accessibility/resourcing in local hospital care. Participants cited their doctor-patient relationship as significantly influencing DSMP. They perceived doctors and nurses as generally uncooperative, prioritizing quick transactions over sincere support due to high patient volumes. These accounts may underscore systemic issues, including provider and medicine shortages, leading to patient dissatisfaction and reluctance to seek further medical treatment.

Some participants noted that doctors are crucial sources of information on critical aspects of DSMP. Participants expressed that information on healthy eating, physical exercise, and foot care was also beneficial; this information was utilized by people with T2DM in their self-management and shared with family members, fostering a supportive home environment for DSMP.

Further, participants in rural areas faced linguistic barriers as doctors primarily communicated in English or used complex medical terms, leading to difficulty in understanding instructions. Consequently, many people with T2DM felt the healthcare system did not meet their personal needs in managing diabetes.

Participants perceived community clinic locations and services as convenient for diabetes care. Yet, the majority noted anecdotally a significant decline in the availability of diabetes medication at these clinics over recent years, with medications largely unavailable, affecting their ongoing treatment. Further, participants faced critical gaps in diabetes care, affecting both testing accuracy and treatment consistency; long distances, transport costs, and considerable waiting times posed a significant risk to their health.

Participants expressed concerns about the lack of government support for people with T2DM in rural areas, noting that despite various allowances offered for other needs, there is a significant gap in providing essential medical equipment. This shortfall forces many patients to incur additional financial burdens or forgo necessary treatments.

Participants were presented with preferences for DSMP intervention design, including setting, delivery mode, program format, content, and facilitator type, as well as barriers to and enablers of engagement. Overall, participants preferred programs delivered at community clinics as they considered them reliable, culturally appropriate central ‘hubs’ to assemble. Alternatively, home-based services were chosen based on safety and convenience concerns, particularly by women or those with mobility issues. Group-based attendance was favored for its potential to enhance learning through shared experiences, with a strong preference for resources with illustrations (rather than only written material) and involving family members for additional support in implementing the advice received. Participants emphasized their preference for programs delivered by those with a medical background (e.g., specialized doctors). However, knowledgeable professionals were also considered trustworthy. Financial constraints were a significant concern, with many participants advocating for free or subsidized programs to alleviate the economic burden of diabetes management. These details are included in Table 3.

The study design was theoretically underpinned by the Theoretical Domains Framework to explore T2DM management and targeted multiple levels of influence (30, 31). To the best of our knowledge, this is the first in-depth formative research, conducted within rural community settings where T2DM is highly prevalent and health services are under significant strain, underscoring the urgent need for model implementation of DSMP. Using a pre-tested interview guide, we provide new insights into the critical factors for self-management of T2DM and community-level preferences for program development. To emphasize the relevance of involving both people with T2DM and their caregivers to address the holistic needs of T2DM self-management practices. We recruited dyads of patients and their caregivers. This approach has not been extensively utilized in previous research. This study is subject to possible limitations. First, the T2DM subjects were from middle-aged to older age groups, and therefore, their responses may not be generalizable to other age groups (such as below 18 years age) with differing priorities and perspectives. Second, we recruited the study participants from rural areas; thus, the results were limited by geographic location. Additionally, the study included people with T2DM who were under the care of a registered physician. This may cause selection bias, as the sample may not represent the broader T2DM population, especially those with limited healthcare access. Despite these limitations, rigorous study methodology enables this to be conceptually transferable. Third, due to the linguistic differences between native Bangla and English, nuances, idiomatic expressions, and cultural contexts in Bangla often lack direct equivalents in English, leading to potential misinterpretations and challenges in accurately conveying participants’ experiences. To address this challenge and ensure an accurate representation of participants’ experiences, we included cultural experts and bilingual translators, in addition to using back-translation, and participant feedback. Fourth, as the study employed a qualitative approach to explore in-depth experiences and perspectives, quantitative data collection was beyond its scope. However, the findings provide a foundational understanding that could inform future mixed-methods research to quantify the impact of the identified barriers and facilitators. Future studies are needed to provide further insight, particularly in urban settings, where providing basic diabetes care (and primary care health care in general) attracts barriers and enablers.