Type 2 diabetes mellitus (T2DM) is a major global health issue, with 463 million adults affected globally, 80% of whom are living in low- and middle-income countries (LMICs) [1]. This health crisis is further exacerbated by conflicts, political instability, and natural disasters that disproportionately affect LMICs [2,3] such as Afghanistan and Pakistan. Afghanistan faces significant challenges, with an estimated 11.07% of adults having T2DM and 10.32% having impaired hyperglycemia [4]. Similarly, in Pakistan, the world’s fifth most populous country, 26.7% of adults (approximately 33 million people) have T2DM [5]. Additionally, many cases of T2DM in both countries are undiagnosed, worsening the health and economic burden of the condition.

Environmental changes, epidemiological transition, genetic predisposition, and behavioral factors such as diet and physical inactivity contribute to the rising incidence of T2DM in LMICs, in countries such as Afghanistan and Pakistan [6,7]. The disease impacts physical health, imposing significant emotional, social, and economic strains on individuals and communities [8]. T2DM resulted in 75.3 million (95% uncertainty interval 63.5‐90.2) global disability-adjusted life years in 2021 and was responsible for 2.6% (95% uncertainty interval 2.3%‐2.9%) of total global disability-adjusted life years [9].

Recent advances in health care have suggested addressing T2DM through specialized care, including a multidisciplinary approach to address diabetes-related complications [10]. However, this approach places a continuous and significant burden on individuals due to high out-of-pocket costs [11,12] for frequent medical visits, expensive medications, and diagnostic tests. Furthermore, the focus is usually placed on the management rather than the prevention of the disease, which further adds strain on already stretched health care systems, making diabetes management less accessible for many. In contrast, community-based participatory approaches are becoming increasingly popular because of their effectiveness in improving diabetes outcomes [13]. Moreover, community-based participatory approaches to health address the underlying causes of the disease and help in the prevention as well as the management of the health issue [14].

Community Health Participatory Research (CHPR) is an inclusive approach that actively involves community members in identifying areas for intervention, defining the problem, and generating potential solutions [15]. These approaches prioritize collaboration, equity, and the cocreation of knowledge to produce more relevant and effective interventions [16]. One CHPR approach is Participatory Learning and Action (PLA). PLA was inspired by Paulo Freire’s philosophy, where community members actively participate in learning about, defining, and addressing their health issues through facilitated sessions and activities [17]. PLA empowers marginalized communities to identify and solve their problems through a cycle of problem identification, solution planning, implementation, and reflection rather than relying on top-down strategies [18]. PLA has been systematized into a group-based intervention and has proven effective in improving neonatal and child health [19,20]. More recently, this approach was adapted to address diabetes in rural Bangladesh through the D-Magic (Diabetes Mellitus: Action Through Community Groups or mHealth Information for Better Control) trial [21]. The PLA intervention that was tested led to a 48% reduction in the prevalence of T2DM compared to the control group (305/3757, 8% vs 493/3821, 13%, respectively), with an adjusted odds ratio of 0.52 (95% CI 0.38‐0.71; P<.001). Additionally, the intervention increased awareness of T2DM prevention and motivated healthy dietary changes [21]. The D-Magic trial successfully used a participatory approach, organizing communities into small groups that held monthly meetings led by a lay facilitator. These groups followed a 4-phase PLA cycle to prevent and control T2DM, identifying barriers to healthy behavior and developing and implementing context-specific strategies, such as raising awareness, group exercise, income generation, and kitchen gardening [21].

The EMPOWER-D (Engagement of Community Through Participatory Learning and Action for Control and Prevention of Type 2 Diabetes and Its Risk Factors) trial is grounded in the conviction that PLA can empower communities to take ownership of diabetes prevention and control efforts. By organizing communities for collective action, PLA aims to increase awareness and motivation for healthy behaviors, thereby enabling individuals to make informed choices about their health. As a result, individuals are expected to gain confidence and motivation to seek health care services and adopt healthier habits. Ultimately, this approach is anticipated to lead to the development of a self-sustainable model for the prevention and control of diabetes, which can also have a positive impact on the prevention and control of other noncommunicable diseases (NCDs) in the region.

The theory of change logic model for our PLA intervention in EMPOWER-D (Multimedia Appendix 1) illustrates the hypothesized pathways through which participatory group activities are expected to lead to improved diabetes prevention and control in fragile LMIC contexts. It was developed by the research team, based on the findings from the D-Magic process evaluation and the context of Pakistan and Afghanistan. This model underpins our adaptation approach and will guide both implementation and evaluation.

Participatory interventions like PLA are highly context-sensitive, requiring adaptation to achieve sustainable outcomes when implemented in new settings or with different populations [22]. Despite its effectiveness and greater acceptability among the people, evidence on how PLA interventions like D-Magic can be adapted for fragile, crisis-affected, and socially diverse LMICs remains scarce. This is particularly relevant in countries with high T2DM burdens, such as Afghanistan and Pakistan, where direct replication could overlook important contextual distinctions [23,24]. These include variations in health care system capacity, cultural differences, gender-segregated communities, urban-rural disparities, and the added burden of conflict and political instability [25]. In Afghanistan, decades of conflict have overshadowed public health efforts, including addressing NCDs like T2DM, likely resulting in underreported prevalence [26]. In Pakistan, although the situation is relatively stable, the country has faced repercussions due to its proximity to unstable regions and a rapidly changing political environment, which adds to the issues of rapid urbanization and health care disparities, particularly between urban and rural areas, and for women in conservative rural regions [27]. These challenges contribute to rising T2DM rates and a fragmented health care system, with private urban centers offering better care than underfunded public clinics [28].

These differences shape how communities perceive health, engage with interventions, and mobilize collective actions. Thus, this study does not aim for simple transferability of D-Magic but for systematic adaptation that will generate new evidence on how participatory approaches can be tailored for fragile, gender-segregated, and highly diverse LMIC contexts.

This work aims to provide the following:

Ethical approval for this study has been granted by the ethics review committees of the Aga Khan University, Karachi, Pakistan (2024-9340-28927); Khyber Medical University, Peshawar, Pakistan (KMU/IPHSS/Ethics/2023/EO/0136); HealthNetTPO, Kabul, Afghanistan (A-12-24-464); and the National Bioethics Committee (NBC-R-1070). Researchers will approach identified participants and inform them about the research verbally and through a participant information sheet in their local language (Pashto, Dari, and Urdu). Participants will be encouraged to ask about the research before deciding whether they would like to participate. For those who consent to participate and provide a written consent, a time and date for an in-depth interview (IDI) or focus group discussion (FGD) will be arranged. Interviews will be conducted in a comfortable and confidential environment to ensure participants’ privacy. Participants will be assigned unique identification codes, and no personal identifiers will be included in transcripts or analysis files. Audio recordings will be stored in password-protected, encrypted folders and deleted after transcription. Deidentified transcripts will be used for analysis. All electronic data will be stored on secure, password-protected devices, and hard copies (eg, consent forms) will be kept in locked cabinets. Only authorized research team members will have access to the data, and findings will be reported in aggregate to protect participant identity. Participants will receive a monetary incentive to cover their travel expenses and compensate them for their time spent participating in the interview.

The full version of the EMPOWER-D trial is registered with ClinicalTrials.gov (NCT07350694 for rural Afghanistan, registered January 20, 2026, NCT06561126 for rural Pakistan, registered August 23, 2024, and NCT06570057 for urban Pakistan, registered August 26, 2024).

This study is part of the research program EMPOWER-D. The EMPOWER-D trial, a project of Centre for Improving Mental and Physical Health Together (C4IMPACT), was launched to specifically address the prevention and control of T2DM in rural and urban communities of Afghanistan and Pakistan.

We will utilize the ADAPT guidance [22] to adapt the intervention systematically. The ADAPT guidance framework emphasizes the importance of involving stakeholders and community members in the adaptation process to ensure that the adapted evidence-based intervention (EBI) is culturally appropriate and contextually relevant.

The framework comprises 5 steps to guide the adaptation process (Figure 1). This study will adapt the D-Magic intervention by carrying out three main activities: (1) identifying the barriers and facilitators in the implementation of community-based interventions (CBIs) and their contextual appropriateness through a qualitative study (using the socioecological model (SEM) [29]) with local stakeholders and a scoping review of the published literature, (2) co-designing the intervention materials, and (3) piloting the adapted materials for feasibility and effectiveness.

The adaptation of intervention materials across the 3 sites (rural Afghanistan, rural Pakistan, and urban Pakistan) will vary according to each context, but the process of adaptation will be the same. The process and findings will be documented using the FRAME (Framework for Reporting Adaptations and Modifications in Evidence-based Interventions) framework [30] (Figure 1), detailing the changes made, the reasons for these changes, and the implementation procedures. The outcomes and implications of the study will be recorded and reported, with recommendations for future research and practice.

Co-design methods will be employed to culturally, relevantly, and context-specifically adapt PLA interventions for the prevention and control of T2DM, using the findings from qualitative and scoping reviews [38]. Relevant stakeholders, including the research team, representatives of the local communities, health care providers, individuals with T2DM or those at risk, and graphic designers (for the preparation of materials), will be engaged in the co-design process to foster a collaborative approach to creating and refining strategies that are effective, sustainable, and acceptable to those who will implement and benefit from them. Co-design methods will ensure that the intervention is grounded in the sociocultural context and the lived experiences of the community, leading to more comprehensive and long-term solutions.

We will conduct a focused pilot of 1‐2 meetings at each site to obtain feedback on the co-designed PLA materials and assess initial feasibility and acceptability before moving to implementation. Given resource and time constraints, this will not be a full-scale pilot of the intervention but rather an opportunity to examine key aspects of how the adapted materials and group processes function in real settings.

In rural Pakistan, the adapted PLA materials will be tested in 2 rural clusters (one in Peshawar and one in Swabi), with two community groups (male and female) per cluster. In rural Afghanistan (Kabul) and urban Pakistan (Karachi), the pilot will be conducted in 1 cluster each, with 2 community groups (male and female). Groups will be facilitated by trained male and female community mobilizers under the supervision of senior staff.

Two pilot meetings will be conducted at each site, and the evaluation will focus on 2 core domains:

Data sources will therefore include the following:

Success criteria will be pragmatic: for example, reasonable attendance (≥50% of invited participants attending at least one of two meetings), visible engagement in discussions, and no major concerns about the cultural acceptability of the materials.

Findings will be summarized in a short pilot report structured around these domains. Any revisions to the PLA materials will be informed by participants, mobilizers, and supervisors’ feedback. All modifications will be documented using the FRAME framework, including the rationale for the change [30].

This study has an active program of community engagement for which we have constituted CAPs at each of the project sites. The CAP members represent both men and women from the community who have knowledge and understanding of the local contexts and cultures. The members of the CAPs include individuals living with diabetes, individuals without diabetes, caregivers of diabetic individuals, youth representatives, social workers, community health workers, persons from the seminary or mosque, school, and local women’s representation. CAP members have been involved in designing the topic guides and their translation into the local languages. They have also been involved in identifying participants for the study and will help in disseminating the study findings. Their involvement will improve research quality and help foster a sense of ownership of outputs by the community.

This study hypothesizes that adapting the D-Magic intervention through systematic use of the ADAPT guidance and the FRAME framework will result in contextually relevant, acceptable, and feasible community-based participatory interventions tailored for the prevention and control of T2DM in Afghanistan and Pakistan. By incorporating sociocultural preferences, local resources, and community participation, the adapted intervention is expected to enhance both its effectiveness and sustainability in these fragmented settings.

We anticipate that the adaptation process will generate interventions that are culturally appropriate, acceptable to participants, and feasible to implement across 3 diverse sites, building a pathway for future implications in other LMICs. The process is designed such that it maintains the internal logic and core of the original PLA intervention, while allowing for necessary contextual modifications. This balance between fidelity and flexibility is expected to strengthen community ownership and improve the potential for sustained impact on diabetes prevention and control.

Our approach builds upon evidence from the D-Magic trial in Bangladesh, which demonstrated the effectiveness and cost-effectiveness of PLA in reducing diabetes risk factors in a collectivist, low-resource setting. Afghanistan and Pakistan share some social structures and health system challenges, making PLA a potentially transferable and scalable strategy. At the same time, this study contributes to the growing literature on how systematic adaptation frameworks can be applied to EBIs to ensure relevance across diversified populations.

A major strength of this study lies in its systematic and participatory approach to adaptation, which focuses on community engagement, ensures transparency in documenting changes, and provides insights into maintaining fidelity. However, we understand that given fragile health systems and potential political instability or security issues, feasibility and fidelity may be affected in ways that are difficult to fully anticipate. As this is a protocol of an adaptation study, the conclusions are based on anticipated rather than observed outcomes.

The adapted PLA toolkit will be disseminated within the project team for its implementation. Key adaptation findings will also be shared with the CAP members. Our qualitative research findings and the scoping review will be disseminated through high-impact, open-access peer-reviewed journals and at local and international conferences. Reports will also be prepared for policymakers and stakeholders in the countries involved.

The findings of this study will emphasize the importance of systematic adaptation processes for shaping intervention delivery, effectiveness, and sustainability in implementation research, particularly in resource-constrained and crisis-affected settings. This study will also provide a replicable framework for other EBIs targeting NCDs and related risk factors in LMICs.

The adaptation of D-Magic for EMPOWER-D is expected to synthesize a culturally grounded intervention that both adheres to its psychosocial origin and responds to the contextual realities of Afghanistan and Pakistan. By applying structured frameworks to guide and document this process, the results are expected to provide insights into how fidelity and contextual relevance can be balanced in complex and diverse, resource-limited settings. The hypothesized findings will inform the design of a definitive trial and contribute to the wider evidence base on the adaptation and implementation of participatory CBIs for NCD prevention and control in LMICs.