This study used a qualitative research design, involving in-depth interviews with key health system stakeholders to explore barriers to the sustainability of the Sehatmandi app. The qualitative design allows a rich, nuanced understanding of participant experiences and perspectives.

The Sehatmandi app was implemented across 189 health facilities, encompassing 115 health facilities in 28 districts of the Badakhshan province and 74 health facilities in 8 districts of the Bamyan province, Afghanistan. For the qualitative study, we selected health facilities across both provinces, ensuring geographic diversity and representation of various facility types. This selection ensured the inclusion of facilities representing varied operational contexts and challenges.

The study population included health facility managers, administrators, and high-level decision-makers.

Semistructured interviews were conducted between June and July 2024 either in a private room at a health facility or virtually via Zoom in cases where physical access was constrained. The interviews were conducted by 2 local qualitative researchers, one from Badakhshan and one from Bamyan, who were familiar with Afghanistan’s sociocultural and health system contexts, enabling strong rapport-building and culturally sensitive engagement. A pretested interview guide comprising open-ended questions was used to elicit in-depth information on participants' experiences with the Sehatmandi app. Interviews were conducted in either Dari or English according to the participant's preference and were audio-recorded with their consent. Transcripts were produced in Dari and subsequently translated into English by bilingual researchers. All translated transcripts were independently verified for accuracy by a third reviewer fluent in both languages. Data collection was concluded upon reaching theoretical saturation, with no new themes emerging, resulting in a total of 24 complete interviews.

Thematic analysis was conducted using a deductive-inductive approach. Data were coded and analyzed using NVivo (version 11; QSR International) by a trained researcher at Aga Khan University (FJ). To enhance credibility and interpretive rigor, the research team engaged in reflexive practices such as iterative review of emerging codes, and findings were validated through peer debriefings and team discussions. A realist epistemological stance underpinned the analysis, assuming that participants’ accounts reflect real-world experiences that can inform and sustain digital health interventions in resource-constrained and conflict-affected settings.

Ethics approval was obtained from the ethical review committee of Aga Khan University, Pakistan (2023-8159-24046) and the institutional review board of the Ministry of Public Health, Afghanistan (A-8-24-444). Informed consent was obtained from all participants prior to the interviews. Participants were informed about the study purpose, voluntary nature of participation, confidentiality, and their right to withdraw at any time without consequence. To ensure privacy and confidentiality, during transcription, all identifiable information was removed from the transcripts. Data were securely stored in password-protected systems accessible only to the research team. No financial compensation was provided.

A total of 24 participants were interviewed across Badakhshan and Bamyan provinces to evaluate the sustainability of the Sehatmandi digital health intervention in Afghanistan. In Badakhshan, 14 participants were interviewed, comprising 10 health managers, 1 HMIS officer, 1 project director, 1 technical officer, and 1 clinic director. In Bamyan, 10 participants were interviewed, comprising 8 directors of health facilities, 1 HMIS officer, and 1 monitoring and evaluation officer.

Thematic analysis of the qualitative interviews revealed 7 major themes: (1) enhanced performance monitoring and accountability, (2) timely reporting, (3) infrastructure and operational challenges, (4) workforce capacity and training needs, (5) financial and political constraints, (6) monitoring and training gaps, and (7) technical and system limitations.

Participants in this study highlighted key factors that supported the implementation and uptake of the Sehatmandi digital health app. Their perspectives reflect real-world experiences from the field, illustrating how Sehatmandi contributed to improved health system performance. The following themes emerged from the interviews, capturing the ways in which the app enhanced performance monitoring and accountability of health care staff.

One manager explained that the system’s offline functionality enabled data entry in areas with limited internet access.

The platform also streamlined the reporting process to higher administrative levels.

Several participants emphasized how the app supported routine tracking of service delivery metrics, enabling them to monitor progress against key performance indicators. One facility manager reflected as follows.

The app further facilitated structured review processes during HMIS committee meetings, which helped institutionalize accountability mechanisms.

At the provincial level, a health information officer highlighted the value of Sehatmandi for centralized oversight.

The system faced substantial obstacles, including infrastructure limitations, financial constraints, and human resource challenges, which impeded its overall efficiency and ability to scale. Despite these difficulties, valuable lessons were learned, providing critical guidance for future digital health initiatives in comparable resource-limited and conflict-affected environments.

The Sehatmandi digital health intervention faced significant infrastructure and operational hurdles. Inconsistent electricity and poor internet connectivity severely disrupted data synchronization and hindered real-time data monitoring. A few users reported that improper handling led to device breakage. Delays in providing replacements further disrupted consistent data collection. Crucial operational requirements such as timely SIM card distribution were frequently unmet.

To enable real-time data monitoring, reliable internet and stable electricity are essential. As a health facility manager emphasized, “All health centers should have access to the internet.” Additionally, durable devices, backup power solutions, and offline data entry options can mitigate disruptions in remote areas. In response, offline data entry was introduced in Sehatmandi, enabling health care administrators to continue data entry without internet connectivity and synchronize the data when connectivity was restored. Moreover, a user highlighted that dedicated local support structure such as an on-site focal person can provide real-time troubleshooting, enhance user confidence, and improve adoption rates.

High staff turnover and limited digital literacy among health workers initially reduced the usability of the Sehatmandi system. Many health care workers were unfamiliar with digital tools, leading to hesitancy in adoption.

Continuous training and capacity building should be performed to enhance user proficiency and adoption. These training sessions should focus on practical skills, troubleshooting, and building user confidence through structured programs and regular refreshers, ultimately increasing staff proficiency and reducing resistance to the technology, leading to sustainable and improved data collection practices.

Taliban's takeover of Afghanistan in 2021 created significant financial challenges. Several international funding agencies, citing security concerns and uncertainties in governance, suspended funding, abruptly halting project operations. This financial instability limited resource access, demotivated staff, and threatened the program’s sustainability.

To ensure sustainability, efforts shall be made to integrate digital health media into national health budgets and explore alternative funding models, including donor support and public-private partnerships. Securing diversified financial resources reduced the dependency on single funding sources and enabled more consistent improvements to the system.

Data reliability suffered from inadequate oversight and delayed evaluations. Supervisory teams provided insufficient follow-up, and quality assurance officers were slow to review submitted reports. A manager from a subhealth center stated, “Proper evaluation...was sometimes lacking. Reports we entered were not reviewed for up to two or three months.”

Robust monitoring and feedback systems are crucial for maintaining data accuracy and enhancing overall program performance. Regular monthly follow-up meetings can enable supervisors to provide timely guidance and address emerging problems, fostering accountability and driving continuous improvement. Consistent operational and technical support is also critical for successful app implementation and expansion. As emphasized by a project director, “continuous follow-up is necessary.” Managers also stressed the importance of timely feedback and regular monitoring, with one reporting, “submitted reports should be checked, timely feedback should be provided, and the program should be monitored and followed up.” Furthermore, a shift to more frequent reporting, as suggested by an HMIS manager on a “daily or weekly basis rather than a monthly” could enhance health worker engagement and activity.

The Sehatmandi app’s user adoption was hampered by technical and system limitations. For instance, an HMIS manager (Bamyan) noted, “The system requires revision to align with the Ministry of Public Health's new HMIS package. Specifically, integrating reports such as Health Management Information, Monthly Activity and Achievement, Nutrition, Expanded Program on Immunization, Maternal and Child Health, and Pharmacy reports would transform Sehatmandi into a comprehensive data tracking system.”

A manager from the Eshkashem Health Facility (Badakhshan) mentioned that “data were submitted in an aggregated form,” while another manager from the Naheya Health Facility (Badakhshan) stated, “We had to report both through Sehatmandi and on paper, which increased the workload.”

Although a system’s design may align with the administrative goals, its successful adoption by users hinges on how clearly the project’s objectives are communicated. To ensure user buy-in, it is essential to not only explain the project’s purpose but also clearly outline data formats and workflows during trainings.

This study evaluates the sustainability of the Sehatmandi digital health intervention in Afghanistan by analyzing perspectives from 24 key stakeholders across Badakhshan and Bamyan provinces. Sehatmandi significantly improved performance monitoring, accountability, and timely data reporting in health facilities, particularly in underserved areas. However, the system faced considerable challenges, including poor infrastructure, financial and political instability, high staff turnover, and limited digital literacy.

Digital performance monitoring tools, similar to Sehatmandi, are increasingly being adopted in LMICs and have shown measurable benefits in strengthening accountability and workforce performance. A study evaluating Ethiopia’s electronic community health information system found that digitizing performance monitoring and linking it to incentives significantly boosted personnel motivation and performance [17]. Similarly, Pakistan’s digital immunization tracking app improved user satisfaction, microplanning capacity, and institutional transparency among district managers [18]. In South Africa, replacing paper processes with mHealth-based monitoring improved referral tracking and supervision effectiveness [19]. These examples reinforce that combining digital performance dashboards with structured feedback, incentives, and supervision strengthens workforce accountability and performance in resource-constrained health systems.

The challenges identified in this study, particularly infrastructure limitations and workforce constraints, are consistent with findings from digital health implementations in other LMICs. Prior studies have shown that unreliable electricity, poor internet access, and low levels of digital literacy among health care workers are among the most significant barriers to the effective use of mHealth tools in rural and conflict-affected areas [20,21]. The integration of offline functionality in the Sehatmandi app reflects lessons learned from similar interventions in resource-constrained contexts such as Pakistan where hybrid online–offline systems substantially improved usability in regions with intermittent or weak connectivity [22,23].

Human resource challenges were also prominent in this study, particularly high staff turnover and the need for ongoing training to build and sustain digital skills. These findings are widely echoed in the digital health literature, where workforce capacity issues are closely linked to problems with system adoption, data quality, and long-term program sustainability [20,21]. Structured onboarding processes, periodic refresher training, and regular supportive supervision were identified by participants in this study as important strategies to address these barriers. Evidence from Ethiopia further reinforces this view, demonstrating that building the capacity of health workers and performance monitoring teams, implementing incentive structures for data utilization, involving the community in data validation, and establishing accountability systems for health data are crucial for improving data usage and quality [24].

Financial constraints further exacerbate these challenges. The announcement of the cessation of Sehatmandi funding in mid-August 2021 resulted in a sharp decline in health service utilization, emphasizing the critical need for sustained funding to enhance health service delivery [16]. Insufficient funds impede critical infrastructure upgrades, and delays in payment systems further threaten the program’s sustainability. These findings are consistent with the broader systemic issues in health care financing within LMICs, where limited budgets and reliance on donor funding create significant bottlenecks [25].

This study offers a number of strengths. First, by including perspectives from 2 distinct provinces, Badakhshan and Bamyan, the findings are geographically and operationally diverse, enhancing the generalizability to similar conflict-affected and low-resource environments. Second, the use of thematic content analysis allowed for a nuanced understanding of the facilitators and barriers to Sehatmandi's sustainability. Third, the rigorous translation and transcription process ensured that the linguistic barriers did not compromise data integrity.

However, there are a few limitations. First, the cross-sectional design provides a snapshot of stakeholder perspectives at a single point in time. This limits the ability to assess temporal changes or the long-term impact. Future research employing longitudinal or follow-up interviews could help track the evolution of digital health adoption over time. Second, this study relies on self-reported data, which may be subject to recall bias or social desirability bias. To mitigate this, participants were assured confidentiality, and interviews were conducted by trained researchers in participants’ local language to encourage openness.

The findings in this study point toward several future directions. First, investment in infrastructure, particularly internet access at health facilities, durable digital devices, and backup power sources, is crucial for real-time data use in health decision-making. Second, structured and continuous digital literacy training programs for health care workers are essential to overcome resistance, reduce errors, and ensure consistent data quality. Third, system integration with existing national HMIS platforms is critical to avoid fragmentation and to offer a holistic view of health service delivery. Fourth, exploring diversified funding sources, including public-private partnerships and embedding the intervention within national health budgets, could help sustain digital health systems in volatile political climates. Lastly, embedding ongoing monitoring and feedback loops into digital health implementation will foster accountability, adaptability, and continuous learning.

Participants reported that the Sehatmandi digital health intervention enhanced health system functions, particularly performance monitoring, accountability, and data-driven decision-making, thereby improving service delivery responsiveness in Afghanistan. However, its long-term sustainability and scalability were hindered by infrastructural, technical, human resource, and systemic integration challenges. Adaptive features such as offline data entry helped sustain its use in connectivity-constrained areas. These findings highlight the need for future digital health initiatives in fragile, resource-limited settings to invest in foundational infrastructure, capacity building, follow-up and monitoring, long-term funding, and seamless interoperability with national health systems. Addressing these gaps will enable digital interventions such as Sehatmandi to play a transformative role in improving health outcomes and strengthening health care resilience in LMICs.