Search strategy

On 30^th December 2023, we searched peer-reviewed literature across 10 databases: PubMed, Medline, Embase, Ovid, Web of Science, Global Health, EBSCO SocINDEX, ABI/INFORM, ALNAP, and Sociological Abstracts. Grey literature was also included and sourced from Google Scholar. Our search was timebound to include only papers published from 2008 to 2023. We rerun the search in November 2024 and determined that no new publications met our inclusion criteria. Three groups of search terms were used, relating to 1) local and international humanitarian NGOs, 2) LMICs, and 3) DA. Synonymous terms for DA were compiled, including beneficiary accountability, accountability to affected populations, humanitarian aid accountability, social accountability, impact evaluation, and answerability. Our full search strategy is included in supplementary materials (S2 Table).

Eligibility criteria

We assessed the papers against the inclusion criteria (S2 Table). We included qualitative case studies that used data collection methods such as focus groups and interviews with NGO beneficiaries and staff, reporting on DA effect outcomes amongst NGOs in LMICs. Studies that reported on high- and upper-middle-income countries, as defined by the World Bank [20], were excluded along with studies from governmental organizations, those focused on upward accountability, and studies lacking effectiveness outcome data.

Screening and quality assessment

Title and abstract screening of the initial search was performed independently by all authors using Covidence software [21]. Relevant abstracts underwent a full-text review against inclusion criteria. Two researchers conducted the full-text screening and critical quality appraisal. Two researchers independently used the Joanna Briggs Institute (JBI) Critical Appraisal Tool for Qualitative Research [22] to assess the risk of bias in each of the included studies (S3 Table). This tool is used to appraise the study design, sampling, data analysis, researcher reflexivity, and interpretation of the data. Studies with scores of eight to ten are considered high quality, five to seven medium, and one to four low. To allow for a comprehensive overview, no studies were excluded based on quality assessment. All discrepancies found in the full-text and quality appraisal screening phase were resolved through discussion and by a third researcher.

Data extraction and analysis

Predefined data was extracted from each study, including study design, methods, findings, and recommendations. Core findings related to DA mechanism effect outcomes and beneficiary experiences were collated into a case study synthesis. The analytic method followed the process of Thomas and Harden (2008), which is derived from the thematic synthesis method that links primary research to implications for practice [23]. Studies were imported to the ATLAS.ti qualitative data analysis software [24] that allows for line-by-line coding used to identify key concepts that are grouped based on similarity and stratified into hierarchical categories. Themes were defined based on a combination of seminal works on DA mechanisms, including frameworks proposed by Edwards and Hulme [7] and the Core Humanitarian Standard Alliance [25], which emphasize accountability to beneficiaries through participatory approaches and transparency. In addition, emergent patterns identified during coding were incorporated. Identified themes were further refined and validated based on their recurrence across multiple studies, particularly those focusing on participation, ownership, transparency, program audits, and social audits—central components in evaluating NGO accountability mechanisms. Thematic coding performed by two researchers was then independently used to identify interrelationships within the data and maintain consistency across findings. Throughout extraction and data analysis, studies were discussed amongst researchers to reach a consensus.

Accountability frameworks

We synthesized a comparative analysis of the core components, advantages, and limitations of accountability frameworks. This analysis provides a foundation for understanding how these frameworks influence NGO performance outcomes (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Accountability frameworks.

https://doi.org/10.1371/journal.pone.0324098.t004

Participation

The mechanism of participation is often used synonymously with accountability as it transcends across many processes and tools (Table 3). We found that effective participation mechanisms in NGO projects, such as those implemented in Haiti’s poverty alleviation and microenterprise development programs, are crucial for achieving active beneficiary involvement [26]. Engagement with beneficiaries through participation enhances economic and social outcomes for the underserved by involving them in business ventures, decision-making processes, and community dialogue. This engagement not only promotes self-sufficiency but also ensures that projects are tailored to meet local needs. Collaboration with local authorities and the community to integrate local knowledge was identified as a critical factor [26]. By facilitating two-way communication, participatory representation allows beneficiaries to express their needs and opinions directly, enhancing the effectiveness of NGO activities via focus group discussions and community consultations [27,28]. However, our review also noted that while digital technologies for feedback are beneficial, they often create an echo chamber effect, which can limit meaningful dialogue with NGOs [29].

A common theme throughout our review is that implementing participation mechanisms presents a complex landscape of strengths and weaknesses. While there is a stated intention to empower communities, participation often remains superficial, characterized more by consultative processes than truly collaborative engagement [30]. Our assessment reveals that several factors contribute to this dynamic, including a paternalistic approach by humanitarian staff and inadequate evaluation of accessible communication tools that hinder meaningful engagement in humanitarian contexts [29]. Many participants perceive NGO initiatives as externally imposed rather than genuinely inclusive, undermining sustainable DA [31].

The literature suggests that the facilitators for DA include effective participation that requires cultivating trust, establishing shared goals, and facilitating effective communication for collaboration between NGOs and affected populations [32]. We found that formal meetings leave communities feeling excluded from policymaking and fund allocation decisions [32]. Thematic barriers to effective participation include insufficient staff training and resources, such as inadequate budget allocation, which limit the successful implementation of participation initiatives [33]. Additionally, there is a lack of effective impact monitoring and follow-up mechanisms, exacerbating implementation challenges [18]. Furthermore, conflicting organizational priorities and commitments complicate participation delivery, as NGOs can prioritize their own agenda over community interests [30]. An emerging theme across various studies was that power imbalances within organizational structures play a significant role when donor priorities overshadow local community voices [30,33,34]. The level of perceived community capability influences involvement in decision-making processes [9]. Limited community awareness of their rights contributes to sustaining this power imbalance and reduces the demand for NGOs to be held accountable [12,35].

Ownership

Ownership emerged as a significant theme in our qualitative systematic review of the humanitarian sector (Table 3). Ownership signifies a deliberate act of empowerment through tools such as knowledge transfer and capacity building, aiming for self-reliance. Ownership and empowerment are frequently equated with accountability or seen as outcomes of DA [9,31,36]. We found that ownership is implemented through direct financial assistance and micro-enterprise interventions in several studies [9,31,36]. These interventions provided technical support while allowing beneficiaries to make financial decisions, shifting power to beneficiaries and placing them at the center of program design and evaluation [9,31,36]. Capacity building and knowledge transfer are effective delivery methods for DA that inherently demand empowerment, by connecting community members’ capacity and knowledge to their goals [9].

An aspect of ownership is the transfer of resources, in the form of information and skills, that validates DA and necessitates community buy-in [37]. Various studies demonstrate that this mechanism allows community members to be agents of change, creating demand and fulfilling service roles traditionally managed by NGOs [9,31,36]. Empowered beneficiaries can appraise the NGO using success criteria they devise themselves [37]. Other mechanisms such as participation, feedback, transparency, and social audits play significant roles in community empowerment and ownership in delivering DA.

Transparency

Current practices identified as necessary for DA include transparency and information-sharing between the disclosure of evaluation reports to upward stakeholders, often requiring legal documentation in the case of financial data [10,34], whereas advocates for DA convey the process of relaying this same information to communities as equally essential (Table 3). This overarching theme indicated that the act of informing does not translate directly to effective delivery of DA unless it is utilized as a tool to advance credibility and trust, resulting in increased community participation [6,35].

For example, the type of information that is shared with beneficiaries includes project aims, NGO activity progress, and the prospect of adding value to the community [38]. However, information regarding financial statements and other areas remains ambiguous to communities, which conflicts with the need for transparency [38]. Qualitative data from several studies reveals that in some circumstances communities are unaware of the roles and responsibilities of NGO staff, a key facet in building trust [34,38,39]. In addition, methods used for communicating information involve the establishment of committees that meet with community members at regular intervals [5], displaying information on news boards [28], social mapping [36], and informal propagation of information by word of mouth [9] that can promote transparency.

Despite regulations that demand financial transparency through submission of annual reports [40], in many circumstances, the sharing of information is voluntary and dependent on the willingness of NGOs to share data with communities [41]. Likewise, we found that quantifying social accountability remains a challenge, further limiting NGO transparency to communities [9]. Concepts that arose related to informal methods of information-sharing include knowledge gaps among individuals living in remote locations and those with limited contact, such as the elderly [11]. Conversely, in some studies the use of formal channels is perceived as impersonal, thereby creating distance between NGOs and communities [35,39]. Moreover, claims of transparency are not always true, with as many as 25% of NGOs in one study declaring publicly accessible financial information that was ultimately not provided [39]. Furthermore, examples abound that reputable organizations may misrepresent information to uphold good standing at the expense of transparency [35,39].

Program audit

Our study revealed that program audits, including self-regulation, play a crucial role in maintaining NGO credibility, legitimacy, and public trust (Table 3) [34]. We observed significant variations in implementation and impact within LMICs. In some cases, self-regulation improves transparency and resource management [39], while in others it may fail to prevent mismanagement, as seen with inflated program figures and financial mismanagement [42]. The effectiveness of self-regulation varies across NGOs. Some organizations leverage these mechanisms to enhance transparency and accountability, whereas others face challenges with enforcement and compliance [40]. The efficacy of these mechanisms can be compromised by power dynamics and the potential for misrepresentation [42]. Several examples indicate that the integrity of self-regulation could be compromised when field staff inflate scores to secure more funding, resulting in a loss of learning and information-sharing opportunities [8,42].

In addition, program audits deliver effective accountability through cultivating sustainable improvements in NGO performance and ethical practices in resource management [34]. A common theme found across studies is the prioritization of quantitative outcomes over process outcomes, which can undermine DA through oversimplified monitoring that inadequately captures the complexity of development programs [34,39]. This suggests a lack of effective assessment tools across the humanitarian sector, particularly for small NGOs [34]. In summary, while program audits are requisite to upward accountability, their role in delivering DA requires balanced internal and external oversight, with a commitment to transparency and quality improvement that aligns closely with beneficiary perspectives to be truly effective [34,35].

Social audit

Feedback and community surveillance mechanisms offer a systematic approach to receiving perspectives from affected populations through social auditing (Table 3) [10]. This integrated mechanism ensures DA by monitoring service delivery and fostering trust in the NGO-beneficiary relationship [4]. While feedback systems can effectively amplify beneficiary voices, they often suffer from low engagement levels among beneficiaries [10]. Factors such as fear of losing aid, lack of rights awareness, and failure to act on feedback by NGOs influence the effectiveness of this mechanism [10]. Community-integrated dialogue with affected populations and building up rights-based awareness are aspects of social auditing found integral to the effective delivery of DA [4,10].

NGOs have experienced varying degrees of success in acquiring honest and critical feedback from beneficiaries [10,12,43]. Feedback channels that consider cultural norms and societal restrictions of beneficiaries were found to be vital in improving the usage of feedback systems, particularly regarding the inequitable representation of female participation [43]. Community-led processes demonstrate that stigma-free environments and privacy-focused behaviors were significant factors in the community’s ability to provide feedback for effective DA [43]. The strength of feedback system implementation includes anonymity when reporting issues of corruption or fraud [42,43]. However, a notable weakness is the limited access to technical modes of delivery. We found instances where incomplete feedback loops hinder affected populations from feeling heard and being updated on how NGOs are acting on their feedback [1,42]. NGOs can believe they implemented feedback mechanisms according to their own definitions and goals, while in reality, it does not reach the community as they can have a different definition or expectation of feedback systems [1,42]. Our review suggests that improving communication within the NGO-beneficiary relationship promotes the effective delivery of DA in social auditing [42].

Barriers to DA

Various studies demonstrated that implementation challenges exacerbate intricate macro-level barriers related to existing power asymmetries between donors and beneficiaries, and fragmentation within the humanitarian sector regarding the effective delivery of DA (Table 5) [12,44]. The obstacles to direct implementation strategies include logistical, practical, and cultural barriers that impact the sustainable outcomes of DA. Geographically and historically, organizations that fund aid are in a different region than the beneficiary community, creating a multifaceted disconnect that can inhibit DA [34,44]. Resource constraints create significant logistical barriers that limit staff recruitment, training, accurate data collection, and effective feedback mechanisms necessary for implementation [35]. The literature suggests that accountability measures increase as NGOs become larger [45,46]. The practical aspects of accountability can also be contingent on establishing clear guidelines and consistent practices within an organization’s administration [1,28,47].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Barriers to DA.

https://doi.org/10.1371/journal.pone.0324098.t005

Power asymmetry is another significant issue, as we observed that management in some NGOs can have a low perception of the importance of DA. In some cases, there is explicit disbelief in the knowledge and skill capacity of beneficiaries to participate in their developmental aid [9,31,48]. Barriers that arise from cultural differences among stakeholders can influence the DA implementation process. Westernized individualistic approaches can create resistance from communities, requiring significant efforts to mitigate this barrier to gain productive input from beneficiaries [11,49].

Beneficiaries lack the social and financial currency that other NGO stakeholders have to enforce accountability [50,51]. This power imbalance necessitates DA, since it is not intrinsic but must be manufactured out of the goodwill of NGOs –a goodwill we ascribe to NGOs intrinsically, earned or not [18,52]. This element of “choice” in whether they participate in DA weakens its practice, as it has been consistently documented that beneficiary demands most often compete with and lose to the demands of more powerful stakeholders due to financial dependency or the array of direct barriers described [8,50]. Various studies indicate that power imbalance within communities leads to low representation of feedback from those fearing loss of aid and favors those with financial or social power, hindering effective DA by obstructing comprehensive participation [8,9,27,31,34]. This imbalance undermines the effectiveness of NGO interventions and erodes trust amongst beneficiaries, exacerbated by centralized decision-making that aligns more with government and private sector requirements than local needs [9,30,40]. This often results in NGOs failing to address grassroots issues and being accountable to their intended beneficiaries [12]. Relying on donor funds and limited local dialogue, we found that NGOs frequently import externally designed models rather than developing locally appropriate solutions [9,12].

An emerging theme intersecting these barriers is fragmentation across the NGO sector relating to DA. The lack of legal protection, universal guidelines, and standardized definitions for beneficiary accountability further limits its effective implementation [8,40]. The pervasive belief as indicated in the literature, that NGOs are not responsible for certain actions, exacerbates this issue [40,53]. Additionally, staff members who lack a clear understanding of DA due to the absence of standardized definitions and committed organizational direction, often develop poor attitudes towards accountability practices [1,8]. They may perceive efforts to improve accountability as punitive rather than constructive. Our study also observed that while the standardization and regulation of DA have the potential to influence power dynamics and fragmentation, these efforts may remain performative without genuine and effective implementation of DA practices [8,40,54].

Assessing the impact of DA mechanisms

The effectiveness of DA mechanisms can be evaluated based upon their impact on the sustainability of local development and organizational effectiveness (Table 6) [9,17]. Key indicators include the level of beneficiary influence in aid processes [18], the responsiveness and adaptability of programs to beneficiary need [28], the use of community-led key performance indicators [2], integration with existing local systems [28], the rate of completion feedback loops [2], and the enhancement of internal process capacity for effective implementation [45]. By incorporating both qualitative and quantitative metrics— such as stakeholder feedback, resource management outcomes, and programmatic adjustments— an evaluation tool would allow NGOs to comprehensively assess the effectiveness of their DA initiatives towards sustainable local development [9,17]. While beyond the scope of this paper, the measurement criteria outlined could form the basis of a future empirical tool to assess DA mechanisms.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Criteria for evaluating the effectiveness of DA mechanisms.

https://doi.org/10.1371/journal.pone.0324098.t006

Cultural and social influence on DA mechanism effectiveness

The effectiveness of DA mechanisms in LMICs is significantly shaped by cultural and social dynamics [1,2]. Western accountability practices often fail to resonate in non-Western contexts, leading to systems that feel disconnected from local needs and values [8,11]. Despite this, the intersection of cross-cultural factors with DA has been insufficiently explored in existing literature [1,9].

Colonial legacies and historical power imbalances further complicate DA in LMICs. Communities that have long been subjected to externally driven development may view DA mechanisms as extensions of unequal power structures, fostering skepticism and disengagement. Decolonizing aid discussions emphasize reshaping power dynamics to empower local communities in decision-making [1,35]. When DA mechanisms overlook these imbalances, they can inadvertently reinforce the inequalities they aim to address.

Our review suggests that NGOs often focus on technical accountability solutions, such as procedural transparency while neglecting the cultural and social factors that shape community engagement with DA mechanisms [1,2]. In many LMICs, social hierarchies and authority norms hinder participation, particularly for marginalized groups like women and minorities [1]. These groups often face significant barriers to having an equal voice in holding organizations accountable.

Tailoring DA mechanisms to local cultural and social contexts is essential. There is a growing call for mechanisms that prioritize local perspectives and decision-making, rather than imposing external standards. By doing so, DA mechanisms can become more relevant, effective, and empowering for the communities they are meant to serve.

Enhancing organizational effectiveness and sustainability through DA

The ultimate goal of DA is to foster sustainable local development and enhance organizational effectiveness. The DA mechanisms explored in this paper offer multiple pathways to achieve these objectives, both as tools and direct processes that contribute to lasting impact. Mechanisms such as transparency, program audits, and social audits play a critical role in nurturing public trust by establishing open communication channels between NGOs and the communities they serve. Full disclosure includes providing beneficiaries with access to financial information, resource management, and project planning that empowers communities to make informed decisions and report any instances of misappropriation or unethical conduct. Feedback systems, especially those allowing for anonymity, further encourage active beneficiary participation in the DA process.

When implemented effectively, these DA mechanisms can help mitigate common challenges in aid delivery, such as failure to follow through or failure to act. A person-centered approach, as opposed to using standard assessment metrics, helps to build trust in the NGO-beneficiary relationship, reinforcing the legitimacy of the organization and facilitating demand-driven aid.

Addressing contextual needs through collaboration with local communities and authorities is essential for prioritizing local developmental aid. This can be achieved through beneficiary engagement at all stages of the aid process encompassing planning, decision-making, implementation, and monitoring. Active participation in these processes not only fosters self-sufficiency and empowerment but also ensures that interventions are culturally relevant and practically feasible.

The strength of DA mechanisms lies in their capacity to integrate transparency, social audits, and program audits within community engagement frameworks. This approach maximizes the return on investment in resource management, develops culturally appropriate interventions with high uptake, and ultimately strengthens organizational legitimacy and public trust. The overarching objectives of DA—beneficiary ownership and empowerment—serve as proxies for achieving organizational effectiveness and sustainability, contributing to the long-term success of development efforts.

Improving DA in NGOs: Key factors and challenges

Quality leadership and positive attitudes of front-line staff are essential to improving the implementation of DA through training and fostering a supportive organizational culture [8,39]. NGOs can enhance DA by prioritizing beneficiary engagement, local leadership, and feedback mechanisms over excessive resources for formal evaluations [6,51]. Trust and cultural humility between NGOs and communities are crucial for promoting active participation and driving change [9,11].

However, NGOs face challenges integrating DA into their frameworks due to limited time, resources, and staff, often leading to over-reporting community involvement to preserve donor relationships [28,39]. DA is often treated as supplementary, with inadequate adaptation to beneficiary feedback, signaling a need for more organizational investment in DA [28]. The literature also supports making DA a requirement for donor funding [42,54].

Criticism of results-based management tools highlights their tendency to oversimplify social change and neglect local contexts, leading to rigid, one-size-fits-all solutions that hinder adaptability and responsiveness to community needs [8,52,55]. A balanced approach to DA is necessary, ensuring that affected populations actively shape outcomes rather than being passive recipients [11,12]. Shifting from the term “beneficiary” to recognizing communities as decision-makers can transform accountability dynamics and improve humanitarian aid effectiveness.

Strengthening DA in humanitarian NGOs: A path forward

For effective DA in humanitarian NGOs, a systematic prioritization of local development through policy change is essential [28]. While initiatives like the Participation Revolution in the Grand Bargain aim to improve resource management, there is still much to be done in achieving demand-driven aid outcomes [56]. Policies should emphasize a rights-based approach, self-reliance, and integration with donor-driven DA to mitigate implementation barriers. Effective DA requires listening to stakeholders, including affected populations, and empowering them with equal voice and power [6,27].

The erosion of trust in humanitarian aid calls for policies founded on transparency, including regular communication with communities about project details and rights [8]. Standardized legal frameworks for engagement and resource management are also recommended [6,10,40]. Recent policy shifts favor bottom-up evaluation models, like the Grounded Accountability Model [57], which incorporates continuous community feedback to develop localized indicators. Although resource-intensive, such models yield high returns in community engagement.

DA and accountability to donors can complement each other. Engaging donors and governments directly with aid recipients cultivates trust and cooperation, ultimately enhancing NGO impact [54]. By viewing upward and DA as interconnected, NGOs can facilitate open communication, balance functional and strategic accountability, and shift focus from reporting to advancing sustainable, long-term transformation [1,9].

Haiti: A case study on the need for DA in the “NGO Republic”

Haiti’s history illustrates the consequences of inadequate DA mechanisms, particularly in post-crisis humanitarian efforts. Often referred to as “The NGO Republic,” Haiti has seen extensive aid investment without significant tangible results, contributing to the country’s ongoing crises of food insecurity, displacement, and gang violence [10,38]. Following the 2010 earthquake, NGOs flooded the country yet failed to meet the needs of the population. Three major DA barriers were identified: 1) Inappropriate allocation of resources, 2) Fragmented response and recovery efforts, and 3) Power asymmetry, where NGO priorities diverged from the needs of the population [10,38].

The influx of NGOs in post-earthquake Haiti displayed a lack of contextual understanding and direction, leading to activities that were misaligned with national health norms and abandoned without proper transition, which undermined health system recovery [10,38]. This lack of transparency and accountability led to distrust, leaving affected Haitians disempowered and uncertain about the future impact of aid. The failure to meet community needs underscores the demand for a shift toward strategic accountability, which emphasizes local leadership and equal participation to drive sustainable, effective delivery of DA [10,38].

Research gaps in the effectiveness of DA

Our literature review reveals a significant research gap: a systematic review of DA mechanism effectiveness has never been conducted. This gap highlights the importance of DA in achieving positive outcomes in aid activities. While numerous studies discuss accountability mechanisms, there is a disproportionate focus on participatory approaches, with a lack of evaluation and reporting on DA practices (Table 3). Participation was cited 97 times, while program auditing was cited 69 times, yet evaluation reporting appeared only 6 times, indicating a lack of transparency and an oversimplification of DA as synonymous with participation.

Moreover, our review found no standardized criteria for measuring DA effectiveness. Table 3 shows that while the effectiveness of DA mechanisms is frequently discussed, the mechanisms themselves and their components are often overlooked. This gap stems from the historical uncertainty about what constitutes DA, making it challenging to evaluate and measure its impact effectively. This gap in research emphasizes the need for clearer definitions, classifications, and methods to assess DA mechanisms and their outcomes.

Strengths and limitations

A key theme across the literature is the bias in interviews regarding DA practices, particularly the lack of representation from beneficiaries’ perspectives. Internal reports often overrepresent satisfaction with NGO activities due to fear of losing aid or feeling unqualified to offer negative feedback. Additionally, there is a gap in research related to the demographic representation of marginalized populations. The variability in data quality (S3 Table) further complicates the analysis, as some studies fail to account for interviewer bias or influence on participant responses. Moreover, the lack of detailed analytical methods makes it challenging to assess the rigor of studies across findings. However, triangulation and peer review were employed to enhance the qualitative review’s rigor, utilizing diverse data collection methods, such as focus groups, surveys, and interviews. The strength of these methods lies in their potential applicability across various humanitarian contexts.

Definitions

Downward accountability: being answerable to affected populations for actions, giving them influence over key decisions which are made throughout the project’s lifetime that is a dynamic process of listening, adapting, and responding while creating sustainable systems. Accountability mechanisms: tools and processes that deliver accountability; tools can be measured at specific points in time with tangible documentation; processes are a course of action where the means of delivery is more tangible than the results. Social auditing: community surveillance, needs assessment, focus groups, surveys. Feedback Systems: complaint mechanisms, involvement of communities in evaluating activities, responsibility to improve quality of program actions. Program auditing: internal assessment, reporting and improving on social performance and ethical behavior, performance evaluation, self-regulation and enforcement. Evaluation reporting: transparency, disseminating information to communities on how finances are used, organizational structure, and program outcomes, communicating motives, resources, power dynamics, limitations of work, and decision making process, aims to maintain community trust, being upfront about access to human rights. Ownership: knowledge transfer, capacity building, sustainability for self-reliance, integration with existing systems, local leadership, beneficiaries making the decisions and those decisions are implemented. Participation: consulting, involving, collaborating, community engagement.