Information and communications technology use to prevent and respond to sexual and gender-based violence in low- and middle-income countries: An evidence and gap map

There is a small but growing evidence base underlying ICT interventions for prevention and response to sexual and gender-based violence in low- and middle-income countries

Methodologically rigorous evidence examining the effectiveness of information and communication technology (ICT) in preventing and responding to sexual and gender-based violence against women and children in low- and middle-income countries is still relatively low, and almost non-existent with respect to children. However, with increasing interventions, future high quality studies appear to be increasing.

What is this EGM about?

The problem of sexual and gender-based violence (SGBV) against women and children is prevalent in low- and middle-income countries (LMIC). Adoption of ICT has increased exponentially throughout the world in the past decade. As new types of ICT interventions are being implemented with the purpose of preventing and/or responding to SGBV against women and children in LMICs, donors and policymakers have been looking for evidence based on rigorous studies that assess the effectiveness of such interventions. This EGM shows the available evidence from individual studies and systematic reviews.

What is the aim of this EGM?

The primary goal of this EGM is to establish a baseline for the state of evidence relating to the use of ICT for preventing and responding to sexual and gender-based violence against women and children in low- and middle-income countries.

What studies are included?

The EGM includes systematic reviews, academic studies and gray literature from January 2005 to July 2021, focusing on: 1.

women and girls

Ten studies met the inclusion criteria. Of these, two are systematic reviews, two are scoping reviews and the remaining six are individual studies (two of which had separately published protocols, which are also included in the EGM).

Studies and reviews were categorized by type of intervention, and outcomes reflected the RESPECT Women Framework for preventing violence against women (World Health Organization, 2019a) and the INSPIRE Seven Strategies for Ending Violence against Children (World Health Organization, 2016).

What are the main findings of this EGM?

No studies focused on using ICT exclusively for children. There is some evidence of impact in the areas of using ICT: 1.

to influence gender norms related to SGBV prevention and/or response;

What do the findings of the EGM mean?

The results indicate a need for more rigorous studies on using ICT for SGBV prevention and/or response conducted in LMIC.

Specific research gap areas include: using harmonized and standardized impact indicators, child-focused interventions, and studies explicitly linking how ICT contributes to intermediate outcomes related to SGBV (identified under the RESPECT and INSPIRE frameworks) and ultimately to prevention and response outcomes.

How up-to-date is this EGM?

The authors searched for studies published up to July 2021.

Introduction

The intervention

Why it is important to develop the EGM

SGBV is a persistent global phenomenon. ICT represents a potential facilitator for SGBV prevention and/or response. Very little evidence-based knowledge exists regarding its use, effectiveness, and feasibility in LMIC. This scarcity of evidence impedes important government, NGO, and activist efforts to better understand ICT's potential to contribute to preventing and responding to SGBV. An EGM is a tool for better understanding this evidence landscape including: description of where ICT work is being done (context); focus population; whether it is to prevent or respond to SGBV; and the quality and the diversity of the body of evidence that exists. An EGM serves to establish a baseline to inform future research efforts.

Systematic reviews that address the evidence related to preventing SGBV and the use of ICT to achieve other health outcomes (e.g., relating to HIV, maternal health, etc.) have been published. However, there are few to no reviews (systematic or EGMs) identifying evidence related to the use of ICT for outcomes related to prevention and/or response to SGBV against women and children in LMIC. Those that exist are limited in scope and focus on higher-income countries (Anderson et al., 2019). (See Section 4.1 “EGMs: Definition and Scope”).

Further, as the use of ICT expands at such a rapid pace both within and outside the context of SGBV prevention and/or response, the need for methodologically rigorous research requires advocacy amongst both the ICT and SBGV research communities. We confirmed that few rigorous studies exist meeting the eligibility criteria, but we also believe that determining a baseline of existing evidence, as few as they may be, is an essential starting point for an advocacy strategy encouraging more rigorous studies to be conducted.

To further underscore the timeliness and importance of this review, we note that emerging reports show a correlation between the Covid-19 pandemic and increasing gender-based violence, especially IPV (Chukwueke, 2020; Mlambo-Ngcuka, 2020). Accordingly, we attempted to explore emerging evidence related to SGBV during the pandemic, and the role ICT can play in prevention and/or response interventions. For example, the Gender-Based Violence Information Management System in Mali generated data demonstrating a 35% increase in SGBV between April 2019 and April 2020 (UNFPA, 2020). This SBGV trend has also been noted in other African countries. At the time of submission of the EGM, the authors did not identify eligible studies exploring this important intersection.

An EGM that locates evidence connected to the operationalization and implementation of ICT for preventing and responding to SGBV is not only warranted, but critical for implementers to comply with the principle of “do no harm” when delivering interventions in potentially vulnerable contexts.

EGM: Definition and scope

EGMs represent a relatively new approach to identifying and classifying available research and evidence for a broad scope or topic, such as ICT for SGBV prevention and/or response. EGMs offer a rigorous and systematic process to “map” what evidence exists and where also gaps in evidence exist. They are ideal for topics where there is little understanding of the research landscape; or those that incorporate several potential interventions with multiple outcomes and may lack widely available studies. EGMs differ from systematic reviews in that they can identify clusters of evidence that may be further analyzed by meta-analysis with each cluster representing a potential systematic review in itself (Saran & White, 2018).

An EGM is ideal for identifying the evidence connected to the use of ICT for the prevention and response of SGBV against women and children in LMIC (those listed by the Organization of Economic Co-operation and Development as “low” and “lower-middle” income) precisely because the scope of SGBV and ICT encompasses several potential intervention designs with multiple target populations and outcomes. An EGM offers a systematic approach to locate and analyze what evidence exists for using ICT to prevent and respond to SGBV and highlight evidence gaps. While EGMs do not include meta-analysis per se—the clusters they can identify are often readily developed into individual systematic reviews. In search strategy and methodological rigor, however, EGMs differ little from systematic reviews.

Systematic reviews require precision with inclusion and exclusion criteria to properly meta-synthesize data. Given that we expected little, but diverse available evidence for SGBV and ICT, a systematic review was deemed likely to include very few studies and miss other key evidence that is necessary to understand the broader context of ICT for SGBV prevention and/or response research in LMIC. An initial EGM, in contrast, allows us to first identify what, if any, evidence clusters do exist, and further develop these into individual reviews, if possible.

Stakeholder engagement

A steering committee of SGBV and ICT experts, primarily including representatives from LMIC, have advised and assisted in the conceptualization of the EGM and contributed to a concurrent landscape review that was conducted in conjunction with the EGM. The committee included experts in ICT, gender and SGBV, and were comprised mostly of women. The committee included academics, civil society actors, and activists. The committee assisted in the identification of stakeholders for key informant interviews, interventions, and approaches that helped guide the EGM. The landscape review helped to identify what types and where ICT for SGBV prevention and/or response interventions are being or have been implemented in LMIC, but not reported in peer review literature. In July 2021, an online participatory workshop was convened, including the steering committee, global specialists, academics, implementers, and donors who work with ICT and SGBV, to review preliminary findings of the landscape review and EGM to develop a framework in the form of an ecological model for analyzing the use of ICT to prevent and respond to SGBV in LMIC (Mechael et al., 2021).

Conceptual framework development and scope

The framework for this EGM was developed through consultation with stakeholders (key informant interviews), a global steering committee of experts, and through the application of existing violence prevention frameworks (e.g., RESPECT and INSPIRE), theories and policies. As a complement to the EGM, a landscape review of interventions using ICT to prevent and respond to SGBV against women and children in LMIC captures work being conducted by activists, civil society, NGOs, and other groups that are not represented or published in traditional academic and research settings (and therefore likely to not be identified through systematic searches of peer reviewed literature).

Because of challenges measuring primary SGBV outcomes, especially prevention, different groups may be targeted as mediators (intermediaries) for achieving intermediate outcomes that contribute to, or are inferred, based upon available evidence to contribute to, SGBV prevention and response primary outcomes. The research team acknowledges that interventions, for the purpose of ultimately preventing and/or responding to violence, may have the objective of achieving an intermediate outcome that contributes to the primary outcome (or impact) of either preventing SGBV or responding to SGBV (e.g., facilitating access to services, including the prevention of the SGBV recurrence for survivors of SGBV). Often there is more than one intermediate outcome, attributable to multiple interventions, which together form a causal pathway leading to the primary outcome (or impact) such as SGBV prevention (e.g., women's economic empowerment and strengthening relationship/life skills) (World Health Organization, 2019a) (see Figure 1).

OPEN IN VIEWER

Figure 1

Intermediate outcomes and causation

We attempted in our searches to identify studies that evaluated intermediate results that may be components of Theories of Change or are components of an ecological model that contribute to, or are prerequisites leading to primary outcomes (or impact), specifically prevention and response. It is difficult to measure primary prevention outcomes or impact (e.g., a reduction in SGBV prevalence) without a longitudinal study.

Established evidence related to the prevention of sexual violence supports inferences that achieving one or more intermediate outcomes will necessarily lead to, or contribute to prevention (reduced SGBV prevalence), or effective response interventions (greater access to services). It was anticipated and acknowledged that studies may tend to more likely evaluate the effectiveness of an intervention in achieving an intermediate outcome related to the prevention or response to SGBV against women and children, referring to the RESPECT and INSPIRE frameworks as guides (assuming that evidence regarding the prevention of violence in general can apply to the prevention of the more specific SGBV). Based upon evidence, we sought to infer that the intermediate outcome will either directly lead to, or is a necessary ingredient for achieving the primary outcome/impact (e.g., SGBV prevention or improved access to quality SGBV services).

We also attempted to search for studies that targeted and measured outcomes connected with groups other than women and children. These targeted intermediary groups include SGBV service providers (social workers, health providers, prosecutors, etc.), potential and actual SGBV perpetrators (including intimate partners), community members, and organizations.

The World Health Organization's RESPECT evidence-based framework for the prevention of violence against women (including SGBV) is useful to conceptualize SGBV intervention outcomes (World Health Organization, 2019a). The RESPECT framework conceptualizes risk and protective factors for violence (including SGBV) across four environmental levels: individual, interpersonal, community, and society. These divisions offer an important roadmap to consider SGBV outcomes that go beyond individual attitudes, knowledge, and behaviors and look towards structural outcomes at both the community and society level. Similarly, the INSPIRE Seven Strategies for Ending Violence Against Children (World Health Organization, 2016) presents evidence-based strategies for preventing and addressing violence, including SGBV, against children. The RESPECT and INSPIRE frameworks have been used to inform the EGM's outcome dimensions (see Figures 2 and 3).

OPEN IN VIEWER

Figure 2

RESPECT Preventing Violence Against Women Framework (WHO, 2019)

OPEN IN VIEWER

Figure 3

INSPIRE Seven Strategies for Preventing Violence against Children (WHO, 2016)

We also drew heavily from the results of an online workshop we held with SGBV and ICT stakeholders held in August 2021.

See Figures 4–6 for the Stakeholder Workshop Identification of ICT Interventions and Outcomes for Preventing and Responding to SGBV.

OPEN IN VIEWER

Figure 4

Conceptual framework for ICT interventions for SGBV (developed at Stakeholder Workshop).

OPEN IN VIEWER

Figure 5

ICT SGBV outcomes identified in Stakeholder Workshop.

OPEN IN VIEWER

Figure 6

Stakeholders connected with using ICT for SGBV prevention and response identified in Stakeholder Workshop.

Dimensions

The conceptual dimensions used for coding the EGM reflect key elements of using ICT for the prevention of, and response to, SGBV for intervention design, organized by type or description of ICT intervention, target population (or user) intervention outcome, study design, and other descriptive factors such as region and sponsor organization.

We organized the EGM data along the axes of intervention context by intervention outcome. The columns represent the context of the ICT interventions (IPV focus, focus on prevention and/or response, discussions of sustainability, and unintended consequences). While the rows lay out different intervention outcomes. We developed the codes for our outcomes by using the WHO RESPECT Framework (World Health Organization, 2019a). Our outcomes include both primary (SGBV incidence/prevalence) and intermediary outcomes, such as women's empowerment and access to SGBV-related services. We also coded for studies that were descriptive in nature and did not include explicit outcomes, as well as coded for feasibility studies which evaluated a potential or developing intervention. Included studies that had more than one outcome were coded more than once as appropriate. However, we attempted to avoid the overuse of double coding.

We segmented our EGM dimensions by study design (RCT, Qualitative, Quantitative, or Mixed-Methods). There was no double-coding for study design. We included many other descriptive filters such target population and type of ICT device used by the intervention.

Analysis and presentation

Data collection and analysis

Description of studies

Synthesis of included studies

Unless otherwise specified, we refer to the primary study with the published results.

See Summary of findings Table 2 (primary studies)

See Summary of findings Table 3 (“systematic” reviews)

Resulting evidence from studies

There is not yet a single body of evidence of methodologically rigorous studies from which generalizations on impact can be made. There is also not yet a cluster of evidence demonstrating the consistent impact of a particular type ICT intervention for preventing or responding to SGBV against women and children in LMIC.

With that said, several studies, based upon the rigor of methodology and outcomes, demonstrated promising results as a baseline warranting further research, particularly in different contexts. One study that demonstrated impact in preventing (and responding) to SGBV was the Decker, Wood, Hameeduddin, et al. (2020) study evaluating how an ICT safety decision aid app (MyPlan), proven effective in HIC contexts, can be adapted to be effective in an LMIC context, specifically Kenya. Intervention groups reported increased usefulness of safety strategies, increased in resilience and significant decreases in the risk for lethal violence compared to the control group. Intervention participants also demonstrated immediate post-intervention safety preparedness relative to control participants. Both intervention and control groups in the Kenya study reported reductions in IPV (Decker, Wood, Hameeduddin, et al., 2020). Current efforts are underway to study the impact of using MyPlan in other LMIC contexts including Somalia. (Glass, 2017)

The Clark et al. (2020) study, an RCT, highlighted how ICT (SMS and IVR), although not studied in isolation with a control, can effectively contribute to a multi-media approach toward reducing SGBV (IPV) against Nepalese women. The Bartels et al. (2019) study demonstrated how ICT (using the SenseMaker M&E tool) could be used to facilitate gathering the impressions of women regarding the quality of, and important criteria for, SGBV/GBV services in Lebanon (humanitarian setting). These impressions can be used to inform the quality of SGBV services (and improve access) although route of exploration was not an objective of the Bartels study. The Mishori et al. (2017) study highlighted how ICT (using MediCapt) could be used as a tool by clinicians and prosecutors for collecting forensic information in cases of SGBV. Whether the forensic evidence collected contributes to medical, legal, and human rights outcomes is under current study (Mishori et al., 2017, p. 150).

The four reviews did not present consistent conclusions regarding impact. Two of the four reviews looked at existing eHealth and/or mHealth interventions/tools and their impact on IPV. The Linde et al. (2020) review, evaluating the effect of “eHealth interventions compared with standard care on reducing IPV, depression, and post-traumatic stress disorder (PTSD) among women exposed to IPV,” as well as secondary outcomes of “type-specific IPV,” concluded that there “is not evidence from randomized trials of a beneficial effect of eHealth interventions on IPV.” The Linde et al. (2020) review recommended “more high-quality trials” and “harmonizing outcome reporting in IPV trials by establishing core outcome set.” Similarly, Anderson, looking at mHealth interventions, concluded that evidence for efficacy of mHealth interventions compared to conventional IPV was “limited” (Anderson, Krause, et al., 2019). The Eisenhut et al. (2020); and Hayes (2014) reviews were more of scoping or landscaping reviews which did not include reviewing methodologically rigorous studies of impact.

Risk of bias in included reviews and quality of evidence for individual studies

The four reviews included were analyzed for risk of bias and quality of evidence using AMSTAR-2 tool for systematic reviews (Shea et al., 2017). Two reviews (Anderson, Krause, et al., 2019; Linde et al., 2020) scored high on the AMSTAR-2 tool presenting low risk of bias. The third review (Eisenhut et al., 2020) demonstrated some risk of bias, as the review did not include information related to study design selection criteria, nor information as to whether or not analysis and extraction were performed in duplicate. However, this review did not search for published peer-reviewed studies in scientific databases, but searched for mobile phone applications specific to SGBV directly though mobile application hosting platforms (i.e., app stores). The forth review (Hayes, 2014) was not truly systematic in nature and, as a result, scored low on the AMSTAR-2 tool. However, (Hayes, 2014) is better characterized as a desk review of several case studies. The authors were cognizant of this distinction from a traditional systematic review in their methodology, which was otherwise sound. The review therefore met the inclusion criteria for this EGM.

Though optional, the six individual studies were analyzed for quality of evidence using several tools. The three included RCTs where analyzed using the Cochrane Tool for Risk of Bias for RCTs. Two RCTs (Brody et al., 2021; Decker, Wood, Hameeduddin, et al. (2020) were at low risk of bias according to the Cochrane analysis, while Clark et al. (2020) presented some concern of bias regarding the measurement of its outcomes. The remaining three studies were all qualitative (feasibility studies) and analyzed using the Critical Appraisal Checklist for Qualitative Research from the JBI Manual of Evidence Synthesis (Aromataris & Munn, 2020). All three studies scored highly for quality of evidence with low risk of bias.

Additional dimensions (if applicable)

Summary of main results

Broadly, there is an, overall dearth of methodologically rigorous studies conducted in LMIC examining the use of ICT for the prevention of, and response to SGBV. While there were several studies addressing SGBV against women, we did not identify any eligible studies specifically addressing SGBV against children (i.e., child sexual assault or “CSA”) nor against LGBTQ populations (i.e., targeted attacks against transgender or gender non-conforming persons). There does appear, however, to be an upward trend in LMIC to study how ICT may be best leveraged to address SGBV against women and to a lesser extent girls. There is also a limited number of ongoing studies in LMIC that use ICT for addressing VAC/CSA, including gamification for adolescents to change underlying attitudes and beliefs and digital parenting interventions. Overall, we found a scarcity of studies measuring primary prevention, with most uses of ICT focused on secondary and tertiary prevention of SGBV (otherwise characterized as response interventions).

While there is no single body of evidence demonstrating the consistent impact of a particular type of ICT intervention for preventing or responding to SGBV against women and children in LMIC, we noted several promising completed individual studies.

One study that demonstrated impact in preventing and responding to SGBV was the Decker, Wood, Hameeduddin, et al. (2020) study evaluating how an ICT safety decision aid app (MyPlan), proven effective in HIC contexts can be adapted to be effective in an LMIC context, specifically Kenya. Intervention groups reported increased use of safety strategies, increases in resilience and significant decreases in the risk for lethal violence when compared to the control group. Intervention participants also demonstrated immediate post-intervention safety preparedness relative to the control. Both intervention and control groups in the Kenya study reported reductions in IPV (Decker, Wood, Hameeduddin, 2020, p. 1).

While safety apps like MyPlan can be strategic in reducing SGBV by helping potential SGBV targets (of both sexes) in evaluating their circumstances and avoid situations in which they have a heightened risk to SGBV, they do not necessarily address the underlying attitudes and norms fueling SGBV. The most promising type of intervention involving ICT for primary prevention would be an intervention that has an objective related to changing gender and social norms (toward SGBV). This would include changing underlying attitudes toward SGBV acceptability by both the perpetrator and the victim/survivor. In some cultures, women and girls (and others) may think that SGBV is something to which they must acquiesce and resign themselves. Interventions that target changing gender and social norms go beyond increasing knowledge and targetthe root causes of most SGBV. Accordingly, interventions targeting young people (adolescents) and developing social norms that reject SGBV, serve as potentially the more promising interventions for primary prevention of SGBV. Ongoing studies around gamification (BREAKAWAY Guatemala Pilot Study) and digital parenting through the Parenting for Lifelong Health program (Digital Parenting Studies) could confirm the strategic use of ICT for contributing to primary prevention of SGBV.

The Clark et al. (2020) study seems to infer that changes to underlying attitudes regarding SGBV require a multi-media approach involving both community discussions, reinforced by ICT messaging.

Formative studies that measure usability, feasibility and elicit participatory input from an ICT intervention's targeted user population are also necessary ingredients for effective ICT design and implementation They are as equally as important as studying and measuring impact because they influence the uptake and sustainability of an intervention, a consideration we did not find often addressed nor emphasized in those studies and reviews we deemed eligible for this EGM. Formative studies are acutely pertinent because local contexts and enabling environments for ICT are varied and cannot be generalized, particularly in LMIC.

One such study (Mishori et al., 2017) described the formative process, including needs and resource assessments, end-user feasibility assessments and field pilot testing, for an ICT-based sexual violence data collection tool (MediCapt) in the DRC. While most studies that turned up in our search described the use of ICT in prevention of, and response to SGBV originated in HIC (albeit not eligible), we identified only one study demonstrating a formative process for how ICT interventions proven effective may be adapted form a HIC to an LMIC context (Decker, Wood, Kennedy, et al., 2020).

While we excluded studies and interventions conducted in HIC from eligibility, we did find that the reviews containing a disproportionate number of studies conducted in HIC included interventions in the areas of addressing intimate partner violence (IPV), and evaluating the effectiveness, acceptability, and “suitability of ICT” for addressing different aspects of awareness, screening, prevention, treatment, and mental health. A number of these studies explored how current mHealth web-based eHealth interventions could be leveraged to address SGBV. While these HIC-based studies published minimum evidence of effectiveness (and in the case of Linde et al. (2020) no evidence of effectiveness in using mHealth/eHealth) in addressing SGBV (with the exception of safety apps), they do provide a foundation and some limited promising results to further explore, particularly for adaptation to LMIC contexts (see, e.g., Anderson, Krause, et al., 2019; Braithwaite & Fincham, 2014; El Morr 2020, Glass, 2017; Glass et al., 2021; Rempel, 2019).

Measuring effectiveness in term of absolute changes in levels of SGBV (e.g., IPV incidence) with validated tools, can be difficult and requires longitudinal resources. Effectiveness studies, however, can also measure evidence-based intermediary outcomes such as those outlined under the RESPECT and INSPIRE frameworks published by the World Health Organization. Validated tools for measuring different impact outcome variables, whether primary (changes in SGBV prevalence), intermediary (e.g., changes in women's empowerment, or changes in reported cases of sexual violence) are varied. Clear consensus for standardized indicators and measurement tools does not yet exist. The research we observed during our review reflected this, and at least one of the reviews explicitly noted the lack of harmonized outcome indicators (Linde et al., 2020). Ironically, the observation about the lack of a lack of homogeneity among the outcome measurements, an obstacle to making analytical comparisons, was also noted in a review of ICT used for SGBV in HIC as well (El Morr, 2020, p. 1)

Of the studies that measured some sort of intermediary “impact,” there appears to be more studies measuring how ICT may increase women's “empowerment,” strengthen relationships and contribute to conflict resolution. There also seems to be a small cluster of studies gauging ICT end-user acceptance and feasibility, both necessary for adoption and use of ICT facilitated interventions contributing to SGBV prevention and/or response outcomes.

See Figure 2 RESPECT Preventing Violence Against Women Framework (World Health Organization, 2019a).

See Figure 3 INSPIRE Seven Strategies for or Preventing Violence Against Children (World Health Organization, 2016).

Areas of “evidence clusters”

As mentioned, even though there was a low number of included studies, certain areas emerged based on a few individual studies which appear promising for future research efforts in LMIC. These areas, however, cannot be characterized as “clusters of evidence.” All of the eligible studies focused on prevention and/or response to SGBV against women and girls, although several limited the target group to women above a certain age. All of these studies were diverse in outcomes, measurements, and design.

It is important to note that the bulk of the evidence from the systematic reviews (that included both HIC and LMIC-based interventions was from HIC settings and reported that the heterogeneity of study design and IPV outcomes made generalization and meta-analysis difficult across studies. Results from Clark et al. (2020) did not report significant changes in IPV between the control and intervention, though the intervention was associated with some decreases of certain forms of IPV at endline. (Clark et al., 2020.) The Clark study, however, did not isolate the digital ICT component for evaluation, but rather evaluated a combined multi-media intervention.

The types of ICT intervention represent a broad spectrum (SMS, IVR, web-based decision-algorithms, etc.) and no one single type of intervention emerges as predominant in the body of eligible studies from LMIC. However, studies from HIC, although not included as eligible for this EGM, seem to suggest that the use of SMS to be multifaceted and leveraged for different purposes related to both prevention and response: underscoring communications for changing gender norms, building capacity of service personnel, connecting survivors to each other and with services, reporting incidents, mental health support, etc. Adapting those interventions proven effective in HIC for LMIC contexts, could be strategic. The MyPlan adaptation protocol could serve as a model guide, as well as SVRI's Adaptation Guide under development (Sexual Violence Research Initiative, 2021).

Targets, in terms of users and beneficiaries, primarily included women. However, including potential perpetrators of SGBV as targets (in terms of users) may be strategic in realizing prevention goals. The Clark et al., 2020 study is an example which included communities and couples. The ongoing Breakaway study, targeting both sexes of adolescents (boys and girls) may be informative for future targeted interventions.

Areas of major gaps in the evidence

Our review reveals that there are few methodologically rigorous studies of ICT-facilitated interventions for the prevention and/or response of SGBV in LMIC, especially relative to the number of studies conducted in HIC. There is a dearth of studies utilizing RCT methodologies, the “gold standard” for demonstrating evidence of effectiveness. While we did identify some evidence looking at how ICT can be facilitated for interventions to prevent and/or respond to SGBV in LMIC, there were overall very few studies. There is not yet a strong consistent body of evidence connected with using ICT for SGBV in LMIC.

The reviews of ICT interventions indicated that there is not yet a consensus nor harmonized set of outcome indicators relating to using ICT in SGBV prevention and/or response. We observed some studies measuring intermediate outcomes as defined under the RESPECT framework, but a more explicit linking of these intermediate outcomes to SGBV primary prevention would strengthen the evidence base for using ICT for SGBV prevention and/or response. We also noted studies that measured changes in attitude regarding IPV, but not with a consistency that would allow for characterizing them as “clusters.” Further, there is a gap in the number of studies that isolate and compare the ICT component in the intervention target group with a control target group in which ICT is not used, such as in the Clark et al. (2020) study.

As evidenced by the lack of included eligible studies, research is needed to examine the role that ICT can play in addressing SGBV against children, adolescents, and LGBTQ populations in LMIC. This gap is highlighted by the increasing proliferation of smart phone use amongst children and youth in LMIC. One such explanation for the dearth of studies examining the use of ICT regarding SGBV against children may be the immense challenges securing an Insitutional Review Board (IRB) approval for any study involving children as subjects. There is also an exponential growth of technology-facilitated SGBV that is not fully understood in LMIC, especially as this relates to human trafficking (see, e.g., Segrave & Vitas, 2017; Van Reisen et al., 2017). As technology-facilitated SGBV violence has become a common trait connected with trafficking, more efforts to study how ICT can be used to counter and prevent it are warranted. Interventions with the objective of addressing technology facilitated violence were excluded from this EGM.

In the same vein, while there are a number of interventions targeting SGBV response (secondary and tertiary prevention), noticeable gaps exist related to how ICT can contribute to addressing the underlying reasons for SGBV for primary prevention (e.g., addressing gender and social norms.).

We identified two studies that examined how ICT can facilitate SGBV-related data collection (Bartels et al., 2019; Mishori et al., 2017). Though they serve as a springboard for studying how ICT can facilitate forensic and testimonial data collection, more impact research is needed to understand how data collection can help lead to better prosecution of perpetrators and ensure the stronger enforcement of laws (which in turn can become deterrents for future SGBV offenses). As such, it is critical to create and analyze valid Theories of Change that link ICT-based data collection with better enforcement of laws and prosecutions of perpetrators.

Other gaps in the research of ICT for SGBV prevention and/or response interventions in LMIC include, but are not limited to:

Studies stating a clear theoretical basis and Theory of Change for the intervention, both deemed essential criteria by practitioners and academics for a study of an intervention to be characterized as methodologically rigorous.

Potential biases in the mapping process

As authors of this EGM, we made conscious effort to reduce risk of bias wherever possible. We developed the EGM's protocol, methodology, search strategy, data extraction and analysis under the guidance of a steering committee of diverse SGBV experts and stakeholders with expertise in ICT-based interventions and LMIC contexts. The steering committee gave continuous feedback regarding the direction of the EGM, including potential areas of bias. Searches and data extraction were undertaken by two independent researchers with discrepancies resolved by discussion to avoid researcher bias. We sought to be comprehensive in the employment of search terms and database selection. We do not claim that this EGM is totally free of limitation or bias, however. While searches were exhaustive, our searches were limited to English, Spanish and Portuguese (our spoken languages). We may have missed evidence published in other languages. While we reviewed all major scientific and academic databases, we may have also missed evidence arquired in smaller, country-specific, or more niche-specific databases. Given the relative scarcity of ICT-based interventions for SGBV in LMIC, we feel confident that the evidence in this EGM is comprehensive. The Landscape Review completed in tandem with this EGM also confirmed the scarcity of evidence for ICT-based interventions in LMIC outside of the context of peer-reviewed content.

Limitations of the EGM

The primary limitation to the EGM is the death of methodologically rigorous studies examining the impact of the use of ICT to prevent or respond to SGBV against either women or children in LMIC.

Stakeholder engagement throughout the EGM process

There were no substantive changes in the stakeholder engagement regarding the Steering Committee throughout the EGM process. The authors also conducted an online workshop with 24 SGBV and ICT stakeholders in July 2021 to elicit input into a conceptual framework, and for framing a “Call to Action” for methodologically rigorous research into the use of ICT for prevention of and/or response to SGBV against women and children in LMIC.

Implications for research, practice, and/or policy

This review shows that there is a scarcity of methodologically rigorous studies reporting on the use of ICT in LMIC for the prevention and response to SGBV against women and children. This includes studies ensuring the participation of at least one LMIC-based researcher as an author.

Our landscape review (separate from this EGM), completed in tandem with this EGM, demonstrated a lgrowing number of ICT-based programs and interventions for the prevention of, and response to SGBV, in LMIC. However, these have not yet been rigorously evaluated. Future research is needed to better locate, document, and most importantly, evaluate the effectiveness and impact of using ICT for SGBV prevention and response in LMIC contexts.

Most striking is the paucity of studies examining the use of ICT in connection with preventing or responding to SGBV (CSA) against children. In light of the exponential increase in the use of ICT by children and adolescents, even in LMIC, greater attention should be given to examining how ICT can be used during adolescence to address gender norms that lead to SGBV, taking into account the challenges in securing IRB approval for studies in which children are subjects. Ongoing efforts to study and pilot gaming apps and how they can lead to changes in gender norms related to SGBV are promising for preventing future SGBV. (BREAKAWAY Guatemala Pilot Study).

While no “clusters” of evidence emerged, several individual studies point to potential emerging areas of future research, including: (1) how ICT can contribute changing gender and social norms related to SGBV and primary prevention; (2) mobile phone applications that promote safety and security; (3) mobile technology for the collection and analysis of survivors' experience with SGBV response services (to inform the design of effective SGBV services); and (4) digital tools that support the collection of forensic evidence for SGBV response and secondary prevention.

The Decker, Wood, Kennedy, et al. (2020) and Decker, Wood, Hameeduddin, et al. (2020) protocol and study around safety decision apps demonstrate how an effective ICT intervention for preventing and of responding to SGBV in an HIC context can be adapted for an LMIC context. Similar efforts to first identity effective interventions in HIC contexts, and then adapt them to LMIC contexts could prove effective in preventing and or responding to SGBV in LMIC.

We also observed a lack of harmonized global impact indicators which further impedes evidence synthesis and overall generalization. Several of the systematic reviews commented on this lack of uniform outcomes and indicators for SGBV and cited this gap as the reason for not completing comprehensive meta-analysis. Future research is needed to understand how the complexity of SGBV may be uniformly measured across outcomes and indicators, especially within the context of ICT-based interventions for SGBV response and prevention. Part of this effort includes articulating, understanding, and testing a clear Theory of Change and theoretical basis with each intervention.

In conclusion, ICT-based interventions contributing to the response and prevention of SGBV in LMIC is an emerging field of study. While there is some very limited evidence showing promise for ICT-based interventions for SGBV prevention and/or response in LMIC, much more methodologically rigorous research is needed to identify how effective they are, and just what roles they can contribute to preventing and responding to SGBV. To address the gaps in the evidence and build towards a more cohesive evidence base, standard metrics are needed to evaluate the intermediary outcomes aligned to evidence-based frameworks such as RESPECT and INSPIRE. As more ICT interventions are used for prevention of, and/or response to SGBV, and with the use standard metrics to evaluate their impact, it is anticipated that clusters of evidence will emerge. There are clear examples of interventions that are designed using evidence-based frameworks that are well-positioned for evaluation. They are from higher-income settings that have been rigorously evaluated that are relevant for LMICs, namely safety decision apps and interventions focusing on changing harmful gender-norms that contribute to SGBV, particularly IPV. Future feasibility and adaptability (from HIC to LMIC) studies will also help researchers and policy-makers better understand best practice for adapting ICT- based interventions for a variety of contexts in LMIC. Such studies will also contribute to the scale-up and sustainability of interventions, issues that are commonly overlooked.