This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on https://www.researchprotocols.org, as well as this copyright and license information must be included.
Decentralized health systems in low- and middle-income countries (LMICs) affected by humanitarian crises lack resources and a qualified workforce to attend to the overwhelming demand for mental health care in emergencies. Innovative approaches that are safe, cost-effective, and scalable are needed to address the burden of traumatic stress caused by emergencies. High mobile phone ownership rates combined with the precision of neural, cognitive, and biometric measures of trauma and their feasible integration with artificial intelligence makes digital app interventions a promising pathway to promote precision diagnosis and high-impact care.
This study aimed to advance methods for the objective diagnosis and treatment of trauma in emergencies across LMICs by examining neural, cognitive, and biometric markers and the efficacy of the eResilience app, a neuroscience-informed mobile health mental health app intervention, via changes in clinical symptomatology, cognitive performance, and brain activity.
Trauma-exposed African refugees residing in Australia were selected for this study. A research software version of the eResilience app with advanced monitoring capabilities was designed for this trial. Participants completed the eResilience app at home during a 7-day period. Clinical, cognitive, and electrophysiological data were collected at baseline, along with posttest measurements to examine biomarkers of trauma and the efficacy of the proposed digital intervention for the treatment of trauma and its potential outcomes, including depression, anxiety, physical symptoms, self-harm, substance misuse, and cognitive impairment. In addition, biofeedback, well-being, and subjective stress data points were collected via the app during the treatment week, followed by clinical interviews at 1, 3, 6, and 12 months after the intervention.
Data collection was conducted between 2018 and 2020. A total of 100 participants exposed to war were screened; 75 (75%) were enrolled and assigned to a trauma-exposed control (38/75, 51%) or posttraumatic stress disorder condition (37/75, 49%); and 70 (70%) completed all baseline, treatment, and posttest assessments. A total of 89% (62/70) of those who completed the intervention opted to enroll in the 3-, 6-, and 12-month follow-ups. Data collection is complete. As of May 2022, the results of all proposed analyses are being prepared for publication. If proven efficacious, this proof-of-concept clinical trial will inform fully powered randomized clinical trials in LMICs to further develop artificial intelligence–powered, app-based diagnostic and prognostic features and determine the app’s cross-cultural efficacy for the treatment of trauma in emergency settings.
This protocol provides researchers with a comprehensive background of the study rationale, a detailed guideline for replication studies interested in examining the feasibility and efficacy of the eResilience app across varied demographics, and a robust framework for investigating low-cost objective diagnostic markers in mental health interventions. Methodological limitations and suggestions are also provided.
Australian New Zealand Clinical Trials Registry ACTRN12616001205426; https://tinyurl.com/yckwc4d7
RR1-10.2196/38223
Nations worldwide face the continuous challenge of addressing the complex and varied consequences of past, current, and evolving humanitarian emergencies. The United Nations Office for the Coordination of Humanitarian Affairs estimates that there are currently 234 million vulnerable people across 56 countries affected by war and conflict, climate change, hunger, and the COVID-19 pandemic [
Accelerated by the COVID-19 pandemic, investments in digital health are increasing markedly. In 2020 alone, >90,000 new health apps emerged in a market that already surpasses 350,000 apps, of which 22% account for mental health apps promoting the management of behavioral disorders via cognitive behavior therapy (CBT) strategies, mindfulness, meditation, stress and anxiety management, and sleep hygiene practices and monitoring [
Although not trauma-focused, an example of a successful digital adaptation of CBT-based care for humanitarian settings is Problem Management Plus (PM+), an initiative by the World Health Organization as part of the mental health Gap Action Programme [
Overall, only 2 studies on app-based interventions designed specifically for trauma in a humanitarian context were identified. A randomized clinical trial examined the efficacy of Sanadak, a CBT-based self-help trauma app in Arabic for Syrian refugees [
This study was focused on 2 primary objectives. First, it aimed to investigate novel neural and cognitive diagnostic and prognostic markers of clinical and subclinical PTSD with the potential for future integration in portable technology. We proposed that objective markers, when integrated into artificial intelligence, could aid lay humanitarian actors in the fast and accurate screening of individuals in need of care, guide best practices, assist with precision impact evaluation, and serve as predictive measures in prevention initiatives. Second, this proof-of-concept study introduced a trauma-focused digital mental health intervention, the eResilience app. Although the intervention’s clinical rationale presents a transdiagnostic approach targeting basic physiology and neurocircuitry affected by stress, hence potentially having clinical utility to address a range of mental health conditions, this study examined its feasibility as a primary treatment for clinical and subclinical PTSD. The examination outcome assessment process for the intervention was based on psychological, cognitive, and neurobiological systems via (1) electrophysiological activity during a state of rest, (2) neuropsychological testing, and (3) clinical symptomatology.
The eResilience app was proposed as a stand-alone intervention. The clinical curriculum is personalized for communities living under the poverty line, categorized by earning <US $5 per day. Moreover, it is low-intensity, without the involvement of specialists; it is trauma-focused, based on core components of complex trauma interventions including safety, self-regulation, self-reflective information processing, relational engagement, and positive affect enhancement [
In alignment with the core components of complex trauma interventions, the curriculum was based on three clinical aims established for the eResilience protocol: (1) the creation of a safe and personal therapeutic space to hone skills of autonomic nervous system (ANS) control and awareness, (2) the enhancement of core cognitive functions affected by trauma, and (3) building or restoring relational engagement. The clinical tools in the curriculum that aim to build ANS awareness and control are rooted in theoretical and biological processes in the sympathetic and parasympathetic branches of the nervous and somatosensory systems. The method includes (1) building body awareness skills, (2) training one’s ability to shift awareness between somatic sensations, and (3) identifying internal and external resources for top-down and bottom-up self-regulation, altogether aiming to promote a sense of safety that has been disrupted by trauma in the body and the environment. The clinical tools selected for building ANS awareness and control included breath work, progressive muscle relaxation, biofeedback, rhythm-focused exercises (music therapy, binaural beats, and bilateral tapping), yoga nidra, guided imagery, and grounding techniques. Therapeutic tasks primarily targeting cognitive function in the prefrontal cortex introduce the practice of exteroception skills, including engaging attention shifting from external surroundings, such as objects and sounds, and a short written task (which may also be adapted to accommodate the needs of individuals with low literacy levels) to identify internal and external resources for coping with adversity, as well as self-reflective information processing, positive affect, meaning making, and gratitude practice aimed at enhancing overall executive function capacities disrupted by trauma. For relational systems, selected tasks were informed by the biological principles and neural circuitry of social behavior and included the introduction of subjective and real-life safe interpersonal experiences, support system identification, building a sense of purpose in the community, and engagement in altruistic behavior, altogether aiming to promote environmental safety, interpersonal relatedness, planning, decision-making, initiation, and pleasure in the context of social interactions.
In total, 3 software versions of the eResilience app have been created. The first prototype, also known as a minimum viable product, along with its first field version designed for groups, was created for and piloted by an international charity in West Africa. For this clinical trial, the eResilience research version 1.0 (
On the basis of the evidence presented for anomalies in electrophysiological and cognitive performance in PTSD, this study questioned whether resting-state electrophysiological activity and neuropsychological performance were reliable objective measures for diagnosing clinical and subclinical PTSD and predicting intervention outcomes. It was hypothesized that, at baseline, (1) the PTSD cohort would present statistically significant differences in quantitative electroencephalograms (EEGs) compared with the controls and that (2) the PTSD group would perform more poorly than the control participants on the cognitive tests. Moreover, in questioning the feasibility of the proposed treatment for clinical and subclinical PTSD, we hypothesized that, at posttest measurements, the intervention would (1) reduce PTSD severity according to the Clinician-Administered PTSD Scale for the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5; CAPS-5), Past-Month Edition, between baseline and posttest measurements (3-, 6-, and 12-month follow-up time points) and (2) improve participant performance in cognitive measures.
The study was approved by the Human Research Ethics Committees of the University of Sydney and the University of the Sunshine Coast, Australia. This trial was conducted in compliance with the ethics committee approval conditions, the National Health and Medical Research Council Statement on Ethical Conduct in Human Research (2007), and the Note for Guidance on Good Clinical Practice (CPMP/ICH-135/95).
This proof-of-concept study was designed as a nonrandomized controlled trial, including 10 time points across the 2 study phases. The first phase involved a 10- to 11-day commitment, comprising intake (time point 1), 1 or 2 baseline home visits to complete clinical interviews (time point 2, within 2 weeks of time point 1), a baseline university laboratory visit (time point 3, after 7 days of time point 2), the eResilience app intervention (time point 4, for 7 consecutive days starting the day of time point 3 upon laboratory data collection completion), a posttest university laboratory visit (time point 5, the day after time point 4 completion), and the 1-month follow-up (time point 6, after 30 days of the time point 5 laboratory visit). The second phase of the study included a new informed consent (time point 7, within 2 months of completing time point 6) to collect 3-, 6-, and 12-month follow-ups (time points 8, 9, and 10). The interventions at each time point are summarized in
Study design. CANTAB: Cambridge Neuropsychological Test Automated Battery; CAPS-5: Clinician-Administered Posttraumatic Stress Disorder Scale for the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; EEG: electroencephalogram; HTQ-4: Harvard Trauma Questionnaire-4; PTSD: posttraumatic stress disorder; SCID-5-RV: Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Research Version; SUDS: Subjective Units of Distress Scale; TBI: traumatic brain injury.
Quantitative EEG signals were collected in a sound-attenuated room during a resting state with 4-minute eyes-open and 4-minute eyes-closed paradigms and a 1- to 5-minute break between recordings to check and fix impedance as needed. The participants were instructed to fixate on a dot in the center of a television monitor during the eyes-open collection and rest quietly during the eyes-closed paradigm.
The participants followed instructions to not consume alcohol, caffeinated drinks, or nicotine on the day of the quantitative EEG recordings. Moreover, they were instructed to wash their hair with only shampoo the evening before the laboratory visits and avoid conditioners, gels, oils, and hair spray to increase electrode adherence to the scalp.
Data were acquired at a sampling interval of 4000 µS and sampling rate of 250 Hz using the Electrical Geodesics’ (EGI) Geodesic EEG System 400 [
Several diagnostic clinical interviews as well as psychosocial and self-report questionnaires were selected for this study. All clinical interviews were conducted by a masters-level clinician with 12 years of experience conducting trauma assessments with African refugee communities (JVP) and a trained clinical research assistant with experience collecting and evaluating trauma assessments of 400 refugees across Uganda (Elsa Goninon). Challenging diagnostic decisions were discussed among senior researchers in the team (CH and BOT).
The CAPS-5 was chosen as a primary outcome measure for PTSD, possessing excellent convergent and discriminant validity (κ=0.84) [
Additional clinical interviews included items 1 to 6 of the Peritraumatic Dissociative Experience Questionnaire [
Finally, 3 instruments were created for this study. The Demographics for Africans in Australia (
The Cambridge Neuropsychological Test Automated Battery [
At time point 3, the participants were individually taken through a 30-minute Microsoft PowerPoint introduction to the eResilience app. The session included handling a take-home kit containing printed safety guidelines, a smartphone with software, soundproof headphones, biofeedback equipment for temperature measurements, and the printed app response booklet (
The eResilience app research version (
Once the intervention was launched, the clinical tasks were presented via visual and audio instructions recorded in a pitch-controlled, text-to-speech software. Advanced features to promote participant safety included well-being, body discomfort, and stress checks added at the start and end of the day and in between blocks of exercises, automating SMS text messages that were sent to researchers on standby whenever a participant reported elevated levels of distress during the intervention. As part of the study design, the software also presented integration for external biofeedback equipment for temperature check-ins as part of the objective stress measurements. The data set collected via software during the intervention week included time-tracking features to register precise individual participation (eg, when a clinical task had been paused, skipped, or not played). Data were secured and accessed via a password-protected researcher account.
Participants were recruited in collaboration with the following local African community organizations in the state of Queensland: the
Participants were eligible if they were Liberian, Congolese, or Sudanese refugees who had migrated to Australia after the age of 18 years. All participants were required to have English proficiency and have fled emergency in Africa between 1989 and 2018. The participants were divided into two groups: a PTSD group, including clinical and subclinical PTSD cases according to the diagnostic specifications outlined in
Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) threshold: Clinician-Administered PTSD Scale for the DSM-5 (CAPS-5) criteria A+B+C+D+E+F+G
DSM-5 subthreshold: CAPS-5 criteria A+(at least one threshold symptom at B, C, D, and E)+F+G
DSM-5 subsyndromal: CAPS-5 criteria A+B+(meets criteria for cluster C or D or E or any 2 of the 3 clusters, but not C+D+E combined)+F+G
DSM-5 other stress or trauma-related disorders: CAPS-5 criteria A+F+G (and no overlap with DSM-5 Subsyndromal cases)
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) threshold: Clinician-Administered PTSD Scale for the DSM-IV (CAPS-4) criteria A+B+C+D+E+F
DSM-IV subthreshold: CAPS-IV criteria A+(at least one threshold symptom at B, C, and D)+E+F
DSM-IV subsyndromal: CAPS-4 criteria A+B+(C or D)+E+F
DSM-IV Harvard Trauma Questionnaire (HTQ) risk screen: HTQ mean score of items 1 to 16 ≥2.5 (symptomatic for PTSD)
DSM-IV threshold: CAPS-IV criteria A+B+C+D+E+F
DSM-IV subthreshold: CAPS-4 criteria A+(at least one threshold symptom at B, C, and D)+E+F
Considering the reported limitations of the Diagnostic and Statistical Manual of Mental Disorders in identifying trauma among refugee populations [
Exclusion criteria for both groups included a self-reported severe medical condition, genetic disorder, or concurrent drug or alcohol abuse or dependency within the month before the intervention trial.
Additional exclusion criteria for the PTSD group included current use of psychotropic medication or use within the 2 months preceding the intervention, concurrent psychotherapy for PTSD, and acute risk of suicide or homicide.
Considering the various levels of trauma symptomatology proposed for the PTSD group, additional exclusion criteria for the trauma-exposed control group included not meeting the threshold, subthreshold, subsyndromal, or other stress- or trauma-related disorders criteria on either the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), or the DSM-5 as well as not meeting PTSD risk or symptom criteria on the HTQ.
All participants enrolled provided oral and written consent and signed a participant consent form at time point 2 (
Participants were contacted 30 days after completion of the 1-month posttest measurement, invited to participate in the study phase 2—involving 3-, 6-, and 12-month follow-up interviews—and given a copy in person or via email of the second participant information statement (
Study measures will be compared between the PTSD and control group participants. Descriptive statistics and 2-tailed
All time-domain quantitative EEG recordings will be plotted into a frequency power spectrum using fast Fourier transform. The grand average for each segment of data for the control and experimental conditions will be calculated at pre- and posttest measurements across both paradigms to generate topographical distributions of spectral power values in the delta, theta, alpha, peak alpha, beta, peak beta, and gamma frequency bands. The results will be presented in the form of power values (uV2).
On the basis of the respective data distribution, McNemar nonparametric tests,
The proposed sample size of 60 for this study (PTSD: 30/60, 50%; controls: 30/60, 50%) was based on the effect size computation of a quantitative EEG measure previously used to investigate the dimensional complexity of the EEG between the PTSD and control groups [
For the recruitment results (
In the first phase of baseline assessments, time point 2, a total of 80 participants consented and were fully assessed, and 75 (94%) chose to enroll in the intervention. All enrolled participants (75/75, 100%) completed the second baseline phase (time point 3) and took the eResilience app home to complete the intervention during the 7-day period (time point 4). In total, 7% (5/75) of the participants withdrew consent before commencing the intervention because of lack of time (4/5, 80%) and a disclosure of alcohol abuse (1/5, 20%); thus, their data will be removed from the analyses. All 70 participants who completed the intervention returned to the laboratory for postintervention assessments (time point 5) and completed the 1-month follow-up assessment (time point 6) via telephone. Upon completion of the 1-month follow-up, the participants were contacted and extended an invitation to participate in the 3-, 6-, and 12-month follow-ups for the study. In total, 89% (62/70) of the participants who completed the trial consented and opted to enroll again (time point 7). Of the 8 participants who did not join the follow-ups, 2 (25%) were part of the PTSD group and opted out without disclosing reasons, and 6 (75%) were unable to be reached because of changes in phone numbers, addresses, or email (PTSD: 2/6, 33%; controls: 4/6, 67%). All participants enrolled in the follow-ups (62/62, 100%) completed the interviews at the 3 time points (time point 8, time point 9, and time point 10). In summary, the final data sample includes 70 participants at baseline, 1-week, and 1-month posttest analyses and 62 participants in the 3-, 6-, and 12-month follow-ups.
Study recruitment results flowchart.
The final sample that completed the treatment phase (n=70) included 44% (31/70) men and 56% (39/70) women aged 18 to 54 years (mean 33.64, SD 10.54 years). The participants were Congolese (Democratic Republic of the Congo; 23/70, 33%), Liberian (22/70, 31%), and Sudanese (25/70, 36%) and altogether represented 26 African ethnic groups. The PTSD cohort (35/70, 50%) included 69% (24/35) of participants who met the full DSM-5 PTSD diagnostic criteria based on the CAPS-5, 14% (5/35) who met the DSM-5 subthreshold criteria, and 17% (6/35) who met the DSM-5 subsyndromal criteria, as detailed in
The participants were asked to complete the GPS on their own at the end of the home visit. Nevertheless, upon noticing that several participants mistakenly responded to the GPS based on lifetime experiences, the researchers conducted a second GPS interview within 24 hours to ensure that all answers corresponded to the symptoms experienced within the previous 30 days.
None of the participants requested referral for the continuation of mental health care. Many participants expressed an interest in downloading the app to continue the intervention tasks when it became available.
No adverse events were reported during the study.
Additional results for all analyses proposed are being prepared for publication as of May 2022.
We anticipate that this proof-of-concept trial will provide evidence of the preliminary efficacy of the proposed intervention in treating trauma and will identify novel cognitive and electrophysiological diagnostic and prognostic markers of clinical and subclinical PTSD. Although the results of this trial are being prepared for publication, the enrollment outcomes were presented in this protocol. On the basis of the sample size estimation, the aim of this study was to recruit 60 participants. However, refugee community members’ interest in joining the trial exceeded the study capacity. Moreover, the dropout rates were unusually low for PTSD clinical trials, with only 3% (1/35) of the PTSD participants dropping out before treatment initiation and none dropping out during or after treatment. In contrast, other intervention studies have indicated dropout rates >20% [
This study design presents strengths and limitations. First, it is essential to acknowledge that our healthy control group, although not presenting PTSD or other psychiatric disorders, was exposed to similar war and conflict-related traumatic events as the PTSD group. The recruitment of nonexposed controls, although most suitable, was not possible considering that all 3 African refugee communities involved in this study migrated to Australia because they were fleeing conflict. Moreover, between both groups, traumatic brain injury, medication washout, and other medical conditions were only self-reported via the approved clinical interviews. Thus, it is possible that trauma exposure among controls, along with brain injury, unreported use of medications, or other undetected medical conditions between both groups, could affect response to the intervention, cognitive test scores, and quantitative EEG recordings. Particularly in the PTSD group, quantitative EEG recordings can also be affected by the dysregulation of sleep often present in traumatic stress; therefore, this must be considered during the analyses.
Although one of the strengths of this study is the combination of objective and subjective diagnostic and prognostic measures, owing to restricted resources, the follow-up phase did not include the collection of quantitative EEG, biometric data, or cognitive tests. In addition, although the proposed design does not include randomization of participants and a no-intervention group as a proof of concept, this feasibility study serves a vital role in subjectively and objectively examining the preliminary efficacy of the eResilience app before directing resources toward large randomized controlled trials. It is also important to note that, although the proposed app intervention was designed for cross-cultural use, the sample recruited for this study, primarily represents African cohorts exposed to war and conflict-related trauma. Finally, the COVID-19 pandemic coincided with the 12-month follow-up mark, potentially causing additional stress after the intervention.
To date, apps are unlikely to address all complexities of mental health disorders and, thus far, evidence supports that they are best used to supplement care rather than as first-line interventions. The development of the eResilience app as a high-impact intervention and the proposed study design are challenged by the fact that the populations for which this intervention is designed are likely never to have access to adequate professional help in their lifetime, and an app such as the one proposed may be the only help they will ever access. Moreover, recognizing that PTSD is not the sole mental health concern among our target populations and that the eResilience app should aim toward a transdiagnostic approach, this proposed study lacks the power and resources to examine outcomes across diverse domains. Another considerable challenge relates to designing a tool suitable for large-scale dissemination while not compromising safety and quality of care.
Notably, one of the promising strengths of this intervention design for individual or group use is the cost efficacy projected to be relatively lower than existing programs. In addition, the simplicity and safety of the designed clinical tools aim to eliminate the need for training and supervision of local lay staff. Finally, the week-long time frame would also further reduce costs in extensive community outreach efforts, thus boosting its scalability potential.
If successful, this trial will provide a foundation for large-scale, cross-cultural studies to further examine the efficacy of the proposed app intervention and biomarker precision as well as to better understand how brain data may inform other cost-effective biometric measures for software integration, serve as a digital aid for the precision diagnosis of PTSD and related disorders, and measure treatment response.
The research software version used in this study is available upon request to all external researchers interested in conducting future trials. Recommendations for future designs include (1) the inclusion of healthy controls that have not been exposed to traumatic stress, particularly as it pertains to the identification of diagnostic markers; (2) the inclusion of objective measures to examine physiological regulation skills at long-term follow-ups; (3) the use of smartphones with sensor technology for biometric data collection, including heart rate variability; (4) the investigation of the efficacy of the app for other disorders as primary outcome measures, such as depression and generalized anxiety; and (5) the expansion of the sample diversity to include non-African communities and demographics affected by other types of humanitarian emergencies besides war and conflict. Finally, although the design and results of this trial may be premature in assessing the cost-efficacy and scalability qualities of the app, we also strongly recommend that those qualities be considered in future study protocols. Further considerations will be discussed alongside the publication of the study results.
This study protocol introduces a novel digital app for the treatment of posttraumatic stress in emergencies. Its nonreliance on the DSM-5 constructs of PTSD and greater focus on the neural substrates of traumatic stress as an attempt to target the cause of symptoms may increase the likelihood of cross-cultural adaptability, promoting long-term sustainable change on a neurobiological basis rather than via the reduction and ongoing maintenance of symptoms. The potential identification of diagnostic markers could yield further advances for the humanitarian sector, contributing toward the cost-effectiveness of large-scale programs by making the best use of limited resources when identifying persons in need and objectively monitoring impact via automated health data feedback. Overall, we conclude that the successful enrollment and low dropout results presented indicate a strong interest from the Liberian, Sudanese, and Congolese refugee communities in Australia to be involved in research to contribute toward the development of novel approaches to address global mental health challenges caused by war and adversity.
eResilience app research version 1.0.
Demographics for Africans in Australia.
Chronological trauma exposure chart.
App survey for 12-months post-test assessment.
eResilience app response booklet.
Participant information statement.
Participant consent form.
Participant Information Statement - Phase 2.
Participant Consent Form Phase 2.
autonomic nervous system
Clinician-Administered Posttraumatic Stress Disorder Scale for the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition
cognitive behavioral therapy
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition
Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition
electroencephalogram
Electrical Geodesics
Global Psychotrauma Screen
Harvard Trauma Questionnaire
Problem Management Plus
posttraumatic stress disorder
This study was self-funded. The authors acknowledge the research staff (Charlotte Jones, Flavia Montoni, Rafael Bras, Warwick Smith, Elsa Goninon, Naomi Irvine, Victoire M Parker, Kiarra Lynch, Tahlia Wyer, Jayme-Leigh Wilson, Patricia Otto, and Emilie Cullum) and the following refugee-led organizations that assisted in recruitment efforts: the Congolese United for Peace and Reconciliation in Australia, Together We are Powerful, the Federation of Liberian Communities in Australia, and the South Sudanese Youth Council.
The first author JVP contributed to study conceptualization, design, and manuscript writing. The second and third authors CH and BOT provided expert supervision and revision. All authors have read and approved the final manuscript.
None declared.