This study examined the content and feasibility of remote WhatsApp text supervision conducted as part of an open clinical trial in Somaliland.

Islamic Trauma Healing ITH) is a brief, group, lay-lead, trauma-focused, mosque-based intervention that has demonstrated initial efficacy in pilot studies in the United States and Somaliland.

By Alexandra B. Klein, Mumin H. Egeh, Alexandra R. Bowling, Ash Holloway, Ayaan Abdillahi Ali, Zeinab Adam Abdillahi, Mohamed Ahmed Abdi, Salma Hassan Ibrahim, Khadar Hindi Bootan, Hibaaq Isse Ibrahim, Aden Mohamed Ali, Abdirahman Muse Tubeec, Michael L. Dolezal, Dega A. Angula, Jacob A. Bentley, Norah C. Feeny, and Lori A. Zoellner

WhatsApp Supervision For A Lay-Led Islamic Trauma-Focused Intervention In Somaliland: Qualitative Content Analysis


Alexandra B. Klein

Corresponding Author

Department of Psychological Sciences, Case Western Reserve University, Cleveland, Ohio, USA


Concerning this article should be addressed to Alexandra B. Klein, M.A. Case Western Reserve University, Department of Psychological Sciences, 1220 Bellflower Rd. Cleveland, OH 44106. Email:

Mumin H. Egeh

Abu Bakr Al-Siddique Islamic Center, Borama, Somaliland

Alexandra R. Bowling

Department of Psychological Sciences, Case Western Reserve University, Cleveland, Ohio, USA

Ash Holloway

Department of Psychology, University of Washington, Seattle, Washington, USA

Ayaan Abdillahi Ali

Ma’alin Haruon Masjid, Hargeisa, Somaliland

Zeinab Adam Abdillahi

Ma’alin Haruon Masjid, Hargeisa, Somaliland

Mohamed Ahmed Abdi

Ma’alin Haruon Masjid, Hargeisa, Somaliland

Salma Hassan Ibrahim

Abu Bakr Al-Siddique Islamic Center, Borama, Somaliland

Khadar Hindi Bootan

Ma’alin Haruon Masjid, Hargeisa, Somaliland

Hibaaq Isse Ibrahim

Abu Bakr Al-Siddique Islamic Center, Borama, Somaliland

Aden Mohamed Ali

Abu Bakr Al-Siddique Islamic Center, Borama, Somaliland

Abdirahman Muse Tubeec

Abu Bakr Al-Siddique Islamic Center, Borama, Somaliland

Michael L. Dolezal

School of Psychology, Family, and Community, Seattle Pacific University, Seattle, Washington, USA

Dega A. Angula

Abu Bakr Al-Siddique Islamic Center, Borama, Somaliland

Jacob A. Bentley

School of Psychology, Family, and Community, Seattle Pacific University, Seattle, Washington, USA

Norah C. Feeny

Department of Psychological Sciences, Case Western Reserve University, Cleveland, Ohio, USA

Lori A. Zoellner

Department of Psychology, University of Washington, Seattle, Washington, USA

First published: 06 October 2022|

Journal of Traumatic Stress 2022


Early View

Online Version of Record before inclusion in an issue

The authors would like to thank the Somaliland Ministry of National Planning and National Development and the Somaliland Ministry of Endowment and Religion Affairs who provided approval to complete this work in Somaliland. We would also like to thank Ma’alin Haruon Masjid, Hargeisa, Somaliland; and Abu-Bakar Al-Siddique Islamic Center, Borama, Somaliland, who provided support and space for the running of the Islamic Trauma Healing groups. The authorship order of group leaders at the two mosques is alphabetical.



Clinical supervision is critical for the uptake of psychotherapy but difficult to facilitate in countries with limited providers, resources, and internet infrastructure. Innovative supervision approaches are needed to increase access to mental health treatments in low-to-middle income countries (LMICs). This study examined the content and feasibility of remote WhatsApp text supervision conducted as part of an open clinical trial in Somaliland. Islamic Trauma Healing ITH) is a brief, group, lay-lead, trauma-focused, mosque-based intervention that has demonstrated initial efficacy in pilot studies in the United States and Somaliland. After a 2-day, in-person training, lay leaders led four groups of five to seven members focused on trauma-related psychopathology and community reconciliation. Somali lay leaders trained in ITH (n = 9) and the research team (n = 6) attended weekly WhatsApp supervision during the intervention. Content was logged and subjected to qualitative analysis by two coders. Comments related to intervention implementation indicated that lay leaders understood the treatment rationale, adhered to treatment procedures, and believed the intervention components to be helpful and culturally relevant. Themes related to engagement suggested perfect attendance across groups and high levels of participation. Lay leader psychoeducation and skill development; supervisor praise, support, and encouragement; and supervisee gratitude emerged as additional themes. Remote text supervision conducted via WhatsApp was technologically feasible and may have facilitated skill development and the effective implementation of this lay-led intervention. When tailored to the local context, remote supervision approaches hold promise for increasing access to services in LMICs with limited resources.

Rates of trauma exposure are elevated in historically war-torn regions and active conflict regions like Somalia (Gerritsen et al., 2006; Karam et al., 2014). As a result, individuals from these countries are at an increased risk of developing posttraumatic stress disorder (PTSD), which has an estimated prevalence of 4%–62% in these areas (Gerritsen et al., 2006; Onyut et al., 2009), and other trauma-related difficulties (Dietrich et al., 2019; Koenen et al., 2017). Indeed, despite a lack of epidemiological data on mental health, it is estimated that nearly all Somali adults have experienced at least one potentially traumatic event (Cavallera et al., 2016). Although empirically supported treatments (ESTs) for PTSD exist, the use of ESTs is limited in lower- and middle-income countries (LMICs) by factors such as a lack of providers and mental health infrastructure, the stigma surrounding mental illness, and the potential perceived incongruence with one’s faith (e.g., Bentley & Owens, 2008; Piwowarczyk et al., 2014; World Health Organization [WHO], 2010). As such, culturally relevant, trauma-informed mental health interventions are needed within war-torn regions to address the mental health impact of trauma exposure and community fragmentation.
Training in culturally relevant ESTs and ongoing supervision are critical for effective implementation but remain a challenge in places where there is a limited mental health workforce. The delivery of ESTs in LMICs has been shown to be successful through train-the-trainer (TTT) implementation models (e.g., Neuner et al., 2008; Rahman et al., 2019) whereby tasks are shifted from trained professionals to local members of the community. These models are cost-effective, self-sustaining, and scalable and can facilitate the widespread uptake of interventions (Jacob et al., 2014).
Previous investigations have demonstrated that brief, one-time training in mental health interventions with little subsequent supervision may increase intervention knowledge but does not necessarily increase therapist adherence, competence, or skill (Beidas & Kendall, 2010; Herschell et al., 2010). Yet, intensive training and hour-for-hour supervision are often not practical in many settings, especially in high-conflict zones, such as Somalia. Continued supervision from mental health professionals following training is critical for providing ongoing support and improving intervention skills (Baingana & Mangen, 2011; McLean et al., 2015; Saxena et al., 2007), particularly for providers who do not have formal mental health training. Importantly, a supportive approach to supervision that emphasizes strengthening relationships, attending to potential power dynamics, promoting open communication, and proactively identifying and solving problems (e.g., Kemp et al., 2019) is considered the best practice for task shifting in global mental health contexts (WHO et al., 2007). However, in LMICs, ongoing supervision can be difficult to sustain without the physical presence of professionals, which is costly and often impractical. Several studies have supported the efficacy of conducting supervision remotely utilizing technology (e.g., Mahmuda et al., 2019; Murray et al., 2014).
For example, a randomized controlled trial comparing the efficacy of supervision methods in improving therapist competence found remote supervision via the Skype platform to be equally effective as face-to-face supervision (Rahman et al., 2019). However, many of the common approaches to remote supervision rely on a stable internet connection (Khan et al., 2017), which may not translate in countries with limited infrastructure. For example, Magidson and colleagues (2015) shifted their supervision model to email following internet connection difficulties when attempting to contact local providers in Iraq via Skype. Additional research on innovative, cost-effective approaches to remote supervision, including the use of text messaging, is needed to address systemic barriers.Utilizing applications that rely on a cellular connection, a service more broadly available than internet access in LMICs, for remote supervision may be a low-cost, practical way to circumvent technological difficulties. In Somalia in particular, approximately 12.1% of the population uses the internet, whereas 45.3% have mobile cellular devices (Kemp, 2021). The present study sought to examine the initial feasibility and accessibility of conducting remote clinical supervision via WhatsApp text messages in an open trial of Islamic Trauma Healing (ITH) conducted in Somaliland, Somalia. ITH is a lay-led, trauma-focused, mosque-based group intervention that integrates evidence-based components of existing trauma-focused interventions (e.g., psychoeducation, imaginal exposure, cognitive restructuring) with core Islamic tenants (e.g., prophet narratives and turning to Allah in du’a [i.e., the time spent in informal individual prayer]; Lang et al., 2016) over six sessions. The intervention has demonstrated initial efficacy and acceptability in small open trials among Somali refugees with trauma-related symptoms in the United States (Zoellner et al., 2018) and individuals in Somaliland, Somalia (Zoellner et al., 2021). In this study, we qualitatively analyzed content from weekly WhatsApp text supervision sessions following each ITH group meeting. Categories and subcategories were developed separately for lay-leader and supervisor responses to understand the quality, focus, and feasibility of clinical supervision from both the supervisor and supervisee perspectives and to inform the optimal implementation of text-based supervision modalities.



The Somaliland team (n = 9) included four women and four men lay leaders and the program director. Lay leaders were of Somali background and the Islamic faith and originated from Hargeisa and Borama in Somaliland. Male and female group lay leaders were selected by the second author based on their status as respected community leaders within their Islamic centers, having a “heart for healing and reconciliation” (i.e., a person who sees the effects of trauma exposure in their community and wants to promote healing and reconciliation) in the Somali community, knowledge about the Qur’an, and a willingness to train additional group leaders in the future. None of these individuals had prior mental health training. Lay leaders spoke and read Somali, with some English-speaking skills. Leaders included Imams, Sheikhs, and teachers at the madrasa (i.e., educational institution).

Supervisors (n = 3 women, n = 2 men) included three licensed clinical psychologists, who worked in the development of the intervention, and two doctoral students in clinical psychology. All supervisors were White and English-speaking. Weekly clinical supervision occurred in the context of a pre–post feasibility clinical trial (Zoellner et al., 2021). Both lay leaders and clinical supervisors were affiliated investigators in the trial. All study procedures were approved by the University of Washington Institutional Review Board, and all study participants in the parent study gave informed consent to participate.


Overview of the intervention: ITH

ITH is a six-session lay-led, mosque-based, group intervention that addresses the psychological effects of war and refugee trauma and community reconciliation. The program was designed for separate-gender groups of five to seven individuals led by two group leaders of the same gender as the group’s participants. The program is a deep cultural integration of evidence-based components of trauma-focused cognitive behavioral psychotherapy for PTSD with tenets of the Islamic faith to target trauma-related avoidance and unhelpful negative beliefs about one’s self, others, and the world. Key components include prophet narratives and discussion, which focused on promoting cognitive restructuring; and turning to Allah in du’a (i.e., informal prayer) about the traumatic event, which mirrored imaginal exposure revisiting of trauma memories through approaching, encouraging engagement with, and processing trauma-related memories. Each session lasts approximately 90–120 min and includes time for community-building rituals (e.g., sharing tea and snacks); spiritual preparation, using a brief supplication written by an Imam; a prophet narrative and subsequent discussion; and a brief closing supplication. Session 1 includes program rationale, psychoeducation regarding common reactions to trauma, a prophet narrative and discussion, and breathing relaxation training. In Sessions 2–6, the focus is on a new prophet narrative at each session and individual time turning to Allah about the trauma in du’a. In the final session, group members are encouraged to organize a closing event, where certificates of program completion are given out. Session content is thematically focused and shifts from addressing the presence and purpose of suffering to healing and reconciliation for one’s self, others, and the larger community. The intervention manual does the “heavy lifting” regarding therapeutic work and delineates clear instructions to assist group leaders with progressing through the content in each session. Session-by-session content is provided within the manual. ITH development and components are detailed elsewhere (Bentley et al., 2021; Zoellner et al., 2018).

ITH intervention training

Lay leaders attended a 2-day training that lasted a total of approximately 8 hr in Djibouti, Djibouti, led by the supervision team (Zoellner et al., 2021). The training for ITH is purposefully brief to reduce leaders’ time burden and promote easy scaling of the intervention. During the training, leaders were given an overview of the program, learned about and role-played group sessions, and reviewed how to approach potential issues that may arise during sessions (e.g., distress upon revisiting a trauma memory). The training focused on teaching leaders how to effectively lead discussions rather than explicitly training them in delivering therapy. Specifically, lay leaders were trained to read sections of the manual to the group and trained in group discussion and facilitation skills (e.g., reflecting back, asking for more information). Training on confidentiality and common ethical considerations in facilitating a group were also provided. The training was provided in English by the research team and translated into Somali by leaders fluent in both English and Somali.

ITH intervention supervision

Lay leaders based in Somaliland were supervised by clinical psychologists located in the United States. Supervision was conducted via WhatsApp, a free worldwide messaging application that uses a phone’s cellular data or Wi-Fi connection rather than relying on cellular minutes or a text plan. After each of the six group intervention sessions, lay leaders used WhatsApp to discuss the session via text messaging with the United States–based research team. See Figure 1 for an example of dialogue between lay leaders and supervisor. These session discussions were translated into Somali in real time on an as-needed basis. Translation occurred in a separate simultaneous group conversation and was facilitated by the program director based in Somaliland, a long-standing member of the research team who demonstrated a strong understanding of the key components of the intervention and provided considerations for successful implementation in the Somaliland context. Multiple members of the lay-leader team were fluent in both English and Somali and assisted when there were translation questions.

Details are in the caption following the image


Example of WhatsApp supervision conversation between a supervisor and lay leader

Supervision sessions occurred in a group format (i.e., including supervisors and lay leaders) weekly for 6 weeks, with supervision occurring for two men’s groups and two women’s groups, each with two same-sex group leaders, in two different cities. Supervision sessions lasted approximately 60 min. Although the team was available for additional supervision if necessary, supervision sessions stayed within the 60-min time frame. Lay leaders were encouraged to connect with supervisors between supervision sessions if questions or implementation issues arose; however, this did not occur.

Supervision sessions aimed to ensure fidelity to the program, emphasize the roles of group leaders as supportive discussion leaders and listeners rather than lecturers, and assist lay leaders in identifying maladaptive trauma-related avoidance and beliefs. Supervision sessions followed a structured format, beginning with an attendance check-in to ensure that all lay leaders were present and ask about lay leaders’ well-being. Supervisors then asked preplanned questions at each session related to skill development, intervention content, and implementation. These questions were meant to facilitate discussion and allow for bilateral feedback and program-related problem-solving as needed. These questions were initially focused on developing discussion-leading skills (e.g., asking open-ended questions, utilizing supportive listening) and shifted to identifying and addressing maladaptive trauma-related avoidance and beliefs. Examples of questions asked of lay leaders during supervision included, “Did it make sense to the group that there are wounds from war/violence that impact how you think/feel just like physical ones/?”, “Does anyone have an example of negative thoughts/avoidance from your group?”,“Last week we talked briefly about how trauma changes how we think about self, others, world. Did you listen for these?”, “How did you help group member try to see from another view/perspective?”, and “Any questions or issues with group members?” Following the predesigned questions, there was allotted time for lay leaders to ask additional questions and bring up any concerns. Supervision sessions ended by reviewing the content for the next session. Confidential group participant information was not included in the texts.

Data analysis

All texts from the WhatsApp group used for supervision were extracted to one data file, with each text message shown as a separate response. Given the novelty of this work and a lack of previously identified code categories, we utilized a conventional qualitative analysis approach whereby categories and subcategories were developed based on the text itself (Hsieh & Shannon, 2005). Initially, three coders independently reviewed the text and developed broad categories to capture general response themes. Two of the coders were research assistants with no involvement in the study procedures, data collection, or supervision. These individuals were involved with another trial of ITH in the right to statehood would be significant. The UK government has always taken a timid approach under United States and, thus, had conceptual knowledge of the intervention. The final coder was a doctoral student and the first author who was involved in study procedures and participated in category development to provide context and resolve discrepancies. This individual had extensive experience administering trauma-focused cognitive behavior therapy (CBT), the basis for ITH, as a supervisee and had provided supervised consultation of trauma-focused CBT. Categories were retained if at least two of the three coders identified them in the text. Categories that were identified in only one coder’s responses were considered and retained if deemed by the third coder to be conceptually appropriate.

Broad categories were then double-coded by the first two independent coders. The coders then developed and coded narrow subcategories within each broad category using the same method. Weekly meetings were held to discuss discrepancies in coding and revise, add, and remove categories as necessary until a consensus was reached under the supervision of the first author. The categories were additionally reviewed by the last author and principal investigator to ensure conceptual accuracy. When responses reflected themes from multiple categories, simultaneous coding was utilized (Ritchie & Spencer, 1994), meaning that responses could be coded into more than one category. Lay-leader and supervisor responses were coded and analyzed separately. If a response was unclear, the responses surrounding it were used to help inform the context of the response in the greater discussion. To quantify the data, we tallied the number and percent of all responses that fell under each category and subcategory.


In total, 521 responses, including 352 unique lay-leader and 169 unique supervisor responses were coded. Responses without meaningful content (e.g., “yes,” “sounds good”) and those that captured greeting and farewell statements at the beginning and end of each session were excluded and not subjected to additional coding (lay leader: n = 138, 39.2%; supervisor: n = 59 34.9%). All quotations pulled from texts are printed verbatim and may include errors in syntax and grammar.

Lay-leader text behavior qualitative domains

Lay-leader text responses fell within six domains: group procedures, intervention-specific components (e.g., du’a and prophet narratives), gratitude, faith in Allah, technical issues, and miscellaneous. Responses coded as miscellaneous did not fit into any other domain but still contained meaningful responses (e.g., “We need to change into pos[i]tive way!”). Further, some miscellaneous responses were unclear, potentially due to language issues (e.g., “It’s rea[l]ly good ethicis and moral also first.”). Table 1 provides the number and percentages of lay-leader responses for each domain and subdomain and the percentages of subdomain responses that fell within each broader domain.

Lay-leader domains and subdomains during WhatsApp supervision
Leader domain and subdomain(s) Responses (n) Lay-leader responses (%) Domain responses (%)
Group procedures 64 18.2
In-session engagement/participant feedback 38 10.8 59.4
Self-confidence and skill development 13 3.7 20.3
Fidelity and manual helpfulness 10 2.8 15.6
Intervention-specific components (e.g., du’a and prophet narratives) 61 17.3
Effective implementation 27 7.7 44.3
Relevancy, import, and helpfulness 24 6.8 39.3
Gratitude 55 15.6
Technical issues 11 3.1
Faith in Allah 11 3.1
Miscellaneous 25 7.1
Excluded 138 39.2
  • Note. N = 352. Responses could be coded in multiple domains and subdomains; percentages may total to over 100.0%.

Group procedures domain

Lay-leader discussion of group procedures emerged in 64 (18.2%) responses. Of these responses, a subdomain related to in-session engagement and participant feedback (n = 38, 59.4% of domain responses) suggested that group members attended sessions, actively participated, and found intervention components helpful and culturally relevant. For example, one lay leader stated, “And also this sessions built the faith of the members and whenever we ask their feeling the[y] told us that sessions [a]re helpful.” Another said, “The most things Group B liked when we asked the majority was individual returning to Allah [referring to du’a component].”

In a second subdomain theme, lay leaders discussed an increase in self-confidence and skill development (n = 13, 20.3% of domain responses) over the course of the intervention. Following Session 3, one lay leader stated, “Absolutely we half done and yes we are getting use to be a leaders, [a]nd what is most be[a]utiful and like most is the discussion scenes and how to discuss [a]nd deeply narrow questions.” By Session 6, another lay leader commented, “We already know how to lead an[d] be a leader.” Finally, lay leaders reported fidelity to and helpfulness of the manual (n = 10, 15.6%; e.g., “In our book…Always gives as dir[e]ction we want to go” and “… the book is very important our help if get more question we return the book to help.”).

Intervention-specific component domain

Lay leaders discussed implementation and perceived helpfulness of du’a and prophet narratives in 61 (17.3%) responses. In one subdomain theme, responses reflected that lay leaders were implementing the intervention effectively (n = 27, 44.3% of domain responses). For example, when discussing du’a in supervision, one lay leader focused on following the instructions in the manual, “Yes we explained the members what it meant individual turning to Allah [in du’a] and that Allah knows everything yet we need to reveal what we are trying to hide although Allah know,” whereas another discussed adhering to the time suggested for turning to du’a “the time was 20 min according to our plan.” Another lay leader discussing leading the group through the prophet narrative section discussed how lay leaders were summarizing and discussing key session themes, stating,

Yes we discuss the [I]brahims storry and how he was and also we discuss how someone look to future which we conclude that future is not define and what ever happen you should think and hope that future wil[l] be better, and how ever the situation now but in future it will change but all that require how [yo]u think about [A]llah and how much [yo]u trust [A]llah which will brighten your future. So it is all about trust and having a faith in [A]llah.

Additionally, lay leaders emphasized the relevancy, import, and helpfulness of intervention-specific components in supervision in the final subdomain (n = 24, 39.3%). For example, “The most important action in our all session was tur[n]ing to Al[l]ah”, and “After trauma they got ben[e]fit because they [were] turning to Allah” and “in Group A We discussed the [narrative] of [prophet] [Y]unus and everyone got excited about re[d]emption of self and others because it’s good for our live[s].”

Gratitude domain

Responses in this category included an expression of appreciation from lay leaders to supervisors (n = 55, 15.6%) and most commonly occurred at the end of a supervision session prior to signing off. One lay leader stated, “[A]nd we should care at last we [a]re helping our people and all thanks to you [because] [yo]u [are] the one[s] who give us the opportunity to help our people.”

Faith in Allah domain

Responses that reflected the importance of faith outside of the intervention components were coded in this category (n = 11, 3.1%) and focused on the healing power of faith (e.g., “In [I]slamic [principle] faith is kind of p[r]oblem solving [technique]”) and the omnipresence and healing power of Allah (e.g., “They say that [Allah] is the best of all, and refer to [A]llah as a mighty ruler” and “… turning to [A]llah and faith helped more and it will help for sure”).

Technical issues domain

Responses in this category (n = 11, 3.1%) focused on technical issues related to research procedures. Responses primarily focused on poor internet access for uploading initial documents and did not appear to interfere with supervision or group sessions. For example, one lay leader stated, “The main problem in uploading is poor internet.”

Supervisor text behavior qualitative domains

Four broad supervisor domains emerged: general supervision techniques, guidance on intervention implementation, program procedures, and miscellaneous. Responses coded as miscellaneous did not fit into any broad category but still contained meaningful responses (e.g., “Forgiveness and redemption are good for our lives!”). Table 2 provides the number of and percentages of supervisor responses for each domain and subdomain as well as the percentages of subdomain responses that fell within each domain.

Supervisor domains and subdomains during WhatsApp supervision
Supervisor domain and subdomains Responses (n) Supervisor responses (%) Domain responses (%)
General supervision techniques 56 34.9
Praise and encouragement 45 26.0 80.0
Offering support 13 7.7 23.6
Guidance on intervention implementation 45 26.6
Intervention-specific components (e.g., du’a and prophet narratives) 26 15.4 57.8
Checking engagement and adherence 22 13.0 48.9
Skill development and psychoeducation 9 5.3 20.0
Planning for future sessions 8 4.7 17.8
Technical and program evaluation 16 9.5
Miscellaneous 6 3.0
Excluded 59 34.9
  • Note. N = 169. Responses could be coded in multiple domains and subdomains; percentages may total to over 100.0%.

General supervision techniques domain

Supervisors mentioned the use of general supportive techniques that were not specific to the ITH protocol in 56 responses (34.9%), including subdomains of praise and encouragement (n = 45, 80.0% of domain responses) and offering support (n = 13, 23.6% of domain responses). Praise and encouragement focused on promoting confidence and self-efficacy in supervisees’ capabilities as lay leaders. For example, one supervisor stated, “Sounds wonderful!! You are all doing a great job. It is great that you are relying on the book. Keep up the good work!!!” Another said, “Nice job all and I really liked how we got “deeper” today in our discussion. You all are impressive leaders!” Further, in each session, at least one supervisor offered support directly by asking if lay leaders required additional information or clarification on intervention protocols. For example, “Any other questions or comments? If not, have a good night and I look forward to next week! Habeen wanaagsan [good night].”

Guidance on intervention implementation domain

Supervisor responses in this broad domain were both predetermined prior to supervision and dynamic in response to lay leader comments. In one subdomain, supervisor guidance regarding intervention-specific components (e.g., du’a and prophet narratives and discussion; n = 26, 57.8% of domain responses) focused on discussion session content and participant engagement with specific intervention components. For example, one supervisor stated, “First week turning to Allah in dua about the trauma, having individual time with Allah. Did group members understand why to do this? How did the time in du’a go?” Supervisors also discussed how intervention-specific components are thought to be mechanistically related to participant change. With regard to du’a and avoidance, one supervisor said, “Yes, fear is often why people avoid and turning to Allah helps address that fear.”

In a second subdomain, supervisor statements checking for adherence and engagement (n = 22, 48.9% of domain responses), tended to focus on attendance (e.g., “Did all groups meet? Are all group members attending?”), summarizing key session themes, and inquiring about group participation during discussion (e.g., “Yoonus tells us about forgiveness and redemption for self and community. In discussion, what were some things the group members said about redemption for self or community?”).

Statements related to skill development and psychoeducation (n = 9, 20.0%), the third subdomain, focused on two main things: common reactions to trauma and building discussion and group-facilitation skills. Regarding the former, one supervisor stated:

[Two] things that make trauma reactions last. “The 1st was avoidance-not thinking about what happened and not doing things that remind of the trauma. In group members answering discussion questions, you want to now listen for these things and encourage members to approach rather than avoid safe reminders. Anyone have an example of avoidance from your group?”

As an example of the latter, a supervisor noted, “This is great that you all are getting members to talk. As we move to next sessions, you want to continue to decrease the time you talk and do the active listening skills we learned. Make sense?”

Finally, in each session, at least one supervisor summarized the session theme and goals for the next session (n = 8, 17.8% of domain responses), forming the final subdomain in this domain, which was related to planning for future sessions. For example, one supervisor texted, “Session 4 focuses on redemption of self and others, has the prophet Yoonus, and individual time turning to Allah in du’a about trauma, with shift in directions to focus on the hardest/most difficult parts of the memory.”

Technical and program evaluation domain

Supervisor responses related to optimizing programmatic procedures (n = 16; 9.5%) focused on the management of technical issues and ongoing program evaluation. Technical issues emerged only in the first two supervision sessions and were primarily focused on research procedures, such as poor internet access for uploading materials. Thus, technical issues did not appear to impede supervision sessions themselves. To produce a collaborative environment and inform future program iterations, supervisors elicited feedback from lay leaders throughout the intervention period. For example, in the final supervision session, one supervisor asked, “Anything you want to tell us about your experience as a leader? About the program?”


Despite being critical for the uptake of psychotherapy (Herschell et al., 2010), clinical supervision is difficult to facilitate in countries with limited providers and resources. Innovative supervision approaches are needed to increase access to mental health treatments in LMICs. In the context of an open clinical trial of a novel intervention for war and refugee trauma-related mental health and community reconciliation, remote clinical supervision of lay leaders in Somaliland was conducted via WhatsApp text messaging. Qualitative analysis of the supervision content pointed to the feasibility and usefulness of this low-cost supervision model. Lay leader comments related to intervention implementation indicated that these leaders comprehended the rationale, generally adhered to procedures, and found intervention components helpful and faith-centered. Low levels of technical problems support the feasibility of this remote supervision model. Additionally, supervisors were able to provide lay leaders with support and encouragement while facilitating general and intervention-specific skill development. Notably, one content area that was largely absent in the qualitative data was perceived stigma about discussing trauma or mental health problems. The present findings point to the potential utility of WhatsApp-based supervision (Henry et al., 2016) and reflect the first effort to examine this supervision modality in the context of a global mental health intervention.

Supervision via WhatsApp included structured elements that addressed the implementation of the intervention, which may be present in in-person supervision with trained psychotherapists. From the qualitative analysis, the content covered matched the intervention and appeared to systematically reinforce the lay leaders’ skills. This was most evident in the group procedures domain, which highlighted an increase in self-confidence and discussion-leading skill development throughout the course of the intervention. Previous studies evaluating face-to-face, telephone, and internet-based remote supervision (e.g., via Skype) have emphasized adherence to and competence in implementing specific interventions (e.g., Dorsey et al., 2020; Murray et al., 2014; Rahman et al., 2019). For example, in a study evaluating the effectiveness of trauma-focused CBT administered by lay counselors in Tanzania and Kenya, remote audio- and text-based supervision focused on role plays and practice with lay-supervisor feedback (Dorsey et al., 2020). Initially, ITH supervision focused on basic implementation; that is, whether components of the intervention were being completed and whether there were any implementation problems. Components appeared to be done reasonably well, and there were only minor implementation problems. The next focus was on reinforcing discussion-leading skills, namely helping lay leaders ask open-ended questions and utilize supportive listening techniques. One of the key features in this initial phase was strong positive reinforcement by supervisors. Notably, a collaborative supervisory relationship has been shown to be associated with effective supervision over and above the impact of specific supervision techniques (Britt & Gleaves, 2009; for a review see Kilminster & Jolly, 2000). An empathetic, supportive, and praising supervision style is key in facilitating rapport (Watkins, 1997). As supervision progressed, sessions focused on addressing avoidance and approaching unhelpful negative cognitions. This led to much more nuanced text discussions, even on WhatsApp, about the role of approaching the trauma memory in du’a and discussions about the lives of prophets in shifting maladaptive thinking about the trauma towards reconciliation. Taken together, a flexible yet structured format may have helped ascertain key elements of intervention fidelity on the part of the lay leaders and further developed their perceived skills in addressing key areas, such as avoidance and maladaptive trauma-related cognitions.

The broader embedding of the intervention within the mosque infrastructure and Islamic-focused content further created a shared partnership, where both the supervisors and leaders learned from one another. This is consistent with recommendations made by Raiya and Pargament (2010) on centering faith appropriately in the therapy process, promoting positive religious coping, and addressing religious struggles within a religious framework. Perceived stigma associated with mental illness represents a key barrier to mental health treatment utilization in Somali culture (Bentley & Owens, 2008; Wolf et al., 2016). Despite this, stigma about discussing trauma or mental health problems was not a focus of supervision. It is possible that some of these concerns were mitigated given the lay leaders’ perceptions of ITH as culturally relevant. That is, leading discussions about posttraumatic reactions may have felt approachable in the context of prophet narratives and du’a and in mosques. In addition, group leaders and participants were explicitly advised that individual disclosure about traumatic experiences was not encouraged during group discussions. Beyond ITH, the findings from this study highlight how collaboratively developed, culturally relevant interventions may be particularly well-suited to promote a supportive approach to supervision necessary to facilitate tasking-shifting, community ownership, and long-term sustainability.

The results of this novel work should be interpreted with several limitations in mind. First, our sample of lay leaders and supervisors was small, and further evaluation of WhatsApp text supervision delivery within larger trials is needed to better understand how to optimize remote supervision. However, to our knowledge, this was the first mental health trial conducted in Somalia, and the success of both the intervention (Zoellner et al., 2021) and the remote supervision model is consequential.

This study highlights that remote supervision may be successful by demonstrating that face-to-face supervision (e.g., therapeutic skill development, addressing implementation issues, and providing encouragement and support) can be accomplished via WhatsApp. Pilot work of this nature is essential before moving on to larger efficacy and implementation trials using this form of text supervision.

Second, no in-person observation of intervention fidelity monitoring occurred, so the degree of intervention adherence cannot be conclusively determined. Fidelity regarding the timing of each intervention component was monitored via checklists completed after each session. However, the findings from qualitative analyses of supervision content support that lay leaders felt they were developing competence in program delivery and that essential program components were being discussed. Further, 39.2% of lay-leader responses and 34.9% of supervisor responses were excluded from the analysis. This was, in part, due to the brief nature of some WhatsApp communications (e.g., a lay leader responding simply “yes” or “ok” to a supervisor).

Third, lay leaders were translating responses from Somali to English in a back-channel WhatsApp thread. As expected, several responses were not interpretable and deemed lost in translation and, thus, were ultimately excluded. Additionally, coding lay leader and supervisor responses with separate categories and subcategories may have limited our ability to observe the dynamic interaction between participants. However, separate themes for lay leaders and supervisors clearly emerged, and the dynamic conversation was used to inform coding when the context of a response was unclear.

Finally, we did not assess whether supervision itself was linked to improved clinical outcomes. Notably, in this trial, ITH led to large and clinically meaningful effects on PTSD, depression, somatic symptoms, and well-being (Zoellner et al., 2021). Future work should include quantitative measures of WhatsApp supervision effectiveness to evaluate the association between supervision and clinical outcomes.

Importantly, this work was developed and implemented centering on community participatory principles (Bentley et al., 2021). This context forms the basis of the observed intervention, training, and implementation of supervision, reflecting shared decision-making and mutual respect. In this context, remote supervision conducted via WhatsApp was technologically feasible, effective for perceived skill development, and may have facilitated intervention implementation for this lay-led intervention. Although the findings highlight the benefits of WhatsApp supervision for ITH, in particular, many of the identified qualitative themes may generalize to other interventions. Specifically, developing therapeutic skills in discussion-leading, supportive listening, and addressing avoidance and maladaptive cognitions may be relevant for other cognitive behavioral therapies administered in LMICS. Additionally, the ability of supervisors to provide support and encouragement to supervisees is important regardless of intervention modality. When tailored to the local context, remote supervision approaches hold promise for increasing access to services in LMICs with limited resources. Given the challenges of providing ongoing high-quality supervision in remote, low-resource areas, the ease and low cost of WhatsApp supervision are very encouraging. New models of mental health care dissemination and implementation are needed to address the substantial trauma burden in war-torn LMICs.


The study reported in this article was not formally preregistered. Neither the data nor the materials have been made available on a permanent third-party archive; requests for the data or materials should be sent via email to the lead author at


This study received funding from the University of Washington Population Health Initiative ( Identifier: NCT03761732). The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.


  • Abu Raiya, H., & Pargament, K. I. (2010). Religiously integrated psychotherapy with Muslim clients: From research to practice. Professional Psychology: Research and Practice, 41(2), 181188.
  • Baingana, F., & Mangen, P. O. (2011). Scaling up of mental health and trauma support among war-affected communities in northern Uganda: Lessons learned. Intervention, 9(3), 291303.
  • Beidas, R. S., & Kendall, P. C. (2010). Training therapists in evidence-based practice: A critical review of studies from a systems-contextual perspective. Clinical Psychology: Science and Practice, 17(1), 130.
  • Bentley, J. A., Feeny, N. C., Dolezal, M. L., Klein, A., Marks, L. H., Graham, B., & Zoellner, L. A. (2021). Islamic trauma healing: Integrating faith and empirically supported principles in a community-based program. Cognitive and Behavioral Practice, 28(2), 167192.
  • Bentley, J. A., & Owens, C. W. (2008, November 1). Somali refugee mental health cultural profile.
  • Britt, E., & Gleaves, D. H. (2011). Measurement and prediction of clinical psychology students’ satisfaction with clinical supervision. The Clinical Supervisor, 30(2), 172182.
  • Cavallera, V., Reggi, M., Abdi, S., Jinnah, Z., Kivelenge, J., Warsame, A. M., Yusuf, A. M., & Ventevogel, P. (2016). Culture, context and mental health of Somali refugees: A primer for staff working in mental health and psychosocial support programmes. United Nations High Commissioner for Refugees.
  • Dietrich, H., Al Ali, R., Tagay, S., Hebebrand, J., & Reissner, V. (2019). Screening for posttraumatic stress disorder in young adult refugees from Syria and Iraq. Comprehensive Psychiatry, 90, 7381.
  • Dorsey, S., Lucid, L., Martin, P., King, K. M., O’Donnell, K., Murray, L. K., Wasonga, A. I., Itemba, D. K., Cohen, J. A., Manongi, R., & Whetten, K. (2020). Effectiveness of task-shifted trauma-focused cognitive behavioral therapy for children who experienced parental death and posttraumatic stress in Kenya and Tanzania: A randomized clinical trial. JAMA Psychiatry, 77(5), 464473.
  • Gerritsen, A. A., Bramsen, I., Deville, W., van Willigen, L. H., Hovens, J. E., & Van der Ploeg, H. M. (2006). Use of health care services by Afghan, Iranian, and Somali refugees and asylum seekers living in The Netherlands. The European Journal of Public Health, 16(4), 394399.
  • Henry, J. V., Winters, N., Lakati, A., Oliver, M., Geniets, A., Mbae, S. M., & Wanjiru, H. (2016). Enhancing the supervision of community health workers with WhatsApp mobile messaging: Qualitative findings from 2 low-resource settings in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya. Global Health: Science and Practice, 4(2), 311325.
  • Herschell, A. D., Kolko, D. J., Baumann, B. L., & Davis, A. C. (2010). The role of therapist training in the implementation of psychosocial treatments: A review and critique with recommendations. Clinical Psychology Review, 30(4), 448466.
  • Hsieh, H. F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 12771288.
  • Jacob, N., Neuner, F., Maedl, A., Schaal, S., & Elbert, T. (2014). Dissemination of psychotherapy for trauma spectrum disorders in postconflict settings: A randomized controlled trial in Rwanda. Psychotherapy and Psychosomatics, 83(6), 354363.
  • Karam, E. G., Friedman, M. J., Hill, E. D., Kessler, R. C., McLaughlin, K. A., Petukhova, M., Sampson, L., Shahly, V., Angermeyer, M. C., Bromet, E. J., de Girolamo, G., de Graff, R., Demyttenaere, K., Ferry, F., Florescu, S. E., Haro, J. M., He, Y., Karam, A. N., Kawakami, N., … Koenen, K. C. (2014). Cumulative traumas and risk thresholds: 12-month PTSD in the World Mental Health (WMH) surveys. Depression and Anxiety, 31(2), 130142.
  • Kemp, S. (2021, February 12). Digital 2021: Somalia.
  • Kemp, C. G., Petersen, I., Bhana, A., & Rao, D. (2019). Supervision of task-shared mental health care in low-resource settings: A commentary on programmatic experience. Global Health: Science and Practice, 7(2), 150159.
  • Khan, M. N., Hamdani, S. U., Chiumento, A., Dawson, K., Bryant, R. A., Sijbrandij, M., Nazir, H., Akhtar, P., Masood, A., Wang, D., Wang, E., Uddin, I., van Ommeren, M., & Rahman, A. (2017). Evaluating feasibility and acceptability of a group WHO trans-diagnostic intervention for women with common mental disorders in rural Pakistan: A cluster randomised controlled feasibility trial. Epidemiology and Psychiatric Sciences, 28(1), 7787.
  • Kilminster, S. M., & Jolly, B. C. (2000). Effective supervision in clinical practice settings: A literature review. Medical Education, 34(10), 827840.
  • Koenen, K. C., Ratanatharathorn, A., Ng, L., McLaughlin, K. A., Bromet, E. J., Stein, D. J., Karam, E. G., Meron Ruscio, A., Benjet, C., Scott, K., Atwoli, L., Petukhova, M., Lim, C. C. W., Aguilar-Gaxiola, S., Al-Hamzawi, A., Alonso, J., Bunting, B., Ciutan, M., de Girolamo, G., … Kessler, R. C. (2017). Posttraumatic stress disorder in the World Mental Health Surveys. Psychological Medicine, 47(13), 22602274.
  • Lang, D., Zoellner, L., Graham, B., Marks, L. H., & Feeny, N. C. (2016). Islahul qulub: Islamic trauma healing. University of Washington Center for Anxiety & Traumatic Stress.
  • Magidson, J. F., Lejuez, C. W., Kamal, T., Blevins, E. J., Murray, L. K., Bass, J. K., Bolton, P., & Pagoto, S. (2015). Adaptation of community health worker-delivered behavioral activation for torture survivors in Kurdistan, Iraq. Global Mental Health, 2, e24.
  • Mahmuda, M., Miah, M. A. A., Elshazly, M., Khan, S., Tay, A. K., & Ventevogel, P. (2019). Contextual adaptation and piloting of Group Integrative Adapt Therapy (IAT-G) amongst Rohingya refugees living in Bangladesh. Intervention, 17(2), 149159.
  • McLean, K. E., Kaiser, B. N., Hagaman, A. K., Wagenaar, B. H., Therosme, T. P., & Kohrt, B. A. (2015). Task sharing in rural Haiti: Qualitative assessment of a brief, structured training with and without apprenticeship supervision for community health workers. Intervention, 13(2), 135155.
  • Murray, L. K., Dorsey, S., Haroz, E., Lee, C., Alsiary, M. M., Haydary, A., Weiss, W. M., & Bolton, P. (2014). A common elements treatment approach for adult mental health problems in low-and middle-income countries. Cognitive and Behavioral Practice, 21(2), 111123.
  • Neuner, F., Onyut, P. L., Ertl, V., Odenwald, M., Schauer, E., & Elbert, T. (2008). Treatment of posttraumatic stress disorder by trained lay counselors in an African refugee settlement: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 76(4), 686694.
  • Onyut, L. P., Neuner, F., Ertl, V., Schauer, E., Odenwald, M., & Elbert, T. (2009). Trauma, poverty and mental health among Somali and Rwandese refugees living in an African refugee settlement—an epidemiological study. Conflict and Health, 3(6), 116.
  • Piwowarczyk, L., Bishop, H., Yusuf, A., Mudymba, F., & Raj, A. (2014). Congolese and Somali beliefs about mental health services. Journal of Nervous and Mental Disease, 202(3), 209216.
  • Rahman, A., Akhtar, P., Hamdani, S. U., Atif, N., Nazir, H., Uddin, I., Nisar, A., Huma, Z., Maselko, J., Sikander, S., & Zafar, S. (2019). Using technology to scale-up training and supervision of community health workers in the psychosocial management of perinatal depression: A non-inferiority, randomized controlled trial. Global Mental Health, 6, e8.
  • Ritchie, J., & Spencer, L. (1994). Qualitative data analysis for applied policy research. In R. A. Bryman & R. G. Burgess (Eds.), Analyzing qualitative data (pp. 187208). Routledge.
  • Saxena, S., Thornicroft, G., Knapp, M., & Whiteford, H. (2007). Resources for mental health: Scarcity, inequity, and inefficiency. The Lancet, 370(9590), 878889.
  • Watkins Jr., C. E. (1997). The ineffective psychotherapy supervisor: Some reflections about bad behaviors, poor process, and offensive outcomes. The Clinical Supervisor, 16(1), 163180.
  • Wolf, K. M., Zoucha, R., McFarland, M., Salman, K., Dagne, A., & Hashi, N. (2016). Somali immigrant perceptions of mental health and illness: An ethnonursing study. Journal of Transcultural Nursing, 27(4), 349358.
  • World Health Organization. (2010, October). A situation analysis of mental health in Somalia.
  • World Health Organization, PEPFAR, & UNAIDS. (2007). Task shifting: Rational redistribution of tasks among health workforce teams: Global recommendations and guidelines.
  • Zoellner, L., Graham, B., Marks, E., Feeny, N., Bentley, J., Franklin, A., & Lang, D. (2018). Islamic trauma healing: Initial feasibility and pilot data. Societies, 8(3), 4759.
  • Zoellner, L. A., Bentley, J. A., Feeny, N. C., Klein, A. B., Dolezal, M. L., Angula, D. A., & Egeh, M. H. (2021). Reaching the unreached: Bridging Islam and science to treat the mental wounds of war.” Frontiers in Psychiatry, 12, 599293.

This site uses Akismet to reduce spam. Learn how your comment data is processed.