Impact statement
In low- and middle-income (LMIC) countries, as well as resource-limited settings in high-income countries (HICs), implementing task shifting and task sharing interventions may increase the capacity of healthcare organizations to provide mental health and psychosocial support (MHPSS) services to people living with HIV/AIDS (PLWH). Training lay health workers (LHWs), such as non-specialist health care workers, community health workers, and peers to deliver MHPSS services can improve access to and engagement in mental health care among PLWH and may provide relief to overburdened health care systems. This global scoping review describes and characterizes the extent to which LHWs have been mobilized to deliver MHPSS services to PLWH. Specifically, it summarizes the literature on this topic over a 10-year period (2013–2022), describing the characteristics of interventions studied, results, implementation barriers and facilitators, study limitations, gaps in knowledge, and areas for future research.
Findings from this review indicate that task shifting and sharing are widely used strategies among LMICs and have been relatively successful at improving mental health outcomes among PLWH. Researchers and practitioners can use this scoping review as a resource to guide the future development of MHPSS task shifting and sharing studies and interventions for PLWH.
Introduction
In 2023, the World Health Organization (WHO) estimated 39.9 million people were living with HIV worldwide, including 1.3 million (3.26%) newly diagnosed; that same year, an estimated 630,000 people died of HIV-related causes globally (WHO, 2024b). Low- and middle-income countries (LMICs) shoulder a disproportionate share (over 80% in 2020) of the global burden of HIV (Allel et al., Reference Allel, Jaoude, Birungi, Palmer, Skordis and Haghparast-Bidgoli2022). African countries are most affected (Shao and Williamson, Reference Shao and Williamson2012; WHO, 2024b), comprising 26 million (65.16%) cases, 640,000 (49.23%) new infections, and 61.90% of HIV-related deaths globally among people living with HIV (PLWH) (WHO, 2024b).
PLWH experience disproportionately high rates of mental health conditions, e.g., depression, anxiety, and post-traumatic stress disorder (PTSD) (LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015). The consequences of not addressing the mental health concerns of PLWH are reduced retention in care (Anderson et al., Reference Anderson, Monda, Haardorfer and Waldrop-Valverde2019) and adherence to antiretroviral therapy (ART) (Crepaz et al., Reference Crepaz, Passin, Herbst, Rama, Malow, Purcell and Wolitski2008; Boarts et al., Reference Boarts, Buckley-Fisher, Armelie, Bogart and Delahanty2009; Pence, Reference Pence2009; Mayston et al., Reference Mayston, Kinyanda, Chishinga, Prince and Patel2012; LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015; Anderson et al., Reference Anderson, Monda, Haardorfer and Waldrop-Valverde2019; Smith and Cook, Reference Smith and Cook2019; Hou et al., Reference Hou, Fu, Meng, Jiang, Guo, Wu, Su and Zhang2020), disease progression (e.g., increased viral loads [Anderson et al., Reference Anderson, Monda, Haardorfer and Waldrop-Valverde2019; LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015], lower CD4 cell counts [Anderson et al., Reference Anderson, Monda, Haardorfer and Waldrop-Valverde2019; LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015; Pence, Reference Pence2009], and virologic and treatment failure [LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015; Pence, Reference Pence2009]), poorer quality of life (QoL) (Crepaz et al., Reference Crepaz, Passin, Herbst, Rama, Malow, Purcell and Wolitski2008; Degroote et al., Reference Degroote, Vogelaers and Vandijck2014; Penner-Goeke et al., Reference Penner-Goeke, Henriksen, Chateau, Latimer, Sareen and Katz2015; Kabunga et al., Reference Kabunga, Kigongo, Udho, Auma, Tumwesigye, Musinguzi, Acup, Akello, Okalo, Nabaziwa, E-M and Halima2024), and increased mortality (Pence, Reference Pence2009; Anderson et al., Reference Anderson, Monda, Haardorfer and Waldrop-Valverde2019). Interventions addressing trauma among PLWH are needed, as they are few and understudied (LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015).
Evidence from mental health and HIV intervention literature supports the implementation of task shifting and task sharing to deliver mental health and psychosocial support (MHPSS) interventions to PLWH. The WHO developed task shifting and sharing as strategies to alleviate the burden on health care providers (e.g., mental health professionals), which are often used in LMICs to improve health care access despite limited resources, e.g., training lay health workers (LHWs), such as non-specialist health care workers (HCWs; e.g., nurses and adherence counselors), community health workers (CHWs), and peers (i.e., fellow patients) to deliver needed services (Padmanathan, Reference Padmanathan and De Silva2013; Javanparast et al., Reference Javanparast, Windle, Freeman and Baum2018).
LHWs work in paid/compensated positions or as volunteers, occupying a broad spectrum of roles to improve community health and competence by providing information, practical assistance, and social support (Eng and Parker, Reference Eng, Parker, DiClimente, Crosby and Kegler2002; Scott, Reference Scott2009). They are often “natural helpers” who have shared or lived experiences and can assist healthcare organizations in several ways by improving: (1) health practices by increasing knowledge of and access to health resources and by facilitating the use of services; (2) institutional awareness and responsiveness to community needs; and (3) coordination of services through collaborative relationships with healthcare providers (Eng and Parker, Reference Eng, Parker, DiClimente, Crosby and Kegler2002).
Although substantial evidence exists of the effectiveness of LHW MHPSS interventions in LMICs (Chibanda et al., Reference Chibanda, Cowan, Healy, Abas and Lund2015), little is known of their impact in high-income countries (HICs) (Sikkema et al., Reference Sikkema, Dennis, Watt, Choi, Yemeke and Joska2015; Javanparast et al., Reference Javanparast, Windle, Freeman and Baum2018). In recent years, implementation research concerning the mobilization of LHWs (particularly CHWs) to strengthen the HIV care continuum has become a priority of the United States (U.S.) National Institutes of Health (NIH) (NIH, 2020a, 2020b, 2020c). Han et al., (Reference Han, Kim, Murphy, Cudjoe, Wilson, Sharps and Farley2018) obtained mixed results when conducting a systematic review of 13 CHW interventions, seven of which were conducted in the U.S., to improve psychosocial health outcomes in PLWH. Although none of the interventions yielded statistically significant improvements in mental health outcomes, most studies examined had various methodological and design flaws that may have biased results (e.g., psychosocial outcomes were not primary outcomes, sample size calculations were not performed in advance, a priori power analyses were not conducted for psychosocial variables, unstandardized psychological outcome measures were used, studies had high attrition, and follow-up was short-term) (Han et al., Reference Han, Kim, Murphy, Cudjoe, Wilson, Sharps and Farley2018). Systematic reviews (Boucher et al., Reference Boucher, Liddy, Mihan and Kendall2020; Berg et al., Reference Berg, Page and Øgård-Repål2021; Øgård-Repål et al., Reference Øgård-Repål, Berg and Fossum2021) of peer interventions for PLWH yielded more promising results: three (Simoni et al., Reference Simoni, Pantalone, Plummer and Huang2007; Brashers et al., Reference Brashers, Basinger, Rintamaki, Caughlin and Para2017; Cunningham et al., Reference Cunningham, Weiss, Nakazono, Malek, Shoptaw, Ettner and Harawa2018) of four U.S. studies (Simoni et al., Reference Simoni, Pantalone, Plummer and Huang2007; Brashers et al., Reference Brashers, Basinger, Rintamaki, Caughlin and Para2017; Cunningham et al., Reference Cunningham, Weiss, Nakazono, Malek, Shoptaw, Ettner and Harawa2018; Merlin et al., Reference Merlin, Westfall, Long, Davies, Saag, Demonte, Young, Kerns, Bair, Kertesz, Turan, Kilgore, Clay, Starrels, Pekmezi and Johnson2019) reported significant improvements in mental health outcomes, including reductions in depressive symptomology/decline in depression and increased use of mental health care.
Studies indicate that peer-delivered interventions are both acceptable and feasible for people in need of mental health services (Repper and Carter, Reference Repper and Carter2011; Miyamoto and Sono, Reference Miyamoto and Sono2012; Padmanathan, Reference Padmanathan and De Silva2013), as well as PLWH (Wewers et al., Reference Wewers, Neidig and Kihm2000; Wolitski et al., Reference Wolitski, Gómez and Parsons2005; Purcell et al., Reference Purcell, Latka, Metsch, Latkin, Gómez, Mizuno, Arnsten, Wilkinson, Knight, Knowlton, Santibanez, Tobin, Dawson-Rose, Valverde, Gourevitch, Eldred and Borkowf2007; Webel, Reference Webel2010; Horvath et al., Reference Horvath, Michael Oakes, Simon Rosser, Danilenko, Vezina, Rivet Amico, Williams and Simoni2013; Enriquez et al., Reference Enriquez, Cheng, Banderas, Farnan, Chertoff, Hayes, Ortego, Moreno, Peterson and McKinsey2014; Steward et al., Reference Steward, Sumitani, Moran, Ratlhagana, Morris, Isodoro, Gilvydis, Tumbo, Grignon, Barnhart and Lippman2018; Merlin et al., Reference Merlin, Westfall, Long, Davies, Saag, Demonte, Young, Kerns, Bair, Kertesz, Turan, Kilgore, Clay, Starrels, Pekmezi and Johnson2019; Boucher et al., Reference Boucher, Liddy, Mihan and Kendall2020). Furthermore, the WHO champions the mobilization of LHWs, especially peers, as an acceptable and even preferred means of delivering services to PLWH (WHO, 2016; Haberer et al., Reference Haberer, Sabin, Rivet Amico, Orrell, Galárraga, Tsai, Vreeman, Wilson, Sam-Agudu, Blaschke, Vrijens, Mellins, Remien, Weiser, Lowenthal, Stirratt, Salif Sow, Thomas, Ford, Mills, Lester, Nachega, Bwana, Ssewamala, Mbuagbaw, Munderi, Geng and Bangsberg2017; Boucher et al., Reference Boucher, Liddy, Mihan and Kendall2020). In LMICs, peer support, mentorship, and counseling interventions have been effective in reducing symptoms of depression and anxiety among PLWH (Mayston et al., Reference Mayston, Kinyanda, Chishinga, Prince and Patel2012; Sikkema et al., Reference Sikkema, Dennis, Watt, Choi, Yemeke and Joska2015; Asrat et al., Reference Asrat, Schneider, Ambaw and Lund2020). For PLWH experiencing mental health issues, peers may be mobilized to greater effect than other LHWs due to their lived experience and ability to better relate to their clients’ concerns, serving as care navigators and role models for recovery (Berg et al., Reference Berg, Page and Øgård-Repål2021; Øgård-Repål et al., Reference Øgård-Repål, Berg and Fossum2021; Krulic et al., Reference Krulic, Brown and Bourne2022). Understanding how different LHWs compare in effectiveness to one another and usual care may influence LHW mobilization for future delivery of MHPSS services.
In some countries, LHWs have been empowered to deliver MHPSS services to PLWH, including therapeutic interventions, such as cognitive behavioral therapy (CBT) (Chibanda et al., Reference Chibanda, Cowan, Healy, Abas and Lund2015; Asrat et al., Reference Asrat, Schneider, Ambaw and Lund2020). However, it is important to note that practice requirements may vary by type of therapy and across countries and implementing institutions/organizations (e.g., licensing/certification, advanced degree/higher education, and specialized training), posing potential barriers to service delivery. Improving awareness of which MHPSS interventions are being delivered to PLWH, by whom (including their qualifications and training), where, and to what outcome will advance our understanding of how LHWs can help close the mental health treatment gap and reinforce steps along the HIV care continuum.
Past reviews have extensively covered task shifting and sharing interventions for PLWH but have not focused on mental health outcomes or have been more limited in scope (e.g., only focused on one type of intervention) or setting (e.g., LMICs only). We were interested in how LHWs have been mobilized in different contexts and settings to deliver MHPSS services, identifying current trends and gaps in knowledge, and informing future research. To this purpose, we conducted a global scoping review (inclusive of both LMICs and HICs) to characterize the recent literature (publications from 2013 through 2022) on MHPSS task shifting and sharing interventions for PLWH.
Additionally, due to the scarcity of empirically supported trauma interventions concerning PLWH identified by LeGrand et al. (Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015), as a secondary interest, we sought to identify if and how trauma influenced intervention design and selection among studies reviewed, and whether a gap remains.
Methods
This paper is guided by the preferred reporting items for systematic review and meta-analyses (PRISMA) extension for scoping reviews (PRISMA-ScR) checklist and explanatory paper (Tricco et al., Reference Tricco, Lillie, Zarin, O’Brien, Colquhoun, Levac, Moher, Peters, Horsley, Weeks, Hempel, Akl, Chang, McGowan, Stewart, Hartling, Aldcroft, Wilson, Garritty, Lewin, Godfrey, Macdonald, Langlois, Soares-Weiser, Moriarty, Clifford, Tunҫalp and Straus2018).
Eligibility criteria
To be included in this review, intervention studies met the following eligibility criteria: (1) they were published in a peer-reviewed journal from 2013 through 2022, (2) they were written in English, (3) the study population was adult PLWH aged 18+, (4) the primary/main study outcome(s) were related to improving mental health, and (5) a task shifting or task sharing approach was used to deliver MHPSS services. Quantitative and qualitative study designs were eligible for inclusion, provided they met all other criteria; however, literature reviews were not. Additionally, papers were excluded if the intervention studied was not fully implemented at the time of publication. Studies targeting youth and adolescents that included some participants aged 18+ were also excluded.
We used a broad definition for mental health outcome, which included scores from the measurement/assessment of a specific disorder (e.g., depression, anxiety, and PTSD) and related constructs (e.g., QoL and resiliency [Chuang et al., Reference Chuang, JY- and Wang2023; González-Blanch et al., Reference González-Blanch, Hernández-de-Hita, Muñoz-Navarro, Ruíz-Rodríguez, Medrano and Cano-Vindel2018; Hu et al., Reference Hu, Zhang and Wang2015]), as well as changes in access to and engagement in mental health services (e.g., linkage to services and a number of mental health visits). For mental health to be considered a primary study outcome, it had to be explicitly stated as the main purpose for the conducted research. If mental health was one of several outcomes but not prioritized, it was not considered a primary outcome.
Pilot studies assessing acceptability and feasibility were only included if improving mental health was the intervention’s primary purpose, and mental health outcomes were also reported. Task shifting and sharing were defined as intervention strategies delegated to or collaboratively mobilizing non-specialist HCWs and other LHWs (e.g., CHWs and peers) to provide services to PLWH. Interventions delivered by research staff or individuals with prior education or training, such as a degree or certificate in mental health, were excluded (neither task shifting nor task sharing).
Search strategy and selection process
An initial search of the PubMed/MEDLINE database was conducted to explore results from a combination of key search components and variations of related terms: i.e., (population, e.g., PLWH) AND (mental health, e.g., depression, anxiety, and PTSD) AND (intervention/program) AND (task shifting/sharing, e.g., lay health). When the study team was satisfied with the relevance of the search returns, search terms and conditions were finalized and translated to other databases (i.e., PsycINFO, Global Health, Web of Science, and SCOPUS) to source additional studies (File S1). Filters were also applied to limit search results to English-language (e.g., “Eng[lang]”) publications from 2013 through 2022 (e.g., 2013/01/01 “[PDAT]”:2023/01/01 “[PDAT]”).
All studies retrieved from these searches were uploaded to Covidence, an online systematic review management platform. The software identified and flagged duplicates for review, which the principal investigator (PI) removed where applicable. The articles were then screened for selection and data extraction. The PI conducted the initial title and abstract review of each study retrieved. Although literature reviews in our search were not included, their reference lists were examined to ensure that relevant studies were still captured.
Two reviewers (the PI and another doctoral student) then read the full text of the remaining articles and independently decided whether each study met the criteria for inclusion. Coding categories for study exclusion criteria were created in Covidence, allowing reviewers to provide their rationale for excluding studies. Although studies may have had multiple reasons for exclusion, Covidence software permitted only one rationale selection. Thus, reviewers selected their primary reason for exclusion, although this was not systematically done. Reviewers then met to discuss and resolve discrepancies in study selection flagged by Covidence. When they disagreed, a third reviewer (a subject matter expert) was consulted and a consensus was reached before a final decision was made to include or exclude the study. Study quality was not assessed and did not influence selection.
Data extraction
After the studies were selected, two reviewers extracted relevant data from a close reading of each publication and entered it into a Microsoft Excel spreadsheet table. Reviewers pulled the following information from each study: (1) publication year; (2) study design and comparison groups; (3) location and setting; (4) data collection period; (5) study aim; (6) sample size and demographic data; (7) intervention characteristics and whether trauma was a consideration in the design or selection of an intervention; (8) LHW characteristics, roles, training, compensation, and intervention responsibilities; and (9) mental health results. Barriers and facilitators to implementation were documented and study limitations were noted to identify gaps for further research.
Data analysis
Data were explored through close reading, paraphrasing, and reorganizing and restructuring tables in Microsoft Excel. The study team then generated summaries for each data item and identified themes and patterns that emerged across studies. Summary tables were created in Microsoft Word to visually represent findings from this review and are presented in the next section.
Results
Study selection
Our literature search retrieved 1,504 papers; after removing duplicates, 885 remained. Although literature reviews were excluded, we identified 14 additional papers from reference lists of relevant reviews among our search returns and screened a total of 899 titles and abstracts. After reviewing titles and abstracts, we read the full text of 104 papers to assess their eligibility and excluded 75. Ultimately, 29 papers were included in our review, two of which reported findings from the same intervention and were considered together as one study (N = 28). See Figure 1 for details of the selection process.
Figure 1. PRISMA flow diagram of papers reviewed (N = 1,518).
Study characteristics
Papers included in our review were published (based on issue/citation date) from 2013 through 2022. The most productive year was 2014 with six papers. However, the second half of the review period accounted for 58.6% of the overall sample, and an increase in productivity in recent years (2020–2022) accounted for 37.9% of papers. Of the 28 papers that reported when data collection took place, recruitment often coincided with baseline data collection, and the average total duration of data collection was 15.6 months (range: 2–38). The characteristics of individual studies are presented in Table 1.
Table 1. Characteristics of included studies (N = 28)
Note: Some mental health-related outcomes were assessed but not included in this table, such as functioning impairment (*), functioning (**), social support (***), perceived social support (****), and perceived functional social support (*****).
Study design
Half of the studies in this review (n = 14) were randomized controlled trials (RCTs), including cluster randomized designs (cRCT; n = 4). Other designs used included non-randomized, single-arm, cross-sectional, prospective cohort, case series, times series, and pre-/post-tests. Most studies elected to use quantitative research methods, however, some used qualitative and mixed methods.
Comparators were baseline and follow-up assessment(s), usual care, enhanced usual care, a limited version of the intervention, or an active control. Studies did not compare nor assess differences in intervention efficacy/effectiveness based on LHW classification. However, Wagner et al., Reference Wagner, Ngo, Goutam, Glick, Musisi and Akena2016 compared results between two non-specialist HCW intervention models – “structured protocol” via trained nurses and “clinical acumen” via primary care providers (mostly nurses) – rather than using a control. Additionally, Myers et al., Reference Myers, Lombard, Lund, Joska, Levitt, Naledi, Petersen Williams, van der Westhuizen, Cuijpers, Stein and Sorsdahl2022 compared two CHW approaches – dedicated care (representing 100% effort of the CHW) versus designated care (additional responsibilities above and beyond the CHW’s usual role) – to treatment as usual.
Study aims
Studies primarily aimed to assess intervention efficacy or effectiveness (n = 21) from pre-post and over time. Twelve (42.9%) of the studies were pilots, some of which also evaluated the acceptability (n = 7) and feasibility (n = 8) of interventions, e.g., attendance, retention, fidelity, and identifying barriers and facilitators. Stated aims also included exploring the associations between intervention-related independent variables and mental health-related dependent variables to identify predictors, as well as the effects of other variables (e.g., mediation and moderation) on intervention effects. One qualitative study, Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020, explored the perspectives and experiences of LHWs delivering the intervention and clients receiving it. Wagner et al., Reference Wagner, Ngo, Goutam, Glick, Musisi and Akena2016 compared the implementation of two depression care intervention models for their advantages and disadvantages in outcomes.
Location
Based on the 2024 fiscal year World Bank income classification (WB, 2024b, 2024c), which is determined by the 2022 calendar year (the publication year of the most recent study in this review) gross national income per capita as calculated using the World Bank Atlas method (WB, 2024a), nearly all studies were conducted in LMICs (n = 27, 96.4%) with over half occurring in the WHO African region (n = 16, 57.1%). The U.S. was the only HIC with a study eligible for inclusion in this review. None of the studies in this review were conducted in the European and Eastern Mediterranean regions. China had the greatest representation among all the countries involved in research on this subject matter (n = 6, 21.4%). The number of studies in each location is reported by country, WHO region, and World Bank income classification in Table 2.
Table 2. Studies by WHO region and World Bank income classification (N = 28)
Note: *Two papers reported findings from the same study. Regional categories are from the World Health Organization (WHO, 2024a) and income-level country classifications are based on the World Bank Fiscal Year 2024 (Calendar Year 2022) – the publication year of the most recent study in this review (WB, 2024b, 2024c).
Setting
The majority of studies took place at clinics/health centers (n = 21, 75%), which were often affiliated with and/or government-funded/operated (n = 10, 35.7%), hospitals (n = 7, 25%), NGOs/nonprofits (n = 5, 17.9%), and universities/research institutes (n = 3, 10.7%). Study sites provided primary care (n = 7, 25%) and/or specialized care, e.g., HIV (n = 13, 46.4%), infectious diseases (n = 2, 7.1%), and prenatal (n = 1, 3.6%) care.
Study sample
The sample size average was 377 (median: 100, range: 9–1,473). Most studies included both men and women. To participate, some individuals were required to have symptoms of mental illness or diagnoses of major depressive disorder (MDD) or PTSD. Other criteria included ART enrollment or the length of time since diagnosis. Some studies included participants who were not PLWH, such as family members (Chen et al., Reference Chen, Shiu, Yang, Wang, Zhang, Zhang, Reynolds, Kennedy, Khoshnood, Chen, Bao, Zhao and Lu2018), community members (Prinsloo et al., Reference Prinsloo, Greeff, Kruger and Ellis2016), diabetes patients (Myers et al. 2020), and other non-PLWH (Yu et al., Reference Yu, Lau, Mak, Cheng, Lv and Zhang2014; Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020). However, their involvement was typically a key component of intervention design and implementation for the benefit of PLWH (e.g., stigma reduction efforts), except for the study by Myers et al. 2020, and their outcomes were reported separately. Only one study, Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020, reported studying PLWH with histories of trauma.
Interventions
Studies assessed novel intervention packages, as well as evidence-based MHPSS interventions and treatment models/frameworks, including problem-solving therapy (PST; n = 5), cognitive behavioral therapy (CBT; n = 4), interpersonal therapy (IPT; n = 3), stepped care model (n = 2), self- and family-based management intervention (SAFMI; n = 1), and measurement-based care (MBC; n = 1). Some studies (Masquillier et al., Reference Masquillier, Wouters, Mortelmans and FlR2014; Thomson et al., Reference Thomson, Rich, Kaigamba, Socci, Hakizamungu, Bagiruwigize, Binagwaho and Franke2014; Williams et al., Reference Williams, Wang, Li, Chen, Li and Fennie2014 and Pokhrel et al., Reference Pokhrel, Sharma, Pokhrel, Neupane, Mlunde, Poudel and Jimba2018) assessed the impact of adherence support interventions for ART on mental health.
With few exceptions, studies did not disclose whether trauma informed intervention design nor selection. Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020 assessed the impact of The Friendship Bench (an intervention for common mental disorders and not “trauma-specific”) on PLWH with trauma histories in Zimbabwe and explored sources of their trauma. Similarly, Meffert et al., Reference Meffert, Neylan, McCulloch, Blum, Cohen, Bukusi, Verdeli, Markowitz, Kahn, Bukusi, Thirumurthy, Rota, Rota, Oketch, Opiyo and Ongeri2021 and Nakimuli-Mpungu et al., Reference Nakimuli-Mpungu, Musisi, Wamala, Okello, Ndyanabangi, Birungi, Nanfuka, Etukoit, Mayora, Ssengooba, Mojtabai, Nachenga, Harari and Mills2020 assessed interventions’ transdiagnostic potential for treating PTSD.
LHW characteristics
Interventions mobilized non-specialist HCWs (n = 14; e.g., nurses, social workers, adherence counselors, and case managers), peers (n = 12; i.e., fellow PLWH), CHWs (n = 10; i.e., local community members with clinical roles and responsibilities), and other LHWs (n = 2; i.e., traditional medicine practitioners or family/friends/colleagues of PLWH) to deliver MHPSS services to PLWH. Summaries detailing characteristics of LHW mobilization (i.e., roles, responsibilities, and training) by worker type can be found in Table 3. Few studies reported specific qualifications/eligibility requirements but most disclosed characteristics of LHWs that could factor into selection (e.g., profession/clinic experience, education/literacy, and/or lived experience). However, LHWs also appeared to be recruited for convenience and proximity to service locations.
Table 3. Characteristics of LHW mobilization by worker type
Note: This is a summary of what was reported across studies but is not representative of every study.
Compensation
Compensation for LHWs was rarely disclosed and usually discussed in general terms. Some LHWs were employed (Garfin et al., Reference Garfin, Shin, Ekstrand, Yadav, Carpenter, Sinha and Nyamathi2019; Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020), paid (Thomson et al. Reference Thomson, Rich, Kaigamba, Socci, Hakizamungu, Bagiruwigize, Binagwaho and Franke2014), or had funding (Petersen et al., Reference Petersen, Hanass Hancock, Bhana and Govender2014). Nakimuli-Mpungu et al., Reference Nakimuli-Mpungu, Musisi, Wamala, Okello, Ndyanabangi, Birungi, Nanfuka, Etukoit, Mayora, Ssengooba, Mojtabai, Nachenga, Harari and Mills2020 estimated the value of voluntary time that LHWs spent facilitating group support psychotherapy (GSP) or group HIV education sessions as $21.62 USD or the equivalent of three full-time equivalent (FTE) days based on their 2017 earning potential in Uganda ($199 USD per month or about $7 USD per FTE day), which they deemed very cost-effective.
Intervention duration and delivery
The duration of the interventions ranged from 1 to 18 months. Interaction between LHWs and participants ranged from daily to once per month. However, contact occurred on a weekly basis in half of the interventions studied. Interventions were mostly conducted in person, although some took place over the phone or the Internet.
Mental health outcomes
Depression and related outcomes (e.g., major depressive disorder, depressive symptoms, and depression severity) were assessed by nearly 90% of studies, which more often used either the Patient Health Questionnaire-9 (PHQ-9; and abbreviated versions) and/or the Center for Epidemiological Studies Depression Scale (CES-D). Other mental health conditions assessed were anxiety (n = 7, 25%) and PTSD/trauma (n = 3, 10.7%). Four (14.3%) of the studies measured or identified suicide risk, ideation, attempts, and/or deaths during the process of data collection. Quality of life (n = 5, 17.9%), social support (n = 5, 17.9%), and other mental health-related outcomes (e.g., hope, cognitive escape, stress, resilience, and well-being) were also assessed. Three studies assessed the frequency of activities along the mental health continuum, such as screening, evaluation and diagnosis, treatment (i.e., prescription), referral, and navigator interactions.
Synthesis of results
Most studies reported statistically significant improvements in mental health outcomes from baseline to follow-up and between groups post-intervention. Conversely, Guy et al., Reference Guy, Woodward, Kannout and Du Bois2022 found that a peer support intervention to improve treatment engagement among African Americans living with HIV and serious mental illness was successful in rendering the mediation effects of internalized stigma on the association between HIV discrimination and cognitive escape coping insignificant.
Usually, the statistical significance of positive effects was sustained when measured over time (up to 12 months). However, Yu et al., Reference Yu, Lau, Mak, Cheng, Lv and Zhang2014 observed that although the effects on depression were sustained, they were no longer significant after 3 months (p = .05). The authors posit that this change could be explained by the study’s relatively small sample size and that the small to moderate effect size (.29) at follow-up may become statistically significant with a larger study sample. Conversely, Masquillier et al., Reference Masquillier, Wouters, Mortelmans and FlR2014 found that peer adherence support did not have a direct effect on the level of hope among PLWH but that better family function significantly increased the positive effect of the intervention at the second follow-up (2–3 years post-intervention).
Other studies found that significance differed based on the outcome being assessed (Yu et al., Reference Yu, Lau, Mak, Cheng, Lv and Zhang2014; Prinsloo et al., Reference Prinsloo, Greeff, Kruger and Ellis2016; Han et al., Reference Han, Hu, Lu, Zhang, Zhu, Luo, Relf, Mulawa, Pei and Wu2020; Davis et al., Reference Davis, Angeles, McNaughton-Reyes, Matthews, Muessig, Northbrook and Barrington2021; Li et al., Reference Li, Mo, Kahler and Lau2021) or on the type of support, participants received (Davis et al., Reference Davis, Angeles, McNaughton-Reyes, Matthews, Muessig, Northbrook and Barrington2021). Abas et al. (Reference Abas, Nyamayaro, Bere, Saruchera, Mothobi, Simms, Mangezi, Macpherson, Croome, Magidson, Makadzange, Safren, Chibanda and O’Cleirigh2018) found that while participants had lower mean depression scores on the PHQ-9 and the Shona Symptom Questionnaire-14 (SSQ-14), the effects were only significant on the PHQ-9. This could be explained by differences in sensitivity, specificity, and cultural relevancy of the instruments used or that the SSQ-14 is a scale for common mental disorders (e.g., anxiety) and not specific to depression.
Studies did not report any adverse effects due to intervention participation, however, Fajriyah et al. (Reference Fajriyah, Demartoto and Murti2018) found that other variables (i.e., stigma and depression) negatively influenced the quality of life. A few studies did not assess the statistical significance of their findings. Duffy et al. (Reference Duffy, Sharer, Cornman, Pearson, Pitorak and Fullem2017) provided only descriptive statistics (e.g., frequency of positive screens), Nyamayaro et al. (Reference Nyamayaro, Bere, Magidson, Simms, O’Cleirigh, Chibanda and Abas2020) used case studies to illustrate findings from a case series, and Verhey et al. (Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020) was a qualitative study.
Implementation barriers/facilitators
The most common challenges were the burdensome amount of time and effort required to implement the intervention (n = 9, 32.1%; e.g., recruitment, training, supervision, and support), cost of implementation/participation (n = 7, 25%), and distance from study site/access to transportation among participants (n = 5, 17.9%). Other challenges faced by participants included concerns about the intervention (e.g., lack of trust/comfort or privacy and confidentiality), literacy issues (e.g., inability to complete assignments or lack of knowledge/understanding of mental health), and poor help-seeking (e.g., fear of being stigmatized, unfamiliarity with the intervention and benefits, scheduling challenges/competing priorities, and willingness to participate).
Clinic-level barriers were mostly structural: infrastructure/capacity (e.g., lack of meeting space, routine data collection/health records system, or trained personnel), health system incompatibility (e.g., lack of collaboration among different service providers or ability to integrate services), and lack of available resources (e.g., availability of psychosocial services/alternative treatment and ability to address clients’ social service needs).
Finally, some studies (n = 9, 32.1%) identified issues with fidelity to the intervention, including compliance with intervention protocol, attendance/retention among participants, and LHW inability to deliver services due to insufficient training or guidance. However, Anderson et al. (Reference Anderson, Monda, Haardorfer and Waldrop-Valverde2019) suggested that continued education (e.g., booster training to review session content) and implementation of “peer supervision models” (e.g., from fellow nurses trained to implement CBT) may help improve fidelity to the intervention. Additionally, Verhey et al. (Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020) emphasized the need to provide LHWs with clear instructions for delivering MHPSS services to PLWH in need of support (e.g., how to address symptoms of PTSD in PLWH once they are recognized), while Petersen et al. (Reference Petersen, Hanass Hancock, Bhana and Govender2014) highlighted the need to provide organizational leaders with a human resource plan (e.g., LHW role, the scope of practice, and supervision structure) and education about the value of interventions being implemented and service requirements.
Studies also reported a variety of measures taken to facilitate implementation. Researchers and LHWs worked to establish trust through relationship building (e.g., care teams visiting participants at home/off-site or by giving participants continuity with the same service provider) and to ensure participants felt supported. Compensation, transportation, and/or food assistance were provided to help defray the costs of participating in interventions, which were often tailored to the target population (e.g., culturally appropriate, community-informed, client-driven/patient-centered, appropriate education/knowledge level, and delivered in local language) and easy for lay people to deliver. Intervention acceptability was reported as a key component of success. A couple of studies also reported using fidelity checks to ensure interventions were being delivered as intended.
Gaps/future research
Studies in this review self-reported several limitations, mainly related to study design. Low sample size was common (many studies reviewed were pilots), so findings were often preliminary and underpowered. Some studies did not have comparison groups or did not use blinding nor randomization techniques when they did. Limitations of data instruments and omissions in data collection/analysis were also reported. A few studies reported potential contamination or confounding.
To address these limitations, a common suggestion for future research was to pursue large-scale efficacy and effectiveness studies (i.e., RCTs and mixed methods) with control/comparison groups. Researchers also proposed new lines of inquiry and adjustments/changes to interventions (e.g., fewer sessions, using a train-the-trainer approach, and other modalities), as well as alternative methods of data collection (e.g., type of data and measures used) and analysis (e.g., mediation/moderation and subgroup analyses). To better understand the full potential of interventions, several studies recommended assessing effects over longer periods and assessing HIV outcomes. It was also suggested that studies try integrating mental health services (e.g., screening and treatment) into HIV/chronic disease service delivery. Studies called for more research on depression and other common mental disorders, alcohol and substance use, and PTSD.
Discussion
Findings indicate that MHPSS task shifting and task sharing intervention studies published from 2013 through 2022 yielded promising results for PLWH. Studies primarily aimed to assess the efficacy and effectiveness of interventions, and nearly all mental health outcomes measured significantly improved. In our sample, the distribution of studies involving non-specialist HCWs, CHWs, and peers was relatively even, and there was no clear indication nor study showing that one type of LHW should be preferred over the other in terms of efficacy/effectiveness. It is likely that service delivery model selection may solely depend on the priority given by implementers to clinic role/experience, lived experience, and community knowledge/standing when delivering MHPSS services based on context and setting. In fact, this review suggests that significant improvements in the mental health outcomes of PLWH can be achieved when MHPSS services are delivered by LHWs of any kind.
This scoping review identified several gaps in the literature. First, as predicted, there was a great divide between LMICs and HICs concerning the production of literature on this research subject. Within LMICs, as income classification increased, so did the number of studies. HICs are the exception with the least number of studies. Although we had hoped to gain more insight into interventions being used in HICs, only one study was eligible for inclusion in our review, highlighting the continued need for more studies to be conducted in these settings to facilitate the bi-directional translation of knowledge and exchange of ideas between LMICs and HICs. This is of particular importance in HICs with a large mental health provider gap, such as in the U.S., where there is a confluence of a national mental health professional shortage (RHIhub, 2025) and high unmet needs (an estimated 46.2% of adults with mental illness do not receive treatment) (SAMHSA, 2024). Mental health providers may be unable to respond to high patient volume at HIV clinics, thus exacerbating problems with access to and utilization of mental health services, which may account for a significant proportion of PLWH with needs not receiving them (an estimated 27%) (Reif et al., Reference Reif, Whetten, Ostermann and Raper2006; CDC, 2024).
This review adds to growing evidence suggesting that task shifting and sharing interventions are valuable tools in resource-limited settings. The lack of studies in HICs may be explained by barriers such as licensing or certification to deliver certain therapies, as well as the associated costs and pre-requisites (e.g., higher education, prior training, and specialization); it may be time to reconsider what qualifications are necessary to deliver MHPSS services to PLWH and to establish pathways empowering LHWs to attain them.
Second, interventions specifically for PTSD among PLWH (Nakimuli-Mpungu et al., Reference Nakimuli-Mpungu, Musisi, Wamala, Okello, Ndyanabangi, Birungi, Nanfuka, Etukoit, Mayora, Ssengooba, Mojtabai, Nachenga, Harari and Mills2020; Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020; Meffert et al., Reference Meffert, Neylan, McCulloch, Blum, Cohen, Bukusi, Verdeli, Markowitz, Kahn, Bukusi, Thirumurthy, Rota, Rota, Oketch, Opiyo and Ongeri2021) composed only a small fraction of studies in this review, and trauma did not otherwise appear to factor intentionally into the design nor selection of interventions in our sample. CBT – a therapy strongly recommended by the American Psychological Association’s Clinical Practice Guideline for the Treatment of PTSD (APA, 2017) – was implemented or adapted by several studies in our sample to treat common mental disorders, such as depression and anxiety. However, Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020, which assessed a CBT-based PST intervention for PLWH with trauma histories, noted that the intervention “was not a trauma-specific intervention that can be equated with recognized PTSD treatments.” The study authors also suggested that future studies integrate PTSD screening and management into LHW-delivered care (Verhey et al., Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020).
Our review suggests that there has been little progress in this area since LeGrand et al., Reference LeGrand, Reif, Sullivan, Murray, Barlow and Whetten2015 and that there is an urgent need for more studies to assess the efficacy/effectiveness of trauma-informed interventions for this population and whether LHWs can also deliver them. Promising findings from a recent systematic review and meta-analysis conducted by Connolly et al. (Reference Connolly, Vanchu-Orosco, Warner, Seidi, Edwards, Boath and Irgens2021) indicate that professionally trained lay counselors in LMICs were associated with a significant, medium-sized improvement in mental health symptoms (where PTSD was a primary outcome in most interventions studied) across populations and settings.
Third, several studies in this review assessed suicidality among PLWH and identified concerning levels of suicidal ideation, risk/attempts, and deaths. Notably, Duffy et al. (Reference Duffy, Sharer, Cornman, Pearson, Pitorak and Fullem2017) reported that 61% of PLWH that screened positive for depression and/or anxiety also experienced suicidal ideation, Verhey et al. (Reference Verhey, Chibanda, Vera, Manda, Brakarsh and Seedat2020) found that suicidal ideation was common among PLWH with a history of njodzi (trauma), and Nakimuli-Mpungu et al. (Reference Nakimuli-Mpungu, Musisi, Wamala, Okello, Ndyanabangi, Birungi, Nanfuka, Etukoit, Mayora, Ssengooba, Mojtabai, Nachenga, Harari and Mills2020) observed 25 suicide attempts over a 2-year period following an 8-week intervention for PLWH with depression, including four suicide deaths (two participants in GSP and two participants in the active control group within 6 months post-intervention). These findings align with a recent study by the Penn Center for AIDS Research, which found that PTSD and MDD were predictive of suicide risk (Brown et al., Reference Brown, Majeed, Mu, McCann, Durborow, Chen and Blank2020), and a systematic review of 43 studies across all six WHO regions from 2010 to 2021, which identified a high risk of suicidality among PLWH (Tsai et al., Reference Tsai, Padmalatha, Ku, Wu, Yu, M-H and Ko2022).
Within our sample, MBC administered by a non-specialist nurse depression care manager reduced suicidal ideation over a 12-week intervention period (Pence et al., Reference Pence, Gaynes, Atashili, O’Donnell, Kats, Whetten, Njamnshi, Mbu, Kefie, Asanji and Ndumbe2014; Gaynes et al., Reference Gaynes, Pence, Atashili, O’Donnell, Njamnshi, Tabenyang, Arrey, Whetten, Whetten and Ndumbe2015) and fewer participants reported suicide risk in GSP delivered by trained LHWs than in an active control group (Nakimuli-Mpungu et al., Reference Nakimuli-Mpungu, Musisi, Wamala, Okello, Ndyanabangi, Birungi, Nanfuka, Etukoit, Mayora, Ssengooba, Mojtabai, Nachenga, Harari and Mills2020). Identifying task shifting and sharing intervention strategies that are effective in improving suicide-related outcomes is an important area for future study.
Fourth, recruitment/selection, compensation, supervision, and assessment of LHWs were rarely reported in detail or at all, making it difficult to identify common practices. Important questions remain, such as: (1) if there is any association between the qualifications of LHWs and their ability to deliver services, (2) how compensation affects turnover/quality of care/availability/reliability, (3) if supervision is associated with greater fidelity to intervention protocol, and (4) if training is sufficient. Understanding what these components are in practice and their impact on the success of MHPSS interventions delivered by LHWs is critical to facilitating the translation of this research into broader use.
Finally, budgets and costs associated with implementation beyond participant compensation were rarely disclosed, if at all. Cost-effectiveness was only assessed by Nakimuli-Mpungu et al. (Reference Nakimuli-Mpungu, Musisi, Wamala, Okello, Ndyanabangi, Birungi, Nanfuka, Etukoit, Mayora, Ssengooba, Mojtabai, Nachenga, Harari and Mills2020). We recommend collecting cost data and conducting this type of analysis in future studies, when possible, as it is an essential consideration for implementation in resource-limited settings.
To our knowledge, this scoping review is the first to focus on MHPSS task shifting and sharing interventions for PLWH globally and inclusive of both LMICs and HICs, which is important for facilitating the translation of knowledge and practices used in different contexts and settings. However, our review had several limitations. We only reviewed English-language studies; thus, important work being done in non-English speaking countries would not have been part of our search returns if published in another language. We omitted acceptability and feasibility papers that described interventions aimed at improving mental health among PLWH but did not report on mental health outcomes for the intended recipients or only reported outcomes based on LHWs’ perspectives. The desired outcomes may have been reported elsewhere but those papers did not appear in our search returns and we did not actively search for other papers related to an intervention nor contact authors to find information about data items that were not reported. Therefore, the interventions described in our review should not be considered the full extent of research on this topic but what was accessible.
To best represent and characterize the available literature on this topic, we also did not exclude studies based on methodological quality nor bias, as these assessments are not typical of scoping reviews, which differ from systematic reviews in their aims (Munn et al., Reference Munn, Peters, Stern, Tufanaru, McArthur and Aromataris2018; Tricco et al., Reference Tricco, Lillie, Zarin, O’Brien, Colquhoun, Levac, Moher, Peters, Horsley, Weeks, Hempel, Akl, Chang, McGowan, Stewart, Hartling, Aldcroft, Wilson, Garritty, Lewin, Godfrey, Macdonald, Langlois, Soares-Weiser, Moriarty, Clifford, Tunҫalp and Straus2018). However, we noted these risks as study limitations where applicable. We observed that many studies reviewed were pilots with low sample sizes; there is a need for larger efficacy/effectiveness studies, preferably RCTs and mixed methods. Following intervention participants over longer periods of time could also improve understanding of the long-term mental health benefits of MHPSS services delivered by LHWs.
Conclusion
There has been growing interest in MHPSS task shifting and task sharing interventions for PLWH in recent years. The findings from this review are promising, however, further research is necessary to improve understanding of the impact of these interventions, such as conducting a systematic review and meta-analysis to estimate effect size, comparing intervention models (i.e., by LHW type and by role in service delivery), testing trauma-informed interventions against trauma-related outcomes, and investigating their potential application in preventing suicide. Gaps in knowledge remain concerning the implementation of MHPSS interventions delivered by LHWs, particularly in recruitment/selection, compensation, and training (e.g., supervision and assessment). Assessing the cost-effectiveness of interventions should be prioritized and may be necessary to persuade HICs to further engage in future research and facilitate the adoption of these service delivery models.
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/gmh.2025.10013.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/gmh.2025.10013.
Data availability statement
The studies included in this review are publicly available online through their publishing journals, although some require an institutional membership or a fee to access them.
Acknowledgements
This study was supported by the Woodruff Health Sciences Center Library at Emory University. We thank Sharon Leslie, MS, for her consulting services while conducting the literature search. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of Emory University. The authors have no conflicts of interest to disclose.
Author contribution
The design of this scoping review and its parameters were developed by C.W.K. (PI) and J.M.S. The literature search was conducted by C.W.K. Studies were reviewed and data was extracted by C.W.K. and M.S.C. with J.M.S. serving as a third reviewer. Data analysis was conducted by C.W.K. All review processes were guided by J.M.S., A.S.K., B.A.W-J., and B.G.D. This article was authored by C.W.K. with editorial support from the other authors.
Financial support
This review was supported by the Hope2Action Study grant (R01MH121962) to J.M.S. and A.S.K. from the National Institute of Mental Health of the National Institutes of Health. This work was further supported by the Center for AIDS Research at Emory University (P30 AI050409). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Mental Health or the National Institutes of Health.
Competing interests
The authors have no conflicts of interest to declare.
Ethics
This study was not subject to ethical review as it did not involve the participation of human subjects.
Open access
Comments
ROLLINS SCHOOL OF PUBLIC HEALTH
DEPARTMENT OF BEHAVIORAL, SOCIAL, AND HEALTH EDUCATION SCIENCES
October 9, 2024
Dear Drs. Bass and Chibanda,
We are excited to share our manuscript, “A global scoping review of task shifting and sharing interventions to improve the mental health of people living with HIV/AIDS”, for consideration for publication in Cambridge Prisms: Global Mental Health. We believe our manuscript aligns with the aims and scope of this publication, specifically your call to increase knowledge within the growing discipline of global mental health. Our study seeks to improve understanding of the intervention research that has been done in this area involving the mobilization of lay health workers for the psychological benefit of people living with HIV/AIDS. All study procedures were approved by the Emory University Institutional Review Board. Furthermore, this manuscript is not currently under consideration, nor has it been published elsewhere. The authors have no conflicts of interest to disclose, and are not required to nor have sought pre-submission approval from funders and collaborating organizations.
If you have any questions regarding the manuscript, please feel free to contact me. We look forward to your consideration and hope to see our manuscript in a future issue of your journal.
Sincerely,
Caroline W. Kokubun, MPH
Department of Behavioral, Social, and Health Education Sciences
Rollins School of Public Health
Emory University
1518 Clifton Rd., Room 549
Atlanta, GA 30322
Phone: 503-746-1524
Fax: 404-727-1369
Email: caroline.kokubun@emory.edu