Community engagement

Community engagement is an interactive process that brings together community members, researchers and other key stakeholders, as equal partners in initiatives that benefit the community. Several overlapping strategies including, formative research, community and traditional leaders’ support, partnership with community stakeholders, community sensitisation and education have been identified as critical for community engagement (Musesengwa & Chimbari, Reference Musesengwa and Chimbari2016). It has been argued that while these strategies are essential, they often prioritise communication aimed at defending science from misinformation and seek to educate the community about science, with less emphasis on building trust and establishing resonance with the public (Dudo & Besley, Reference Dudo and Besley2016). This often results in communities not driving the public health agenda (South & Phillips, Reference South and Phillips2014).

In recent years, an engagement strategy that has gained considerable traction and appears to address all the above issues while fostering genuine collective team building between communities, user groups and researchers is the theory of change (ToC) approach (De Silva et al. Reference De Silva, Breuer, Lee, Asher, Chowdhary, Lund and Patel2014). The ToC approach facilitates stakeholder participation, including communities through a collective iterative process, which enables explicit identification of causal pathways, and potential for linking indicators to the design of interventions (Hailemariam et al. Reference Hailemariam, Fekadu, Selamu, Alem, Medhin, Giorgis, DeSilva and Breuer2015; Breuer et al. Reference Breuer, Lee, De Silva and Lund2016). For instance, engaging key stakeholders in participatory planning of mental health services through ToC workshops has been effective in India, South Africa, Nepal, Uganda and Ethiopia through the Programme for Improving Mental Health Care (PRIME) (Hailemariam et al. Reference Hailemariam, Fekadu, Selamu, Alem, Medhin, Giorgis, DeSilva and Breuer2015).

A component not often addressed in ToC workshops is that of scaling up interventions, particularly in the absence of funding.

In Zimbabwe, a deliberate focus on the subject of scale-up post-funding has been included in ToC workshops related to the development of a local intervention – The Friendship Bench (Chibanda et al. Reference Chibanda, Mesu, Kajawu, Cowan, Araya and Abas2011). After successfully engaging both local and international stakeholders through a series of ToC workshops (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Cowan and Lund2016c ), a separate component specifically aimed at exploring the feasibility of scale up was undertaken with the involvement of the community (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Rusakaniko, Cowan and Lund2016d ). Through this scale-up strategy a range of requirements were put forward with specifics on how to address each of them. These included a thorough needs assessment involving key stakeholders and community members, and identifying key priority areas. Other requirements included skills assessment to identify core competencies and current gaps of LHWs, nurses and doctors and consultations to determine funding mechanisms both locally and externally. Our final ToC narrative described key requirements for scale-up, which included early buy-in from key stakeholders particularly the community. Emphasis was placed on the use of existing human resources, obtaining financial support locally, and developing a coherent sustainability plan endorsed by the public, government and private sector (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Rusakaniko, Cowan and Lund2016d ). A particularly powerful component of the ToC approach that seemed to unify participants and contribute towards building consensus for scale-up was the creation of the ToC map, which provided a visual illustration of the causal pathway leading to scale-up. The ToC map further illustrated expected benefits for the community as a result of the scale-up. Community members and LHW described the ToC map as an effective visual tool for building consensus on an evidence-based intervention (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Cowan and Lund2016c ).

EBT approach

Treatment packages for MNS adapted from the West have been effective in several clinical trials conducted in Africa (Mynors-Wallis et al. Reference Mynors-Wallis, Gath, Lloyd-Thomas and Tomlinson1995; Cooper et al. Reference Cooper, Landman, Tomlinson, Molteno, Swartz, Murray, Cooper, Landman, Tomlinson, Molteno, Swartz and Murray2002b ; Bolton et al. Reference Bolton, Bass, Neugebauer, Verdeli, Clougherty, Wickramaratne, Speelman, Ndogoni and Weissman2003; Kaaya et al. Reference Kaaya, Blander, Antelman, Cyprian, Emmons, Matsumoto, Chopyak, Levine and Smith Fawzi2013b ; Nakimuli-Mpungu et al. Reference Nakimuli-Mpungu, Okello, Kinyanda, Alderman, Nakku, Alderman, Pavia, Adaku, Allden and Musisi2013; Chibanda et al. Reference Chibanda, Shetty, Tshimanga, Woelk, Stranix-Chibanda and Rusakaniko2014b ); however, translation of results into routine practice and scale up has been a challenge (Olagunju et al. Reference Olagunju, Ogundipe, Erinfolami, Akinbode and Adeyemi2013; Shidhaye et al. Reference Shidhaye, Lund and Chisholm2015). Clinical trials often do not reflect the real world, particularly in relation to human resources and financial realities on the ground. For instance, interventions in clinical trial settings are often delivered by skilled cadres such as psychologists, Ph.D. candidates, nurses and doctors (Chibanda et al. Reference Chibanda, Cowan, Healy, Abas and Lund2015b ; Verhey et al. Reference Verhey, Chibanda, Brakarsh and Seedat2016), and are usually well funded. Furthermore, ensuring fidelity outside of a clinical trial setting particularly where LHWs are the delivering agent can be challenging.

Translating EBT into local context should therefore take into consideration existing human resources such as LHWs (Patel et al. Reference Patel, Maj, Flisher, De Silva, Koschorke, Prince, Zonal and Member Society2010). Although requiring more training and supervision, LHWs can offer sustainable options through task shifting (Ventevogel, Reference Ventevogel2014) and there is growing evidence supporting the use of this cadre (Cooper et al. Reference Cooper, Landman, Tomlinson, Molteno, Swartz and Murray2002a ; Chibanda et al. Reference Chibanda, Mesu, Kajawu, Cowan, Araya and Abas2011; van't Hof et al. Reference van't Hof, Stein, Marks, Tomlinson and Cuijpers2011; van Ginneken et al. Reference van Ginneken, Tharyan, Lewin, Rao, Meera, Pian, Chandrashekar and Patel2013; Nakimuli-Mpungu et al. Reference Nakimuli-Mpungu, Wamala, Okello, Alderman, Odokonyero, Mojtabai, Mills, Kanters, Nachega and Musisi2015). LHWs often understand the local context and can facilitate a balanced mix between emic and etic features (Patel & Mann, Reference Patel and Mann1997) or cross-cultural methods (Kleinman, Reference Kleinman2003). For instance screening tools should be culturally acceptable and user friendly for LHWs without compromising on sensitivity and specificity (Akena et al. Reference Akena, Joska, Obuku, Amos, Musisi and Stein2012b ). In this regard, several tools when translated into African languages and validated against recognised gold standards have been found to be appropriate for use in Africa, such as the SRQ-20, GHQ-12, the HADS-D and PHQ-9 (Ali et al. Reference Ali, Ryan and De Silva2016). In addition, there is emerging evidence of validated visual tools for illiterate populations (Akena et al. Reference Akena, Joska, Musisi and Stein2012a ).

During the formative phase of the Friendship Bench (Chibanda et al. Reference Chibanda, Verhey, Gibson, Munetsi, Machando, Rusakaniko, Munjoma, Araya, Weiss and Abas2016b ), LHWs were involved in the validation of screening tools. This exercise included translation/back-translation of tools, reviewing the gold standard, protocol training and deciding on a cut-off score based on sensitivity and specificity. We have found that LHW involvement in what is traditionally perceived to be technical work is critical for enhanced buy-in and likelihood of continued use of validated screening tools by LHWs after the study period.

Second, during the development of the psychological intervention, LHW participation contributed to the team's understanding of local concepts of CMD, including indigenous idioms of distress used to navigate through the therapy sessions (Chibanda et al. Reference Chibanda, Cowan, Verhey, Machando, Abas and Lund2016a ). For instance, locally conceived terms for PST such as kuvhura pfungwa (opening the mind through empathy and listening), kusimudzira (uplifting through acknowledging the problems and facilitating a process aimed at selecting a single problem to focus on) and kusimbisa (strengthening through the process of brain storming for solutions) contributed towards the development of a local mental health lexicon and cultivated community ownership of the intervention (Chibanda et al. Reference Chibanda, Cowan, Verhey, Machando, Abas and Lund2016a ). The final intervention, which was tested through a cluster randomised controlled trial (Chibanda et al. Reference Chibanda, Bowers, Verhey, Rusakaniko, Abas, Weiss and Araya2015a ) showed effectiveness of all outcome measures after a 6-month follow-up (submitted). The clinical trial findings prompted the health authorities to recommend scaling up of the Friendship Bench (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Rusakaniko, Cowan and Lund2016d ).

Government endorsed scale up

A common challenge facing MNS researchers is how to integrate or scale up interventions that are effective (Eaton et al. Reference Eaton, McCay, Semrau, Chatterjee, Baingana, Araya, Ntulo, Thornicroft and Saxena2011; Shidhaye et al. Reference Shidhaye, Lund and Chisholm2015). Using task shifting, promoting community-based recovery, and use of training as a continuous way of strengthening clinical competencies are some suggested strategies for integrating and sustaining MNS care in primary care settings (Patel et al. Reference Patel, Maj, Flisher, De Silva, Koschorke, Prince, Zonal and Member Society2010; Ventevogel, Reference Ventevogel2014).

Political buy-in has been described as an equally important component of the process (Collins et al. Reference Collins, Insel, Chockalingam, Daar and Maddox2013; Rebello et al. Reference Rebello, Marques, Gureje and Pike2014). In Zimbabwe, political buy-in for the scaling up of the Friendship bench to over 70 primary health care clinics has largely been driven by local health authorities and the community (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Rusakaniko, Cowan and Lund2016d ). Three key-selling points aimed at achieving the above have been: (i) Emphasis on the added value of treating CMD as a comorbid condition in existing priority public health programmes such as HIV/AIDS, maternal and child health, and other NCDs; (ii) an emphasis on using existing resources; and (iii) the novelty of providing therapy in an outdoor setting on a bench – the Friendship Bench (Chibanda et al. Reference Chibanda, Mesu, Kajawu, Cowan, Araya and Abas2011), thus allowing for ease of access and scale up.

Over a 6-year period these three selling points have been highlighted, particularly during formal and informal meetings with both communities and policymakers.

Policymakers have regularly been invited as speakers or participants where these selling points have been emphasised. Furthermore, beneficiaries of the Friendship Bench intervention with co-morbid conditions such as HIV, hypertension, diabetes have been invited to speak about their experience of receiving care through the Friendship Bench and how this has helped address challenges related to their non-mental health co-morbid medical conditions. In addition, through this approach, partnerships aimed at strengthening both MNS and other priority public health conditions such as HIV/AIDS have been formed with Médecins sans Frontières (MSF) resulting in a win–win situation (Chibanda et al. Reference Chibanda, Verhey, Munetsi, Rusakaniko, Cowan and Lund2016d ).

While the strategies described above have shown positive results at national level, the foundation for these successes has been built through a number of earlier programmes. These include the US Government's Medical Educational partnership Initiative (MEPI) linked award for Improving Mental Health Education and Research in Zimbabwe (http://www.nectar-uz.ac.zw/IMHERZ), which contributed to the establishment of an academic and research development plan for the country. And recently the African Mental Health Research Initiative (AMARI) grant awarded by the Wellcome Trust as part of its DELTAS scheme which aims to Develop Excellence In Leadership Training And Science. Prior to this the Grand Challenges Canada (GCC) transition to scale grant (0763-05) had provided the platform for the scale up. These and other grants have deliberately been designed to integrate MNS care within existing health services with support from the government, particularly through salaries for LHWs and their immediate supervisors.

Challenges

Measuring fidelity of over 300 LHWs delivering the Friendship Bench intervention in over 70 primary care clinics has particularly been a major challenge. The use of mobile phone technology as described in the original Friendship Bench trial protocol (Chibanda et al. Reference Chibanda, Bowers, Verhey, Rusakaniko, Abas, Weiss and Araya2015a ) has mitigated communication challenges between LHWs and their supervisors. With a 90% national mobile phone coverage the use of platforms such as SMS, WhatsApp and Slack have enabled regular fidelity checks, however, the reliability, sustainability and ethical implications of such platforms in the long term will need further research.

LHWs on the Friendship Bench are predominantly trained in screening and managing common mental disorders such as depression and anxiety disorders among adults. However, care packages focusing on children and adolescents are urgently needed because most MNS begin during this period (Patel et al. Reference Patel, Flisher, Hetrick and McGorry2007). Furthermore, other prevalent conditions such as substance use disorders and post-traumatic stress disorders will need to be integrated in the existing care package. How these additional components will impact on the current Friendship Bench structures is not known.

In recent years, a strategy that has been explored as a means of addressing these challenges is the common elements treatment approach (CETA) (Murray et al. Reference Murray, Dorsey, Haroz, Lee, Alsiary, Haydary, Weiss and Bolton2014; Murray & Jordans, Reference Murray and Jordans2016; Pacichana-Quinayaz et al. Reference Pacichana-Quinayaz, Osorio-Cuellar, Bonilla-Escobar, Fandino-Losada and Gutierrez-Martinez2016). The CETA method relies on an internal stepped care approach (Murray et al. Reference Murray, Dorsey, Bolton, Jordans, Rahman, Bass and Verdeli2011), where LHWs are trained in the common elements care of several MNS, which then enables them to switch from one condition to the next based on presentation (Murray et al. Reference Murray, Dorsey, Haroz, Lee, Alsiary, Haydary, Weiss and Bolton2014). However, such an approach may require a lot more training, support and supervision of LHWs and their immediate supervisors.

Finally, social-economic and political instabilities are often the main obstacle to implementing health programmes in Africa and Zimbabwe is no exception. Although salary support of LHWs and their supervisors throughout the country has contributed to the programme's success, a key mitigating factor to the country's social-economic and political volatility has been the strong focus on making the community the key driving force behind the Friendship Bench.