Skip to main content Accessibility help
×
Home

Information:

  • Access
  • Open access

Actions:

      • Send article to Kindle

        To send this article to your Kindle, first ensure no-reply@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Contextualizing and pilot testing the Mental Health Gap Action Programme Intervention Guide (mhGAP-IG) to primary healthcare workers in Kilifi, Kenya
        Available formats
        ×

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Contextualizing and pilot testing the Mental Health Gap Action Programme Intervention Guide (mhGAP-IG) to primary healthcare workers in Kilifi, Kenya
        Available formats
        ×

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Contextualizing and pilot testing the Mental Health Gap Action Programme Intervention Guide (mhGAP-IG) to primary healthcare workers in Kilifi, Kenya
        Available formats
        ×
Export citation

Abstract

Background

Little data exists about the methodology of contextualizing version two of the Mental Health Gap Action Programme Intervention Guide (mhGAP-IG) in resource-poor settings. This paper describes the contextualisation and pilot testing of the guide in Kilifi, Kenya.

Methods

Contextualisation was conducted as a collaboration between the KEMRI-Wellcome Trust Research Programme (KWTRP) and Kilifi County Government's Department of Health (KCGH) between 2016 and 2018. It adapted a mixed-method design and involved a situational analysis, stakeholder engagement, local adaptation and pilot testing of the adapted guide. Qualitative data were analysed using content analysis to identify key facilitators and barriers to the implementation process. Pre- and post-training scores of the adapted guide were compared using the Wilcoxon signed-rank test.

Results

Human resource for mental health in Kilifi is strained with limited infrastructure and outdated legislation. Barriers to implementation included few specialists for referral, unreliable drug supply, difficulty in translating the guide to Kiswahili language, lack of clarity of the roles of KWTRP and KCGH in the implementation process and the unwillingness of the biomedical practitioners to collaborate with traditional health practitioners to enhance referrals to hospital. In the adaptation process, stakeholders recommended the exclusion of child and adolescent mental and behavioural problems, as well as dementia modules from the final version of the guide. Pilot testing of the adapted guide showed a significant improvement in the post-training scores: 66.3% (95% CI 62.4–70.8) v. 76.6% (95% CI 71.6–79.2) (p < 0.001).

Conclusion

The adapted mhGAP-IG version two can be used across coastal Kenya to train primary healthcare providers. However, successful implementation in Kilifi will require a review of new evidence on the burden of disease, improvements in the mental health system and sustained dialogue among stakeholders.

Background

The mental health Gap Action Programme (mhGAP) was first published by the World Health Organization (WHO) in 2008, in response to the recognition of a need to scale up delivery of mental health services by utilising available resources within a health system (WHO, 2008). The mhGAP Intervention Guide (mhGAP-IG) for mental neurological and substance use disorders (MNSD) is a protocol for clinical decision making that was developed through a systematic review of evidence and international consultations to aid in the implementation of the mhGAP (WHO, 2015a). Since its inception, over 90 countries have taken up its use, although literature is dominated by only a few countries as reported in a systematic review on the implementation of the mhGAP (Keynejad et al., Reference Keynejad, Dua, Barbui and Thornicroft2018). This review also indicated that the largest proportion of studies (46%) used the guide for training courses, while other uses included clinical implementations, formation of economic models and as rating scales. From this review, only three studies reported local adaptation of the guide and of these, only one study in Nigeria provided a step by step account of the mhGAP-IG contextualisation (Abdulmalik et al., Reference Abdulmalik, Kola, Fadahunsi, Adebayo, Yasamy, Musa and Gureje2013). Some of the lessons learnt from Nigeria that informed contextualisation of the guide to their setting included the large gaps in the training curriculum of primary healthcare workers and the importance of establishing a clear referral pathway with proper clinical supervision for the trained health workers. However, this study contextualised version one of the guide (WHO, 2010) which have since been revised (WHO, 2016a). More recently, an evaluation of the implementation of the revised guide in Fiji reported that although the guide were deemed valuable and easy to use by the healthcare providers, there was need to improve planning and leadership in the implementation (Charlson et al., Reference Charlson, Chang, Kubuabola, Schess, Latu, Hunter, Tukana, Qaloewai and Shidhaye2019). However, data on adaptation of the revised guide from low- and middle-income countries are lacking. Lack of data on local adaptation of the revised guide creates a missed opportunity for exploring the real-world challenges of implementing the guide such as language barriers and poor infrastructure which may explain the persistently large treatment gaps (Fekadu et al., Reference Fekadu, Medhin, Lund, Desilva, Selamu, Alem, Asher, Birhane, Patel, Hailemariam, Shibre, Thornicroft, Prince and Hanlon2019) in countries that have successfully implemented these guide.

In Kenya, version 1 of the mhGAP-IG has been contextualised (Mutiso et al., Reference Mutiso, Gitonga, Musau, Musyimi, Nandoya, Rebello, Pike and Ndetei2018), and used for clinical practice in both rural and urban areas of eastern Kenya with good specificity (86%), but low sensitivity (46%) (Musyimi et al., Reference Musyimi, Mutiso, Ndetei, Henderson and Bunders2017). The low sensitivity may be attributed to lack of recognition of some disorders by local communities (Bitta et al., Reference Bitta, Kariuki, Gona, Abubakar and Newton2019). However, there are no studies on adaptation and implementation of version two of the guide mhGAP-IG (mhGAP-IG V2). The present study adapted and pilot tested guide mhGAP-IG V2 in coastal Kenya. The pilot testing was conducted among primary healthcare providers to assess the ease of use of the guide and the changes in level of knowledge after training.

Methods

The contextualisation process was initiated in 2016 by the Kenya Medical Research Institute- Wellcome Trust Programme (KWTRP) in collaboration with the Department of Health of the County Government of Kilifi. The KWTRP is a research institute, formed in 1989 as a collaboration between the Kenya Medical Research Institute, Wellcome Trust and Oxford University from the UK. The institute conducts medical research of both communicable and non-communicable diseases in an established demographic and surveillance system (Scott et al., Reference Scott, Bauni, Moisi, Ojal, Gatakaa, Nyundo, Molyneux, Kombe, Tsofa, Marsh, Peshu and Williams2012). The department of health in Kilifi was established in 2010 after the implementation of the devolved system of government in Kenya, which saw the establishment of 47 independent administrative units called counties, including Kilifi (KIlifi, 2019). Permission to conduct this study was obtained from the Scientific Ethics and Review Unit of the Kenya Medical Research Institute under protocol number 038/3260. The steps of the contextualisation process are described below.

Step I: situation analysis of the state of mental health services in Kilifi Kenya

To understand the Kenyan context, we conducted a narrative synthesis of publicly available data. Our data sources included: (i) the Ministry of Health, where we obtained data on the structure of the formal health system, expenditure on health and human resource for health (GOK, 2014, 2015); (ii) published peer reviewed articles for data about alternative pathways to mental healthcare in Kenya (Mbwayo et al., Reference Mbwayo, Ndetei, Mutiso and Khasakhala2013; Burns and Tomita, Reference Burns and Tomita2015; Musyimi et al., Reference Musyimi, Mutiso, Loeffen, Krumeich and Ndetei2018); (iii) Kenya National Bureau of Statistics for Kenya's demographics (KNBS, 2009); (iv) the Constitution Of Kenya for data about mental health legislation (GOK, 2015) and (v) the Mental Health Atlas and published literature for data about the burden of mental illnesses in Kenya (Jenkins et al., Reference Jenkins, Njenga, Okonji, Kigamwa, Baraza, Ayuyo, Singleton, Mcmanus and Kiima2012; Kariuki et al., Reference Kariuki, Chengo, Ibinda, Odhiambo, Etyang, Ngugi and Newton2015; WHO, 2015b; Otiende et al., Reference Otiende, Abubakar, Mochamah, Walumbe, Nyundo, Doyle, Ross, Newton and Bauni2017; Bitta et al., Reference Bitta, Bakolis, Kariuki, Nyutu, Mochama, Thornicroft and Newton2018). We then conducted a situational analysis of the state of mental health systems in Kilifi using the brief version of the World Health Organization's Assessment Instrument for Mental Health Systems (WHO-AIMS) (WHO, 2009). This instrument quantitatively assesses a mental health system based on six main domains: (i) policy and legislative framework; (ii) mental health services; (iii) mental health in primary care; (iv) human resources; (v) public education and links with other sectors; and (vi) monitoring and research. Data were either binary (i.e ’yes or “no’) or numerical (e.g. the number of staffs, or the year a policy was developed etc). The data were entered in a standardised WHO- AIMS 2.2 Excel spreadsheet. Descriptive statistics were conducted using Stata version 13 (Stata Corp, Texas, USA) and a binomial distribution was used to generate confidence intervals.

Step II: strategic stakeholders meeting to promote involvement

The aim of these meetings was to discuss the feasibility of implementing the mhGAP-IG in Kilifi County. The meetings, which lasted about an hour, were conducted with the National Director of the Department of Mental Health Substance Use, County Director of Health and County Head of Community Strategy. These stakeholders were identified through consultation with KWTRP's community liaison team as well as through previous engagements between KWTRP's researchers and the County government. At both the county and national level, we asked the stakeholders for potential facilitators and barriers to the successful implementation of the mhGAP-IG in Kilifi. Additionally, we asked the County level stakeholders about their anticipated roles and those of the research institute (KWTRP) in implementation. These were informal introductory meetings and consent was not obtained to record the conversations or to quote the findings verbatim, but we obtained oral consent to take notes during the meetings and analyse the findings. We used a constant comparison approach with content analysis to summarise the key findings of the discussions.

Step III: focus group discussions with primary healthcare providers, traditional health practitioners and expert meetings to contextualise the guide

Focus group discussions were conducted with primary healthcare providers and traditional health practitioners to understand the local terms, presentations and management of the priority MNSD in Kilifi. The discussions aimed to provide a qualitative situational analysis of the understanding and management of the disorders in Kilifi. This would inform adaptation of the mhGAP-IG training materials, (e.g. to make role plays and examples used during trainings locally relevant). Traditional health practitioners were included because previous studies in Kilifi (Kendall-Taylor et al., Reference Kendall-Taylor, Kathomi, Rimba and Newton2009) show that they are an integral part of the pathways to healthcare among people with the disorders in Kilifi. Purposive sampling was used to identify primary healthcare providers who were selected to ensure maximum variations in age, sex and geographical locations. Traditional health practitioners were identified through a database established in a previous study (Kendall-Taylor et al., Reference Kendall-Taylor, Kathomi, Rimba and Newton2008). Informed consent was obtained from all participants. Using vignettes which were adapted from a primary healthcare providers' mental health services handbook (Patel, Reference Patel2018) to represent each of the priority conditions, an experienced moderator who was fluent in both English, Kiswahili and Kigiriama (the local language) asked participants the following questions: (i) What are the local terms for these disorders? (ii) How do people with these disorders present in the community? (iii) How do you manage these disorders? An example of a vignette for psychosis is provided in Table 1. The questions were used as a framework for content analysis where two investigators independently coded the data using an inductive approach and then compared and discussed their findings until consensus was reached.

Table 1. An example of a vignette used for psychosis

To contextualise the guide, the mhGAP-IG contextualisation questionnaire (Abdulmalik et al., Reference Abdulmalik, Kola, Fadahunsi, Adebayo, Yasamy, Musa and Gureje2013) was used by specialists who included three psychiatric nurses, a clinical psychologist, two epilepsy specialists and two clinical officers with mental health experience. The master chart, which will be displayed in facilities and will be used as the first assessment tool, was translated to Kiswahili (Appendix 1) which is the lingua franca 7 in Kenya. The specialists reviewed the recommendations of the guide against the availability and levels of skill and competence of the primary healthcare providers who would implement this guide and against the availability of resources at present and their likelihood of being available in the future. Only essential changes were made. These were changes that were unlikely to improve soon or which were not relevant to the setting. For instance, we revised the drugs recommended in the guide to align with those in the County's essential drugs list Table 2.

Table 2. Availability of essential medicines in Kilifi County based on Kenya's essential medicines list for mental illnesses

a Drops with dosing pipette.

b Use only in patients refractory to, or intolerant of, 1st generation antipsychotic.

c Use only in patients refractory to, or unable to tolerate, other antipsychotics.

d Only use in patients >8 years.

e Only use in anxiety with agitation.

f Only use in the management of alcohol dependence.

g Use under close supervision within substance dependency treatment programmes.

h Ascorbic acid 500 mg, nicotinamide 160 mg, pyridoxine hydrochloride 50 mg, riboflavin 4 mg, thiamine hydrochloride 250 mg/7 ml.

i Ascorbic acid 500 mg, nicotinamide 160 mg, pyridoxine hydrochloride 50 mg, riboflavin 4 mg, thiamine hydrochloride 250 mg/10 ml.

Step IV: pilot testing of adapted guide among primary healthcare providers and post-pilot feedback workshop among specialists as well as primary healthcare providers

The pilot sites were three primary healthcare facilities within Kilifi County. Convenience sampling was used to select participants of the pilot to represent different cadres, age groups and sex. A 1-week training on the use of the guide was facilitated by a total of seven trainers with each trainer facilitating one module in the guide. The mhGAP-IG training manuals (WHO, 2016b) guided the training sessions. The training included both theoretical skills of the common signs and symptoms and management modalities of the priority disorders as well as hands on skills such as communication and physical examination skills as specified in the training manuals. For each module, pre and post-training knowledge questionnaires, which are part of the training materials, were administered to the healthcare providers to measure their changes in level of knowledge. The pre- and post-training scores on levels of knowledge were compared using a Wilcoxon signed-rank test. A feedback meeting among the specialists and representatives of the trainees was conducted to review the training.

Results

Step I: situation analysis of the state of mental health services in Kilifi Kenya

The Kenyan context

Kenya is an East African country with a population of approximately 47.6 million people and about 44 officially recognised tribes although there are about 70 different ethnic groups. Kenya's basic governance system is two tiered and includes a National level, headed by a president and a County level headed by a governor. There are 47 counties in Kenya and each independently sets priorities and operations in terms of revenue generation and resource allocation. The National Government is responsible for national referral facilities and health policy and legislation; while the County government is responsible for human resource for health and infrastructural development and support. Kenya has an essential healthcare package which comprises five levels namely: community level, (dispensaries or clinics) health centres, primary hospitals, secondary hospitals and tertiary facilities. Approximately 60% of all health services are provided by the public health sector while the remainder 40% are provided by private and faith-based sectors. The ratio of healthcare workforce per 100 000 population is 20.4 for doctors, 158.2 for nurses and midwives, and 5.0 for other healthcare providers such as nutritionists, occupational therapists and counsellors. Additionally, there is an informal health sector composed of traditional health practitioners and faith healers mainly in rural areas as well as unlicensed biomedical practitioners mostly based in urban areas. As a result, the pathways to healthcare are tortuous and sometimes shift between formal and informal care, especially for mental and neurological disorders.

Kenya contributes data to the Mental Health Atlas project through the Department of Mental Health and Substance Use. According to these data, in 2017, mental illnesses contributed approximately 2467 disability-adjusted life years (DALY) per 100 000 population and depressive disorders were one of the top ten leading causes of DALY. Mental health infrastructure is poorly developed, and disproportionately distributed in favour of major cities situated in a few counties. There is only one standalone national referral mental specialist hospital in Kenya, 15 psychiatric in-patient units in general hospitals with about 12 000 patients, 97.8% of whom are involuntarily admitted. There are 30 mental health outpatient units, which receive approximately 741 visits per 100 000 population annually. The mental health expenditure per person per year is approximately 0.03 dollars, which is only 0.01% of the total national government expenditure on health. There are 92 psychiatrists, four of whom are child and adolescent mental health specialists. The mental health workforce is about 0.19 per 100 000 population. The Kenya Mental Health Act of 1989 is the sole stand-alone law for mental health in Kenya. It is outdated with sections that are inconsistent with the current international standards such as lack of an independent oversight authority that assess compliance of this legislation with international laws on human rights. Kenya has a mental health policy that was developed in 2015 and includes a mental health plan with specific indicators against which implementation can be evaluated. Key among the priority areas is human resource development and one strategic action relevant to this study is the provision of in-service training for health workers, to build capacity for management of mental illnesses.

The Kilifi context

Kilifi County is one of six coastal counties of Kenya, with the others being Mombasa, Kwale, Tana River, Lamu and Taita Taveta. It is a largely rural area on the coast of Kenya and comprises six sub-counties namely, Kilifi, Ganze, Malindi, Magarini, Rabai and Kaloleni and covers a total surface area of 12 609.7 km2 (2.1% of the country). The population of the County is estimated to be 1.4 million. The main communities include the Mijikenda, Bajuni, Swa-Arab, Indian, European decent and other migrant Kenyan communities. The main economic activities are agriculture (cashew nuts and horticulture), fishing, forestry and mining. Kilifi County has 156 public health facilities (one teaching and referral hospital, four sub-County hospitals, 14 health centres and 137 dispensaries) and 151 private facilities including a neurology clinic run by the Neuroscience Department of the KEMRI-Wellcome Trust Research Programme in collaboration with Kilifi county Department of Health. This clinic provides free follow-up care mainly for research participants with epilepsy and neurodevelopmental disorders but also for members of the public who are referred from other facilities. There are three public outpatient facilities attached to general hospitals that offer psychiatric outpatient care which manage about 180 psychiatric patients per 100 000 populations based on the number of outpatient visits only. The doctor-patient ratio in Kilifi County is low at 10 per 100 000 population whereas the nurse-patient ratio is 40 per 100 000 population compared to WHO's recommendation of 333 per 100 000 population and 250 per 100 000 population, respectively. Health services are managed by the county Health Services Department, which reports to the national Ministry of Health. The county director of health works closely with the County health management team and provides leadership to the department of health. Kilifi County Hospital serves as the mainstream teaching and referral hospital in the County. It provides both outpatient and inpatient services and has a total bed capacity of 184.

Results of the analysis using WHO-AIMS have been presented extensively elsewhere (Bitta et al., Reference Bitta, Kariuki, Chengo and Newton2017). In summary, there were no regional policy and legislative frameworks specifically for mental health, hence prompting the coastal region to use national laws and regulations. Infrastructure for specialised mental healthcare was limited to three psychiatric outpatient units attached to general hospitals. In-patient psychiatric care was provided in general hospitals based on the availability of beds. Based on Kenya's essential drugs list for managing psychiatric conditions, less than half of the drugs were available in Kilifi county as shown in Table 2. The supply of drugs was erratic. Specialised human resource for mental health comprised three psychiatric nurses and one clinical officer for a population of approximately 1.2 million people, giving a provider patient ratio of 1:300 000. There are no efforts to integrate mental health with primary care and monitoring and research in the region is mainly conducted by KWTRP. Community surveys have been conducted for neurological disorders, mainly epilepsy and acute seizures, some mental illnesses such as behavioural and emotional problems in children and suicide as a fatal outcome of mental illness.

Step II: strategic stakeholders meeting to promote involvement

As we did not obtain consent to record and use verbatim data obtained during these meetings, we do not provide quotations in this section. Although we had originally aimed to reach three key stakeholders, we conducted a total of six meetings over a period of 1 year with the following stakeholders :(i) National Director of the Department of Mental Health and Substance Use; (ii) Chief Officer of Health in Kilifi County; (iii) the County Head of Community Strategy; (iv) the County Coordinator of non-communicable diseases and research and (v) two medical superintendents from Kilifi County Hospital and Malindi Sub-County Hospital. These stakeholders were identified by snowballing from the three key stakeholders that we had initially targeted. With regards to the barriers to successful implementation, the stakeholders identified low levels of human resource at the primary care facilities, lack of skilled personnel to offer supervision to the trained healthcare providers and lack of clarity on the roles of the research institute and the County government. Potential facilitators of successful implementation included the positive 30 year working relationship between the County government of Kilifi and the research institute (KWTRP). Additional findings from the County and national levels and key outputs of these meetings were summarised in Table 3.

Table 3. A summary of the stakeholder meetings and key outputs of the meeting

a Since this was the first meeting, we did not tape-record the conversations as this may have created a communication barrier between the researchers and respondents.

Step III: focus group discussions with primary healthcare providers, traditional health practitioners and expert meetings to contextualise the guide

In total, eight focus group discussions, four for each group, were conducted at which point information saturation was achieved. In total there were 60 participants and each group had seven or eight participants. There were local terms for all the priority disorders except dementia which was not recognised by either of the groups. Child and adolescent mental and behavioural problems were not recognised as mental illnesses but as consequences of poor parenting. Most of the common presentations listed in the mhGAP were identified by the participants. However, among the traditional health practitioners, names of unrelated conditions were used simultaneously (e.g. tetanus and epilepsy), suggesting potential misdiagnosis. Management for the disorders among traditional health practitioners depended on the perceived cause of illness and on their expertise. It included referral to biomedical facilities in cases where a patient was not responding to traditional treatment. Management among primary healthcare providers ranged from no care at all, to unstructured counselling to medication to alleviate acute symptoms (e.g. diazepam for agitation in a patient with depression). A detailed summary of these findings including quotations to reflect these findings is published elsewhere (Bitta et al., Reference Bitta, Kariuki, Gona, Abubakar and Newton2019). During translation of the master chart, there were no Kiswahili equivalent words for some disorders such as dementia and child and adolescent mental and behavioural disorders. Only the most essential changes were made to the guide as shown in Table 4. The changes included clarifications of the referral pathways to indicate that specialists in the setting referred to psychiatric nurses and clinical officers specialised in psychiatry as well as adaptation of suggested pharmacotherapy to tally with the drugs that are available in the County (e.g. fluoxetine was replaced with amitriptyline).

Table 4. Examples of suggested contextualisation of the mhGAP-IG

a There were no suggested changes.

Step IV: pilot testing of adapted guide among primary healthcare providers and post-pilot feedback workshop among specialists as well as primary healthcare providers

In total, 16 primary healthcare providers were trained. They included one medical officer, two clinical officers and 13 nurses of different cadres. Overall, there was a significant difference in the mean knowledge scores pre-training v. post-training scores 47.0% (s.d. = 3.9) v. 60.8% (s.d. = 4.9) p < 0.001. In the individual modules, there was significant improvement in knowledge for all modules except for the epilepsy and essential care and practice modules as shown in Table 5. The stakeholders recommended exclusion of the child and adolescent and dementia modules. During the feedback workshop, the final version of the guide was approved after including the changes highlighted in Table 4.

Table 5. Pre- and post-training scores of the pilot training among primary healthcare providers

Discussion

The process of contextualising mhGAP-IG V2 in one of the poorest regions of Kenya has shed light on the gaps and opportunities that a primary healthcare system offers in a devolved system of governance. This study also offers unique insight into the process of collaboration between a research programme (KWTRP) and the County government in adopting and implementing the guide. Our study identified strengths such as willingness of primary healthcare providers to participate in the intervention and challenges such as an excessively strained human resource that were previously identified in other settings in Kenya (Mutiso et al., Reference Mutiso, Gitonga, Musau, Musyimi, Nandoya, Rebello, Pike and Ndetei2018). We also identified opportunities and challenges that have not been previously reported in the contextualisation of the mhGAP-IG in Kenya, such as a good pre-existing collaboration between a research institute and a County government. We discuss how lessons learnt from each stage informed contextualisation.

Situation analysis of the state of mental health services in coastal Kenya

The situational analysis contributed knowledge to the already available evidence that mental health systems in many low-and middle-income countries are poorly resourced (Roberts et al., Reference Roberts, Mogan and Asare2014; Luitel et al., Reference Luitel, Jordans, Adhikari, Upadhaya, Hanlon, Lund and Komproe2015; Spagnolo et al., Reference Spagnolo, Champagne, Leduc, Melki, Guesmi, Bram, Guisset, Piat, Laporta and Charfi2018). This may have direct negative consequences on the implementation efforts of the mhGAP-IG. For instance, even with correct diagnosis, management of MNSD may fail due to erratic supply of drugs which will lower patients' adherence to medication and lead to poor outcomes. Additionally, lack of trained healthcare providers to identify MNSD and experts to manage MNSD significantly disrupts the referral chain recommended in the mhGAP-IG. For this setting, the human resource limitations can be addressed by ensuring that ‘gate keepers’ of health (i.e. healthcare providers at peripheral facilities) are trained on correct diagnosis and management of priority MNSD so that the few available experts can focus on managing complicated cases, rather than primary diagnosis. In the Kenyan system where drug supply is based on demand (Aronovich and Kinzett, Reference Aronovich and Kinzett2001) the creation of demand through increasing case detection and drug prescription may increase drug supply.

Strategic stakeholders meeting to promote involvement

Studies on implementation of interventions (Frieden, Reference Frieden2014; Spiel et al., Reference Spiel, Schober and Strohmeier2018), including the mhGAP-IG (Dos Santos et al., Reference Dos Santos, Cumbe, Gouveia, De Fouchier, Teuwen and Dua2019) have demonstrated that gaining stakeholder support is not only necessary for successful implementation of interventions, but also for adoption of evidence generated from the intervention into policy and practice. The engagements in this study provided vital linkages to strategic implementation partners. For instance, we obtained crucial linkages to the national government, the county hospitals and the community health teams which facilitated wider participation in the process. Additionally, the devolved system of governance made stakeholder engagement easier, since both health and administrative services were concentrated within a small geographic region. The engagements identified a key challenge unique to this context: there was a lack of clarity on the roles and frameworks of operation between the County government and the KWTRP (which was facilitating this process). This may have arisen as it is the first time KWTRP was facilitating the implementation of a mental health intervention that directly involved County government health staff on a large scale. However, this was resolved after clarification by the KWTRP that they would only play a facilitatory role. The positive longstanding working relationship between KWTRP and Kilifi County officials that saw KWTRP establish a health and demographic surveillance system (Scott et al., Reference Scott, Bauni, Moisi, Ojal, Gatakaa, Nyundo, Molyneux, Kombe, Tsofa, Marsh, Peshu and Williams2012) may have contributed to collaboration. Although the nature of this collaboration is beyond the scope of this paper, it is important to note that expectations from different partners in the implementation process may influence the outcome of the process. For this setting, before implementation, there will be need for further engagement with the stakeholders through theory of change workshops (De Silva et al., Reference De Silva, Breuer, Lee, Asher, Chowdhary, Lund and Patel2014) to address barriers that may hinder task sharing such as competing responsibilities for management of other illnesses and low human resource.

Focus group discussions with primary healthcare providers, traditional health practitioners and expert meetings to contextualise the guide

Most priority disorders were recognised by both traditional health practitioners and primary healthcare provider. However, dementia and child and adolescent mental and behavioural problems were not regarded as priority conditions by primary healthcare providers and they were not recognised by traditional health practitioners. Lack of recognition of these disorders may be due to lack of knowledge about these illnesses rather than due to a low burden of disease as shown by a 13% prevalence of behavioural and emotional problems in children in Kilifi (Kariuki et al., Reference Kariuki, Abubakar, Kombe, Kazungu, Odhiambo, Stein and Newton2017). Exclusion of this module by countries that have implemented the guide including South Africa, Nepal and Uganda (Lund et al., Reference Lund, Tomlinson and Patel2016) created a missed opportunity of evaluating the barriers and facilitators of successful implementation, especially in sub-Saharan Africa.

Inclusion of traditional health practitioners in the contextualisation process promoted the process of referral because evidence from this setting suggested that traditional health practitioners are an integral part of the healthcare system (Kendall-Taylor et al., Reference Kendall-Taylor, Kathomi, Rimba and Newton2008). The unidirectional willingness of the traditional health practitioners to cooperate with biomedical practitioners has been reported in other parts of sub-Saharan Africa (Akol et al., Reference Akol, Moland, Babirye and Engebretsen2018). Traditional health practitioners play a critical role in the referral pathway and can offer psychosocial support. It is critical to address the underlying causes of the strained relationship with the biomedical practitioners, throughout the implementation process.

During the translation of the master chart, which is part of the guide, there were no equivalent Kiswahili words for most of the mental and neurological illnesses. This resulted in the use of phrases which lengthened the guide. The delicate balance between a concise guideline and fidelity of the translations highlighted the salient difficulties of cross-cultural psychiatry, specifically the translation of materials to local languages in many low and middle-income countries. However, because it is common to observe evolution in the terms that describe mental and emotional experiences, this situation is likely to improve over the years as mental health awareness increases. The translation of the guide into Kiswahili may therefore improve in the long term.

Pilot testing of adapted guide and feedback workshop with the national and county government as well as implementation of suggested changes

Successful pilot testing of the guide among the healthcare providers indicated that the guide can be used in coastal Kenya for diagnosis and management of common mental disorders. Like in other settings (Akol et al., Reference Akol, Nalugya, Nshemereirwe, Babirye and Engebretsen2017, Robles et al., Reference Robles, Lopez-Garcia, Miret, Cabello, Cisneros, Rizo, Ayuso-Mateos and Medina-Mora2020), there was an improvement in the post-test scores in most of the modules except for the epilepsy module. An improvement in the post-training scores of the adapted guide showed that the guide improved the level of knowledge of the healthcare providers about diagnosis and management of mental and substance use disorders. There was no improvement in the epilepsy module perhaps because the study population was involved in epilepsy research studies (Mbuba and Newton, Reference Mbuba and Newton2009; Ibinda et al., Reference Ibinda, Mbuba, Kariuki, Chengo, Ngugi, Odhiambo, Lowe, Fegan, Carter and Newton2014; Kariuki et al., Reference Kariuki, Chengo, Ibinda, Odhiambo, Etyang, Ngugi and Newton2015)and had higher baseline levels of knowledge about the disorder as shown by the high pre-training scores. Feedback from the specialists provided useful insights on the use of a step-wise approach in training healthcare providers based on their qualifications and level of experience.

Although the specialists recommended the exclusion of child and adolescent and dementia modules from the final version, evidence that emerged after the adaptation process showed that the burden of child and adolescent mental and behavioural problems is high (Kariuki et al., Reference Kariuki, Abubakar, Kombe, Kazungu, Odhiambo, Stein and Newton2017) and this could be applicable to dementia. Therefore, future implementation plans will consider this new evidence and will include adaptation of these module. This will provide a unique opportunity for the evaluation of the implementation of these modules, especially in sub-Saharan Africa where very few studies have implemented this module (Akol et al., Reference Akol, Nalugya, Nshemereirwe, Babirye and Engebretsen2017).

Study limitations

We did not adapt the psychological interventions' components of the guide for use by lay health workers therefore patients requiring specialised services such as cognitive behavioural therapy, will still need to be referred to specialists. However, at present, there is a lot of effort by the government (GOK, 2019) to improve mental healthcare in the country and we anticipate that the human resource for mental health is likely to improve including increase in the number of specialists capable of providing specialised psychological interventions such as cognitive behavioural therapy. Additionally, to utilise the robust community strategy of provision of health services in Kilifi County, the study team plans to pilot psychological interventions for lay health workers, that are currently under development.

The pilot did not evaluate knowledge retention among primary healthcare workers after training. However, implementation of the guide, which will be facilitated by KWTRP, will include close monitoring and supervision by master trainers who will be trained healthcare providers within the County.

Conclusions

The process of contextualising the mhGAP-IG in coastal Kenya involved extensive planning and consultation at multiple levels of government with numerous stakeholders. The key findings from the process included the challenges of different levels of experience in MNSD among healthcare staff , difficulties in translation due to lack of words in Kiswahili which is the local language and policy limitations which affected supply of drugs and availability of specialised human resource. Successful pilot testing of the adapted guide provided a basis for large-scale implementation of the mhGAP-IG in coastal Kenya, which is currently in the planning phases. However, the process demonstrated the need to continue engagements with policy makers to improve resource allocation towards mental health and sustain the implementation efforts. Implementation will also require enough resources to provide sustained training and supervisory support to the primary healthcare providers for successful outcomes. Lastly, barriers to access of care such as stigma and other social determinants of mental health were not addressed during the consultative meetings and these may hinder the uptake of services during the implementation process. Participants of the consultative process may have excluded these barriers in the discussions due to social desirability associated with the use of qualitative methods of data collection. However, the research team from KEMRI Wellcome Trust is currently conducting research to provide baseline data on the levels of stigma among people with MNSD.

Acknowledgements

We would like to acknowledge the National and County government representatives from the department of health for their support and for providing linkages with key stakeholders in the implementation process especially Dr Simon Njuguna who is the current national director of the department of mental health and substance use, Mr Samuel Tunje from the Malindi sub-County substance use rehabilitation clinic for providing helpful expert inputs at different stages of the contextualisation process. We would also like to thank members of the mhGAP office at the World Health Organization for sending us the most up to date training materials and for providing technical support during the trainings. This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Conflict of interest

None.

References

Abdulmalik, J, Kola, L, Fadahunsi, W, Adebayo, K, Yasamy, MT, Musa, E and Gureje, O (2013) mhGAP contextualization questionnaire [Online]. Available at https://journals.plos.org/plosmedicine/article/file?type=supplementary&id=info:doi/10.1371/journal.pmed.1001501.s001 (Accessed 14 November 2019).
Akol, A, Moland, KM, Babirye, JN and Engebretsen, IMS (2018) “We are like co-wives”: traditional healers' views on collaborating with the formal child and adolescent mental health system in Uganda. BMC Health Services Research 18, 258.
Akol, A, Nalugya, J, Nshemereirwe, S, Babirye, JN and Engebretsen, IMS (2017) Does child and adolescent mental health in-service training result in equivalent knowledge gain among cadres of non-specialist health workers in Uganda? A pre-test post-test study. International Journal of Mental Health Systems 11, 50.
Aronovich, DG and Kinzett, S (2001) Kenya: Assessment of the health commodity supply chain and the role of KEMSA. John Snow, Inc 1616 North Fort Meyer rive, Arlington, VA 22209, USA.
Bitta, MA, Bakolis, I, Kariuki, SM, Nyutu, G, Mochama, G, Thornicroft, G and Newton, C (2018) Suicide in a rural area of coastal Kenya. BMC Psychiatry 18, 267.
Bitta, MA, Kariuki, SM, Chengo, E and Newton, C (2017) An overview of mental health care system in Kilifi, Kenya: results from an initial assessment using the World Health Organization's Assessment Instrument for mental health systems. International Journal of Mental Health Systems 11, 28.
Bitta, MA, Kariuki, SM, Gona, J, Abubakar, A and Newton, C (2019) Priority mental, neurological and substance use disorders in rural Kenya: traditional health practitioners' and primary health care workers' perspectives. PLoS ONE 14, e0220034.
Burns, JK and Tomita, A (2015) Traditional and religious healers in the pathway to care for people with mental disorders in Africa: a systematic review and meta-analysis. Social Psychiatry and Psychiatric Epidemiology 50, 867877.
Charlson, F, Chang, O, Kubuabola, I, Schess, J, Latu, C, Hunter, E, Tukana, I, Qaloewai, S and Shidhaye, R (2019) Implementation of the Mental Health Gap Action Programme (mhGAP) within the Fijian healthcare system: a mixed-methods evaluation. International Journal of Mental Health Systems 13, 4343.
De Silva, MJ, Breuer, E, Lee, L, Asher, L, Chowdhary, N, Lund, C and Patel, V (2014) Theory of change: a theory-driven approach to enhance the medical research council's framework for complex interventions. Trials 15, 267.
Dos Santos, PF, Cumbe, V, Gouveia, ML, De Fouchier, C, Teuwen, D and Dua, T (2019) Implementation of mhGAP in Mozambique: integrating epilepsy care into the primary health care system. International Journal of Mental Health Systems 13, 36.
Fekadu, A, Medhin, G, Lund, C, Desilva, M, Selamu, M, Alem, A, Asher, L, Birhane, R, Patel, V, Hailemariam, M, Shibre, T, Thornicroft, G, Prince, M and Hanlon, C (2019) The psychosis treatment gap and its consequences in rural Ethiopia. BMC Psychiatry 19, 325.
Frieden, TR (2014) Six components necessary for effective public health program implementation. American Journal of Public Health 104, 1722.
GOK (2014) Health Sector Human Resources Strategy 2014–2018 [Online]. Kenya: Ministry of Health, Government of Kenya. Available at http://www.health.go.ke/wp-content/uploads/2016/04/Kenya-HRH-Strategy-2014-2018.pdf (Accessed 22 February 2019).
GOK (2015) Kenya Mental Health Policy 2015-2030. Available at https://healthservices.uonbi.ac.ke/sites/default/files/centraladmin/healthservices/Kenya%20Mental%20Health%20Policy.pdf (Accessed 22 February 19).
GOK (2019) Taskforce formed to assess Mental health issues of Kenyans.
Ibinda, F, Mbuba, CK, Kariuki, SM, Chengo, E, Ngugi, AK, Odhiambo, R, Lowe, B, Fegan, G, Carter, JA and Newton, CR (2014) Evaluation of Kilifi epilepsy education programme: a randomized controlled trial. Epilepsia 55, 344352.
Jenkins, R, Njenga, F, Okonji, M, Kigamwa, P, Baraza, M, Ayuyo, J, Singleton, N, Mcmanus, S and Kiima, D (2012) Prevalence of common mental disorders in a rural district of Kenya, and socio-demographic risk factors. International Journal of Environmental Research and Public Health 9, 18101819.
Kariuki, SM, Abubakar, A, Kombe, M, Kazungu, M, Odhiambo, R, Stein, A and Newton, C (2017) Burden, risk factors, and comorbidities of behavioural and emotional problems in Kenyan children: a population-based study. The Lancet. Psychiatry 4, 136145.
Kariuki, SM, Chengo, E, Ibinda, F, Odhiambo, R, Etyang, A, Ngugi, AK and Newton, CR (2015) Burden, causes, and outcomes of people with epilepsy admitted to a rural hospital in Kenya. Epilepsia 56, 577584.
Kendall-Taylor, N, Kathomi, C, Rimba, K and Newton, CR (2008) Traditional healers and epilepsy treatment on the Kenyan coast. Epilepsia 49, 16381639.
Kendall-Taylor, N, Kathomi, C, Rimba, K and Newton, C (2009) Comparing characteristics of epilepsy treatment providers on the Kenyan coast: implications for treatment seeking and intervention. Rural and Remote Health 9, 12531266.
Keynejad, RC, Dua, T, Barbui, C and Thornicroft, G (2018) WHO Mental Health Gap Action Programme (mhGAP) intervention guide: a systematic review of evidence from low and middle-income countries. Evidence-based Mental Health 21, 3034.
Kilifi (2019) County Government of Kilifi [Online]. Kilifi. Available at https://www.kilifi.go.ke/index.php?com=1#.XfdbTJMzY2w (Accessed 16 December 2019).
KNBS (2009) National Housing and Population Census 2009 [Online]. Nairobi: Kenya National Bureau of Statistics. Available at https://www.knbs.or.ke/?wpdmpro=single-and-grouped-ages-in-years-by-county-and-district (Accessed 16 December 2019).
Luitel, NP, Jordans, MJ, Adhikari, A, Upadhaya, N, Hanlon, C, Lund, C and Komproe, IH (2015) Mental health care in Nepal: current situation and challenges for development of a district mental health care plan. Conflict and Health 9, 3.
Lund, C, Tomlinson, M and Patel, V (2016) Integration of mental health into primary care in low- and middle-income countries: the PRIME mental healthcare plans. British Journal of Psychiatry 208 (suppl. 56), s1s3.
Mbuba, CK and Newton, CR (2009) Packages of care for epilepsy in low- and middle-income countries. PLoS Medicine 6, e1000162.
Mbwayo, AW, Ndetei, DM, Mutiso, V and Khasakhala, LI (2013) Traditional healers and provision of mental health services in cosmopolitan informal settlements in Nairobi, Kenya. Afr J Psychiatry (Johannesbg) 16, 134140.
Musyimi, CW, Mutiso, VN, Loeffen, L, Krumeich, A and Ndetei, DM (2018) Exploring mental health practice among traditional health practitioners: a qualitative study in rural Kenya. BMC Complementary and Alternative Medicine 18, 334.
Musyimi, CW, Mutiso, V, Ndetei, DM, Henderson, DC and Bunders, J (2017) Mental health outcomes of psychosocial intervention among traditional health practitioner depressed patients in Kenya. Culture, Medicine and Psychiatry 41, 453465.
Mutiso, VN, Gitonga, I, Musau, A, Musyimi, CW, Nandoya, E, Rebello, TJ, Pike, KM and Ndetei, DM (2018) A step-wise community engagement and capacity building model prior to implementation of mhGAP-IG in a low- and middle-income country: a case study of Makueni County, Kenya. International Journal of Mental Health Systems 12, 57.
Otiende, M, Abubakar, A, Mochamah, G, Walumbe, D, Nyundo, C, Doyle, AM, Ross, DA, Newton, CR and Bauni, E (2017) Psychometric evaluation of the major depression inventory among young people living in Coastal Kenya. Wellcome Open Research 2, 113.
Patel, V (2018) Where There Is No Psychiatrist: A Mental Health Care Manual, Royal College of Psychiatrists.
Roberts, M, Mogan, C and Asare, JB (2014) An overview of Ghana's mental health system: results from an assessment using the World Health Organization's Assessment Instrument for Mental Health Systems (WHO-AIMS). International Journal of Mental Health Systems 8, 16.
Robles, R, Lopez-Garcia, P, Miret, M, Cabello, M, Cisneros, E, Rizo, A, Ayuso-Mateos, JL and Medina-Mora, ME (2020) WHO-mhGAP Training in Mexico: increasing knowledge and readiness for the identification and management of depression and suicide risk in primary care. Archives of Medical Research 50, 558566.
Scott, JA, Bauni, E, Moisi, JC, Ojal, J, Gatakaa, H, Nyundo, C, Molyneux, CS, Kombe, F, Tsofa, B, Marsh, K, Peshu, N and Williams, TN (2012) Profile: the Kilifi Health and Demographic Surveillance System (KHDSS). International Journal of Epidemiology 41, 650657.
Spagnolo, J, Champagne, F, Leduc, N, Melki, W, Guesmi, I, Bram, N, Guisset, AL, Piat, M, Laporta, M and Charfi, F (2018) Tailoring a training based on the Mental Health Gap Action Programme (mhGAP) Intervention Guide (IG) to Tunisia: process and relevant adaptations. Global Mental Health (Cambridge) 5, e17.
Spiel, C, Schober, B and Strohmeier, D (2018) Implementing intervention research into public policy-the “I(3)-approach”. Prevention Science 19, 337346.
WHO (2008) mhGAP: Mental Health Gap Action Programme: Scaling Up Care for Mental, Neurological and Substance Use Disorders. Geneva.
WHO (2009) The World Health Organization Assessment Instrument for Mental Health Systems (WHO-AIMS) [Online]. Available at https://www.who.int/mental_health/evidence/WHO-AIMS/en/ (Accessed 12 January 2016).
WHO (2010) mhGAP Intevention Guidelines Version 1. mhGAP Intervention Guide for Mental, Neurological and Substance Use Disorders in Non-Specialized Health Settings: Mental Health Gap Action Programme (mhGAP). V1. Geneva.
WHO (2015 a) Update of the Mental Health Gap Action Programme (mhGAP) Guidelines for Mental, Neurological and Substance Use Disorders, 2015. Geneva.
WHO (2015 b) Mental health atlas. Bulletin of the World Health Organization 93, 516.
WHO (2016 a) mhGAP Intervention Guidelines Version 2. mhGAP Intervention Guide for Mental, Neurological and Substance Use Disorders in Non-Specialized Health Settings: Mental Health Gap Action Programme (mhGAP): Version 2.0. Geneva.
WHO (2016 b) mhGAP Training Manuals – for the mhGAP Intervention Guide for mental, neurological and substance use disorders in non-specialized health settings, version 2.0 [Online]. Available at https://www.who.int/mental_health/mhgap/training_manuals/en/ (Accessed 21 February 2017).