Healthcare seekers across the world utilise multiple systems of care.^1,2,3,4 Traditional health practitioners (THPs) are often consulted or even preferred, alongside mainstream, allopathic practitioners, because they are potentially more accessible, available, affordable and culturally acceptable for holistic care.^5,6,7,8,9,10 Indeed, a recent World Health Organization (WHO) global survey found that in most countries (93 of 133), THP services are highly utilised.¹¹ The WHO defines THPs as practitioners who use indigenous or native approaches, knowledge, culture and spiritual therapies, as well as beliefs that incorporate plant-, animal- and/or mineral-based medicines to maintain well-being.¹²

This definition provided by WHO is often used to encompass all traditional healers, and other alternative and complementary healers. As such, in the global literature, there is usually a conflation of traditional healers with complementary and alternative medicine (CAM). The latter often refers to traditional medicine or healing practiced outside of its origin and traditional culture, thus less culture-specific.¹³ Furthermore, CAM is distinguished by being far advanced in policy development and regulations.^{14,15,16,17,18} On the other hand, traditional healing practices are localised, deeply rooted in spirituality and still marginalised as a healthcare system in most parts of the world, when compared to CAM.¹⁹ For this study, the focus is on traditional healers and any reference to THP only refers to indigenous traditional healers and not CAM.

Over the past few decades, the WHO has sought to support countries in harnessing the contribution of THPs in the healthcare system. For example, the WHO’s Traditional Medicine Strategy released in 2002 and updated in 2013 highlights the context, which can foster a more integrated health system, where THPs are fully part of the health system.^20,21 Despite the important progress that has been made in this area, there are still significant limitations.^{11,22,23,24,25} As early as 1978, the WHO stressed that partnership between THPs and biomedical health practitioners (BHPs) would be key in the process of integration.^12,20 Since then, increased attention has been placed on how to foster practitioner-level collaborations.^11,26,27,28 This work has involved investing time and resources into studying traditional medicines and herbs, institutionalisation of THPs and patients views on such collaborations, as well as the establishment of healthcare centres for THPs and BHPs collaborations in well-resourced parts of the world.^15,16,17

In Africa, disease outbreaks such as human immunodeficiency virus (HIV) have led to increased literature on collaborative partnerships between THPs and BHPs.^29,30 King and Homsy²⁹ give a summary of these attempts in the area of HIV in their review. Sima and colleagues³¹ provide an example of recent attempts of THPs and BHPs collaboration from TB cases in Ethiopia. In these collaborations, THPs are trained to refer clients to BHPs, as such they have been criticised for being one-sided, leaving practitioners and policymakers with little to learn from these attempts.^30,32,33,34

For collaboration to be reciprocal, it requires a process where THPs and BHPs form partnerships which, at the very least, constitute an act of client referral between two practitioners.^12,20,22 This definition does not provide a clear indicator of how these referrals should occur and does not mention other components that should be present in a collaboration. D’Amour et al.³⁵ maintain that collaboration extends beyond forming partnerships, and involves mutual trust and respect, having a set of common goals, sharing of values, responsibilities, power and decision-making, as well as interdependency in addressing patients’ needs. This interdependency facilitates co-treatment or bidirectional referral of patients between practitioners.

Similar to D’Amour and colleagues, Kraus³⁶ describes collaboration in healthcare as:

Therefore, genuine collaboration necessitates a two-sided effort whereby the healing methods of one are brought to the fore and the most effective one is chosen to cure the patient’s identified problem at that time.³⁷

As seen from these definitions, collaboration must constitute reciprocity, and any one-way partnership cannot be referred to as collaboration. Recent studies by Ampomah et al.,³⁸ Mutale,³⁹ and Baheretibeb, Wondimagegn and Law⁴⁰ point to the continued paucity of collaboration reports between THPs and BHPs. These scholars repeat the same barriers that have been highlighted in most studies in this area. These mainly include philosophical difference in how the practitioners from the two systems view health and approach treatment, countries’ resources and political willingness to incorporate THPs into the health systems, and administrative barriers in ensuring partnerships are formed and sustained between THPs and BHPs. Despite these challenges, THPs and BHPs are encouraged to formulate functional collaborations as the service users continue to use them both, concurrently.³⁹

As stated by Kayombo et al.⁴¹ and others earlier, information on the availability of collaboration for patient care is missing. In response to this, this study sought to scope and understand whether collaborations of this nature potentially exist; if so, which salient factors are associated with the success (or lack of) of these collaborations. Understanding these factors is useful for the conceptualisation of pragmatic collaboration attempts, as well as the development of collaboration frameworks, which will inform future policy and practice. We therefore mapped literature on collaborations between THPs and BHPs globally, across all health conditions and based on the definition of collaboration described earlier.

The description of studies, intervention characteristics and findings of studies have been added in Online Appendix 2.

As seen in Figure 3, research on collaboration between THPs and BHPs came from Southern America (n = 2),^50,51 Northern America,^49,54,55 Africa (n = 2),^48,57 China (n = 2)^53,56 and New Zealand (n = 1).⁵² Out of the 10 included studies, 4 were conducted in high-income countries (HICs)^49,52,54,55 and 6 in low- and middle-income countries (LMICs).^{48,50,51,53,56,57}

This scoping review, covering the past five decades, demonstrates that there is a dearth of evidence on collaboration practices between THPs and BHPs globally. With only 10 studies reporting THP and BHP bidirectional collaboration, our study shows that while literature on collaboration between THPs and BHPs exists, as seen from the number of studies screened and later excluded in this review, only a fraction of this literature reports actual reciprocal collaborations between the two healthcare systems. This study was able to show the availability, as well as describe the structure, dynamics and overall contributing factors to success or lack of success of collaborations, globally.

The available publications on collaboration were sparse, with only four studies published in the last decade in this area. However, we caution that it is likely for collaboration endeavours to go undocumented, as we found that in publication information of two studies by Bouchard⁵¹ and Durie.⁵² The reported interventions were implemented in 1990 and 1980, but only published in 2009. In the two studies reported from Africa, neither reported real-life collaboration programmes, but rather experimental research studies. Likewise, these collaborations had a short time frame with no plan of sustaining the programmes. Street et al.⁵⁹ mentioned how these research efforts do not consider the scale up of intervention programmes. As the studies included were research projects, occurring outside of the daily functioning of the health facilities, making them difficult to imagine as a day-to-day clinical practice, it is difficult to make implementation recommendations from these studies. This finding indicates that more studies are needed to understand reasons for continued lack of collaboration programmes in this region. The evidence from other countries was just as limited, with only five studies presenting collaboration cases in America, two studies in Asia and one study in New Zealand.

It was surprising to see little literature reporting collaboration in countries such as China. While we excluded records on the CAM practitioners, we still expected to find data on Chinese THPs considering that the country has taken huge strides in integrating traditional Chinese healers into the health system. According to Canaway¹⁵ and Pinkoane,⁶⁰ the integration of THPs in China has mainly happened at the governance or constitutional level, as seen with the transformation of the education to include both systems. The focus has also been in the integration of THP medicine, compared to the formation of reciprocal collaborations between THPs and BHPs in this setting. Park and Canaway²⁸ state that Chinese practitioners are still trying to forge partnerships. Zhou and Nunes⁵⁶ and recently Boum⁶¹ confirm that there are interpersonal and training differences that make it difficult for the two practitioners to collaborate, despite the seemingly enabling environment.

Out of all studies analysed here, THP-BHP collaboration was mostly found for mental health services. Our findings revealed that there seems to be a focus of collaboration towards severe disorders within mental health, such as psychosis. Mental health conventional practitioners have long recognised the potential of THPs within the space of mental health management.^{13,62,63,64,65} Our evidence shows that, while THPs and BHPs seem to be doing better in initiating collaborations within mental health, more work needs to be performed to strengthen the administration within these collaborations. For example, we found that THPs complained about inability to record their diagnosis on client files, as spiritual illnesses are not accommodated. The current Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR)⁶⁶ tool does make mention of culture-bound illnesses; however, it is possible that this tool places different spiritual related illness under one diagnosis, mostly referred to as culture-bound syndrome.

This review allowed for an understanding of collaboration structures, processes and collaboration dynamics, as well as an overview of challenges with forming and sustaining collaboration. Martín-Rodríguez et al.⁶⁷ write that for collaboration to efficiently happen, changes need to happen at systematic, organisational and interactional levels. Collaboration especially in HICs was mostly mandated. This mandate to create person-centred integrated healthcare came from the health ministries and the healthcare seekers. This was followed by infrastructural changes, availability of leadership support and financial resources.

Durie,⁵² who reported a case study with least reported challenges, described how collaboration rather started at policy level, where there was a political shift to embrace indigenous ways of healing and foster culture-sensitive healing in New Zealand. In their study, integration filtered down to university level where health sciences students were taught about Maori traditional healing methods, which enabled them to appreciate working with THPs when government enforced collaboration.

When it comes to the organisation of collaborations, reports from all studies showed that THPs had to leave their community-based practices to be placed at a hospital healthcare centre or a research site, where they would work. Shortell et al.⁶⁸ and Coddington, Fischer and Moore⁶⁹ warn that this acute care approach is a cultural barrier to collaboration and contradicts the concept of population-based healthcare delivery. It also corroborates with the findings by Shuval and Mizrach,⁷⁰ as well as Shange,⁷¹ who emphasise BHPs’ inclination to control THPs and excluding THPs in collaborations within health settings. While the usefulness of having collaborating practitioners co-located cannot be denied, it is important to observe that unless THPs are included at senior decision-making level in these spaces, THPs become employees with little say on patient care. This could lead to dissatisfaction and less engagement by THPs in the plans towards collaboration. Therefore, discussions on power imbalances are important in coordinating collaboration.⁷¹

Although collaborations were sometimes enabled by systematic changes, such as political and government support, their organisation, cross-referrals were ad hoc and the client was the one navigating these referrals on their own. The role of the client in collaboration processes is something that is barely discussed in the literature. Kaptchuk and Miller⁷² state that having a client being mostly responsible for coordinating different treatments indicates weakness in the administration processes and the relationship between the practitioners, which are key ingredients in collaborations. A study by Ampomah³⁹ concluded that a proper coordination of cross-referrals is a priority in enabling collaboration between the two practitioners. Guidelines and policies on collaborations should provide templates that practitioners can use while attempting collaborations.³⁹

This study shows that the work of collaboration is complex and nonlinear. The substance of collaboration seemed to be linked with the time taken in investing in policies, resources and structures. It was noticed that attempts to collaborate are being made in LMICs, with 6 of the 10 studies coming from these counties. Furthermore, collaboration programmes are small scale and located in rural areas, where health resources are usually limited. This speaks to the recognition of THPs in expanding healthcare service delivery in areas where poor resources for health are reported. This is not to say these collaborations were out of challenges, but the support for them to occur ensured continuity even in the face of problems. As we move towards plans for universal health coverage and person-centred care, the need for this investment is crucial.

We believe that this is probably one of the first studies to summarise the evidence of existing collaboration attempts between THPs practitioners and BHPs, globally for all health conditions. We applied an extensive search method in seven databases, which also included searching for grey literature to insure we include the most relevant studies. The search period was also widened to map the literature published since 1978–2023, August. We notice that using only three databases in the latter two searcher is a possible limitation for this study. The reasons for this is because during initial search we found that of the eight databases, three sources yielded more results, and as such we were guaranteed to find the relevant literature.

We used the Rayyan software for screening data because this software allows for ‘blind’ screening, and we were able to minimise bias. This also increased the credibility of our screening process. It is also possible that searches in this topic existed under different terminologies that were not documented in the review. However, we included MeSH terms to help address this.

From the studies reviewed, we were able to understand the structures that need to be in place to enable bidirectional collaborations to occur, the processes that take place within these collaboration programmes, as well as the factors that may hinder sustainability or scale up of these programmes. While the scoping approach does not allow for in-depth and detailed analysis of each of the collaboration cases presented here, it did map key aspects of these collaborations. A qualitative evidence synthesis, alongside primary studies that would provide an analysis of each programme, focusing on its mechanisms, collaboration influencers, processes and outcomes, is needed. Furthermore, capturing the experiences of the practitioners within those collaborations would be useful, particularly for programme implementers looking into replicating already existing collaboration programmes. Using case study methods in reporting these collaboration programmes will allow for better insights into the contextual components and dynamics of these collaboration programmes. This study confirmed that practitioner-level collaborations are few and their dynamics are similar in various settings, in that they depend on the formulation of dialogues and relationships, the practice of cross-referrals, opportunities for joint training, and treatment plan meetings. If countries are to achieve the integration of the two systems as per the WHO’s goal, more work is needed to move policy into implementation. Investing time in researching interpersonal, contextual and administrative hurdles to initiating and maintaining collaboration is one way to move policy plans into implementation.