Public-private partnership models for rehabilitation service delivery: A scoping review

Background Public-private partnership (PPP) for the delivery of health services is known to improve access to healthcare, yet little is known about its utilisation for rehabilitation services, particularly in sub-Saharan Africa (SSA). Objectives As a first step to generating evidence to develop a PPP model for physiotherapy service delivery in South Africa, our study mapped and described available research evidence on PPP models for rehabilitation services in the global literature. Method The Arksey and O’Malley framework guided our scoping review. Published research on rehabilitation and PPP was searched in five databases from 2000 to August 2022 using keywords, Medical Subject Headings (MeSH) and Boolean terms. Two reviewers independently completed the titles, abstracts and full-text screening of the articles and data extraction from the included articles. A narrative synthesis was conducted, and summaries of the findings are reported. Results Nine articles were included from a total of 137 obtained from the evidence searches. Of these, five were from Australia and the others from Hong Kong, Denmark, Bangladesh and the Netherlands. All the included articles showed evidence of PPP models for physiotherapy service delivery. Conclusion Our study suggests that PPP models for physiotherapy service delivery exist, particularly in high-income countries (HICs). It also highlights limited research in low- and middle-income countries (LMICs). Clinical implications There is a need for primary studies to generate further evidence and develop innovative PPP models for rehabilitation services for the populations who need them most as part of efforts towards improving access to healthcare in LMICs.

. People living with disability, for example, have limited access to rehabilitation services in sub-Saharan Africa (SSA) (Naidoo & Ennion 2019). This has been attributed to a lack of funding for the construction of rehabilitation centres and the high financial costs associated with rehabilitation (Naidoo & Ennion 2019), among others, such as transportation costs to the facility, distance and poor knowledge by PLWDs of where these services are, or knowledge about rehabilitation. Furthermore, referral pathways are erratic, and the availability of rehabilitation services is compromised, resulting in avoidable complications for patients due to inadequate follow-up (Health 2015;Naidoo & Ennion 2019;Sherry 2014;Visagie, Scheffler & Schneider 2013;Visagie & Swartz 2016). The rehabilitation challenges are exacerbated by a lack of infrastructure, particularly in SSA, where district hospital rehabilitation units are poorly maintained and scarce (Health 2015). There has been little to no research on appropriate rehabilitation development indicators at a tertiary, specialised or primaryhealthcare (PHC) levels (Health 2015). Currently, communitieswith limited resources rely on community-based rehabilitation (CBR), friends, family and other community groups (World Health Organization 2011).
A public-private partnership (PPP) model has been identified as a critical strategy for improving public health systems and mitigating the rising costs of a private healthcare sector that is already expensive and unsustainable (Myezwa & Van Niekerk 2013;Raman & Björkman 2015). A PPP is an agreement between a government institution and a private party in which (1) the private party performs an institutional function and/ or uses state property in terms of output specifications, and (2) substantial project risk (financial, technical, operational) is transferred to the private party, with the private party benefiting from unitary payments from government budgets and/or user fees (Manuel 2007).
There has been an increasing interest in implementing PPPs in some SSA countries to improve public health systems (Kula & Fryatt 2014;Loxley 2013;Mabunda, London & Pienaar 2018;Thadani 2014;Walwyn & Nkolele 2018). In Lesotho, for instance, the private sector helped to renovate and redevelop Queen Mamohato Memorial Hospital in Maseru, a public institution (Lang 2016). Moreover, the government collaborated with a local church in Uganda to build Ruharo Mission Hospital (Asasira & Ahimbisibwe 2018). Consequently, PPPs can leverage resources by forming mutually beneficial partnerships with private healthcare providers to create an effective, efficient and responsive public health sector by transferring private sector technical skills, innovation and resources (Raman & Björkman 2015;Suchman, Hart & Montagu 2018;Whyle & Olivier 2016). Globally, governments are tasked with improving their healthcare systems to meet the needs of most citizens who rely on public health systems. As a first step towards generating evidence to develop a PPP model for physiotherapy services in South Africa, our study systematically mapped and described available research evidence on PPP models for rehabilitation service delivery in a global context.

Method
Our scoping review followed the five steps outlined by Arksey and O'Malley in their methodological frameworkidentifying the research question; identifying relevant studies; study selection; charting the data; and collating, summarising and reporting results (Arksey & O'Malley 2005).

Identifying research question(s)
Our review answered the following question: 'What research evidence exists on PPP models for rehabilitation services delivery in the global context?' The population, concept and context for our study's question are defined in Box 1.

Identifying relevant studies
Relevant peer-reviewed articles and grey literature were sourced from PubMed, EBSCOhost (Academic search complete, CINAHL with full text, Health Sources), Cochrane Library, SCOPUS and Google Scholar from 01 January 2000 to 12 August 2022. In consultation with an expert librarian, a search strategy was developed using keywords ('medical rehabilitation', 'physical therapy', 'physiotherapy', 'occupational therapy', 'speech therapy', 'public-private partnership', 'public-private mix', 'public-private cooperation', 'public-private coordination', 'public-private collaboration', 'contract out', 'contracting out', 'Private finance initiative contracts'), Boolean terms (and/or) and Medical Subject Heading (MeSH) terms. The first author (S.M.M.), a physiotherapist, conducted the literature search with support from an expert librarian. EndNote X 20 was used to manage all citations.

Inclusion criteria
• Peer-reviewed publications. • Grey literature. • Any language. • Publication between 2000 and August 2022. • Articles that focused on rehabilitation services delivery (occupational, physiotherapy, speech therapy and audiology, psychologists, social workers and dieticians and/or nutritionist).

BOX 1:
Population, concept and context framework for the main review question.

Population
Patients of all ages who use rehabilitation services such as occupational therapy, physiotherapy, speech therapy and audiology.

Concept
Public-private partnership: This refers to a contract between a private party and a government agency for providing a public service in which the private party bears significant risk and management responsibility (Forrer et al. 2010).

Context Global
• Articles that involved PPP for rehabilitation service delivery. • Articles presenting evidence on access to rehabilitation services. • Articles presenting evidence on referral pathways in rehabilitation. • Articles presenting PPP models and/or frameworks for rehabilitation service delivery. • Primary study designs. • Frameworks or models.

Exclusion criteria
• Studies that focused on PPP policies, other healthcare services such as for human immunodeficiency virus (HIV), malaria, tuberculosis and others. • Studies that had no link with the PPP policies. • Articles published before 2000.

Study selection
The EndNote library was cleaned by identifying and removing all duplicate articles and shared with the review team. The screening tools were piloted by two reviewers independently using 10% of the articles. The necessary adjustments based on the feedback received were made to ensure the screening tools were accurate and reliable. Based on the eligibility criteria, two reviewers independently screened and categorised the articles into either the 'include' or 'exclude' group at the titles and abstracts and the full-text articles screening stages. A third reviewer resolved differences in the responses of the two independent reviewers at both screening stages. The University of KwaZulu-Natal library service was used to retrieve fulltext articles that were not open-access publications for screening.

Charting data
A data extraction form was developed to chart relevant data from the included articles. Two reviewers independently extracted data from the included studies using a pilot-tested form. A third reviewer was employed to resolve any discrepancies. To answer our study's question, we extracted the following relevant data: author information and publication year, study objective or aim, study setting or location (country), type of rehabilitation service delivery (service domain), services provided, type of model and targeted population.

Collating, summarising and reporting the results
A narrative synthesis summarised all relevant data to answer our study question. A flowchart diagram (Page et al. 2021) was used to present the results of our study selection, while a table presents the characteristics of the included studies and results. In addition, a narrative summary of our findings is presented.

Study selection
A total of 137 articles was obtained from the database searches. Of the 137 articles, 18 duplicates were identified and removed. Subsequently, 119 articles were screened, and 110 which did not meet our inclusion criteria during the titles, abstracts and full-text screening phases were excluded from the data extraction and synthesis ( Figure 1).

Study findings Business models
Of the nine included articles in the review, four from Australia reported research findings that involved a business model (Carr et al. 2013;Farquhar et al. 2020), Bangladesh (Al Imam et al. 2022), and the Netherlands (Ijntema et al. 2022).
In Australia, Carr et al. investigated the feasibility of a business model that used three different funding sources (local health district, Medicare local, and private) to establish a physiotherapy service in rural communities in southwestern New South Wales (Carr et al. 2013). The pilot initial results showed a high level of uptake for previously unavailable services in primary care, residential aged care and acute and/or subacute care across a wide geographical area (Carr et al. 2013). Oversubscription of available services in some communities in the latter stages of the pilot was also reported (Carr et al. 2013). Farquhar et al. investigated how the success of a business public-private physiotherapy service delivery model in a rural setting was defined by stakeholders from each organisation and identified the barriers and enablers (Farquhar et al. 2020). All participants deemed the model successful (Farquhar et al. 2020). Critical success factors were vital to improved access to local services and satisfied stakeholders (Farquhar et al. 2020). Three mechanisms were identified to implement the service delivery model (Farquhar et al. 2020). The first mechanism was the provision of human and other resources, which included the workforce model and the use of various partnership resources (Farquhar et al. 2020). The second mechanism was stakeholder engagement, which required both motivated and consistent stakeholders (Farquhar et al. 2020). The third mechanism was streamlined processes, which included the contract and referral schedule content, streamlined administration processes for contracting and accounting, processes for managing private therapists in a public setting and communication processes (Farquhar et al. 2020).
In the Netherlands, Ijntema et al. (2022) delineated the relationships between business model efficiency and novelty and physiotherapy primary healthcare organisation (PTPHO)centred outcomes in Dutch healthcare while accounting for managed competition contracts (Ijntema et al. 2022). The study showed that managed competition and business model efficiency do not affect PTPHO outcomes (Ijntema et al. 2022). The managed competition contract does not appear to moderate the business model efficiency and PTPHO-centred outcomes relationship (Ijntema et al. 2022). Business model novelty correlates positively with PTPHO-centred outcomes (Ijntema et al. 2022). A managed competition contract is discovered to moderate the relationship between business model novelty and PTPHO-centred outcomes (Ijntema et al. 2022).

New graduate physiotherapists training publicprivate partnership model
In Australia, Schmidt and Dmytryk investigated the shared education process and the value of learning opportunities in each sector (public and private) (Schmidt & Dmytryk 2017). The study showed that the model offered a novel approach   http://www.sajp.co.za Open Access

Denmark
To investigate whether a multidimensional PPP intervention, focussing on improving the quality and efficiency of sickness benefit case management, reduced the sickness benefit duration and the duration until self-support.
Quasi-experimental study (differencein-difference design) Physiotherapy Implementing Falck's rehabilitation model with integrated case management, job coaching and healthcare services, with diagnoses and treatment by physiotherapists and others.
Social insurance officers, the job coach and medical expert were located in the same building where they cooperate to establish return-to-work interventions that better integrate medical and employment-oriented aspects.

All individuals
Al Imam et al.

Bangladesh
To report the lesson learnt in establishing a social business model of early intervention and rehabilitation services for children with CP and adults with disabilities in a rural sub-district. to providing comprehensive new graduate physiotherapy education in a rural setting (Schmidt & Dmytryk 2017). Although the study only included two small organisations, the experience demonstrated that the public and private sectors working together to provide education leveraged the strengths of both sectors (Schmidt & Dmytryk 2017).

Informal public-private partnership model for recruitment and management of new graduate physiotherapists
In Australia, Schmidt and Dmytryk studied the value and potential of the PPP model in recruiting and managing new graduate physiotherapists (Schmidt & Dmytryk 2014). The findings revealed that the ability to attract high-quality applicants to difficult-to-fill positions reduced the risk of new-graduate attrition due to social isolation and enhanced networking between sectors, and enhanced staff skill development through diverse clinical and nonclinical experiences. These were all organisational benefits of a shared public-private role (Schmidt & Dmytryk 2014). The model was based on management flexibility and had the potential to be extended to other areas and professions (Schmidt & Dmytryk 2014). Dedicated funding, targeted recruitment strategies and increased planning to ease the transition into the workplace were identified as potential facilitators to improve the model even further (Schmidt & Dmytryk 2014).

Individual funding model
In Australia, the study by Dew et al. described the benefits and barriers to using IF to access physiotherapy services in rural areas (Dew et al. 2013). Individual funding improved access to and choice of physiotherapy providers (Dew et al. 2013). The barriers identified were the lack of information and advice, limited local service options and capacity, higher costs and fewer services, and complexity of self-managing packages (Dew et al. 2013). Carers expressed a desire for easily accessible information, a local point of contact for support and guidance, adequate financial compensation to offset additional travel expenses, and coordinated eligibility and accountability systems to address the barriers to using IF in rural and remote areas (Dew et al. 2013). At the same time, service providers needed coordinated cross-sector approaches, local workforce planning to address therapist shortages, certainty about service viability and growth, and transparent policies and procedures for implementing IF to address barriers to using IF in rural and remote areas (Dew et al. 2013).

Falck's rehabilitation model
In Denmark, Larsen et al. investigated whether a multifaceted PPP intervention aimed at improving the quality and efficiency of sickness benefit case management reduced sickness benefit duration and time to self-sufficiency (Larsen et al. 2017 (Larsen et al. 2017).

Social health model
In Hong Kong, Schoeb's study examined the challenges confronting Hong Kong in light of the health social model (Schoeb 2016). The Hong Kong Special Administrative Region combines a British colonial history with a Chinese cultural context. It provides its residents with a dual system that includes a comprehensive and efficient public healthcare system and private hospitals and practitioners (Schoeb 2016). Due to an ageing population, staff shortages at all levels, an underdeveloped PHC system and increasing demand for health services are all looming challenges (Schoeb 2016).

Discussion
Our study mapped and described research evidence on PPP models for rehabilitation services in the global literature. The results showed a variety of PPP models, such as business (Al Imam et al. 2022;Carr et al. 2013;Farquhar et al. 2020;Ijntema et al. 2022), recruitment (Schmidt & Dmytryk 2014), new graduate training (Schmidt & Dmytryk 2017) and IF (Dew et al. 2013), exist for physiotherapy service delivery. The PPP arrangement facilitated access to physiotherapy services, especially in rural areas. However, most PPP models were developed and implemented in HICs, with a few in LMICs.
Like HICs, the private sector plays an essential role in healthcare services delivery, including rehabilitation services in most LMICs (Bhattacharyya et al. 2010), partly because the public health system is unable to provide the needed health services to all who need them (Bhattacharyya et al. 2010). Aside from infrastructure challenges for rehabilitation in most LMICs, there is also a problem with human resource capacity for public hospitals or clinics, such as inadequate medical rehabilitation professionals (Fusheini & Eyles 2016). Governments, particularly in low-income countries, struggle to retain the few trained rehabilitation professionals partly due to remuneration concerns (Rockers et al. 2012). Nonetheless, the United Nations advocates universal health coverage as stipulated by sustainable development goal 3.8. All member countries are urged to achieve this goal by 2030 (United Nations 2016). Therefore, we encourage governments in LMICs to partner with private rehabilitation providers to expand access to PLWDs, particularly those who are poor and cannot afford the services from private providers that are not available in public facilities. Most often, it is argued that private health services are generally expensive. Still, it is our opinion that through a PPP arrangement acceptable to all key stakeholders, their services may be affordable to even the less privileged person living with a disability requiring rehabilitation.
Although our study identified several PPP models for physiotherapy service delivery, their implementation and sustainability may be challenging in some settings, considering the economic differences among countries. We believe that further research, such as primary studies or implementation research, would be required to address contextual issues before their adoption and full implementation. Moreover, such studies may help generate further evidence to develop an innovative PPP model(s) for various rehabilitation service delivery relevant to context and potential to address service gaps in public hospitals or clinics.
Although our study has several strengths, including the inclusion of global literature, it may have inherent limitations. Few (five) databases were searched. Aside from that, our study was focused on only PPP models. Despite these limitations, our study is novel. It could provoke more research in this area, especially PPP arrangements for other rehabilitation services such as occupational therapy, speech and audiology.
In conclusion, our findings indicate that PPP models for physiotherapy service delivery exist, mainly in HICs. It also emphasises the scarcity of research on PPP models and rehabilitation services in LMICs. To improve access to healthcare in LMICs, we recommend primary studies to generate additional evidence and develop innovative PPP models for rehabilitation services for the populations who need them the most.