Empirical research discloses implementation as a significant impediment to change [1, 2]. This paper contributes to implementation studies by testing an implementation theory, Contextual Interaction theory in a new domain, health. Our study explored policy implementation experiences of 71 actors involved in universal health coverage (UHC) policy implementation, in one South African pilot district using the CIT lens.
South Africa is the focus of this study because the South African government began piloting policies aimed at achieving UHC in 2012. Policies aimed at achieving UHC have been and are being rolled out in the selected 10 pilot districts. World-wide, there is growing UHC enthusiasm at global and national levels . Almost everyone world-wide believes that no-one should be pushed into poverty for accessing health care. Many also believe in a world in which everyone can live healthy productive lives regardless of who they are or where they live. This outcome will depend to a great extent on how UHC policies are and will be implemented. For example, the South African Ministry of health`s role in providing guidance has been characterized by good policies without equivalent emphasis on implementation, monitoring and assessment of these policies through-out the system . When filtering public policy through the circumstance of reality, it is important to bear in mind that intention may not correspond with outputs and outcomes . In 2017, the South African NHI pilot phase, health systems strengthening came to an end, though health strengthening initiatives and improvements of service delivery platforms continue to be implemented . Many lessons have been learnt during the past five years including implementation bottlenecks such as supply chain, infrastructural and resource challenges and underspending of conditional grants [3, 6]. In addition, there is a call that it is essential that once data and information from pilot districts becomes available, it should be fully utilized, evaluated and results be published to allow for government and public to engage with the process as well as foster accountability .
To understand implementation, it is important to define the concept as used in the scope of this study. Pressman and Wildavsky  defined implementation as to carry out, accomplish, fulfil, produce, complete a policy. Mazmanian and Sabatier  define implementation as the carrying out of a basic policy decision. O`Toole  defines implementation as what develops between establishment of an apparent intention on the part of government to do something or to stop doing something and the ultimate impact in the world of action. For the purpose of this study we define policy as translating public policy intention into results .
Policy implementation is a complex process incumbent upon a number of factors. It is important to not only understand why policies succeed but also why they fail . Many scholars find it useful to treat implementation as a distinctive point for analysis within the policy process with an ability to shed light on the whole [1, 11, 12].Three generations of implementation theories shape the policy implementation field. The first generation, illustrated by the work of Pressman and Wildavsky is also known as top-down. Researchers analysed the implementation of a regional employment programme and came to the conclusion that amongst other things, implementation failed due to poor policy planning. This generation assumed that implementation happens as a pure application of earlier decisions and consisted of solitary case studies featuring negative reports of the way governments implement their own programmes. Policy making was assumed to be automatically followed by successful policy (14). This generation of theories is also likened to what Elmore  describes as forward mapping. Forward mapping assumes that implementation is controlled from the top and stresses factors that emphasize control such as funding formulas and authority relationships. The closer one is to the source of policy, the greater their influence and authority . According to Goggin et al (ref), the first-generation studies have been criticized for being a theoretical, case specific and noncumulative and overly pessimistic. They however have been valuable in shifting the focus and showing how a law becomes a programme and demonstrating the complex and dynamic nature of implementation among others .
The second generation theories, according to Sabatier  started in the late 1970s to early 1980s, where the bottom uppers started with an analysis of the multitude of actors who interact at an operational (local ) level on a particular issues. The focus was on strategies pursued by various actors with the aim of fulfilling their objectives. Efforts were focused on setting out to record and illustrate the complexities of implementation and to show that just because a policy exists, does not guarantee that it will be implemented . The bottom up approach emphasizes the role of the smallest actor in the implementation process all the way to the top . Lipsky work on Street Level Bureaucrats  is one bottom up example. The study illustrates how civil servants at lower levels implement policies according to what they think is best and not according to directives. This is not a result of defying authority or going against rules set by those at the top, but it is a way in which actors at the bottom cope with implementation of policies that are otherwise out of their scope because of the way they are planned . Walker and Gilson used the street level bureaucrat`s framework in South Africa, investigating, “how a group of nurses in primary care clinics experienced the implementation of free care and other South African national health policies post 1996.” They too, came to a conclusion that, civil servants faced with policy implementation use their discretion based on their views and values . After years of debate between proponents of top down and bottom up approaches, most researchers concede the merits of both perspectives with several scholars recommending a synthesis of the two into a unifying model [9, 11].
Van Meter and Van Horn  highlight the systems model of implementation” in which they show that there is more to what influences the performance of a programme or a policy that separate factors. The separate factors include policy standards, resources, support for policies in a political environment, economic and social conditions, characteristics of implementing agencies, communication of policy standards incentives to promote compliance and policy dispositions of implementing officials. These factors highlight the importance of context in policy implementation process leading us to the third generation of implementation research and how it combines both top down and bottom up approaches. The third-generation implementation researchers see implementation as an ongoing process, regardless of result. Goggin et al  promote a third generation of research to illuminate the variability within implementation scenarios using more stringent scientific methods. Ad-hoc explanations do not capture the essence of why policy implementation fails, are often incomplete and insufficient in telling and explaining the whole story  Implementation theory and research have outgrown the search for a single theory of implementation and have entered a new era that recognizes multiple theories appropriate to various implementation research questions .
Justification of Contextual Interaction Theory
Contextual Interaction Theory (CIT) is a third generation theory  developed in the Netherlands during the late 1990s and has been applied in several studies [2, 20] but not widely used in the health policy domain. CIT is relatively simple, with broad applicability as it analyses the very core of implementation: the motivation, information, resources and power of target and implementer. CIT emphasizes the policy target and implementer, whether they exist as local implementers or higher-level administrators. A useful theory must condense reality into less detailed but informative elements 
Scharpf (23 ) writes that overly parsimonious theories ignore either actors or institutions in pursuit of the other. CIT is parsimonious, distilling a sea of options for implementation variables into core variables of motivation, information, resources and power. These variable are not arbitrarily chosen as three important variables among others but because they have high explanatory power and exist at the core of interaction process [2, 11]. CIT is therefore a deductive and realistic approach that allows implementation to be effectively analysed .In implementation science, theoretical approaches serve three main purposes: to describe and or guide the translation of research into practice-process models, to understand or explain factors that influence policy implementation (implementation theories) and to evaluate implementation (evaluation frame works) . CIT utility has been proven in all three areas elsewhere [2, 11, 23, 24]
Description of CIT
The theory focuses on motivation, information, power, resources and interactions of these. CIT also allows this study to focus not only on CIT variables but also context, structural and outer. CIT has been used in the South African context before. Applying CIT to analyse UHC policy implementation is a new domain. Based on the amount of information the actors have on policy, their level of motivation, the amount of resources they have to implement and the amount of power they have to mobilize needed resources and various interactions of these, a policy can be successfully implemented. In this study we analysed how much Information, Motivation, Power, and Resources they had with regards to UHC policies they were tasked to implement. We also analysed the data inductively.
One of CIT`s key assumption is that factors influencing implementation are interactive. The influence of any factor whether positive or negative depends on the particular context. The theory distinguishes a set of core constructs or concepts related to the actors involved which jointly contribute to implementation. Core constructs are;
The level of importance the actors place on a policy and the degree to which policy contributes to their goals and objectives affects implementation. If actors have low motivation, they may ignore implementing the policy. Examining motivation helps to understand the perspectives of implementers, their belief system, value priorities and perception of the importance and magnitude of specific problems often revealing root causes of implementation barriers .
Successful policy implementation requires that those involved have sufficient information including technical knowledge of the matter at hand, levels and patterns of communication between actors. For example, do those responsible for implementation actually know with whom they should be working and who the policy should benefit? Do they know which department is assigned to lead the implementation and how the programme will be monitored? How is information and communication between actors coordinated? Have guidelines been developed and are they readily available ? [11, 20].
Who is empowered to implement policy and to what degree? Power may derive from formal sources such as a legal system e.g. appointment or from informal sources such as charisma or being an elderly.
Having adequate resources for the intended action is important for actors to realize policy implementation goals. Resources provide the capacity to act . The relevance and availability of resources influence the actors motivation which in turn influences the whole policy implementation process .
interactions predict the level of collaboration among and between actors which in turn influence policy implementation. They must be considered to further analyze barriers to implementation. These interactions can take different forms from cooperation, passive cooperation, forced cooperation, opposition or joint learning. In turn actors collaboration depends on how they perceive the problem being addressed as a priority, how convinced they are that there is an acceptable solution, that taking action now is in own best interest and if they have implementing capacity [11, 20]. Specifying the above constructs facilitates the development of tools to measure the level at which each of the core construct contributes or hinders implementation . These central CIT tenets guided our analysis and interpretation of findings in this study.