Transferring performance-based financing practices in Burundi health system: the role of actors’ interactions

INTRODUCTION

For the past twenty years, African countries have experienced a massive deployment of performance-based financing practices (PBF-P) aimed at reforming their health systems (HS) affected by performance issues. Inspired by the ideas of new public management, the transfer of PBF-P aims at replacing traditional organizational systems to introduce participatory management, contracting and good governance practices within national HSs in Africa.¹ Basically, the concept of transfer has existed for a very long time in the literature to define an exchange relationship between actors.^2–6 Faced with the ambivalence of this concept and the complexity of its process, we define transfer as an operational process of exchanging managerial practices between two or more entities with a view to carrying out an activity, from the detection of the need to its satisfaction.

Thus, the transfer of PBF-P objective is to replace the incentive and compensation systems with a financial incentive system linked to specific results.⁷ This process has been set-up by the diffusion entrepreneurs and several other global health actors.^1,8 These actors have played key roles in the many cases of transfer process (TP) of PBF-P in Africa, especially in Burundi health system (BHS). The diversity of their profiles and the multi-level of their involvement have led to several forms of contextualization of the implementation of PBF-P.

Recent research in transfer has identified several processes concerning either policies or practices transfer. Some authors have been interested in policy transfer with an emphasis on the determinants of public policy transfer, dissemination mechanisms and the definition of phenomena such as imposition, influence, convergence, translation, transposition, inspiration or imitation._5,6,9 Others have focused more specifically on the introduction, the implementation, and the diffusion of PBF-P as well as the diffusion models of managerial practices in HSs.^8,10,11 The actors’ role has been highlighted in the transfer of PBF-P as they have intervened in different ways at different levels, and with various functions.^1,8,9

However, the role of actors’ interactions in the TP of PBF-P remains less explored in context and in depth to understand their interactional dynamics. In the field of social relations, actors’ interactions include the practices of reciprocity, adjustment, interpretation and influence.^13–15 In other words, these are social relationships that are established between two or more actors who intervene in a mutual field of expectations, influence, interpretation and adjustment.

This article aims to fill this gap. Faced with the observations we made, we propose in this paper to explore the role of the interactions of the actors who intervene in the transfer case of PBF-P within BHS. Our approach was first to present the theoretical framework exploring the PBF transfer and interaction concepts. Then, we developped the case study, starting with the methodological approach which followed an exploratory, interpretive, abductive and qualitative logic to propose the roll-out model for the TP of PBF-P in BHS.

Actors’ interactions in the TP of PBF-P

Recent research in the transfer of PBF-P have shown the invlovement of many actors whose multiple profiles and various intervention levels raise questions about the role of their interactions in the deployment of this process. Indeed, the introduction of a new practice in a new setting requires a "problematisation" step.^16,17 This step corresponds to an operation of identification that is essential to any action of setting improvment. It aims to establish, hypothetically, the identity of all the actors involved in the TP as well as the mechanisms that link them. Being aware of the multidimensional nature of PBF actors and the different forms of complexity observed, it is normal to think that the interaction of these actors would affect the deployment of PBF-P.

As for the implementation and dissemination of PBF-P, several strategies have been used. Its analysis leads to "actors’ interactions". These strategies include conducting perception studies, holding multi-stakeholder meetings, and organising study tours for the exchange of experiences between stakeholders.^10,18-20 These strategies allowed actors to interact in analysing the transferability and applicability of PBF-P. Moreover, the institutional set-up of PBF-P also constituted a strategy that created actors’ collaboration. Indeed, the implementation of some practices in new settings implies the need for convergence between both origin and destination settings.¹⁶ It was through this convergence that all the theoretical and practical aspects have been defined by actors involved for an effective implementation of PBF-P. The choice to start with the pilot projects were made by involving local actors to determine their content and duration which varied from one country to another.

Literature in the transfer of PBF-P has identified national differences in pilot projects set-up as a determining factor of the deployment effectiveness of the PBF TP.^12,21 The development of pilot projects was an operational element of convergence, on the one hand between external and internal actors in HSs, and on the other hand, between the internal actors of the same HS.^21,22 The philosophy of these pilot projects was to conduct feasibility experiments before the generalization of PBF-P to avoid the risks of failure as it has been observed in Uganda and Tchad.^23,24 But, it should be noted that the implementation of the PBF reforms was also an opportunity for many countries to involve a synergy of actors, especially non-state actors, and to strengthen the public-private partnership.²⁵

Faced with the success stories of PBF-P implemention in HSs, some countries, like Burundi, have attempted to disseminate these practices in sectors other than health. This has been due in particular to the existence of a kind of enthusiasm created by the personal opinions of some actors involved. This aspect of diffusion often leads to the development of institutional arrangements that should an understanding of the PBF principles before their transfer. In the idea of restructuring the transfer of PBF-P that pose critical issues in Africa, a closer and inclusive collaboration of all actors involved would remain one of the effective strategies to this end.⁸

Finally, it can be underlined that the actors’ interactions were observed in the different phases of the transfer deployment of PBF-P. The deployment of this complex process is conditioned by a set of factors affected by the intervention of highly interactive actors. From the literature review and the observed case of the PBF-P TP in BHS, we have put forward three explanatory proposals to discuss in a serie of concordance verification to state our main final proposals: The interactions of the actors involved, their causes, and their effects affect the deployment of the TP of PBF-P.

METHODS

We conducted a case study within BHS. Indeed, PBF-P are implemented in BHS since 2006. This TP reveals three observations. The first is the processual complexity. Indeed, the new principle of separation between regulator, funder, provider, auditor and community voice functions run up against existing practices from the public sector bureaucracy not yet reformed. This 'heterogeneous' set is operating in the same HS while the complexity is generated from the diversity of the functions created. This complexity raises the question of the integration of PBF-P in the presence of divergent logics of actors involved in the TP. Secondly, there is an organisational complexity. This is linked to the difficulty of migrating from the "public" logic to managerial practices while actors involved were facing the problem of decision-making due to the multiple and complex choices offered in addition to the multiplicity of their intervention levels. Finally, we noted that the lack of consensus on PBF-P could be "real" according to the implementation reports leading to ambivalence on the effectiveness of PBF because of the recurring and contradictory debate between the promoters and detractors of these practices. These observations raised question about the actors’ interactions role in the TP of PBF-P within the "model" transfer in Africa.

However, as literature converges on the role of multiple and various actors involved, the understanding of the TP of PBF-P requires a contextual and in-depth study to clarify the interactions’ role which could promote or hinder the TP. Then, we tried to answer the following question: How do the interactions of the actors affect the deployment of the transfer of managerial practices in the case of PBF in Burundi?

To answer this question, we used the qualitative case study method. This choice is justified by the specificity and complexity of the TP of PBF-P. This method is particularly suited to questions about the more or less implicit interactions linked to a contemporary phenomenon. Through an interpretative position, we have conducted an abductive reasoning to explore the role of the actors’ interactions towards a better understanding of the PBF-P transfer phenomenon.

For data collection, three techniques were used both for the triangulation of data and the verification of the empirical saturation of the data corpus.²⁶ Firstly, we used the semi-structured interview in the main mode. To this end, thirty-two interviews were conducted between September 2019 and December 2020 with PBF experts in BHS using a previously developed guide. Then, we conducted observations in contextual mode. This technique allowed our deep immersion as observer researcher to note in logbook behaviors and issues in co-presence situations of transfer actors. Finally, we used documentation in a complementary mode to understand the history of PBF-P. Several documents consisting of PBF implementation and study reports, archives and physical artefacts were analysed for the purpose of this research.

This combination of different techniques in this case study allowed us to collect rich and varied information from several sources for triangulation and empirical saturation.²⁶ Although we were available to interviewees for any clarifications on the question guide, the use of these methods reduced our influence on their ideas of the respondents. During the data collection, we avoided developing any other relationship beside the trusting relationship that allowed our acceptance in the research field.

In order to interpret data, we went through the purification of the collected documents to make a content analysis. We also transcribed and coded the interviews with the NVivo12 software in order to arrive at the thematic analysis. A serie of iterations were carried out between the data and the corpus of literature to extract the most significant verbatims that allowed us to deduce the trends of the results. The aim of this process was to conceptually extend the theoretical framework of the transfer of PBF-P from the confrontation of the explanatory proposals we made with the empirical observations and the literature. After conceptualization, the imaginary formulation of empirical observations constitutes, in its turn, the heart of a constellation which produces different attempts to construct reality.

RESULTS

After analysing the collected data, we identified on the one hand, five key factors that play a role in shaping the interactions between PBF transfer actors. On the other hand, we discovered four typologies of actors’ interactions in the TP of PBF-P within BHS.

Regarding the intervening factors, we identified firstly, factors related to the setting of the TP of PBF-P. These factors included the political and economic situation in Burundi while the transfer is occurring. Concurrently, the intervening factors included the social climate that prevailed in the institutions including the HS, the level of technological skills as well as the environmental and legal conditions. Secondly, we identified factors related to the organisational specifics of BHS with regard to the transfer of PBF-P. These specifics were related to governance mechanisms, availability of resources and the current performance status. Third, we identified factors related to the nature of the transferred PBF-P.

This nature was thus related to the historical background of PBF-P as well as their innovative and dynamic characters. Fourth, we identified factors related to the international, national and subnational levels of the TP operationalisation of PBF-P. Fifth, we identified factors related to the actors perceptions on the TP of PBF-P. These factors included the significance of the TP of PBF-P, the perceived values of the actors involved in the TP as well as the adopted behaviours in relation to the TP of PBF-P (Figure 1).

With regard to the typologies of actors’ interactions, we first discovered cooperative interactions, which were designed by a set of negotiated forms of collaboration and exchange between the actors involved. During these interactions, the objectives sought by the actors involved were linked to the search for consensus in the implementation of PBF-P within BHS.

Thus, several types of partnerships were concluded and various contracts were also signed. The second typology was the coercive interactions. This was a form of collaboration and exchange that was imposed on the actors involved by performance constraints. Faced with these constraints, the actors were obliged to collaborate in order to implement PBF-P. To this end, several working sessions were organised to face the permanent search for performance which aimed at adapting the dynamics of PBF-P in BHS. The third typology was selective interactions. This type of interactions were forms of collaboration and exchange that were dictated by the choice of priorities of the actors involved. Thus, a common fund was created to facilitate the pooling of mobilised funds. In return for the joint mobilisation of resources, joint management and evaluation committees were set up for monitoring. Their interactions made it possible to ensure good allocation of resources for the common interest. Finally, the fourth typology was progressive interactions. This kind of interactions was a gradual and contingent form of collaboration and exchange. The progressive interactions were adapted by actors in co-presence and according to the circumstances and the evolution of the different settings. This collaboration was the basis for several forms of adjustment in the implementation of PBF-P within BHS (Figure 2).

DISCUSSION

This paper aimed to explore the role of actors’ interactions in the TP of PBF-P within BHS. Results are discussed using an abductive reasoning approach that allowed us to formulate the main proposals from the theoretical insights and empirical observations.27 We formulated empirical observations and checked the concordance of the generated empirical observations with the explanatory proposals and the literature.

Formulation of empirical observations

From research results, six empirical observations can be generated: The transfer setting (political, economic sociocultural, technological environmental, legal) affect actors’ interactions (cooperative, coercive, selective, progressive), organizational specifics (governance mechanisms, resource availability, performance status) affect actors’ interactions (cooperative, coercive, selective, progressive), nature of the transferred practices (historical background, innovative character, dynamic character) affect actors’ interactions (cooperative, coercive, selective, progressive), levels of operationalization (international, national, subnational) affect actors’ interactions (cooperative, coercive, selective, progressive), actors’ perceptions (significance of the transfer, perceived value, adopted behavior) affect actors’ interactions (cooperative, coercive, selective, progressive), interactions of transfer actors (cooperative, coercive, selective, progressive) affect the deployment of the TP of PBF-P. After the formulation of empirical observations, they have been compared to the explanatory proposals and the literature.

Consistency of explanatory proposals with empirical observations

This verification was carried out by proceeding case by case between the explanatory proposals and the empirical observations using an adductive approach. In this reasoning approach, an explanatory proposition that was not confirmed had to be reformulated to confirm the selected proposal which were then compared with the literature.

This is how the first explanatory proposition which proposes that the interactions of the actors involved affect the deployment of the TP of PBF-P. In addition to the first explanatory proposition, five other explanatory propositions were generated and retained. These are are: setting factors affect the deployment of the TP of PBF-P, organizational specifics affect the deployment of the TP of PBF-P, the nature of the transferred practices affects the deployment of the TP of PBF-P, the operationalization levels affect the deployment of the TP of PBF-P and the perceptions of the actors involved affect the deployment of the TP of PBF-P. Using the same approach, the explanatory propositions according to which the causes and the effects of the actors’ interactions affect the deployment of the TP of PBF-P were confirmed.

Beyond this verification, two new empirical observations were generated from empirical observations to verify their consistency with the literature. On one hand, we found that actors’ interactions could influence the relationship between intervening factors and the deployment of the TP of PBF practice. On the other, we found that actors’ interactions could explain the relationship between intervening factors and the deployment of the TP of PBF-P. Finally, ten main explanatory proposals were reformulated and retained for verification with the literature (Figure 3).

Consistency of explanatory proposals with the literature

Results suggested that factors related to the setting, the organizational specifics, the nature of the transferred practices, the operationalization levels and the perceptions of the actors affect the deployment of the TP of PBF-P. Furthermore, results proposed that cooperative, coercive, selective and progressive interactions affect the deployment of the TP of PBF-P. These results partly corroborate some authors who argued that transfer of HR practices is a social process where the governance mechanisms, characteristics of the HR mnagement systems, the social relationship, and the transfer approach will influence the outcome of the process.⁶ However, those authors were not interested in the role of actors’ interactions in the deployment of the TP and their research was limited to companies that already had relationships with each other.

In the logic of consistency checking, our results corroborate also others authors who concluded that actors could shape a subset of factors involved in the TP of public policies by taking certain decisions regarding transferability, adoptability and process design, albeit within the boundaries of the environment.⁵ Their results have inspired the problematic of this research because the actors’ behavior during the TP of public policies.

Furthermore, our results have confirmed the literature which identified four interaction cotegories which are correlative, autonomous, confrontational and regulatory interactions.₂ However, our results suggested the reformulation of these categories due to the differences observed on the definitions of the categories. The specificity of our study lies in the complexity of the transfers of PBF-P and the need for their study in context and in depth.

In addition, the presence of several actors in their diversity was seen as the starting point for their interactions with a view to transferring PBF-P in BHS. This multidimensional aspect of the actors’ presence has been already discussed in the literature.^1,8 Some of them indicated that social interactions between "diffusion entrepreneurs" contributed to the mutual attractiveness for the strengthening of PBF communities of practice.⁸ In this logic, the spirit of collaboration between actors allowed closer adaptation of PBF-P that deviated from the initial context and planning.¹¹ In these circumstances the collaboration and the exchange between the actors resulted in compromises which were established between the actors to reconcile the fundamentals of PBF-P and the need to introduce relevant and acceptable managerial practices in the political context as it was the case in BHS.²⁸ The above estimation perfectly confirms our proposal that the interactions of the actors involved affect the deployment of the TP of PBF-P.

Littérature showed that the complexity of the interactions between factors associated with HSs should lead global health actors to be very cautious about their roles in the implementation of PBF-P.^1,17 Authors highlighted the need to take into account both the factors involved and the complexity of their interactions.²⁹ This idea is inline with with our proposals according to which factors linked to the setting, organisational specifics, the nature of the transferred practices, the operationalization levels and the perceptions of the actors affect the deployment of the TP of PBF-P.

Moreover, the proposal that the operationalization levels affect the TP concurs with literature which pointed out that the process of policy transfer should be examined through a structure and agency approach with three dimensions: global, international and transnational levels, the macro-level and the interorganizational level.³⁰ They argued that "a context of interaction may therefore take place through the organization of seminars, fact-finding missions, conferences and the exchange of specialist policy advice documents (for example the drafting of legislation)".³⁰ In other words, the process of transferring happens through contextual framework where actors interact with expected results. This contextual framework includes several forms of TP where agents act as channel for the development of consensual knowledge as argued by authors.³⁰ As the role of the actors involved is crucial for this process, this idea perfectly concurs with authors who underlined that actors can shape the other factors involved in the TP of public policies.⁵ These arguments are in line with our results, which sproposal which suggests that the causes and effects of the interactions of the actors involved condition the deployment of the transfer of PBF-P.

Through empirical observations, our results suggested that actors’ interactions would play both moderating and mediating roles in the deployment of the TP of PBF-P. The moderating role is explained by the fact that the interactions of PBF actors could affect the relationship between the determining factors of the transfer and the deployment of this process. While the mediating role is explained by the fact that the interactions of PBF actors could explain the nature of the relationship between the determining factors of the transfer and the deployment of this process. In the literature review, this relationship has not yet been investigated; hence the specific contribution of this research. However, some researchers have shown that interaction ties and actor relations can influence the success or failure of the TP of HR practices6 or public policies.⁵ In their contexts, these authors limited their researches to considering the actors’ interactions as direct factors which determine the transfer of policies and practices. However, the transfer and the organizational structure complexity of BHS, the diversity of the actors involved and the multiplicity of their intervention levels made us to suppress the moderating and mediating effects of the actors’ interactions within the transfer case of PBF-P. The proposals of authors strongly support our proposals according to which actors’ interactions play both a moderating and mediating role in the deployment of the TP of PBF-P.^5,6 The consistancy check of explanatory proposals with the literature allowed us to validate all the ten proposals as the main contribution of this paper. In view of the previous discussion and the retained proposals, we can schematize the deployment of the TP of PBF-P within BHS (Figure 3).

CONCLUSION

This paper contributes to better understanding the transfer phenomenon of PBF-P, in some specific and unknown aspects not yet analysed in the existing literature. More specifically, we have identified the factors involved and the typologies of actors’ interactions in the deployment of the TP of PBF-P. Based on our results, ten explanatory proposals were retained to generate a roll-up model of the TP. By highlighting the role of interactions, the contributions of the actors involved in the TP of PBF-P are reassessed and reoriented.

However, this research had some limits relating to the defect of the case study as method. This lies in its main characteristic: the field relations that are omnipresent, rich but complex, fascinating and tough at the same time. While defining adaptation strategies to ensure scientific rigor, we have chosen a case under study that was considered like exemplary because the TP of PBF-P in BHS was listed among the "pioneers" of this process in Africa. This allowed us to remove a lot of bias related to the choice of the case. Furthermore, this case can be generalized to theoretical propositions, but not to contexts other than the above specific context.

As extension, we suggest testing the proposals formulated with a larger sample in the same context of BHS and/or in other HSs in Africa by using quantitative methods. Thus, future researchers may be interested in validating our model for its application to other managerial practices. As long as the actors’ interactions can cause the (agency) conflicts due to the success or failure of the TP of managerial practices in certain contexts, this openness for future research would make it possible to conclude on the supposed role of both moderator and mediator of the actors interactions in the deployment of the TP of PBF-P in BHS. This validation could help to improve the decision-making about the TP in general.

ACKNOWLEDGEMENTS

Authors would like to thank the people interviewed for their collaboration in collecting date and colleagues for performing such kind of crucially important study which is a part of the doctoral thesis.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Turcotte-Tremblay AM, Gautier L, Bodson O, Sambieni NE, Ridde V. Le rôle des acteurs de la santé mondiale dans l’expansion du financement basé sur la performance dans les pays à faible et à moyen revenu. J Gest d’économie médicales. 2018;36(5):261.
2. 2. Degenne A. Tipos de interacciones, formas de confianza y relaciones. Redes Rev Hisp para el análisis redes Soc. 2009;16(1):63.
3. 3. Joyeux-Prunel B. Les transferts culturels. Hypothèses. 2003;6(1):149.
4. 4. Bracke D. Vers un modèle théorique du transfert : les contraintes à respecter. Rev Sci Edu. 1998;24(2):235.
5. 5. Minkman E, Buuren A, Bekkers V. Policy transfer routes: an evidence-based conceptual model to explain policy adoption. Policy Stud. 2018;39(2):222-50.
6. 6. Björkman I, Lervik JE. Transferring HR practices within multinational corporations. Hum Resour Manag J. 2007;17(4):320-35.
7. 7. Van Haeperen B. Que sont les principes du New Public Management devenus ? Reflets Perspect la vie économique. 2012;LI(2):83.
8. 8. Gautier L, Tosun J, De Allegri M, Ridde V. How do diffusion entrepreneurs spread policies? Insights from performance-based financing in Sub-Saharan Africa. World Dev. 2018;110:160-75.
9. 9. Delpeuch T. Comprendre la circulation internationale des solutions d’action publique : panorama des policy transfer studies. Crit Int. 2009;43(2):153.
10. 10. Bonfrer I, Soeters R, Van de Poel E. Introduction Of Performance-Based Financing In Burundi Was Associated With Improvements In Care And Quality. Health Aff. 2014;33(12):2179-2187.
11. 11. Bodson O, Barro A, Turcotte-Tremblay AM, Zanté N, Somé P-A, Ridde V. A study on the implementation fidelity of the performance-based financing policy in Burkina Faso after 12 months. Arch Public Heal. 2018;76(1):4.
12. 12. Ndayishimiye R, Niyondiko D, Bigawa AB. The diffusion of managerial practices in performance perspective: a review of transfer models in health systems. Int J Community Med Public Heal. 2020;7(7):2825.
13. 13. Le Breton D. Les grands axes théoriques de l’interactionnisme. In: L’interactionnisme symbolique. Quadrige. Presses Universitaires de France. 2012;45-98.
14. 14. Marc E, Picard D. Interaction. In: Vocabulaire de Psychosociologie. ERES; 2016;191.
15. 15. Bajoit G. Le concept de relation sociale. Nouv Perspect en Sci Soc. 2009;5(1):51-65.
16. 16. Callon M. Éléments pour une sociologie de la traduction: la domestication des coquilles Saint-Jacques et des marins-pêcheurs dans la baie de Saint-Brieuc. L’Année Sociol. 1986;36:169-208.
17. 17. Sieleunou I, Turcotte-Tremblay AM, Yumo H. Transferring the Purchasing Role from International to National Organizations During the Scale-Up Phase of Performance-Based Financing in Cameroon. Heal Syst Reform. 2017;3(2):91-104.
18. 18. Zitti T, Gautier L, Coulibaly A, Ridde V. Stakeholder Perceptions and Context of the Implementation of Performance-Based Financing in District Hospitals in Mali. Int J Heal Policy Manag. 2019;8(10):583-92.
19. 19. Gautier L, Coulibaly A, De Allegri M, Ridde V. From Amsterdam to Bamako: a qualitative case study on diffusion entrepreneurs’ contribution to performance-based financing propagation in Mali. Health Policy Plan. 2019;34(9):656-66.
20. 20. Sieleunou I, Turcotte-Tremblay A-M, Fotso J-C, et al. Setting performance-based financing in the health sector agenda: a case study in Cameroon. Global Health. 2017;13(1):52.
21. 21. Chimhutu V, Tjomsland M, Songstad NG, Mrisho M, Moland KM. Introducing payment for performance in the health sector of Tanzania- the policy process. Global Health. 2015;11(1):38.
22. 22. Kondo KK, Damberg CL, Mendelson A. Implementation Processes and Pay for Performance in Healthcare: A Systematic Review. J Gen Intern Med. 2016;31(S1):61-9.
23. 23. Ssengooba F, McPake B, Palmer N. Why performance-based contracting failed in Uganda – An “open-box” evaluation of a complex health system intervention. Soc Sci Med. 2012;75(2):377-83.
24. 24. Kiendrébéogo JA, Shroff ZC, Berthé A, Yonli L, Béchir M, Meessen B. Why Performance-Based Financing in Chad Failed to Emerge on the National Policy Agenda. Heal Syst Reform. 2017;3(2):80-90.
25. 25. Peerenboom P, Basenya O, Bossuyt M, Ndayishimiye J, Ntakarutimana L, van de Weerd J. La bonne gouvernance dans la réforme du financement du système de santé au Burundi. Sante Publique (Paris). 2014;26(2):229.
26. 26. Pirès A. Échantillonnage et recherche qualitative: essai théorique et méthodologique. La Rech Qual Enjeux épistémologiques méthodologiques. 1997:113-69.
27. 27. Logique DA. épistémologie et méthodologie en sciences de gestion. In: Conférence de l’AIMS. 1999:23.
28. 28. Falisse JB. Au-delà du modèle voyageur ? Usage stratégique et hybridation du financement basé sur la performance (FBP) dans la santé au Burundi. Polit africaine. 2019;156(4):83.
29. 29. Schloemer T, Schröder-Bäck P. Criteria for evaluating transferability of health interventions: a systematic review and thematic synthesis. Implement Sci. 2018;13(1):88.
30. 30. Evans M, Davies J. Understanding Policy Transfer: A Multi-Level, Multi-Disciplinary Perspective. Gest y Polit Publica. 1999;77(2):361-85.