Regional Reference Centers: analysis of the national implementation of training processes on alcohol and other drugs

An analysis of the national implementation of Regional Reference Centers (RRCs) was carried out. A survey with 97 open-ended and closed-ended questions was answered by the coordinators of 51 RRCs from the five Brazilian regions. A multi-case ex-post-facto study was facilitated by the triangulation of descriptive statistical analysis of closed-ended questions and thematic analysis of open-ended ones. The effective national scope of the RRCs was based on the increased capillarity of public education institutions. Methodological flexibility was highlighted as a convenient feature in the course design, allowing content relevant to the local reality, with proposed intervention projects in the territory, affecting the coordination of the intersectoral network on drugs. Worth highlighting was the need for higher administrative and financial autonomy and the relevance of continuity of the training processes due to the critical role of RRCs in local and regional drug policies. The analysis flags axes that can support the construction of indicators for the evaluation of alcohol and other drugs educational processes, as well as the elaboration of training principles and guidelines for this area.


Introduction
The emphasis on the formation and continuing education as a function of the SUS has been affirmed since the Eighth National Health Conference held in 1986 and enshrined in the Organic Health Law of 1990. In the first two National Conferences on Human Resources for Health, held in 1986 and 1993, the importance of guiding the training for the enhancement of health work was emphasized. The Third National Conference on Labor Management and Health Education, held in 2006, evidenced the need to include "drug addiction" as a training subject.
Permanent Health Education (EPS) adopts the daily technical-social actions of health workers in their teams as starting points of training processes [1][2][3] , orienting itself towards the acquisition of knowledge and the implementation of health practices that strategically contribute to advancing the SUS 4 . EPS-based training policies should be consistent with the principles of human rights and citizenship promotion, adopting the indissociability between theory and practice 5 , with the potential to minimize stigmas related to drug users 6 .
The National Drug Policy (PNAD) has considered the need to involve educational institutions to empower, articulate, multiply, and strengthen drug use-related actions since 2005. However, the federal government only conducted a national training policy in 2010. As a political response to the so-called "crack epidemic" at the time, the Integrated Plan to Fight Crack and Other Drugs (PIEC) was established, setting funding for the creation of Regional Reference Centers (RRCs).
In 2010, the PIEC Steering Committee, through Public Notice N°002/2010/GSIPR/ SENAD 7 (hereafter referred to as "2010 Public Notice"), made the first public call for proposals for financial support for the implementation of RRCs. The 2010 Public Notice pre-delimited the object of the courses, target audience, workload, a minimum number of students, objectives, and content. Initially, the National Secretariat for Drug Policy (SENAD) awarded decentralized funds to 51 RRCs. Both this Public Notice and the other two subsequently launched advocated that the proposals be EPS-based.
In 2011, the "Crack, it is possible to defeat it" Program was launched, whose "Prevention" axis prioritized training, dissemination of information and guidance, aimed at diverse audiences and resorting to classroom and distance learning.
Thus, the second public call was launched, targeting both the RRCs already implemented and the proposition of new Centers. The Public Call Notice N° 01/20128 (hereafter referred to "2012 Public Notice") followed the general parameters of the 2010 Public Notice, including, besides health and social assistance, the judiciary, the public prosecutor's office, and public security.
The Public Notice Nº 08/2014-SENAD/MJ 9 ("2014 Public Notice"), launched two years later, the third and last to date, broadened the scope of the target audience to "policies with interface in the prevention and care of drug use-related problems". The object, content, objectives, target audience, and methods of the courses, unlike the 2010 and 2012 Public Notices, should be agreed upon with intersectoral collegiate bodies. The 2014 Public Notice required theoretical and practical sessions in the formative processes, and a total workload ranging from 40 to 80 hours.
Thus, we can observe that training in the field of alcohol and other drugs was one of the structuring components of the policies and programs that emerged in the government cycle inaugurated after the "Crack, it is possible to defeat it" Program. Worth mentioning is the effort made by the federal government in that period to ensure conditions and funds for drug training to a wide range of workers. The "Crack, it is Possible to Defeat it" aimed to reach 463,000 health professionals and drug policy strategic areas 5 .
A recent systematic review of Brazilian mental health assessments concluded that the qualification of human resources is a crucial aspect, among others, for improving the provision of services 10 . Nevertheless, drug policy design and implementation require multi-sectoral responses, with scientific evidence considered in prioritizing the government agenda 11 . However, evaluations of implementation, effectiveness, and results in this field of public policies are rare in the Brazilian reality.
The study reported here is the first nationwide assessment of Brazilian RRCs. It sought to characterize and analyze, from the perspective of its coordinators, the process of implementation at the national level of RRCs in public institutions of technical and higher education.

Methods
An implementation analysis was conducted as an evaluative research modality, which stands out as a learning opportunity about the political-insti-tutional framework of the development of public policies, in this case, referring to the training in the field of alcohol and other drugs 12 . Implementation analyses can identify critical aspects of the success or failure of the goals set and understand the processes that have enabled a certain level of achievement of expected results 13 .
An ex-post-fact multiple case study was facilitated by collecting data from a survey questionnaire containing 97 questions in total (69 open-ended and 28 closed-ended) answered by the coordinators of the Brazilian RRCs. Coordinators were requested to inform the public notices to which they submitted, approved, and developed course proposals, either partially or in full. Concerning the 2010 and 2012 Public Notices, which pre-defined the courses, closed-ended questions assessed the relevance of target audiences, content, methods, and workload sufficiency, as well as the means of dissemination, educational resources, and post-course follow-up strategies. Open-ended questions requested additional comments. As for the 2014 public notice, which did not pre-delimit the courses, open-ended questions asked for information on the three most relevant courses: title, syllabus/content, methods, evaluation/follow-up strategies, target audience and the number of vacancies. Other open-ended questions requested comments on the relevance of target audiences, content, and workload. Then, coordinators were asked to make a general assessment of the RRC implementation: repercussions, feasibility of resource execution, timely articulations, current status of the RRC, and recommendations for improvement. In the end, questions were included aiming at the sociodemographic characterization of the coordinator.
We resorted to the free online tool Google Forms. A pilot study was conducted with the participation of three coordinators whose modification suggestions were fully accepted. The research project was submitted for analysis by the Research Ethics Committee before forwarding it to the coordinators, with a recommendation for development. The invitations to participate in the survey were sent by e-mail between April and June 2018, and contained data about the survey and the link to access the questionnaire to be completed. An available list that included a total of 59 RRCs deployed in the country was adopted.
The following axes were adopted to analyze the implementation of the Brazilian RRCs: (1) characterization of the coordinators and Centers; (2) characterization and analysis of the compo-nents of the training processes; (3) analysis of the challenges and potentialities of administrative and financial management; and (4) self-assessment of locoregional impacts. The analyses were triangulated from the same data source 14 in order to develop these four axes: descriptive statistical analysis of closed-ended questions, and thematic analysis of open-ended questions.
The descriptive statistical analysis (frequency, mean, and standard deviation) used the functions available in the Google Spreadsheet generated from the data entered in the questionnaire. The open-ended questions, in turn, were submitted to thematic analysis 15 subsidized by the use of Atlas.ti version 7 software. Each of the open-ended questions was transformed into a primary Atlas.ti document containing the answers given individually by the coordinators in order to perform the thematic analysis. The two-stage coding generated a well-structured code system 16,17 . Excerpts from the answers given to the open-ended questions were included in the presentation of results as an illustration of the analyses carried out, using a code to mention the coordinators, organized by sequential receipt of the questionnaire ("Cnn"), to ensure anonymity.
Fifty-four coordinators from 51 RRCs participated in the study, and information on three RRCs was provided by two coordinators each. A significant response rate of 86% was achieved taking the RRC as a unit of analysis. Tables 1 and 2 systematize the results regarding Axis 1 (characterization of coordinators and RRCs). The results of axes 2, 3, and 4 will be shown in the next section. Table 1 shows that the most significant number of RRCs were implemented in the Northeast (n = 16), followed by the Southeast (n = 14). The North region had the lowest number of RRCs deployed (n = 5), but very close to the Midwest (n = 6). Data are not surprising concerning the Southeast region, as it is the region with the most significant number and some among the most traditional public technical and higher education institutions. However, it is surprising to see the high number of proposals developed in the Northeast and North regions of the country, given the historical inequality of regional access to continuing education actions.

Characterization of coordinators and RRCs
Regarding the proposal submission public notice also in Table 1, practically half of the RRCs either submitted and developed proposals for the 2014 public notice (n = 13), or all three public notices (n = 12). One hypothesis for the high adherence to the 2014 Public Notice was the good receptivity to the methodological flexibility of the formative processes, indicated even in the thematic analysis of open-ended questions. Concerning the 12 RRCs that submitted proposals to the three Public Notices, no relevant differences were found between them concerning the region or specific characteristics of their coordinators. However, coordinators' mean work time in the mental health, alcohol, and other drugs fields in these 12 RRCs is 23.18 years (sd = 7.05), against 18.23 (sd = 8.82) of the group, suggesting that they are more experienced coordinators.
The age range of the coordinators is relatively well spread among the delimited decades, with a mean age of 50.54 years (sd = 9.98). Concerning the gender of coordinators, approximately two thirds (n = 36) were women. This frequency is consistent with the fact that 58.7% of the coordinators took their undergraduate courses in the areas of Psychology and Nursing, which have more female graduates historically.

Characterization and analysis of training process components
Data in Table 2 express the predominantly positive evaluation of coordinators regarding the target audience, content, methodology, and workload of the developed training processes. The assessment of the relevance of target audiences and seminars (as an element of the methodology) focuses overwhelmingly on "totally agree" and "agree". However, there is also a positive assessment, with a clear concentration around "agree", but including "disagree" (besides "strongly agree") regarding the adequacy of the expected workload and the content and methods to the target audience.
The thematic analysis performed contributed to elucidate these evaluations. First, concerning the virtually unanimous assessment of the relevance of the intended target audiences, we noted that the coordinators emphasized the role of interprofessional and intersectoral education. Notwithstanding the fact that several training courses were considered since the first public notice, the coordinators reported efforts to include in the courses participants in sectors not foreseen in the 2010 and 2012 Public Notices: education professionals, workers of the Judiciary, public security workers, among others: "The participation of criminal judges, prosecutors, and appellate judges contributed to the continuity of learning when, after the course, they put the acquired knowledge into practice" (C25). Again, concerning the target audience, the leadership of the students in the teaching-learning process stands out, using the active methodologies and a vast repertoire of educational resources: "The training process was carried out with active methodologies with content discussion together with the participants where real case studies were used" (C09).
Seminars and events in different formats, open to students and other audiences, were also  The courses offered by the RRC have had a significant impact on regional drug policies. evaluated in a notoriously positive way. Seminars were perceived as spaces that promote the reception of new students and meeting those who already attend them; sharing work experiences and intervention projects in the territory, and a space for articulation between partners and analysis of the situation of alcohol and other drugs policies: "we held a closing seminar in which the participants of the courses lead on, as they presented the intervention projects that were ongoing or being started" (C47). Again, concerning the methods, two other aspects may have contributed to a less homogeneously positive assessment of the methodology's adequacy vis-à-vis the target audiences. The first refers to the difficulties for greater involvement of the students: (a) obstacles to the removal for face-to-face participation, due to the questioning of previous agreements on change of management and the non-guarantee of removal for training due to the type of work relationship ("Drop-out also occurred during times of government transition or local political instability" (C12)); and (b) problems with the travel of students to the RRC host city to attend staged classroom activities ("there were many enrolled, but not all continued the course for various reasons, including distance", (C40)). The second aspect refers to the adjustments made to the preset workload for the courses (in particular in the proposals submitted to the 2010 and 2012 Public Notices) to effectively consider practical activities: "The workload could be up to 60 hours, and part was intended for practice. [...] We consider it very complicated to meet this issue, due to the difficulty of workers leaving their jobs for so long. Thus, we redesigned the courses for 20h theory and 20h practice" (C04).
The thematic analysis also allowed understanding the inhomogeneous assessment of the content's adequacy to the needs of the target audience. Thus, we noted, at first, that the approach to comprehensive care involves dealing with drugs as a taboo issue, encompassing, besides the transmission of knowledge, educational practices that promote awareness: "Although the improvement course is aimed at professionals who deal directly with the problem, dealing with the drug problem, in particular, the situation of drug users and their place as a subject of rights is still a big taboo" (C23).
Also, the adherence of students to the formative processes required managing the tension between the offer of general contents for beginners and specific ones for the most experienced ones: "Sensitized professionals, with professional practices using some of the mentioned approaches, demanded more content, deepening" (C12).
Concerning the evaluation strategies used in the formative processes, two aspects were highlighted in the thematic analysis: (a) the participatory nature of the evaluations, coherently with the valorization of the leadership of students ("The presence and participation in the course and elaboration of a problem-solving proposal related to the practice of professionals was considered in the evaluation", C28), and (b) the prevailing adoption of formative assessment, allowing the detection of weaknesses and learning needs throughout the courses, with the elaboration of intervention projects as a final product ("Besides the instrument that evaluated the activities, participants were assessed by presenting territorial analysis projects, for their applicability and involvement", C41). The elaboration of intervention projects facilitated the consideration of contents that are more in tune with the local reality, such as family, network articulation and drug use among indigenous, young and old people: "we incorporated aging-related modules into the course content, as well as drug use by this population contingent" (C03).

Analysis of management challenges and potentials
The coordinators' assessment of the proposed management, contained in Table 2, is at odds with the mostly positive assessment presented in the previous section on the adequacy of the target audience, content, methods, and workload of the courses implemented. Thus, when the smooth implementation of decentralized resources is stated, both positive ("totally agree" and "agree"), with 60.4% (n = 29), and negative ("totally disagree" and "disagree") assessments, with 39.6% (n = 19) are identified.
The thematic analysis provided an opportunity to proceed with an in-depth financial implementation. Part of the coordinators stated that they had fully implemented the resources and that their availability was timely and sufficient: "The project managed to implement all the committed resources, and there was no difficulty in meeting the budget target" (C42). A second part reports partial implementation of resources, due to (a) the slow availability of resources by SEN-AD, either in the initial budget decentralization or in the transfer of installments after decentralization; and (b) the slow internal administrative flow to the host institutions, for example, due to the need to commit spending in the budget year: "In our public IES, funds were credited in the institution's account and followed all the procedures, such as bid and specific outsourcing of some services [...] we experienced some difficulties and sometimes impediment to use these funds" (C12). A third party, in turn, mentions the impossibility of implementing one or more of the approved proposals, due to legal or institutional constraints, or impediments related to the support foundation: "The first course was started and close to its completion, the fund managing university foundation went bankrupt" (C28).
Besides the financial implementation, the coordinators highlighted elements of the administrative implementation of the courses. They believe that the way the educational institution received the RRC in its internal bodies was decisive for its development: "The [University] assumed the RRC, and given the confidence that the population has in the scientific institution, all managers signed the technical cooperation agreement whose counterpart would be to free professionals at work and allow research" (C25).
Also, the coordinators highlighted obstacles to the development of the proposals: (a) overlaps in the training agenda, with simultaneous provision of classroom training processes (such as those proposed by the RRCs and the Care Paths, for example), with the understanding that the SENAD could have had a mediating role ("At the time, a competition of offerings by the different managing levels was in place [...] including the Care Paths project, which as per the municipal managers, hindered the removal of professionals from duty", C31); and (b) lack of more direct technical-educational advice to the RRCs by SENAD to support the development of the formative processes ("I think it is indispensable to improve the follow-up and guidance process to run this vital national project", C04).
Finally, regarding the current situation of the RRCs, the coordinators point out two distinct situations: (a) Centers that are inactive, not currently developing any actions; and (b) Operating Centers. Among the latter, we report demotivated teams due to lack of funds, and relocation of the initial objectives: "Well, the curricular and extra-curricular continuing education demand persists, but is unfunded; it is complicated to attract teachers with the same commitment" (C04). As a result, they have focused their activities on academic-scientific events, at postgraduate education level or in research activities: "The RRC has been promoting annual Seminars/Colloquiums and conducting local research. However, we do not have any funds to offer systematic extension courses, such as that facilitated by the 2014 Public Notice" (C23).

Self-assessment of locoregional impacts
Regarding the perceived impacts, Table 2 expresses a positive assessment of the coordinators on the repercussions of the courses and the promotion of articulations in locoregional drug policies, addressing "totally agree" and "agree" options. The thematic analysis allowed identifying factors related to such impacts, from the coordinators' perspective.
We understand that the very opportunity to update knowledge and practices through the courses produces impacts, due to: (1) the affinity of the courses offered with drug policy reform, based mainly on the ethics of comprehensive care and human rights, and (2) the adoption of theoretical-practical activities, supporting the elaboration and development of singular therapeutic projects and in loco clinical-institutional oversight activities to strategic services, such as CAPSad: "In our region, CAPSad opening was taking place in several municipalities. These courses were important for the training and support of new teams and update and exchange space for professionals" (C54).
Concerning the teaching-service relationship, the following impacts were highlighted: (a) opportunities for reflection on prevailing prevention and care practices through the development of intervention projects in the territories; (b) elaboration of new training proposals, such as a lato sensu postgraduate course, due to the rapprochement of RRCs teams of university professors who previously worked in isolation; and (c) resumption of ties with the academic formation by the students due to the internalization allowed by the offer of formative processes in municipalities far from the major centers: "The regional articulations were significant concerning University reference to the municipalities and the bodies involved in drug policy in the region, so much so that the public notice was completed, and we continue to be asked for courses, research, extensions, among others" (C05).
The articulations in and of the network facilitated through the courses offered were also highlighted. Coordinators believe that this led to a more in-depth dialogue with managers, which minimized stigmas related to drug users, and supported sustaining changes in locoregional drug policies: "Managers, professionals and service users, as well as faculty and safety staff, had to talk to enable the holding of the Intervention Project in their workplaces" (C53).
The RRCs also mediated conflicts (travel, removal, among others) between government levels in order to increase participation opportunities: "A new agreement of the course coordination was required with SENAD and the State Drug Secretariat for the development of the program in other municipalities" (C09). Also, some RRCs worked with the care network, proposing intersectoral forums on alcohol and other drugs, thus contributing to the integration of local public policies: "Forums that gathered professionals and users were implemented, establishing care networks" (C36).
Finally, four suggestions for improvement to broaden impacts were generally pointed out:

Discussion
The analysis allowed observing the effectively national scope of the implemented RRCs, based on the expanded capillarity of public technical and higher education institutions. The seminars held by the RRCs and the intervention projects in the territory stood out for having transcended their most direct pedagogical function by influencing the articulation of the intersectoral network. The relevance of the methodological flexibility in the course design was also evidenced, allowing the consideration of specific contents of the local reality of the drug policies. Coordinators stressed the importance of greater administrative and financial autonomy concerning proposal management. Finally, the role of RRCs as stakeholders in locoregional drug policies was highlighted.
We observed that RRCs that submitted and developed proposals for the three Public Notices were coordinated by people with more experience in the field of mental health, alcohol, and other drugs. Gaining experience with administra-tive and financial execution is crucial in program implementation, given the relevance of developing the institutional capacity of coordinators as implementing agents 13 . Thus, we suggest that the findings of this research be taken into account when designing future public notices. The fact that two-thirds of the coordinators were women confirms the studies that point out the increased presence of women as leaders of research groups, among the best qualified academic strata of Brazilian science 18,19 .
Regarding the regional distribution of the implemented proposals, the expanded public network of federal higher education institutions (IFES) and federal institutes of education, science and technology (IFET) from 2003 to 2011 20 may have contributed to the implementation of a significant number of RRCs in the North and Northeast regions of the country. Another element that possibly also intervened was innovations in drug policies in some of the capitals and leading regions of northeastern Brazil, such as the "More Life" and "Comprehensive Care to Drug Users and Family Members" (ATITUDE) programs of the Pernambuco State Government 21 , and the pioneering projects of the Center for Drug Abuse Studies and Therapy (CETAD-UF-BA), such as the "Street Office" (later elevated to SUS services) and "Citizenship Point" 22 .
The adoption of analysis axes that prioritized a data aggregate analysis was necessary due to the changes observed in each new public call for proposals, especially in the 2014 Public Notice, which introduced greater methodological flexibility. Despite being well evaluated, these changes created difficulties for disaggregated analysis, at every public notice, making their comparability difficult. Thus, implementation studies allow highlighting the political and institutional conditions of the management and development of public policy interventions 13 ; however, they provide limited data regarding impacts and results achieved, which would require the construction of a concomitant evaluation matrix at the onset of the implementation of the proposal.
The analyses carried out corroborated the obstacles to the more effective impacts of training policies in the field of alcohol and other drugs signaled by other studies: (a) high turnover of SUS and SUAS teams due to the weak work ties and the "seasonal political game" 5 (p. 181); (b) intra-and intersectoral fragmentation of public policy actions; (c) devaluation of phenomena in the field of alcohol and other drugs by managers with the presence of "repressive logic residues" 5 (p. 181); and (d) negligible investment in the management of health education at the municipal level. The coordinators stressed the relevance of questioning moral positions and the mediation of conflicts 23 in pedagogical work and with managers.
The lack of a well-defined normative and evaluative parameter hinders judgments. However, the research developed indicates quantitative and qualitative evaluation indicators. Briefly, indicators "[are] some kind of flags that seek to synthetically express an aspect of (variable) reality" 24 (p. 106). Although indicators must be conceived in a contextualized way and in the light of the singular reality to be evaluated, the naturalistic generalization of the results of this case study enables future evaluations of process and results to be supported by it.
Giving a voice to a stakeholder with an essential responsibility for the implementation at the local level allows moving beyond the "black box" of assessments that tightly divide the moments of formulation and implementation into a classic policy cycle perspective 13 . Again concerning the method, the easy application of the instrument was considered an additional advantage, bypassing the lack of interactivity inherent to the use of surveys.
The use of coordinators as the only informants of the analysis did not allow confrontation and comparison with the perspective of other critical stakeholders for the implementation of training proposals, such as teachers, students, and managers of educational institutions and public policies at their various levels and sectors. However, it allowed incorporating a segment with a relevant intrinsic motivation for participation 25 , enabling a practically census-like sampling. Further studies should consider the perspective of the other stakeholders involved, such as managers, participants, and other publics reached.

Final considerations
There is a common place among several publications related to the field of drug policy: most authors recognize the inadequacy of training on this topic [1][2][3][4][5] . As a result, extensive efforts have been made to overcome this critical node for psychosocial care to drug users. In this context, the role of RRCs in drug policies was evidenced by the role played not only in their most immediate goal of providing knowledge update for the agents involved in prevention and care actions but also in acting effectively in the articulation of locoregional intersectoral networks. Despite the significant impact, RRCs are currently experiencing the repeated problem of discontinuity in the planning and implementation of public policies in our country when management changes 26 .
As this is an assessment of implementation, the analysis focused on the components of the formative processes, not addressing the proposals themselves, concerning consistency with the continuing or permanent education proposals. However, the articulation of the proposals with "practical activities" (usually related to services), especially those deriving from the proposals presented to the 2014 Public Notice, suggests their coherence with continuing education. However, this analysis would require longitudinal follow-up to gain greater consistency.
The Brazilian drug policy adopted a new direction in 2016. Since then, no funds have been made available to secure continuing training in the same perspective as it has been built. Support modalities must be resumed to sustain the functioning of both the implemented RRCs and the new Centers, given the persisting needs that prompted their proposal. Nevertheless, the experience accumulated by RRCs should be considered in proposing principles, guidelines, and priorities for a national policy of continuing education in the field of alcohol and other drugs, yet to be guided in our country.