The Evidence-Informed Policy Network (EVIPNet) in Chile: lessons learned from a year of coordinated efforts

ABSTRACT Informing the health policymaking process with the best available scientific evidence has become relevant to health systems globally. Knowledge Translation Platforms (KTP), such as the World Health Organization’s Evidence Informed Policy Networks (EVIPNet), are a recognized strategy for linking research to action. This report describes the experience of implementing EVIPNet in Chile, from its objectives, organizational structure, strategy, activities, and main outputs, to its evolution over the course of its first year. Lessons learned are also covered. Of the activities initiated by EVIPNet-Chile, the Rapid Response Service proved to be a good starting point for engaging policymakers. Capacity building workshops and policy dialogues with relevant stakeholders were also successful. Additionally, EVIPNet-Chile developed a model for engaging academic institutions in policymaking through a network focused on preparing evidence briefs. A number of challenges, such as changing methods for producing rapid evidence syntheses, were also identified. This KTP implementation model located in a Ministry of Health could contribute to the development of similar initiatives in other health systems.

Decision makers use a number of different inputs to address a wide range of questions about the relevance and size of a policy issue, the impacts of different policy options, and the implementation of these options in their health systems. Research evidence is one of these key inputs in the policymaking process, and could contribute to making better decisions. Although efficient use of available resources has special relevance for Lowand Middle-Income Countries (LMICs), it also pertains to High-Income Countries (HICs) where policymakers strive to make the best use of resources.
In this context, the World Health Organization (WHO) 2004 Annual Report included a chapter on the need for linking research to action (1). Later, in May 2005, the World Health Assembly encouraged countries to "establish or strengthen mechanisms of knowledge-transfer to support public health development, health-related policies, and evidence-based health systems" (2). Afterwards, this call was reinforced by the 2008 Bamako "Call to Action on Research for Health" (3), and more recently, by the 2013 World Health Report (4). The 2013 report issued a strong call for "closer collaboration between researchers and policymakers, i.e., research needs to be taken outside the academic institutions and into public health programs that are close to the supply of and demand for health services" (4)-in order to obtain Universal Health Coverage.
Despite these global calls to use scientific knowledge in policymaking, research evidence has not been systematically used to make recommendations (5). Although policymakers are actually using the available evidence to make decisions, there is still a gap between the availability of scientific knowledge and its systematic use across different levels of the health system, including the policymaking process (6).
Evidence Informed Health Policymaking (EIHPM) aims to ensure that the decision-making process is systematically and transparently informed by the best available scientific evidence (7). The community and civil society could also become part of EIHPM by representing stakeholders in a variety of health policies, transforming policymaking into a participative process (8).

KNOWLEDGE TRANSLATION PLATFORMS AND EVIPNET
Knowledge Translation (KT), defined as "a dynamic and iterative process including synthesis, dissemination, exchange, and application of knowledge in order to improve population health, provide more effective health services and products, and strengthen health systems," is closely related to EIHPM (9). In this sense, the development and implementation of Knowledge Translation Platforms (KTP)-partnerships among policymakers, researchers, civil society organizations, and other stakeholders that promote the use of evidence in policymaking-provide the infrastructure for country-level efforts to link research to action (10).
A number of experiences in LMIC countries-e.g., Evidence Informed Policy Network (EVIPNet) in Cameroon (11), Regional East African Community Health Policy Initiative (REACH-PI) in Uganda (11), and Zambia Forum for Health Research (ZAMFOHR) in Zambia (12)-and in high-income countries-e.g., McMaster Health Forum in Canada (13), Center on Knowledge Translation for Disability and Rehabilitation Research (KTDRR) in the United States of America (14), and EVIPNet in Europe (15)-have shown the feasibility of implementing this type of initiative. The EVIPNets are KTPs sponsored by WHO with variable levels of activity across the different WHO Regions. EVIPNet promotes partnerships at the country level among policymakers, researchers, and civil society, in order to facilitate both policy development and implementation using the best scientific evidence available. EVIPNet comprises networks that bring together countrylevel teams that are coordinated at both the regional and global levels.

PROMOTING USE OF EVIDENCE FOR POLICYMAKING IN CHILE
Chile has some experience using evidence in policymaking. The processes of formulating and evaluating the National Health Objectives for 2000-2010 (16) and the National Health Plan for 2011-2020 (17) were informed by local and international evidence. In 2004, a law that established "Explicit Guarantees in Health" for 80 prioritized health conditions took into account local prevalence and incorporated specific guidelines for clinical practice for each condition. In 1997, the Ministry of Health (MoH) created a Health Technology Assessment (HTA) unit. In 2012, a National Committee was officially formalized to propose an institutionalization plan for HTA. Currently, this Committee is cooperating with the HTA unit to expand its scope to develop evidence-based recommendations for coverage decisions through the recently created Financial Protection System for High Cost Diagnostics and Treatments.
Between 2010 and 2013, the Secretariat of EVIPNet Americas organized a number of training activities in this WHO Region, aiming to promote EIHPM and to build capacity on preparing evidence briefs for policy (18) and organizing policy dialogues (19). Some of these activities were carried out and coordinated by a group of Chilean researchers interested in the field (20).
Despite these initiatives, the scientific evidence was still not being used systematically in every policy decision in Chile. Moreover, research was not always an important input for decision makers and health policy. Finally, in 2014, the MoH opened a full-time position exclusively dedicated to establishing an EVIPNet working group in Chile. This paper describes the development and evolution of EVIPNet-Chile and identifies some of the lessons learned and challenges met after a year of experience. It describes the process of developing this initiative within the MoH, including objectives, organizational structure, strategy, activities, and main outputs during the period from October 2014 to October 2015.

Governance, structure, and stakeholders
The structure of a KTP within a country could have an important impact for EIHPM. However, there is not a unique "organizational solution" to where the platform should be located in the national health system. There are examples of KTPs in universities, non-governmental organizations, government institutions, and public agencies (21).
EVIPNet-Chile is coordinated by a Secretariat hosted by the Cabinet of the Minister at the MoH. This hosting enables a direct connection with the highest authorities of the MoH and other secretariats, with policymakers, local governments, public health care providers, civil society, and other stakeholders in health policymaking. In addition, EVIPNet-Chile Secretariat facilitates a fluid dialogue with universities and research groups ( Figure 1).
As a first step, in 2015, EVIPNet-Chile initiated a partnership with the Network of Schools and Departments of Public Health at a number of universities in Chile. This partnership-called the EVIPNet-Chile Network-is supported by the Country Office of the Pan American Health Organization (PAHO). It aims to expand the capacity of EVIPNet-Chile, to organize activities and prepare evidence syntheses to policymakers, and increase the numbers of those trained (critical mass) in EIHPM concepts across the country.

Strategy
EVIPNet-Chile aims to support the health policymaking process, in order to improve public health and reduce health inequities in Chile. It has three main strategic objectives: i. To use the best available evidence to produce outputs that can inform decisions in the health policymaking process. ii. To promote the systematic use of evidence in the health policymaking process. iii. To promote collaboration among policymakers, researchers, and civil society organizations.
In order to accomplish these objectives, EVIPNet-Chile has outlined five main types of activities/products (Table 1): i. Rapid Response Service for rapid evidence synthesis ii. Website (one-stop shop) iii. Capacity-building workshops iv. Evidence briefs for policy v. Policy dialogues The overall strategy of EVIPNet-Chile is organized in a matrix where each activity/product is aligned with each objective. The matrix shown in Table 2 is a visual display of the relationship between objective and activity/product. This matrix allows identification of a portfolio of activities/products to address specific objectives. For instance, when a team is particularly interested in promoting collaboration among policymakers, researchers, and civil society, its time would be best invested in writing rapid evidence syntheses, preparing evidence briefs for policy, and/or organizing policy dialogues, rather than website development or capacity-building workshops. This matrix also allows more efficient monitoring and evaluation of EVIPNet-Chile activities, allocating indicators, results, and targets to each activity-objective.

Activities
Although there is a variety of existing activities in which a KTP might engage, in Chile the aforementioned five were chosen as a good starting point, and a balanced combination that would address all the strategic objectives.
Rapid Response Service (preparing rapid evidence syntheses). On a daily basis, decision makers require urgent evidence-informed answers to a number of policy questions. The Rapid Response Service aims to improve evidence accessibility to them by significantly reducing the time needed to summarize the existing research. The EVIPNet-Chile Secretariat prepares rapid evidence syntheses (summaries of the impact of specific interventions) in less than 20 working days, based mainly on relevant systematic reviews. The depth of the evidence analysis depends on the time available to deliver the product (Figure 2).
Although the types of questions that the Rapid Response Service answers are mainly related to the impact of health policies or health system interventions, decision makers' needs are frequently broader. Therefore, many requirements start with an unstructured question that needs to be clarified and framed before the Rapid Response Service can provide an answer. In its first 12 months of operation, the Rapid Response Service   (Table 3). The process of preparing a rapid evidence synthesis shares several steps with that of preparing a systematic review (22). However, there is not an agreed-upon method to write a rapid evidence synthesis, and methodological approaches vary considerably (23)(24)(25). The methods used by the EVIPNet-Chile Secretariat to produce rapid evidence syntheses have been continuously modified to consider the specific challenges faced in preparing each of them. Nonetheless, the process of developing a handbook is underway and will provide a step-by-step guide to produce this type of product systematically. -stop shop). One-stop shops are a useful strategy for improving accessibility to research evidence by health policymakers. There are a number of successful initiatives in this area, such as Health Systems Evidence in Canada (26), PDQ-Evidence (27), and CIPHER in Australia (28).

Website (one
Following EVIPNet Global templates, EVIPNet-Chile developed a website   accessible from within the MoH. This website provides access to the rapid evidence syntheses prepared by the team (Table 3), as well as evidence briefs for policy, policy dialogue summaries, and workshop presentations. In addition, the EVIPNet-Chile website allows users to access relevant KT resources, such as news, publications, multimedia, and events, and offers a specialized search engine for locating relevant literature. There is a slightly different version of the website accessible from outside the MoH. From its launch in February 2015, through October 2015, the EVIPNet-Chile website had been visited more than 1 000 times from within Chile alone. Capacity-building workshops. Since 2014, EVIPNet-Chile has organized four half-day workshops within the MoH, and one with the EVIPNet-Chile Network. These workshops are designed to build capacity on the systematic and transparent use of evidence in policymaking. There are two types of workshops: • Driven by the EVIPNet-Chile Secretariat: workshops that address the general concepts of EIHPM and the use of systematic reviews. These workshops are built mainly from contents developed by the team at the McMaster Health Forum (13). • User-demanded: workshops that address specific training needs. For example, methods to produce rapid evidence syntheses, requested by the HTA Unit at the MoH, and methods for preparing evidence briefs for policymakers in the EVIPNet-Chile Network.
These workshops have reached more than 30 of the almost 50 eligible professionals at the MoH and 20 members of EVIPNet-Chile Network across the country.
Preparation of evidence briefs for policy. These evidence summaries take a policy problem/issue as a starting point, describe its underlying factors, and frame a number of relevant policy options while identifying barriers and facilitators to their implementation. Then the focus turns to finding and distilling the full range of research evidence relevant to the various features of the problem/ issue, such as the impact of the different options or the effectiveness of implementation strategies (18).
The problem identified for an evidence brief should be broad and also, a political priority. In addition, it should be developed together with the MoH "issue-owner," i.e., those charged with managing the issue.
As mentioned before, participation in the EVIPNet-Chile Network is a key element in this process. Evidence briefs for policy are mainly prepared by academic teams that have attended a capacity-building workshop organized by the EVIPNet-Chile Secretariat. The PAHO Country Office has also played a crucial role in this partnership by supporting the preparation of these products. The policy issues addressed by the evidence briefs for policy are related to the MoH priorities and communicated to the network by the EVIPNet-Chile Secretariat.
Organizing policy dialogues. Structured discussions about an evidence brief for policy or a rapid evidence synthesis can contribute to the development of evidence-informed health policies, helping to clarify the problem and solutions, and to develop a shared understanding among stakeholders (19,29).
Over the course of the first year, EVIP-Net-Chile organized three policy dialogues based on three rapid evidence syntheses: human milk banks, female condoms, and acupuncture (see Table 3). Health care providers, ministries, nongovernmental organizations (associations of patients and workers), academic institutions, and scientific societies have attended these meetings, contributing to the deliberations and adding inputs to the process, beyond scientific evidence.

DISCUSSION
Implementing KTPs is an effective strategy to inform decision-making processes with evidence, providing an efficient route to improving health policies, especially in LMICs (30). EVIPNet-Chile provides a clear experience of how a KTP can be established and institutionalized within a MoH and is a good example of a working KTP with a portfolio of different activities and products. After this first year, lessons learned from the experience in Chile can be summarized in five main points.
First, EVIPNet-Chile provides a concrete example of how a KTP can be established and institutionalized within a MoH. This model has allowed a closer relationship with policymakers and more permanent relationships within the Government, a main objective of a KTP (21).
Second, the balanced portfolio of initiatives that a KTP decides to incorporate is very important to comprehensively addressing its objectives. Although EVIPNet-Chile has defined five main activities/products, there are others that could be used to further develop EIHPM, such as summaries of systematic reviews (31), citizen panels (32), and communities of Practice (33).
Third, the Rapid Response Service has been the most used and valued activity in our MoH because it is perceived as providing timely evidence for policymaking. A rapid evidence synthesis is a very efficient instrument for engaging policymakers with KT, since it quickly gets an evidence-informed answer to a specific policy question. In this sense, Rapid Response Services can be a very good starting point when introducing a KTP within a MoH. Although EVIP-Net-Chile has used mainstream methods, mainly based on the use of systematic reviews, the specific processes have continuously changed over the study period. We are now using rapid evidence syntheses as an input for policy dialogues; this represents an innovation to prior KT research (19,29).
Fourth, collaboration with academic groups arose from the EVIPNet-Chile Secretariat's need for human resources; previously, their availability was not enough to prepare evidence briefs for policy. A lack of resources for any KTP, especially initially, can be used as an opportunity to expand the network by engaging other stakeholders in the process.
Lastly, it is important for KTP professionals to be up-to-date in terms of current technologies and methodologies that could be used to better link research to action in policymaking. The EVIP-Net-Chile Secretariat has been paying close attention to new developments and innovations in this area. Some examples are Epistemonikos (evidence matrix systematic reviews with primary studies cited) (27), Health Systems Evidence new web interface (34) (enhanced features), and RevMan (to develop meta-analysis for rapid evidence synthesis) (35).

Barriers and challenges
There are important barriers that we addressed during this first year. For instance, although the EVIPNet-Chile Network has been positively evaluated to date, it is necessary to consolidate the relationships among KT, the policymaking process, and academic institutions. This network has been a major advance in terms of preparing evidence briefs for policy, but the work could be expanded into other new activities/products in the future.
Secondly, as a new program inside a Ministry, it is necessary to make a cultural change for promoting the use of evidence in policymaking processes. This is generally a hard process, considering the existing different uses of evidence. In this sense, Rapid Response Service has been a useful tool, consolidating an important position within the MoH.
Lastly, uses of evidence in the different stages of the policymaking process have also been challenging. The balanced combination of rapid evidence syntheses, evidence briefs for policy, and policy dialogues has successfully responded to the needs of policymakers and relevant stakeholders in a variety of scenarios and stages of the policymaking process.
A number of challenges should be considered when implementing a KTP. First, as a part of the KT process, it is necessary to have a plan for incorporating new developments on EIHPM. For example, text mining (36) and living evidence (37) are current evolving topics worth incorporating in the foreseeable future in our portfolio of methods. Acting on what has been learned from past experiences with evidence briefs is also imperative (38).

Conclusions and recommendations
A KTP can certainly be enriched by collaboration with international organizations, such as PAHO and WHO, but also by engaging with other countries of Latin America that are working on similar initiatives. Also, closer contact with non-governmental organizations working in this field, such as the Cochrane Collaboration, would be a major improvement, especially in terms of developing methodologies. The Rapid Response Service was the most challenging part of this effort because there is no standard method for producing the summaries; regardless, the service has been widely demanded within the MoH.
Finally, in order to concretely evaluate the impact of EVIPNet-Chile on the policymaking process, it is necessary to conduct rigorous evaluations. A good starting point could be an analysis of the strategy matrix presented in Table 2, defining indicators related to each activity. Such analysis would determine how well the objectives are being addressed by the work of EVIPNet-Chile, and how its implementation has specifically improved evidence-informed policymaking in our health system.

Conflict of interests. None declared.
Disclaimer. Authors hold sole responsibility for the views expressed in the manuscript, which may not necessarily reflect the opinion or policy of the RPSP/ PAJPH and/or PAHO.