Explore the vast range of titles available.

Dialogue policy learning (DPL) is a key component in a task-oriented dialogue (TOD) system. Its goal is to decide the next action of the dialogue system, given the dialogue state at each turn based on a learned dialogue policy. Reinforcement learning (RL) is widely used to optimize this dialogue policy. In the learning process, the user is regarded as the environment and the system as the agent. In this paper, we present an overview of the recent advances and challenges in dialogue policy from the perspective of RL. More specifically, we identify the problems and summarize corresponding solutions for RL-based dialogue policy learning. In addition, we provide a comprehensive survey of applying RL to DPL by categorizing recent methods into five basic elements in RL. We believe this survey can shed light on future research in DPL.

Background In the aftermath of the COVID-19 pandemic, health policymakers have been reflecting upon sustainability and resilience issues in health systems worldwide. Promoting sustainability and resilience requires policy changes built upon evidence and on the views of health stakeholders and experts. This study aimed to engage health stakeholders in designing and discussing policy recommendations with a high potential to improve sustainability and resilience in the Portuguese healthcare system. Methods As part of the Partnership for Health System Sustainability and Resilience initiative in Portugal (PHSSR-PT), this study proposes a novel policy dialogue that combines participatory methods—workshops and Web-Delphi processes—with content analysis tools—namely Dialogue Mapping—to promote agreement and help health stakeholders and experts to identify and discuss policy recommendations with high potential to improve health systems’ sustainability and resilience. Departing from the COVID-19 pandemic as a critical event and drawing on evidence and data, a group of health stakeholders and experts ideated and agreed upon high-value policy recommendations across seven domains: Governance, Financing, Workforce, Medicines and Technology, Service Delivery, Population Health and Environmental Sustainability. Results 40 top-level Portuguese health stakeholders and experts successfully collaborated to generate and discuss the benefits, risks and implementation issues of 69 policy recommendations, out of which 43 were selected as having gathered a high level of agreement on their potential to improve system sustainability and resilience. The adopted policy dialogue promoted high convergence. Many of these 43 recommendations were shown to entail interconnectedness with other policy recommendations. Conclusions This study provides actionable insights to advance discussions on sustainability and resilience in Portugal. It shows that Governance and Population Health policy recommendations are critical to improve sustainability and resilience in several domains, and there is a high level of agreement on the need to adopt many recommendations, but questions remain about their implementation. The study also shows that new ways of engaging health stakeholders and experts can be adopted to promote dialogue, consensus and transparent discussion in policy processes.

Background Policy dialogue, a collaborative governance mechanism, has raised interest among international stakeholders. They see it as a means to strengthen health systems governance and to participate in the development of health policies that support universal health coverage. In this context, WHO has set up the Universal Health Coverage Partnership. This Partnership aims to support health ministries in establishing inclusive, participatory, and evidence-informed policy dialogue. The general purpose of our study is to understand how and in what contexts the Partnership may support policy dialogue and with what outcomes. More specifically, our study aims to answer two questions: 1) How and in what contexts may the Partnership initiate and nurture policy dialogue? 2) How do collaboration dynamics unfold within policy dialogue supported by the Partnership?  Methods We conducted a multiple-case study realist evaluation based on Emerson’s integrative framework for collaborative governance to investigate the role of the Partnership in policy dialogue on three policy issues in six sub-Saharan African countries: health financing (Burkina Faso and Democratic Republic of Congo), health planning (Cabo Verde, Niger, and Togo), and aid coordination for health (Liberia). We interviewed 121 key informants, analyzed policy documents, and observed policy dialogue events. Results The Partnership may facilitate the initiation of policy dialogue when: 1) stakeholders feel uncertain about health sector issues and acknowledge their interdependence in responding to such issues, and 2) policy dialogue coincides with their needs and interests. In this context, policy dialogue enables stakeholders to build a shared understanding of issues and of the need for action and encourages collective leadership. However, ministries’ weak ownership of policy dialogue and stakeholders’ lack of confidence in their capacity for joint action hinder their engagement and curb the institutionalization of policy dialogue. Conclusions Development aid actors wishing to support policy dialogue must do so over the long term so that collaborative governance becomes routine and a culture of collaboration has time to grow. Public administrations should develop collaborative governance mechanisms that are transparent and intelligible in order to facilitate stakeholder engagement.

Background A policy dialogue is a tool which promotes evidence-informed policy-making. It involves deliberation about a high-priority issue, informed by a synthesis of the best-available evidence, where potential policy interventions are discussed by stakeholders. We offer an ethical analysis of policy dialogues – an argument about how policy dialogues ought to be conceived and executed – to guide those organizing and participating in policy dialogues. Our analysis focuses on the deliberative dialogues themselves, rather than ethical issues in the broader policy context within which they are situated. Methods We conduct a philosophical conceptual analysis of policy dialogues, informed by a formal and an interpretative literature review. Results We identify the objectives of policy dialogues, and consider the procedural and substantive values that should govern them. As knowledge translation tools, the chief objective of policy dialogues is to ensure that prospective evidence-informed health policies are appropriate for and likely to support evidence-informed decision-making in a particular context. We identify five core characteristics which serve this objective: policy dialogues are (i) focused on a high-priority issue, (ii) evidence-informed, (iii) deliberative, (iv) participatory and (v) action-oriented. In contrast to dominant ethical frameworks for policy-making, we argue that transparency and accountability are not central procedural values for policy dialogues, as they are liable to inhibit the open deliberation that is necessary for successful policy dialogues. Instead, policy dialogues are legitimate insofar as they pursue the objectives and embody the core characteristics identified above. Finally, we argue that good policy dialogues need to actively consider a range of substantive values other than health benefit and equity. Conclusions Policy dialogues should recognize the limits of effectiveness as a guiding value for policy-making, and operate with an expansive conception of successful outcomes. We offer a set of questions to support those organizing and participating in policy dialogues.

In recent decades, the number of people living with one or more chronic diseases has increased dramatically, affecting all sectors of society, particularly the labour market. Such an increase of people with chronic diseases combined with the aging of working population affects income levels and job opportunities, careers, social inclusion and working conditions. Both legislation and company regulations should take into account the difficulties that workers experiencing chronic diseases may face in order to be able to formulate innovative and person-centred responses to effectively manage this workforce while simultaneously ensuring employee wellbeing and continued employer productivity. The European Joint Action “CHRODIS PLUS: Implementing good practices for Chronic Diseases” supports European Union Member States in the implementation of new and innovative policies and practices for health promotion, diseases prevention and for promoting participation of people with chronic diseases in labour market. Therefore, a Toolbox for employment and chronic conditions has been developed and its aim is to improve work access and participation of people with chronic diseases and to support employers in implementing health promotion and chronic disease prevention activities in the workplace. The Toolbox consists of two independent instruments: the Training tool for managers and the Toolkit for workplaces that have been tested in different medium and large companies and working sectors in several European countries.

Sound economic and social policies are important if countries wish to prosper and achieve sustainable development. It is far from guaranteed, however, that policymakers select and implement good policies, which provides a rationale for external policy support. Indeed, many organizations are engaged in supporting policy reform processes in recipient countries. This study investigates the limits and opportunities of supporting policy reform by focusing on four dimensions of support: conditional financing, policy dialogue, technical evidence and political institutions. Four findings follow from a review of the literature. First, without commitment on the recipient side, conditional financing is unlikely to induce policy reform. Second, when external actors acquire a seat at the policy dialogue table, it is important to detect (and influence) the beliefs policymakers hold. Third, outside parties should bring sound evidence to the table about the costs, benefits, and effectiveness of their policy proposals. Finally, supporting changes in political institutions without considering general equilibrium effects can be counterproductive. The study concludes with a discussion and some avenues for future research in this field.

Background The prevalence of childhood obesity (CO) and related complications is high and alarmingly increasing in Iran. This study applied a mixed Delphi & Policy Dialogue approach to exploring and prioritizing policy options to control childhood obesity in Iran. Methods This study is organized in three Delphi phases followed by a policy dialogue session. This study applied the advocacy collation framework and evidence-informed policy-making approach to enhance the chance of a feasible and acceptable policy package. The first step consisted of interviews with 30 experts and primary stakeholders. Based on their answers and a comprehensive literature review, a list of presumed effective policy options to combat CO in Iran was made. Then, panelists were asked to score each policy option using a five-point Likert scale in seven constructs. To maximize the spread of opinions, panelists were chosen to represent three perspectives: policy-makers at different levels, presidents of various organizations who would implement potential policy options, and academics. Twenty-one stakeholders were invited to discuss the policy options in a policy dialogue section. Results We introduced 27 policy options and asked stakeholders to rank them using seven criteria on a five-level Likert scale. Totally, 41 experts participated in round 2 (66.2% response rate), and 33 experts took part in round 3 (72% response rate). Participants believed that healthy schools, creating healthy environments in kindergartens and other child care centers, subsidizing healthy foods, educating healthy lifestyles in mass media, and increasing access to physical activity facilities are the most effective and feasible policies in controlling CO. After the policy dialogue, the healthy school remained the most prioritized policy. a policy package to combat CO in Iran was designed with the participation of all stakeholders. Conclusion The advocacy collation framework and the evidence-informed policy-making approach were used to draft a policy package to combat CO, increasing the acceptability and feasibility of the developed policy package. Graphical Abstract Highlights Childhood obesity is a gateway to several diseases in the near and distant future. The prevalence of childhood obesity is high and is increasing alarmingly in Iran. Several policy options are proposed by scientists and other stakeholders to combat childhood obesity in Iran. Stakeholders believed that healthy schools, creating healthy environments in kindergartens and other child care centers, subsidizing healthy foods, educating healthy lifestyles in mass media, and increasing access to physical activity facilities with priority given to deprived areas are the most effective and feasible policies in controlling childhood obesity.