LESSONS FROM THE FIELD
Creating windows of opportunity for policy change: incorporating evidence into decentralized planning in Kenya
Créer des opportunités de changement politique : prise en compte des éléments factuels dans la planification décentralisée au Kenya
Crear de oportunidades de cambio normativo: incorporación de evidencia a la planificación descentralizada en Kenya
Lori S Ashford1; Rhonda R Smith; Roger-Mark De Souza; Fariyal F Fikree; Nancy V Yinger
Population Reference Bureau, Suite 520, 1875 Connecticut Avenue, NW, Washington, DC, 20009, USA
ABSTRACT
PROBLEM: Because researchers and policy-makers work in different spheres, policy decisions in the health arena are often not based on available scientific evidence.
APPROACH: We describe a model that illustrates the policy process and how to work strategically to translate knowledge into policy actions. Several types of activity agenda-setting, coalition building and policy learning together can create a window of opportunity for policy change.
LOCAL SETTING: Activities were undertaken as part of the Kenyan Ministry of Health's new decentralized planning-process. The objective was to ensure that the results of a national assessment of health services were used in the preparation of district-level health plans.
RELEVANT CHANGES: Following the intervention, 70 district-level, evidence-based work plans were developed and approved by the Kenyan Ministry of Health.
LESSONS LEARNED: Substantial investment and effort are needed to bring stakeholders together to work towards policy change. More in-depth evaluation of these efforts can aid understanding of how systematic approaches to policy change can be replicated elsewhere.
RÉSUMÉ
PROBLÉMATIQUE: Les chercheurs et les responsables politiques interviennent dans des sphères différentes et les décisions politiques dans le domaine de la santé ne sont souvent pas fondées sur les données scientifiques disponibles.
APPROCHE: Le modèle décrit présente les procédures politiques et les stratégies pour transposer les connaissances en mesures concrètes. Plusieurs types d'activité établissement d'un programme, regroupement des partisans du changement et transposition des connaissances en éléments décisionnels aboutissent conjointement à la possibilité de modifier la politique suivie.
CONTEXTE LOCAL: Des activités ont été entreprises dans le cadre du nouveau processus de planification décentralisée du Ministère kényen de la Santé. Il s'agissait de s'assurer de l'utilisation des résultats d'une évaluation nationale des services de santé pour préparer des plans sanitaires à l'échelle du district.
CHANGEMENTS PERTINENTS: A la suite de l'intervention, 70 plans de travail élaborés à partir d'éléments factuels à l'intention d'un district ont été approuvés par le Ministère kényen de la Santé.
ENSEIGNEMENTS TIRÉS: Un investissement et des efforts considérables s'imposent pour amener les différentes parties intéressées à collaborer en faveur d'un changement de politique. Une évaluation plus approfondie de ces efforts permettrait de mieux comprendre comment des approches systématiques favorisant l'évolution des politiques pourraient être appliquées de manière similaire à d'autres contextes.
RESUMEN
PROBLEMA: Dado que los investigadores y las instancias normativas trabajan en diferentes esferas, ocurre con frecuencia que las decisiones de política adoptadas en el terreno de la salud no se basan en las pruebas científicas disponibles.
MÉTODOS: Describimos un modelo que ilustra el proceso de acción normativa y la manera de trabajar estratégicamente para traducir los conocimientos en medidas de política. Combinando varios tipos de actividad -establecimiento de la agenda, formación de coaliciones y aprendizaje de políticas- se pueden generar oportunidades para introducir cambios normativos.
CONTEXTO LOCAL: Como parte del nuevo proceso de planificación descentralizada del Ministerio de Salud de Kenya, se emprendieron diversas actividades con el objetivo de que, a la hora de preparar los planes de salud de distrito, se utilizaran los resultados de una evaluación nacional de los servicios de salud.
CAMBIOS DESTACABLES: Tras la intervención, el Ministerio de Salud de Kenya elaboró y aprobó 70 planes de trabajo de distrito basados en la evidencia.
ENSEÑANZAS EXTRAIDAS: Es preciso dedicar mucho esfuerzo e inversiones para reunir a los interesados directos y hacerles colaborar al objeto de introducir cambios de política. Una evaluación más detallada de esos esfuerzos permitiría comprender mejor la manera de reproducir en otros lugares las iniciativas sistemáticas de introducción de cambios normativos.
Introduction
The translation of knowledge into action in the policy arena requires substantial time and investment because of the many stakeholders and competing interests involved. Published work in the field of public policy weighs heavily in favour of taking a "multiple-streams" approach rather than a linear approach to policy reform,1,2 yet few organizations commit the resources needed to promote and sustain policy change in all of its complexity. In this article, we describe a theoretical model that illustrates how scientific evidence can be brought in to the policy-making process. We include a case study that shows how research results were used during policy planning in Kenya.
A model to guide policy reform
A first step in closing the gap between knowledge and practice is to translate scientific evidence for policy decisions that will ultimately lead to policy reforms and changes in health-care practice. People often envision the policy process as linear, moving step-by-step from the identification of problems to the implementation of policy solutions. In reality, the policy process is complex, and policy-makers draw information from a variety of sources, scientific and otherwise.
Policy-makers may be influenced by prominent individuals who have personal agendas, competing ideologies, or long-standing practices that they are reluctant to change. For new information to percolate through the policy environment and become part of policy-makers' thinking, it must sometimes be fed through a murky process of exchange and challenges.2
Recognizing that decision-making in the policy arena is neither linear nor necessarily grounded in science, the Population Reference Bureau (PRB) developed an interactive model to describe how policy change comes about (Fig. 1). The model was adapted from the health policy literature to provide a realistic yet streamlined vision of the complexities of the policy process.3,4
In this model, the policy process is a complex interaction of three spheres, where problems are identified and viable solutions are articulated. The third sphere is the political environment, where events take place independently, not necessarily related to problems or their solutions. When these three spheres come together, there is a window of opportunity for policy change. Three types of activities can help create this window of opportunity: focusing attention on issues to get them on the policy agenda (agenda-setting); creating or strengthening coalitions that sustain attention around an issue (coalition building); and increasing the knowledge that policy-makers have about issues (policy learning).
Researchers have identified four key factors that contribute to agenda-setting: the presence of clear measurable indicators to describe the problem; the actions of political entrepreneurs, or "champions"; the occurrence of events that focus attention on an issue; and the presence of feasible policy options.5,6 Together these factors can determine which issues rise to the attention of policy-makers. Coalitions of advocates including research and policy organizations, the media, the academic community, nongovernmental organizations (NGOs), and individuals with political connections can mobilize and sustain attention around a particular issue. Policy learning, also known as knowledge acquisition, occurs over a period of time. Translating technical or scientific facts into political or social facts is vitally important in generating wider understanding of, and potential support for, policy reforms.7
In the full articulation of the model, capacity building is a circle that surrounds all the other activity spheres and is a fundamental element of the policy change process. Providing national-level partners with tools for policy communication facilitates the effective use of data and research results for policy change and programme improvement.
How does the model work in practice? The model makes clear the dynamic nature of the policy process and the need to work on more than one front to have an effect. For example, the release of new survey results or research findings needs to be accompanied by attention-generating events that help place issues on the policy agenda. By fostering alliances that have enough local clout to push for policy change, issues can be advanced at the national level. Evidence on health problems and issues (brought to light through surveys and other research) and on solutions to health problems (based on current knowledge and documentation of best practices) is at the core of the three spheres of activities.
Applying the model in Kenya
Scarce human and financial resources, as well as long-held organizational practices and bureaucratic inertia, often impede efforts to move new information into the policy process. Experiences in Kenya show that the likelihood of new evidence being used is increased by integrating agenda-setting, coalition building, and policy-learning activities into routine planning and management exercises.
In 2001, the Population Reference Bureau (PRB) supported efforts to disseminate nationwide the results of the Kenya Service Provision Assessment (KSPA), conducted in conjunction with the national Demographic and Health Survey (DHS). The KSPA surveyed a representative sample of 388 health facilities in Kenya, collecting information about the functioning of health services, including service availability, infrastructure, quality, and level of use. A key objective of the dissemination efforts was to ensure that the findings from the survey reached key policy and programme officials in accessible formats for planning, management, and policy purposes.
Several activities contributed to the policy-learning and agenda-setting process at the sub-national level, where the design and implementation of health plans and programmes were to take place. These included, prior to dissemination of the KSPA results, holding 16 district seminars to present and discuss the findings of the national DHS, creating region-specific presentations and 16 district fact sheets, and generating media coverage by calling local media contacts and writing a press alert with key survey findings. Information from the DHS laid the groundwork for understanding the challenges facing health services.
Later, through a series of regional planning seminars, PRB worked with a team from the Kenyan Government to disseminate and use the results of the KSPA. Historically, local officials would simply organize a series of seminars to present the data. Throwing off the yoke of tradition, officials from the Ministry of Planning and the Ministry of Health forged a new understanding to work together to apply the information in a more innovative way. Recognizing the value of using the new data as a planning tool, the team held four regional seminars to link the dissemination of the KSPA results to the Ministry of Health's annual process for developing district-level work plans. The seminars took place in the context of the Kenyan Government's new decentralized programme-reform initiative a planning process that used the KSPA data as a way to identify priority issues.
The seminars fostered coalition-building by including teams from most of Kenya's districts. The teams included district medical health officers, nurses, health educators, NGO representatives, private physicians, hospital representatives, and personnel from the two government ministries (health and plannning). In working groups, participants identified the top five problems arising from the KSPA that could be addressed at the district level in the areas of maternal health, child health, family planning, and sexually transmitted infections. District officers were charged with developing draft action plans that followed the format of the Ministry of Health planning forms, so that proposed solutions could be put into practice.
A primary outcome of the planning seminars was the development of 70 evidence-based health plans. According to officials from the Ministry of Health, it was important for district staff to see the links between the priority needs identified in the surveys and the activities they prepared plans for, because it gave credibility to the ministry's reform planning effort.
Thus, the national-level policy-learning and agenda-setting efforts provided visibility and impetus for policy change; the regional seminars fostered additional learning and coalition building; and concrete actions followed when district planners incorporated the data into health plans. From the national dissemination effort to district-level planning, capacity building of ministry personnel was of key importance in achieving these results. Most of the plans were approved and funded by the Kenyan Ministry of Health in 2002.
Lessons and challenges
For the momentum on any particular policy issue to be sustained, stakeholders must be kept engaged and informed over time. This means distilling and disseminating the latest research findings and programme experiences, and developing the skills of local organizations to communicate this information effectively. For information to be effectively disseminated and used, continued input from and cooperation among a large number of agencies need to be supported.
Related to the need to sustain policy reform activities is the need to measure the outcomes of these actions. Many evaluations assess participants' satisfaction with dissemination seminars and new knowledge gained through training programmes. Beyond this, a more comprehensive investigation of the policy process and policy reforms is needed. Evaluations might include in-depth interviews with key informants such as selected policy-makers, opinion leaders and technical experts involved in a particular policy issue. Ideally, standardized interviews would be conducted before and after a policy intervention to assess whether there had been a change in knowledge and attitudes, and whether a particular policy action resulted from the change in beliefs.
Conclusion
Published work in the area of health policy illuminates a process through which data and information can influence issues: but the process cannot be navigated without considerable effort. Those committed to issues have to work hard to bring evidence and concerns to the attention of policy-makers and work tirelessly on various fronts, bringing people together to learn about the issues and agree on concrete changes. That effort though not without costs can be made more efficient by using systematic approaches, such as those discussed here, to bring about change. 
Funding: This paper was written with funding from the Bill & Melinda Gates Foundation through the Advancing Global Health Policies through Information Project.
Competing interests: none declared.
References
1. Weiss CH. The many meanings of research utilization. Public Admin Rev 1979;39:426-31.
2. Walt G. How far does Research Influence Policy? Eur J Public Health 1994;4:233-5.
3. Kingdon JW. Agendas, alternatives, and public policies. Ann Arbor: University of Michigan; 1984.
4. Porter RW. Knowledge utilization and the process of policy formation: toward a framework for Africa. Washington, DC: Academy for Educational Development, Support for Analysis and Research in Africa Project; 1995.
5. Shiffman J, Generating political will for safe motherhood in Indonesia. Soc Sci Med 2003;56:1197-207.
6. Birkland TA. After disaster: agenda setting, public policy, and focusing events. Washington, DC: Georgetown University Press;1997.
7. Porter RW. Knowledge utilization and the process of policy formation. Washington, DC: USAID, Africa Bureau, SARA Project; 1995. Available from: http://pdf.dec.org/pdf_docs/PNABX317.pdf
(Submitted: 31 January 2006 Final revised version received: 27 April 2006 Accepted: 28 April 2006)
1 Correspondence to Ms Lori Ashford (email: lashford@prb.org).