Elsevier

Public Health

Volume 124, Issue 3, March 2010, Pages 159-166
Public Health

Original Research
Stakeholder analysis for health research: Case studies from low- and middle-income countries

https://doi.org/10.1016/j.puhe.2009.12.006Get rights and content

Summary

Objectives

Future Health Systems: Innovations for Equity (FHS) is working in six partner countries in Asia and Africa, focusing on strengthening the research–policy interface in relation to specific health system research projects. These projects present an opportunity to study the influence of stakeholders on research and policy processes.

Study design

Qualitative stakeholder analysis.

Methods

Stakeholder analysis was conducted in each FHS country using a structured approach. A cross-country evaluation was performed concentrating on six key areas: chosen research topic; type of intervention considered; inclusion/exclusion of stakeholder groups; general stakeholder considerations; power level, power type and agreement level of stakeholders; and classification of and approaches to identified stakeholders.

Results

All six countries identified a range of stakeholders but each country had a different focus. Four of the six countries identified stakeholders in addition to the guidelines, while some of the stakeholder categories were not identified by countries. The mean power level of identified stakeholders was between 3.4 and 4.5 (1 = very low; 5 = very high). The percentage of classified stakeholders that were either drivers or supporters ranged from 60% to 91%.

Conclusion

Three important common areas emerge when examining the execution of the FHS country stakeholder analyses: clarity on the purpose of the analyses; value of internal vs external analysts; and the role of primary vs secondary analyses. This paper adds to the global body of knowledge on the utilization of stakeholder analysis to strengthen the research–policy interface in the developing world.

Introduction

Health research and the translation of findings to action is critical to improving population health in the developing world.1, 2 Such research translation is essential to meeting the health-related millennium development goals.3 Positivity towards incorporating new research findings into intervention improvements is recognized as a common ingredient in successful health interventions in various developing world settings.4 The importance of a systematic approach to priority setting for such research has also been articulated.5 Research quality and its impact on policy and implementation can be enhanced when multiple stakeholder perspectives, particularly from consumers derived from real-life situations, are taken into consideration.6

What do we mean by stakeholders and why is it important to consider them in attempts to improve population health in the developing world? Stakeholders can be defined as ‘organizations and individuals that are involved in a specific activity because they participate in producing, consuming, managing, regulating, or evaluating the activity’.7 Taking into account stakeholder perspectives, varying from an individual residing within a community to national governments to global organizations, allows health interventions to be seen from multiple angles. This has several advantages. First, understanding the perspective of key decision makers provides information on the likelihood of policy changes required for intervention implementation. Second, consumer ideas, concerns and expectations related to the intervention can predict the likelihood of successful intervention implementation. Third, understanding multiple stakeholder perspectives allows intervention refinement incorporating innovative ideas. Fourth, strategies to influence key stakeholders can be formulated. Lastly, sharing perspectives between key stakeholders may enhance solidarity around a particular intervention.

A number of conceptual frameworks linking the research–policy interface have been developed. One such conceptual framework describes key elements as: processes of research generation and decision-making; the stakeholders; the products; the mediators; and the context.8 Stakeholder engagement throughout research generation and policy-making is emphasized as critical to strengthening the research–policy interface. Mediators, individuals or institutions who foster linkages between different stakeholders are described as perhaps the most crucial component of the framework encouraging strong research–policy linkages.8 Importance of key influential mediators in initiating change is also supported by literature on ‘tipping points’, emphasizing the centrality of ‘connectors’ in precipitating change.9 Another research–action framework focuses on country-level assessment of linkages.10 The framework has four elements: general climate; research production; a mix of push and pull factors; and evaluation approaches. The critical role of a wide range of stakeholders in linking research to action is acknowledged. The Future Health Systems framework explores health systems research and its influence on policy processes in low-income countries, and articulates four ‘streams of influence’ on the research–policy interface: contexts; stakeholders; accountabilities; and processes.1 Importance of interests, values and power of stakeholders at the research–policy interface is emphasized.

The utility of active participation of stakeholders in ensuring health research translation into sound public health policy is recognized in theory.11, 12 There is a gradually emerging pool of global experiences (Africa, Asia and South America) reported in the literature. The utility of stakeholder-focused approaches to analytical techniques in healthcare decision-making, such as cost-effectiveness analysis13 and health technology assessments,14 have been articulated. Practical studies from Brazil, Burkina Faso, Indonesia, Lithuania, South Africa, Uruguay and Pakistan demonstrate the importance of stakeholder engagement for strengthening linkages between research and policy with respect to varied research agendas.8 An analysis of health policy and systems research agendas in developing countries found varying stakeholder agendas, with decision-making power residing with donors.15 An ex-ante assessment of transfer of global scientific knowledge on respiratory health to rural Nepal recognized the importance of social system stakeholders in transfering knowledge to practice.16 An exploration of use of evidence in decision-making for mother to child transmission of human immunodeficiency virus in Uganda demonstrated openness of stakeholders towards incorporating evidence into decision-making.17 An innovative assessment of the policy environment for evidence-based primary healthcare changes in Chile articulated key stakeholder thoughts and perceptions.18 A stakeholder analysis on a proposed USAID maternal and child health project in India demonstrated the ability to forecast project success.19 An assessment of the policy environment surrounding the Framework Convention on Tobacco Control in Thailand and Zimbabwe provided insights into the role of global stakeholders in setting contexts for national activities on this key public health issue.20

Building on this existing knowledge, the overarching goal of this paper is to add to the global body of knowledge on utilization of stakeholder analysis to strengthen the research–policy interface in the developing world. Four key objectives are: to review methodological issues in conducting stakeholder analyses in low- and middle-income countries; to present preliminary findings from six baseline country stakeholder analyses in Asia and Africa on specific research areas; to report on an evaluation template for cross-country comparison of stakeholder analyses; and to explore utility of the approach used for future use in low-income countries.

This paper is based on the work of Future Health Systems: Innovations for Equity (FHS), which aims to generate knowledge that shapes health systems to benefit the world's poor.21 FHS brings policy makers from six countries together with leading public health and development research institutions to test strategies in three areas: financing of health care to reduce people's risk of poverty; improving access to health services; and strengthening the health systems research–policy interface to promote the interests of the poor. Consortium partners are based in Afghanistan, Bangladesh, China, India, Nigeria and Uganda, and the lead agencies are Johns Hopkins University Bloomberg School of Public Health, USA and the Institute of Development Studies, UK. While each country has separate studies which are specific to local context and need, most studies do involve influencing policy at various levels, either as a direct or indirect outcome of the proposed work. Details of each study can be found on the consortium website.21

Evolution in the use of stakeholder analyses in health and development has been reviewed elsewhere.22 Multiple methods are utilized and can be adapted to conduct stakeholder analyses23; this is in keeping with an emphasis on ‘methodological pluralism’ for public health interventions, particularly engagement with stakeholders.24 The stakeholder analyses process can be divided into three phases (planning, conducting and analysing) and key considerations in each of these phases are articulated by Varvasovszky and Brugha.25 Reflection on all three phases is important early in the stakeholder analyses process, and multiple amendments are required due to the iterative nature of analyses. As a starting point, clarity on purpose of analysis is essential.25, 26 This purpose may be different early in the planning phase (rapid situational analysis) than later on in project/intervention implementation. If the purpose of the stakeholder analysis is to provide a comprehensive analysis in order to provide new information on policy-making processes, a retrospective study may be beneficial.

The time dimension of analyses has implications on the methodology employed. Approaches and methods are different depending on whether the focus is on understanding current policy formation on a particular issue, as opposed to likelihood of a particular policy or project being successful in the future. Future-oriented stakeholder analyses require application of prospective methods, often broad in scope.19, 25 Further, time commitment needs to be ongoing throughout the project and must be aligned with availability of resources.

Stakeholder analysis setting, in terms of context as well as the level of analysis, is a key consideration.25 The sociocultural context of where the analyses take place requires special attention for effective planning and implementation. In particular, communication channels among different stakeholders, as well as between stakeholders and those carrying out analyses, will be unique to individual sociocultural environments. Adapting approaches based on understanding this environment will enhance the validity of findings. The level of analysis also needs clear articulation, and analyses have successfully been carried out focusing on various levels: local27; district16; national28; and international.20

Analyses can be carried out by individuals or teams, either from within or external to the project or issue on which stakeholder analyses are focusing.25 Advantages to an internal analysis are familiarity with importance of key stakeholders and cognizance of knowledge flow pathways in local contexts; however, a disadvantage is that internal analysts may have preformed opinions and relationships with key stakeholders, which may bias findings. Objectivity of external analysis may lead to more robust findings.23

Available data sources for analyses can be categorized into primary and secondary.25 Primary sources are stakeholders themselves; secondary sources include documents, reports, statements and opinions of others regarding stakeholder viewpoints. In early stages of analyses, secondary sources may provide a rapid means of gaining knowledge on stakeholder viewpoints. Numerous methods can be utilized to gather primary data on stakeholders (semi-structured interviews of key informants, telephone interviews, focus group discussions, consensus methods and nominal group processes).29 Information collection, whether from primary or secondary sources, should be iterative, allowing continuous development of the knowledge base of stakeholders.

Stakeholder identification warrants careful judgement, needing to be neither under-inclusive (limiting breadth of perspectives) nor over-inclusive (attenuating necessary focus).26 Three key criteria have been articulated for inclusion of key stakeholders: the potential to weaken, strengthen or influence support for the intervention or policy.23 An initial list can be constructed by brainstorming relevant issues; further additions to the list can utilize a snowball technique where stakeholders identify further stakeholders.18, 28 An external analyst may also suggest stakeholders not initially identified by internal analysts. Factors increasing the likelihood of gaining access to key stakeholders, such as recommendations or introductions, may also alter stakeholder viewpoints.25 Access to community stakeholders may be particularly difficult, warranting a focus on continuous community engagement and long-term partnerships.27, 30 Communication between stakeholders is a dynamic process that can shape future stakeholder positions26; this is often ignored in stakeholder analyses. Mapping information flow and influences between stakeholders can provide valuable direction to strategic approaches to key stakeholders.25 Accountabilities governing relationships between different stakeholders with disparate interests, values and power warrant careful examination.1 Further, coalitions between key stakeholders, particularly in the community, may be facilitated by understanding these inter-relationships.31

Systematic organization and presentation of findings from stakeholder analyses is essential. Matrix tables can be constructed to summarize stakeholder perspectives using key headings important to the particular context.26 Possible headings include: involvement in the issue; interest in the issue; level of influence; position adopted; and impact of issue on actor.28 Matrix cells under each column may utilize free-text descriptions, ordinal scales or categorical variables. Forcefield matrices displaying two dimensions of stakeholder perspectives (e.g. level of influence and level of agreement) can be very useful in categorizing stakeholders into types. Other methods to present findings include maps illustrating networks and positions.25 Findings can be used to strategize management of stakeholders,23 matching stakeholder types with strategic approaches.25 The approach to stakeholders that are supportive, mixed, non-supportive and marginal is to involve, collaborate, defend and monitor, respectively.25 Suboptimal and inappropriate approaches may lead to: unnecessary attention to some stakeholders, leading to wasted resources; missed opportunities for gaining support; and placing the proposed intervention or policy at risk.25

Many limitations exist in the use of stakeholder analyses. First, analyses can be over-utilized, making it essential to remain cognizant that such analysis is not an end in itself, and is a representation of reality (not reality itself).23 Second, due to its cross-sectional nature, analyses are time dependent: in rapidly changing contexts, this can be a significant drawback, particularly if strategic approaches to stakeholders are shaped from outdated findings.25 Third, analyst-focused limitations include: ascertainment bias representing a systemic failure to represent equally all the stakeholders that warrant representation; biases in interpretation of analysis findings; and the Hawthorn effect (i.e. changed stakeholder responses as a consequence of analyst activities).32 Fourth, stakeholder-focused limitations include: response bias, where some stakeholder types are more likely to participate; conflict between individual stakeholder views and those of the organization they represent; levels of true disclosure; and unknown inter-stakeholder influences in shifting positions.25 Finally, analysts themselves, particularly internal analysts, can be stakeholders in the process; this can often be overlooked.

Section snippets

Methods

Given the above methodological considerations, a multi-step guideline for stakeholder analysis by FHS country teams was constructed. The purpose of such analyses was to generate findings from baseline stakeholder analyses to inform initial strategies. This involved both primary and secondary sources. The guidelines can be summarized into 12 points (Table 1). Three steps require further explanation. Step 3, identification of stakeholders, requires systematic consideration of 11 categories of

Results

A brief description of the highlights of each country-level stakeholder analysis is presented here. It is anticipated that these findings will be reported in detail in due course.

The FHS Bangladesh team conducted a stakeholder analysis on rural informal healthcare providers with a focus on designing skill enhancement interventions and establishment of links between formal and informal healthcare providers. A wide array of stakeholder groups were considered with a particular focus on

Discussion

Three important common areas emerge when examining the execution of the FHS country stakeholder analyses: clarity on the purpose of the analyses; value of internal vs external analysts; and the role of primary vs secondary analyses. The general purpose of stakeholder analyses in each FHS partner country was to assist with adoption or implementation of proposed research on future health systems. The specific purpose of the baseline stakeholder analyses was to identify groups that can affect

Ethical approval

Ethical approval in each country was sought and granted. FHS team members in each country carried out their own data collection.

Funding

Department for International Development, UK.

Competing interests

None declared.

References (32)

  • L. Kapiriri et al.

    The relationship between prevention of mother to child transmission of HIV and stakeholder decision making in Uganda: implications for health policy

    Health Policy

    (2003)
  • A.A. Hyder et al.

    Exploring health systems research and its influence on policy processes in low income countries

    BMC Public Health

    (2007)
  • S.B. Syed et al.

    Exploring evidence–policy linkages in health research plans: a case study from six countries

    Health Res Policy Syst

    (2008)
  • M. Claeson et al.

    The case for mobilizing new research behind the health millennium development goals

  • C. Medlin et al.

    Improving the health of populations: lessons of experience

  • B. Bloom et al.

    Priorities for global research and development of interventions

  • A. Mills et al.

    Strengthening health systems

  • L. Novick et al.

    Public health administration: principles for population-based management

    (2001)
  • COHRED Working Group on Research to Action and Policy

    Lessons in research to action and policy: case studies from seven countries

    (2000)
  • Gladwell M. The law of the few: connectors, mavens, and salesmen. The tipping point: how little things can make a big...
  • J.N. Lavis et al.

    Assessing country-level efforts to link research to action

    Bull World Health Organ

    (2006)
  • P. Kunasol et al.

    Public health sciences and policy in developing countries

  • Y. Nuyens et al.

    Knowledge translation: linking the past to the future

    Bull World Health Organ

    (2006)
  • M.L. Berger et al.

    Cost-effectiveness analysis: from science to application

    Med Care

    (2005)
  • A. Kazanjian

    Reflections on the social epidemiologic dimension of health technology assessment

    Int J Technol Assess Health Care

    (2004)
  • M.A. Gonzalez-Block

    Health policy and systems research agendas in developing countries

    Health Res Policy Syst

    (2004)
  • View full text