Prioritizing Surgical Care on National Health Agendas: A Qualitative Case Study of Papua New Guinea, Uganda, and Sierra Leone

Background: Little is known about the social and political factors that influence priority setting for different health services in low- and middle-income countries (LMICs), yet these factors are integral to understanding how national health agendas are established. We investigated factors that facilitate or prevent surgical care from being prioritized in LMICs. Methods and Findings: We undertook country case studies in Papua New Guinea, Uganda, and Sierra Leone, using a qualitative process-tracing method. We conducted 74 semi-structured interviews with stakeholders involved in health agenda setting and surgical care in these countries. Interviews were triangulated with published academic literature, country reports, national health plans, and policies. Data were analyzed using a conceptual framework based on four components (actor power, ideas, political contexts, issue characteristics) to assess national factors influencing priority for surgery. Political priority for surgical care in the three countries varies. Priority was highest in Papua New Guinea, where surgical care is firmly embedded within national health plans and receives significant domestic and international resources, and much lower in Uganda and Sierra Leone. Factors influencing whether surgical care was prioritized were the degree of sustained and effective domestic advocacy by the local surgical community, the national political and economic environment in which health policy setting occurs, and the influence of international actors, particularly donors, on national agenda setting. The results from Papua New Guinea show that a strong surgical community can generate priority from the ground up, even where other factors are unfavorable. Conclusions: National health agenda setting is a complex social and political process. To embed surgical care within national health policy, sustained advocacy efforts, effective framing of the problem and solutions, and country-specific data are required. Political, technical, and financial support from regional and international partners is also important. Institutional review board approval was obtained from the Committee on Human Research at the University of California, San Francisco, the PNG National Department of Health Ethics Committee, the Mulago Hospital Research and Ethics Committee, the Uganda National Council for Science and Technology, the King’s College London Research Ethics Committee, and the Sierra Leone Ethics and Scientific Review Committee. We obtained written informed consent from all those who agreed to be interviewed. A unique identifier (e.g., Uganda informant 1, UG01) was assigned for each interviewee to ensure anonymity and protect confidentiality. Three case countries were selected from the 139 World Bank–defined LMICs (July 2013 country listings) [15]. In selecting case countries, we sought to include countries at different stages on the pathway to building comprehensive national surgical systems. Potential case countries were evaluated based on objective evidence of key surgical indicators [4], including met and unmet need for surgical care, surgical case volume per population, surgical providers per population, and the presence of a national health plan that specifically included surgical care. Such information was not available for many LMICs, and the case countries selected reflect those for which there were at least some objective country-level data on surgical care. We then further narrowed down case country selection on the basis of existing professional relationships with investigator institutions and local collaborators, recognizing that access to key actors within each country required drawing on local health, surgical, and political networks. The three case countries selected—PNG, Uganda, and Sierra Leone—represent diverse geographic, political, and cultural settings. Despite the diversity, they share one characteristic: all three countries have poor general health indicators and have experienced documented challenges in providing surgical care to meet population needs (Table 1). These challenges include a high burden of surgical conditions [16,17], low ratios of surgical providers per 100,000 people, low operative volumes [18,19], and inadequate surgical capacity [20–22]. Documented changes over time in these key surgical indicators showed improvement in one case country (PNG), largely unchanged performance in the second case country (Uganda), and worsened performance in the third case country (Sierra Leone). Key health indicators are drawn from the World Bank Group 2014 World Development Indicators [23]. aIncludes general surgical and obstetric procedures because disaggregated data are not available for all countries. bThe Lancet Commission on Global Surgery recommends a minimum global target of least 20 surgical providers per 100,000 population and 5,000 surgical procedures per 100,000 population. cPublic sector only; no data available on private sector operative volumes. PNG figures are for general surgical procedures only and do not include obstetric operations, unlike for Sierra Leone, Uganda, and the UK. dThese figures are for England only. The total number of surgeons does not include obstetricians as they are represented by a separate college and workforce count in the UK. eIncludes surgeons and obstetricians. fIncludes surgeons, medical officers, and clinical officers providing surgical care. GDP, gross domestic product. Our study explored the differing responses of each country to its unmet surgical need over the past 25 years—spanning the decade prior to the institution of the MDGs and the years after the MDGs were in place. It sought to place these responses in the context of both contemporaneous and historical social, political, and economic events in each country. The differences between case countries allowed us to conduct a comparative analysis of the factors that have facilitated or obstructed surgical care from receiving national political attention, priority, and action. To analyze factors influencing priority for surgical care in each of the three case countries, we drew on a previously published conceptual framework developed by Shiffman and Smith (Table 2) [24]. This framework has been used to examine political priority at the global level for maternal mortality, health systems strengthening, mental health, and neonatal survival as well as national political priority for safe motherhood in LMICs [2,24–27]. A health problem is defined as having political priority when (1) political leaders publicly and privately express sustained concern for the problem; (2) political leaders and governments, through an authoritative decision-making process, enact policies that offer widely embraced strategies to address the problem; and (3) financial resources commensurate with the problem’s gravity are mobilized and allocated to address the problem [2,24]. This framework has been modified from the original framework of Shiffman and Smith [24] to include factors shaping national political priority that are more specific to surgical care. The framework identifies determinants of political priority for health initiatives and organizes these into four overarching categories through which to understand the generation of political support and collective action for an issue. These are (1) actor power, (2) ideas, (3) political contexts, and (4) issue characteristics. The cumulative impact of the presence of different factors across these four categories is broadly understood to improve the likelihood that an issue will receive priority [24]. To examine how surgical care is prioritized at a national level, we used process tracing, a qualitative case study research methodology that is commonly used in political science [28]. Process tracing triangulates multiple sources of information to minimize bias and to identify and test causal mechanisms of a theory. It analyzes change and causation, and focuses on describing and analyzing sequences of independent, dependent, and intervening events or variables [29]. This methodology is unique in its ability to reveal social and political processes within a real-life context, while also accounting for historical influences. In each of the three case countries, we conducted interviews with key health and political actors involved in events and processes related to health agenda setting. Using the conceptual framework outlined in Table 2, we developed a semi-structured interview guide with open-ended questions (the guide is in S1 Text). Questions were designed to query the attitudes, values, beliefs, and knowledge of the key informants (KIs) about how and why different health issues, including surgical care, were prioritized within their country. In total, 29 interviews were conducted in PNG, 32 in Uganda, and 12 in Sierra Leone between March 1 and July 31, 2014, lasting on average 1 h in length. All interviews were conducted in English by A. J. D. (Sierra Leone), K. C. L. (Uganda), and J. B. (PNG), assisted by the local investigators A. E. E., T. B. K., O. L., S. L., A. D., and G. K. The interviews were audio recorded and transcribed. KIs included health care providers (surgical and non-surgical providers from the public, private for-profit, and non-governmental organization [NGO] sectors), health educators and policy-makers, politicians, civil society members, and health funders (S1 Table lists the professions of the KIs). KIs were identified from professional networks, by review of the published and grey literature on surgical care and health agenda setting in each country, and by asking other KIs. We also carried out archival research on the history of health and surgical care within each country, using both peer-reviewed and grey literature. Sources of grey literature included documents published by the government of each country (including the ministry of health [MoH] and the ministry of finance) or NGOs, written accounts of the history and current state of surgery by independent authors, conference presentations, and accounts published by other news outlets. All interviews were transcribed by one of the authors or a commercial transcription service. Transcripts were analyzed with the assistance of Dedoose, a qualitative data analysis and research software program [30]. Each interview was coded independently by one of three authors (A. J. D., K. C. L., or J. B.). A random sample of interviews from each country underwent double-coding, to ensure reliability. Any differences were resolved through discussion between the two coders and, where differences could not be resolved, through adjudication by a third coder. The data analysis process was influenced by a grounded-theory approach [31] to analyze the KI interviews and describe the determinants of national political priority for surgical care in each country. This approach involved reading through the transcript, discussing general themes and concepts, and grouping the concepts into the previously described framework. During the data analysis process, information on timelines and events revealed through the interviews was triangulated with primary sources, academic and grey literature, and historical accounts. In addition, the local investigators in each country, as well as one independent actor from each country who had a long history of involvement in the health or surgical sector, specifically reviewed the manuscript for historical accuracy prior to submission. Two of these independent actors were also KIs in their respective countries. This study was designed, analyzed, and reported in accordance with the Consolidated Criteria for Reporting Qualitative Studies (COREQ) (interviews and focus groups) [32]. Quotations from individual KIs are used in this manuscript to illustrate and support the broader findings reported in the text.