The political economy of expedience: examining perspectives on military support to Sierra Leone's Ebola response.

The 2013-2016 West Africa Ebola Epidemic is the largest outbreak of Ebola in history. By September, 2014 the outbreak was worsening significantly, and the international president of Médecins Sans Frontières called for military assistance. In Sierra Leone, the British and Sierra Leonean militaries intervened. They quickly established a National Ebola Response Centre and a constituent network of District Ebola Response Centres. Thereafter, these inherently militarised centres are where almost all Ebola response activities were coordinated. In order to examine perspectives on the nature of the militaries' intervention, 110 semi-structured qualitative interviews were conducted and analysed. Military support to Sierra Leone's Ebola response was felt by most respondents to be a valuable contribution to the overall effort to contain the outbreak, especially in light of the perceived weakness of the Ministry of Health and Sanitation to effectively do so. However, a smaller number of respondents emphasised that the military deployments facilitated various structural harms, including for how the perceived exclusion of public institutions (as above) and other local actors from Ebola response decision making was felt to prevent capacity building, and in turn, to limit resilience to future crises. The concurrent provision of life-saving assistance and rendering of structural harm resulting from the militaries' intervention is ultimately found to be part of a vicious cycle, which this article conceptualises as the 'political economy of expedience', a paradox that should be considered inherent in any militarised intervention during humanitarian and public health crises.

Improving health system readiness to address violence against women and girls: a conceptual framework.

BACKGROUND: There is an increasing focus on readiness of health systems to respond to survivors of violence against women (VAW), a global human rights violation damaging women's health. Health system readiness focuses on how prepared healthcare systems and institutions, including providers and potential users, are to adopt changes brought about by the integration of VAW care into services. In VAW research, such assessment is often limited to individual provider readiness or facility-level factors that need to be strengthened, with less attention to health system dimensions. The paper presents a framework for health system readiness assessment to improve quality of care for intimate partner violence (IPV), which was tested in Brazil and Palestinian territories (oPT). METHODS: Data synthesis of primary data from 43 qualitative interviews with healthcare providers and health managers in Brazil and oPT to explore readiness in health systems. RESULTS: The application of the framework showed that it had significant added value in capturing system capabilities - beyond the availability of material and technical capacity - to encompass stakeholder values, confidence, motivation and connection with clients and communities. Our analysis highlighted two missing elements within the initial framework: client and community engagement and gender equality issues. Subsequently, the framework was finalised and organised around three levels of analysis: macro, meso and micro. The micro level highlighted the need to also consider how the system can sustainably involve and interact with clients (women) and communities to ensure and promote readiness for integrating (and participating in) change. Addressing cultural and gender norms around IPV and enhancing support and commitment from health managers was also shown to be necessary for a health system environment that enables the integration of IPV care. CONCLUSION: The proposed framework helps identify a) system capabilities and pre-conditions for system readiness; b) system changes required for delivering quality care for IPV; and c) connections between and across system levels and capabilities.

Integrated delivery of family planning and childhood immunisation services: a mixed methods assessment of service responsiveness.

BACKGROUND: Postpartum women represent a considerable share of the global unmet need for modern contraceptives. Evidence suggests that the integration of family planning (FP) with childhood immunisation services could help reduce this unmet need by providing repeat opportunities for timely contact with FP services. However, little is known about the clients' experiences of FP services that are integrated with childhood immunisations, despite being crucial to contraceptive uptake and repeat service utilisation. METHODS: The responsiveness of FP services that were integrated with childhood immunisations in Malawi was assessed using cross-sectional convergent mixed methods. Exit interviews with clients (n=146) and audits (n=15) were conducted in routine outreach clinics. Responsiveness scores across eight domains were determined according to the proportion of clients who rated each domain positively. Text summary analyses of qualitative data from cognitive interviewing probes were also conducted to explain responsiveness scores. Additionally, Spearman rank correlation and Pearson's chi-squared test were used to identify correlations between domain ratings and to examine associations between domain ratings and client, service and clinic characteristics. RESULTS: Responsiveness scores varied across domains: dignity (97.9%); service continuity (90.9%); communication (88.7%); ease of access (77.2%); counselling (66.4%); confidentiality (62.0%); environment (53.9%) and choice of provider (28.4%). Despite some low performing domains, 98.6% of clients said they would recommend the clinic to a friend or family member interested in FP. The choice of provider, communication, confidentiality and counselling ratings were positively associated with clients' exclusive use of one clinic for FP services. Also, the organisation of services in the clinics and the providers' individual behaviours were found to be critical to service responsiveness. CONCLUSIONS: This study establishes that in routine outreach clinics, FP services can be responsive when integrated with childhood immunisations, particularly in terms of the dignity and service continuity afforded to clients, though less so in terms of the choice of provider, environment, and confidentiality experienced. Additionally, it demonstrates the value of combining cognitive interviewing techniques with Likert questions to assess service responsiveness.

Framing universal health coverage in Kenya: an interpretive analysis of the 2004 Bill on National Social Health Insurance.

In 2004, President Mwai Kibaki of Kenya refused to sign a popular Bill on National Social Health Insurance into law. Drawing on innovations in framing theory, this research provides a social explanation for this decision. In addition to document review, this study involved interpretive analysis of transcripts from 50 semi-structured interviews with leading actors involved in the health financing policy process in Kenya, 2014-15. The frame-critical analysis focused on how actors engaged in (1) sensemaking, (2) naming, which includes selecting and categorizing and (3) storytelling. We demonstrated that actors' abilities to make sense of the Bill were largely influenced by their own understandings of the finer features of the Bill and the array of interest groups privy to the debate. This was reinforced by a process of naming, which selects and categorizes aspects of the Bill, including the public persona of its primary sponsor, its affordability, sustainability, technical dimensions and linkages to notions of economic liberalism. Actors used these understandings and names to tell stories of ideational warfare, which involved narrative accounts of policy resistance and betrayal. This analysis illustrates the difficulty in enacting sweeping reform measures and thus provides a basis for understanding incrementalism in Kenyan health policy.

Targeting anticorruption interventions at the front line: developmental governance in health systems.

In 2008, Vian reported an increasing interest in understanding how corruption affects healthcare outcomes and asked what could be done to combat corruption in the health sector. Eleven years later, corruption is seen as a heterogeneous mix of activity, extensive and expensive in terms of loss of productivity, increasing inequity and costs, but with few examples of programmes that have successfully tackled corruption in low-income or middle-income countries. The commitment, by multilateral organisations and many governments to the Sustainable Development Goals and Universal Health Coverage has renewed an interest to find ways to tackle corruption within health systems. These efforts must, however, begin with a critical assessment of the existing theoretical models and approaches that have underpinned action in the health sector in the past and an assessment of the potential of innovations from anticorruption work developed in sectors other than health. To that end, this paper maps the key debates and theoretical frameworks that have dominated research on corruption in health. It examines their limitations, the blind spots that they create in terms of the questions asked, and the capacity for research to take account of contextual factors that drive practice. It draws on new work from heterodox economics which seeks to target anticorruption interventions at practices that have high impact and which are politically and economically feasible to address. We consider how such approaches can be adopted into health systems and what new questions need to be addressed by researchers to support the development of sustainable solutions to corruption. We present a short case study from Bangladesh to show how such an approach reveals new perspectives on actors and drivers of corruption practice. We conclude by considering the most important areas for research and policy.

New partnerships, new perspectives: The relevance of sexual and reproductive health and rights for sustainable development.

In the light of the opportunities presented by the Sustainable Development Goals (SDGs) debate is being reignited to understand the connections between human population dynamics (including rapid population growth) and sustainable development. Sustainable development is seriously affected by human population dynamics yet programme planners too often fail to consider them in development programming, casting doubt on the sustainability of such programming. Some innovative initiatives are attempting to cross sector boundaries once again, such as the Population Health and Environment (PHE) programmes, which are integrated programmes encompassing family planning service provision with broader public health services and environmental conservation activities. These initiatives take on greater prominence in the context of the SDGs since they explicitly seek to provide cross-sector programming and governance to improve both human and planetary wellbeing. Yet such initiatives remain under-researched and under promoted.

Globalisation and transitions in abortion care in Ghana.

BACKGROUND: Access to safe abortion is a globally contested policy and social justice issue - contested because of its religious and moral dimensions regarding the right to life and personhood of a foetus vs. the rights of women to make decisions about their own bodies. Many nations have agreed to address the health consequences of unsafe abortion, though stopped short of committing to providing comprehensive services. Ghana has a relatively liberal abortion law dating from 1985 and has ratified most international agreements on provision of care. Policy implementation has been very slow, but modest efforts are now being made to reduce maternal mortality caused by unsafe abortions. Understanding whether globalisation has played a role in this transition to practice is important to institutionalise the transition in Ghana and to learn lessons for other countries seeking to implement policies, but analysis is lacking. METHODS: Drawing on 58 in-depth key informant interviews and policy document analysis we describe the development of de jure law and policies on comprehensive abortion care in Ghana, de facto interpretation and implementation of those policies, and assess what role globalization played in the transition in abortion care in Ghana. RESULTS: We found that an accumulation of global influences has converged to start a transition in the culture of abortion care and service provision in Ghana, from a restrictive interpretation of the law to facilitating more widespread access to legal, safe abortion services through development of policies and guidelines and a slow change in attitudes and practices of health providers. These global influences can be categorised as: a global governance architecture of reproductive rights-obligations which creates pressure on signatory governments to act; and global communication of ideas and mobility of health providers (particularly through cross-cultural training opportunities and interaction with international NGOs) which facilitate global cultural interaction on the benefits of safe abortion services for reducing consequences of unsafe abortions. CONCLUSION: Globalisation of information, debate and training experience as well as of international rights frameworks can together create a powerful force for good to protect women and their children from the needless pain and death resulting from unsafe abortions.

"Five hundred years of medicine gone to waste"? Negotiating the implementation of an intercultural health policy in the Ecuadorian Andes.

BACKGROUND: In Ecuador, indigenous women have poorer maternal health outcomes and access to maternity services. This is partly due to cultural barriers. A hospital in Ecuador implemented the Vertical Birth (VB) policy to address such inequities by adapting services to the local culture. This included conducting upright deliveries, introducing Traditional Birth Attendants (TBAs) and making physical adaptations to hospital facilities. METHODS: Using qualitative methods, we studied the VB policy implementation in an Ecuadorian hospital to analyse the factors that affect effective implementation of intercultural health policies at the local level. We collected data through observation, in-depth interviews, a focus group discussion, and documentation review. We conducted 46 interviews with healthcare workers, managers, TBAs, key informants and policy-makers involved in maternal health. Data analysis was guided by grounded theory and drew heavily on concepts of "street-level bureaucracy" to interpret policy implementation. RESULTS: The VB policy was highly controversial; actors' values (including concerns over patient safety) motivated their support or opposition to the Vertical Birth policy. For those who supported the policy, managers, policy-makers, indigenous actors and a minority of healthcare workers supported the policy, it was critical to address ethnic discrimination to improve indigenous women's access to the health service. Most healthcare workers initially resisted the policy because they believed vertical births led to poorer clinical outcomes and because they resented working alongside TBAs. Healthcare workers developed coping strategies and effectively modified the policy. Managers accepted these as a compromise to enable implementation. CONCLUSIONS: Although contentious, intercultural health policies such as the VB policy have the potential to improve maternity services and access for indigenous women. Evidence-base medicine should be used as a lever to facilitate the dialogue between healthcare workers and TBAs and to promote best practice and patient safety. Actors' values influenced policy implementation; policy implementation resulted from an ongoing negotiation between healthcare workers and managers.

Factors shaping political priorities for violence against women-mitigation policies in Sri Lanka.

BACKGROUND: Although violence against women (VAW) is a global public health issue, its importance as a health issue is often unrecognized in legal and health policy documents. This paper uses Sri Lanka as a case study to explore the factors influencing the national policy response to VAW, particularly by the health sector. METHODS: A document based health policy analysis was conducted to examine current policy responses to VAW in Sri Lanka using the Shiffman and Smith (2007) policy analysis framework. RESULTS: The findings suggest that the networks and influences of various actors in Sri Lanka, and their ideas used to frame the issue of VAW, have been particularly important in shaping the nature of the policy response to date. The Ministry of Women and Child Affairs led the national response on VAW, but suffered from limited financial and political support. Results also suggest that there was low engagement by the health sector in the initial policy response to VAW in Sri Lanka, which focused primarily on criminal legislation, following global influences. Furthermore, a lack of empirical data on VAW has impeded its promotion as a health policy issue, despite financial support from international organisations enabling an initial health systems response by the Ministry of Health. Until a legal framework was established (2005), the political context provided limited opportunities for VAW to also be construed as a health issue. It was only then that the Ministry of Health got legitimacy to institutionalise VAW services. CONCLUSION: Nearly a decade later, a change in government has led to a new national plan on VAW, giving a clear role to the health sector in the fight against VAW. High-level political will, criminalisation of violence, coalesced women's groups advocating for legislative change, prevalence data, and financial support from influential institutions are all critical elements helping frame violence as a national public health issue.

Towards an understanding of resilience: responding to health systems shocks.

The recent outbreak of Ebola Virus Disease (EVD) in West Africa has drawn attention to the role and responsiveness of health systems in the face of shock. It brought into sharp focus the idea that health systems need not only to be stronger but also more 'resilient'. In this article, we argue that responding to shocks is an important aspect of resilience, examining the health system behaviour in the face of four types of contemporary shocks: the financial crisis in Europe from 2008 onwards; climate change disasters; the EVD outbreak in West Africa 2013-16; and the recent refugee and migration crisis in Europe. Based on this analysis, we identify '3 plus 2' critical dimensions of particular relevance to health systems' ability to adapt and respond to shocks; actions in all of these will determine the extent to which a response is successful. These are three core dimensions corresponding to three health systems functions: 'health information systems' (having the information and the knowledge to make a decision on what needs to be done); 'funding/financing mechanisms' (investing or mobilising resources to fund a response); and 'health workforce' (who should plan and implement it and how). These intersect with two cross-cutting aspects: 'governance', as a fundamental function affecting all other system dimensions; and predominant 'values' shaping the response, and how it is experienced at individual and community levels. Moreover, across the crises examined here, integration within the health system contributed to resilience, as does connecting with local communities, evidenced by successful community responses to Ebola and social movements responding to the financial crisis. In all crises, inequalities grew, yet our evidence also highlights that the impact of shocks is amenable to government action. All these factors are shaped by context. We argue that the '3 plus 2' dimensions can inform pragmatic policies seeking to increase health systems resilience.

Numbers, systems, people: how interactions influence integration. Insights from case studies of HIV and reproductive health services delivery in Kenya.

Drawing on rich data from the Integra evaluation of integrated HIV and reproductive-health services, we explored the interaction of systems hardware and software factors to explain why some facilities were able to implement and sustain integrated service delivery while others were not. This article draws on detailed mixed-methods data for four case-study facilities offering reproductive-health and HIV services between 2009 and 2013 in Kenya: (i) time-series client flow, tracking service uptake for 8841 clients; (ii) structured questionnaires with 24 providers; (iii) in-depth interviews with 17 providers; (iv) workload and facility data using a periodic activity review and cost-instruments; and (v) contextual data on external activities related to integration in study sites. Overall, our findings suggested that although structural factors like stock-outs, distribution of staffing and workload, rotation of staff can affect how integrated care is provided, all these factors can be influenced by staff themselves: both frontline and management. Facilities where staff displayed agency of decision making, worked as a team to share workload and had management that supported this, showed better integration delivery and staff were able to overcome some structural deficiencies to enable integrated care. Poor-performing facilities had good structural integration, but staff were unable to utilize this because they were poorly organized, unsupported or teams were dysfunctional. Conscientious objection and moralistic attitudes were also barriers.Integra has demonstrated that structural integration is not sufficient for integrated service delivery. Rather, our case studies show that in some cases excellent leadership and peer-teamwork enabled facilities to perform well despite resource shortages. The ability to provide support for staff to work flexibly to deliver integrated services and build resilient health systems to meet changing needs is particularly relevant as health systems face challenges of changing burdens of disease, climate change, epidemic outbreaks and more.

The Current Status of Research on the Integration of Sexual and Reproductive Health and HIV Services.

Integration of services for sexual and reproductive health (SRH) and HIV has been widely promoted globally in the belief that both clients and health providers benefit through improvements in quality, efficient use of resources, and lower costs, helping to maximize limited health resources and provide comprehensive client-centered care. This article builds on the growing body of research on integrated sexual SRH and HIV services. It brings together critical reviews on issues within the wider SRH and rights agenda and synthesizes recent research on integrated services, drawing on the Integra Initiative and other major research. Unintended pregnancy and HIV are intrinsically interrelated SRH issues, however broadening the constellation of services, scaling up, and mainstreaming integration continue to be challenging. Overcoming stigma, reducing gender-based violence, and meeting key populations' SRH needs are critical. Health systems research using SRH as the entry point for integrated services and interaction with communities and clients is needed to realize universal health coverage.

Barriers and Facilitators to Integrating Health Service Responses to Intimate Partner Violence in Low- and Middle-Income Countries: A Comparative Health Systems and Service Analysis.

This systematic review synthesizes 11 studies of health-sector responses to intimate partner violence (IPV) in low- and middle-income countries. The services that were most comprehensive and integrated in their responsiveness to IPV were primarily in primary health and antenatal care settings. Findings suggest that the following facilitators are important: availability of clear guidelines, policies, or protocols; management support; intersectoral coordination with clear, accessible on-site and off-site referral options; adequate and trained staff with accepting and empathetic attitudes toward survivors of IPV; initial and ongoing training for health workers; and a supportive and supervised environment in which to enact new IPV protocols. A key characteristic of the most integrated responses was the connection or "linkages" between different individual factors. Irrespective of their service entry point, what emerged as crucial was a connected systems-level response, with all elements implemented in a coordinated manner.

Public accountability needs to be enforced -a case study of the governance arrangements and accountability practices in a rural health district in Ghana.

BACKGROUND: Improving public accountability is currently high on the global agenda. At the same time, the organisation of health services in low- and middle-income countries is taking place in fragmented institutional landscapes. State and non-state actors are involved in increasingly complex governance arrangements. This often leads to coordination problems, confusion of roles and responsibilities and possibly accountability gaps. This study aimed at assessing the governance arrangements and the accountability practices of key health actors at the level of a Ghanaian health district with the aim to understand how far public accountability is achieved. METHODS: We adopted the case study design as it allows for in-depth analysis of the governance arrangements and accountability relations between actors, their formal policies and actual accountability practices towards the public and towards stakeholders. Data were collected at a rural health district using in-depth interviews, observation and document review. In the analysis, we used a four-step sequence: identification of the key actors and their relationships, description of the multi-level governance arrangements, identification of the actual accountability relations and practices between all actors and finally appraisal of the public accountability practices, which we define as those practices that ensure direct accountability towards the public. RESULTS: In this rural health district with few (international) non-governmental organisations and private sector providers, accountability linkages towards management and partners in health programmes were found to be strong. Direct accountability towards the public, however, was woefully underdeveloped. This study shows that in settings where there is a small number of actors involved in organising health care, and where the state actors are underfunded, the intense interaction can lead to a web of relations that favours collaboration between partners in health service delivery, but fails public accountability. CONCLUSIONS: It is clear that new formal channels need to be created by all actors involved in health service delivery to address the demand of the public for accountability. If the public does not find an adequate response to its genuine concerns, distrust between communities and service users on one hand, and providers, international non-governmental organisations and District Health Management Teams on the other is likely to increase to the detriment of all parties' interests.

The emergence of the vertical birth in Ecuador: an analysis of agenda setting and policy windows for intercultural health.

Maternal mortality continues to claim the lives of thousands of women in Latin America despite the availability of effective treatments to avert maternal death. In the past, efforts to acknowledge cultural diversity in birth practices had not been clearly integrated into policy. However, in Otavalo (Ecuador) a local hospital pioneered the implementation of the 'Vertical Birth'-a practical manifestation of an intercultural health policy aimed at increasing indigenous women's access to maternity care. Drawing on agenda-setting theory, this qualitative research explores how the vertical birth practice made it onto the local policy agenda and the processes that allowed actors to seize a window of opportunity allowing the vertical birth practice to emerge. Our results show that the processes that brought about the vertical birth practice took place over a prolonged period of time and resulted from the interplay between various factors. Firstly, a maternal health policy community involving indigenous actors played a key role in identifying maternal mortality as a policy problem, defining its causes and framing it as an indigenous rights issue. Secondly, previous initiatives to address maternal mortality provided a wealth of experience that gave these actors the knowledge and experience to formulate a feasible policy solution and consolidate support from powerful actors. Thirdly, the election of a new government that had incorporated the demands of the indigenous movement opened up a window of opportunity to push intercultural health policies such as the vertical birth. We conclude that the socioeconomic and political changes at both national and local level allowed the meaningful participation of indigenous actors that made a critical contribution to the emergence of the vertical birth practice. These findings can help us advance our knowledge of strategies to set the agenda for intercultural maternal health policy and inform future policy in similar settings. Our results also show that Kingdon's model was useful in explaining how the VB practice emerged but also that it needs modifications when applied to low and middle income countries.

Agenda setting and framing of gender-based violence in Nepal: how it became a health issue.

Gender-based violence (GBV) has been addressed as a policy issue in Nepal since the mid 1990s, yet it was only in 2010 that Nepal developed a legal and policy framework to combat GBV. This article draws on the concepts of agenda setting and framing to analyse the historical processes by which GBV became legitimized as a health policy issue in Nepal and explored factors that facilitated and constrained the opening and closing of windows of opportunity. The results presented are based on a document analysis of the policy and regulatory framework around GBV in Nepal. A content analysis was undertaken. Agenda setting for GBV policies in Nepal evolved over many years and was characterized by the interplay of political context factors, actors and multiple frames. The way the issue was depicted at different times and by different actors played a key role in the delay in bringing health onto the policy agenda. Women's groups and less powerful Ministries developed gender equity and development frames, but it was only when the more powerful human rights frame was promoted by the country's new Constitution and the Office of the Prime Minister that legislation on GBV was achieved and a domestic violence bill was adopted, followed by a National Plan of Action. This eventually enabled the health frame to converge around the development of implementation policies that incorporated health service responses. Our explicit incorporation of framing within the Kindgon model has illustrated how important it is for understanding the emergence of policy issues, and the subsequent debates about their resolution. The framing of a policy problem by certain policy actors, affects the development of each of the three policy streams, and may facilitate or constrain their convergence. The concept of framing therefore lends an additional depth of understanding to the Kindgon agenda setting model.

The Costs of Delivering Integrated HIV and Sexual Reproductive Health Services in Limited Resource Settings.

OBJECTIVE: To present evidence on the total costs and unit costs of delivering six integrated sexual reproductive health and HIV services in a high and medium HIV prevalence setting, in order to support policy makers and planners scaling up these essential services. DESIGN: A retrospective facility based costing study conducted in 40 non-government organization and public health facilities in Kenya and Swaziland. METHODS: Economic and financial costs were collected retrospectively for the year 2010/11, from each study site with an aim to estimate the cost per visit of six integrated HIV and SRH services. A full cost analysis using a combination of bottom-up and step-down costing methods was conducted from the health provider's perspective. The main unit of analysis is the economic unit cost per visit for each service. Costs are converted to 2013 International dollars. RESULTS: The mean cost per visit for the HIV/SRH services ranged from $Int 14.23 (PNC visit) to $Int 74.21 (HIV treatment visit). We found considerable variation in the unit costs per visit across settings with family planning services exhibiting the least variation ($Int 6.71-52.24) and STI treatment and HIV treatment visits exhibiting the highest variation in unit cost ranging from ($Int 5.44-281.85) and ($Int 0.83-314.95), respectively. Unit costs of visits were driven by fixed costs while variability in visit costs across facilities was explained mainly by technology used and service maturity. CONCLUSION: For all services, variability in unit costs and cost components suggest that potential exists to reduce costs through better use of both human and capital resources, despite the high proportion of expenditure on drugs and medical supplies. Further work is required to explore the key drivers of efficiency and interventions that may facilitate efficiency improvements.

Fertility decisions and contraceptive use at different stages of relationships: windows of risk among men and women in accra.

CONTEXT: Despite declining fertility in Ghana, modern contraceptive use-even in urban areas-is low for reasons that remain unclear. Few studies have explored what drives fertility decisions and contraceptive use among contemporary urban residents within a relationship context. METHODS: In-depth contraceptive life history interviews were conducted among a purposive sample of 80 sexually active women and men living in Accra. RESULTS: Contraception is viewed favorably, although the timing and choice of method strongly depends on the type and stage of relationship. At sexual debut and at first sex with a new partner, sex is usually unprotected. Many women show agency in subsequently negotiating condom use; men also show motivation to practice contraception. As relationships stabilize, couples abandon condoms and adopt traditional methods, out of fear that modern methods could affect fertility. After a first birth, couples prefer modern contraceptives to space children, but side effects often lead women to switch methods or discontinue use; women in supportive relationships are more likely than those in unsupportive relationships to continue use of modern contraceptives despite side effects. After reproductive goals have been realized, couples revert to using traditional methods to avoid further exposure to "chemicals." CONCLUSIONS: Contraceptive programs may be more successful if they target messages according to stage of relationship, involve men and work with people's desires to use traditional methods at certain times to ensure that they can do so safely.