Although abortion is legal in Zambia, very few women know they have the right to have an abortion or know where to seek this care. Unsafe abortions have caused unnecessary deaths in Zambia even among women who have the right to a safe and legal abortion. In this national study we used the opportunity to introduce abortion with medication and improve understanding of the law and abortion care in and around 28 health facilities in the country.

In this study routine statistics from more than 13,000 women were collected from health facilities and combined with survey data from women seeking abortions, trained health care workers and community members to illustrate the results of the national efforts to introduce and scale up abortion services.

This study is an example of tackling a difficult issue, involving a wide range of different groups and partners and proposing a national model. This study also provides an example of work and lessons for other countries around Africa.

Although abortion is technically legal in Zambia, the reality is far more complicated. Zambia has what has been called a “paper law” with numerous barriers to care at both the policy and implementation levels [1]. Unsafe abortion remains a very real problem - causing death and disability across the country. Although evidence on the incidence and consequences of unsafe abortion in recent years is scarce, numerous studies from the late 1990s identify the methods used across the country, such as ingesting toxins like detergent and inserting cassava sticks in the cervix [2–4]. While information and utilization of legal abortions is becoming more common, most experts feel that unsafe abortion is persisting [5–7]. Between 2003 and 2008 in Zambia’s major hospitals, almost one-third of all gynecologic admissions were due to complications of unsafe abortion, researchers estimated that 6 in every 1,000 of these women died as a result of their complications [8].

The impact of recent maternal health interventions on the incidence of unintended pregnancy and unsafe abortion remains unclear. Early childbearing is still very common, almost 60% of Zambian women have borne a child by age 19 [9]. Rural young women in poverty bear the brunt of morbidity from early childbearing, facing twice the risk of complications during pregnancy as older women and severely limited opportunities for continued education that help lift women and their families from poverty [9]. Regardless of education, socioeconomic status or place of residence, most Zambian women have more children than they originally wanted [9].

Until recently, legal abortion services have not been widely available in health centers or hospitals, compelling women to continue to turn to illegal providers and unsafe methods and confirming that just the existence of a law, with insufficient political will or guidance for implementation may not be effective in improving maternal health [6, 10–12]. Unsafe abortion – the termination of an unintended pregnancy either by persons lacking the necessary skills or in an environment lacking the minimal medical standards, or both – contributes to 8-13% of global maternal mortality [13, 14]. Although Zambia’s Maternal Mortality Ratio (MMR) declined in the preceding 7 years from 591 to 398 per 100,000 births in the 2014 Demographic and Health Survey, it remains high for the African continent; as many as 30% of these deaths could be the result of unsafe abortion according to an audit of maternal deaths at the University Teaching Hospital (UTH) in Lusaka [5].

Medical abortion (MA) using the World Health Organization’s recommended mifepristone and misoprostol combined regimen, is a highly effective method of pregnancy termination, with low complication rates up to and even after 9 weeks of gestation [15–17]. The use of medications offers women a non-surgical method of induced abortion, and has been proven to be highly acceptable to both clients and providers [18, 19]. Particularly in low-resource settings, with limited health care access and few trained surgical abortion providers, MA can increase safe abortion access, and decrease maternal morbidity and mortality associated with unsafe abortion by providing a safe and effective non-surgical option. However, at the time of this study, mifepristone was registered in only two African countries - Tunisia and South Africa; since 2009 only five additional African countries have registered mifepristone for importation - Zambia, Ghana, Mozambique, Ethiopia and Kenya [20]. Restrictive abortion laws, the expense of the medication, and reluctance by policymakers to shepherd mifepristone to broader markets and tackle policies necessary for wider distribution have contributed to few developing countries making this drug available.

In 2008, concerned with alarming maternal death rates and faced with daunting Millennium Development Goals (MDGs), the Ministry of Health (MOH) supported a national strategic assessment of unsafe abortion in the country. Findings revealed high demand for safe abortion in the country, the existing use of a wide variety of unsafe and traditional abortifacients, persistent high levels of provider stigma, logistical barriers to accessing safe abortion for women across the country and lack of specific provisions for youth [21]. On the heels of this assessment, Ipas, an international non-governmental organization (NGO) working in the field of reproductive health and rights, in collaboration with the Ministry of Health (MOH) and the University Teaching Hospital, conducted a comprehensive pilot project to introduce medical abortion (MA) with mifepristone and misoprostol in Zambia and to demonstrate a model for strengthening and scaling up safe abortion services to the extent allowed by the law. The project evaluation was designed using an implementation science model to document the factors and components that could be used to move a national comprehensive abortion care (CAC) program from limited to extensive service provision for replication in other parts of the continent.

This paper first describes the process and results of galvanizing access to medical abortion in an environment where abortion has been legal for many years, but provision of safe services has been severely limited. Secondly, results are presented to highlight the successes and the challenges of bringing medical abortion to an operationally complex context, a health system being rebuilt after enormous investment in HIV prevention and treatment, using implementation science to document the results of 2 years of implementation. Finally, these findings not only provide a case study of medical abortion in Zambia, but also offer important lessons and recommendations for expanding access to safe abortion in similar settings across Africa.

In response to the findings of the Strategic Assessment, a collaborative intervention between the MOH, UTH and Ipas was established with funding from the Consortium for Research on Unsafe Abortion in Africa with the aim of introducing medical abortion in Zambia for the first time while addressing the multiple concerns of the Strategic Assessment team. An implementation science model was proposed to evaluate and determine effectiveness as well as gleaning lessons learned about bringing safe and legal abortion services to scale. The intervention involved the provision of CAC services in 28 public health facilities in Zambia for a 2-year period, from August 2009 to September 2011. The study focused on three main areas: building health worker capacity to provide PAC and induced abortion through the first 14 weeks of pregnancy in public sector facilities and introducing abortion with medication, working with pharmacists to provide improved information and access to MA, and community engagement and mobilization to increase knowledge of abortion service availability and rights among the general public through stronger health system and community partnerships. The research was reviewed and approved by the University of Zambia Biomedical Research Ethics Committee on 10 August, 2009.

The health services package of activities included development and distribution of national standards and guidelines for safe abortion and the reduction of unsafe abortion, provider training in CAC, mentoring and supervision of providers, site renovations, registration of a medical abortion product, and the provision of MVA and medical abortion in 28 facilities of Lusaka and the Copperbelt. Health care workers were trained to provide induced abortion with either MVA or the combined product of 200 mg of mifepristone orally at the facility and 800 mcg of misoprostol 24–48 h later, either by returning to the facility or at home, depending on her preference and to treat incomplete abortion with MVA. Additionally, project staff worked with partners to build the capacity of the Planned Parenthood Association of Zambia (PPAZ), Youth Vision Zambia, Society for Women and Aids in Zambia, and Women in Law and Development in Africa and Women and Law in Southern Africa to create awareness about the law and services among their membership and reduce stigma in communities. The purpose was to create an enabling environment, provide more accurate information, appropriate referrals and greater community awareness to build support for access to sexual and reproductive health and rights, including safe abortion care. Figure 1 provides additional details on the three areas of intervention.

Perhaps the first and most critical lesson learned from this case study is the importance of engagement and support from local policy and implementation stakeholders. The impetus for the current efforts on CAC stemmed from national-level awareness of the impact of unsafe abortion on the maternal mortality rate and the increased global attention brought to this issue by the MDGs. This awareness presented an opportunity for partnerships, which resulted in both the strategic assessment on unsafe abortion and the subsequent implementation science study on the scale-up of CAC services. Global efforts, such as the MDGs, WHO global policy and technical updates, and the Sustainable Development Goals (SDGs) should be utilized to catalyze national efforts and interest in advancing abortion care and eliminating unsafe abortion. The strategic assessment itself united multiple local stakeholders and provided critical evidence in support of MA scale-up, which further spurred stakeholders to action. Without support from the MOH and local stakeholders, provision and scale-up of induced abortion services would have been impossible even after the evaluation findings.

Although the intervention was focused on induced abortion, the introduction and improvement of PAC services was an important extra benefit. Before the intervention, only four of the hospitals and two of the health centers provided PAC, and only the UTH provided induced abortion in any systematic way. While most PAC cases were treated in hospitals throughout the study, the intervention succeeded in expanding PAC to health centers – closer to where women live – in an effort to decentralize this important service and allow over-crowded hospital rooms to treat serious cases more efficiently. Introducing misoprostol for PAC in health centers could help in further decentralizing PAC to the health center level but by far the most important factor would be a reconsideration of guidelines to allow midlevel providers to train and provide CAC services without physician signatures. An expansion of the scope of MLPS in the provision of CAC would allow for the expansion of services to the most remote areas of the country and potentially improve contraceptive acceptance while decreasing unintended pregnancies and repeat abortions.

The study demonstrated that trained mid-level providers can deliver safe abortion services. However, values clarification appeared necessary to change attitudes among many groups (e.g., providers, facility managers, the general public, pharmacists and pharmacy workers) to improve abortion care and increase awareness of these services. Health workers still face opposition in their facilities and the need remains for management to create an environment more supportive about abortion care. Additionally, more open discussion about both safe and unsafe abortion at the community level, through volunteers and mass media, will continue to normalize abortion care and place it in the context of the package of essential maternal health care. Additional efforts are needed to support and normalize abortion care services and integrate them into the rest of the public health system.

The study results demonstrated that some pharmacists and pharmacy workers have a role to play in safe abortion services and are willing to play it. The mystery client interactions show an increase in pharmacists and pharmacy workers’ willingness to provide information, drugs for MA, and health facility referrals over the 2 years of the study. Mystery clients also reported fewer hostile interactions and increased sympathy from pharmacists and pharmacy workers.

Contraception plays a crucial role in reducing the need for abortion services, and almost half of women who received abortion care left with a contraceptive method. Postabortion contraceptive acceptance was better at health centers than at hospitals; however, contraceptive counseling and method provision should be better integrated into all abortion services. If these women had tried to induce their abortions on their own or had gone instead to unsafe providers, it is unlikely that they would have been counseled for contraception or have received a method. However, when asked directly after their procedures, there were still women who wanted a method who left the facility without one and many who received a method reported “wanting to learn more” about family planning.

At endline, many respondents did not know that abortion was legal and responded negatively to questions about their attitudes toward abortion. The lack of clarity and communication about the national abortion law and policies serves only to confuse the populace and prevent or delay women from seeking services to which they are legally entitled. The community engagement and mobilization component of the intervention improved community members’ knowledge about the legal abortion services available but had less impact in changing feelings and deep-rooted stigma about abortion services. Further efforts must be made to inform women of their right to and availability of safe abortion and to change community perceptions about this highly stigmatized topic.

Finally, to maintain and continue to make progress on access to safe abortion, there is a need for a broader coalition of partners to build support. This should include the MOH but not be limited to them. Champions for this project in the MOH have been highly mobile and have often moved to positions where they lose their ability to speak freely on sensitive topics. By creating a more robust base of support among policymakers, the safe abortion movement will not be as vulnerable to frequent political reappointments. Furthermore, there is a continued need to build a vertical but integrated CAC service within other maternal health or gynecological services. Placing abortion firmly within the constellation of safe and legal reproductive health services is the only means by which to ensure improved access for all women.

The experience of introducing MA and scaling up safe abortion in Zambia offers important lessons for other countries seeking to improve maternal outcomes by improving access to comprehensive abortion care. The Zambian experience can provide an avenue for safe abortion services in other contexts where abortion rights are provisionally legal but practically restrictive. By building local support among stakeholders, coordinating efforts of implementers and using evidence to guide and inform decision-making, study partners were able to scale-up access to abortion services in an otherwise barren landscape of care. However, as the results of the evaluation show, there is still much work to be done to ensure that the level of care remains high, that access across the country continues to grow, and that stigma for abortion services, be jettisoned in favor of putting women’s lives first.