Experience from a multi‐country initiative to improve the monitoring of selected reproductive health indicators in Africa

Universal access to sexual and reproductive health remains part of the unfinished business of global development in Africa. To achieve it, health interventions should be monitored using programmatic indicators. WHO's Strengthening Measurement of Reproductive Health Indicators in Africa initiative, implemented in Ghana, Nigeria, Kenya, Uganda, and Zimbabwe, aimed to improve national information systems for routine monitoring of reproductive health indicators. Participating countries developed action plans employing a two‐pronged strategy: (1) revising, standardizing, and harmonizing existing reproductive health indicators captured through routine information‐systems; and (2) building data‐collection capacity through training and supervision at select pilot sites. Country teams evaluated existing and new indicators, and outlined barriers to strengthening routine measurement. Activities included updating abortion‐care guidelines (spontaneous and induced abortions), providing training on laws surrounding induced abortions, and improving feedback mechanisms. The country teams updated monitoring and evaluation frameworks, and attempted to build recording/reporting capacity in selected pilot areas. Barriers to implementing the initiative that were encountered included restrictive induced‐abortion laws, staff turn‐over, and administrative delays, including low capacity among healthcare staff and competing priorities for staff time. The areas identified for further improvement were up‐scaling programs to a national level, creating scorecards to record data, increasing collaborations with the private sector, conducting related costing exercises, and performing ex‐post evaluations.

To reduce maternal mortality rates and improve maternal health, women must have access to high-impact interventions and quality reproductive services. Health-service interventions with the aim of reducing unmet need, preventing unsafe abortion, and managing complications due to abortion, contribute towards meeting these targets, particularly universal access to reproductive healthcare. However, to ensure these goals are met, interventions-established and implemented at a national level-should also be monitored and strengthened through the use of programmatic indicators. Global, regional, and national monitoring of carefully selected programmatic indicators would help in detecting periodic changes, identifying gaps to achieving goals, and directing actions toward achieving success.
Maternal mortality (and its causes) and unmet contraception needs were among the key monitoring indicators for assessing progress in achieving Millennium Development Goal 5, improving maternal health. Global and national monitoring can help detect changes in these indicators, aiding understanding of progress toward Millennium Development Goal 5 and its two targets: (1) reducing maternal mortality, and (2) achieving universal access to reproductive health ( Figure 1). 3 For almost all indicators, the slowest progress rates have been in African countries. 1 Despite the continent having experienced a notable decrease in unmet family-planning needs (−1.9% between 2000 and 2015, second only to −2.0% in Latin America and Caribbean regions), 4 24% of women in Africa remain without access to family planning; this is the highest percentage worldwide. 4 In 2007, to assist countries in this endeavor, WHO and the United Nations Population Fund developed a guide for national-level monitoring of progress toward universal access to reproductive health. 5 The guide provided a framework (hereafter referred to as the 2007 framework) to monitor progress toward universal access to sexual and reproductive healthcare, consisting of a set of indicators supporting the collection of reproductive healthcare data. The 17 indicators presented in the 2007 framework underwent a rigorous consultation process with the goal of ensuring they addressed the main aspects of sexual and reproductive healthcare and could be used to monitor progress in terms of removing barriers, improving health systems, and positively addressing needs and risks within this area. 5 Additionally, the 2007 framework was intended to help guide decision-making to identify and implement activities that could accelerate progress toward universal access to reproductive healthcare.
This underscores the increased need for quality indicators-crucial for intervention programs-particularly, in Sub-Saharan Africa. 6,7 The 2007 framework, which remains relevant owing to renewed global commitment to reducing maternal mortality and morbidity, formed the basis for the initiative described in the present manuscript.

| PROGRAM DESCRIPTION
Between November 2012 and May 2015, the WHO's Department of Reproductive Health and Research led the Strengthening F I G U R E 1 Unmet need for family planning by country for 2015. Unmet family-planning need was calculated as the percentage of women aged 15-49 years who were married or in unions who wished to delay childbearing but were not using contraception.
Measurement of Reproductive Health Indicators in Africa initiative, providing implementation support to the Ministries of Health of Ghana, Nigeria, Kenya, Uganda, and Zimbabwe. Each country developed an action plan to strengthen information systems through the inclusion of selected reproductive health indicators (regarding family planning and safe abortion) to better monitor actions taken to reduce maternal mortality and improve maternal and reproductive health.
Technical teams from the Ministry of Health in each study country and WHO developed plans to strengthen national information systems by: (1) revising existing reproductive health indicators, (2) incorporating new family-planning and safe abortion care indicators, including post-abortion contraceptive use by method, (3) piloting both revised and new indicators at a sub-national level (through training and supervision), and (4) strengthening feedback mechanisms at both the sub-national and facility level.
Each country's plan outlined context-specific activities that supported the two main aims of the initiative: to strengthen reproductive health indicators (their definition, collection, and monitoring) and to ensure that efforts yielded improved information for decision making.

| PROGRAM PLANNING
In November 2012, a multi-country workshop was convened in Nairobi, Kenya, to introduce the reproductive health-program monitoring indicators (based the 2007 framework) related to family planning and safe abortion. 8 Country teams were comprised of Ministry of Health program officers, monitoring and evaluation officers, and WHO reproductive health technical staff (from both country offices and headquarters). During the meeting, participants received an overview of program implementation, monitoring and evaluation frameworks, and reproductive health indicators-including their characteristics, data sources, quality, and analysis (both quantitative and qualitative).
The countries present at the workshop were Ethiopia, Ghana, Kenya, Nigeria, Sierra Leone, Swaziland, Tanzania, Uganda, Zambia, and Zimbabwe. Based on the aforementioned discussions, countries conducted systematic evaluations of the status of existing reproductive health indicators (Appendix S1). This assessment was based on data sources, data quality, and strategies for improvement. At the same workshop, countries also identified and prioritized new indicators (drawn from the 2007 framework) that were based on data sources, feasibility, and requirements for routine reporting. Additionally, the country teams outlined barriers to strengthening the routine measurement of those indicators deemed to be high priority during the evaluation process.
All country teams present at the workshop were invited to submit a concept note and five countries were selected to submit full proposals for the initiative (implemented between August 2013 and December 2015). These countries were Ghana, Kenya, Nigeria, Uganda, and Zimbabwe. Across these five countries, the main objective of the proposals was to apply country-specific strategies aimed at improving the monitoring of reproductive health services and related indicators. Once the proposals were finalized and approved, the country teams conducted activities and prepared programmatic updates approximately every 6 months using a standardized reporting form (Appendix S2). A mid-point review meeting was held in Kampala, Uganda, in September 2014, where countries were able to present and share learning experiences from their strategies, update work plans, and to prepare for the final implementation phase.

| RESULTS
Based on the activities at the Kenya workshop in 2012, country teams prepared proposals that prioritized indicators and identified data sources. The majority of the core indicators prioritized, such as unmet need for family planning, adolescent birth rate, total fertility rate, and contraceptive use, were already collected as part of Demographic Health Surveys, and countries relied on these surveys to measure these indicators. The prioritized indicators that were not collected as part of Demographic and Health Surveys were identified (Table 1), and country teams identified major barriers to tracking these indicators. These barriers were grouped under the three main areas: (1) policy, national standards, leadership (including financial costs), and inadequate commitment to monitoring and evaluation activities at the sub-national level; (2) weaknesses in data collection and information systems, including inadequate distribution logistics for data collection tools, a lack of systematic feedback mechanisms, low data quality, and weak linkages among data sources; and (3) issues related to health workers, including excessive staff reporting obligations, inadequate staff numbers, and high staff turn-over.

| Nigeria
As part of the initiative described in the present manuscript, the

| Uganda
Within Uganda, there was a supportive family-planning policy envi-

| Zimbabwe
Prior to the initiative described in the present manuscript, Zimbabwe had a national integrated data-collection system. It included datacollection using death-notification forms (completed within 7 days of a maternal death), and sexual and reproductive healthcare data sheets and registers (employed at the facility level; data sheets were used to report aggregated information on a monthly basis and registers were

| DISCUSSION
The five countries that participated in the present initiative revised and updated national sexual and reproductive healthcare-related indicators, and monitoring and evaluation plans, with a specific focus on family planning and safe induced-abortion services. Each country team tailored their implementation plans following the initial analysis and discussions. Country teams reported that effective collaboration with relevant stakeholders was important throughout the implementation phase. Pilot programs were implemented in at least two sub-national areas in each of the five countries to investigate the application of updated data-collection tools; alongside this, sub-national and facility-level healthcare workers received training in implementing the data-collection tools.
The During the implementation phase, country teams highlighted the harmonization of data-collection tools as a priority. This harmonization was reported as being important, not only nationally, but also across partner organizations to ensure that data collected at the facility level were then included in higher level databases. Efforts to synchronize data collection were necessary to avoid overlap and superfluous collection, trends that can occasionally be propagated by donors and other stakeholders. 12 Population-based surveys (such as the Demographic Health Surveys) are important but they should not be used as the sole source for core reproductive health indicators. During the implementation phase of the present initiative, the feasibility of using routine health information systems to track indicators and progress toward targets was highlighted. The development of more robust feedback loops is also crucial for improving data collection, accountability, and support to staff in the implementation of monitoring and evaluation activities to improve service provision. This could be initiated through the implementation of a bulletin, as was reported by some country teams in the present initiative, and/or the inclusion of indicators in the Reproductive, Maternal, Neonatal, Child and Adolescent Health scorecard, 13 which is currently produced quarterly in a number of countries in the African region.
In the present initiative, effective co-ordination between coun- During the course of the initiative, some limitations were encountered. Although participating countries used a two-pronged approach in implementing the project (working at the national level to build consensus, and at the sub-national level to implement changes), it was difficult to draw comparisons across the five countries. Each participating country had its own priorities and activities that were dependent upon national political conditions (particularly in terms of policies surrounding induced abortion), and interventions during the implementation phase varied depending on national-level circumstances.
Additionally, high staff turnover, particularly in monitoring and evaluation positions (owing to personal reasons or, elections within higherlevels of government) made it difficult to sustain field activities. During the initiative, elections in Kenya resulted in delays to activities of longer than 6 months and high staff turnover in Ghana resulted in training at sub-national facilities having to be repeated. Finally, another limitation was a lack of knowledge regarding the impact of indicators themselves; although the initiative aimed to improve data collection and use, country teams did not track changes in specific healthcare measures such as contraceptive-use rates.
The conceptual framework developed ( Figure 2) provided a standard approach to measuring the outcomes of reproductive health programs in Africa. Effective implementation was facilitated by basing projects on a common protocol with a clear conceptual framework that allowed countries to tailor programs to specific settings. In terms of future steps, it was intended that country teams could: The need for robust health information systems is highlighted by the momentum and political will surrounding the sustainable development goals. 7 Specifically, there is a need to strengthen existing national health information systems to routinely collect complete and accurate data to enable the timely implementation of decisions, particularly in terms of induced-abortion and family planning services in Sub-Saharan Africa, where demand for limiting births is increasing. 15 The collection and monitoring of data and indicators at the local, national, and global level would also present the necessary tools for increasing accountability. A two-pronged strategy to standardize reproductive health indicators, including stakeholder support at the national level, and increased data-collection capacity and supervision at the sub-national level, could strengthen the routine gathering of information on family planning and safe induced-abortion services.

AUTHOR CONTRIBUTIONS
MB, ÖT, and LS guided the conception, design, and implementation of the project, and prepared the manuscript. NM and AAA assisted in the conception and design of the project, provided technical assistance, and edited and revised the manuscript. All authors approved the final manuscript and agreed to be accountable for all aspects of the work.

ACKNOWLEDGMENTS
The project described in this manuscript was funded by the WHO