CLAimHealth: COLLABORATING, LEARNING, AND ADAPTING FOR IMPROVED HEALTH ACTIVITY Evaluating the Effectiveness of After-Action Reviews for Adaptive Management in Selected USAID/Philippines Health Project Interventions March 16, 2022 DISCLAIMER The views expressed in this report do not necessarily reflect the views of the United States Agency for International Development or of the United States government. This report was produced for review by the United States Agency for International Development. It was prepared by Panagora Group for the USAID/Philippines Collaborating, Learning and Adapting for Improved Health (CLAimHealth) Activity, IDIQ No. AID-OAA-I-15-00025, Task Order No. 72049218F00001 as a technical deliverable per Task Order Section F.6 (b) and therefore uses USAID branding as outlined in the CLAimHealth Branding and Marking Plan. Recommended citation: USAID/Philippines Collaborating Learning and Adapting for Improved Health (CLAimHealth) Activity. Evaluating the Effectiveness of After-Action Reviews for Adaptive Management in Selected USAID/Philippines Health Project Interventions. March 2022. Panagora Group Contact: Mary Ann Lansang, MD Chief of Party 11/F, Ramon Magsaysay Center – Tower Manila1004, Philippines ACKNOWLEDGMENTS Panagora Group and CLAimHealth gratefully acknowledge the contributions of the Planetary and Global Health Program of the St. Luke’s Medical Center College of Medicine – William H. Quasha Memorial (SLMCCM – WHQM) in producing this report. The team is composed of Ramon Lorenzo Luis R. Guinto, MD DrPH (Principal Investigator), Rafael Jairah D. Matoy, Jr., MPH, RN (Research Fellow), and Michael B. Salise, MHSS, RND (Research Associate). We also acknowledge the contributions of the following resource persons from the three implementing partners that participated in this implementation research: ReachHealth • Dr. Laurentiu Stan, Chief of Party • Dr. Easter Dasmariñas, Deputy Chief of Party • Dr. Ophelia Mendoza, MERLA Director • Mr. Glenn Bede Benablo, Regional Manager • Ms. Algin Gultia, Program Officer for Luzon, Social and Behavior Change & Gender RenewHealth • Dr. Regina Hechanova-Alampay, Chief of Party • Dr. Ivanhoe Escartin, Deputy Chief of Party • Ms. Aimee Malinga, Monitoring, Evaluation and Learning Director • Dr. Jason Manaois, Learning and Development Specialist TBIHSS • Dr. Soliman Guirgis, Chief of Party • Dr. Lalaine Mortera, Deputy Chief of Party • Mr. Anup Kumar, Strategic Information Lead • Ms. Maria Althea Sabrina Perez, Monitoring and Evaluation Specialist • Dr. Bea Barbara Carrascal, Public-Private Mix Specialist The SLMCCM – WHQM team also wishes to express their gratitude to Dr. Pilar Ramos-Jimenez and Ms. Justine Co of CLAimHealth for their generous collaboration. TABLE OF CONTENTS ACKNOWLEDGMENTS ............................................................................................................................................... iii ACRONYMS....................................................................................................................................................................... ii EXECUTIVE SUMMARY.................................................................................................................................................. 1 1. INTRODUCTION........................................................................................................................................................ 4 2. LEARNING QUESTIONS AND OBJECTIVES...................................................................................................... 5 2.1. Primary Learning Question........................................................................................................................... 5 2.2. Secondary Learning Questions..................................................................................................................... 5 2.3. Objectives ......................................................................................................................................................... 5 3. AFTER-ACTION REVIEWS: A BRIEF REVIEW..................................................................................................... 6 3.1. Forms of AAR........................................................................................................................................................ 6 3.2. The AAR Process.................................................................................................................................................. 7 3.3. Evaluating AARs..................................................................................................................................................... 8 3.4. Pause and Reflect Among Different USAID Implementing Partners........................................................ 8 4. METHODOLOGY......................................................................................................................................................10 4.1. Overall Research Design...................................................................................................................................10 4.2. Selection of Cases...............................................................................................................................................10 4.3. Data Collection....................................................................................................................................................10 4.4. Data Analysis........................................................................................................................................................13 4.5. Analytical Framework.........................................................................................................................................13 4.6. Limitations of the Study.....................................................................................................................................16 4.7. Ethical Considerations .......................................................................................................................................16 5. RESULTS .......................................................................................................................................................................17 5.1. Implementing Partner 1: Improved Health for Underserved Filipinos: Family Planning and Maternal and Neonatal Health Innovations and Capacity Building Platforms (ReachHealth)...................17 5.2. Specific Intervention 1: ReachHealth’s Family Planning (FP) Ayuda Express...................................19 5.3. Implementing Partner 2: RenewHealth - Expanding Access to Community-based Drug Rehabilitation................................................................................................................................................................25 5.4. Specific Intervention 2: RenewHealth’s Motivational Interviewing Training ...................................28 5.5. Implementing Partner 3: TB Innovations and Health Systems Strengthening (TBIHSS) Project 31 5.6. Specific Intervention 3: TBIHSS’s Philippine Private Sector Diagnostics Consortium (PPSDC) 35 6. DISCUSSION...............................................................................................................................................................39 6.1. The 7 P’s of AARs in the IPs.............................................................................................................................39 6.2. Achievement of Implementation Outcomes Related to the Implementation of AARs......................41 6.3. AARs as Applied to Specific IP Interventions...............................................................................................42 6.4. Impact of AARs on Implementation Outcomes Related to the Intervention.......................................43 7. CONCLUSION...........................................................................................................................................................45 8. RECOMMENDATIONS............................................................................................................................................46 8.1. For USAID ............................................................................................................................................................46 8.2. For Implementing Partners...............................................................................................................................46 8.3. For Implementation Researchers....................................................................................................................46 REFERENCES....................................................................................................................................................................47 ANNEX A. SURVEY QUESTIONNAIRE ..................................................................................................................49 General Questionnaire About AAR Practices.....................................................................................................49 Questionnaire About the Specific Intervention...................................................................................................52 ANNEX B. LIST OF DOCUMENTS REVIEWED....................................................................................................54 i LIST OF TABLES AND FIGURES Table 1. AAR Formats...................................................................................................................................................... 7 Table 2. Group-Based Pause and Reflect Interventions........................................................................................... 9 Table 3. Selected USAID/Philippines HP Activities, Their Justification, and Specific Interventions.............11 Table 4. AAR Effectiveness Evaluation Framework Based on Widely-Used Implementation Outcomes..15 Table 5. AAR Good Practices and Resulting Intervention Adaptations Across Three IPs............................43 Figure 1. AARs as a Type of P&R activity .................................................................................................................... 8 Figure 2. Study Design....................................................................................................................................................10 Figure 3. Seven P’s Framework ....................................................................................................................................14 Figure 4. AAR Planning Roadmap ................................................................................................................................14 Figure 5. FP Ayuda Express’ Implementation Phases..............................................................................................20 Figure 6. Social Media Cards from FP Ayuda Express............................................................................................21 Figure 7. Sample Chat Inquiry in Facebook Messenger..........................................................................................21 Figure 8. AAR Sessions Held During FP Ayuda Express’ Implementation Period ...........................................23 Figure 9. Sample Content Based on Client Testimonials.......................................................................................24 Figure 10. General Phases of KKDK-CBDR Training, Including MI ....................................................................28 Figure 11. Steps in the AAR Process for MI Training .............................................................................................29 Figure 12. AAR Sessions Held During the Implementation of MI Training .......................................................30 Figure 13. Eisenhower Decision Matrix .....................................................................................................................34 Figure 14. AAR Sessions Held During PPSDC’s Implementation PeriodError! Bookmark not defined. ii ACRONYMS 7P’s people, preparation, process, performance, purpose, period, and place AAR after-action review AMELP Annual Monitoring, Evaluation, and Learning Plan CBDR community-based drug rehabilitation CLA collaborating, learning, and adapting CLAimHealth Collaborating, Learning and Adapting for Improved Health DOH Department of Health DOH-CHD Department of Health Center for Health Development FP family planning HP Health Project IP implementing partner KKDK Katatagan Kalusugan at Damayan ng Komunidad L&D learning and development LGU local government unit M&E monitoring and evaluation MI motivational interviewing MEL monitoring, evaluation, and learning MERLA monitoring, evaluation, research, learning and adapting NTP National Tuberculosis Control Program NTRL National Tuberculosis Reference Laboratory OH Office of Health P&R pause and reflect PACT Partnership among Academes and Communities: Teaming up for Improved Family Planning Services in Bicol PDSA Plan-Do-Study-Act Cycle PhilCAT Philippine Coalition Against Tuberculosis PhilSTEP Philippine Strategic Tuberculosis Elimination Plan iii PIR program implementation review POPCOM Commission on Population and Development PPM public-private mix PPSDC Philippine Private Sector Diagnostics Consortium PWUD people who use drugs ReachHealth Improved Health for Underserved Filipinos: Family Planning and Maternal and Neonatal Health Innovations and Capacity Building Platforms RenewHealth Expanding Access to Community-Based Drug Rehabilitation in the Philippines TBIHSS Tuberculosis Innovations and Health Systems Strengthening UHC Universal Health Coverage URC University Research Co. 1 EXECUTIVE SUMMARY Through its Office of Health (OH), the United States Agency for International Development (USAID) is implementing a project in the Philippines to enhance the capacity of the national government, local government units (LGUs), and the private sector to provide quality, efficient health service delivery. USAID OH utilizes a collaborating, learning, and adapting (CLA) approach to ensure that USAID￾financed activities are on track to achieve their objectives. One of the tools in the CLA approach that USAID recommends is the After-Action Review (AAR), which is an assessment conducted after a major activity or after an identifiable event within a project or major intervention. It allows team members and leaders to discuss what happened and to reassess direction; review successes and challenges; and identify action steps to meet goals. USAID’s Collaborating, Learning, and Adapting for Improved Health activity (CLAimHealth), one of the eight ongoing activities of the USAID/Philippines Health Project, aimed to document and assess AAR processes across USAID/Philippines activities. This report presents the findings of implementation research that examined AAR practices and experiences among three of USAID’s implementing partners (IPs): 1. Improved Health for Underserved Filipinos: Family Planning and Maternal and Neonatal Health Innovations and Capacity Building Platforms (ReachHealth) 2. Expanding Access to Community-Based Drug Rehabilitation in the Philippines (RenewHealth) 3. TB Innovations and Health Systems Strengthening (TBIHSS) Project. The objectives of this implementation research project are as follows: 1. Document the AAR processes of three IPs under the USAID/Philippines Health Project, highlighting what is working well and what is not working well, and showing how they contribute to adaptive management (i.e., generating short learning loops or activating the Plan-Do-Study￾Act (PDSA) cycle). 2. Generate evidence on how AAR practices contributed to adaptive management and the further improvement of a specific intervention implemented by the IP over time. 3. Assess the effectiveness of AAR practices in creating learning and improving performance, using six implementation outcomes: fidelity, adaptation, accountability, integration, inclusivity, and sustainability. 4. Provide recommendations on how AARs can be more effective in improving the Health Project’s interventions. This implementation research team adopted an exploratory, retrospective, embedded multiple case study research design in which IPs are the main cases and a specific intervention from each IP serves as the ‘case within a case.’ They studied the following specific interventions: • ReachHealth’s Family Planning (FP) Ayuda Express • RenewHealth’s Motivational Interviewing Training • TBIHSS’s Philippine Private Sector Diagnostics Consortium (PPSDC) 2 To develop the individual case studies, the research team used three data collection methods: holding preliminary meetings with IPs, providing survey questionnaires to key IP representatives, and reviewing documents obtained from the IPs. The team conducted data analysis in two stages: ‘within-case’ analysis and ‘between-case’ analysis. These were informed by three frameworks: 1. The Seven P’s Framework, which looked at the anatomy of the AAR process (people, preparation, process, performance, purpose, period, and place) 2. The AAR Planning Roadmap, which depicts the different phases of the AAR process (pre-, during-, and post-AAR phases) 3. The AAR Effectiveness Evaluation Framework, which is based on six widely-used implementation science outcomes (fidelity, adaptation, accountability, integration, inclusivity, and sustainability). The case studies of ReachHealth, RenewHealth, and TBIHSS showed that projects are conducting AARs in different forms, reflecting their diversity in terms of period, place, people involved, preparation, and process. They include stand-alone pause and reflect (P&R) sessions, embedded P&R sessions in routine meetings (such as program implementation reviews), and post-session debriefings after a particular training event. Despite the different terminology and methodologies that IPs use to conduct AARs, , their AAR activities share a common purpose of strengthening and continuously improving project implementation. AARs involve internal team members and external partners in identifying and assessing enabling factors as well as barriers, gaps, and challenges in implementation. Projects then use such valuable information to plan and execute course correction strategies and manage emerging challenges. The repeated AAR activities are meant to eventually help the IP achieve its ultimate goal through continuous process improvements. As a sign of adaptation, AAR activities underwent a notable shift from in-person to virtual sessions in response to restrictions imposed by the COVID-19 pandemic. AAR sessions also helped enhance the sense of accountability among team members. There are additional indications that these AAR practices promoted inclusivity through their participants and processes; were integrated into CLA plans and routine operations; and were adequately resourced to ensure they could be sustainably conducted over time. When applied to the three specific interventions that the selected IPs implemented, AARs and the learnings they generated were an integral part of improving processes and implementation. AARs served as a major knowledge sharing opportunity for both internal and external stakeholders to assess the implementation of each intervention, identify gaps and challenges that require adaptive actions, and determine best practices and lessons learned. The research team used different types of AARs to review the studied interventions: P&R sessions for ReachHealth’s FP Ayuda Express; post-session debriefings for RenewHealth’s Motivational Interviewing Training; and Governing Council Meetings for TBIHSS’s PPSDC. AARs enabled teams to successfully address concerns in a timely manner such as increasing responsiveness to clients in FP Ayuda Express; improving the content and delivery in motivational interviewing training; and expanding the membership of PPSDC. Central to the usefulness of AARs is the considering feedback from end-users to development actions for modifying and improving interventions. Moreover, applying AARs in the specific interventions also helped create a sense of ownership and accountability among different stakeholders, promote inclusivity, and develop strategies for eventual sustainability and integration of the intervention as part of the health system. The three case studies in this research (i.e., the three IPs and their specific interventions) demonstrated that there are different understandings of AARs and of the ways to conduct them. This is not necessarily a problem, provided that these activities still have the key elements of an AAR and fulfill the same purpose. The cases that the research team examined demonstrated how adopting AAR practices helps 3 in adaptive management, which is the basic purpose of AARs and the reason for their and effectiveness. Activating rapid learning cycles allows the IPs to immediately integrate lessons learned from failures and successes into implementation. This an essential ingredient for iteration, improvement, and innovation. The AAR practices documented in the case studies also revealed numerous other benefits: enhancing engagement between different stakeholders (both internal and external); promoting accountability and ownership for the intervention, which is vital for its long-term sustainability especially beyond the IP’s implementation period; and enhancing the responsiveness of interventions by incorporating the voices of clients and users. Based on the lessons generated from this research, CLAimHealth makes the following recommendations: For USAID • Clarify the definition of AAR, its types and dimensions, and existing tools and guidelines. • Explicitly and actively promote the use of AARs as part of the CLA approach to raise awareness and understanding, and increase uptake. • Create platforms for joint learning about AAR practices across IPs. • Encourage and support continuous documentation of AAR practices and their effectiveness through case study development and implementation research. For implementing partners • Adopt existing AAR design templates to ensure quality, enhance repeatability, and ease documentation and reporting. • Consider the use of existing tools for prioritization (e.g., the Eisenhower Decision Matrix), tracking progress (e.g., the Responsibility Matrix), and building accountability. • Improve documentation of AAR practices, agreements, and outcomes, including through case study development and implementation research. • Share good practices on conducting AARs with other IPs. For implementation researchers • Examine similar IPs/organizations/interventions, but with different exposures to AAR practices; and consider including a control that does not practice AAR at all. • Conduct more quantitative investigations to build evidence on the causal impacts of AAR practices on the ultimate outcomes of activities and interventions. • Gather the perspective of external stakeholders on how important and useful to them are the AARs conducted by the IPs (especially external AARs) 4 1. INTRODUCTION Through its Office of Health (OH), the United States Agency for International Development (USAID), is implementing a project in the Philippines to enhance the capacity of the national government, local government units (LGUs), and the private sector to provide quality, efficient health service delivery. The USAID Health Project (HP) contributes to the attainment of the Philippines’ National Objectives for Health and is carried out under a Development Objective Agreement with the Government of the Philippines. Through activities that its implementing partners (IPs) manage, the HP contributes to health system strengthening (HSS) by raising demand for and access to family planning (FP), adolescent reproductive health (ARH), maternal and child health services, tuberculosis (TB) treatment and prevention, and community-based drug rehabilitation (CBDR). To ensure that USAID-financed IPs are on track to achieve their objectives, the USAID OH utilizes a collaborating, learning, and adapting (CLA) approach. The CLA approach informs the implementation framework that IPs employ and enables them to be responsive to challenges as they emerge. One of the tools in the CLA approach that USAID recommends is the After-Action Review (AAR). According to USAID, an AAR is an assessment conducted after a major activity, or after an identifiable event within a project or major activity. It allows team members and leaders to discuss what happened and why to reassess direction; review successes and challenges; and identify action steps to meet goals (USAID, 2006). AARs foster a culture of learning, knowledge sharing, collaboration, and adaptive management, which is defined as “an intentional approach to making decisions and adjustments in response to new information and changes in context” (USAID, 2021). In terms of organizational learning, AARs provide the opportunity for teams to continuously assess their performance and learn from successes and failures. In general, AARs follow a sequence of four steps: planning, preparation, implementation, and follow-through. As such, AARs are intended to generate short learning loops for adaptive management among OH and IPs, functioning to facilitate the iterative plan-do-study-act (PDSA) cycle, and facilitate the use of data and documentation over time to inform next steps and follow-through (Taylor, et al., 2014). There are different types of AARs, including formal reviews that are regularly scheduled, more informal pause-and-reflect (P&R) sessions, and individualized meetings in the field (e.g., spot inquiries), and others (WHO, 2019). USAID’s Collaborating, Learning, and Adapting for Improved Health activity (CLAimHealth), one of eight activities in the USAID/Philippines Health Project, intended to document and assess AAR processes across USAID/Philippines activities. This report presents the findings of implementation research that examined AAR practices and experiences of three IPs: 1. Improved Health for Underserved Filipinos: Family Planning and Maternal and Neonatal Health Innovations and Capacity Building Platforms (ReachHealth) 2. Expanding Access to Community-Based Drug Rehabilitation in the Philippines (RenewHealth) 3. TB Innovations and Health Systems Strengthening (TBIHSS) Project. 5 2. LEARNING QUESTIONS AND OBJECTIVES 2.1. Primary Learning Question Are the after-action reviews that USAID/Philippines’ Health Project implementing partners conduct effective in generating short learning loops for adaptive management, and quality improvement of health interventions and technical assistance? What is working well and what can be enhanced? 2.2. Secondary Learning Questions • How are government and local implementing stakeholders actively involved in the AAR process, including adaptive management? Should AARs involve appropriate representatives from national and local partners, or ensure their inputs are solicited? • In terms of AAR follow-through, how are agreed-upon adaptive management actions brought down to the level of the right partners for action and implementation? 2.3. Objectives The objectives of this implementation research project are as follows: 1. Document the AAR processes of three (3) IPs under the USAID OH Philippines Health Project, highlighting what is working well and what is not working well, and showing how they contribute to adaptive management (i.e., how well they generate short learning loops or activate the PDSA cycle). 2. Generate evidence on how AAR practices contributed over time to adaptive management and further improvement of a specific intervention implemented by the IP. 3. Assess the effectiveness of AAR practices in creating learning and improving performance, using six implementation outcomes: fidelity, adaptation, accountability, integration, inclusivity, and sustainability. 4. Provide recommendations on how AARs should be implemented by other IPs. 6 3. AFTER-ACTION REVIEWS: A BRIEF REVIEW Given the complexity of health problems, it is imperative to ensure that health interventions are responsive to needs. Considering the dynamics of public health issues, practitioners must constantly capture learnings within their organizations to find creative ways to develop interventions that address challenges as they emerge. One example of activities that harvest lessons learned during program implementation is after-action reviews (AAR). An AAR is an assessment conducted after a major activity or an identifiable event within a project or intervention. It allows team members and leaders to track progress, correct unintended impacts, and ensure that planned outcomes are achieved (USAID, 2006; UNDP, 2007). Originally, an AAR was conducted at the end of military training for leaders to reflect on and quickly adopt critical analysis provided to them by observers (National Police Foundation, 2020). Nowadays, it fosters a culture of learning, knowledge sharing, collaboration, and adaptive management in organizations because it is a way for them to reflect and learn while performing (Baird, et al., 1999). AARs are structured around one basic analytical methodology that aims to respond to four main questions (WHO, 2019): 1. What was supposed to happen? 2. What actually happened? 3. Why was there a difference? 4. What can we learn from this? The simplicity of this framework allows various organizations to assess the processes of their activities, analyze how these processes performed compared to their planned assumptions, and identify reasons for variance or determine actions for improvement (WHO, 2019; CLIC, nd) 3.1. Forms of AAR There are three forms of AARs that various organizations usually practice: formal, informal, and personal/individual. Formal AARs are resource-intensive and usually involve planning, coordinating, and preparing supportive training aids, an AAR site, support personnel, and time for facilitation and report preparation (WHO, 2019). Informal AARs are usually carried out by the people who are responsible for an activity. If necessary, the discussion leader or facilitator can be identified beforehand or chosen by the team itself (WHO, 2019). Finally, individual or personal reviews usually concern individual reflection on the course of action or activities of immediate past events (USAID, 2013). The type of AAR to be used may vary depending on the activity to be reviewed, the level of various team members’ participation in the project, the scope of the project, and a general assessment of which approach might work best (Hengeveld-Bidmon, 2015). AARs may also be conducted through different formats, which offer collective learning and operational improvement after a public health response (WHO, 2019). Table 1 lists different formats that can be used in conducting AARs. 7 Table 1. AAR Formats (WHO, 2019) Format Description Best Used For Debrief An informal facilitator-led discussion with no more than 20 participants focusing on a limited number of functional areas (3 or less), taking place over half a day with more focus on learning within a team Smaller responses/activities Working group An interactive format that consists of guided group work and plenary sessions with up to 50 participants focusing on more than 3 functional areas, taking place over 2 to 3 days Responses that involve multiple sectors Key informant interview A longer and more in-depth review of an event that begins with a literature review and feedback surveys, followed by semi-structured interviews and focus group discussion, taking place over several weeks Complex and larger responses where those involved can no longer be brought together, or where confidentiality and non-attribution are necessary for honest and open feedback Mixed-method A blended approach of the above three formats consisting of a working group AAR with contents supplemented from key informant interviews Responses with a large scope where the majority (but not all) of those involved can be brought together 3.2. The AAR Process In general, AARs follow a sequence of four steps: planning, preparation, implementation, and follow￾through (Serrat, 2008). As such, an AAR is intended to generate short learning loops for adaptive management among organizations or institutions, facilitate the iterative Plan-Do-Study-Act (PDSA) Cycle, and encourage the use of data and documentation over time to inform next steps and follow￾through (Taylor, et al., 2014). Guidelines for conducting an AAR are available, such as the ones developed by the World Health Organization (WHO) and USAID. Sexton and McConnan (2003) also proposed a checklist that is structured similarly to the WHO guidelines. That checklist is divided into five components: 1) before the AAR event; 2) during and after the event; 3) disseminating event findings; 4) monitoring and follow-up; and 5) development activities. Each component lists several tasks accompanied by a subtle reminder to ensure smooth task execution and achievement of objectives. Having a database is also useful to connect AAR experiences for future improvement. Effectively managing data generated by AAR activities helps stakeholders understand the context and mechanisms that drive successful and unsuccessful practices; identify and share best practices; and drive individual and organizational improvement. Weak data management structures, data sharing restrictions, and concerns about staff privacy and job security are challenges to collecting information from AAR processes. Hence, it is necessary to offer incentives that encourage teams and their individual members to report and document AAR findings. It is also important to frame reporting systems as more than a way to ensure accountability, but also to facilitate organizational learning and system improvement, as is the case with the aviation industry's critical incident registry (WHO, 2019). 8 3.3. Evaluating AARs While an AAR itself has an evaluative purpose, there is limited literature on how to evaluate the AAR process and its outcomes. One comparative case study by Sexton and McConnan (2003) studied the AAR practices of three organizations: World Vision International (WVI), the British Red Cross Society (BRCS), and the Joint Emergency Food Aid Programme (JEFAP). Drawing lessons from an extremely small sample of past AAR-type events, the study found out that even the simplest approach to learning is difficult to execute because of an environment with a complex arrangement. The AARs’ most critical components—which are reflective and deeper probing work around identifying critical issues--tend to be overlooked more often than not. Another study used a bare-bones meta-analysis approach to examine the effectiveness of AARs (Keiser and Arthur, 2021). Based on 61 studies, it showed that AARs led to an overall improvement in multiple training evaluation criteria. The study suggested that a highly structured AAR is more effective than a less structured AAR in the military, but either highly- or lowly-structured AARs displayed comparable effectiveness in the healthcare setting. During the COVID-19 era, a more recent study in a tertiary care hospital by Sorbello et al. (2021) evaluated the AAR process by conducting an AAR itself. Overall, the hospital’s COVID-19 response was deemed to be successful and timely, underscoring the importance of three components: governance, health personnel, and a multidisciplinary approach. Meanwhile, the participants in the AAR cited poor communication management as a major factor that negatively affected the hospital's response. 3.4. Pause and Reflect Among Different USAID Implementing Partners As mentioned earlier, USAID is endorsing the incorporation of CLA practices into IP activities (USAID Learning Lab, nd). Pause and reflect (P&R) is one CLA method that USAID promotes, emphasizing that pausing and reflecting regularly identifies what is working and what needs adapting. It also allows organizations to consider the impact of changes in the operating environment or context (USAID, 2018). The USAID CLA Toolkit for Adaptive Management suggests that P&R can be done individually or in groups. Individual-based P&R can be done through journaling or blogging. Meanwhile, examples of group-based P&R sessions, which include AARs, are listed in Table 2. Because AARs are seen as one type of P&R activity, the formats for AAR (according to WHO) and the types of P&R (according to USAID) can be visualized in the manner shown in Figure 1. Figure 1. AARs as a Type of P&R activity (Adapted from USAID, 2018 and WHO, 2019) 9 Table 2. Group-Based Pause and Reflect Interventions Activity Description Portfolio Review A periodic review of all aspects of a USAID Mission/Office's assistance objective, projects, and activities, often held prior to preparing the annual Joint Operational Plan After-Action Review An assessment originated in the U.S. military, conducted after a project or major activity to allow team members and leaders to discover (learn) what happened and why, reassess direction, and review successes and challenges Hot Wash A more informal equivalent of an AAR also originating from the U.S. military, comprising a facilitated discussion immediately following an exercise to capture feedback on issues, concerns, and proposed improvements from implementers Liberating Structures A series of innovative approaches to help groups tap into their collective know-how and creativity Appreciative Inquiry A change management approach that focuses on identifying what is working well, analyzing why it is working well, and then doing more of it Communities of Practice A group that shares a common interest and interacts often (face-to-face or online) to learn from each other and advance their work 10 4. METHODOLOGY 4.1. Overall Research Design The implementation research team adopted an exploratory, retrospective, embedded multiple case study research design in which IPs were the main cases and a specific intervention from each IP serves was the ‘case within a case.’ The team employed widely-accepted case study research principles and methods as described in Yin (2018). Below is a diagram summarizing the overall design of the research project (Figure 2). Figure 2. Study Design 4.2. Selection of Cases The study team selected three IPs for this implementation research: ReachHealth, RenewHealth, and TBIHSS (Table 3). They were chosen for several reasons: they expressed interest and willingness to participate in the research; their participation ensured diversity among participating IPs in terms of thematic focus (public health domain or disease area), size, and age of the project; and they had implemented a specific intervention (a ‘case within a case’) that depicted significant change or improvement over time. 4.3. Data Collection The research team used three data collection methods to develop the individual case studies of the three IPs and their specific interventions: 1. Preliminary meetings with IPs to gain an overview of their work, CLA approaches, and AAR practices. 2. A survey questionnaire for key IP representatives, which included specific questions pertaining to various elements of the 7 P’s framework, the AAR Planning Roadmap, and the AAR Effectiveness Evaluation Framework (all discussed below). See Annex A for the questionnaire. 3. A review of documents obtained from the IPs to validate and supplement the findings from preliminary meetings and survey questionnaires. The list of documents can be found in Annex B. 11 Table 3. Selected USAID/Philippines HP Activities, Their Justification, and Specific Interventions Selected USAID HP IP (Case) Brief Description of the IP Justification for Selection Specific Intervention (Case within a Case) ReachHealth Begun in 2018, ReachHealth is a five-year project that aims to strengthen and improve access to critical health services in the Philippines by addressing unmet need for family planning services, teen pregnancy, and maternal and neonatal health. Implemented by RTI International, the project supports the country by identifying and responding to local root causes of poor family planning and maternal and neonatal health outcomes. ReachHealth continues to address gaps and challenges related to family planning and maternal and neonatal health services as well as provide technical assistance on interventions and innovations that contribute towards achieving health targets. One of the ‘older’ USAID/Philippines OH Project projects, ReachHealth demonstrated extensive experience in CLA, particularly in the implementation of AARs and P&R. ReachHealth is also a big project covering many project sites nationwide. Its components range from family planning to COVID-19 response and are primarily focused on delivering interventions to local communities. Hence, this is a good case of a big project with multiple delivery-oriented components. Their monitoring, evaluation, research, learning and adapting (MERLA) team also demonstrated great enthusiasm to participate in this research and shared substantive knowledge during the first meeting. FP Ayuda Express. In response to COVID-19 restrictions, ReachHealth conducted a series of P&R sessions to adapt Enhanced Usapan, a community-based group discussion with immediate on-site service provision of methods suitable to clients’ goals of limiting or spacing childbirths. The intervention was then converted into Ayuda Express, which uses existing popular social media platforms among Bicolanos to communicate FP information and services. 12 Table 3. Selected USAID/Philippines HP Activities, Their Justification, and Specific Interventions (cont.) Selected USAID HP IP (Case) Brief Description of the IP Justification for Selection Specific Intervention (Case within a Case) RenewHealth Launched in 2019, RenewHealth is a five-year project implemented by University Research Co. (URC) that seeks to expand access to quality community-based drug rehabilitation (CBDR) and encourage voluntary drug demand reduction in the Philippines. The project’s theory of change is as follows: If 1) there is increased support for CBDR and help-seeking among persons who use drugs (PWUDs); 2) evidence-based and culturally appropriate treatment tools and interventions are created; and 3) the capacity, policies, systems and resources of communities and government agencies to provide community￾based drug recovery services are enhanced, then PWUDs and their families will access treatment, thus reducing drug dependence in the country. One of the ‘youngest’ USAID/Philippines OH Project projects, RenewHealth is also the first USAID project of its kind in the world that focuses on mental health and substance abuse. This means that there was no foreign model to emulate or existing guidelines to adopt when the project began. Within a short period of time, the project was able to roll out innovative interventions for different parts of the treatment cascade, most especially community-based drug rehabilitation. This was the result of a continuous learning process, including repeated P&R sessions. Motivational Interviewing Training. This training for community facilitators evolved over time through repeated P&R sessions. Future rounds of the training produced significant changes in trainee outcomes (i.e., post-test results). TBIHSS Launched in 2018, TBIHSS is a five-year project implemented by FHI360 that is designed to use a dynamic, strategic, and cutting-edge approach to accelerate the fight against tuberculosis (TB) and to institute health processes and systems to help the Philippines achieve TB elimination targets by 2035. It assists the Department of Health’s National TB Control Program (NTP) in actively identifying, developing, testing, and scaling up innovative technologies and approaches across the TB continuum of care. It also helps maximize programmatic outcomes and strengthen TB health systems. USAID/Philippines OH and CLAimHealth recommended adding a TB-focused project given TB’s national importance. Moreover, as one of the longest-running USAID/Philippines OH projects, there are many lessons from failures and successes that can be learned in relation to the implementation of AARs, especially when it comes to developing innovations. The Philippine Private Sector Diagnostics Consortium (PPSDC) is a group of private laboratories and hospitals that aims to ensure quality TB diagnostics are made available at concessional prices to accredited laboratories and hospitals that will in turn ensure these tests are made available to the public at affordable prices. AAR sessions have helped PPSDC evolve and grow over time. 13 Initially, the study team planned to make key informant interviews, participant observation of upcoming AAR sessions, and review of recordings of previous AAR sessions a part of the research methodology. However, they had to adjust or modify these data collection tools (for instance, through greater use of the survey questionnaire and reliance on existing documents) due to timing limitations brought about by the COVID-19 pandemic, year-end workload borne by the IPs, unavailability of recordings and upcoming events, and other factors. To enhance the validity of the case studies, each IP was also given the chance to review the case study draft and provide additional inputs and corrections if necessary. 4.4. Data Analysis The team analyzed data in two stages, using ‘within-case’ analysis and ‘between-case’ analysis, which are usually done in multiple case study research. ● The ‘within-case’ analysis focused on individual IPs as well as the specific intervention in each IP. This was done by writing an individual case study for each IP using: a general description of an IP’s AAR practices using existing frameworks (see below); documentation of the specific intervention with a summary of events and discussion of lessons learned; and insights related to the implementation outcomes enumerated in the AAR Effectiveness Evaluation Framework. ● The ‘between-case’ analysis examined the three IPs and their specific interventions collectively and comparatively. Using information and insights generated through the ‘within￾case’ analysis, the study team identified common themes across the three IPs and their specific interventions, including AAR characteristics, barriers and challenges encountered, and lessons learned in relation to the implementation outcomes enumerated in the AAR Effectiveness Evaluation Framework. The team summarized common themes as well as shared and unique lessons from all cases in a case study database using Microsoft Excel. 4.5. Analytical Framework To systematically analyze the data, the study team used the following frameworks, which also informed development of the survey questionnaire (see Annex A): The Seven (7) P’s’ Framework (Figure 3; adapted from Graves, 2017) describes the anatomy of the AAR process. The 7 P’s are: people, preparation, process, performance, purpose, period, and place. They resemble the five W’s and one H: who, what, where, when, why, and how. This framework is the basis for developing a general description of each IP’s AAR practices. The AAR Planning Roadmap (Figure 4; WHO, 2019) depicts the different phases of the AAR process: pre-, during-, and post-AAR phase. This framework is essential for tracing the entire journey of an organization and/or intervention on the road to learning and continuous improvement, and if AAR practices contributed to goal achievement. This framework from WHO also resembles USAID’s AAR Technical Guidance (USAID, 2006) and another checklist of AAR good practice (Sexton & McConnan, 2003) that was discussed earlier in the literature review. 14 Figure 3. Seven P’s Framework (Adapted from Graves, 2017) Figure 4. AAR Planning Roadmap (WHO, 2019) 15 The AAR Effectiveness Evaluation Framework (Table 4) is developed by the research team to allow assessment of the IPs and their AAR practices based on widely-used implementation science outcomes. Effectiveness simply pertains to the extent to which an intervention achieved its goal. For an AAR, being effective means that it has enhanced organizational learning and also resulted in improvements to an intervention. In this study, effectiveness is assessed in two dimensions: 1) an ‘internal’ dimension, which pertains to the improvement of organizational learning, including the effective implementation of AARs themselves; and 2) an ‘external' dimension, which pertains to how AAR practices contributed to the improvement of the intervention or the activity as a whole. To assess these two dimensions, the study team selected six widely-used implementation outcomes. They applied internal dimension learning questions to general AAR practices in the IP and used external dimension learning questions to identify the contribution of AAR sessions to implementation improvement in the three specific interventions. Table 4. AAR Effectiveness Evaluation Framework Based on Widely-Used Implementation Outcomes Implementation Outcomes Definition Internal Dimension Learning Question External Dimension Learning Question Fidelity The degree to which an intervention was implemented in relation to the way it was designed in an original protocol, plan, or policy Does the AAR process adhere to the original process designed by the management and/or defined in national or international guidelines/practices? Does the AAR process help the activity/intervention stick to its original implementation design? Adaptation The ability to revise an existing intervention to suit new circumstances Does the AAR process adapt to changing circumstances, for instance shifting to an online platform? Does the AAR process facilitate the activity’s/intervention’s ability to adjust to changing circumstances when necessary? Accountability The process is able to generate a sense of responsibility and ownership among stakeholders, who then feel compelled to execute agreed-upon adaptive actions in a timely manner Does the AAR process have mechanisms to ensure follow-up and dissemination of agreed-upon adaptive actions in a timely manner? Does the AAR process compel or motivate the staff to execute agreed-upon adaptive actions in a timely manner? Integration The process is embedded in everyday operations or organizational structures rather than fragmented and siloed Is the AAR process embedded or tied into the organization’s structures and operations? Does the AAR process aid in ensuring incorporation of the activity/intervention into the health system (i.e., local, provincial, regional, national)? 16 Table 4. AAR Effectiveness Evaluation Framework Based on Widely-Used Implementation Outcomes (cont.) Implementation Outcomes Definition Internal Dimension Learning Question External Dimension Learning Question Inclusivity The process allows relevant stakeholders to participate (rather than excluding them); hence their voices are heard Does the AAR process allow the participation of different personnel within the organization as well as other external stakeholders? Does the AAR process promote the inclusion of diverse stakeholders in implementation of the activity/intervention? Sustainability The process can be executed over time with adequate resources and even when there is a change in leadership or personnel Has the AAR process been continuously performed since it was first introduced and are there indications for its continuous implementation in the future? Does the AAR process help ensure continued implementation and adoption of the activity/intervention even beyond the activity’s timeframe? 4.6. Limitations of the Study While this study was extensively conducted, it still has some limitations. First, because of the major mobility restrictions imposed by the COVID-19 pandemic, the study team had to rely only on information and data it could retrieve by digital means. Second, while the study’s retrospective design generated a wealth of insights about the way IPs conduct AARs, a prospective approach (for instance, by observing actual AAR sessions or conducting intervention research that uses AAR methods) could have also produced rich insight through first-hand observation. Time and resource limitations also forced the research team to examining only three IPs and a specific intervention from each. Finally, throughout the short project period, timing limitations experienced by both the investigators and IPs—including limitations related to COVID-19 and numerous holidays—greatly reduced the possibility of conducting additional interviews and meetings with the IPs. However, the team addressed this limitation through close liaison with the IPs to ensure that information presented in the case studies is as accurate as possible. 4.7. Ethical Considerations This implementation research did not require ethics review approval as it mainly utilized survey questionnaire responses based on facts and not the personal views of select staff members from each IP, as well as documents submitted to the research team. 17 5. RESULTS 5.1. Implementing Partner 1: Improved Health for Underserved Filipinos: Family Planning and Maternal and Neonatal Health Innovations and Capacity Building Platforms (ReachHealth) Overview of AAR Practices To ensure its project goals are achieved, ReachHealth has utilized different CLA tools such as AARs. ReachHealth conducts several CLA activities that embed AAR components like the Mid-Term Technical Review and Program Implementation Review (PIR). It also conducts dedicated P&R sessions, which can be held in groups as well as one-on-one sessions with persons-in-charge of activities. The documentation revealed examples of internal P&R sessions where the senior management team reviewed the status of family planning (FP) in the implementation of a hospital program as well as the postpartum family planning (PPFP) performance of hospitals. These AAR practices allow the team to identify enabling factors, gaps, and challenges in implementation, and agree on priority actions for addressing the gaps determined during the exercise. Before the AAR Event Technical staff. No particular individual officer is assigned as being responsible for preparing and organizing AAR activities. The organizer depends on the activity for which the AAR is being conducted. For example, the MERLA director is in charge of implementing the Mid-Term Technical Review, while the regional managers are responsible for regional PIRs. Budget allocation. There is no explicit line item in the budget that mentions AAR, but ReachHealth has an allocated budget for CLA/MEL activities, which also include AAR/P&R sessions. The shift to online meetings during the COVID-19 pandemic may have also led to reduced spending on CLA and AAR sessions. Venue selection. The selection of the venue usually depends on the location of the intervention being reviewed. However, during the COVID-19 pandemic, these AAR/P&R sessions have been shifted to virtual meetings. ● Before the pandemic, all regional PIRs were held in areas nearest to where most staff members were. The cities of Cebu and Davao hosted the Visayas and Mindanao teams respectively, while the IP's national office in Ortigas hosted the Luzon team. The senior management team traveled to the regions to conduct the PIR. ● During the pandemic, ReachHealth shifted to utilizing online video conferencing platforms (i.e., Zoom) to conduct AARs and other activities as a way of ensuring the safety of participants and team members. Timing/scheduling. Some AARs are conducted as part of routine CLA activities such as the Mid-Term Technical Review and PIRs, which are already pre-scheduled at regular intervals. Meanwhile, AAR/P&R sessions that are dedicated to a specific activity to be reviewed are held right after the event to ensure the attendance of participants. AAR design. There is no predetermined AAR design because it is usually developed by the assigned responsible person for the activity. The assigned team member who developed the first draft AAR design would share it with the senior management team (for a national-level AAR) or with the regional 18 manager (for regional AAR) for review. There are also instances in which participants were tasked to do preliminary activities in preparation for a P&R session. Participants were clustered into small groups and given guide questions for group discussions. These preparatory activities allowed the participants to review their data and to generate learnings that would be presented at the P&R session afterward. Agenda setting. The process followed for preparing the agenda is the same as the process for preparing the design, with the staff member responsible for conducting the activity being the one who prepares the first draft that will later be reviewed by a higher body. Usually, the topics included in the agenda are issues or concerns emerging from a certain implementation period, learnings from routine monitoring data, lessons learned from a period of implementation, root cause analysis, adaptive responses that were identified, and translation of learning into actions. Participant selection. Selecting participants for AAR activities depends on the type of AAR being conducted and the activity being reviewed. For example, for the Mid-Term Technical Review, all project technical staff members were asked to attend and participate in all preliminary activities; PIRs would usually include USAID and other IPs; and implementation reviews of grantee projects that are about to end would include USAID and relevant partners. ● Internal AARs usually include ReachHealth’s expanded senior management team, technical advisors and regional advisors, MERLA national team, regional MERLA staff, regional representatives, and communications staff. Participants whose attendance is required are usually indicated in the AAR design. ● External AARs usually include ReachHealth’s team, representatives from partner government agencies such as the Commission on Population (POPCOM) and the Department of Health Center for Health Development (DOH-CHD), the Provincial Health Office, focal persons from partner health facilities, and family planning coordinators. During the AAR Event Facilitation. It is not standard practice for ReachHealth to prepare trigger questions during an AAR. Discussions and reflections just naturally arise, resulting from questions asked by senior management team members (often addressed to the regional technical staff) or by USAID representatives. In the case of the Mid-Term Technical Review for example, each regional team was even asked to prepare a root cause analysis of their performance. Participation. The responsibility of time management is given to the activity facilitator/moderator. In many instances, the original time allotted for the activity is extended when the team feels that more in-depth discussion is needed to thresh out the issues at-hand. This allows greater participation of more attendees to ensure that diverse insights and reflections are captured. AAR content. P&R sessions usually revolve around reviewing the status of a specific activity within a period of implementation; identifying items that went well and those that need improvement; and sharing and discussing experiences and learnings from both internal and external stakeholders. Participants discuss enabling factors for good practices, gaps and challenges, and opportunities. They are then asked to identify adaptive solutions that will address the challenges that the team encountered. For example, in one P&R session for makeshift teleconsultation installation and implementation, participants were asked to share experiences and identify strengths and challenges encountered in different types of facilities. Participants were given guide questions (“What were the challenges encountered?”; “What are the concerns that we still need to improve on?”; “What are the proposed solutions to 19 address these?”; and “What activities can we undertake to ensure sustainability of the teleconsultation mechanism?”) that allowed them to generate action points for improving access to and utilization of teleconsultation services, and to give recommendations for sustainability measures. Prioritization. The team prioritizes action points in an iterative manner (as opposed to using certain tools or frameworks), but often based on the need, importance, and urgency of addressing an issue. Task assignment. The team identifies who is responsible for a certain task based on the nature of the action to be taken. For example, if the action is about data, then the MERLA officer will be in charge; and if it is an action for a regional team, then the corresponding regional manager will be responsible. Documentation. The last step before closing an AAR session is presenting agreements and recommendations made during the meeting. For future reference, the team documents the highlights of the AAR, discussion and action points made, agreements reached, and next steps agreed upon. The team’s assigned moderator usually prepares the written report and reviewed by a member of the senior management team. A photo of the attendees is usually taken as proof of the activity. Post-AAR evaluation. There is no indication that AAR sessions are also evaluated, either immediately or later. After the AAR Event Follow-up and monitoring. ReachHealth monitors the progress of recommendations and action points in several ways. In the case of agreements resulting from the Mid-Term Technical Review for example, the IP has a learning tracker that is updated every quarter. In most instances, the technical person-in-charge monitors the progress of recommendations and action points. Dissemination of AAR outcomes. ReachHealth disseminates the outcomes of AAR activities to key partners through other external and internal P&R sessions, as well as through regular staff meetings and senior management team meetings. Progress reporting. In reporting back to key partners on progress, the project team compares key indicators from previous quarters to present quarters in accordance with the outcomes of the P&R sessions. 5.2. Specific Intervention 1: ReachHealth’s Family Planning (FP) Ayuda Express Background of the Intervention FP Ayuda Express is a virtual adaptive solution that has used existing popular social media to communicate and disseminate family planning-related information and services in the Bicol Region during the COVID-19 pandemic. Clients can ask questions on topics related to family planning and avail themselves of family planning-related services by sending queries through FP Ayuda Express Chat via Facebook Messenger or by calling the POPCOM Helpline. Chat moderators or call respondents then link these clients to the nearest health facilities and ensure they receive their family planning method of choice. This virtual platform is a result of the joint implementation of PACT (Partnership among Academes and Communities: Teaming up for Improved FP Services in Bicol,) consisting of the Family Health Cluster Division of DOH-CHD Bicol Region, POPCOM Bicol Region, seven academic partners in Albay and Camarines Sur, and USAID’s ReachHealth Project. The intervention is part of ReachHealth’s plan to 20 integrate its Enhanced Usapan program in the community health education curriculum of select academic institutions in the Bicol Region. Originally, select academic institutions with midwifery and nursing courses were to be developed as training centers in conducting Usapan sessions among partner communities. However, the COVID-19 pandemic disrupted implementation of face-to-face demand generation activities and house-to-house tracking of unmet needs of the community. The situation called for innovative solutions to ensure continuous delivery of family planning-related services that can be implemented with limited face-to-face interactions without compromising the health of the community and the team. As part of ReachHealth’s adaptive response, the program shifted to online delivery and launched FP Ayuda Express, a virtual platform that serves as a teleconsultation facility for family planning services in the region. The implementation of FP Ayuda Express was divided into four phases (Figure 5). Phase 1 focused on designing the intervention and engaging partners. In this phase, the concept was presented to the regional CHD, regional POPCOM, and academic institutions. Through a series of consultation meetings with key stakeholders, the team was able to develop the design of FP Ayuda Express and conducted pre￾testing activities to ensure the intervention would respond to the community’s needs. Further, this phase was also the time when the team selected and trained students as chat moderators, created the chat guide and algorithm for chat moderators, conducted an orientation on FP Ayuda Express for students and clinical instructors, and formulated the content plan for its Facebook page. Phase 1 began in May 2020 and was completed in June 2020. Figure 5. FP Ayuda Express’ Implementation Phases 21 Phase 2 comprised the launch and implementation of the intervention. It focused on developing social media cards for the FP Ayuda Express Facebook page and addressing inquiries from clients. Figure 6 shows two examples of social media cards uploaded on Facebook that inform clients what the intervention is and how to access it. Once clients send their inquiry via Facebook Messenger, they are given topics that may be of interest to them. Sample topics include family planning services, family planning methods, and specific information for mothers practicing the lactation amenorrhea method (LAM). Figure 7 shows how the FP Ayuda Express Chat works. Phase 2 was completed in December 2020. Figure 6. Social Media Cards from FP Ayuda Express Figure 7. Sample Chat Inquiry in Facebook Messenger 22 Phase 3 involved monitoring the intervention. To ensure the intervention could meet its objectives, the FP Ayuda Express team documented implementation and implemented a referral and feedback mechanism. Insights from the documentation and feedback collected from this mechanism were discussed during monthly P&R meetings with key partners. This phase ran from July 2020 to April 2021. Finally, Phase 4 of the intervention involved sustaining and scaling-up FP Ayuda Express. In this phase, the team developed a playbook on FP Ayuda Express and distributed it to partners. The team also developed its sustainability plan, conducted an orientation on scaling up the intervention, and identified potential scale-up locations. Moreover, ReachHealth proposed actions for 2022, including expanding and developing the new FP Ayuda Express platform for private lying-in clinics (PLICs) in Central Luzon. After more than a year of implementation, FP Ayuda Express has tallied a total of 2,908 followers on its Facebook page and provided family planning information to over 850,000 individuals online. In 2021 alone, the platform directly reached more than 500,000 people and over 5 million via local radio. After-Action Review as Applied to the Intervention For FP Ayuda Express, AARs are conducted through the intervention’s monthly M&E meetings in the form of a P&R session. These sessions serve as a venue for the ReachHealth team and key stakeholders to discuss updates on project implementation, and raise emerging issues and concerns for the team to address. They also allow the team to share lessons learned during a period of implementation, which is helpful for general improvement of the project. As one member during a session shared, “The regular meeting is very valuable to troubleshoot certain issues and to improve services for the good of the people.” The AAR is usually called to order by a presiding officer of the day and includes the participation of the ReachHealth team, academic partners, representatives from DOH CHD Region V and POPCOM Region V, and student volunteers. Generally, during the implementation of an AAR, the team follows a structure that involves presenting and discussing implementation updates on FP Ayuda Express Chat and Call as well as the latest analytics on FP Ayuda Express Chat, which is vital for improving the service delivery of the intervention. After the presentation and discussion of updates and analytics, each partner will then share experiences, updates, and accomplishments. Discussions on issues and challenges follow, and agreements or recommendations are then identified. The AAR is conducted monthly unless a specific date is agreed upon by the team. As shown in Figure 8, AARs provide an opportunity for the project team to discuss what went well and what needed improvement. The team was able to identify specific gaps through AARs, which led them to revise their chat guide, change their medium of language into Tagalog, and modify the automatic response on Facebook Messenger. The team was also able to develop a client tracking mechanism to record those who have been successfully referred to facilities and received appropriate family planning services, and identify appropriate family planning-related messages to increase engagement and reach a wider audience. 23 Figure 8. AAR Sessions Held During FP Ayuda Express’ Implementation Period Impact of AARS on Implementation Outcomes Fidelity. Initially, the intervention was slated to be implemented in-person, but had to be adapted to online mode because of the COVID-19 pandemic. Thus, it could be said that fidelity to the intervention’s original design significantly decreased. However, while the delivery mode may have changed, the actual purpose and content of the intervention remained intact. The monthly P&R sessions could have helped come up with adaptive actions in response to changing demands and circumstances, and also ensure that the team and the intervention remained faithful to the overall goals that were set at the outset. Adaptation. As expected, regular AAR sessions have helped the implementation team develop adaptive actions in response to changing circumstances and client demands. AARs serve as a platform for deliberating insights gathered from clients (in this case, the users of the online platform). For instance, the team learned that testimonials previous clients posted on the Facebook Page achieved the highest reach and engagement. This resulted in the creation of more content based on stories of successful family planning clients to motivate and encourage behavior change among women with unmet needs for FP. Figure 9 shows an example of content based on a successful client’s testimony that garnered 1,500 likes and an inquiry from a follower. 24 Figure 9. Sample Content Based on Client Testimonials During one of the P&R sessions, the team also discussed feedback that some clients may be intimidated by the use of English on the online platform. This led to a shift to conversational Tagalog as the platform’s primary medium for its automatic responses to ensure that clients can easily understand the questions and are comfortable in communicating with chat moderators. During a P&R session, the team also discussed the fact that the platform constantly received non-family planning-related health topics. To address this concern, the team devised a mechanism that would properly refer clients to appropriate agencies and organizations. The monthly P&R session also provided an opportunity for the team to revisit the online platform’s algorithm. However, there have been ongoing discussions in modifying the algorithm to accommodate other relevant audiences such as male members of households, as well as other topics related to adolescence and gender-based violence. Documentation of P&R sessions indicate that the platform is continuously evolving in response to expanding client needs. Accountability. As part of the IP’s P&R system, the team also developed an FP Ayuda Express responsibility matrix to keep track of interventions throughout implementation. It includes the person or partner organization that is responsible for a particular task and the target date for completion. The team utilizes the matrix to determine the status of each activity assigned to a specific team member or partner organization. Subsequent P&R sessions are used to revisit previous agreements and check whether the team already implemented tasks and incorporated recommendations. One occasion where a P&R session helped enhance accountability occurred when the team found out that there were clients who prefer not to disclose personal information in the chat; that there were unclear queries; and that there were reports of delayed responses from dispatchers. To address these findings, the team assigned specific team members to fix the identified issues, which later led to enhanced efficiency and confidentiality on the online platform. Integration. The monthly P&R sessions included internal staff members as well as representatives from the DOH-CHD, who are familiar with the state of the local health system. Hence, these external stakeholders can help the implementing team identify ways to address challenges faced by clients in 25 accessing family planning services in rural health units and barangay health centers. Moreover, their participation allows stakeholders to explore ways to incorporate the intervention into the health system or the school system, rather than implementing it only on an ad hoc basis. Inclusivity. The monthly updates facilitated knowledge sharing among PACT members, allowing them to identify opportunities for engaging more potential family planning clients. PACT members included representatives from partner government agencies such as POPCOM Region V and DOH CHD Bicol Region, as well as responsible parenthood and family planning (RPFP) experts, clinical instructors from seven partner academic institutions in Bicol Region, chat moderators, and call responders. The team regularly conducted P&R sessions with diverse PACT members to garner their inputs and feedback, as well as to ensure their commitment to executing specific action points. For instance, specific feedback from chat moderators and clinical instructors allowed the team to revisit its algorithm for addressing chat inquiries and revise the chat guide to properly respond to these inquiries. Sustainability. The involvement of academic partners, regional DOH and POPCOM, and students during monthly P&R sessions is important to promote inclusivity and to also generate a sense of ownership, which is important for sustaining and supporting the intervention beyond the project implementation period. Since these partners were actively included in both the implementation of the intervention and accompanying AAR sessions, there is a higher chance they will continue implementing the intervention, track progress, and adapt to changing circumstances, even with little support from USAID and ReachHealth. Lessons learned and best practices documented in the AARs during the implementation period can also be useful for future implementers as the FP Ayuda Express continues to expand. 5.3. Implementing Partner 2: RenewHealth - Expanding Access to Community-based Drug Rehabilitation Overview of AAR Practices AARs are part of RenewHealth’s processes for learning and developing collaborative solutions. They allow the project to realign its interventions (e.g., trainings) with changing circumstances and improve rapidly. RenewHealth conducts AARs in the form of post-session debriefings. Such AAR sessions provide an opportunity for staff members, partners, and communities to critically assess whether interventions and innovations are being implemented properly and impacting the project’s performance indicators (for instance, in relation to community-based drug rehabilitation). In general, the AAR process in RenewHealth comprises the following components: a post-activity evaluation with participants using a questionnaire; discussion of the results of the post-activity evaluation by internal staff; a post-session debriefing with internal staff, allowing them to share general experiences with an intervention (i.e., what went well and what could be improved) ; and progress sharing with external stakeholders, including AAR outcomes. Before the AAR Event Technical staff. Under the guidance of the chief of party (COP), the learning and development (L&D) officer manages RenewHealth’s training programs. The L&D officer is also responsible for conducting AAR activities after training activities, monitoring their progress, and sharing outcomes with the internal staff and external stakeholders. Meanwhile, since the M&E officer normally administers the online survey, they are also responsible for presenting the results and leading the discussion to analyze them. Budget allocation. There is a budget allocated for CLA activities, which include workshops, field focus group discussions (FGDs), and conferences. While AAR activities are not mentioned explicitly, it can be assumed that they are also covered by the CLA budget. Because most face-to-face activities such as CLA 26 activities were adjusted to online due to the COVID-19 pandemic, some of the budgets are also allocated to online technologies such as Zoom accounts, which are also used for AAR sessions. Venue selection. The COVID-19 pandemic has had a substantial impact on AAR sessions. All AAR activities that were normally held in person have been switched to virtual meetings. ● Before the pandemic, AARs typically occurred shortly after a session or event had ended. Such was done for everyone's convenience and to ensure that feedback was obtained as soon as possible before staff members forget the specific details from the intervention that was being reviewed. ● During the pandemic, all AAR activities have been conducted using Zoom, an online platform that is widely used among organizations. The AAR activities generally take place during an allotted time, usually lasting for an hour. A copy of the recording is kept for future reference and is also shared internally. Timing/scheduling. AARs are carried out immediately after an intervention has finished. It is essential to obtain feedback and insights from internal staff members while the memory of the activity is still fresh, allowing participants to recall what happened. AAR design. Although there is no predetermined AAR design and written guidelines have not been developed, there is a generic structure that RenewHealth follows during post-session debriefings. The first part is a discussion of the post-activity evaluation result. Staff members are encouraged to comment on issues and challenges collected from the evaluation. This is followed by a discussion on what went well and what can be improved. Each staff member is asked for insights. The post-session debriefing wraps up with the identification of next steps (e.g., action points for improving the next activity), all of which are collectively agreed upon by the team. Agenda setting. Since an AAR session is conducted right after a particular training event, there is no specific agenda prepared beforehand. However, a post-session debriefing usually follows the generic structure earlier described. It is a free-flowing discussion that mainly tackles results of the post-activity evaluation that is regularly collected by the end of an intervention. During the post-session debriefing, the team assesses the intervention in terms of challenges, issues, and improvements. Everyone is encouraged to give their insights during the discussion. Participant selection. As described above, the AAR sessions are primarily internal staff meetings where results of the post-activity evaluation survey that was administered with training participants are deliberated. The only participation of external stakeholders such as training participants is through the survey, which is a one-way feedback process. Hence, there is no opportunity for a more interactive discussion or bidirectional exchange between project staff members and training participants. During the AAR Event Facilitation. For the post-session debrief, internal staff members are normally asked a set of questions, including what went well, what didn't go well, and what else could be done better. Post-activity evaluation results are complemented with a presentation of data relevant to the intervention such as the number of invited participants who attended a training, and the number of drop-outs or those who were unable to complete the training. Reasons for such occurrences are explored during the discussion. 27 Participation. Post-session debriefings are conducted immediately after an intervention, while all staff members engaged in the intervention are still present and can participate in the discussions. A facilitator usually asks the participants to voluntarily share their insights on the evaluation results. To stimulate lively discussions, the facilitator utilizes follow-up questions and paraphrasing techniques. Sometimes, the facilitator calls on participants who they believe have not participated yet in the discussion. AAR content. Generally, the results of the post-intervention evaluation survey administered to training session participants indicate what is working and what is not. Hence, these responses provide the basis for internal team discussions on what needs to be changed and improved for subsequent training interventions. Prioritization. While there is no straightforward approach to prioritizing action points, suggestions from the participants are given a lot of attention and the utmost importance. For instance, the team was quick to adjust the session duration with speakers from one hour to two hours based on feedback from participants in motivational interviewing training. Meanwhile, other suggestions may require a longer time to be implemented such as requests for the development of materials using the local language. The time required for suggestions to materialize is also taken into account when prioritizing action points. Task assignment. Particular tasks that emerge from the post-session debriefing are assigned to specific team members who possess the relevant skills and/or are responsible for the relevant task area (e.g., refining information, education, and communications or IEC materials, or addressing technical issues related to the CBDR online portal). Documentation. The AAR session discussions are included in the minutes of the meetings. The lessons learned and highlights of the discussion are written in separate sections for easy reference. The draft is shared with the members of the team for refinement, recontextualization, and clarification. Once everyone’s inputs are garnered, the document is then finalized. Post-AAR evaluation. There is no indication that AAR sessions are also evaluated, either immediately or later. After the AAR Event Follow-up and monitoring. The L&D officer is responsible for monitoring progress in terms of the execution of action points that were agreed upon during the post-session debriefing. Dissemination of AAR outcomes. While agreements made during the AAR session are shared with the internal staff, it is not a practice to share the results of an AAR activity with external stakeholders. Instead, the changes to the intervention (e.g., training) that were decided upon are automatically implemented and observed by future participants. Documentation of AAR outcomes are only kept internally as reference for further improvement of subsequent activities. Reporting of progress. The L&D officer is primarily responsible for sharing progress with the team on implementing post-session debriefing action points. Occasionally, progress reports are also shared with external partners. These reports are normally prepared by the program development officer and approved by the L&D officer. For instance, the project team updated the Bureau of Jail Management and Penology (BJMP) regarding implementation of a facilitators training. The report included discrete information such as a list of participants who participated in the training as well as a list of those who did not complete the training. It also contained a discussion of the limitations of online simulation as well as feedback from trained coaches regarding what went well and what needed to be improved. 28 5.4. Specific Intervention 2: RenewHealth’s Motivational Interviewing Training Background of the Intervention RenewHealth spearheaded the Motivational Interviewing (MI) Training project to assist LGUs in providing evidence-based treatment to people who use drugs (PWUD). MI is a counseling skill designed to evoke change talk among clients. This skill is vital for community facilitators implementing community-based drug rehabilitation (CBDR) interventions as they interact with their clients/PWUDs. Facilitators implementing the intervention must undergo training/simulation, where they need to demonstrate basic competence in MI and facilitation skills. Currently, there are no internationally-accredited MI trainers in the Philippines. This project aims to build a cadre of MI coaches and facilitators. RenewHealth developed a learning management system to deliver the MI course. The course consists of four modules: 1. Introduction to Motivational Interviewing 2. Motivational Interviewing Fundamental Skills: Questioning and Reflective Listening 3. Motivational Interviewing Fundamental Skills: Reflective Listening & Rolling with Resistance 4. Motivational Interviewing Fundamental Skills: Evoking Change Talk & Readiness for Change MI was initially embedded in the five-day training program, named KKDK-CBDR (Katatagan Kalusugan at Damayan ng Komunidad). Training was delivered face-to-face before the COVID-19 pandemic, which subsequently restricted travel across sites and limited the size of meetings and trainings. These restrictions made it difficult for project team members to go to sites to pilot-test, train, and observe community facilitators. As a result, MI training is currently delivered via the CBDR Learning Portal, developed by RenewHealth. The program utilized a blended design with four hours of asynchronous training and four hours of synchronous training. The asynchronous modules are designed to deliver knowledge, whereas the synchronous training is used to enable participants to practice skills in MI. The roll-out strategy is depicted in Figure 10. Figure 10. General Phases of KKDK-CBDR Training, Including MI MI training has four levels of evaluation. This ensures the quality of training and allows for gauging if there is a need to incorporate adaptive changes for the next training . Levels 1, 2, and 3 are conducted immediately after each training. ● Level 1: Reactions - This level is composed of quantitative and qualitative components. For the quantitative part, participants are asked to use a Likert scale to rate the training’s usefulness, ease of learning, visual impact, and ease of use. On the other hand, the qualitative component collects opinions from the participants on which content is most useful and asks if they have any suggestions for improvement. ● Level 2: Learning - This level gauges the acquisition of knowledge by comparing pre- and post￾tests taken by participants. 29 ● Level 3: Behavior - This level checks the confidence level of the participants based on a self￾assessment before and after the training. ● Level 4: Outcomes - This level is conducted three months after the training to check participants’ retention of knowledge and skills, as well as if they have effectively applied what they learned in the real world. After-Action Review as Applied to the Intervention The AAR conducted for MI training is a post-session debriefing. After the end of the training, an online survey is immediately sent to the participants. The survey essentially covers levels 1, 2, and 3 of the MI evaluation. It is an anonymous survey to encourage everyone to participate and provide honest feedback, which is later used for the post-session debriefing with internal staff. The post-session debriefing usually takes place right after the end of training. It generally follows the structure shown in Figure 11. The M&E officer leads these sessions. Attendance at the training is first discussed. Specific indicators are presented, including the number of invitations sent, the number of people who actually participated, the drop-out rate, and the completion rate. This is followed by a review of results from the post-evaluation activity. Staff members are encouraged to provide comments and suggestions on how to address the issues and challenges that were identified from the survey results. Lastly, a general discussion on the activity is triggered by a question: “What went well and what can be improved?” All the staff attendees are encouraged to give their insights on what could have been done better and how this can be applied to the next training session. Figure 11. Steps in the AAR Process for MI Training Implementation of AARs in the form of post-session debriefings that utilize results from post-evaluation surveys contributed to the gradual improvement of MI Training. Figure 12 shows some milestones in the training events and their accompanying AARs, as well as the changes instituted in subsequent trainings. For instance, in batches 1 and 2, the need to increase interaction time with the lecturer was repeatedly mentioned in survey feedback. This feedback was no longer observed in batches 3, 4, and 5, which may indicate that the issue had been addressed by then. 30 Figure 12. AAR Sessions Held During the Implementation of MI Training While several issues were resolved through repeated AARs, a few concerns also remained and were repeatedly raised in all batches through the survey. Examples of these concerns included the need to provide more realistic scenarios and the need for a Tagalog version of the video material to make it easier to understand. These issues may have not been successfully addressed right away given that there is only a one-month period between training batches. Revising the learning materials comprises a complex process of drafting and testing before fully integrating a new version into the CBDR online portal for training use. This requires time, effort, and resources Impact of AARs on Implementation Outcomes Fidelity. Initially, the intervention was slated to be implemented in-person, but because of the COVID-19 pandemic, it had to be adapted to online mode. Thus, it can be said that that fidelity may have significantly decreased in relation to the intervention’s original pre-pandemic design. However, while delivery may have changed, the actual purpose and content of the intervention remained unchanged. The online version is still patterned after the original KKDK-CBDR training. This version was gradually improved to suit the demands of training participants through repeated post-session debriefings on the results of post-evaluation surveys. Nonetheless, the post-session debriefings also ensured that the team and the intervention remained faithful to the intervention’s initial overall goals. Moreover, despite unpredictable circumstances, the training series proceeded as scheduled. Adaptation. The shift from in-person to online mode presented new challenges that were unfamiliar to training participants. Hence, AARs are essential to ensuring the intervention adapts to changing circumstances. One major modification that was made based on the results of the AAR/post-session debriefing pertains to the length of the practice session. The session was stretched to two hours from its original one-hour length. After the practice session, an open forum was also added to ask participants about issues related to the online portal and immediately troubleshoot their concerns. Furthermore, orientation on the use of and access to the portal was also lengthened to allow a detailed walk-through with the participants. 31 Accountability. The post-session debriefings also motivate the team members to be responsive to the demands and needs of their clients (i.e., training participants). Responsiveness is a hallmark of accountability and a reflection of a commitment to quality. For example, the team heeded the participants’ suggestion to extend the practice session from one to two hours. AAR sessions and participant surveying aimed at improving training processes also likely had a positive effect on trainers’ motivation to recognize and pursue ways to improve their personal effectiveness. Notably, facilitators consistently obtained scores of 4.5 out of 5 in all training batches. Integration. Community facilitators are the primary health system agents for delivering CBDR services to the community. Through repeated AAR sessions and evaluations by participants, the MI training has been continuously improved over time. The hope is that better training will produce better community facilitators, who are embedded in the local health system. However, since AARs in this case do not include participants from government partners, there may be little discussion conducted regarding how to eventually integrate the MI training program into the local government’s CBDR programs. Inclusivity. After each training, the team holds post-training huddles or debriefings to discuss what went well and what needed improvement. While the feedback from participants is utilized during the post￾session debriefings, they are not included in the actual discussions by the team. Other local government partners are also not included. Sustainability. Similar to integration, the instructional design that has been improved over time through repeated AAR sessions will hopefully produce well-equipped community facilitators who will continue to provide CBDR services in the community. However, the absence of local government partners in the AAR sessions may limit the potential for the training to be sustained and owned by them in the long term. It is important to note though that RenewHealth has more than two years remaining; thus discussions about long-term ownership and sustainability can still be initiated. 5.5. Implementing Partner 3: TB Innovations and Health Systems Strengthening (TBIHSS) Project Overview of AAR Practices AAR processes in TBIHSS are embedded in the entire project life cycle. Two types of AARs are practiced in TBIHSS: topic or theme-based P&R sessions, and periodic P&R sessions that are held as part of routine internal team meetings. The topic- or theme-based AARs are triggered, for instance, by policies that are being introduced by the government. According to TBIHSS staff members, the passage of the Universal Health Care (UHC) Law in 2019 was a major trigger event, pushing the project to organize P&R sessions. This is important because the IP must align its agenda and activities with the government's emerging priorities. On the other hand, internal review meetings trigger the implementation of P&R sessions to assess progress, achievements, and challenges. In this case, the purpose of an AAR is to brainstorm strategies to further strengthen progress already achieved while planning and executing course correction strategies to manage emerging challenges. Before the AAR Event Technical staff. The strategic information (SI) team is responsible for conducting AAR activities. It leads the coordination of TBIHSS' P&R sessions, for both internal and external events. The team consists of representatives from three units: the M&E unit; the operations research unit; and the knowledge management unit. There is no single person responsible for conceptualizing the P&R sessions. Instead, team members collaboratively work to plan, manage, and coordinate the sessions. 32 Budget allocation. AAR activities are budgeted during the annual work planning process. The budget includes a contingency fund for unplanned events as well as all pre-planned AAR events for the financial year. The budget includes an allocation for software and other technologies for enabling remote meetings, publications for documenting P&R sessions, field trips for project monitoring, and fringe provisions such as lunch for in-person P&R sessions. Venue selection. The COVID-19 pandemic significantly affected venue selection. Generally, all P&R sessions that are usually held in-person needed to be changed to virtual meetings beginning in early 2020. ● Before the pandemic, the paramount consideration in choosing a venue for AAR events was finding a location that could guarantee the attendance and safety of participants. Sometimes, the participants were also consulted on their preference for the venue. There were times when AAR activities were held in the TBIHSS office, DOH central and regional offices, and rented facilities where interventions were being conducted. ● So as to continue conducting AARs during the pandemic, session organizers made two types of adaptions that address mobility restrictions: (1) investing in digital tools such as Zoom and online polling application software, and (2) orienting and building the capacity of partners and staff members on using remote technologies. Timing/scheduling. The scheduling of AAR activities adopts a more iterative approach. Rather than a “one-size-fits-all” approach, AAR scheduling is based on the stage of activities under review. This includes considering whether an immediate review is pertinent (i.e., if there is an issue that needs to be urgently addressed) and the level of invited participants (e.g., including busy senior officials may make scheduling more difficult). AAR design. A predetermined template is used for TBIHSS’s monthly technical team review-cum-planning meeting, an internal meeting that has a strong AAR component. Other teams in the organization participate in these meetings. Having such a template allows structured and standardized documentation of the agenda and the proceedings. In the meeting minutes they reviewed, the CLAimHealth study team found that TBIHSS technical team leads are expected to provide updates, discuss issues and challenges, garner comments and suggestions, and arrive at decision points. TBIHSS uses another template for an AAR involving external partners. It consists of six main components: rationale background, learning objectives, methodology, expected participants, expected output, and session flow. The learning objective directs the overall discussion of the meeting. It contains key learning questions and sub-questions that must be answered during the meeting. Documentation is crucial for following up and monitoring the action points that were agreed upon during the review and planning meetings. Agenda setting. AAR session agendas are developed in consultation with different stakeholders such as internal project staff members, and local and national government partners. The overall aim of these sessions is to share knowledge and strategies on how to strengthen linkages in TB programs at various levels (i.e., national, regional, local, community). Some preparatory work is initiated to inform the AAR session’s agenda. For instance, an online assessment form hosted on Google Forms or Survey Monkey is sent to participants of a recently-concluded training. The questions are both quantitative and qualitative. The qualitative portion elicits feedback on what the participants liked the most about the training and suggestions for further improvement. 33 Participant selection ● For internal AAR sessions, invitees are usually the TBIHSS project staff, particularly relevant technical teams. ● For external AAR sessions, the following partners and counterparts are usually invited: Department of Health, particularly the National Tuberculosis Program, as well as other relevant bureaus such as the Bureau of Local Health Systems and Development (BLHSD), the Disease Prevention and Control Bureau (DPCB), the Epidemiology Bureau (EB), the Health Facility Development Bureau (HFDB), the Health Policy Development and Planning Bureau (HPDPB), and the Health Promotion Bureau (HPB); the Philippine Health Insurance Corporation (PhilHealth); DOH’s Regional Offices/Centers for Health Development (CHD); Provincial Health Offices (PHO); USAID/Philippines’ Office of Health; implementing partners of other USAID/Philippines OH activities; and private sector partners such as professional medical societies and the Philippine Coalition Against Tuberculosis (PhilCAT); manufacturer of TB diagnostics (e.g., Cepheid) During the AAR Event Facilitation. Two approaches are often used to stimulate discussion during the AAR activity: open-ended trigger questions and presentation of quantitative data. These approaches ensure an in-depth discussion for exploring the root cause of a problem during the review. Open-ended trigger questions are prepared and used during the plenary sessions to encourage participants to open up and share their thoughts and opinions. For more fact-based deeper discussions, quantitative data is used to trigger discussions and reflection on current performance. Such data-based information encourages participants to dig deeper into the topics. During the COVID-19 pandemic, AAR sessions needed to shift to virtual meetings. Hence, participants were actively engaged using online polls for open-ended discussions during plenary sessions. Participation. Part of the AAR’s organization plan is to assign a timekeeper so that the session adheres to schedule. It was noted that timekeeping during in-person events was relatively better controlled than virtual sessions. Proper time allocation ensures that all participants can actively participate to express their thoughts and opinions, and obtain valuable information from others. In addition and as a backup plan, online platforms such as a Google Sheets are opened during the plenary session for individuals who prefer to offer their opinions in writing. Such virtual technologies also help the AAR session remain on￾schedule. AAR content. Conducting AARs is part of TBIHSS’s commitment to building a culture of learning within the organization. Central to this is reviewing lessons learned from both successes achieved and challenges encountered. Hence, implementation failures and the factors behind them are continuously assessed in AAR sessions. Lessons learned from adversities as well as their underlying reasons were carefully analyzed based on solid evidence and rather than mere anecdotes. This approach helps inform the development of adaptive measures to avoid adverse consequences and improve project performance. Sustained documentation of AAR sessions, including main discussion points and key decisions, has led to the creation of a repository of good practices and lessons learned that also feeds into subsequent AAR activities. Prioritization. The Eisenhower Decision Matrix (Figure 13) is used to prioritize the long list of action points that come from an AAR session by categorizing them according to their urgency and importance. 34 Figure 13. Eisenhower Decision Matrix Task assignment. Minutes of external AAR sessions did not clearly indicate who was responsible for action points and next steps. Meanwhile, for internal AAR sessions, tasks were assigned according to the individual or team's specialty. For instance, tasks related to community engagement and demand generation of TB testing from one AAR were assigned to the demand generation team. Documentation. As part of the AAR organization plan, a dedicated person is pre-assigned to document the salient points from the session. To support this process, informed consent is obtained at the start of the event to audio-record the sessions. Post-AAR evaluation. There is no indication that AAR sessions are also evaluated, either immediately or later. After the AAR Event Follow-up and monitoring. Individual teams regularly monitoring specific action points. Meanwhile, overall progress is monitored and discussed during subsequent meetings. Additionally, the M&E team monitors the quantitative aspect of implementation progress, comparing data about past performance and current status. Dissemination of AAR outcomes. A few days after the event and AAR session, documentation is shared with participants through e-mail or web link. When AARs are published, a printed copy of documents such as case studies or communication materials is delivered to the participants. Reporting of progress. As a standard practice, progress in implementing action points from AAR sessions is routinely reported back to stakeholders. The progress report is initially shared for comments from stakeholders in advance of the next activity and its accompanying AAR session. This practice makes them aware of the project's status and helps them prepare pertinent points they would like to raise during a future governing council meeting. 35 5.6. Specific Intervention 3: TBIHSS’s Philippine Private Sector Diagnostics Consortium (PPSDC) Background of the Intervention In 2018, the Philippines made a commitment to find and treat around 2.5 million Filipinos with TB by 2022. Currently, one in three Filipinos with TB prefers to seek testing and care from the private sector. However, the costs of private sector TB services are catastrophic for Filipino households with an average income. GeneXpert testing, one of the globally recommended primary diagnostic tests for TB, costs an average of $150 when performed by private facilities. To address this gap in health services, TBIHSS spearheaded the creation of the Philippine Private Sector Diagnostics Consortium (PPSDC) in 2019, a platform that allows TB testing to be made available to the general public at a cost that is 70 percent below commercial pricing. Convened by the Philippine Coalition Against Tuberculosis (PhilCAT), the PPSDC comprises various private hospitals and laboratories, suppliers, distributors, and procurement agencies in the Philippines. The aim of PPSDC is to ensure quality TB diagnostics are made available to the general public at affordable prices, particularly at concessional pricing through pooled procurement with accredited laboratories and hospitals. To ensure PPSDC’s success of in achieving its objectives, the Governing Council was established to oversee the consortium. Formalized through an MOU, PPSDC has so far invited Cepheid and Macare to join the Consortium for GeneXpert systems and cartridges, and established Philippine Pharma Procurement, Inc. (PPPI) as the default procurement agency. It also has an MOU with the DOH and the Research Institute of Tropical Medicine (RITM), which oversees the National TB Reference Laboratory (NTRL). This has served to formalize recognition of PPSDC members as members of the country’s TB laboratory network, and as members of their respective local laboratory networks at the regional and provincial/city levels. After-Action Review as Applied to the Intervention The Governing Council conducts a periodic meeting, which TBIHSS considers to be the PPSDC’s regular AAR activity. The schedule of the Governing Council’s meeting (i.e., AAR session) is set regularly on a quarterly basis. It is attended by the PPSDC’s members, which include both the private and public sectors. Since the time of PPSDC’s inception, eight council meetings have been conducted. These meetings provide an opportunity for members to look at current project activities and critically assess them to identify key challenges and gaps. The meetings serve as a platform for each member and partner to contribute evidence and provide criticism for further improvements. Meetings follow a general structure that involves the following: checking attendance; presenting the agenda; giving a brief recap on the previous council meeting; giving updates on membership, procurement, and the number of tests conducted; discussing other concerns; and wrapping up by presenting the next steps that were identified during the meeting. As the meeting progresses, questions and suggestions are encouraged from the participants to stimulate active discussion. 36 Figure 14 shows PPSDC’s implementation period, which includes Governing Council meetings conducted since the beginning of the project. The findings from Governing Council meetings were helpful, particularly for increasing the number of members. During the second council meeting, it was noted that a strategy for membership expansion needs to be developed. Immediate actions agreed upon during these meetings included setting the criteria for targeted membership expansion, and mapping the level 2 and 3 hospitals and laboratories in Regions 3 and 4 to be recruited. Over time, new members have been added at every council meeting. By the end of 2021, PPSDC membership had increased to 21 private hospitals and laboratories. Fourteen (14) PPSDC members participated in nine pooled procurements for 18 GeneXpert machines and 16,450 cartridges. These initiatives helped lower cost in relation to the regular commercial price, ultimately making TB testing more affordable to indigent patients in the country. Within this framework, PPSDC members participate in training on GeneXpert systems, biosafety, and recording reporting procedures. They also participate in post-training competency assessments as well as data quality assessment activities organized by the NTRL in coordination with the regions. To date, 136 personnel from member labs have been trained. Of these, 35 staff members from seven different laboratories achieved a 100 percent score on the competency evaluation. One data quality check (DQC) event was organized to identify challenges, best practices, and thematic areas for improvement. Impact of AARs on Implementation Outcomes Fidelity. No significant deviation from PPSDC’s original vision and design was observed over time. Hence, it can be said that there is a high degree of fidelity in this intervention. The changes that occurred in the Consortium could be viewed as enrichment to the original design rather than a major pivot since the actions that were agreed upon during Governing Council meetings were geared towards PPSDC’s sustainability and expansion. Adaptation. Since the PPSDC is still in its infancy, its internal team and external stakeholders agree that organizational and process improvements will be required, especially to achieve the Consortium’s goal of expanding its membership. Documentation from Governing Council meetings indicates numerous adjustments that were agreed upon in order to recruit more partners and adapt to changing circumstances. This included, for example, shifting GeneXpert training from in-person to online delivery due to COVID-19 restrictions. Accountability. During the third Governing Council meeting, the roles and responsibilities of PPSDC members were clearly defined and delineated, which has helped enhance accountability within the Consortium. The attendance of Consortium members is also checked at every meeting because their presence and participation are key expectations. Additionally, members actively participate in discussions and give their insights, which demonstrates their commitment to exercising their roles. All action points are carefully documented, indicating the individuals and organizations that are responsible for carrying out agreed-upon tasks, which is vital for accountability building. Integration. The initiative to form the PPSDC began as part of TBIHSS’ technical assistance to increase support to the private sector in addressing TB in the country. One of the main goals is to ensure the initiative operates in conjunction with the National TB Program’s goals. The PPSDC’s operations are anchored in the Philippine Strategic TB Elimination Plan (PhilSTEP) as well as the Public-Private Mix (PPM) national action plans. Because of this, reaching out to the local health system is easier during the implementation process. The MOU between PhilCAT and DOH to formally introduce PPSDC members to the different regions is another key agreement that aids in enhancing integration. 37 Figure 14. AAR Sessions Held During PPSDC’s Implementation Period 38 Inclusivity. Since the project's inception, PPSDC has been constituted by hospitals and laboratories from the public and private sector. Individually, they have no power to negotiate directly with suppliers and distributors to lower commercial pricing for GeneXpert machines and cartridges. During the third Governing Council meeting, a strategy was developed for membership expansion in level 2 and 3 hospitals and laboratories in Regions 3 and 4A to join PPSDC. Sustainability. The council meetings always take into account PPSDC’s sustainability beyond the TBIHSS project. One aspect that has been considered is directly involving members in operations while the project is ongoing. The four committees created as part of the Governing Council included representatives of different laboratories and hospitals. In this manner, members develop a sense of ownership towards PPSDC during project implementation, increasing the chance of sustainability in the long run. By design, 2 percent of each cartridge’s purchase price is used to subsidize the Consortium’s operational budget, thus serving as seed money to sustain its functions beyond project support. 39 6. DISCUSSION USAID has promoted the CLA approach by developing a design for conducting AARs through the Agency’s Innovation Lab. The design is available to all IPs (USAID, 2006) but is not a strict requirement. IPs are allowed to design their own approach or use other AAR designs such as the ones their respective organizations have developed. For instance, TBIHSS is guided by the AAR designs that its implementer, University Research Company (URC), developed. Generally, AARs are conducted in different forms because IPs do not abide by a standard definition of an AAR. In fact, not all IPs are familiar with the term ‘AAR.’ However, when AARs were described to the IPs in this study, they recognized them as a familiar process that their organization conducts in various ways. For example, TBIHSS and ReachHealth conduct AARs in the form of P&R sessions, while RenewHealth utilizes post-session debriefings after each activity. AAR styles also differ across the three IPs. Although TBIHSS and ReachHealth conduct P&R sessions, each has its own way of executing them. TBIHSS conducts P&R sessions that can either be topic- or theme-based, and are either triggered by new government policies or are periodic (i.e., prompted by routine internal meetings). Meanwhile, ReachHealth P&R sessions come in the form of Mid-Term Technical Reviews and Program Implementation Reviews; monitoring and evaluation meetings for specific interventions (such as FP Ayuda Express); or one-on-one review sessions with the person-in-charge of interventions. Moreover, RenewHealth AARs gather feedback from training participants/attendees, which the IP’s team subsequently analyzes through post-session debriefings. Despite differences in terminology and methodology, IPs’ AARs share the common purpose of strengthening and continuously improving the implementation of activities and interventions. AARs involve different partners in identifying and assessing enabling factors as well as barriers, gaps, and challenges. They then allow the IPs to plan and execute course correction strategies to manage emerging challenges. The repeated AARs also eventually help achieve a project’s ultimate goal through continuous process improvements. Even if there are different interpretations of what an AAR is, the different interventions reported by and observed in the three IPs remain faithful to the definition and characteristics of AARs, as described in existing toolkits. 6.1. The 7 P’s of AARs in the IPs Period. The timing of AARs across the three IPs differs from immediate to routine. On some occasions such as a training event, an AAR is conducted immediately afterwards because all key participants are present, and can provide feedback and insights while their memories are still fresh. Meanwhile, a success factor for more routine AAR sessions is the availability of key partner stakeholders, especially those who hold key positions in organizations and government agencies. Place. Before the pandemic, organizers selected AAR venues based on convenience to the participants to better guarantee their attendance. TBIHSS often considered the preference of the participants, while ReachHealth and RenewHealth often held their AARs in the venue where the intervention being reviewed had been held. However, venue selection was significantly affected by the emergence of the COVID-19 pandemic in early 2020. Generally, all AAR sessions that were usually held in-person needed to be moved to virtual meetings, utilizing video conferencing platforms such as Zoom and Google Meet to ensure the safety of participants and team members. People. Responsibility for the planning and implementation of AARs varies across the three IPs. RenewHealth specifically assigns this task to the learning and development officer. In TBIHSS, the strategic information team is responsible for preparing all internal and external AAR activities. The 40 members of the team collaboratively plan, manage, and coordinate the sessions. For ReachHealth, the responsibility for conducting the AAR goes to the leader of the intervention being reviewed. When it comes to participants or attendees, AARs are not limited to the internal staff. A diverse group of people are invited, including representatives from various partner stakeholder organizations who are actively engaged in the activity or intervention being reviewed. However, the extent of their involvement varies across the IPs. In RenewHealth, participants in an intervention like a training event were asked to answer a survey, which the internal team later analyzed during their AAR session. When conducting a PIR with an AAR component, ReachHealth only invited those who played a key role in implementing the activity. Purpose. In general, AAR activities in all three IPs tend to focus on identifying challenges encountered during a certain implementation period; identifying potential solutions to known problems; and distilling lessons learned. The three IPs have slight differences in terms of developing specific agenda items. TBIHSS consults with different stakeholders about specific items that need to be added to the agenda prior to the AAR session. In RenewHealth, while there may be a general topic to be discussed, the actual discussion tends to be more free-flowing and unanticipated topics may also emerge. ReachHealth, on the other hand, observes a more structured hierarchical process. Usually, officers (e.g., the MERLA officer and provincial technical officer) will prepare the first draft of the agenda, which will be later reviewed by a higher body (e.g., a member of the senior management team or a regional officer). Preparation. The budget for AAR activities is taken from allocated funding for approved CLA/MEL activities. TBIHSS includes a budget for its AAR activities during the annual work planning process, as well as contingency funds for unplanned events. RenewHealth and ReachHealth have no specific budget line allocation for AAR activities. RenewHealth did not usually need a dedicated budget for AARs because they conducted them immediately after events. During the COVID-19 pandemic, all budgets for CLA activities were reduced. Moreover, budgets for CLA activities including AAR were focused on procuring software and other technologies needed for remote meetings. IPs do not have written guidelines on how to conduct AARs. TBIHSS has different templates for both internal and external AARs, which are reflected in the documentation of past sessions. While RenewHealth and ReachHealth don’t use predetermined designs for their AARs, these IPs tailor designs to the activity or intervention they are reviewing. Common topics include issues or concerns, routine monitoring data, and lessons learned from a specific period of the intervention’s implementation. Process. Proper use of time during an AAR session is crucial to ensuring active and inclusive engagement of the participants. In TBIHSS and ReachHealth, a timekeeper is assigned to ensure proper time allocation for the different parts of the AAR session. This practice helps ensure that every participant is given equal time to express their thoughts. To stimulate the discussion, different approaches are utilized such as open-ended questions and root cause analysis. A structured discussion on what went well and what didn’t go well is usually included in the agenda. One good practice across the three IPs is using data during the AAR session. A presentation of data related to the activity of intervention being reviewed is initially used as a starting point for discussion. For instance, in RenewHealth, team members jointly analyze data generated from post-intervention evaluations. This allows them to distill lessons learned from implementation shortcomings. Another type of data that is often used is learnings from previous AAR sessions, which help participants recall previous realizations and agreed-upon actions, and monitor progress over time. Using these different types of data, participants are then asked to come up with adaptive and innovative solutions, and identify risks to avoid adverse consequences in the future. 41 All discussions and agreements made during the AAR sessions are carefully documented. In TBIHSS, a dedicated documenter is identified in the AAR organization plan. Meanwhile, for RenewHealth and ReachHealth, a draft of the documentation is shared with team members for review and refinement, ensuring the accuracy of the data and insights captured. Performance. The ‘performance’ of an AAR session (i.e., outcomes such as agreed-upon action points) is reflected in its main outcomes. In all three IPs, AAR sessions generate lessons learned and action points for improving implementation and solving problems. Since these outcomes may be numerous, IPs prioritize them according to their importance and urgency, and the level of effort required; often in an iterative and unstructured fashion. However, TBIHSS is the only IP that uses a matrix (the Eisenhower Decision Matrix) to identify the action points that are most urgent and require the highest priority. Another way to measure an AAR’s performance is by evaluating its actual implementation, and if participants found it useful and satisfying. However, a post-AAR evaluation is not common practice among the three IPs, perhaps because it may be deemed unusual to ‘review a review.’ An AAR’s performance is also followed-up and monitored by a responsible team across the three IPs. For instance, in ReachHealth uses an AAR monitoring tracker to quickly check the progress of approved action points. Typically, IPs also report progress on implementing these action points to other external stakeholders, for instance by sharing a soft copy of the AAR session’s outcome. However, in the case of RenewHealth, AAR outcomes are only kept internally as a reference for further improvement of subsequent activities. 6.2. Achievement of Implementation Outcomes Related to the Implementation of AARs Fidelity. Among the three IPs, only TBIHSS has a pro forma template to guide AAR session flow. This helped their AAR session stick to the overall design they initially set. Similarly, the other two IPs’ AAR sessions closely adhere to the initial design, even though there is no predetermined template. When compared with existing guidelines such as the WHO’s AAR Planning Roadmap, the most apparent missing step in the AAR process of all IPs is post-AAR evaluation. The IPs do not typically conduct this step upon completion of an AAR session. Adaptation. During the COVID-19 pandemic, AAR interventions underwent a notable shift from in￾person to virtual sessions. Aside from that, IPs reported that they sometimes adjusted the time allotted for particular sessions to accommodate participants' insights. For example, in some instances, RenewHealth extended the post-session debriefing to allow more time for discussion and obtain more insights from the internal staff about the participants' suggestions for improvement, as reflected in the post-activity evaluation survey. Accountability. All of the IPs carefully document AAR sessions, which allows them to record agreed-upon action points, including the names of staff members who are responsible or assigned to execute them. The progress of implementing action points is usually discussed at the subsequent AAR session. ReachHealth in particular developed an AAR monitoring tool for easily monitoring action points. Integration. Rather than being an ad hoc activity, the AAR process is integrated into the IPs’ project cycle. The AMELPs of all three IPs also mention AARs (or P&R), embedding the practice in the organization's CLA system. Moreover, regular implementation of AAR-like sessions, either as part of an initiative with routine activities or as automatic aftermath of a specific event, may also indicate the level by which AARs are integrated into an IP’s operations. 42 Inclusivity. While internal AARs include staff members who are relevant to the intervention under review, external AARs involve various people from outside the team; especially key external stakeholders such as partners from the government, private sector, and civil society. For example, TBIHSS usually has internal general meetings involving the organization’s different teams (e.g., Monthly Technical Team Review-cum-Planning Meetings, which have a strong AAR component). This allows the teams to discover for themselves what happened and why, and how to build on strengths and improve on areas of weakness, as well as explore ways by which they might collaborate more effectively. In terms of external AARs, ReachHealth conducts regional P&R sessions (combined with a harmonization workshop) that include representatives from partner government agencies. Sustainability. AMELPs from the three IPs indicate that interventions for monitoring, evaluation, research, and learning are planned and budgeted. Budget is not necessarily specified for AARs, though it is presumably included in the overall AMELP. The presence of documentation of past AARs that were conducted repeatedly or routinely may also be reflective of the three IPs’ commitment to the sustained implementation of AARs as part of their CLA approach for continuous improvement. 6.3. AARs as Applied to Specific IP Interventions The case studies showed that, in the three specific interventions implemented by the selected IPs, incorporating learnings generated through AARs was an integral part of the process and of improving implementation (Table 5). AARs also served as a major knowledge-sharing opportunity. Internal and external stakeholders assessed the implementation of each intervention (e.g., to gauge whether it was aligned with the intervention’s original design), identify gaps and challenges that require adaptive actions, and determine best practices and lessons learned. Different forms of AARs were used to review the studied interventions: P&R sessions for ReachHealth’s FP Ayuda Express; post-session debriefings for RenewHealth’s motivational interviewing training; and Governing Council Meetings for TBIHSS’s PPSDC. In all three interventions studied, the implementation of repeated or routine AAR activities allowed the implementing teams to address issues and concerns that emerged during certain implementation periods. Most concerns were successfully addressed by the team because adjustments that were agreed upon during an AAR session were accomplished within the agreed timeline. For example, improvements in the content and delivery of RenewHealth’s motivational interviewing training were immediately realized because of the team’s commitment to the adaptive actions that were decided upon during AAR sessions. Specifically, in the first and second batches, participants repeatedly mentioned a need to lengthen the time with master coaches for interactions. In response, the team extended the duration of a particular session from one hour to two hours. This issue did not reappear in subsequent AAR sessions. Moreover, project teams also allocated time for discussion in subsequent AAR sessions to recall agreements made in previous sessions. This practice allows IPs to ensure that the adjustments that were previously agreed upon were successfully incorporated into the intervention. Furthermore, it helped monitor and assess whether these adjustments were effective in improving interventions. 43 Table 5. AAR Good Practices and Resulting Intervention Adaptations Across Three IPs IP and specific intervention ReachHealth’s Family Planning (FP) Ayuda Express RenewHealth’s Motivational Interviewing Training TBIHSS’s Philippine Private Sector Diagnostics Consortium (PPSDC) AAR good practices Conduct P&R sessions with internal staff members and external stakeholders utilizing analytics obtained from online platforms Hold internal team post￾session debriefings to analyze post-training evaluation surveys Convene regular Governing Council meetings where Consortium members can discuss the status of interventions and related challenges Intervention improvements/ adaptations The language used in the chat guide was changed from English to Tagalog, and content was improved to highlight success stories The length of orientation and practice sessions were extended in response to participants’ request Improved strategies and criteria were agreed upon to continuously expand membership It is also important to note that there were instances when issues and concerns re-emerged after being addressed. For example, the FP Ayuda Express team often received inquiries on non-FP related topics that were not covered by the virtual platform during its early implementation run. These issues and concerns were discussed during subsequent AAR sessions. Eventually, this led to the creation of a referral system to link these clients to relevant offices that could assist them. Therefore, it is important to continuously revisit the implementation process through AARs and assess which aspects need further improvement so similar issues and concerns no longer re-emerge. 6.4. Impact of AARs on Implementation Outcomes Related to the Intervention Fidelity. Based on the case studies of the three specific interventions examined, there seemed to be no major deviations from the original vision of the interventions. Rather, the changes that have been instituted as a result of AAR activities are better viewed as enrichment to the original design. The AAR sessions also served as regular checks on whether the intervention is faithful to the initial plan and is achieving the goals that were established at the onset. Adaptation. Common concerns of end-users were the primary consideration for modification and improvement. For FP Ayuda Express, for instance, testimonials from previous clients posted on the Facebook Page achieved the highest reach and engagement, which resulted in producing more content based on successful family planning clients. For motivational interviewing training, the team lengthened its allocated time for practice sessions to cover more topics. This was done in response to participant feedback that was collected through post-evaluation surveys. The COVID-19 pandemic also called for adaptive measures to continue implementation despite limitations and restrictions. For example, PPSDC shifted its interventions online by integrating its training platform to its Consortium website. Accountability. Team members and external stakeholders involved in an intervention are assigned tasks and responsibilities, and there are various mechanisms to ensure accountability and monitor progress 44 throughout implementation. For instance, FP Ayuda Express developed a responsibility matrix to track the execution of agreed-upon actions. This matrix includes the person or organization responsible for a particular action and the target date for completion. In PPSDC, the PPM specialist oversees overall AAR interventions (i.e., Governing Council meetings), including design and execution. In motivational interviewing training, facilitators constantly receive high marks on post-evaluation surveys, indicating their responsiveness to training participants’ needs. Integration. Securing the participation and support of local and national government partners helps ensure that these specific interventions are integrated into the health system. For instance, reaching out to the local health system is easier for PPSDC because its operations are anchored to the Philippine Strategic Tuberculosis Elimination Plan (PhilSTEP) and the Public-Private Mix (PPM) national action plans. It was also supported by an MOU between PhilCAT and DOH that introduces PPSDC members to different regions in the country. Improvement to motivational interviewing training will hopefully capacitate more community facilitators, who will be integrated into local health systems to manage substance abuse clients. In FP Ayuda Express, the consistent participation of representatives from regional DOH and POPCOM aids the implementing team in identifying ways to address challenges related to accessing family planning services, and in incorporating the intervention into the health system. Inclusivity. Internal staff members and external stakeholders participated in AARs for specific interventions, emphasizing shared responsibility towards achieving common goals. For example, FP Ayuda Express’ P&R sessions served as an opportunity for knowledge-sharing across different stakeholders, and resulted in improvements to that intervention’s algorithm and mechanism. For motivational interviewing training, the feedback generated from training participants allowed the implementing team to identify points of improvement that were later applied to subsequent training events. For PPSDC, the inclusion of more private hospitals and laboratories was made possible by developing an inclusive strategy for membership expansion. Sustainability. AARs are helpful in transitioning ownership for interventions to the government and securing needed resources, particularly budgets for continuous implementation. In FP Ayuda Express, lessons learned and best practices that were documented during the implementation—including the P&R sessions—are useful for expanding and developing similar platforms in new locations. For motivational interviewing training, local government has yet to be included in long-term sustainability discussions (e.g., via AAR sessions), but it is hoped that newly-equipped trainees will provide services to PWUDs and strengthen CBDR in the local health system. PPSDC’s AAR activities (i.e., Governing Council meetings) are always anchored to sustainability. With this mindset, members are directly involved in operations, allowing them to develop a sense of ownership towards the Consortium. 45 7. CONCLUSION This study of AAR practices among three IPs implementing the USAID/Philippines Health Project contributes to a deeper awareness and understanding of the AAR process, which is an essential component of the CLA approach. This study also documented examples of how AARs helped create a culture of learning within USAID/Philippines and helped improve the performance of public health interventions. IPs have different understandings of AARs and different ways of conducting them. This is not necessarily a problem, provided these activities still have the key elements of an AAR and fulfill the same purpose. The cases that the study team examined demonstrated how adopting AAR practices facilitates adaptive management, which is the basic purpose of AARs and the reason behind their effectiveness. Activating rapid learning cycles allows IPs to immediately integrate lessons learned from failures and successes into implementation. This is an essential ingredient of iteration, improvement, and innovation. The AAR practices documented in these case studies revealed additional benefits beyond direct effects on adaptive management, learning, and process and performance improvement. One additional benefit is that AAR processes also help enhance engagement between different stakeholders, both internal and external. In the examples examined across the IPs, stakeholder engagement manifested in diverse ways, including garnering participant feedback through post-training surveys, co-designing meeting agendas, and active participation in consultative meetings. As a result, the outcomes of these AAR practices generally reflect the insights and experiences of a wide range of actors, which is a good indicator of inclusivity. Thanks to their inclusive nature, AAR processes also help promote accountability and ownership for interventions, which is vital for its long-term sustainability, especially beyond a project’s implementation period. AAR sessions provide a platform for discussing how interventions can be integrated into the health system. Finally, the practice of AARs among IPs aids in enhancing the responsiveness of interventions to clients and users. These activities provide an opportunity for the voices of clients and users to be considered in reaching agreements on how to improve the delivery of public health interventions. 46 8. RECOMMENDATIONS 8.1. For USAID USAID plays a crucial role in ensuring that AAR practices help strengthen the implementation of public health interventions by activating the PDSA cycle and enhancing adaptive management within implementing organizations. CLAimHealth recommends the following to optimize the adoption of AAR practices across USAID and its IPs: ● Clarify the definition of AAR, its types and dimensions, and existing tools and guidelines. ● Explicitly and actively promote the use of AARs as part of the CLA approach to raise awareness and understanding, and increase uptake. ● Create platforms for joint learning about AAR practices across IPs. ● Encourage and support continuous documentation of AAR practices and their effectiveness through case study development and implementation research. 8.2. For Implementing Partners IPs are not ‘starting from scratch’ when it comes to adopting and practicing AARs. In fact, they are practicing them in different ways. To further expand and enhance the use of AAR among IPs, CLAimHealth recommends that IPs: ● Adopt existing AAR design templates to ensure quality, enhance repeatability, and ease documentation and reporting. ● Consider the use of existing tools for prioritization (e.g., the Eisenhower Decision Matrix), tracking progress (e.g., the Responsibility Matrix), and building accountability. ● Improve documentation of AAR practices, agreements, and outcomes, including through case study development and implementation research. ● Share good practices on conducting AARs with other IPs. 8.3. For Implementation Researchers The study generated rich insights about how different IPs conduct AARs and how AARs differ for specific interventions. There are other arenas that this study only partially covered, which future implementation research can address. To advance scholarship on AARs, CLAimHealth recommends that implementation researchers: ● Examine similar IPs/organizations/interventions, but with different exposures to AAR practices; and consider including a control that does not practice AAR at all. ● Conduct more quantitative investigations to build evidence on the causal impacts of AAR practices on the ultimate outcomes of projects and interventions. ● Gather the perspective of external stakeholders on how important and useful to them are the AARs conducted by the IPs (especially external AARs) 47 REFERENCES Baird, L.S., Holland, P., & Deacon, S. (1999). Learning from action: Imbedding more learning into the performance fast enough to make a difference. Organizational Dynamics, 27, 19-32. https://doi.org/10.1016/S0090-2616(99)90027-X\ CLIC. (nd). After action review (AAR) – evaluate all activities (CPS toolkit). Cumbria, United Kingdom: Cumbria Learning and Improvement Collaborative (CLIC). https://www.theclic.org.uk/resources/toolkits/toolkit-2/appendices/tools-in-detail/a-module￾1/after-action-review Graves, T. (2017). How to build continuous-learning into architecture-practice. https://www.slideshare.net/tetradian/how-to-build-continuouslearning-into-architecturepractice Hengeveld-Bidmon, E. (2015). An After-Action Review of a Research Project Team: A Teaching Case. Mack Center on Nonprofit and Public Sector Management in the Human Services. University of California, Berkeley. https://mackcenter.berkeley.edu/sites/default/files/an_after_action_review_of_a_research_project _team.pdf Keiser, N. L., & Arthur, W., Jr. (2021). A meta-analysis of the effectiveness of the after-action review (or debrief) and factors that influence its effectiveness. Journal of Applied Psychology, 106(7), 1007– 1032. https://doi.org/10.1037/apl0000821 National Police Foundation. (2020). How to conduct an after-action review. https://www.policefoundation.org/publication/how-to-conduct-an-after-action-review/ Serrat, O. (2008). Conducting After-Action Reviews and Retrospects. Washington, DC: Asian Development Bank. https://www.adb.org/sites/default/files/publication/27570/conducting-after￾action-reviews.pdf Sexton, R. and McConnan, I. (2003). A Comparative Study of After-Action Review (AAR) in the Context of the Southern Africa Crisis. United Kingdom: ALNAP. https://reliefweb.int/report/lesotho/comparative-study-after-action-review-aar-context-southern￾africa-crisis Sorbello, S., Bossi, E., Zandalasini, C., Carioli, G., Signorelli, C., Ciceri, F., Ambrosio, A., Zangrillo, A., & Odone, A. (2021). After Action Reviews of COVID‐19 response: Case study of a large tertiary care hospital in Italy. The International Journal of Health Planning and Management, 36(5), 1758– 1771. https://doi.org/10.1002/hpm.3258 Taylor, M. J., McNicholas, C., Nicolay, C., Darzi, A., Bell, D., & Reed, J. E. (2014). Systematic review of the application of the plan–do–study–act method to improve quality in healthcare. BMJ Quality & Safety, 23(4), 290. https://doi.org/10.1136/bmjqs-2013-001862 UNDP. (2007). Knowledge management toolkit for crisis prevention and recovery practice area. United Nations Development Programme. https://reliefweb.int/report/world/knowledge-management￾toolkit-crisis-prevention-and-recovery-practice-area 48 USAID. (2006). After-Action Review Technical Guidance. Washington, DC: USAID. https://pdf.usaid.gov/pdf_docs/PNADF360.pdf USAID. (2013). Pre-action review in Jordan. United States for International Development. https://jordankmportal.com/resources/ldlyl-l~-lmrj-at-lty-tsbq-tkhdh-ljrt USAID. (2018). Facilitating Pause and Reflect. United States Agency for International Development. Retrieved from https://usaidlearninglab.org/sites/default/files/resource/files/cla_toolkit_adaptive_management_facil tiating_pause_and_reflect_final_508.pdf USAID, (2021). ADS Chapter 201, Program Cycle Operational Policy. Retrieved from https://www.usaid.gov/sites/default/files/documents/201.pdf USAID Learning Lab. (nd). Understanding CLA. https://usaidlearninglab.org/qrg/understanding-cla-0 WHO. (2019). Guidance for After Action Review (AAR). Copenhagen: WHO Europe. https://apps.who.int/iris/rest/bitstreams/1212880/retrieve Yin, R. (2018). Case study research and applications: design and methods. 6th Edition. USA: Sage Publications. 49 ANNEX A. SURVEY QUESTIONNAIRE General Questionnaire About AAR Practices Questions Answers AARS IN GENERAL 1 Can you describe the different types of AAR activities that are being done by your IP? 2 What kinds of activities trigger an AAR activity? 3 Can you share with us a list of all AARs you did for the past year, and kindly indicate the activity reviewed, date of both activity and AAR, and venue (including zoom)? Kindly share the agenda or minutes of the meeting for each as well. Kindly share photos of previous AAR activities as well. 4 Can you briefly share some experiences when your conduct of AARs led to successes? How about failures – do you have experiences when, despite the conduct of AARs, a project or intervention cannot anymore be salvaged? AAR PREPARATION 5 Is there a permanent individual or team that is always responsible for the conduct of AAR activities? If yes, who is this individual, or who are the people usually included in the team? If no, who leads the preparations for the AAR activity? 6 Do you allocate a regular budget for the conduct of AAR activities? Is this included in the annual budget? If yes, can you share a copy of the budget? If no, how are AAR activities budgeted? 7 What are your considerations in choosing a venue for the AAR activity, especially before the pandemic? How about now during the pandemic – are your AAR activities all virtual? Briefly describe the situation 8 Are the AARs you conduct sufficiently timely and immediate to the event being reviewed? 50 Questions Answers If no, when is the AAR conducted relative to the activity being reviewed? 9 Do you have a predetermined AAR design based on your internal guidelines? If yes, can we have a copy of the AAR design or guidelines? Perhaps you can also share some written agenda of previous AAR activities? If yes, are there any instances when you cannot adhere to the design planned (before and during the COVID-19 pandemic)? Kindly share some examples – why did that happen? If no, can you briefly describe how the AAR design is developed? 10 How is the agenda prepared prior to the conduct of AAR? Can you briefly describe it? Can we ask for a copy of the AAR’s agenda – perhaps samples from previous AARs? 11 Who are usually invited to these AAR activities – any external stakeholders? How do you determine the set of participants to be invited for the AAR activity? Can we have copies of attendance sheets from previous AARs? ACTUAL CONDUCT OF AAR 12 Can you describe the overall design of your IP’s AAR activities? What happens during the AAR activity? Kindly share copies of the agenda of previous AARs. 13 How do you stimulate discussion and reflection during the AAR? Are there trigger questions prepared on hand? 14 Does the AAR look beyond surface issues and encourage participants to keep ’digging’ to discover the real root cause of a problem? Kindly describe. 15 How do you manage the time for the AAR ensuring that all participants could actively participate and obtain valuable information? 16 How do you prioritize the action points/recommendations generated from the AAR? 17 During the AAR activity, how do you distill lessons from both successes and failures? 18 How do you check through the results of past AARs to identify any lessons that might be useful to the current AAR? 19 During the AAR activity, how do you identify who is responsible for which actions? 51 Questions Answers 20 How do you document the salient points discussed during the AAR activity? If yes, kindly share some sample documentation of AAR activity reports/minutes. 21 How do you assess the AAR event? Do you conduct a quality or audit check or elicit feedback from participants? Kindly share documentation of feedback or evaluation of past AAR activities, if any. AFTER THE AAR 22 How do you monitor progress on the recommendations and action points? 23 Do you report back to participants and stakeholders on progress made during follow-up? How? 24 How do you disseminate the outcomes of the AAR activity to other staff and stakeholders? 52 Questionnaire About the Specific Intervention Questions Answers General Questions 1 Tell us a bit about this intervention – what is it all about? What are its components and activities? 2 How was the intervention conceived and designed? 3 How was the intervention implemented? What is its current status and outcomes so far? 4 Kindly share background documents about the intervention – examples include concept note and workplan (including original and updated versions), reports pertaining to implementation, reports regarding status and outcomes AARs as Applied to the Intervention 5 This intervention was selected for this study because it underwent AARs throughout its implementation. Kindly share agenda/reports/minutes of AAR activities conducted throughout the implementation of this intervention. 6 Do you recall the first AAR conducted during the implementation of this intervention? What were the key findings and recommendations of the AAR? Were the suggested changes incorporated into the intervention? Kindly share agenda/reports/minutes of this AAR activity 7 Do you recall the subsequent (i.e., second, third, etc.) AAR activities conducted during the implementation of this intervention? What were the key findings and recommendations of the AAR? Were the suggested changes incorporated into the intervention? Kindly share agenda/reports/minutes of this AAR activity Assessing the Role of AARs in Achieving Implementation Outcomes 8 Since we are interested in AARs, can you tell us how the conduct of AARs modified or improved this intervention? What adjustments were made? How did your AAR process facilitate the adjustments for your intervention? 9 Meanwhile, can you tell us how your AAR process also helped the intervention to stick to its original implementation design? 53 Questions Answers 10 Can you tell us how your AAR process compelled or motivated the staff to execute the agreed-upon adaptive actions? How about the other stakeholders in the health system? 11 In the conduct of AAR processes for this intervention, were diverse stakeholders within and outside the project included? Who are they? Kindly describe the experience. 12 Can you tell us how your AAR process also aided in ensuring the incorporation of the intervention into the health system (i.e., local, provincial, regional, national)? 13 Can you tell us how the AAR process helped ensure the continued implementation and adoption of the project/intervention even beyond the project’s timeframe? What are the indications? (i.e., budget, policy, manpower capacity, interest) 54 ANNEX B. LIST OF DOCUMENTS REVIEWED Document USAID OH Implementing Partner ReachHealth RenewHealth TBIHSS Annual Monitoring, Evaluation and Learning Plan (AMELP) 3 3 3 Annual Report 3 5 2 2019 Progress Report 2 1 0 2020 Progress Report 3 3 2 2021 Progress Report 3 3 3 Work Plan 4 4 3 Case Study 0 0 3 Fact Sheets 0 0 5 Activity Design / Agenda of AAR 13 0 1 Health Project (HO) Research Undertaking 0 0 1 TBHISS evaluation on Online Training for GeneXpert 0 0 1 FP Ayuda Express Reach (Facebook, Chat Inquiries, POPCOM Helpline) 1 0 0