1

2

3

Table of Contents

Conference Program ..................................................................................................................................... 4

Preamble; ...................................................................................................................................................... 8

Theme: .......................................................................................................................................................... 8

Abstracts; ...................................................................................................................................................... 9

Abstract no. 1: Kenya Task Sharing Policy and Guidelines for Health Care Services ................................ 9

Abstract no. 2: Does leading an Improvement project equip participants with additional leadership behaviors? ............................................................................................................................................... 10

Abstract no. 3: Title: Factors Leading to Ownership and Sustainability of Health Information Systems _ a case of Kenya rHRIS .............................................................................................................................. 11

Abstract no. 4: African health professions regional collaboration initiatives (ARC) on leadership & governance in ECSA region ..................................................................................................................... 13

Abstract no.5: Integrating Information Systems, Policies and People to Reach Every Child with Immunization Services: Best Practices from the BID Initiative in Tanzania ............................................ 17

Abstract no. 6: The electronic Maternal and Perinatal Deaths Notification System (eMPDNS) to reduce maternal mortality. A pilot in Manicaland and Masvingo provinces of Zimbabwe ................................ 20

Abstract no. 7 Improved maternal and newborn health in developing countries: the role of quality improvement .......................................................................................................................................... 23

Abstract 8: Lessons from scaling up program delivery approaches: the case of Maternal Newborn Health and Nutrition programming in Ethiopia ...................................................................................... 24

Abstract 9: PROGRAMMATIC EXPERIENCES AND LESSONS FROM A COMMUNITY-BASED MATERNAL AND NEWBORN HEALTH AND NUTRITION (CBMNH-N) PROJECT IN KENYA: “LINDA AFYA YA MAMA NA MTOTO PROJECT” ................................................................................................................................... 25

Abstract 10: INTRODUCING ZINC THROUGH THE PUBLIC AND PRIVATE SECTOR FOR CHILDHOOD DIARRHOEA CASE MANAGEMENT IN ETHIOPIA:EXPERIENCES, RESULTS AND LESSONS LEARNED ....... 26

Abstract 11: Addressing mistreatment of women during facility-based childbirth: how implementation research has informed action toward the achievement of respectful maternity care in East Africa .... 27

Abstract no. 12: Building the Next Generation of Surgical Work Force in Kenya ................................... 29

Abstract no. 13: Best Fit Algorithm for TB Detection among People Presumed to have TB in the EAPHLN-Operational Research project sites in Kenya: An Hypothetical Approach ............................... 30

Abstract no 14: Does choice of spot, morning or both sputum samples determine optimal performance of a diagnostic tool? .......................................................................................................... 31

Abstract 15: Students’ baseline knowledge about HIV and AIDS at the University of Swaziland .......... 33

4

Conference Program

East, Central and Southern Africa Health Community – ECSA-HC

10th Best Practices Forum and 26th Directors Joint Consultative Committee Meeting

Theme: “Promoting Multi-Sectoral Collaboration for Health Through Sustainable Development Goals”.

Venue: Mount Meru Hotel, Arusha, Tanzania

10th to 12th April 2017

Saturday 8th April 1017 x Pre-Conference Meeting x Consultation Meeting on ECSA-HC Strategic Plan

Sunday 9th April 2017 � Pre-Conference Meetings � SECURE End of Project Dissemination Meeting � Meeting of Heads of Finance from ECSA-HC Member States

Sunday, 14.00 – 18.00

Registration for the Main Conference

Day 1: Monday 10th April 2017 08.00 – 09.00 Registration 09.00 – 10.00 Opening Session

(See Separate Programme) 10.00 – 10.20 Keynote Address: Promoting Multi-Sectoral Collaboration for Health Through Sustainable Development Goals –

Prof. Yoswa Dambisya, ECSA-HC 10.20 – 10.30 Evaluation of the Best Practices Forum: An Introduction – ECSA Secretariat 10.30 – 11:00 TEA BREAK 11.00 – 12.00 Plenary Session 1 (Chair: Swaziland)

Sub theme 1 Keynote: Good Governance and Leadership Practices in the Health sector – Dr Peter Eriki, ACHEST � African Health Professions Regional Collaboration Initiatives on Leadership and Governance in the ECSA

Region – Agnes Waudo, ARC Secretariat Discussion

12:00 – 13:00 Plenary Session 2 Health Workforce for Mitigating the Impact of Emerging and Re-emerging Diseases

� Building the next Generation of Surgical Workforce in Kenya to Address Emerging Disease Conditions – Prof P. Jani, COSECSA

� Addressing the Eye Health Workforce Crisis in the ECSA Region – Dr Ibrahim Matende, COECSA � Brief on the Formation of The ECSA College of Obstetricians and Gynecologists – Prof. Josephat Byamugisha

Discussion 13:00 – 14:00 LUNCH 14:00 – 17.00 PARALLEL SESSIONS Parallel Session 1: Good

Governance and Parallel Session 2: Mitigating the Impact of

Parallel Session 3: Global Health

Parallel Session 4: Global Fund Regional Tuberculosis

5

Programme for 10th Best Practices Forum

Leadership Practices in the Health Sector

Emerging and Re-emerging Diseases

Diplomacy (See separate programme)

Lab Manager’s Meeting (See separate programme)

Chairperson: Mauritius Facilitator: Alphonce/Edward Panel Discussion: Health Data Collaborative

� Kenya Task-sharing and Policy Guidelines for Healthcare Services – Agnes Waudo

� Factors Leading to Ownership and Sustainability of Health Information Systems _ a case of Kenya rHRIS – Rose Kiriinya

Discussion Recommendations

Chairperson: Kenya Facilitator: Willy Were

� Does choice of spot, morning or both sputum samples determine optimal performance of a diagnostic tool? – M. Mwangi

� Best Fit Algorithm for TB Detection among People Presumed to have TB in the EAPHLN-Operational Research project sites in Kenya: An Hypothetical Approach – W.A. Githui

� Students’ baseline knowledge about HIV and AIDS at the University of Swaziland – Prof. P. Dlamini

Discussion Recommendations

Chairperson: TBD Facilitator: Stephen Muleshe

End of Day 1

Day 2: Tuesday 11th April 2017 08:30 – 09:30 Plenary Session 3 (Chair: Tanzania)

Sub-theme Keynote: Multisectoral Responses to Non-communicable Diseases – Dr Abdikamal Alisalad, WHO/Afro � Towards Operationalizing Multi-Sectoral Approaches to NCD Prevention in Africa – Dr Pamela Juma,

APHRC Discussion

09:30 – 10:30 Plenary Session 4 Sub-theme 4 Keynote: Accountability for Women’s Children’s and Adolescent Health Post 2015 – Ms Felister Mayala Bwana, UNFPA

� The Electronic Maternal and Perinatal Deaths Notification System (eMPDNS) to Reduce Maternal Mortality. A pilot in Manicaland and Masvingo Provinces of Zimbabwe – Dr Fortunate Machingura

� Addressing Mistreatment of Women During Facility-based Childbirth: How Implementation Research has Informed Action Toward the Achievement of Respectful Maternity Care in East Africa – Mary Mwanyika Sando

6

Discussion 10:30 – 11:00 TEA BREAK 11:00 – 13:30 PARALLEL SESSIONS Parallel Session 5: Multi-sectoral

Responses to Non-communicable Diseases

Parallel Session 6: Accountability for Women’s Children’s and Adolescent Health Post 2015

Parallel Session 7: Global Fund Regional Tuberculosis Lab Manager’s Meeting (See Separate Programme Attached)

Chairperson: Zimbabwe Facilitator: Rosemary Mwaisaka

� Best practice to regulatory framework and road map to nutrition practice in Kenya – David Omondi

� Africa Maize Fortification Strategy – Ronald Afidra

� Lessons from scaling up program delivery approaches: the case of Maternal Newborn Health and Nutrition programming in Ethiopia – Girma Mamo Discussion Recommendations

Chairperson: Malawi Facilitator: Walter Odoch

� Improved maternal and newborn health in developing countries: the role of quality improvement – Dr F. Manzi

� Integrating Information Systems, Policies and People to Reach Every Child with Immunization Services: Best Practices from the BID Initiative in Tanzania – Robert Kindoli

� Programmatic Experiences And Lessons From A Community-Based Maternal And Newborn Health And Nutrition (Cbmnh-N) Project In Kenya: “Linda Afya Ya Mama Na Mtoto Project” – Jacqueline Kung’u

� Development of Guidance Note of Task Sharing on Cesarean Section by Associate Clinicians – Emily Peca Discussion Recommendations

Chairperson: TBD Facilitator: Stephen Muleshe

13:30 – 14:30 LUNCH 14:30 – 16:00 Parallel Groups Report Back 16:00 – 16:30 Evaluation of the Best Practices Forum TEA BREAK End of Day 2 18:00 Cocktail Hosted by SECURE Health Programme

7

26th Directors’ Joint Consultative Committee meeting Promoting Multi-Sectoral Collaboration for Health Through Sustainable

Development Goals Programme for the Director’s Joint Consultative Committee (DJCC) Meeting

Day 3: Wednesday 12th April 2017

08:00 – 09:00 Drafting of Recommendations Exhibitions

09:00 – 10.30 Plenary Session 5 (Chair: DJCC Chairperson) Presentation and Discussion of the BPF Recommendations - Chief Facilitator

10:30 -10:45 Closing of the Best Practices Forum x Remarks by the Director General x Remarks and Closing by the DJCC Chairperson

10:45 -11:15 TEA BREAK

11:15 – 13:00 Closed Session of the DJCC

13:00 – 14:00 LUNCH 14:00 – 16:00 Closed Session of the DJCC (Cont’d)

END OF BPF & DJCC PROGRAMME

8

Preamble; The East, Central and Southern Africa Health Community (ECSA-HC) is ho the 10th Best Practices Forum and 26th Directors Joint Consultative Committee from 10 to 12 April 2017 in Arusha, Tanzania. The following are the main theme and subthemes for the meeting.

Theme: The theme: Promoting Multi-Sectoral Collaboration for Health through Sustainable Development Goals.

The Conference will address its Theme through the following sub-themes:

1. Good Governance and Leadership Practices in the Health sector 2. Mitigating the Impact of emerging and re-emerging diseases. 3. Multi-Sectoral responses to Non-communicable Diseases. 4. Accountability for Women’s, Children’s and Adolescent Health post-2015

The Conference will be organized in two parts as follows;

i. The Regional Forum on Best Practices; 10 to 11 April 2017 ii. The Directors Joint Consultative Committee; 12 April 2017

The two meetings bring together senior officials from Ministries of Health, Health Researchers, Heads of Health Training Institutions from Member States and diverse collaborating Partners and Experts from the region and beyond.

The aim of the BPF and DJCC meetings is to share best practices and research evidence, identify health policy issues and making recommendations to the Health Ministers Conference to pass resolutions in line with this year’s theme: Promoting Multi-Sectoral Collaboration for Health through Sustainable Development Goals.

The ECSA-HC invited the abstracts of best practices and scientific papers that are relevant to the conference sub themes. The scientific papers and best practices do consist of case studies and evidence based programme experiences that are innovative, unique or have added value and new thinking in health. The abstracts and scientific papers do form the basis for the recommendations that will be presented to the Health Ministers for further deliberation and adoption as resolutions.

These abstracts will thereafter be published in the upcoming ECSA Best Practices Series, for wider dissemination in the region.

9

Abstracts;

Abstract no. 1: Kenya Task Sharing Policy and Guidelines for Health Care Services Main author; Agnes Waudo1 Co-Authors:, Jessica Gross2, Keith Waters2, Rose Kiriinya6,David Njoroge,3 Jamlick karumbi3, Kethi Mullei5, Nicholas Muraguri3, Martha Rogers4, Kenneth Hepburn4 and Andre Verani2

ARC Secretariat1; CDC Atlanta2; MOH Kenya3; Emory University4; Independent consultant5 Emory University Kenya Project6

Background: In Kenya, approximately 30% of nurses were involved in prescribing or initiating patients on antiretroviral therapy (ART) in 2010 according to Service Provision Assessment data, yet Kenya lacked a comprehensive, national framework for task sharing, as recommended in the WHO Task Shifting Global Recommendations and Guidelines. The absence of a national task sharing policy was identified as a major barrier to the scale-up of pediatric ART coverage nationally.

Methods: To support the development of an evidence-based national task sharing policy, the following methods were utilized: 1) a desk review of policies, guidelines, scopes of practice, task analyses, grey literature and peer-reviewed research 2) mapping health workforce distribution by cadre and county to HIV burden and pediatric ART coverage 3) the formation of a policy advisory committee 4) the utilization technical working groups (TWGs) to draft the policy.

Results: The desk review revealed a strong scientific evidence-base for nurse initiated and managed ART (NIMART) in several studies across Africa, as well as in the new 2013 WHO Consolidated Guidelines on the Use of Antiretroviral Drugs, which indicated that trained nurses and midwives can initiate and manage ART. The mapping exercise showed pediatric ART coverage rates (~40%) were approximately half of adult rates (~80%) for counties with the highest HIV prevalence like Siaya, Kisumu, Homabay, and Migori, and that nurses constituted 50%-65% of these counties’ health workforce. A policy advisory committee comprised of national and county level government officials, private, faith-based, academic, regulatory, service delivery, professional association and development partners reviewed the desk review and mapping exercise results and guided the formation of five TWGs: 1) introduction and evidence 2) policy and regulation 3) education and training 4) service delivery 5) implementation, monitoring and evaluation to undertake the work of policy development.

Discussion: High-level support from the Ministry of Health facilitated broad multi-stakeholder participation in the policy development process. The TWGs documented the evidence supporting task sharing, developed a national policy framework analyzing existing and emerging laws, policies and schemes of service, conducted cadre-specific task analyses to identify additional pre-service or in-service training requirements with the authorization of additional tasks, developed a general framework and authorization guidelines for task sharing, and identified implementation and evaluation implications for the national policy.

Conclusions: The development of a national task sharing policy establishes a framework in Kenya for innovative service delivery models for essential health services, including nurse initiated ART as supported by evidence from the desk review, to the scale up pediatric ART coverage, in line with the Kenya Health Policy 2014-2030, the Kenya Vision 2030 and the Sustainable Development Goals.

10

Abstract no. 2: Does leading an Improvement project equip participants with additional leadership behaviors? Corresponding Author: Fleur Kitsell – PhD; Health Dean and IGH Programme Lead, Health Education England, Wessex, UK Email: Fleur.Kitsell@hee.nhs.uk Additional Contributors: Leanne Saddler; Andrew Boyd ; Alex Monkhouse; Anna Gerrard-Hughes; Alice Beardmore-Gray ; Bhavna Oza Background/Significance The importance of leadership development opportunities for healthcare workers is recognised and many programmes have been developed which aspire to equip participants with such skills. The IGH scheme is one such programme, which uses an Improvement Project as the focus for the development of leadership behaviours for all participants. Healthcare workers from all professions in the NHS are recruited and complete a 6-month placement with an overseas partner in a resource-poor setting; during their placement they work in partnership with local health-care workers and together lead and implement an Improvement project, chosen by the in-country partner. The NHS volunteers complete Project Management and Improvement methodology learning prior to undertaking their placement, and share that learning with their in-country partners; they also have regular support from a UK-based mentor. The Western Cape (South Africa) Government’s vision document ‘Healthcare 2030’ includes the key principle ‘to encourage staff to feel empowered to use their judgement and expertise to make the lives of their patients better’. In February 2016 the IGH programme volunteers developed a focussed shared programme of work with 12 senior nurses in Operation Managerial roles in a government hospital in the Western Cape area, supporting them to identify and lead their own Improvement project. Research: State main question/hypothesis Does leading an Improvement Project equip participants with additional Leadership Behaviours? Programme: In 2015 an evaluation of the leadership development of 111 NHS volunteers (known as IGH Fellows) was carried out, and in late 2016 an evaluation of a particular project working with the Deputy Head of Nursing and 12 Nurse Operating Managers in George Hospital, South Africa; who were supported to lead their own Improvement project work was carried out. The evaluation for the IGH programme was based on whether participation in the programme had any impact on participants’ self-assessment of their ability to leadership behaviours (as described in the NHS

11

Abstract no. 3: Title: Factors Leading to Ownership and Sustainability of Health Information Systems _ a case of Kenya rHRIS Authors: Rose Kiriinya1, Kevin Lanyo1, Victor Were1, George Mburu1, Agnes Waudo1, and Martha Rogers2

1-Emory University Kenya Health Workforce Project, 2- Emory University, Atlanta Corresponding Author: Rose Kiriinya rose.kiriinya@emorykenya.org Background: Information and communication technologies (ICTs) have the potential for transforming governance and transparency in the health sector, translating to quality health services across the African continent through increased efficiency and greater accountability (World Bank, 2004). Although there are many on-going projects across Africa that attempt to improve the health sector through the use of ICTs, most remain pilots, few are evaluated and even fewer are designed or assessed for scalability. According to ICTs for health in Africa (2012), WHO recommends sustainable ICT interventions that are part of the user’s daily engagement so that they can become accepted as the norm. By increasing economies of scale and return on investment, the scalability and sustainability of ICT-based interventions are accelerated. Most of the health information systems projects in Africa are developed through donors support and are usually short term in nature and the risk of failure to sustain them is a challenging task (Mursu et al. 2000). In such cases, the HRIS projects are not maintained once the project ends.

In Kenya, health professional regulatory agencies play a significant role towards ensuring excellence in the four characteristic areas (availability, accessibility, acceptability and quality) used by WHO to examine health workforce (WHO Report 2013). The agencies are mandated to regulate training, registration, and practice of health professionals in Kenya.

Methods: The health regulatory bodies under the Ministry of Health, Kenya, in conjunction with Emory University Kenya Project [Centre for Disease Control & Prevention (CDC) funded], collaborated to support the establishment of a state of art regulatory information systems (rHRIS) to provide accurate and reliable data for HRH planning and management. To develop the system, rigorous sensitization of all relevant stakeholders was carried out. Regulatory processes and user requirements were identified through business process analysis (BPA). This further helped to streamline the processes with the functions. Standard operating procedures (SOPs) were developed in line with the identified processes. A web-based, open-source system was designed and developed based on the specifications from BPA. The regulatory agencies were

12

fully involved in the design and development of the system. Each functional unit had control of the features that were developed in respective modules.

A joint technical advisory committee (JTAC) comprising of the technical and information technology officers from the agencies was formed to spearhead the implementation of the system at the respective agency. In addition, a joint regulatory collaborative (JRC) comprising of registrars and the chief executive officers of various regulatory agencies was established to share a common platform for maintenance and sustainability of the system. At all levels of system development and implementation, capacity building in the areas of data and system management, and data for decision making was done.

A sustainability strategy was developed and monitored through periodic gap analysis. To address the identified areas of improvements, a strategic plan was developed. In addition, a transition plan was developed and implemented in a phased approach, which required the agencies to take up system maintenance costs gradually.

Lessons learnt • To get the support of the top management, sensitization at both ministry of health and

agency levels is needed • Involvement of the users in system design and implementation ensures ownership of the

system by users • Use of open source technology helps to lower the cost of sustaining the system due to

minimal license fees • BPA helps to align the system into day to day work process • Continuous capacity building on data and system management, and data use helps to

improve data quality and encourage data use for decision making • Collaboration through JRC & JTAC enhances commitment to system ownership, and sharing

on the cost of system maintenance • To align with government policies, collaboration with the Ministry of Health is paramount • Agency participation in best practices fora encourages ownership & data use. • Documentation at each stage of project implementation is necessary for knowledge

retention and transfer. • There is need to develop a detailed transitioning plan from project to agency ownership and

management. • There is need to develop a sustainability plan early in the project cycle, adequately fund

implementation plan, and evaluate progress • There is need to have a systems that the users can derive value in e.g. increased revenue

collection, reduced lead time, etc. Conclusion/Recommendation: The accrued benefits of efficient health regulatory services, entrenchment of the system into real work processes, increase in council’s/board’s revenue base, and institutionalization of system maintenance have demonstrated a good example of a sustainable HRIS investment.

13

Abstract no. 4: African health professions regional collaboration initiatives (ARC) on leadership & governance in ECSA region Author: Agnes Waudo

Africa Health Professions Regional Collaborative (ARC) Secretariat, Nairobi Kenya

Contacts: Email agnes.waudo@emorykenya.org

BACKGROUND

Most of the Sub- Saharan African countries have similar health workforce issues related to health workforce shortages. WHO estimates that 57 countries worldwide have a critical shortage of health workers, equivalent to a global deficit of about 2.4 million doctors, nurses and midwives. Thirty six of these countries are in sub-Saharan Africa. They would need to increase their health workforce by about 140% to achieve enough coverage for essential health interventions to make a positive difference in the health and life expectancy of their populations. Despite the shortages, many of these countries depend on nurses and midwives as the backbone of health care. Considering that strong health systems provide remarkable improvement in the provision of care, enhancing nursing and midwifery policy and regulation has shown great impact on nursing education and practice through national regulatory frameworks in the African Health Professions Regional Collaborative (ARC) supported countries.

METHODOLOGY

ARC is an innovative, south-to-south collaboration, in which national nursing and midwifery leadership team’s work together to improve professional regulation in the east, central, and southern Africa (ECSA) region. In this regional collaborative, each national leadership team comprises:

x The Chief Nursing Officer, x The Registrar of the Nursing and Midwifery Council, x President of the National Nursing and Midwifery Professional Association, x Senior representative of the nursing and midwifery academic sector.

ARC brings together Nursing and Midwifery teams from each country to own and lead improvements on nursing and midwifery services in their countries. The Nursing and Midwifery leadership are assisted to design projects and compete for funding; expected to successfully execute projects; manage finances and report on outcomes. ARC faculty provides technical assistance on designing and implementation of projects. Summative meetings and two Learning Sessions are organized yearly for learning, networking and south to south sharing among countries. Secondly, it allows for regional dialogue and technical

14

exchange to help advance context-appropriate and harmonized approaches to nursing and midwifery in the ECSA region.

COUNTRY KEY SUCCESSES FOR ARC PHASE 1

x Improved in-country collaboration among nursing pillars x Fostered strong country ownership of projects x Track record in grant and project management x 7 countries established Continuing Professional Development (CPD) programs to

enhance HIV expertise among nurses and midwives. x 12 countries improved and advanced their CPD programs. x 5 countries reviewed and revised the Scope of Practice (SOPs) x 3 countries updated their nursing laws and regulatory frameworks. x 1 country decentralized its nursing council services, enabling nurses to develop HIV

care strategies specific to the local context. x 1 Country established a Nursing and Midwifery council x 2 countries developed entry to practice exams to enable nurses to qualify for HIV

service tasks. x 1 country established a specialty certification in nurse-led HIV care

CHALLENGES

x Limited resources: The project has limited funding which results in only a number of countries being awarded the annual grant.

x Time constraints: The projects are led by the Nursing and Midwifery leaders in their respective countries. These leaders have competing tasks in their official positions and may not have sufficient time to run the ARC activities.

x Changes in quad membership: The Quad is made up of National leaders from their respective countries and any changes in their positions affect the Quad and the continuity of the project.

x National Activities: Impact of National activities led to readjustments of time-frame e.g. political changes, changes in national programs, economic challenges.

x Bureaucracy: Limitations of the Nursing Leadership to make autonomous decisions related to nursing matters.

LESSONS LEARNT

x Resources: Countries learnt to make use of available limited resources. x Networking: It is beneficial to liaise with countries that have similar projects x Stakeholders: It is important to involve the Ministry of Health and other stakeholders

in projects for buy-in and sustainability x Fundraising: Countries learnt to explore alternative sources of funding

15

x Collaboration: There is great impact on the work achieved at the National level due to professional collaboration of quad members for quality service delivery geared towards achieving sustainable development goals.

CONCLUSION

Good Governance and leadership in the health sector begins with the National leaders of a country. ARC has an approach that is designed to empower the Nursing and Midwifery leaders as the influencers and policy makers of the health sector in their respective countries. ARC project has built the capacity of the Nursing and Midwifery leaders in various skills such as leadership, governance, fundraising as well as project management, finance management and project sustainability to address priority needs in development and improvements of regulation of the health workforce.

16

17

Abstract no.5: Integrating Information Systems, Policies and People to Reach Every Child with Immunization Services: Best Practices from the BID Initiative in Tanzania Author(s): Kindoli Robert J. (BID Initiative), Mwanyika H. (BID Initiative), Mtenga H. (BID Initiative) & Apollo T. (BID Initiative)

Corresponding Author: Robert Kindoli Email: rkindoli@path.org

Background

Routine immunizations and new vaccine introductions have proven to be one of the most cost-effective investments to improving people’s health around the globe and prevents an estimated 2.5 million deaths each year. Over the last ten years, increased attention and investments in immunization have reduced mortality rates, particularly among children under five years of age. However global stakeholders and national governments quietly acknowledge that achieving higher immunization rates requires reliable, easily accessed, and actionable data. The BID Initiative, led by Ministry of Health, Community Development, Gender, Elderly and Children (MOHCDGEC) in partnership with PATH, and funded by the Bill & Melinda Gates Foundation, is grounded with the belief that better data, plus better decisions, will lead to better health outcomes. The initiative’s objective is to improve immunization data quality and availability, and increase data use for evidence based decisions at all levels of health system.

The Problem

Routine immunizations and new vaccine introductions have proven to be two of the best investments for improving childhood health globally. Immunization rates in Tanzania have increased significantly during the past decade, reaching above 90 percent at national level. Reaching the last mile (10% of remaining children) and maintaining higher rates requires accurate, timely, and actionable data. Tanzania’s immunization program (IVD) is constrained by limited and often unreliable and inaccurate data that may not reflect the actual realities on the ground. The following were mentioned to be priority challenges for the country:

x Inaccurate or uncertain denominators x Difficulty identifying children who do not start immunization x Defaulter tracing x Poor data visibility at the facility level x Complex data collection tools x Insufficient supply chains and logistics x Inadequate data management and use capacity

Interventions

MOHCDGEC proposed the development of a standards-based Electronic Immunization Registry system to track children and their vaccination records. The Electronic Immunization Registry system integrates labor and delivery services and immunization services such that all newborns can be accounted in the

18

immunization services. It also captures home deliveries through mobile phones to increase access to immunization services including those children whom have yet to access any formal health care. The Electronic Immunization Registry system is a hybrid system having an electronic and a paper version mainly to take into account the Tanzania context given the infrastructure and internet challenges.

Peer learning and networking to connect health service providers to share experiences, challenges and approaches also address different implementation issues through such interventions as WhatsApp groups. Data use job aids, data use mentorship, and targeted supportive supervision were designed, tested and implemented in parallel with the trainings on Electronic immunization registry. These interventions foster sustainable positive system changes in the way immunization services are planned, executed, supervised, and reported.

Methodology

With derived philosophy based on the principles for digital development, the development of BID Initiative interventions employed the Collaborative Requirement Development Methodology (CRDM) approach with the User Advisory Group (UAG) in Arusha region as the demonstration region. This iterative process proved to be effective in accounting for user experiences and perspectives in addressing routine immunization challenges at their work place across all levels of health system. So the developed interventions reflect the reality on the ground and can be sustained by the people who will actually use them. The implementation took an iterative process, with many turning points based on the lessons learnt during implementation process.

Best Practices or lessons learnt

Collaborative Requirement Development Methodology Approach – Tanzania context

Tanzania collaborated with health service providers and decision-makers in a UAG to test and develop the BID Initiative interventions. Inclusion of the two groups together in defining what works best for each of them brought a different dynamic in intervention designing. Most often the two groups had different perspectives on what works best, and the intervention designed, tested, and implemented are a reflection of the compromising state. The UAG is chaired by the Regional Medical Office and is made up of nurses, community health workers, and district and regional health and immunization officers. The group also included representatives from MOHCDGEC and continues to meet on regular basis.

Onsite Training: the best approach to deliver training and hands on experience in implementation

Classroom based trainings have been the core approach in health workforce trainings traditionally. The BID Initiative is using on-the-site training approach to rollout the interventions at facilities level and train healthcare workers. This approach entails training healthcare workers at their place of work and this gives them a chance to continue providing services to their client instead of centralized training where provision of services may be stopped or even health facility to be closed to allow healthcare workers to attend training. Similarly, on-the-site training provides an opportunity to all healthcare workers

19

interested in this training to take part without additional costs. From BID’s experience, this approach is cost effective and more useful for hands on skills building.

20

Abstract no. 6: The electronic Maternal and Perinatal Deaths Notification System (eMPDNS) to reduce maternal mortality. A pilot in Manicaland and Masvingo provinces of Zimbabwe Author(s) and Affiliations: Fortunate Machingura (University of Manchester, Ministry of Health and Child Care (MoHCC) Zimbabwe); Gibson Mhlanga (MoHCC); Henry Chidawanyika (MoHCC) Corresponding Author: Fortunate Machingura Corresponding Author’s email: fortunate.machingura@manchester.ac.uk Background/Significance Despite global progress in reducing maternal mortality, urgent action is needed to achieve the ambitious SDG 2030 target, and ultimately eliminate preventable maternal deaths, identify areas of success, remaining challenges, and frame policy discussions. For most high maternal mortality countries in Sub-Saharan Africa that experience an estimated 99% of global preventable maternal deaths, the rates of reduction that are needed to reduce the global maternal mortality ratio (MMR) to less than 70 per 100,000 live births by 2030 seem too ambitious. In particular, Zimbabwe’s MMR measured in deaths per 100,000 live births increased from 612 in 2005-06 to a peak of 960 in 2010-11, which was followed by a decline to 651 in 2015 (ZDH 2005-6, 2010-11, 2015). Despite the decline, MMR remains unacceptably high and is partly driven by the poor quality of care, lack of routine statistics on who and why a preventable maternal death occurred. Even where data are available, they often suffer from problems in timeliness, accuracy, and comparability. Although maternal mortality estimates can be collated based on statistical models and small-scale community research projects to increase awareness of the problem and especially in the context of inadequate and unreliable vital registration data, they do not provide information needed for targeted and timely response and quantification and determination of death causing factors and their avoidability. Also, the challenge of the Zimbabwe health system has been to weave these disparate data sets of variable reliability and completeness into a national assessment of health conditions and the causes of maternal deaths to guide policy, resource allocation and contextually-attuned maternal health interventions. But Zimbabwe has piloted and rolled out an electronic Maternal and Perinatal Deaths Notification System (eMPDNS) that builds on the principles of the Maternal Death Surveillance and Response (MDSR) public health surveillance system. The eMPDNS is promoting routine identification and timely notification of maternal and perinatal deaths notifications that occur in all health facilities while providing the links with the health information system and quality improvement processes from the local to national level in the shortest time possible. The intervention has provided the inspiration and guidance on how to accomplish the acceleration necessary to reduce preventable maternal deaths substantially. Research question To what extent does the eMPDNS provide information that efficiently guides immediate as well as longer-term actions to reduce preventable maternal mortality by obtaining and strategically using information to guide health interventions and monitoring their impact. Programme/ intervention The eMPDNS, which uses the national DHIS digital platforms, was piloted in Manicaland and Masvingo provinces of Zimbabwe in 2015. In 2017, the system is being rolled out nationally. It uses innovative mobile systems in the notification of community maternal deaths leveraging on the high mobile FM, GM, HC, LC, Abstract eMPDNS - Zimbabwe Penetration (over 90%) in Zimbabwe. The intervention aimed to provide accurate and timely information to effectively guide immediate and long-term interventions to reduce maternal deaths and to count every maternal death while also assessing the actual magnitude of the mortality burden and the impact of actions to reduce it. Methodology (analysis approach, period, setting, including location) eMPDNS pilot location and period In the 2015 pilot phase, the eMPDNS captured, transmitted and analyzed maternal and perinatal deaths notifications that occurred in two provinces of Zimbabwe. It worked alongside the already existing paper death notification forms, information flows and data reviews. Initial data capturing was done on the eMPDNS using mobile tablets at the site where the death had occurred. At the district level data was automatically uploaded into

21

a web-based electronic database and forwarded to the next level. Customized reports were generated from the electronic database. The health worker at facility level who collects and captures eMPDNS data and managers who review and report health data were trained on how to generate and use eMPDNS data. Training on the eMPDNS involved;

� What data to collect for action and defining comparable data elements across districts. � Frequency of local maternal death reviews for facility deaths and the early measures to be instituted. � National guidelines for MDR provided information on processes, availability of data and tools, data

transmission, frequency, required channels, feedback mechanisms, and clear standard operating procedures for MDR at each level, including legal information.

� A user manual was developed as part of training to help users on how to capture and analyze data. An

online open source DHIS2 Manual provided further user information � Data Capture, Analysis, and Interpretation of aggregated findings from reviews involved; � Identifying causes of death, high-risk groups, causal factors, and emerging data trends and to prioritize

health problems to guide the public health response. � The translation of eMPDNS data into information meaningful for decision-makers, the health workers, and

the public for monitoring and evaluating responses. Findings The eMPDNS has and still continues to provide valuable information, which if acted on, can prevent future deaths. It emphasizes the link between information, response and prompts a broader understanding of the wider social determinants of preventable maternal deaths. Its efficient implementation can positively impact on the quality of care, strengthen health systems and improve maternal and perinatal health outcomes But, an assessment of eMPDNS in Mashonaland Central, Manicaland, Midlands and Matebeleland province revealed some gaps. Only 57% of health workers interviewed correctly defined a maternal death. Of the visited facilities only 37% had a maternal death review committee, though all facilities reported regularly reviewing maternal deaths. Only 33% of the facilities reported including community health workers as part of the committees. 46% of the of the facilities reported they conducted verbal autopsies, though no standard tools were being used. Village health workers (VHWs) are responsible for reporting maternal deaths in the community. Of the interviewed 31% were able to give the correct definition of maternal death. The majority 93% of the FM, GM, HC, LC, Abstract eMPDNS - Zimbabwe VHWs were not aware of the existence of a committee at the health facility that discusses institutional and community maternal deaths. Challenges cited by the VHWs on reporting maternal deaths were

� Failure to access information on maternal death from the community � Lack of training on maternal death reporting and notification � Lack of resources and standard reporting tools � Delays in reporting as they have to report the death at the health facility physically

Although the Empnds is now an integral part of the DHIS, it only functions at the health facility level and is not linked to the community surveillance system. Maternal death reviews are hospital based with limited number of hospital staff and no involvement of the community-based workers and no verbal autopsies being done. Thus the challenges of underreporting of maternal deaths, particularly community deaths, challenges in undertaking maternal death reviews and poor linkages of the institutionally based surveillance and the community remain. Research: State knowledge contribution and lessons learned Through ICT innovations, maternal deaths surveillance and response systems have the potential to deliver real-time data, frequent monitoring of maternal mortality level, trends, and causes; provided investments are made on

22

the use of e-health to enhance timeless, completeness of reporting, data accuracy, and analysis as part of the system. Going forward, Zimbabwe needs to develop a community-based health information system that links the eMPNDS with the community surveillance system, through the use of smartphones or tablets in the notification of maternal deaths. Interoperability between the operation of the tablets to the national eMPNDS should be encouraged to ensure availability of real-time information to monitor and inform management decisions at all levels (health facility and community). Also, midwives should be capacitated to conduct verbal autopsies to facilitate task shifting in the review of the community maternal death and response. This approach is significant as it is likely to free up medical doctors who are few and stretched. It would harness the motivation of the community to prevent maternal deaths, improves identification of review of deaths, improve the quality of Maternal Death Review meetings and stimulates action in communities and health facilities – essential elements for tracking progress and accounting for women’s health to achieve the set SDG targets

23

Abstract no. 7 Improved maternal and newborn health in developing countries: the role of quality improvement Authors; Fatuma Manzi, Petro Arafumin, Claudia Hanson, Tara Tancred Corresponding author: Fatuma Man fmanzi@ihi.or.tz Senior Scientist - Health System, Policy and Economic Evaluations Ifakara Health Institute Background In southern Tanzania, uptake of facility births is high at 80%. Paradoxically, maternal and newborn mortality also remain high. These mortality rates are due to the poor quality of maternal and newborn health services. Quality improvement engages stakeholders in a bottom-up approach that aims to identify context-specific problems and create strategies to address these problems. Here we report the experience of applying quality improvement called QUADS in two districts to improve maternal and newborn health in southern Tanzania. Objective The QUADS intervention aims to integrate quality improvement within the district level health system and sustain it at the regional level for quality maternal and newborn health care, better care seeking, and improved health outcomes. Methodology QUADS applies quality improvement at the community, health facility, and district levels. At each level, problems contributing to poor care-seeking and poor quality of care related to maternal and newborn health are addressed. In “learning sessions”, teams develop solutions to the issues they identify. Sessions are repeated quarterly, with new issues being added each time. Results In the first year of implementation, we have trained all CHMTs in two initial districts in quality improvement with the goal that they integrate it in their routine support of health service delivery. Health workers in 1/3 of the facilities in the two districts have been oriented in quality improvement. Thus far, skills around problem identification in maternal and newborn health, data collection and recording, and the use of data to inform their improvement work have been emphasized. Policy QUADS has helped to improve the level of understanding of key maternal and newborn health issues in each district. In general, the program continues to expand the knowledge base for quality improvement for maternal and newborn health in southern Tanzania, which may contribute positively to Tanzania’s national quality improvement strategy.

24

Abstract 8: Lessons from scaling up program delivery approaches: the case of Maternal Newborn Health and Nutrition programming in Ethiopia Authors: Girma Mamo1, Abebe Gebremariam2, Alemu Yosef1, Ermias Mekuria1, Jacqueline K. Kung’u1 Affiliations: 1Micronutrient Initiative; 2Maternal and Newborn Health in Ethiopia Partnership Corresponding author: Girma Mamo Background: Community Based Maternal and Newborn Health and nutrition (CBMNHN) pilot project was designed and implemented by MaNHEP, Emory University and the Micronutrient Initiative to meet the unique needs of underserved pastoralist communities in Afar, Ethiopia and to effectively deliver a package of maternal newborn health and nutrition (MNHN) interventions through a three pronged approach: integrated training, collaborative quality improvement (CQIs), and behaviour change interventions (BCI). There is evidence for positive effects of the CBMNH-N interventions on the knowledge and practices related to antenatal care, IFA supplementation, nutrition and care, delivery, postnatal care, and breastfeeding, after taking the potential confounders into account. Objective: To describe the evidence and lessons learned from the CBMNH-N project that is informing scaling up of MNHN approach in Ethiopia Methods: We conducted landscape analysis of all CBMNH-N components using Program Development Guide developed by Micronutrient Initiative, including: project impact and performance reports, government plans, discussions with project staff and different levels of MoH managers, and reports from field supervision. Recommendations for scale-up were developed based on the prevailing MNHN care needs, socioeconomic characteristics and livelihoods of the community, delivery capacity, availability of relevant distribution platforms and health development plans, as well as relative operational costs. Results: The resulting scale-up plan maintains the three pronged approach to deliver the community based MNHN in areas similar to pilot sites. However, more responsibility placed on Primary Health Care Units (PHCUs) while receiving technical and monitoring support from the pilot partners. In expansion areas where there is more MoH structure and capacity, the MNHN delivery adapted facility-based approach. Technical, financial and monitoring support will be provided but is expected to be required for a shorter period. Relevant recommendations will be compiled and disseminated to implementers at all levels from the ongoing CQI and project evaluations. Conclusion: This process allowed us to identify and incorporate the most critical project components to maximize impact while reducing the cost of initial implementation to an affordable range. This approach will also allow us to continue to re-evaluate, improve and expand the approach as capacities increase.

25

Abstract 9: PROGRAMMATIC EXPERIENCES AND LESSONS FROM A COMMUNITY-BASED MATERNAL AND NEWBORN HEALTH AND NUTRITION (CBMNH-N) PROJECT IN KENYA: “LINDA AFYA YA MAMA NA MTOTO PROJECT” Ndedda C1, Belina S2, Ofware P2, Gachuno O3 and Kung’u JK1 1Micronutrient Initiative, 2Amref Health Africa, 3Pronto International and University of Nairobi Key words: Community-based, Maternal and newborn health, Multi-stakeholder Background: A multi-stakeholder CBMNH-N project in Kakamega County was implemented 2012-15 to address deteriorating maternal health and nutrition indicators in the County. Objective: To draw implementation lessons from a multi-stakeholder CBMNH-N project conducted in Kakamega County to inform future maternal health and nutrition programming. Methods: At the end of project implementation, we reviewed project design and coordination, detailed partner implementation plans, monitoring data, field events, documented anecdotes, activity reports and evaluation reports to make inferences on determinants of success. We complemented these with key informant interviews and a brainstorming session with members of the county implementation team and a focus group discussion with members of the public to clarify perceptions on success, what went well and what required improvements. These findings were then used for a strengths, weaknesses, opportunities and threats (SWOT) analysis to identify key issues across the project lifecycle. Results: Identified strengths included establishment of coordination structures at national and county levels to a) ensure project plans and implementation approaches were technically of high quality and in line with national technical guidelines, b) coordinate all relevant implementing partners’ work, c) integrate national and county level project monitoring and supervision and thus national and county Ministries of Health, d) synchronize work plans and enhance cohesion across stakeholders. In addition, the presence of the Governor’s wife as a champion increased the profile of maternal and newborn health in the County. One opportunity that was seized by the County Government during this project was to increase funding for community health and nutrition. Weaknesses and threats included some conflicting policies across partners, such as methods of procurement, though the coordinating committee was able to remedy these issues. Conclusion: These findings highlight cross-stakeholder coordination and local champions as key components to facilitate program expansion.

26

Abstract 10: INTRODUCING ZINC THROUGH THE PUBLIC AND PRIVATE SECTOR FOR CHILDHOOD DIARRHOEA CASE MANAGEMENT IN ETHIOPIA:EXPERIENCES, RESULTS AND LESSONS LEARNED Eyob Assegida*, Abdulaziz Adisha, Jacqueline K. Kung'ua Author Affiliation aMicronutrient Initiative; *Correspondence: eassegid@micronutrient.org Background It is estimated that the successful scaling up of zinc treatment for childhood diarrhea could potentially save 400,000 under five deaths per year. Prior to 2012 zinc was not available in Ethiopia at large scale. The aim of this project was to assess the availability and use of zinc for treatment of childhood diarrhoea in Ethiopia. Methodology Public and private stakeholders (Ministry of Health, UNICEF, social marketing [DKT], Micronutrient Initiative) collaborated to jointly promote the co-use of zinc and oral rehydration solution (ORS). The results are based on three components: 1) Desk review of key government of Ethiopian diarrhea management documents to assess compliance with international guidelines, mandatory tracking of diarrhea indicators, and information on Zn and ORS delivery, packaging and cost. 2) Program performance based on project progress reports monitoring visits (60 public health facilities, 38 private health facilities in four representative population regions: Oromia, Tigray, Amhara and SNNPR) and 3) Representative Survey (national LC-LQAS 2014) that assessed the use of Zinc and ORS. Results The Lemlem® ORS packs modified to include zinc were rebranded as LemlemPlus®, ensuring private and public sector access. 1) Government documents comply with international guidelines but lack program performance indicators. 2) Zinc and ORS are now available separately or in co-packs through private and health sector .LemlemPlus was available in 45% of Ethiopia’s 6,759 private pharmacies. Monitoring visits revealed that diarrhea management charts/ booklets in 100% of public health facilities but behavior change intervention (BCI) activities were severely lacking. 3) LC LQAS found that Zinc and ORS treatment of diarrhea (children <5yrs) rose from none pre-2014 to ~16% in 2014 (95%CI: 10.4, 21.7). Conclusions The public private partnership facilitated successful program implementation and scale up. Additional support for comprehensive implementation and evaluation of the BCI components is needed. Keywords Diarrhoea, oral rehydration Salt, Zinc, Ethiopia, scale up

27

Abstract 11: Addressing mistreatment of women during facility-based childbirth: how implementation research has informed action toward the achievement of respectful maternity care in East Africa Author(s) and Affiliations: Mary Mwanyika-Sando, Africa Academy for Public Health; Dorothy Temu-Usiri, URC/TRAction Consultant; Wema Moyo IHI; Charity Ndwiga, Population Council; Theopista John Kabuteni, WHO/Tanzania Corresponding Author’s email: epeca@urc-chs.com Background/Significance Evidence from industrialized and low and middle income countries indicates women who give birth in health facilities can face humiliating and undignified conditions. Examples include physical abuse, non-confidential care, non-consented care, non-dignified care, discrimination, abandonment and unfair requests for payment. These negative experiences contribute to poor health outcomes and reinforce mistrust of institutional care. Disrespect and abuse, also referred to as mistreatment, can affect care-seeking across the continuum of services— especially when families are faced with limited resources and difficult choices. The WHO vision for quality maternal and newborn care identifies the provision of care and experience of care (how women are treated), as critical components of quality that can influence service utilization, and ultimately, health outcomes. Respectful Maternity Care (RMC) is a fundamental issue related to human rights, quality of care, service utilization, gender-based violence (GBV), and the ethical imperative. Ensuring RMC is a complex challenge needing a health systems strengthening approach that will respond to individual and health system level constraints such as health provider workload, infrastructure and supply challenges, gender dynamics and health system management. Efforts to measure and address mistreatment of women during childbirth will help us be accountable for women’s, children’s and adolescent health goals as part of the post-2015 Sustainable Development Agenda. Research: State main question/hypothesis How do we measure and address mistreatment of women during facility-based childbirth? Programme: State intervention/activity tested Three multifaceted implementation research projects from Tanzania and Kenya sought to measure prevalence of mistreatment, design approaches to address it (mistreatment), and research the implementation and impact of these approaches. Methodology (analysis approach, period, setting, including location) Two studies in Tanzania (one rural and one urban) and one study in Kenya (in 13 sites across the country) used mixed methods to quantify prevalence of mistreatment of women reported after facility birth and used qualitative methods to understand the drivers of mistreatment. Then, through participatory processes with key stakeholder groups, packages of interventions were developed and tested to address mistreatment and advance respectful care. Data was collected during implementation and again at end-line to understand if the interventions were successful and why. Data (if relevant) Data included qualitative and quantitative base-line and end-line data, and intervention implementation data. Findings The measures used to capture prevalence of mistreatment across the three studies indicated about 15-20% of study participants had reported experiencing disrespect and abuse upon facility exit. Drivers of disrespect and abuse range from socio-cultural factors to health systems issues such as workforce and infrastructure (to name a few). The implementation research results from the Staha and Heshima Projects are promising. The Staha Project

28

showed a significant decrease of disrespect (76% decrease in the intervention site compared to 28% in the control site) and women’s satisfaction with care increased in the intervention facilities. Further, the Heshima Project contributed to a 7% reduction in reports of humiliation and significant decreases in reports of physical and verbal abuse, breaches of confidentiality and detainment. These results suggest multi-level, multi-sectoral and multi-component approaches with consultative processes and linkages between levels can produce a respectful environment of care. Research/Program: State lessons learnt and knowledge contribution Action is required at policy, health system (including health workforce and infrastructure) and community levels to mitigate the prevalence of mistreatment of women during childbirth. Gains can be made with limited investments and can start by raising awareness and reflecting a commitment to respectful care at all levels: national, district and community. Countries such as Tanzania and Kenya are already taking steps to incorporate the implementation research findings into their policies and programs to advance respectful high quality care. This knowledge will contribute to implementing the WHO’s Quality of Care Standards for Maternal Newborn Health Care—and ultimately support the achievement of the sustainable development goals.

29

Abstract no. 12: Building the Next Generation of Surgical Work Force in Kenya Author; Pankaj G Jani, Vice President , COSECSA Corresponding author contacts: Email pjani53@gmail.com Background and purpose: To address the acute shortage of Surgeons in Kenya 2 out of every 3 people do not have access to safe, affordable or timely surgical care and 90% of this occurs in LMIC’s. The combined estimated annual unmet surgical need of 40,949,235 surgical cases in Africa accounts for 29% of the global annual unmet need of surgical cases. Overall, Africa accounts for 24% of the global disease burden but only 3% of the global health workforce. One of the main barriers to surgical care-defined as the provision of safe pre-operative, operative, and post-operative surgical and anaesthesia services-in resource-limited settings is the shortage of trained health workers. There is evidence that provision of basic surgical services can be highly cost-effective in low and middle-income settings. With assistance, COSECSA has the capacity to improve access to surgery by increasing the number of surgeons in the 12 COSECSA member countries including Kenya. Methods: The training of surgeons to provide safe Emergency and Essential surgery at a suitable accredited Hospital thro the COSECSA membership competency based training program, based on, in hospital training, in theatres, wards, and clinics, plus lectures, grand rounds, M&M’s, audio teaching discs, video training materials, IT learning and skills lab training is proposed. The competency-based training approach has been successfully implemented in maternal and child health area and places emphasis on acquisition of competency, knowledge, attitude and skills, rather than just on learning new information. Competency in the new skill or activity is assessed objectively through evaluation of the learner’s performance and an embedded coaching process ensures that learners receive feedback at various points during learning in addition to the education component of training. The major output targets of training are to provide staffing of 3 Membership level and 1 Fellowship level trained surgeons at each of the district hospitals across the 12 member states. The number of doctors graduating has greatly increased in the region to enable the program train doctors and each well-functioning hospital with 3 or more dedicated trained surgeons (Trainers) could become a training facility under the COSECSA training program. This is realistically achievable if spread over the next13 year period so that by 2030, emergency and essential surgery is accessible at the D.H. Data on the number of doctors graduating from Kenya and each COSECSA member country, (potential trainees), the functional hospitals with available trainers, the no. of D.H.’s, total population, plans for training and all the possible problems to achieve this feat will be presented. With the support from the willing international surgical community this feat may advance humanity to the next level. Results: COSECSA has successfully increased the surgical workforce in the ECSA region with 50% are of all surgical trainees being COSECSA trainees at present and 20% of all surgery performed in the region is by COSECSA surgeons. Conclusion/significance: COSECSA has successfully increased the surgical workforce in the ECSA region and can train an adequate number of surgeons for the provision of Essential and Emergency Surgery at D.H. in the ECSA region, with assistance from the international community.

30

Abstract no. 13: Best Fit Algorithm for TB Detection among People Presumed to have TB in the EAPHLN-Operational Research project sites in Kenya: An Hypothetical Approach Authors: *Githui WA1, Mwangi M2, Wanzala P2, Sang WK3, Kiptoo M4 Institutional Affiliations: 1Kenya Medical Research Institute-Centre for Respiratory Disease Research 2Kenya Medical Research Institute-Centre for Public Health Research 3Kenya Medical Research Institute-Centre for Microbiology Research 4South Eastern Kenya University *Corresponding author e-mail: wgithui@yahoo.com Introduction: Direct smear microscopy is a simple, fast and cost effective tuberculosis (TB) diagnostic tool but it does not positively detect all TB cases in people presumed to have TB, especially among HIV positives, due to its low sensitivity. Culture is the gold standard for TB diagnosis but takes a long time. Introduction of GeneXpert, a robust DNA-based technique, has brought new hope in the management of TB. There is need to establish the best fit algorithm for TB detection among people presumed to have TB. Objective: To determine the Best Fit Algorithm for TB detection among People presumed to have TB in EAPHLN-Operational Research (OR) project sites in Kenya. Methods: A cross-sectional study was conducted between February 2013 and October 2016. People presumed to have TB, aged 18 years and above who consented were enrolled from selected public health facilities in Kenya, A structured questionnaire was administered and HIV status determined. Two sputum specimens (spot and morning) were collected from each patient over two consecutive days giving a total of 5318 sputum samples collected from 2659 people presumed to have TB. The specimens shipped to the TB research laboratory at KEMRI, Nairobi and processed for ZN, LED-FM, GeneXpert and LJ culture in accordance with standard procedures. Data management and analysis was done using MySQL and IBM SPSS respectively. Diagnostic test values (sensitivity, specificity, positive/ negative predictive values) for ZN, FM and GeneXpert were determined using culture as a gold standard. The analysis of diagnostic test values was based on combined results for the spot and morning samples, the patient being the unit of analysis. Results: Cumulatively, sensitivity of microscopy was higher in HIV negative for ZN,73.5% (95%CI: 66.1-80.9)%, and FM, 73.5% (95%CI: 65.5-81.5)%, than in HIV positive (55.9% (95%CI: 43.38-68.6) and 57.7% (95%CI:44.3-71.1)%, for ZN and FM, respectively). Sensitivity of GeneXpert (80.0%(95%CI:59.8-100)% was comparable to that of HIV negative (81.8% (95%CI:65.7-97.9)%)%). Generally, sensitivity of GeneXpert was significantly higher in HIV positive than that of ZN and FM microscopy. However, specificity was lower, in HIV positives, than that of microscopy. Performance of all diagnostic tools was lower and comparable in patients with history of previous TB treatment than in patients with no history of previous treatment. Specificity and predictive values remained constant for all the tools. However, there was incremental benefit in the sensitivity of FM (60.7% (95%CI:42.6-78.8)%) in patients with history of previous TB treatment than that of ZN (50.0% (95%CI:33.2-66.8)%) and GeneXpert (50.0% (95%CI:19.0-81.0)%). The hypothetical best fit algorithm chart constructed based on the two analytical attributes (HIV status and TB treatment history), indicate that the highest weighted sensitivities occurred for GeneXpert compared to ZN and FM. The pattern of weighted sensitivity for ZN and FM microscopy was similar but with marginal probabilities. Conclusion: HIV positive people presumed to have TB should optimally benefit from being tested for TB using GeneXpert. However, GeneXpert may not add value in detecting TB in patients with history of previous treatment; but use of LED-FM microscopy shows potential to detection TB in this population

31

Abstract no 14: Does choice of spot, morning or both sputum samples determine optimal performance of a diagnostic tool? Authors: *Mwangi M1, Wanzala P1, Sang WK2, Kiptoo M3, Githui WA4 Institutional Affiliations: 1Kenya Medical Research Institute-Centre for Public Health Research 2Kenya Medical Research Institute-Centre for Microbiology Research 3South Eastern Kenya University 4Kenya Medical Research Institute-Centre for Respiratory Disease Research *Corresponding author e-mail: rewelmwangi@yahoo.com Background: Sputum smear microscopy Ziehl-Neelsen (ZN) using spot and morning samples has been a corner stone in TB diagnosis especially in Kenya. Led Emitted Diodide (LED) fluorescent microscopy (FM) and Xpert MTB/ RIF (GeneXpert) were recently introduced in selected health facilities. Currently there are no guidelines on the choice of specific samples to be used to determine optimal performance of a diagnostic tool. There is a need to advance the diagnostic accuracy of TB diagnostic tools for appropriate patient management. Three different choices of sputum samples, have been used in evaluating performance of TB diagnostic tools in different studies with varying results. These methods include; (1) combined spot and morning samples results, where the patient is the unit of analysis, (2) separate analysis for spot and morning samples, and (3) pooled spot and morning samples, where the sample is the unit of analysis respectively. Objective: To determine whether the choice of sputum samples affects optimal performance of a diagnostic tool Methods: A cross-sectional study was conducted between February 2013 and October 2016 in nine selected public health, facilities in Kenya. People with presumptive TB aged 18years and attending the facilities were eligible for the study were enrolled. Two sputum samples (spot and early morning) were collected over two consecutive days. The samples from study sites were appropriately packaged and shipped to the TB research laboratory in KEMRI, Nairobi, where samples were received and processed for ZN, FM-LED, GeneXpert, and LJ culture in accordance with standard procedures. Culture was used as a gold standard for MTB. A total of 3412 sputum samples (spot and morning) were collected from 1706 people with presumptive TB. All samples were processed for LJ culture, 3366 for ZN, 3370 for FM and 1572 for Genexpert. Results: Sensitivity and specificity of ZN microscopy determined using combined spot and morning samples (69.4%[95%CI: 63.9%-75.0%] and 93.2%[95%CI: 91.9%-94.5%] respectively) was comparable to that of separate spot (67.1%[95%CI: 60.8%-73.4%] and 94.3%[95%CI: 93.1%-95.5%]) and separate morning (63.4%[95%CI: 56.7%-70.0%] and 93.2%[95%CI: 2 | P a g e 92.0%-94.5%]) samples. The results were also comparable to those of pooled spot and morning samples (65.3%[95%CI: 60.7%-69.9%] and 93.8%[95%CI: 92.9%-94.6%]). Sensitivity and specificity of FM-LED microscopy determined using combined spot and morning samples (71.1%[95%CI: 65.6%-76.5%] and 93.8%[95%CI: 92.5%-95.1%] respectively) were comparable to those of separate spot (73.4%[95%CI: 67.4-79.3] and 94.4%[95%CI: 93.3%-95.6%]) and separate morning (69.5%[95%CI: 63.1%-75.8%] and 93.5%[95%CI: 92.2%-94.7%]) samples. Similarly, the results were comparable to those of pooled spot and morning samples (71.5%[95%CI: 67.1%-75.8%] and 93.9%[95%CI: 93.1%-94.8%]). Sensitivity and specificity of GeneXpert determined using combined spot and morning samples (83.1%[95%CI: 76.6%-89.5%] and 88.4%[95%CI: 86.0%-90.9%] respectively) were comparable to those of separate spot (88.1%[95%CI: 81.8%-94.4%] and 89.3%[95%CI: 87.0%-91.7%]) and separate morning (84.3%[95%CI: 77.3%-91.4%] and 88.5%[95%CI: 86.1%-90.8%]) samples. The results were also comparable to that of pooled spot and morning samples (86.2%[95%CI: 81.5%-91.0%] and 88.9%[95%CI: 87.2%-90.6%]). Conclusion:

32

There was no significant difference in sensitivities specificities, ppv and npv between the three different choices of sputum samples used in the three diagnostic tools. However, there was clinical significance if absolute differences between combined samples versus separate spot and morning samples were taken into consideration. The 95% confidence interval for sensitivities specificities, ppv and npv were narrow for the pooled samples compared to those of combined and separate spot and morning samples, thus providing improved precision in estimation when pooled samples are used

33

Abstract 15: Students’ baseline knowledge about HIV and AIDS at the University of Swaziland Author: Priscilla S. Dlamini Contacts: Email psdlamini@uniswa.sz Introduction: University of Swaziland have several programmes meant to educate and raise students’ awareness and provide support related to HIV and AIDS. This study phase one aimed at assessing the knowledge, practice and attitude of the newly enrolled students towards HIV and AIDS thus creating a baseline data by which the effectiveness of these HIV and AIDS programmes can be determined. Method: Phase one of the quantitative descriptive cohort study was conducted, targeting first year students enrolled for the August 2013 intake in all faculties. Data were collected using a self-administered structured questionnaire during orientation. The sample size was 628. Results: Up to 34.5% of the participants thought HIV and AIDS are two different viruses transmitted sexually. About (23.4%) did not know that one can have HIV but be asymptomatic and 23.9% did not know that such a person can transmit the virus to others. About 20.3% and 34.3% did not know that HIV positive mothers can infect their babies during pregnancy or breastfeeding respectively. About (18.2%) did not know that male circumcision (MC) reduces men’s risk of contracting HIV, while 40.5% knew that MC does not reduce women’s risk of contracting HIV. The majority 87.9% knew that there is no cure for HIV, while 4.5% and 2.7% believed that traditional health practitioner and natural products can cure HIV respectively. Conclusion: The majority of the participants knew basic facts about HIV. However, a significant proportion still lacks crucial information and has misconceptions regarding HIV transmission, manifestations and treatment. MC and PMTCT were poorly understood. With the low levels of knowledge on HIV and AIDS, one wonders on the status of knowledge transmission at high schools.