Assessment of the Zambian Health Management Information System

Transcription

1 Assessment of the Zambian Health Management Information System Project SSG/9 ACP ZA 9/5 Arthur Heywood Erik Nielsen Stanislaw Orzeszyna September 2005 c o n s u l t a n t s EURO HEALTH GROUP Tinghøjvej 77 DK 2860 Søborg Denmark Phone: (+45) Fax: (+45) [email protected] internet:

2 Content Content... 2 Abbreviations... 3 Executive summary... 5 Background... 6 Background... 7 Poverty reduction strategy... 7 National Health Strategic Plan Health Sector Reform Health Management Information System Objectives of the Assignment Methodology Activities Findings Millennium development Goals National MDG progress report Health system Indicators Population Data Vital Registration Data sources Surveys Epidemiological surveillance Routine HMIS Human Resources Information and Communication Technology (ICT) HMIS Software Administrative and Management Systems SWOT Analysis Strengths Weaknesses Opportunities Threats Key strategic issues Capacity development Implementation of DART principles Information and communication Technology strengthening Effective use of information HMIS staff retention The Way Forward Revised HMIS Capacity development Improved ICT Increased use of information Annex 1 Terms of Reference Annex 1 Terms of Reference Comments on the Terms of Reference and Key Issues Annex 2 Persons met Annex 3 Timetable...70 Annex 4 Questionnaires Annex 5 Database Annex 6 - Millennium Development Goals Annex 7 Review of the implementation of the 1996 HMIS Annex 8 HMIS Forms Annex 9 Presentations of results

3 Abbreviations ACP ADB AFP AIDS ARI ART ARVs AU AWP CARITAS CBHMIS CBHG CBoH CBHIS CHAZ CHW CMAZ CP CRS CSO CSP DANIDA DART DFID DHB DHIO DHMT DHO DHS DPT EC ECHO EDF EPI EU FAMS GIS GRZ HAST HBC HC HIS HIU HIV HMB HMIS HRD HSR HSSP IDA IFMIS African, Caribbean, Pacific Group of States African Development Bank Acute Flaccid Paralysis Acquired Immuno Deficiency Syndrome Acute Respiratory Infection Anti-Retroviral Treatment Anti-Retroviral Drugs African Union Annual Work Plan Catholic Health Initiatives Community Based Health Management Information System Community Based Health Groups Central Board of Health Community Based Health Information System Christian Health Association of Zambia Community Health Worker Churches Medical Association of Zambia Cooperating Partners Catholic Relief Service Central Statistical Office Country Support Paper Danish International Development Agency Decentralised, Action-oriented, Responsive and Transparent Department for International Development District Health Board District Health Information Officer District Health Management Team District Health Office Demographic and Health Survey Diphtheria, Pertussis, Tetanus European Commission European Commission Humanitarian Aid Office European Development Fund Expanded Program on Immunization European Union Financial & Administrative Management Systems Geographic Information System Government of the Republic of Zambia Integrated HIV, AIDS, STI and TB Program Home Based Care Health Centre Health Information System Health Information Unit Human Immunodeficiency Virus Hospital Management Board Health Management Information System Human Resources Development Health Systems Research Health Sector Support Program International Development Agency Integrated Financial Management Information System Page 3 of 118

4 IP ICT KISS LAN LDC LIMS M&E MCH MDG MoFED MoH MoU MTR NDP NGO OPD ORT PEMFAR PEPFAR PLWHA PPAZ PRSP STI SWAP TA TB TBA UCI UN UNCTAD UNDP UNICEF WHO ZMK ZNBTS In-Patient Information and Communication Technology Keep it Simple and Sustainable Local Area Network Least Developed Countries Logistic Information Management System Monitoring and Evaluation Maternal and Child Health Millennium Development Goals Ministry of Finance and Economic Development Ministry of Health Memorandum of Understanding Mid Term Review National Development Plan Non-Governmental Organisation Out-Patient Department Oral Rehydration Therapy Public Expenditure Management And Financial Accountability Review Presidents Emergency Plan for AIDS Relief People Living With HIV/AIDS Planned Parenthood Association of Zambia Poverty Reduction Strategy Paper Sexually Transmitted Infection Sector Wide Approach to Planning Technical Assistance Tuberculosis Traditional Birth Attendant Universal Childhood Immunization United Nations United Nations Conference on Trade and Development United Nations Development Programme United Nations Children s Fund World Health Organization Zambian Kwacha Zambia National Blood Transfusion Service Page 4 of 118

5 Executive summary Under the Poverty Reduction Strategy Support (PRBS), the European Union provides substantial funds to Zambia for implementation of poverty reduction policies. The PRSP emphasises health and has a number of indicators to monitor the health sector in line with the National Health Strategic Plan strategies and objectives. Strengthening of the National Health Management Information System (HMIS) will lead to improved poverty reduction monitoring. This document contains the results of an assessment of the Health Management Information System (HMIS) conducted by the Euro Health Group (EHG) for the European Union in July and August The second part of the assignment was to formulate a detailed, quantified plan of action to strengthen the HMIS in the country. This plan of action is contained in a separate document. The HMIS assessment was carried out in close collaboration with the Ministry of Health (MoH) and the Central Board of Health (CBoH) during all phases of preparation, field work, data analysis and report preparation. Preliminary findings of the consultation were presented and discussed at two meetings (13 July and 3 August 2005) of the Monitoring and Evaluation Subcommittee of MoH and at two meetings (29 July and 5 August 2005) of the Implementation Review Subcommittee of MoH. The methodology and the results of the mission were also discussed with developmental partners. To assess the HMIS, a framework of the WHO Health Metrics Network was adapted. Particular emphasis was made on the Millennium Development Goals indicators. Other major reference points included the Poverty Reduction Strategy Paper, draft National Health Strategic Plan and National Development Plan. The assessment found that the Zambian HMIS is basically functional at all levels of the health system and compares favorably with most African HMISs. There is a defined indicator set and regularly collected data elements. Data collection and reporting tools are available at all facilities and district offices and the flow of information are clearly set out using the one channel principle. There is regular quarterly analysis of routine data with reasonably good coverage for service-based indicators of underweight children, measles immunization, supervised deliveries, malaria and TB incidence and TB DOTS coverage. However, many MDG indicators are not monitored, there is poor integration of vertical programs and administrative information into the routine HMIS, in general quality of data is not checked regularly and coherently and the system of vital registration is weak. Most staff are inadequately trained in HMIS procedures and there is not much faith in the results coming out of the HMIS. Many of the other subsystems need attention and it is proposed that specific plans are made for HMIS for Hospitals, Finances, Human Resources and community based activities. In addition, the MoH should work closely with the CSO to ensure that the results from the census and household surveys are widely available and that vital registration is strengthened. Five key strategic issues have been identified to strengthen the HMIS: 1. Capacity development of all cadres of staff, including intensive skills development through in-service and pre-service training programs, upgrading of manuals and study of best practice sites. Page 5 of 118

6 2. Return to the 1996 HMIS principles of Decentralization, Action oriented, Responsive and Transparent health information system, and introduction of the information pyramid. 3. Information and communication technology strengthening, through making the database more flexible and strengthening of decentralised information centres that are linked by internet to a central data warehouse. 4. Effective use of information through integration of vertical systems, with improved central coordination between stakeholders and sectors so that the information from HMIS can be used to assess output-oriented performance. Improved action research capacity is needed to improve feedback and dissemination and reduce overlap and duplication. 5. HMIS staff retention, particularly District Information Officers, is needed by improving skills and status and ensuring sustainability of systems, procedures and staff. Major ideas for the Way Forward include: A review of the present system, with adoption of a few key performance indicators that ensure that MDG and NHSP indicators are regularly analysed by all levels and local self-assessment is routinely done to satisfy the needs of health providers, vertical programs and service managers An intensive capacity development program targeted at all users as well as managers, planners and policy makers Strengthening of the ICT, focusing on a revised database and data warehouse, with improved equipment and capacity at provincial and district level information centres. Improved use of information for local action that responds to local needs as well as the needs of program monitoring and management improvement. Promote local action research skills to increase local in-depth analysis of existing records and increased use of Sentinel sites to get quality routine data and up- to- date analysis at local level and improved dissemination and feedback of surveys A MOH project will provide Zambian leadership to the HMIS through use of international standards and regular monitoring and evaluation Page 6 of 118

7 Background Poverty reduction strategy In July 2002, the Zambian Government officially launched its first Poverty Reduction Strategy Paper (PRSP) for the period The main goal is poverty reduction through sustained economic growth and employment creation. In the health sector, in line with existing National Health Strategic Plan, the PRSP has the following programmatic priorities 1 : Provision of the basic health care package, to reduce morbidity and mortality and contribute to poverty reduction. An important component is cost sharing through fee paying; user fees should not, however, constitute a barrier to the poor accessing public health services; New approaches to allocation of financial and human resources to districts to give more weight to poverty issues; community participation and accountability of resources will be promoted; the Sector Wide Approach will be retained; Restructuring of the procurement system, to ensure that purchasing of drugs is done more efficiently and on a need basis; Improving access to health care in hard to reach and under-served areas as well as for vulnerable groups; The following areas are public health priorities: o o o o o o o Malaria - within the framework of the Roll Back Malaria initiative HIV/AIDS, TB, and STI (HAST) - through the National HIV/AIDS Strategic Framework, Integrated Reproductive Health - which includes family planning, safe motherhood, Adolescent health, abortion and post-abortion care, infertility, and (sexual) violence against women, Child health - which will aim at reducing morbidity and mortality due to childhood diseases, Epidemics - improved public health surveillance and control of epidemics, Hygiene, sanitation, and safe water The areas of focus of support services are following: o Development of physical infrastructure and provision of medical equipment o Human resource development o Strengthening of existing management systems The overall goal of interventions in nutrition is to achieve sustainable food and nutrition security among the poor and to eliminate all forms of malnutrition in order to have a well-nourished and healthy population that can contribute to national economic development. The strategies are: 1 Zambia Poverty Reduction Strategy Paper , Ministry of Finance and National Planning, Lusaka, March 2002, Page 7 of 118

8 o To incorporate nutrition objectives into development policies and programmes; improve human resources and institutional finance; and establish strong nutritional networks, o To strengthen nutrition care practices for the poor, the HIV/AIDS infected, and vulnerable groups, o To prevent and control specific macro and micronutrient deficiencies and promote appropriate diets and lifestyles throughout the life cycle. PRSP Cross-Cutting Issues HIV/AIDS - The prevalence and incidence of HIV/AIDS has reached alarming levels in Zambia. The human toll of AIDS is a tragic reality being experienced by families, communities, and the nation at large. There is no aspect of life that has not directly or indirectly been negatively influenced by the AIDS epidemic. AIDS has become the major cause of illness and death among young and middle aged adults, depriving households and society of a critical human resource base and thereby reversing the social and economic gains made since independence. Interventions for HIV/AIDS under PRSP Reduce New HIV/AIDS Infections: Reduce Socio-Economic Impact of HIV/AIDS: Expansion of Access to Quality VCT Plus Services: Community Home Based Care: Anti-Retroviral Viral Treatment (ART): Improve the Quality of Life of Orphans and Vulnerable Children (OVC): Improving STI Management and Treatment in Zambia to Reduce STI infection: Expansion of Access to Quality Prevention of Mother to Child Transmission: Prophylaxis against TB: Drugs for Opportunistic Infections: Evaluation - A number of clinical, epidemiological, behavioural, and impact studies related to HIV/AIDS will be carried out. A sentinel surveillance system for HIV and populationbased studies will continue to be used to monitor the trend of the HIV epidemic. A system of collecting information from health facilities that is already in place to capture cases of AIDS, TB, and other STIs will be used. Lastly, data from various programmes and ministries will be collated and analysed at the national level. Gender - Issues of gender play a very important role in developing a sustainable strategy for poverty reduction. In addition to economic factors, the socially and culturally ascribed gender roles have contributed to high poverty levels among women. Women are usually not part of decisions made on resource allocation at household and other levels. PRSP Evaluation Strategy Poverty monitoring involves tracking key indicators over time and space with a view to seeing what changes have taken place to the indicators following the implementation of the Page 8 of 118

9 PRSP. The central objective of the PRSP is to reduce poverty and evaluations will enable the assessment of the impact on poverty of PRSP interventions. There are three main aspects of the PRSP evaluation strategy: evaluation of the implementation process: to attempt to establish whether programme implementation is done according to design and whether things are working or not; evaluation of outcomes to establish what the results are in relation to the original or revised programme goals and whether particular outcomes are the result of the interventions under the PRSP or a component of it. PRSP monitoring indicators for the health sector Objectives Objectively verifiable indicators (unit) Overall: To improve the health status of all people in Zambia, especially the poor To increase the life expectancy of the population To encourage lifestyles that support health To create environments that support health Life expectancy % Infant mortality rate Maternal mortality rate Under-5 mortality rate Infants aged months fully immunised by 12 months Sexually active adults practising safe sex Children under-5 with low weight for age Reduce by 8% Children who are stunted % % % % To achieve equity in access to health opportunities To provide assured quality health services To provide quality policy and technical guidance to service providers Rural households within 5 km of a health facility Health workers per 1,000 population Cots and beds allocated per 1,000 population Health facilities with 80% of established posts filled Health facilities with essential drugs always available Health facilities rehabilitated New health posts constructed (according to health sector plan) Policies developed Policies reviewed % Page 9 of 118

10 National Health Strategic Plan A 5-year National Health Strategic Plan (NHSP), for the period is coming to an end. The preparation of NHSP takes into account the PRSP/NDP process, the MDGs and new challenges to the basic health care package such as treatment for People Living with HIV/AIDS (PLWHs). Measure Infant Mortality Rate Maternal Mortality Ratio HIV/AIDS Prevalence Summary of Key Performance Indicators Indicator 92 per 1,000 live births 729 per 100,000 live births 16% (16-49 years) Malaria Incidence 396 per 1,000 TB Incidence Life Expectancy at Birth (Yrs) Health Annual Budget Proportion Health Expenditure per Capita Per Capita Income 545 per 100,000 population 50 (CSO) 10% (average) USD 280 USD The NHSP directly addresses the Health Management Information System. It states that current weaknesses of HMIS are, among others: Lack of evidence based decisions Undue influence of sectional interests Weak demand for data Difficulty of monitoring progress Duplication of effort Underuse of data with absence of key data elements Lack of reliable data on inequalities Weak capacity to analyze existing data The NHSP identifies the following strategies and objectives for HMIS: Strategy 1: Create a harmonized framework for the national health management information system Objective 1.1 Define key health indicator standards, along with stratifiers (gender, rural/urban, socioeconomic, etc.) Objective 1.2 Develop platforms for providing essential health information Objective 1.3 Determine data and analytic capacities Objective 1.4 Develop frameworks or guidelines for information use Strategy 2: Strengthen country health information systems Objective 2.1 Establish collaboration with academic and research Institutions for analytic rigor to make information supplied reliable and acceptable Strategy 3: Improve access to and use of health information Page 10 of 118

11 Health Sector Reform Health reforms have been implemented in Zambia since 1992 under the framework of the Sector Wide Approach (SWAP). Resources from government and other stakeholders are pooled so they can be used efficiently. The main success of the health reform processes in Zambia is decentralization down to the district level. In 1996, the Central Board of Health (CBoH) was established as the implementing body. However, failure to implement the delinkage of public health workers from the civil service led to the end of the dualism of the Ministry of Health and Central Board of Health. A new structure of the Ministry of Health is in its final stages of preparation. It will comprise a monitoring and evaluation unit addressing in a comprehensive way the Health Management Information System. There was however a change in attitude generated by the new health reforms whereby a number of initiatives (mainly in the private and NGO sector) were seen as being effective. This showed the possibility of implementing a new national HMIS system which was flexible and more responsive to the needs of all levels of the health system. Health Management Information System Before 1994 the Health Management Information System (HMIS) in Zambia was overly centralized and fragmented, and resulted in the collection of large amounts of data which was not useful for decision making and action. A number of problems were identified with the previous system that included fragmentation, duplication, centralization, delay, unreliability. The system was not producing results and was donor driven, rather than being used for local decision making. Improvement of the HMIS was therefore an important part of overall health reforms and a comprehensive plan was drawn up in 1996 to ensure a functional, district-based HMIS. A review of the development and implementation of the 1996-HMIS is presented in Annex 7. Page 11 of 118

12 Objectives of the Assignment The objective of the consultancy is to make an assessment of the existing health management information system in Zambia and to prepare a work plan for its strengthening. The global objective is to improve the efficiency and effectiveness of health care delivery through the strengthening of HMIS. Support to HMIS is assumed to strengthen poverty reduction monitoring. The specific objective of the consultancy is to develop a detailed, quantified action plan for a strengthened, revised and comprehensive Health Management Information System, focused on coverage of essential health services and on the health status, but open to integration with other components that may be developed at a later stage. The assessment follows the following broad areas: 1. The relevance of the currently operational HMIS to the general M&E framework of the health sector. 2. NDP and HSSP information needs from the HMIS 3. Critical information gaps to the NDP monitoring of the Health sector 4. HMIS capabilities in generating gender and age responsive information on key health indicators and the reporting formats and frequency 5. Investment needs for strengthening the HMIS on: infrastructure (hardware and software), and personnel (training, conditions of service, retention ) to operate the HMIS 6. Mechanisms for quality assurance and control of the HMIS. This report is a product of an assessment of the health management information system carried out by EHG consultants in close collaboration with the Ministry of Health, Central Board of Health and development partners in July-August The second product of the consultancy, a Plan of Action for HMIS strengthening constitutes a separate report. Page 12 of 118

13 Methodology As stated in the Terms of Reference (TOR) the assignment should result in two products, (i) an assessment of the state of the country health information system, its ability to collect, analyse and generate reliable data on specific indicators, and the extent and timing of data dissemination and use, (ii) a detailed, quantified action plan that will build the rationale for HMIS reform, the vision to which Zambia should aspire and options and strategies to help Zambia turn that vision into a reality. It should be quite clear that the content and quality of the action plan, must depend on the results from the assessment of the current HMIS. The short time allotted to the assessment allowed us only to do a thorough assessment of the HMIS ability to collect, analyse and generate data on specific indicators. Although we do have some impressions of the quality and the reliability of data, we cannot substantiate or justify our impressions by exact findings. However it is recognized that the quality of the data is an important issue that has to be addressed by the project for strengthening the HMIS. What follows below is a brief outline of the methodology applied for the assessment of the HMIS, the underlying principles for the exercise, and the key areas of concern. Consultative and participatory One of the basic principles that has guided our work is that it must be a consultative and participatory process, involving as many users of the HMIS as possible, from the facility health workers at ground level, through managers at district, provincial, and national level, to external donors and other cooperative partners. The current HMIS was designed in 1996 and built on the DART-formula, i.e. Decentralised, Action-oriented, Responsive, and Transparent. We have tried to build our assessment using the same principles. The current HMIS has not been revised since its inception. As the ultimate goal of the assignment is to build the rationale for a reform of the existing HMIS we have seen it as an important outcome of the assessment that awareness was raised among the different stakeholders about the need for revising the system. We consulted as many stakeholders, within and outside the ministry, as practically possible within the timeframe given. We interviewed health staff in all 9 provinces, and in 17 of the 72 districts in the country, and a considerable number of health institutions. We visited rural and urban districts, public, private, semi-public, small health centres with very few staff and the biggest hospital in the country, UTH, with more than beds and around employees. During our field visits we were accompanied by staff from the MoH and/or CBoH. When visiting districts and health facilities we were accompanied by the Data Management Specialist from the relevant Provincial and/or District Health Management Board. In that respect the assessment process was also used to facilitate capacity building with in-depth discussions with local staff about findings and the way forward. Page 13 of 118

14 Finally, we were aware during all our work that we should be working with and not for the Ministry of Health. We emphasized direct and active participation of the ministry in the exercise, enabling the ministry to take responsibility and ownership of the plan of action that is envisaged as the offshoot of the assessment. During our work in Lusaka we had daily meetings with the MOH official responsible for the HMIS and a number of officials of MoH and CBoH. We presented our interim reports at the following meetings: 1. Presentation of Inception Report M&E Meeting 13 th July 2. Presentation of 1 st Draft of Assessment Report IRS Meeting 29 th July 3. Presentation of 2 nd Draft of Assessment Report M&E Meeting 03 rd Aug. 4. Presentation of Final Draft of Assessment Report IRS Meeting 05 th Aug. Acceptability The involvement and participation of the ministry in the process of compilation of the report on the assessment was imperative to ensure acceptability and to achieve a realistic plan for strengthening the HMIS. An effective, reliable and timely HMIS is crucial for ensuring optimal health care delivery and as a tool to strengthen poverty reduction monitoring. Its acceptability depends on its ability to meet local needs and be an effective management and planning tool. UN Millennium Development Goals (MDG) Another guiding principle for our assessment has been that the HMIS should monitor the Zambia health sector s commitment to and achievements in fulfilling the Millennium Development Goals (MDGs). The United Nations Millennium Declaration is the commitment to making the right to development a reality for everyone and to freeing the entire human race from want. It was signed by 147 heads of state in September They acknowledged that progress is based on sustainable economic growth, which must focus on the poor, with human rights at the centre. The objective of the Declaration is to promote "a comprehensive approach and a coordinated strategy, tackling many problems simultaneously across a broad front." The declaration calls for halving by the year 2015, the number of people who live on less than one dollar a day. This effort also involves finding solutions to hunger, malnutrition and Page 14 of 118

15 disease, promoting gender equality and the empowerment of women, guaranteeing a basic education for everyone, and supporting the Agenda 21 principles of sustainable development. To help track progress, the United Nations Secretariat and the specialized agencies of the UN system, as well as IMF, the World Bank and OECD defined a set of time-bound and measurable goals and targets for combating poverty, hunger, disease, illiteracy, environmental degradation and discrimination against women. Each year, the Secretary-General will prepare a report on progress achieved towards implementing the Declaration, based on data on the 48 selected indicators. Of the 48 selected indicators 15 (i.e. 31,25%) are directly linked to health services (see list of health MDG indicators in Annex 6). Consequently, monitoring and evaluation of health services is a very important part of the monitoring of the achievement in fulfilling the MDG. HMIS Assessment tool A tool was developed, based on the WHO HMN assessment framework for assessing the Zambian health information systems (HIS) 1. The aim of this tool is: 1. To provide a standardised baseline assessment of the country s health information system 2. To provide a basis for a country HIS development plan, 3. To enable monitoring of progress of the country over time. Objectives of the tool are: Determine performance of sub-systems and overall HIS, and the level of functional integration. To assess performance of HIS with respect to data quality, collection, processing, analysis and use of information Identify on-going development processes and major problems/shortcomings as well as identifying options for addressing such issues. Develop local skills to perform similar assessments in the future Provide clear criteria by which country information systems can be judged This tool provides a standardised baseline assessment for Zambia s monitoring of poverty reduction as defined in the MDGs. However, the tool also incorporates other important components of health system like for instance legal aspects, infrastructure, data sources and utilisation of available data as illustrated. 1 Unpublished paper: Jorn Braa, Arthur Heywood, Sundeep Sahay, Calle Hedberg: Health Metrics Network. A tool for assessing country health information systems (HIS). Page 15 of 118

16 Assessment palette Health System indicators HIS Context Monitoring of MDGs Data Sources Human Resources Legal Framework Information Processes The team focused the assessment on indicators for monitoring poverty reduction, and particularly on the MDG indicators. The assessment tool may serve for developing a standardised diagnostic method for monitoring of progress of the Health Management Information System development. To summarise the methodology, the guiding principles for the assessment have been o o o o o o Consultative and participatory, Decentralised, Action-oriented, Responsive, and Transparent, Acceptable, Awareness raising, Capacity building, Mainly focusing on poverty reduction and MDG indicators Page 16 of 118

17 Activities As illustrated the list of activities in the technical proposal approved by the MoH for the assessment contained 8 steps. What follows is a brief description of the activities carried out during the assessment. Inception 1. Inception and Document Review During the inception and document review, the following activities were carried out: 1. The Terms of Reference (ToR) were clarified and agreed upon with the project steering committee (the MOH Monitoring and Evaluation Subcommittee). The ToR including comments are found in Annex Focus and priority areas for the assessment were defined as the ability of the current HMIS to monitor poverty reduction within the framework of the MDGs. 3. Existing data, reports, studies etc. relating to the HMIS were reviewed and key areas and focus points defined for inclusion in the field assessment (see Annex 7 for details). 4. Instruments and logistics for data collection were agreed on. The general outline of the Health Metrics Network tool was adapted to the realities of Zambia s health sector. Three different questionnaires were developed, one to be used at national and at provincial level, the second one for district level and the last one for health facilities. The questionnaires are found in Annex Geographic areas for data collection were agreed upon and the team members allocated to these. One staff member from either MoH or CBoH was appointed to accompany each of the team members on the field work. The composition of the teams and the areas they visited, are found in Annex Finally a plan was elaborated for the logistics of the field visits and the provinces informed about the visits. Presentation of inception report Field visits Lusaka assessment Draft assessment report Draft plan of action National workshop Final report 2. Presentation of Inception Report Concurrently with the preparations for the field visits the team was reviewing existing documentation on the design and implementation of the 1996 HMIS plan. The review was presented in a steering committee, held on Wednesday 13 th PowerPoint used as the frame for the presentation is included in Annex 9. July. The Page 17 of 118

18 3. Field visits The field visits took place between 14 th and 23 rd of July visiting 8 provinces and 16 districts. Lusaka Provincial and District Health Management boards were visited on 25 th July. Thus, a total of 17 of the 72 districts in the country were visited, which gives a coverage rate of 23.6%. Of the 17 districts visited, 7 were urban, 8 rural, and 2 a mixture of urban and rural. The sample is seen as representative for the country as a whole. Type of district No. % Urban 7 41,2% Rural 8 47,1% Mixed urban/rural 2 11,8% Total districts included in the survey ,0% Total no. of districts in the country 72 Country coverage rate 23,6% A total of 33 health facilities were interviewed and included, including public, private, mission, urban and rural health centres, 1 st, 2 nd and 3 rd level hospitals. The table below gives the details. Type Public % Mission % Private % 1st level hospitals 1 3,0% 1 3,0% 0 0,0% 2nd level hospitals 10 30,3% 5 15,2% 0 0,0% 3rd level hospitals 1 3,0% 0 0,0% 0 0,0% Urban Health Centres 4 12,1% 0 0,0% 2 6,1% Rural Health Centres 8 24,2% 0 0,0% 0 0,0% Hospital Affiliated Health Centres 1 3,0% 0 0,0% 0 0,0% Total HCs included in the survey 25 75,8% 6 18,2% 2 6,1% The field visits were spread all over the country, from the densely populated Copperbelt and Lusaka provinces to the remote and sparsely populated provinces in the north, northwest and west of the country. Results of field visits As each of the provincial and district questionnaires contained 124 records and the facility level questionnaire 150+ records, the amount of data collected during the field visit was huge. As it was impossible to analyse this amount of data manually, a database was developed in MS-Access. The database can in the future be utilised to assess the performance of the Zambia HMIS, or other assessments using a similar tool. The outputs from the database include a number of reports on performance as well as analytical tools that can be used to compare the capacity of the different sub-national users in collection and use of data. Page 18 of 118

19 With the database the Zambia health sector (and all other users of the tool) has a tool that can be utilised to continuously assess and evaluate the performance of the HMIS at all levels. A selected assortment of analytical reports is contained in Annex 5 and the database program engraved on an enclosed CD disc. The structure of the database follows the HMN framework as illustrated below (see next page and Annex 5). It should be noted that the assessment tool neither assess health services nor the health information system, but only the health management information system. We are aware that even if the health management information system is running perfectly it does not imply that the health information system is perfect. However, we believe that a well functioning health management system can add value to the health information system and the performance of the health sector. Database results of the interviews The database reports on the performance of the health information system for the whole country, for the ministry of health, by province, by district and by facility. Further, the database breaks the data down into 4 levels and in addition reports on all the micro-data collected from the interviews with the health service authorities and providers. At Level 0 the overall performance of Zambia HMIS by province and MoH is illustrated in the table and graph below. On average the provinces score is 40.7% of the total score, with relative high variations, with Copperbelt as the best province at 55.6%. Entity Score CEN 40,7% COP 55,6% EAS 47,3% LUA 26,5% Overall performance of Provincial HMB and MoH CEN MOH 60,0% 50,0% 40,0% COP LUS 44,2% NOR 26,3% Prov. Av. 30,0% 20,0% EAS NOW 39,1% SOU 43,3% WES 43,0% WES 10,0 % 0,0% LUA Prov. Av. 40,7% MOH 48,7% SOU LUS NOW NOR Zambia HMIS performance ratio Page 19 of 118

20 Tool for assessment of health management information system LEVEL 0 LEVEL 1 LEVEL 2 LEVEL 3 Prevalence of underweight children (under five years of age) Under five mortality rate Infant mortality rate Proportion of 1 year old children immunized against measles Maternal mortality ratio Monitoring of MDG indic ators MDG Indicators Proportion of births attended by skilled health personnel HIV prevalenc e among year old pregnant women Contraceptive prevalence rate Prevalence of malaria Prevalence of tuberculosis Proportion of TB cases cured under DOTS Zambia HMIS Proportion of population with access to affordable essential drugs on a sustainable Health System Indicators Total health expenditure per capita Health workers per 1000 population Census Vital registration Data Sources Household surveys conducted Surveillance system for acute diseases Monitoring of other indic ators Surveillance system for chronic diseases - HIV/AIDS/ STI/ TB Health service statistics / HMIS Health systems information Legal Framework Legal framework and context HIS context Infrastructure Data availability and management Policy advocacy and dissemination Human Resources Capacity Building Implementation and action Information Proc esses Planning and priority setting Page 20 of 118

21 At Level 1 the data is broken down to monitor MDG and other indicators. Again the Copperbelt gets the highest scores with 50.0% on MDG indicators and 59.2% on other indicators. Entity MDG Others CEN 36,8% 43,2% COP 50,0% 59,2% EAS 39,6% 52,2% MDG/non-MDG by Provincil HM Board and Min. of Health CEN MOH 60,0% 50,0% COP LUA 32,6% 22,6% LUS 42,4% 45,4% Prov. Av. 40,0% 30,0% 20,0% EAS NOR 29,2% 24,6% NOW 28,5% 45,8% WES 10,0% 0,0% LUA SOU 44,4% 42,5% WES 41,7% 43,9% SOU LUS Prov. Av. 38,3% 42,2% NOW NOR MOH 38,3% 49,8% MDG indicators Other indicators At Level 2 the data from the assessment of monitoring of other indicators is further broken into the 7 HMN-frames as illustrated. CEN COP EAS LUA LUS NOR NOW SOU WES P. Av. MOH Monitoring of MDGs 36,8% 50,0% 39,6% 32,6% 42,4% 29,2% 28,5% 44,4% 41,7% 38,3% 46,9% Health System Indicators 37,5% 33,3% 45,8% 0,0% 0,0% 0,0% 45,8% 66,7% 70,8% 33,3% 27,1% Data Sources 40,1% 62,5% 46,4% 24,0% 46,4% 25,0% 42,2% 33,3% 35,4% 39,5% 42,7% Legal Framework 0,0% 66,7% 0,0% 0,0% 0,0% 0,0% 0,0% 33,3% 66,7% 18,5% 0,0% HIS Context 33,3% 41,7% 50,0% 27,1% 33,3% 20,8% 33,3% 33,3% 37,5% 34,5% 56,3% Human Resources 52,4% 66,7% 70,2% 31,0% 61,9% 31,0% 50,0% 42,9% 40,5% 49,6% 57,1% Inform. Processes 53,8% 69,2% 56,4% 23,1% 64,1% 35,9% 61,5% 56,4% 53,8% 52,7% 73,1% To illustrate the performance a variety of graphs can be applied, looking either horizontally (looking at the same systems across all provinces) or vertically (looking at all systems in the same province). The two horizontal examples below show performance on Health Systems Indicators and Data Sources (see Annex 5 for full report). The graphs show considerable variations in the capacity to meet the HMN framework goals with Copperbelt Province performing reasonable well taking in consideration that the HMIS was developed almost 10 years ago and the indicator set has not been revised since then and thus not adapted to demand for monitoring changes. Page 21 of 118

22 Health Systems Indicators 70,0% CEN MOH 60,0% COP 50,0% 37,5% Average 33,3% 27,1% 40,0% 30,0% 20,0% 33,3% 45,8% EAS 10,0 % 70,8% WES 0,0% 0,0% 0,0% 0,0% LUA SOU 66,7% 45,8% LUS NOW NOR Data Sources CEN 70,0% MOH 60,0% 62,5% COP 42,7% 50,0% 40,0% 40,1% Average 30,0% EAS 39,5% 20,0% 46,4% 10,0% WES 35,4% 0,0% 24,0% LUA 33,3% 25,0% 46,4% SOU 42,2% LUS NOW NOR Page 22 of 118

23 Looking at the results from a vertical point of view, the performance of each province broken down into the elements of the HMN framework may be illustrated. The three examples below are Central, Luapula, and Western Province, respectively. Central Province M onitoring of M DGs 70,0% 60,0% formation Processes 53,8% 50,0% 36,8% 40,0% Health System Indica 30,0% 37,5% 20,0% 10,0 % 0,0% 52,4% Resources 0,0% 40,1% Data Sou 33,3% HIS Cont ext Legal Framework Luapula Province M onit oring of M DGs 70,0% 60,0% nformation Processes 50,0% 40,0% 30,0% 32,6% Health System Indicato 23,1% 20,0% 31,0% 10,0% 0,0% 0,0% 0,0% 24,0% an Resources Data Sources 27,1% HIS Context Legal Framework Page 23 of 118

24 Western Province M onitoring of M DGs 70,0% 60,0% Information Processes 53,8% 50,0% 40,0% 41,7% 70,8% Health System Indicators 30,0% 20,0% 10,0% 0,0% man Resources 40,5% 35,4% Data Sources 37,5% HIS Context Legal Framework 66,7% Again, there are considerable variations in how well the provinces perform in relation to the seven groups of indicators. It should be taken into consideration, though, that these graphs do not illustrate the performance of provinces in relation to each other. The final graph illustrates the performance of the national level, i.e. the combined MoH and CBoH. Central Level (MoH & CBoH) M onitoring of M DGs 80,0% 70,0% Informat ion Processes 73,1% 60,0% 46,9% 50,0% Health System Indicators 40,0% 30,0% 20,0% 27,1% 10,0 % 0,0% 57,1% uman Resources 0,0% 42,7% Data Sources 56,3% HIS Cont ext Legal Framework Page 24 of 118

25 Findings Millennium development Goals MDGs are generally not regularly assessed in the Zambian HMIS. Much data is collected that could potentially be used to analyze the MDGs; however, much of that data is not used optimally and many of the indicators are not calculated by provincial or district managers. The following table presents the internationally agreed MDGs and comments on how each one is measured and interpreted in Zambia. MDG Indicator MDG Indicator 4: Prevalence of underweight children under five years of age MDG Goal 1: Eradicate extreme poverty and hunger MDG Target 2: Halve, between 1990 and 2015, the proportion of people who suffer from hunger MDG Indicator 13: Under five mortality rate MDG Goal 4: Reduce child mortality MDG Target 4: Reduce by two third, between 1990 and 2015, the under five mortality rate MDG Indicator 14: Infant mortality rate MDG Goal 4: Reduce child mortality MDG Target 4: Reduce by two third, between 1990 and 2015, the under five mortality rate MDG Indicator 15: Proportion of 1 year old children immunized against measles MDG Goal 4: Reduce child mortality MDG Target 4: Reduce by two third, between 1990 and 2015, the under five mortality rate MDG Indicator 16: Maternal mortality ratio MDG Goal 5: Improve maternal health MDG Target 6: Reduce by threequarters, between 1990 and 2015, the maternal mortality ratio Zambian Measurement Indicator is measured quarterly during performance assessment using routine data. Disaggregated by facility but not by gender or more detailed age group No record review or surveys No reflection on whether the available facility data represents the whole population Not done (except DHS), though Lusaka is starting to look at some facilities. Data is institution based; Death registration is negligible DHS and other survey results are generally not known Not done. Data is only institution based; death registration is negligible DHS and other survey data generally not known Indicator measured quarterly during performance assessment using routine data. Disaggregated by facility but not by gender or more detailed age group No quality control by record review or surveys Includes private data in Lusaka and Copperbelt because vaccines are given free Not calculated, though institutional maternal deaths and live births are available. Some districts and provinces do maternal death audits but generally do not relate to Page 25 of 118

26 MDG Indicator 17: Proportion of births attended by skilled health personnel MDG Goal 5: Improve maternal health MDG Target 6: Reduce by threequarters, between 1990 and 2015, the maternal mortality ratio MDG Indicator 18: HIV prevalence among year old pregnant women MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 7: Have halted, by 2015 and begun to reverse, the incidence of malaria and other major diseases MDG Indicator 19: Contraceptive prevalence rate MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 7: Have halted, by 2015 and begun to reverse, the spread of HIV/AIDS MDG Indicator 21: Prevalence of malaria MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 8: Have halted, by 2015 and begun to reverse, the incidence of malaria and other major diseases MDG Indicator 23: Prevalence of tuberculosis MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 8: Have halted, by 2015 and begun to reverse, the incidence of live births, Maternal Death registration negligible, Cause, time before death and age are not calculated, DHS and other survey results not known. Indicator measured quarterly during performance assessment using routine data. Disaggregated by facility but not by more detailed age group or type of personnel (TBAs, CBEs or midwives) No record review or surveys No reflection on whether facility data represents the whole population Not done at any level and results of national surveys generally unknown, Data available at some laboratories but not analysed by this age group, No use of sentinel sites to get this indicator data does not go to provinces, The number of year old pregnant women not generally known. Not done at any level only New acceptors, which is a meaningless indicator, Data available on number of each type of contraceptive distributed, so one could easily calculate couple year protection, which is a proxy indicator of CPR, DHS results unknown. Incidence at facilities is measured, NOT prevalence in communities, Done on a quarterly basis, using routine facility based data, Some districts doing surveys run by National level but data not at provinces, Quality of data questionable because of IMCI syndromic approach (fever = malaria), Minimal use of sentinel site data and record reviews, No reflection on whether data from the health facilities represents the whole population. Incidence at facilities is measured, NOT prevalence in communities, Quarterly analysis of routine data in all districts, Little further breakdown of incidence by Page 26 of 118

27 malaria and other major diseases MDG Indicator 24: Proportion of TB cases cured under DOTS MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 8: Have halted, by 2015 and begun to reverse, the incidence of malaria and other major diseases MDG Indicator 46: Proportion of population with access to affordable essential drugs on a sustainable basis MDG Goal 8: Develop a Global partnership for development MDG Target 8: provide access to affordable essential drugs in developing countries gender and age, though data is there. Not regularly analysed most respondents claim that ALL TB cures are on DOTS so there is no need to differentiate them from others, Problem with cohort system up to 60% of patients on DOTS do not have initial sputum, so not in cured cohort, Data not integrated into routine HMIS so provinces not regularly informed. Not analysed anywhere, Analysis of drug packs opened per patient (utilisation) and out of stock per facility (availability) relates to service provision but do not address per capita issues. National MDG progress report There was an assessment of MDGs for the whole country 1 to compare data from 1990 and The following results are given: Indicator Probability of meeting target % underweight Unlikely Under five mortality Potentially Infant Mortality Potentially <1 measles coverage Potentially Maternal Mortality Unlikely Skilled Birth Attendances Unlikely ANC HIV positive 15-24yrs Potentially Malaria new cases (NOT prevalence) Potentially There are great gaps in the data and no mention is made of TB prevalence or DOTS, drug availability or CPR. While no mention is made of the precise data sources, the results come mainly from the ZDHS, Annual Health statistical bulletins, ANC sentinel surveillance and Roll Back Malaria surveys. A sustained ongoing assessment of MDGs from health facilities at the periphery using routine data and record review would undoubtedly give important support to improving the reliability and coverage of these figures and make MDGs more real for front-line service providers. 1 MDG progress report 2003 by GRZ and UNDP Page 27 of 118

28 Health system Indicators Data for these health system indicators are generally available, but they are not calculated because there is no demand for them in the HMIS and it is difficult to calculate them with the current database. Total health expenditure per capita Not analysed anywhere though figures are generally available, FAMS not integrated into routine HMIS, Analysis on Health budget by facility not done. Health workers per 1000 population (includes physicians, nurses, midwives) Not analysed anywhere, though figures are said to be available from the HRregisters, Reported on the basis of health workers per patient (workload), Data available on health workers per facility (accessibility), Generally no reflection on the reliability of the data based on projections from CSO. Population Data The census report is generally available only at provincial level, and even there it was usually not electronically available, easily accessed or much used. Provincial managers claim that the census report has been distributed to the district but it was rarely found at the districts visited. The projections seen are not disaggregated by gender and age. At district level there are projections of population by district and by facility and these are generally used as a first line for estimation of coverage of services. There is not much faith in these census figures due to poor counting by enumerators and many provinces and districts use local headcounts for these populations, in order to avoid very high coverage figures. Vital Registration Vital registration systems seem not to be working, with registration of births and deaths (even maternal deaths) negligible. In Lusaka, death registration is enforced by not allowing burial without a death certificate. In all instances it is believed that vital registration is the responsibility of the council or local government. Page 28 of 118

29 Data sources A number of different data sources exist but these are not integrated, tend to be uncoordinated, parallel to each other and data are not easily available to managers or planners. Surveys Household surveys Household surveys on selected MDG indicators (DHS, Malaria and HIV surveys) are done nationally every few years. The Demographic and Health Survey is also done regularly by the CSO and Centres for Disease Control. Provinces are hardly involved in their planning or implementation, though they are used as data capturers. Research and survey skills at provincial and district level are therefore weak. Feedback from these surveys is universally poor, and microdata are rarely available to provincial managers for planning or monitoring. Other surveys Other surveys Service Availability Mapping (SAM), the JICA facility surveys, Service Provision Assessment (SMA) for HIV, Health Manpower etc. are conducted in most provinces by national and international organisations, often in a top-down and non-transparent manner. Again, feedback is weak and micro-data is not available for planning at provincial or district level. Epidemiological surveillance The response system seems to be well developed in all provinces, with special teams set up for event management, contact tracing and follow up of laboratory results. No assessment was made of the effectiveness of these teams. The Integrated Disease Surveillance (IDS) reporting system covers notifiable diseases and is largely parallel to the routine HMIS because of the need for immediate reporting for reaction to disease outbreaks. The IDS is partially integrated into the HMIS which provides quarterly data on malaria deaths and vaccine preventable diseases. However HAST surveillance remains outside of the HMIS, with provincial coordinators, seriously limiting provincial capacity to plan and manage this epidemic in an integrated way at a local level. Sentinel sites are present in most provinces, but only a limited number of districts. Provinces are hardly involved in their support or supervision and these sites seem to be regarded as national projects run by vertical programs rather than as a source of accurate and reliable data for local management Data from sentinel sites is not available for provincial or district planning. Routine HMIS The routine HMIS was the main focus of this assessment. It has been running since 1996 and is well established in all provinces, districts and facilities, with good coverage and reporting Page 29 of 118

30 rates. It covers routine service activities, special programs and integrates some aspects of epidemiological surveillance. Laboratory data, environmental health and administrative data (finances, human resources, transport etc) are not collected routinely in the HMIS. Hospital data is also extremely weak and poorly collected. Although data on human resources (through the HR-registers), finances (through the FAMS), drugs and assets are available they are not integrated in the routine HMIS and data is poor quality and incomplete. Indicators on performance in terms of human resources, finances, drugs and inventory are neither estimated nor reported as part of the routine HMIS. Financial data tends to be late and inaccurate. Data Collection Data is collected through a defined set of patient record cards, tally sheets and registers which are used to complete monthly reports, sent quarterly to the district offices. Staff appears to be happy with using these forms and there is a variable response to how much time they take, though there is a consensus that they are useful in collecting data. There appears to be little problem with supply of cards, tally sheets or registers. Private sector collection varies considerable in many provinces, they hardly exist, in Copperbelt they mostly report whereas in Lusaka they are not included except for immunisation. While they are obliged by law to report notifiable diseases, this is rarely enforced by the medical council. National Indicator Set There is a well defined national indicator set, consisting of Performance Indicators for the MoH and CBoH 1 and Indicators for the NHSP 2. While a few numerators and denominators for the quarterly indicators are well known at all levels, there is little use of other indicators and development of relevant indicators for local management is weak. Much of the data reported is not used to calculate these indicators and there are many more data elements collected than indicators. There are also major overlaps, with parallel reporting systems collecting the same information through different channels. Data definitions are not widely available and data quality is not regularly checked or crossreferenced by register reviews, surveys or other quality control mechanisms. There are some major gaps in that some program areas are completely excluded from the indicator set these include environmental health, mental. 6 of the 15 health-related MDG indicators are calculated routinely by the HMIS. These are described in detail in section 1 MDGs. Many of the other routine GRZ HMIS indicators are close to the MDG indicators and could, with a little adaptation, measure the MDGs at all levels. However child and infant mortality indicators will need to be captured outside the routine HMIS. Private sector data The mission / NGO sector is far from fully integrated into the GRZ health services and not all their data is incorporated. The mission hospitals are only reporting on the grants they receive 1 MoH and CBoH Performance Indicators April GRZ MoH Indicators for the NHSP December 2002 Page 30 of 118

31 from the GRZ while all other inputs to their functions and management operations are excluded from being reported. As such they do not report on donations, neither in money nor in kind nor in human resources, from their mother-churches. Some NGO s may in some case report on some of their activities but the general picture is that they are far from fully integrated in the routine HMIS. The private (for profit) sector is very weak outside the main urban areas and does not appear to be a major issue. On the Copperbelt the private hospitals and clinics are integrated into the routine HMIS on issue of service delivery but are not reporting on financial nor on human resource issues. In the urban areas, private sector data is generally not captured and private facilities and their staff do not appear on provincial or district databases. Data collation Data is collated quarterly at district level, where it is entered into a uniform computer program developed in While there are some data entry quality control mechanisms, these are weak and can easily be circumvented, leading to potentially poor data quality. Data quality is not cross-checked by record reviews or against surveys. Reporting Most districts enter data onto computers and produce standardised quarterly reports to provincial level. These are mainly sent by diskettes or memory sticks, though a few districts send by . Even on the Copperbelt where internet facilities are easily accessible, the district reports are submitted by hand, the reason being that the submission of the reports is seen as an opportunity for the Province Data Management Specialist to discuss and give feedback on the quality of the district HMIS reports. Feedback Quarterly performance assessments (PAs) are regularly conducted at all levels, with written feedback to units that have been supervised. Information from the HMIS is extensively used in these PAs and it is the major source of direct supervision for most units. Apart from these PAs, written feedback is weak at all level of the system and tends to consist of pages of tables with minimal analysis or interpretation. Analysis and Interpretation Self-assessment is generally weak, though some districts (e.g. Mungwe in Northern Province) provide notable exceptions to this statement. Indicators are calculated quarterly and compared using standard guidelines to performance of previous quarters, with general comments made about quarterly variations. Graphs are widely used for displaying information on notice boards, though many feedback reports do not really use graphs enough. Maps are used to show place locations, but GIS is not widely used for display of comparative health service information. Graphs and maps are not commonly displayed for the public. Information use Use of information for annual planning and monitoring is generally good because this is related to fund disbursement. Page 31 of 118

32 However, for the HMIS the 1998 guidelines are not up-to-date and there is little selfassessment or use of information for analysis of performance and little incentive for units that make good use of information. Planning, Monitoring and Evaluation Information is widely used for planning, with an excellent series of regularly updated planning manuals produced for Districts, Training institutions, 1 st level hospitals, 2 nd and 3 rd level Hospitals as well as Health Centres, Health posts and communities 1. These manuals are comprehensive and provide step-by-step guidance on the implementation of the planning process and emphasises the use of integrated local information, with detailed worksheets and comprehensive annexes. Action plans exist at all levels and most look at different problems identified through the HMIS, with budgets and resource allocation based on perceived priorities. Plans are made around indicators, with good links to the appropriate targets. These plans appear to be followed closely, using the HMIS to monitor them regularly through quarterly reports and performance assessments. National benchmarks are available and most provinces (and some districts) have set local benchmarks against which to monitor progress. Strategic plans are prepared for provinces and within these, there are elements of monitoring and evaluation, but there is little emphasis on integration of different HMIS data sources. However, the strategic plans are not strategic in the sense of setting targets for the long-term development but merely a medium-term (3 years) collection of planned interventions Access to information The statistics act was last revised in 1964 and there is not free access to information. A recent circular from the Permanent secretary of the MoH discouraged dissemination of information without specific permission. Information Dissemination Dissemination of information is generally weak, with limited copies of documents available, particularly at the lower levels. The locally produced documents that are available are generally of poor quality with limited analysis. Quarterly Indicators Selected Quarterly HMIS Indicator reports are produced regularly for submission to the Health Sector Support Steering Committee meeting. The 11 indicators presented are intended to show steering committee members and donors that services are being provided so that funds can be released for the subsequent period. The indicators are defined, some trends are graphed and the main body of the document is a series of tables of quarterly indicators by province and district, with some comments. There is very little analysis or interpretation of the data completeness or quality, or even indepth trend analysis. Many of the indicators are in fact meaningless in terms of whether the health system is performing well or not. Some provinces 5 have done innovative analysis, calculations and graphs, but these are all manual and the ideas and information have not been widely disseminated. 5 Central and Luapula, with support from health advisors. Page 32 of 118

33 Annual reports Most provinces produce annual reports or statistical bulletins bi-annually. Most of those seen are merely pages of tables with very little analysis or interpretation. They do not cover all subsystems and most give little impression of what is really happening within the province with regard to MDGs, national special programs or local priorities. Very few districts or facilities produce annual reports. Horizontal reporting to other health related ministries and program areas at the same level is generally weak and information sharing with politicians and the media is also weak. Human Resources Each province has a Data Management Specialist (DMS) and each district has a District Information Officer (DIO) post, (not recognised by the Ministry of Finance) which is generally filled. Capacity of these officers in core health information sciences (Epidemiology, Demography and Statistics) is generally weak and all recognised the need for additional training at all levels. The presence of health advisors in provinces is a temporary support to this capacity, but this capacity does not go down to district level. Guidelines exist for data collection and analysis and copies are available in most facilities visited. These date back to 1998, have not been revised and do not reflect the current needs of local management or the higher levels. Using these guidelines, most districts and provinces have done induction training of new appointees and some conduct ad-hoc in-service training of facility staff off their own initiative. Capacity Development There is almost no structured and objective-oriented HMIS capacity building activities carried out at any level and formal capacity development activities have hardly occurred since 1998, when the system was introduced. Cascade training never cascaded downwards, and trained trainers have gone off to other, better managed and organised tasks. There has been no systematic training of district HIS staff, district program managers or health facility staff in the past five years. Rapid turnover of staff means that, while staff knows how to fill in and use the data collection tools and can complete quarterly reports, capacity for self-assessment and epidemiological thinking is limited. There is inadequate understanding of WHY information is collected, HOW to analyse or interpret information or WHAT to do with information. There is only limited understanding of the importance of reliable data for policy settings, operational management purposes, planning of interventions and general management. Page 33 of 118

34 Information and Communication Technology (ICT) All provinces and most districts have adequate HMIS computer infrastructure, have the HMIS software installed and are able to enter HMIS data. Most districts have additional computers provided by vertical programs, but these do not have HMIS software and are not accessible to HMIS staff or staff from other programs. Most hospitals also have computers, but do not use them for the HMIS. Quarterly reports are normally sent to provinces by diskette or memory stick, rather than by . Most districts have telephone connections. While all districts have internet accounts, most of these are dysfunctional and they have no access to or the internet. No districts or provinces are using satellite technology to access the internet. There is a widespread network of Single side Band radios at facility level, and many mission facilities use modems connected to these to access using the Bush Mail service provider. Technical IT support is generally weak within the GRZ system. Some provinces and districts have maintenance contracts with private providers who keep the systems functioning; others rely on the GRZ MoH technical services available when technical problems arise. In all cases, funding for IT maintenance and support is limited. Maps and electronic mapping information is widely available through the EPI-MAP program. Some provinces use GIS to map service data, though few districts had the skills to do this. GPS coordinates for facilities are available at a central level for every facility through the JICA facility survey, but these are not widely available at district level for use on GIS. It is hoped that data will be available in the near future. There is no integrated data warehouse or repository running at national or provincial level and data is not available via the web or on the ministry intranet. This means that it is extremely difficult to access data from different data sources at one point. HMIS Software The HMIS software was developed by a consultant in 1996 and has hardly been changed since. Source code is not readily available and any changes needed have to go to the original programmer. The result is that the software has not been adapted or modified since its development and nobody at the periphery has the skills or authority to undertake such modifications. It is MS Access based and has entry screens that reflect the report forms and some error traps. The overall structure is rigid a classic black box system that is extremely difficult to work with for local adaptation, e.g., to change entry or reporting formats and would need technical programming skills not available at local level There are only limited links to spreadsheets to enable further manipulation of data and no provision for pivot tables for thorough analysis. There is no integration with the FAMS, Human Resource databases or any of the vertical programs and sub-systems. This makes production of integrated reports almost impossible. This HMIS program is not acceptable in the 21 st overhauled. century and needs to be radically Page 34 of 118

35 Administrative and Management Systems In addition to the routine HMIS a number of administrative and management systems are maintained at national, provincial and district level. Most of these systems are manual, however, attempts to computerize some of the systems has been ongoing for some time, in particular the Financial and Administrative Management Systems (FAMS). None of the administrative or management systems are even partly integrated with the routine HMIS, implying that monitoring of these indicators, e.g. on finances, human resources, drugs, supplies, and assets, is excluded from the monthly and quarterly reporting procedures. Financial and Administrative Management Systems (FAMS) The Financial and Administrative Management Systems (FAMS) was initiated at district level as early as Computerized accounting procedures, through Navision Financial were introduced at CBoH in 1997 and later the same year a preliminary district version was installed in two districts for pilot testing. Considerable resources have been invested to the FAMS development at CBoH. Recently Navision Attain has replaced Navision Financial as the software platform for FAMS and there are plans for introducing the computerized FAMS in all provinces and districts. A pilot programme has been going on for some time in a few selected provinces and districts. Funds for computerization of the MoH have been secured and the MoH-accountants have received basic training in Navision. However, the training done so far has only been for one week which is far from sufficient, and the existing plans for further training are well below what is needed. As CBoH in the near future will merge with MoH, the two accounts sections will become one section. The new accounts section will be placed under the Chief Accountant at MoH with two Principal Accountants, one responsible for GRZ-accounting and the other one for donoraccounting. Chief Accountant Principal Accountant (GRZ) Principal Accountant (DONOR) The Principal Accountant for donor accounting will have two subordinated Senior Accountants, one for Management and the other one for financial issues, each one with a number of subordinated accounts assistants. The total number of persons in the accounts department will be 35. Page 35 of 118

36 For the accountants at MoH it will be a totally new experience to work with computerized accounting packages. Only a few of the accountants have any practical and real experience with computerized accounting. The current accounting culture is based on manual systems and procedures and just to change the attitude will require considerable resources. There seem to be a belief in the department that the Navision software will enable the Account Managers to get immediate, timely and reliable financial report just by a snap of the finger. However, the functionality of a system depends on the users capability and understanding and in this regard the need for training, both formal and on-the-job training, seem to be underestimated. The district offices sign an annual contract with the CBoH where the obligations of the two parties are laid down, among others the quarterly disbursement schedule. The districts pass on the grants to the health institutions according to a predefined distribution key, based on agreed criteria, e.g. catchment population and level of services. This is done in form of a standing imprest for the health facilities in accordance with the activities contained in the annual action plan and budget. All district accountants are using the same manual FAMS, with a cash book and a system of ledger cards for the health facilities. Accounting procedures are laid down in a Financial Procedures Manual. The system is not very complicated, it complies with general accepted accounting principles, and is well-documented. However, the capacity of the account sections at subnational level varies considerably, and where the staff turnover is high the capacity is low due to general lack of feedback, supervision, support, and training. Although a plan for expanding the Navision to all provinces and districts in principle has been approved, the plan should be reviewed. Firstly, because funds for the investment have not yet been secured, and secondly, because a FAMS based on the Navision accounting software will be extremely expensive, both in terms of investment and recurrent costs, thirdly, because technical skills to maintain the software are only available outside the ministry, thus making the ministry dependent on outside service providers, which is an expensive and not sustainable situation, fourthly because the Navision software will demand extensive support to the users at all levels, both as formal training and on-the-job training, and finally because less complicated, more user-friendly, and less expensive packages would make the FAMS more sustainable. Finally, it should be considered how far down the FAMS should use the same common software platform, like Navision. Should the ministry, the province, the districts, the health centres, maybe even the individual health workers, all use the same? Or would it not be more reasonable to base the choice of software on individual needs? And for some users not introduce software at all, but base the system on manual, or semi-manual, procedures? Other administrative and management systems Recently a 6 month internal project to the Logistics Management Information Systems (LMIS) commenced. The objective is to lay down the procedures for the management of drugs and supplies, including development of a database, elaboration of reporting structures, compilation of procedures manuals, and training of staff. An important aspect of this assignment is to reach agreement on the integration of the LMIS and the HMIS. Currently there is no integration at all. A partly computerised Human Resource Management System has been introduced at central level, but has not yet reached the districts. No part of the human resource management system Page 36 of 118

37 has been integrated with the HMIS, thus making it very time consuming to estimate indicators on human resources, manpower and workload. Apart from the Copperbelt Province, data on human resources in the private sector is not collected or reported. Mission hospitals only report on human resources paid for by public funds and do not include volunteers. At district and provincial level assets are not included in the routine HMIS or FAMS. Assets not paid for (donated by external donors) are not costed. Integration between HMIS and other administrative and management systems Currently there is no integration at all between the routine HMIS and other public administrative and management systems. Districts and provinces report on FAMS figures but no ratios combining HMIS and FAMS data are estimated. The same goes for ratios on human resources, assets, drugs and supplies. The lack of comprehensive monitoring is aggravated by the fact that only in exceptional cases is data from outside the public sector collected and reported. To improve the monitoring of the performance of the health sector it is necessary to integrate data and include indicators from all sources and management systems. Currently it can be done manually (but is not) and by spending considerable amount of time as it would imply collecting and handling data from different disintegrated systems. The only sustainable solution would be to get all systems linked electronically into one integrated database, and thus allow for crosscutting exchange of relevant data. FAMS Hospitals Private sector Human Resource HMIS LMIS ASSETS NGOs etc Mission Hospitals Page 37 of 118

38 SWOT Analysis Strengths The HMIS is basically functional and compares favourably with many African HMISs. It produces and reports on data according to a defined indicator set and uses it to make annual plans right down to facility level. Good 1996 plan with excellent basic principles This is an excellent plan that deserves to be fully implemented. The principles of the plan namely DART, uniform pipeline, links to NHSP - are good and should be followed in any future HMIS. Data Pipeline and staff in place Data is collected from most facilities and sent in a systematic way to districts, provinces and national level. A great deal of data exists that could potentially be used to monitor health priorities. In spite of rapid turnover, most staff appears able to use the tools to collect and collate data. DIOs are able to enter data onto the computers and to report it to higher levels. Indicator dataset used The national indicator set is well defined and there is a related dataset linked to it that is well known and used on a regular basis for action plans, performance assessment and supervision. There is regular feedback reports produced through performance assessments and financing is linked to reporting through the selected quarterly indicators Procedures manuals developed and used The Procedures Manuals from 1996 are in place at all levels and are used by new HMIS users as reference files. The manuals are, however, not updated for all new features introduced since the commencement of the 1996-HMIS. Weaknesses There have been serious slippages in terms of implementing the 1996 plan due to poor support, limited funding, weak capacity development, a low demand for quality information (particularly related to monitoring MDGs and the NHSP) and poor ICT use. Lack of HMIS leadership stagnation and frustration The HMIS has been severely neglected at higher levels and lacks top level leadership and proactive management. The steering committee has not materialized and the proposed review process has not been implemented, with the result that the indicator set is set not aligned to MDGs and new challenges such as HIV, STIs and TB. Vertical programs have become frustrated with the HMIS and have set up parallel reporting systems that are bypassing service managers, reporting directly to program coordinators. Page 38 of 118

39 However, these coordinators do not coordinate with other management structures and do not effectively share information. Limited capacity development program The capacity development program was initially implemented, but since 1998 no systematic training has occurred at any level. Facility level staff fills in the forms, often without understanding them and report by rote without analysis. They do limited self assessment and minimal analysis of information. a. DHIOs demotivated, underskilled DHIOs, who should be the cornerstone of the HMIS, do not have establishment posts and have no career structure. They have usually not received adequate training and as a result are underskilled and demotivated, and want to leave for greener pastures. b. Lack of training of new HMIS staff There is no formal training program in place for new DHIO staff. They are generally just orientated about their assignment and the functionality of the HMIS. Poor use of information There is a vicious cycle of poor quality data which is not trusted and comes only quarterly so can not be used for operational management, with the result that data quality gets worse! a. Quality control is weak and little emphasis is put on ensuring data quality at any level. Feedback is infrequent and not very useful. b. Program and service managers are not using the DHIS to analyse or interpret their information and are not getting adequate program-specific information. c. Policy makers and donors receive quarterly data of dubious quality on which to approve funding. This is not properly analysed and fund disbursement is usually rubber stamped, without critical interpretation of data with regard to performance of services. Poor ICT infrastructure ICT is weak outside the central ministry and the computer and equipment that is available is underused because of a lack of technical and managerial skills. Internet and the worldwide web are hardly available in the provinces and are generally not used to transmit data. The software program is rigid, hard-coded and inflexible and can not be adapted locally to suit local needs. This has led to poor integration, weak analysis and interpretation of data and means that data is not available at one point at any level: to get information one has to beg vertical program coordinators whose databases are not coordinated or linked to management. Page 39 of 118

40 Opportunities Return to original HMIS principles With minimum effort and disruption of the current system, the original 1996 principles of DART, single information pipeline, integration of vertical systems into an HMIS umbrella could be easily achieved. In each district a District Information Office should be established, where all data from all subsystems relating to the district is electronically available in one place, and from where all data is disseminated. A Regular review process should be set up, led by a representative steering committee that will provide appropriate leadership of the HMIS. This should be complemented by strong, decentralised information management that focuses on action-led, output-oriented performance assessment that responds to needs and provides real incentives for well-performing districts and facilities. System development The HMIS needs to be reviewed in the light of recent developments, other subsystems strengthened and links created between the different subsystems. 1. Revised indicator set and minimum dataset linked to MDGs. The national indicator set needs to be slightly modified to fit in with MDGs and emerging priorities, a dataset with definitions developed from it. Provinces and districts should be encouraged to develop locally appropriate indicators and targets to suit local needs and capacity. 2. The HMIS needs to be revised This revision, which should be done with minimal disruption of the current system, should be based on the current needs of managers at all levels and will require revision of reporting frequency to facilitate local action and to satisfy some valid vertical program requirements. There should be increased emphasis on annual record reviews to get disaggregated data such as age, gender, rather than burdening the routine system with this requirement. Management of the HMIS needs to be strengthened and clear job descriptions developed that will form the framework for skill-based training.. 3. Completion of planned subsystems Subsystems for Hospitals, Human resources, Community HMIS, finance, assets, drugs need to be assessed and plans drawn up for action as part of the MoH 5 year strategic plan The MoH should liaise closely with CSO to make the census figures more reliable, strengthen the vital registration system and to ensure wide dissemination of the DHS 4. Integration of all administrative and management systems with HMIS Administrative and management systems like e.g. FAMS, LMIS, human resources, assets, need to be linked to the routine HMIS to facilitate uncomplicated reporting. Page 40 of 118

41 Capacity development program This is the key priority to get the HMIS moving, and a major capacity development program is needed for all levels of the system that should include appropriate, adult-oriented management, epidemiology, statistical and technical training at all levels for all users of the HMIS. a. In-service DIOs, health service and program managers, facility staff b. Pre-service DIOs, district managers, clinical officers Strengthened Information, Communication Technology ICT is weak outside the urban centres and there is a need to bridge the digital divide using modern ICT a. The current database needs to be overhauled to make it flexible, responsive and linked to other data sources. b. Districts and provinces need to set up appropriately equipped and staffed information centres with adequate ICT in the form of dedicated computers with good and internet connectivity. These could serve as general MDG support centres, linking other MDG-related ministries and NGOs. c. A Web-based data warehouse needs to be developed that will be able to link information from multiple data sources and platforms and integrate them at one point. This could also easily be expanded to support other organizations monitoring MDGs. Improved use of available information A great deal of data already exists both at the centre and peripherally that could potentially be used to monitor MDGs and the NHSP from the most peripheral level. With improved coordination, better information management, strengthened ICT and increased demand by managers for better quality information, existing data could relatively easily be converted into useful information and disseminated widely. The National M&E unit needs to be proactive in documenting and electronically sharing information between services and sectors. At the core of this improved use should be an improved local research capacity focused on improved performance of the system simple health systems research and action-oriented analysis that will stimulate the information culture and create a demand for quality data that can be easily converted into information that is useful for planning, policy making and monitoring of activities. Threats Loss of DIOs and DMSs DIOs are uniformly looking for greener pastures and say they would leave if they had the opportunity. They do not have establishment posts, have not had adequate training and are generally not skilled to do the job. They feel neglected and abused. Unless this situation is reversed, this crucial cornerstone of the HMIS will be continually eroded and the HMIS can not hope to function. Page 41 of 118

42 Vertical program inflexibility Many vertical programs are driven by powerful outside donors who have specific information needs and inflexible reporting requirements. They also have reliable disbursement of funds so that MoH staff tend to emphasise these programs rather than their routine MoH activities. While it is often impossible to totally change these requirements, high level pressure needs to be brought on these programs to coordinate information needs and minimize information overload on peripheral health workers. Every effort must be made to ensure that indicators are harmonized and both gaps and overlap minimized. Unless this happens, there will continue to be a donor-dominated information agenda with multiple reporting channels, with overlaps and gaps and overload on the front-line data collectors. Political commitment to HMIS change HMIS and the development of an information culture have had minimal effective political or financial support over the past 10 years. A change of philosophy and approach to ensure the use of quality information to make HMIS performance-oriented will require strong national leadership and political direction from the highest level within the ministry. If this does not happen, it is unlikely that the situation will improve. GRZ Funding While donor funding is a short-term stopgap, it is essential that the GRZ MoH commits adequate long-term resources (human, financial and equipment) to strengthen HMIS at all levels. Page 42 of 118

43 Key strategic issues The following strategic issues revolve mainly around improving quality of data and information dissemination by improving and integrating systems and strengthening capacity of staff to implement the HMIS and monitor the MDGs and NHSP 1. Capacity development Capacity for implementation of the HMIS is critically limited at all levels of the system, amongst HMIS staff who are supposed to produce analysed information, amongst managers who should use the information at the level of facility staff who produce the raw data. Training was started in 1998 and one manual was produced, but there has been little implementation of the training program envisaged, though some ad-hoc induction training has taken place. This is due to poor management, inadequate funding and lack of commitment to the HMIS by government and donors alike. The result is that, while the basic data capture and reporting skills are present, there is little attention to data quality and staff is not skilled in self-assessment or the epidemiological thinking needed for the analysis, interpretation and use of information for action. There is thus an urgent need for a comprehensive training program for people already in post and those studying to become health workers (undergraduate and post-graduate). Links need to be set up with other African countries that offer appropriate HMIS-related short courses at universities, technical colleges and training institutions. Courses should be run incountry where possible, but key trainers and decision makers should be sent for short courses appropriate where possible. Manuals A set of appropriate and action oriented HMIS manuals is needed to be developed as the basis for training at all levels. In-service training programs Programs need to be developed with clear curricula, training methodologies and exercises for: Facility level staff for data collection, quality control and self-assessment, HMIS staff at all levels to ensure quality data collation, effective data processing, appropriate analysis and interpretation and intelligent feedback, dissemination and use of information, Program managers at national, provincial and district level to use information in the monitoring of program activities and epidemiological thinking. Pre-service training Pre-service training in HMIS was proposed in 1996 but never implemented. HMIS curricula need to be developed and courses integrated into pre-service training for Monitoring and evaluation officers (Data management specialists and DIOs) at provincial and district level before they take up their posts. Incentives for Page 43 of 118

44 participating in the Institute of Management and Information Systems (IMIS) course need to be investigated District Health Managers at the District Health Management Course at Kabwe and the Health Management course at Chainama hills Undergraduate Curricula on HMIS principles and basic epidemiology should be introduced into the courses for nurses, clinical officers, doctors and other health professionals. This will need a considerable and ongoing commitment from MoH and donors, as well as proactive advocacy and management of the training program. Sustainability of systems, procedures and staff In order for systems to be sustainable, procedures should not be static, as has been the case with the current HMIS, which procedures have not adapted to changing environments and requirements. Procedures should be based on directives and information needs of managers, policy setters, etc. Managers should work with procedures and suggest changes according to internal and external requirements. A set of flexible and regularly updated guidelines and procedures manuals should be developed that also take technical changes into account. Although all changes have to be disseminated immediately the manuals should be technically flexible in the sense that only outdated parts should be replaced by new procedures, and not the whole manual. All users should have access to all and at all times updated manuals through the web and should be allowed to download and to produce and print their own updated versions. Users should be trained properly in how to search information, how to keep manuals updated, and how to introduce local procedures without corrupting the general uniform procedures. This needs proper training of staff - both HMIS and technical staff. 2. Implementation of DART principles The 1996 HMIS plan 1 has a number of excellent basic principles which have not been properly followed. These DART principles need to be revisited if the HMIS is to be effective. A few others such as the information pyramid and the concept of the district information centre should be introduced. Decentralisation A key to good district management is the presence of a functional and robust DHIS that enables local monitoring and analysis of coverage and quality. Data analysis and selfassessment should be carried out at the level where data is collected and information should be used for decision making and action at that level. Data should be collected for local management and not merely for upward reporting for higher levels and to ensure that donors pay out money. This will need revision and simplification of reporting and analysis tools to suit the needs of workers 1 Health Management Information System: Design and Implementation Plan for a DART- HMIS by HMIS Unit 1996 Page 44 of 118

45 Local managers and program coordinators have the right to demand high quality data from the routine HMIS and that adequate resources are allocated for this purpose. Action oriented Data should be collected for local action, self assessment and decision-making, not for filing. The HMIS should collect information for action according to the information pyramid and there should be a clear differentiation between the scope of decisions taken by each level. 1. Health Management Boards require operational information for day-to-day management and supervision; 2. Different health information needs also exist for the community, health post, health centre, hospitals as well as the Regional Boards of Health. 3. Specialised programs have some information needs which fall outside the boundaries of the routine HMIS. These should be collected (and paid for) through integrated use of sentinel surveillance, surveys and other techniques, rather than burdening the routine system. 4. Central Board of Health requires information for longer term strategic management and support, and for setting national policy. Well performing units should be rewarded and poorly performing units given technical and managerial support to improve skills and infrastructure Responsive Data collection should react to changing needs, with data reported in an appropriate timeframe according to its use, and be flexible in terms of adaptation to changing local needs. This responsiveness should be ensured by a high level internal steering committee, e.g. the Monitoring & Evaluation Subcommittee, and regular reviews of the HMIS involving all role players that ensure that each level collects standardised data in a flexible, locally empowering way. The current HMIS system is technically inflexible and has not been regularly reviewed; as a result it has become stagnant and rigid, unable to incorporate new challenges such as HIV/AIDS, the Millennium Development Goals the NHSP etc. Failure to respond to needs of program managers is a major reason why vertical programs have been forced to adopt parallel systems to collect timely and useful information not provided by the HMIS. Transparent A see-through system was envisaged where obtaining information would be easy and dissemination facilitated by the newly created Provincial and National Resource Centers. All stakeholders should be able to easily access anonymous analysed information on key basic programs and the public should be honestly informed of progress and achievements of the health sector through regular press releases and annual publication of comprehensive reports that critically analyse information and give a realistic picture of successes and constraints. In addition, correlation of data collected by the various subsystems will be greatly facilitated by an integrated, centralized and web-based data warehouse to which health workers, managers and the public should have selective access. Page 45 of 118

46 3. Information and communication Technology strengthening Though computers exist at most provincial and district offices, there is an urgent need to improve the use of modern Information and Communication Technology (ICT) at all levels through provision of equipment and developing capacity of staff. Key ICT issues include Revision of the HMIS database to make it into a modern, flexible, and user-friendly system that can be locally adapted to accommodate current information needs (data elements and indicators) and be easily locally adapted to suit future needs. There is no need to start from scratch and there should be an adaptation of existing internationally functioning programs to the needs of Zambia. Strengthened information centres with adequate and internet communication at districts, hospitals and provinces. Appropriate options should be investigated, including landline telephone, satellite dishes (mobile or fixed) and the radio-based bush mail service provider Development of an integrated web-enabled data warehouse to enable system integration and access at all levels, however without compromising anonymity and security of sensible data. Improved interactive Web-site for the ministry at all levels, to facilitate uploading reports, documents and other feedback and to ensure controlled access to information. Develop an ICT policy for the Ministry of Health, including security policies and electronic medical record standards. 4. Effective use of information While the information pipeline is well established and plentiful data exists there is minimal interaction with the data by health staff and little of this is turned into useful information for decision making, performance assessment or resource allocation. Front-line health workers are overburdened with collecting excessive data that is not used and managers receive inappropriate, and therefore useless, information. Most workers and managers do not have adequate information skills to perform tasks required of their jobs. There is an urgent need to 1. Revise the national indicator set to align it with the NHSP , international MDGs, priority programs and key performance indicators of outputs for each level. 2. Reduce the number of data elements reported in line with these indicators. This will in turn reduce the data collection workload of front-line health workers and the amount of data reported. 3. Review the periodicity of reporting so that information that needs monthly action is reported monthly, and information needing less frequent action is reported less frequently. 4. Complete the interfaces between HMIS and the subsystems originally proposed in the 1996 plan but not implemented namely Finances, Procurement, Human Resources and Assets. Theses interfaces should be developed and as a result the subsystems integrated with the new HMIS database. Page 46 of 118

47 Integration of vertical systems There are currently a number of vertical systems collecting and reporting data in systems that are parallel to the National HMIS and have no links to it. These include systems for vertical programs (Malaria, ARVs, and Disease Surveillance etc) as well as management systems (FAMS, LIMS etc). This results in duplication of efforts by front-line staff, overlap of information as well as gaps in the collection of essential information. The reasons for this are numerous, but mainly relate to the inability of the HMIS to provide adequate and timely information that suit the needs of the various stakeholders. The situation needs to be urgently addressed through detailed discussions with the various stakeholders running these parallel systems, to identify their information needs and indicators in use and to see how these can be harmonised with the Routine HMIS using the information pyramid principle development of interfaces between these parallel systems and the HMIS to incorporate key indicators in the routine HMIS and the data warehouse. Research capacity There is room for improvement of capacity for research, including basic epidemiological investigation. All levels need improved skills in simple participatory action research methods to enable health workers and managers to monitor progress towards the MDGs, to use data from sentinel surveillance sites better, to participate actively in surveys conducted at the periphery, to investigate disease outbreaks, to study epidemiological phenomena and to improve the critical content of routine reports. There is need for improved skills in report writing and development of standard formats, or where standards formats exist, a review of these. Some formats may improve by including narrative summaries, in stead of or to supplement tables. Examples of Practical local research would include Annual district-level reports on progress towards MDGs Analysis of data from sentinel sites, disease surveillance reports etc. Record reviews to cross-check routine data quality and to get more disaggregated data on gender, age etc. Improved projections of Population data linked to headcounts Improved analytical content of the annual reports Assessment of output and performance by priority health programs. Improved coordination Improved information use requires that coordination be strengthened vertically (between levels of the health services) and horizontally (with other stakeholders, CSO, health- related ministries and NGOs) Information from HMIS should flow seamlessly between users with overlap and duplication eliminated and the gaps reduced. This will be greatly facilitated by application of the information pyramid principle. Page 47 of 118

48 This will require a strengthened M&E unit of MoH to provide necessary leadership and technical skills and a top level M&E steering committee with involvement of all stakeholders including district and provincial staff. Coordination of surveys such as the DHS, facility survey, Service availability mapping and other future surveys should be improved in future by ensuring that all surveys pass through the national M&E steering committee. Feedback and dissemination Because of high staff turnover, poor pre-service training in HMIS and a generally low level of information skills, regular Performance Assessment is essential for staff to get orientation and feedback. Feedback from performance assessments, though it is done often and at all levels, is generally of poor quality and is often not in an appropriate format that provides appropriate information. Feedback on results of national surveys is weak and peripheral staff is often used as data capturers but not given the results of the surveys, microdata or even overall reports. For example, The quarterly summary of selected HMIS indicators, on which donors base their funding, is merely a list of tables showing quarterly indicators, and appears not to be adequately analysed or checked, though large amounts of money are paid out on the basis of these results, The SAMS and JICA surveys were conducted over six months before the HMIS assessment of July 2005, and no province had received even preliminary reports. The M&E unit should ensure that feedback of information is improved through improved documentation and coordination of surveys and research results, routine HMIS reports and an improved website that makes uploading of data easy, ensuring that survey reports and publications are distributed to district level, both paper-based and electronically, press releases published, and electronic versions uploaded to the website. 5. HMIS staff retention Staff retention is a problem throughout the MoH, affecting all cadres at all levels. District Health Information Officers are seen as the cornerstone of the HMIS, as without them the entire system will collapse. While most districts and all provinces have information officers / data specialists in place, there is an unacceptably high turnover of district health information officers, and the officers in place are demotivated, under-skilled and are all looking for greener pastures. Extensive discussions with the DIOs revealed a great deal of dissatisfaction based around lack of recognition and authority within the DHMT, poor skills for analysis and interpretation of information and lack of established posts. Lessons need to be learned from previous effective retention schemes (e.g. for doctors) that have improved job security, working conditions, career pathways and recognition. Page 48 of 118

49 What is needed appears to be a pilot of the health staff retention plan to include DIOs based on Creation of established DIO posts with clear job descriptions and career prospects. Adequate training for the required tasks, including computer and network skills, epidemiology demography and statistics skills. Appropriate tools such as improved internet access. The Way Forward The MoH has to see improved information as a top priority and set up and support a strong, decentralised program to strengthen data quality and increase use of information in management decisions and monitoring of performance at all levels. There are many subsystems that need to be addressed in order for the overall monitoring of health system performance to be improved. Some of these will involve working with CSO to improve the relevance and accuracy of the census, strengthening the vital registration system and ensuring widespread dissemination of household survey results. Others will involve established sub-systems within the MoH such as the integrated Disease Surveillance, as well as very weak systems such as hospital information system, administration, human resources and the community. This report is concerned mainly with strengthening the Routine HMIS reporting of health service data and its conversion into indicators that can be used for local management of services and improvement of quality of care to patients. The new HMIS project should be founded on the 1996 HMIS plan with the principles of Decentralisation, action Orientation, Responsiveness and Transparency at its centre, and incorporation of the information pyramid and integration of vertical program needs This will require the following outputs, further elaborated in the action plan. 1. Revised HMIS While acknowledging that the data pipeline is in place and that the HMIS is basically functional, there is need for a substantial revision to return to the 1996 DART principles and to adapt to the new challenges of the past 10 years Decentralised action by managers and output-oriented self-assessment Action orientation, based on local decision making to monitor the NHSP and achieve MDGs; simplified tools for data collection and analysis. Responsive to needs of all health workers and priority programs for the NHSP and able to change as the needs arise. Transparent data collection and dissemination and controlled access via a web-based data warehouse; Review of policies and legislation This revision should be done gradually and with minimal disruption to the existing positive aspects of the HMIS, but with a clear focus on improving monitoring of MDGs and local priorities. In order to maintain a balance between local flexibility and international Page 49 of 118

50 standardisation, it is necessary to collect data in a way that ensures a minimum set of information for each level, with data from the lower levels (community and facility) including all data that is mutually agreed as being necessary for higher levels (national and international) as in the diagram below This will require development of a revised integrated national set of indicators so that NHSP targets and international MDGs are regularly analysed by all levels. Relevant national and provincial level indicators will be added that adhere to international standards, while encouraging local flexibility, resulting in a hierarchy of standards. Information pyramid different information needs for different functions Indicators, Procedures, NHSP MDGs datasets & International IS use of Information for ACTION: Monitoring National Inf. Systems of Health Community Local Action Operational Planning Provincial Information Systems District Information Systems Community Information Systems District Province National International Hierarchy of standards, where each level has freedom to define their own standards as long as they align with the standards at the level above. In this way, through ongoing discussion around the minimum indicator set, each level maintains the flexibility to collect data it considers important, and every level will have essential information required for decision making and action reported to it. Using this same pyramidal approach, action for important but neglected subsystems such as Hospitals, community information, administration and human resources need to be investigated and action plans developed. 2. Capacity development Human Resource capacity must be strengthened with a comprehensive training program to promote a culture of information for HMIS staff, program managers and facility health workers at all levels. This should include both technical skills and the DART information principles all levels and all categories of health workers, including district, program and facility managers a focus on in-service skills development for health and information formal in-service and pre-service courses, undergraduate and post-graduate training involving universities, training colleges and technical colleges. Best practice study tours and courses both within Zambia and in other African countries Page 50 of 118

51 3. Improved ICT Appropriate ICT for the 21 st century is desperately needed for Zambia MoH, to enable improved communication between levels and dissemination of information outside of the health sector. The focus should be on a modern and flexible ICT platform revision of the HMIS database to make it user-friendly, flexible and able to integrate needs of new challenges and the vertical programs strengthening of district and provincial information centres with computers, internet and access Setting up an integrated data warehouse that will allow access to relevant healthrelated data. 4. Increased use of information Use of information for action is essential at all levels in order to improve quality of data and to stimulate critical self-assessment. The MDGs and NHSP should be the core focus of all assessment, with regular analysis and self assessment around them. Some ways of increasing use will include promoting local research skills to increase local in-depth analysis of existing records encouraging annual district-based progress reports for NHSP , MDGs and other priorities, based on record review and local surveys, improved use of data from Sentinel sites to get quality routine data and up-to- date analysis at local level improved dissemination and feedback of national level surveys Page 51 of 118

52 Annex 1 Terms of Reference TERMS OF REFERENCE Strengthening of HMIS for poverty reduction monitoring BACKGROUND Together with the National Transitional Development plan, Zambia s PRSP should form the centrepiece of Government policy, The preparation of the successor policy framework, the National Development Plan, is currently ongoing. It aims at fostering economic growth while at the same time having a direct impact on poverty. It is closely linked to HIPC, since Zambia s external debt has been the dominant feature of its economy for many years and HIPC relief is essential to macroeconomically sustainable financing of essential public services. Improving public finance management features highly in it, and is a central concern of co-operating partners in Zambia. The WB-supported Public Expenditure Management and Financial Accountability Review (PEMFAR) of 2003 provides a framework shared by Government with several cooperating partners for developing a consolidated reform programme. The European Commission-funded Poverty Reduction Budget Support (PRBS 1) aims at supporting the implementation of poverty reduction policies. Additional funding for the budget helps to provide the financial resources for essential public services, since policies can only be implemented with continued external financial support. Providing funds through the budget allows the Government to unify planning and to prevent the duplication, uncertainty and delay that characterise donor-led and - implemented projects. PRBS01 provides 110 million in direct non-targeted budget support for Zambian fiscal years 2004, 2005, and In addition, 4 million are availed for poverty reduction monitoring, including specifically support to Ministry of Health and Education for strategic planning and results monitoring. Support to HMIS is seen as a way to strengthen poverty reduction monitoring through the reinforcement of one of its essential building blocks. Particular emphasis is to be placed on harmonisation of this support with other ongoing or planned efforts on management information systems in Education and Health, and under PEMFA programme, whether Government-supported of externally sponsored. Reliable and timely information is an essential foundation of public health action. If sound health statistics are not available, decision makers are unable to identify problems and needs, track progress, evaluate the impact of interventions, and make evidence-based decisions on health policy, programme design, and resource allocation. MoH and CBoH have during the health reform implementation since the 1990s developed systems that have contributed to better management of the health sector such as Financial and Administrative Management System (FAMS), Health Management Information Systems (HMIS) and tools for contract management. Zambia therefore does have a functioning HMIS based on routine facility-based data on health status. Page 52 of 118

53 It was launched in 1994 and designed to cover the entire district health care system. Country-wide coverage (all 72 districts) has been ensured for at least the past 5 years. In about 85% of the districts computerised systems for data aggregation and analysis are operational at the district level. Parallel systems for vertical programmes or support services co-exist. Recently, the GFATM was added. The activities of the GFATM are now pretty much included in the district plans but they keep their own requirements for district level reporting. A separate hospitals HMIS is also developed but hospitals are desperately in need of having computerised systems in place. Private facilities are not captured. Staff in charge of the HMIS are not appropriately trained for their tasks, not least due to high-turnover. Quality of data is preoccupying though not alarming according to the recent MTR of the National Health Strategic Plan. Therefore, providing accurate and comprehensive measurement of health status and health service coverage still presents a series of significant technical and operational challenges. Government aims at strengthening a HMIS firmly founded on sound empirical data, adapting existing and possibly developing new measurement tools, apt for use in resource-constrained settings. The CSO faces serious resource limitations that have affected the production of statistics for quite some time. The CSO remains under-resourced and overly dependent on donors uncertain support for its operations. At the national level, data sharing and coordination of the overall statistical system are not adequately in place. The Census and Statistics Act 1964 establishes the legal basis for the CSO to compile official statistics. The Statistics Act mandates the Director to organize a coordinated scheme of social and economic statistics relating to Zambia. However, there is no systematic program to coordinate the production of official statistics. In practice, coordination is undertaken primarily on a bilateral basis with other government agencies. Resources: Staff and computing resources are not adequate for compiling a coherent set of timely national accounts statistics. The CSO s ability to recruit and retain staff is weak due to the uncompetitiveness of the remuneration and the unavailability of financing. Funding for the statistical program has not been secure and has weakened the agency s ability to undertake planned activities. The CSO works with the Southern Africa Development Community (SADC) to implement international standards that should better serve users needs. The CSO invites comments on statistics during meetings with users, such as the monthly press conferences. However, the relevance of the statistics to users in the private sector is not assessed in a systematic manner. The CSO s mission statement recognizes quality as the cornerstone of the agency s activities. However, resource constraints have limited its ability to conduct evaluations Professionalism: The recruitment and promotion of staff follow the regulations of the public service. Staff are appointed, and can only be dismissed, by the Public Service Commission. Given that the CSO is a department of the MOFNP, the Census and Statistics Act stipulates that the Director shall collect statistics whenever the Minister directs, which implies that the Director may not have complete independence on statistical matters. However, in practice, the CSO operates independently. Decisions Page 53 of 118

54 on sources of data, statistical techniques, and dissemination are determined solely by statistical considerations. Transparency: The terms and conditions for compiling the statistics are contained primarily in the Statistics Act. DESCRIPTION OF THE ASSIGNMENT Design of Assignment This consultancy will be a first phase that will look at the needs assessment and preparation of a workplan. Phase two will be the actual implementation of the workplan. Beneficiary The beneficiary of these consultancies will be the Ministry of Health, with its goal to improve the health status of the people of Zambia, in order to contribute to socio-economic development Global and specific objectives The global objective is to improve the efficiency and effectiveness of health care delivery through the strengthening of HMIS, seen as a combination of people, equipment and procedures organised to provide health information enabling concerned stakeholders to make timely and informed decisions and use of information at various levels. The specific objective of the consultancy is to develop a detailed, quantified action plan for a strengthened, revised and comprehensive Health Management Information System, focused on coverage of essential health services and on the health status, but open to integration with other components that may be developed at a later stage. Requested services The Consultant will assess the state of the country health information system, its ability to collect, analyse and generate reliable data on specific indicators, and the extent and timing of data dissemination and use. As indicative reference, a WHO-produced scheme is suggested: Page 54 of 118

55 It will then lead the identification and development of a detailed, quantified action plan that will be based on a consultative process that will build the rationale for HMIS reform, the vision to which Zambia HMIS should aspire and options and strategies to help Zambia turn that vision into a reality The action plan will describe the principles and processes needed including assessment, planning, implementation, monitoring and evaluation, data generation, analysis and synthesis, and data dissemination and use. The plan will include outcomes and milestones at one and two years. What follows are five broad areas to be analysed, and a number of questions related to them, providing a non-exhaustive check-list of issues to be considered, 1. REVIEW THE RELEVANCE OF THE CURRENTLY OPERATIONAL HMIS TO THE GENERAL M&E FRAMEWORK OF THE HEALTH SECTOR. Reconsider the minimum necessary scope of the health information system in the Zambian context. Analyse and prioritise proposals around the various HMIS subsystems (considering that some subsystems are closely linked to on going government-wide processes): Health Status 1. routine service reporting (including clinical, environmental and community-based health actions) 2. Households surveys (Harmonizing surveys and providing a platform for survey coordination), including HIV related behavioural surveillance systems designed according to UNAIDS Guidelines for second generation HIV surveillance. 3. specific programme reporting (e.g. : PMTCT, ART, VCT for HIV/AIDS, DOTS detection and cure rates for tuberculosis, or ITN coverage for malaria) 4. possible approaches to vital registration (births, deaths, and causes of death) Finance and Administration, (including consideration of the place for National Health Accounts) Assets Page 55 of 118

56 Human Resources Drugs and supplies 2. REVIEW THE NDP AND HSSP INFORMATION NEEDS FROM THE HMIS The MoH is embarking into the preparation of the new National Health Strategic Plan just as the Government of Zambia is defining its National Development Plan as successor to the Poverty Reduction Strategy Paper. These policy developments must be the constant benchmarks in proceeding to the definition of an action plan on the HMIS. Aggregate information produced at central level must be redefined and enriched with information on best and worst performers, monitoring equity Consider the institutional setting for the HMIS in a broader context than just health, as building block of the National Development Plan monitoring, in particular this entails considering the role of the Sector Advisory Groups, and at a lower level the District and Provincial Development Coordinating Committees (DDCC and PDCC) and how the HMIS would relate to this process. 3. REVIEW AND DETERMINE CRITICAL INFORMATION GAPS TO THE NDP MONITORING OF THE HEALTH SECTOR, AND MAKE RECOMMENDATIONS ON HOW TO STRENGTHEN THE SYSTEM Evaluate the pertinence, reliability and time invested in the current system of multiple tally sheets and related guidelines, Shift the burden of processing information as much as possible to the computerised level, and design reporting formats that are simple and user friendly Assess the feasibility of introducing a dimension of community health information to combine local knowledge and health related perceptions, care seeking patterns, quality of care perceived by patients and families, coverage of essential services, better identify key determinants of health, and in turn promote responses to community health problems (OVCs, VCT). 4. REVIEW AND DETERMINE THE HMIS CAPABILITIES IN GENERATING GENDER AND AGE RESPONSIVE INFORMATION ON KEY HEALTH INDICATORS AND THE REPORTING FORMATS AND FREQUENCY 5. REVIEW AND QUANTIFY THE INVESTMENT NEEDS FOR STRENGTHENING THE HMIS ON: INFRASTRUCTURE (HARDWARE AND SOFTWARE), AND PERSONNEL (TRAINING, CONDITIONS OF SERVICE, RETENTION ) TO OPERATE THE HMIS Evaluate the current relationship with CSO and other Government agents, in terms of capacity to respond to HMIS needs, effectiveness and timeliness of feedback. Risks that have to be tackled: failure to take into account the social and professional cultures of the health care entities involved and to recognise that education of users and computer staff is an essential precursor. Eg the HMIS initially increases the workload; training focuses on how rather than why. underestimation of the complexities of routine clinical and managerial processes that must feed into the system, also due to complexities specific to the health sector. There should not be any collection of information that is not used. Consensus building: keeping in balance expectations of those who commission the system, the developers, and the staff at the health facilities, in order to avoid Page 56 of 118

57 pushing requests beyond levels sustainable by staff. User perceptions can kill the implementation of an otherwise solid application. Long implementation in the context of fast turnover: task teams and heads of departments are rarely in post for the whole period of project implementation Capacity by management and project to step back and take a dispassionate view of the cost effectiveness of the interventions being introduced. 6. REVIEW AND DETERMINE MECHANISM FOR QUALITY ASSURANCE AND CONTROL OF THE HMIS, PERIODICALLY Incorporate periodic quality control supervision (data quality audit) and possibly household surveys, to assess reliability of data. Build feedback information into the system, in order to provide comparison, reinforce conformity thus data reliability. operate in the perspective that by 2011, the WHO Health Metric Network comprehensive Consensus Technical Framework (CTF) will be the universally accepted standard for the collection, analysis, reporting and use of health information by all developing countries and global agencies Expected results A comprehensive, detailed, quantified action plan ready for implementation in line with EDF procedures (ie taking in full account administrative requirements), based on a three-years timeframe and assorted with a capacity assessment and possibly proposed capacity inputs. EXPERTS PROFILE The team of experts should have at least 10 years of experience in the field of health systems in developing countries and significant experience in Least Developed Countries in Sub-Saharan Africa, and should include: one expert in epidemiology and statistics (Team Leader), having complementary specific experiences one expert in HMIS and Public Health with knowledge in IT one institutional/organisational development expert Working languages : English. Eligibility: must be EU citizens or citizens of one of the Africa-Caribbean-Pacific countries signatories to the Cotonou Agreement of 2000 LOCATION AND DURATION Starting date : not later than 22 nd June, 2005 Finishing date of the assignment : indicatively end of August (final report available) Schedule and number of days for the assignment per expert Task Location Team leader / stat-epidemio HMIS expert Inception mission Lusaka Presentation of Lusaka inception report Field visits Zambia Organisation development Page 57 of 118

58 Lusaka-based Lusaka activities Draft report base preparation National workshop Lusaka Final report base TOTAL REPORTING Content, language, format and number of reports: Inception report after day-7 max 20 pages. Final: o Situation analysis max 50 pages (plus annexes, including field missions findings) with max 5 pages executive summary; o Proposed Action plan: max 50 pages (plus annexes). Date of submission draft final report and action plan: indicatively 26 July 2005 Date of consensus-building National Workshop (to be organised by consultants in Lusaka): indicatively 19 August 2005 Presentation of final report and action plan: indicatively 31 August 2005 Stakeholders to be consulted: MoH, CBoH, NAC, CSO, PHOs, selected DHDs, selected level 1 and 2 hospitals, co-operating partners, NGOs, CHAZ, Documents to be consulted: References 1 CBoH., Quarterly Statistical Bulletin, First Quarter 1999, May, CBoH., Quarterly Statistical Bulletin, Third Quarter 1998, November, CBoH, HMIS Technical Review Meeting Report, March, CBoH, HMIS Annual Statistical Bulletin, July, CBoH, HMIS Annual Statistical Bulletin, July, CBoH, Quarterly Statistical Bulletin, Second Quarter 1998, August, CBoH, Progress Report 1998 and Plan of Activities, December, CBoH, HMIS Annual Statistical Bulletin, November, Chishimba P, & Musama K., HMIS Data Management and Usage Guide - A Draft, August, Chishimba, P., The HMIS: From Data to Decision-making December, Chishimba, P & Sigande E.,Compilation of VBA Source Code for the HMIS Data Validation, March, Chitengu R. & Chishimba, P, A Workshop Report on the Consensus Building for the HMIS Data Management and Usage Guide PAIDESA, February, Church, M., Field Report to Basics, May, Church, M., Field Report to Basics, February, Church, M., Field Report to Basics, November, Church, M., Field Report to Basics December, Church, M., Field Report to Basics, September, Church M & Simwanza A., Establishment of Information Culture: Achievements Page 58 of 118

59 and Challenges in Zambia, March, Heydelberg E. etal, Review of HMIS/FAMS/ITG Cascade Training - a Mission Report, September, HMIS Development Team, HMIS Plan and Status Report for 1997 Qtr3 to 1998, October, HMIS Development Team, A Report on First Round, HMIS/FAMS/ITG Field Visits in Pretest Districts, September, HMIS Development Team, HMIS Newsletter, November, HMIS Development Team, Design and Implementation Plan for a DART-HMIS, November, HMIS Development Team, Review of the Field Assessment, HMIS/FAMS/ITG Pretest Phase, October, HMIS Development Team, HMIS Consultative Workshop, Senenga Safaris, January, Koot J & Church M., HMIS Annual Statistical Report for , June, Koot, J., HMIS Review Discussion Paper, March, MOH, National Health Strategic Plan , May, Mwendapole A & Chishimba, P, A Workshop Report on the Training of Data Management Specialists, Barn Motel, December, Nsonga B. & Chishimba, P, Tips and Tricks of Data Validation in HMIS, February, Pembamoto B.G., The Systems Requirements Document, September, Simwanza, A, An Overview of the HMI,S July, Tembo M. M., A Synthesis Report of Various Comments on the HMIS Data Management and Reference Guide, October, Young, A., A Report on Various Meetings Regarding HMIS in June of 1997, June, 1997 Page 59 of 118

60 Comments on the Terms of Reference and Key Issues (presented to the Steering Committee during the Inception Report Meeting on 13 th July, 2005) The Terms of Reference (ToR) clearly describe the assignment in relation to objectives, requested services and expected results. The first step in the development of any information system is to define the expected output, which will then determine e.g. human resources, equipment, and procedures needed to carry out the processes and the last step the input to the system. According to the ToR the main focus is to update the health management information system (HMIS) for strategic purposes at national level (monitoring of the National Development Plan and the Health Sector Strategic Plan). However, we assume that the local level will continue to be the prime users of the information system for data analysis and planning, as is the approach in the current HMIS, and in line with the continuing health reform decentralization process. It should further be noted that the ToR formulates the concept of HMIS broader than was originally laid down in the Design and Implementation Plan from where a distinction between Monitoring and Evaluation, Health Management Information System and Health Information System was made, i.e. A Monitoring and Evaluation System comprises all activities that contribute to increased knowledge of inputs, processes, outputs, outcomes, and impacts of health services. Routine reporting, sentinel reporting, performance audits, surveys, research, and censuses are methods used to collect the information. The primary objective of the MES is to provide information to set policy and to monitor its effects. A Health Management Information System includes routine activities that increase knowledge of inputs, processes, outputs, outcomes, and impacts of health services. Performance audits, surveys, and research are not included. The primary objective of the HMIS is to provide operational information on policy implementation. A Health Information System concentrates on information related to diseases and health status, mainly outputs, outcomes, and impact of health services with little attention for inputs and processes. Medical statistics and disease surveillance belong to a health information system. The primary objective of the HIS is to provide information on the health status of a population; it is used both to set policy and to manage operations. The graphics below illustrates the 1996-rationale behind the design of the current HMIS in place in the MoH. 1 Health Management Information System, Design and Implementation Plan for a DART- HMIS, HMIS Unit, 6 November 1996 Page 60 of 118

61 A HMIS includes routine activities The primary objective of the HMIS is to provide operational A HIS concentrates on diseases Medical statistics and disease surveillance belong to HIS The primary objective of the HIS is to provide information on HMI Input Process Outputs Outcomes Impacts Medical statistics Disease surveilance HI Censuse Reseach Routine reporting Surveys Performance audits Sentinel reporting M& A MES comprises all activities that contribute to increased knowledge The primary objective of the MES is to provide information to set policy and to monitor its The current approach to HMIS in the Ministry of Health (MOH) and Central Board of Health (CBOH) is that of a routine system, focusing on operational issues, while the requested services as laid down in the ToR go beyond this point and include as well elements of M&E as HIS like for instance essential health services and health status. The broader formulation also incorporates other management areas, or subsystems, under the control and management of the MoH and/or CBoH, namely Financial Administration and Management System (FAMS), Assets Management, Human Resources Management and Drugs and Supplies Management. We understand that it is the intention of the MoH to incorporate all systems under the same information system umbrella, as illustrated in the graphics below. MES HMIS HIS FAMS HRMS Drugs & Supplies Page 61 of 118

62 The purpose of a well-functioning HMIS, like other information systems, is to provide decision makers with factual and reliable information on which to monitor operations and to make decisions. It is important to note, though, that one does not necessarily improve effectiveness by strengthening information systems. Other factors like policy, culture, availability of resources will also have an impact. A well-functioning HMIS is necessary for a well-functioning health care delivery by providing factual and timely data on the state of the system, however it is not sufficient as it is only one of a number of health sub-systems, like e.g. logistics, infrastructure, decisionmaking processes, and organizational structure. It should also be noted that even if the decision makers do get timely and reliable monitoring data from the HMIS, they may not be able to interpret the data, draw correct conclusions and make sound decisions. Utilization of data will therefore also need to be focused on in the assessment. It appears from the TOR that the consultant, together with the HMIS-users (or beneficiaries) through the process-oriented approach, should be facilitating the process of reaching broad consensus among the stakeholders on the vision for a future Zambian HMIS and how to turn it into reality. To best achieve this, a very sensitive attitude and approach to both exercises, i.e. the needs assessment and the development of an action plan, is required and it is important to keep in mind that the ultimate goal of the assignment is to build the rationale for a HMIS reform, and not the HMIS itself. We understand that this consultancy is a first phase that includes the needs assessment and the preparation of a work plan and that the following phase, phase 2, will be the actual implementation of the plan. We shall in the work plan provide recommendations for the needed MoH support for strengthening the existing HMIS and implementation of the 3 year plan of action i.e. the long/short term, external/local support, procurement needs, institutional /organizational/structural adjustments, capacity development/training needs, HMIS needs including rationale, vision, strategy and approach in a quantified, budgeted and time related manner. We interpret the recognized need for a revised and improved HMIS that there are weaknesses and problems in the existing system that have to be addressed and solved. Through the assessment the areas for improvement should be outlined and the strategy for the development of an improved HMIS plan adopted. We understand that the expected outcome of the assignment is a plan of action, including broad consensus on the rational for a HMIS reform, and not the HMIS itself. Page 62 of 118

63 Annex 2 Persons met Name Function Mr Nicholas Chikwenya Dr Davis Chimfwembe Mr Rutenda Chitembure Dr Juan Garrey Dr Marco Gerrittsen Donors Coordinator, Ministry of Health [email protected] Director, Ministry of Health chimfwembemakasa@hotm ail.com Senior Statistician, Ministry of Health Health Officer, European Commission, Brussels Senior Health Advisor, Royal Netherlands Embassy [email protected] [email protected] Marco.Gerritsen@MinBuZ a.nl Mr Chipalo Kaliki Central Board of Health [email protected] Mr Thomas Königsfelt Danida Mr Roy Maswenyeho Chief Accountant, CBoH [email protected] Mr Belem Matapo World Health Organization [email protected] Mr Chipo Mpamba Dr V. Mukonka Central Board of Health Director, Central Board of Health Mr Evaristo Musaba Chief Accountant, MoH [email protected] Dr Cosmas M. Musumali HSSP Mr Chrispine Mwiiya Ministry of Health [email protected] Dr Felix V. Phiri Mr Emilio Rossetti Mr Mwembe Sichula Manager, Planning, Ministry of Health Delegation to Zambia, European Union Delegation to Zambia, European Union Mr Christopher Simoonga Assistant Director, Management Information, Ministry of Health EMILIO.ROSSETTI@ CEC.EU.INT [email protected] t [email protected] Ms. Virginia Simushi Ministry of Health [email protected] Ms Maria Skarphedinsdottir Ms Petra Staal Department for International Cooperation Programme Officer, Health & HIV/AIDS, Royal Netherlands Embassy m- [email protected] Page 63 of 118

64 Page 64 of 118

65 Page 65 of 118

66 Page 66 of 118

67 Page 67 of 118

68 Page 68 of 118

69 Page 69 of 118

70 Annex 3 Timetable Date Activity 6 July 2005 Travel to Lusaka 7 July 2005 Meeting with Mr Emilio Rossetti, Delegation of the European Union to Zambia 8 July 2005 Meeting with: Mr Emilio Rossetti, Mr Rutendo Chitembure, MoH, Mr Chipalo Kaliki, CBoH, Mr Christopher Simoonga, MoH Meeting with Voyagers to arrange for the logistics of the field visit 9 July 2005 Working in MoH, on documents review, preparation of Inception Report, preparation of tools for fields visits, preparation of the logistics of the field visits 10 July 2005 Working in MoH, on documents review, preparation of Inception Report, preparation of tools for fields visits, preparation of the logistics of the field visits 11 July 2005 Meeting with Mr Simoonga Meeting with Dr Mukonka Meeting with Dr Chmfwembe Working in MoH, on documents review, preparation of Inception Report, preparation of tools for fields visits, preparation of the logistics of the field visits 12 July 2005 Presentation of the Inception Report to the Monitoring and Evaluation Subcommittee of MoH Preparation of tools for fields visits, preparation of the logistics of the field visits 13-23July 2005 Field visits to All Provinces, selected districts and selected health facilities 25 July 2005 Meeting with Dr Marco Gerrittsen, Mr Simoonga, Mr Kaliki at the Royal Netherlands Embassy Processing of field data, preparation of Assessment Report and Plan of Action 26 July 2005 Meeting with Mr Simoonga, Processing of field data, preparation of Assessment Report and Plan of Action 27 July 2005 Meeting with Ms Maria Skarphedinsdottir, of DfID Meeting with Danida Meeting with the representatives of the MoH Processing of field data, preparation of Assessment Report and Plan of Action 28 July 2005 Processing of field data, preparation of Assessment Report and Plan of Action 29 July 2005 Presentation of a Draft Assessment Report to the Implementation Review Sub- Committee of the MOH Preparation of Assessment Report and Plan of Action 30 July 2005 Preparation of Assessment Report and Plan of Action 31 July 2005 Preparation of Assessment Report and Plan of Action 1 August 2005 Preparation of Assessment Report and Plan of Action 2 August 2005 Preparation of Assessment Report and Plan of Action 3 August 2005 Presentation of a Draft Assessment Report and a draft Plan of Action to the Monitoring and Evaluation Sub-Committee of the MOH 4 August 2005 Preparation of Assessment Report and Plan of Action 5 August 2005 Presentation of a draft Plan of Action to the Implementation Review Sub- Committee of the MOH Travel from Lusaka Page 70 of 118

71 Questionnaire July 2005 Annex 4 Questionnaires During the assignment 3 questionnaires were developed, one for provinces and the national level, one for districts, and one for facilities. The example below is the province/national level questionnaire. The two others are contained in the database that was developed to handle the collected data. Province: Date of interview: Strengthening of HMIS for Poverty Reduction Monitoring, Zambia Health Management Information System Assessment at Provincial/District Level 9 Interview with the Director of the Provincial Health Management Board (or person delegated) Good morning, my name is. I am a consultant to a European Union funded project on Strengthening of the Zambian Health Management Information System for Poverty Monitoring. I am accompanied by an official of the Ministry of Health. The task of our team is to assess the current status of the HMIS in Zambia in order to prepare a plan of action for a substantial strengthening of the existing system. I will appreciate it if you could spend some time with me answering questions related to your perception how in your province health information is generated and used for decision making Let us first discuss the data collected and information available at the provincial level. I will cite 12 indicators pertinent to Millennium Development Goals and two related to health systems. For each of these indicators please describe four aspects. For each indicator, please provide concrete evidence of the most recent assessment document. 9 This questionnaire was modified from: Health Metrics Network, Strengthening Country Health Information Systems: Assessment and monitoring tool. Draft 1.0 July 2005 Page 71 of 118

72 Questionnaire July Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 4: Prevalence of underweight children (under five years of age) MDG Goal 1: Eradicate extreme poverty and hunger MDG Target 2: Halve, between 1990 and 2015, the proportion of people who suffer from hunger How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by age, gender and district (Disaggregation) What method was used to collect data (Measurement method) One None >80% children 50-80% children 30-50% children <30% All three Two One None Survey Record review Routine HMIS None 2 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 13: Under five mortality rate MDG Goal 4: Reduce child mortality MDG Target 4: Reduce by two third, between 1990 and 2015, the under five mortality rate How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by age, gender and district (Disaggregation) What method was used to collect data (Measurement method) One None >80%of deaths 50-80% of deaths 30-50% of deaths <30% All three Two One None Survey Record review Routine HMIS Nil Page 72 of 118

73 Questionnaire July Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 14: Infant mortality rate How often has the indicator been estimated during the last five years (Periodicity) Three or more Two One None MDG Goal 4: Reduce child mortality What is the coverage of data on which the last estimate was based (Representativeness) >80%of deaths 50-80% of deaths 30-50% of deaths <30% MDG Target 4: Reduce by two third, between 1990 and 2015, the under five mortality rate Is data disaggregated by age, gender and district (Disaggregation) What method was used to collect data (Measurement method) All three Two One None Survey (DHS or equivalent) Vital registration / record review Routine HMIS 4 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 15: Proportion of 1 year old children immunized against measles MDG Goal 4: Reduce child mortality MDG Target 4: Reduce by two third, between 1990 and 2015, the under five mortality rate How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by age, gender and district (Disaggregation) What method was used to collect data (Measurement method) Three or more Two One None >80% vaccinated 50-80% vaccinated 30-50% vaccinated <30% vaccinated All three Two One None Cluster survey Health cards review at facilities Routine HMIS None Page 73 of 118

74 Questionnaire July Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 16: Maternal mortality ratio How often has the indicator been estimated during the last five years (Periodicity) Three or more Two One None MDG Goal 5: Improve maternal health What is the coverage of data on which the last estimate was based (Representativeness) >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities MDG Target 6: Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio Is data disaggregated by mother's age and district (Disaggregation) What method was used to collect data (Measurement method) Two Two One None Maternal death audit Maternal mortality register Routine HMIS None 6 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 17: Proportion of births attended by skilled health personnel MDG Goal 5: Improve maternal health MDG Target 6: How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by mother's age and district (Disaggregation) Three or more Two One None >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Two Two One None Page 74 of 118

75 Questionnaire July 2005 Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio What method was used to collect data (Measurement method) Survey Record review Routine HMIS None 7 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 18: HIV prevalence among year old pregnant women MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 7: Have halted, by 2015 and begun to reverse, the incidence of malaria and other major diseases How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by age and district (Disaggregation) What method was used to collect data (Measurement method) Three or more Two One None >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Two Two One None ANC Survey HIV antibody testing HIV+ lab testing HIV clinical case reporting None 8 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 19: Contraceptive How often has the indicator been estimated during the last five years (Periodicity) Three or more Two One None Page 75 of 118

76 Questionnaire July 2005 prevalence rate MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 7: Have halted, by 2015 and begun to reverse, the spread of HIV/AIDS What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by gender, age and district (Disaggregation) What method was used to collect data (Measurement method) >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Three Two One None Household survey Record review Routine HMIS None 9 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 21: Prevalence of malaria MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 8: Have halted, by 2015 and begun to reverse, the incidence of malaria and other major diseases How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by gender, age and district (Disaggregation) What method was used to collect data (Measurement method) Three or more Two One None >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Three Two One None Household survey Record review Routine HMIS None 10 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 23: Prevalence of tuberculosis How often has the indicator been estimated during the last five years (Periodicity) Three or more Two One None Page 76 of 118

77 Questionnaire July 2005 MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 8: Have halted, by 2015 and begun to reverse, the incidence of malaria and other major diseases What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by gender, age and district (Disaggregation) What method was used to collect data (Measurement method) >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Three Two One None Household survey Record review Routine HMIS None 11 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 24: Proportion of TB cases cured under DOTS MDG Goal 6: Combat HIV/AIDS, malaria and other diseases MDG Target 8: Have halted, by 2015 and begun to reverse, the incidence of malaria and How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by gender, age and district (Disaggregation) What method was used to collect data (Measurement method) Three or more Two One None >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Three Two One None Survey Record review Routine HMIS None 12 Criteria High (3) Mid (2) Low (1) None (0) Score MDG Indicator 46: Proportion How often has the indicator been estimated during the last five years (Periodicity) Three or more Two One None Page 77 of 118

78 Questionnaire July 2005 of population with access to affordable essential drugs on a sustainable basis MDG Goal 8: Develop a Global partnership for development MDG Target 8: provide access to affordable essential drugs in developing countries What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by gender, age and district (Disaggregation) What method was used to collect data (Measurement method) >80%of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Three Two One None Survey Record review Routine HMIS None Let us now consider two indicators relevant to the health system 13 Criteria High (3) Mid (2) Low (1) None (0) Score Health system Total health expenditure per capita How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by district, health institution and financial source (Disaggregation) What method was used to collect data (Measurement method) Three or more Two One None All financial sources All public, some private data Only public Otherwise Three Two One None Provincial health accounts FAMS Estimates None 14 Criteria High (3) Mid (2) Low (1) None (0) Score Page 78 of 118

79 Questionnaire July 2005 Health system Health workers per 1000 population (includes physicians, nurses, midwives) How often has the indicator been estimated during the last five years (Periodicity) What is the coverage of data on which the last estimate was based (Representativeness) Is data disaggregated by district, health institution and profession(disaggregation) What method was used to collect data (Measurement method) Three or more Two One None All All public, some private data Only public Otherwise Three Two One None Human resources registers in province FAMS Estimates None Page 79 of 118

80 Questionnaire July 2005 Let us now discuss data sources Data sources Item High (3) Mid (2) Low (1) None (0) Score 15 Census 16 Vital registration Most recent census report available and widely distributed Census population projections by age and sex available for districts or below for current year Census projections are used for the estimation of coverage of health services Coverage of vital registration of deaths (in percent) Coverage of birth registration (in percent) Coverage of maternal deaths registration (in percent) At Facilities At Districts At Province Country only Population data by sex and age by facility Population data by sex and age by district Population data by sex and age by province None available Facilities Districts Province Country >80 % % % <30 % >80 % % % <30 % >80 % % % <30 % Data sources Item High (3) Mid (2) Low (1) None (0) Score 17 Household surveys conducted Province in Health surveys provide key information for monitoring the 11 health MDG Microdata from recent surveys are available > <3 At Facilities At Districts At Province Not at all The data allow disaggregation by socio-economic variables: wealth quintiles, education, urban-rural Survey results are used for provincial health planning All None All plans Most Few None Data sources Item High (3) Mid (2) Low (1) None (0) Score Page 80 of 118

81 Questionnaire July Surveillance system for acute diseases The response system includes management of events, tracking of lab results, contact tracing The disease surveillance system is well integrated into the health information system Information on malaria deaths rates is disaggregated by age groups: under 5 and over 5 years All None Fully integrated Partially Vertical reporting >80% of districts 50-80% of districts 30-50% of districts Not at all <30% of districts 19 Surveillance system for chronic diseases HIV/AIDS/ STI/ TB There is an integrated HAST reporting system There are sentinel surveillance sites that report regularly and with high quality data There is regular supervision of the staff at the surveillance sites The system provides information on whether facilities are providing TB treatment in line with DOTS strategy Annual ANC/HIV prevalence survey is implemented >80% of districts >80% of districts 50-80% of districts 50-80% of districts 30-50% of districts 30-50% of districts <30% of districts <30% of districts Monthly Quarterly Semi-annually Not at all >80% of facilities >80% of districts 50-80% of facilities 50-80% of districts 30-50% of facilities 30-50% of districts <30% of facilities <30% of districts Data sources Item High (3) Mid (2) Low (1) None (0) Score 20 Health service statistics / HMIS The private sector facilities are reporting health service statistics Clinics have health workers trained in HMIS in past 3 years > 80% of private facilities 50-80% of private facilities > 80% of clinics 50-80% of clinics 30-50% of private facilities 30-50% of clinics <30% of private facilities <30% of clinics Each district has a trained health information specialist who has at least two years of training The province has adequate telephone connection with the districts > 80% of districts > 80% of districts 50-80% of districts 50-80% of districts 30-50% of districts 30-50% of districts <30% of districts <30% of districts Page 81 of 118

82 Questionnaire July 2005 The province has adequate internet connection with the districts The provincial annual statistical report has been published annually in the past 5 years Districts produce their own annual reports of analysed health service statistics Vertical reporting systems such as HAST and vaccination are integrated into HMIS > 80% of districts 50-80% of districts 30-50% of districts <30% of districts 5 years 3-4 years 1-2 years Nil > 80% of districts 50-80% of districts 30-50% of districts <30% of districts Fully Partially Hardly Not at all Data sources Item High (3) Mid (2) Low (1) None (0) Score 21 Health systems information There is a current provincial database of health facilities There is a current provincial database of human resources for health All facilities All staff All public, most private All public, some private Only public Not at all Only public Not at all The data of health facilities contains the GPS coordinates of the majority of facilities >80% of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Data sources Item High (3) Mid (2) Low (1) None (0) Score 22 Health systems information continued The private sector information is included in the provincial routine HMIS database Maps and electronic mapping information is available for most districts > 80% of private facilities > 80% of districts 50-80% of private facilities 50-80% of districts 30-50% of private facilities 30-50% of districts <30% of private facilities <30% of districts There is provincial information on the availability of core services such as Cesarean section, ARV therapy, and emergency blood transfusion >80% of facilities 50-80% of facilities 30-50% of facilities <30% of facilities Page 82 of 118

83 Questionnaire July 2005 Let us now discuss shortly the health information legal framework and context, human resources and information processes in your Province 23 Legal framework and context Area of interest There is a written provincial HIS strategic plan in active use that emphasizes integration of different data sources, in a modified form, at most districts Highly adequate (3) Yes, it exists and is being implemented Adequate (2) The strategic plan exist, but the resources to implement it are not available Present but not adequate (1) The strategic plan exists, but it is not used or is not prointegration Not adequate at all (0) No Score 24 Human resources Districts have a designated health information officer position (filled) Privincial capacity in core health information sciences (epidemiology, demography, statistics) HIS staff at subnational/district level are able to modify and improve their HIS (e.g. new data elements & indicators) when changed circumstances (e.g. new programmes, new information needs) make this relevant Capacity building activities has occurred over the past year at district level for HIS staff (statistics, software and database maintenance, and/or epidemiology) Capacity building activities has occurred over the past year at district level program managers (epidemiology, report writing, information management) Capacity building activities has occurred over the past year at district level health facility staff (data collection, self-assessment, analysis, presentation) > 80% of districts 50-80% of districts 30-50% of districts Highly adequate Adequate Partially adequate Yes, The majority have good knowledge, but still needs significant external support and further training Significant capacity building occurred as part of a long-term government-driven HRD plan Significant capacity building occurred as part of a long-term government-driven HRD plan Significant capacity building occurred as part of a long-term government-driven HRD plan Significant capacity building, but largely depending on external (e.g. donor) support and input Significant capacity building, but largely depending on external (e.g. donor) support and input Significant capacity building, but largely depending on external (e.g. donor) support and input Huge variations in such skills are typical Limited building Limited building Limited building capacity capacity capacity <30% of districts Not adequate No, such skills are sorely lacking No No No Page 83 of 118

84 Questionnaire July 2005 Area of interest Written guidelines exist defining how facility supervisors and district managers should use information and integrate it into overall health service management Acceptable rate of staff turnover at provincial level Highly adequate (3) Adequate (2) Yes Written guidelines exist and are used, but not integrated into overall service supervision Not a problem Moderate but acceptable Present but not adequate (1) Written guidelines exist but are not implemented/used Not adequate at all (0) No guidelines exist High but manageable Unacceptably high Score 25 Infrastructure All key indicators, with numerators and denominators, are known and understood by programme staff at national, sub-national and district levels Technical IT support (networks, installation, repairs, general hardware/software maintenance) is available and functional with acceptable response times Routine, semi-permanent, and survey data are generally captured at the district level and submitted electronically Integrated systems equivalent the national HIS database or data warehouse are running at provincial and/or district levels Yes Technical IT support available with response/repair/r eplacement times usually less than 3 days Data is captured locally but stored in or automatically submitted to national servers Equivalent system integration at provincial and/or district levels and provincial/district managers have access to the national data warehouse via the Intranet/Internet enabling vertical collaboration via ICT Good knowledge/understan ding, but need backstopping Technical IT support available, but response/repair/r eplacement times are usually from 3 days to 2 weeks Data is captured and submitted by or similar Equivalent system integration at subnational and/or district levels Limited knowledge/understan ding, need continuous support Technical IT support available, but response/repair/r eplacement times are often 2 weeks or more Data is captured and submitted on diskettes or other physical media Limited system integration at subnational and/or district levels No Technical IT support generally not available No, generally reports are on paper No system integration at subnational and/or district levels Page 84 of 118

85 Questionnaire July Data availability and management Area of interest Minimum indicators and data set identified and systematically collected The indicator set contains all the MDG healthrelated indicators Highly adequate (3) Adequate (2) Yes Indicators identified but not systematically collected Present but not adequate (1) Discussions under way to identify essential indicators All indicators 0-7 MDG indicators None Not adequate at all (0) No Score 27 Data analysis and use Policy advocacy and dissemination Implementatio n and action Key data and indicators from across programme areas are readily available through an integrated database framework Integrated HIS summary reports covering (at least) key MDG health indicators and programme areas are distributed regularly (at least every 3 months) to all other relevant parties (public agencies, elected bodies, the media and the general public HIS information is readily available in a written annual (or biannual) report that pulls together and analyses critically health information from all subsystems Managers at all levels are able to, and actually use information from HIS for local programme management, planning and monitoring At least five problems/challenges from different program areas have been addressed through a written action plan based on HIS data/ indicators The effects of the written action plans have been demonstrably monitored using integrated HIS data and indicators from different subsystems Web-enabled data warehouse exist, with at least partial public access via the World Wide Web Regular integrated reports at least quarterly to the National Assembly and all other relevant ministries Yes All resource allocation (budgets, staff allocations) are supposedly based on HIS data/indicators Web-enabled data warehouse exist, but only internal ministry access Regular integrated reports at least quarterly, but mainly targeting the National Assembly and Cabinet Report made but analysis weak Programme planning and monitoring done, but not resource allocation Yes Written action plan, but no clear use of HIS data/indicators Yes, documented Yes, but not documented Data warehouse exist, but not webenabled Occasional reports, but frequently quarterly less than Report out of date or poor quality Information used for monitoring, but no real planning done Addressed yes, but not via a written action plan No data warehouse No reports No report integrated All key decisions are centralised No Partially No Page 85 of 118

86 Questionnaire July 2005 Planning priority setting and Area of interest Information on prevalence or incidence of STI and HIV are systematically used to advocate less-risk behaviour in the general public as well as in targeted vulnerable groups (such as young females, people with multiple partners, gay men, intravenous drug users, sex-workers and their customers) Health workers analyse all health statistics in theirs province/district and compare with national benchmark in order to adjust operation plan for consecutive period Health data and indicators are demonstrably used in the planning process, e.g. for annual integrated development plans, medium-term expenditure frameworks, and long-term strategic plans HIS data and indicators are used in developing action plan, in re-prioritizing activities, and for better targetting. Health workers analyse all health statistics in theirs province/district and compare with national benchmark in order to adjust operation plan for consecutive period All indicators in the national Essential/Minimum Indicator Set are linked to the relevant short (1 year), medium (3-5 years), and long-term (10-15 years) targets Highly adequate (3) Such indicators are systematically used and tailored to fit the risk profile and situation facing each vulnerable group Adequate (2) Such indicators are regularly used, but generally not tailored to each vulnerable group; Yes, Some health statistics are analysed and any discordant activities were adjusted accordingly. Yes, systematically Commonly used for used with methods diagnostic purposes and targets aligned to describe health between different problems/challenges, planning frameworks but no synchronised use of HIS data/indicators between different planning frameworks; Yes, systematically Commonly used for used both in problem either situation identification and in analysis or identifying setting project's project targets target Yes, Some health statistics are analysed and any discordant activities were adjusted accordingly. All indicators have relevant targets 40-80% of indicators have targets Present but not adequate (1) Only used on an adhoc basis; Health statistics are analyzed and reported. Health information is used occasionally; Health information is used occasionally; Health statistics are analyzed and reported. Under 40% of indicators have targets Not adequate at all (0) Not used No, compilation and analysis were done at central level Never used. Never used. No. No targets Score Page 86 of 118

87 Questionnaire July 2005 Data analysis and use Area of interest There are written guidelines available in the Province for analysing and using of HIS from all/related sub-system in planning and/or budgeting processes There are written procedures for dissemination of reports/ information horizontally to all programme areas and management at the same level at least on a quarterly basis HIS data and indicators, that are collected by any of public agencies, are in principle regarded as belonging in the public domain, i.e. it should be available to all interested citizens There is an integration of HIS data/information from different sources and programmes at all levels Management teams are producing regular written feedback from a) National to sub-national managers b) Sub-national to district c) District to facility Health managers are generally demanding complete and validated HIS information delivered on time Graphs are widely used to display information at subnational / District offices Maps are widely used to display information at subnational / District offices Highly adequate (3) Yes, up-to-date streamlined guidelines are in use Yes, written procedures exist and are largely followed Public access and availability are guaranteed by law/regulations and fully implemented Yes, integration is pursued at all levels including facility levels All sub-national units receive regular written feedback Adequate (2) Yes, but there are several often contradictory sets of guidelines and regulations from different ministries There are written procedures, but they are not fully implemented Public access accepted in principle and largely implemented Integration is pursued from the district level and upwards : 40-80% of subnational units receive regular written feedback; Yes General strong demand from managers, but they do not have the skills and experience to evaluation completeness and quality Yes Up-to-date graphs displayed, but only for some programmes Yes Up-to-date maps displayed, but only for some programmes Present but not adequate (1) Yes, but they are outdated and/or not suitable There are no written procedures, but dissemination are common practice Public access accepted in principle, but not implemented in practice Integration is only pursued at the (sub-) national level Under 40% of subnational units receive regular written feedback Demand from managers are adhoc, usually as a result of external pressure (e.g. questions from politicians or the media) Some graphs, but not up-to-date Some maps, but not up-to-date Not adequate at all (0) None No written procedures and negligible horizontal dissemination Access is strictly controlled No No feedback Negligible demand from managers No graphs No maps Score Page 87 of 118

88 Questionnaire July 2005 Area of interest There are incentives for good information performance, such as awards for the best service delivery performance, for the best/most improved district, or for the best HIS products/utilisation Highly adequate (3) Adequate (2) Yes Institutionalised use of incentives in some areas Present but not adequate (1) Sporadic use of incentives only Not adequate at all (0) No Score Page 88 of 118

89 Annex 5 Database The database, which was developed during the assignment, is designed in MS-Access and thus relational, open-source, flexible and user friendly. It is functional at 4 levels, from health facility through district and province to national level. The database structure follows the standard principles of relational databases with a set of basic data tables and a set of diagnostic data tables. Data input Basic data tables contain basic information about the health management entities (name, address, number of staff, number of HMIS staff, date of interviews, etc.), with automatic update links to higher level entities, as illustrated, Health facility (a) Health facility (b) District HMB (i) Health facility (c) Provincial HMB (x) Health facility (d) Health facility (e) District HMB (ii) MoH/CBoH Health facility (f) Health facility (g) District HMB (iii) Provincial HMB (y) basic information about the interviews (date conducted, time, duration, interviewer, persons attending, etc.), with automatic links to basic entity information, as illustrated, Assessment Date 1 Assessment Date 2 Assessment Date 3 District HMB Results Date 1 Results Date 2 Results Date 3 three different master questionnaires, one for central level and provinces, the second one for districts, and the third one for facilities, containing questions and Page 89 of 118

90 answers which, when revised, will automatically change the diagnosis tables, as illustrated below, three different master answer tables, with criteria for scores, following the structure of the questionnaires, as illustrated. Interview 1 Assessment Date 1 Results Date 1 Master Questions District HMB Master Answers Results Date 3 Interview 2 Assessment Date 3 The variable results (diagnosis data) tables contain questions derived automatically from the master questionnaires and linked to the health entity being interviewed, answers to questions and scores derived from the master answer tables, which constitute the results of the assessment. As the results are presented as ratios, i.e. as a percentage of ideal score, interviews conducted at different dates and with different questions are comparative. Data processing The data processing is governed via a menu structure, i.e. a main menu and a number of submenus. Page 90 of 118

91 The data from an assessment interview is entered by clicking on one of the first 4 buttons, e.g. a click on Provincial Health Management Offices will open the data entry form for the interviews conducted at Provincial Health Management Boards. The variable data from an assessment interview is entered by clicking on the Diagnosis Button which will open the main data entry form for this specific health entity. Page 91 of 118

92 The data is entered by a click in the answer box (or two neighbour boxes) that characterizes the current situation in the health board under review. The score at the right is calculated automatically when the cursor is moved away from the question assessed. By scrolling down all the questions in the questionnaire may be accessed and assessed. The box at the bottom will change in accordance with the heading number at the left. The example shows Heading 1 Prevalence of underweight children (under five years of age) MDG Goal 1:.. When all questions have been answered, the form is closed by click on x in the upper right corner and the basic interview form re-appears. By click on the print button at the interview form a print preview of the data entry will open as illustrated below. The results may then be reconciled and the necessary changes made before printing out the complete result report. Page 92 of 118

93 Data output The output from the database is a number of assessment reports on the HMIS performance for the health entities that have taken part in the assessment. The access channel to the provincial assessment reports is as follows Main Menu Reports Analytical Province Report and thereafter click on the wanted report. The options are MDG Indicators/Other indicators by province HMN Frames by province HMN Headings by province HMN questions by province Province by HMN frames Province by HMN headings Province by MDG indicators Below some examples of the format of the reports. MDG Indicators/Other indicators by province HMN Frames by province HMN Headings by province Page 93 of 118

94 HMN questions by province Information on interviews (date, time, persons attending, etc) can be opened by following the menu channel Main Menu Reports List of interviews and thereafter click on the wanted report. Page 94 of 118

95 Below the first part of the list of interviews with the provincial health management boards. A list of provinces, districts and health facilities included in the database may be obtained by following this menu channel Below an example. Main Menu Information System Basics MoH structure Page 95 of 118

96 Annex 6 - Millennium Development Goals Millennium Development Goals, Targets and Indicators related to health Goal 1. Eradicate extreme poverty and hunger Target 2. Halve, between 1990 and 2015, the proportion of people who suffer from hunger Indicators 4. Prevalence of underweight children under five years of age (UNICEF-WHO) Goal 4. Reduce child mortality Target 5. Reduce by two thirds, between 1990 and 2015, the under-five mortality rate Indicators 13. Under-five mortality rate (UNICEF-WHO) 14. Infant mortality rate (UNICEF-WHO) 15. Proportion of 1 year-old children immunized against measles (UNICEF-WHO) Goal 5. Improve maternal health Target 6. Reduce by three quarters, between 1990 and 2015, the maternal mortality ratio Indicators 16. Maternal mortality ratio (UNICEF-WHO) 17. Proportion of births attended by skilled health personnel (UNICEF-WHO) Goal 6. Combat HIV/AIDS, malaria and other diseases Target 7 Have halted by 2015 and begun to reverse the spread of HIV/AIDS Indicators 18. HIV prevalence among pregnant women aged years (UNAIDS-WHO- UNICEF) 19. Condom use rate of the contraceptive prevalence rate (UN Population Division) c 19a. Condom use at last high-risk sex (UNICEF-WHO) 19b. Percentage of population aged years with comprehensive correct knowledge of HIV/AIDS (UNICEF-WHO) d 19c. Contraceptive prevalence rate (UN Population Division) Target 8. Have halted by 2015 and begun to reverse the incidence of malaria and other major diseases Indicators 21. Prevalence and death rates associated with malaria (WHO) 22. Proportion of population in malaria-risk areas using effective malaria prevention and treatment measures (UNICEF-WHO) e 23. Prevalence and death rates associated with tuberculosis (WHO) 24. Proportion of tuberculosis cases detected and cured under DOTS (internationally recommended TB control strategy) (WHO) Target 17. In cooperation with pharmaceutical companies, provide access to affordable essential drugs in developing countries Indicators 46. Proportion of population with access to affordable essential drugs on a sustainable basis (WHO) Page 96 of 118

97 Annex 7 Review of the implementation of the 1996 HMIS Principles of HMIS A number of principles were proposed to improve the system to overcome these problems DART The main characteristics proposed for the design of the 1996 HMIS were known as DART: Decentralised --whoever collects data must analyze it: Analysis and self-assessment was carried out at the level where data was collected and was expected to be used for decision making and action at that level. Data was not merely collected for upward reporting for higher levels. Action oriented--data must be collected for decision-making, not for filing The old paradigm of collecting "as much data as possible" through the HMIS was abandoned. There was a clear differentiation between the scope of decisions taken by each level Health Management Boards require operational information for day-to-day management and supervision; the Central Board of Health requires information for longer term strategic management and support, and for setting national policy. Different health information needs also exist for the community, health post, health centre, hospitals (first level, district, second and third level) as well as the Regional Boards of Health. Specialised or vertical programs were encouraged to satisfy information needs which fall outside the boundaries of the routine HMIS through sentinel surveillance, surveys and other techniques. Responsive Data collection must react to changing needs Data should be reported in an appropriate timeframe according to its use, and be flexible in terms of adaptation to changing local needs. Transparent a see- through system Obtaining information should be easy and dissemination facilitated by the newly created Regional and National Resource Centers. In addition, correlation of data collected by the various subsystems was greatly facilitated. Data flow The essential principle of using one single channel to move information from one level to another was observed. From the community level, data was communicated (through a Community Based Health Management Information System still not developed) to health posts or directly to the health centre. At the health centre data are aggregated and a subset of information sent to the District Health Office; Districts combine health centre data with information from level I hospitals and other health providers in the area (NGO and private). Districts and Level II hospitals report to the Regional Health Office From province an agreed upon set of information was sent to the Central Board of Health. The resource centres at the Regional and Central level were responsible for making information accessible to interested stakeholders; this includes donors and research institutions. The CBoH combines this information with that received from Level III hospitals to get a composite view of the entire health system. Page 97 of 118

98 Scope of the routine system The 1996 Health Management Information System was designed to raise flags at local level, not to answer management questions. The HMIS should not analyze a problem and tell the manager how to rectify the situation; instead the HMIS serves as a signal that attention was required, in the same way that a warning light on the dashboard of a car warns the driver that something in the engine is not functioning. The HMIS was to alert managers that something unexpected may be occurring in three areas: 1 Health status and disease patterns, 2 Service delivery, 3 Resources to support service delivery. The HMIS plan clearly differentiates the different roles of the overall Monitoring and evaluation framework, and two sub-components, the HMIS and the HIS. The priority for this plan was the routine HMIS which focuses on operational activities and it is repeatedly stated that the routine HMIS was not designed for the collection of all data or for the reporting of all indicators. these are to be collected by non-routine methods. The proposed additional methods were to include Performance Audits, sentinel surveillance, medical records reviews, verbal autopsies and rapid assessment methods including Lot Quality Assurance Sampling (LQAS) and other related research such as focus groups, and Delphi panels. Three units under the new Directorate were to have a direct responsibility for information. These are the HMIS unit, the Health Systems Research and Epidemiology Unit, and the Quality Assurance and Management Audit Unit. Together, these other units were to be responsible for additional information collected by alternative monitoring and evaluation exercises, while the HMIS unit focused on a limited set of routine data. HMIS FRAMEWORK The rationale for the development of the overall HMIS and its subsystems was derived from the framework known as the Health Inputs, Processes, Outputs, and Outcomes (HIPPOPOC) model.70 national indicators were incorporated in the HMIS and provide information for each of the subsystems outlined below. Health Status This measures the outputs of the health system (curative care, preventive activities, and health promotion) as well as outcomes in health status. Zambia's National Strategic Health Plan delineated six priority areas for health services and the health status indicators proposed focus on these priority areas. 1. Safe Motherhood (including antenatal and postnatal care, and family planning); 2. Child Health (including immunisation and nutrition); 3. AIDS and Sexually Transmitted Diseases 4. Malaria 5. Tuberculosis (TB) 6. Environmental Sanitation (including water and latrines). Finances This focuses on financial inputs into the health system. It allows managers to measure costs involved in delivering the six priority thrusts for health services as well as providing basic accounting information... The reporting format has been developed under FAMS and will not change. Page 98 of 118

99 Human Resources This subsystem allows for the understanding of staffing patterns, movements and training requirements. Drugs and Supplies Information was used to measure utilisation and stock management. The distribution system was to change from a "push" to a "pull" system to allow Districts to determine their own needs for drugs and supplies. The Central level was responsible for supplying those needs and assuring that the Essential Drugs Stores are sufficiently stocked. This subsystem also provides information on the rational use of pharmaceuticals. Assets Information on the infrastructural and equipment inputs (including transport) allows the District and Central levels to plan and budget for maintenance and rehabilitation/upgrades. Indicators help measure progress against minimum physical quality standards already developed for health centres. Data from these five subsystems was linked with the subsystem on Health Systems Performance. With this linkage, financial reporting can be linked to health status and service delivery to measure cost-effectiveness. Also staff workload can be linked to utilisation of institutions The Notifiable Disease list was revised to include only those diseases which require immediate action by the Central Board of Health. Other diseases posing a more localized public health threat were to be added at the District's discretion. The HMIS focused on the interpretation of data for decision-making. A number of tools were to be developed for assessing needs, understanding the services the health sector provides, improving coverage rates, and planning for unmet needs. A variety of analytic tools were developed for use at different levels of the system. A quarterly reporting schedule was proposed which was less taxing than the monthly system. Reporting was to occur according to the use of information for planning and action. In exceptional cases such as epidemics, shorter reporting periods were foreseen. HMIS implementation Manual System The implementation strategy was to make the new HMIS system operational nationwide by January The implementation of the manual HMIS was to occur in three phases: development, pretest and roll out. Development phase This phase took place from January-March 1997 and was reviewed in April The main purposes of this stage were the development of analytic tools; ensuring data quality and to provide guidance in using information through training. There were very few health centre managers who had training in management. The HMIS training provided the first and perhaps, the only exposure to basic management principles. Two districts in Western Province (Kaoma and Mongu) were chosen for the Development Stage, on the basis that the Dutch had been supporting a successful Primary Health Care Page 99 of 118

100 (PHC) project in the province and the two districts had been test sites for a number of reform activities including the Financial and Administrative Management System (FAMS). Needs Assessment A nation wide needs assessment was done to ascertain the shortcomings in the previous system and to determine the potential data needs of users. They found that before 1997 the data collection system was fragmented with data collection needs mostly influenced by donorsupported programmes such as Family Health Programme, Universal Child Immunization, Nutrition, Malaria Control, etc. About 37 different forms were in circulation in health facilities - all required by different departments at the central office. A Systems Requirements document was produced in September 1995 with support from the Centre for Disease Control (the Data for Decision-Making Project) The next step involved consultations with programme managers in order to determine their data requirements in the new system. In June 1996, negotiations were held to ascertain the usefulness of the requested data. This process was heavily supported politically and hence moved forward rapidly Development of Forms and Indicators After the final compilation of the data needs, the task of producing an indicator list and data collection tools was contracted out to a technical health practitioner who translated these needs into indicators and data collection forms into a computer software program. Numerous attempts were made to organize meetings at the Central Board of Health to discuss the final indicator list, but attendance was always very poor and programme managers did not have a detailed look at the forms and indicator list, so the very people meant to be the final users of the information did not buy in to the system. Again, political pressure prevailed and implementation went ahead hastily and without adequate consultation. Community-based Information System A synergy needed to be created between data collection for facility and community levels. An attempt was made to conceptualize a community -based health information system, with a view of defining minimum information needs for the CBIHS and to identify methods and tools. The CBIHS however, was abandoned due to lack of funds and policy drive as Central Board of Health felt it was necessary to deal firstly with the facility-based system before moving to community level. Pre-test PHASE This phase ran from May to October 1997, with a post-implementation review in October The selection of pre-test districts was left to Regional Directors; at least one urban district and the four remaining districts in Western province should be included. The system was thus introduced in 12 districts with various levels of experience in data management. Training Strategies A cascaded training strategy was adopted which involved the formation of a national core training team to train district trainers who would in turn train facility staff and brief DHMTs. Page 100 of 118

101 a) National Core Training Team From April to May 1997, 22 officers recruited from national, provincial and district level of the Ministry of Health as well as Central Statistical Office were trained in basic epidemiology and the HMIS as national core training team. b) District Trainers In each of 15 districts, three staff members, (two from the district office and one from the health centre) were trained as district trainers. The national co-training team used this as a practice session following the national training and adjusted some materials based on the feedback from these sessions. c) Health Centre Staff This training ran concurrently with the district training during the second quarter of From each facility a minimum of two staff members were trained, with more for the bigger centres. This included the in-charges, their deputies and sometimes their classified daily employees (CDEs), who were included because in most rural facilities they play a vital role in health service provision and data collection. Upon completion of the training, new tools were to replace the old ones, and by the beginning of the third quarter of 1997 all the health facilities in the 15 districts had embarked on the new HMIS. d) Support visit findings After implementation for one quarter, a team of core and district trainers provided on-the-spot supervision to identify problems the health centres faced in implementing the new system. They checked for data completeness in the forms and registers, increased workload and use of the self-assessment forms, among others. These visits provided a backbone for ensuring that the system was being implemented uniformly throughout the test districts. It also provided an opportunity for making any necessary changes in content and strategy. Some of the major findings from these visits were: e) Use of HMIS Tools and Instruments With the exception of hospitals, the system was generally well -understood and implemented up to the stage of data aggregation. Most of the problems observed in the health centres were easily corrected. The concept of Triple A analysis, embedded into the HMIS, provided the first ever chance for most of the facility staff to understand that data collected by them could be used to identify and analyse problems in their local programmes. System Ownership In some districts, districts trainers had initiated their own strategies to ensure that all staff at the district health office were orientated in the meaning and use of the HMIS. However, some districts lagged behind. Diagnostic Capabilities A Case Definitions Manual was developed to help clinicians make a proper diagnosis. The new case definitions were well understood, except in hospitals. As a result of the many problems observed in hospitals, special attention was required for this level, which lead to the concept of Hospital-Affiliated Health Centres. Page 101 of 118

102 Review of the Training Strategy A review was conducted into the cascade type of training which used during the HMIS training. Concerns had been raised regarding whether the correct messages were filtering down to the lower levels. The review found that there was a variation between districts, with stronger districts able to handle the new training well, whereas others could not handle it Roll out Phase The roll out involved introduction and implementation of the system in the remaining 57 districts and was initially scheduled to take place in September 1997 but only started in September/October 1998, with follow up visits carried out between January and March 1999 because the Ministry of Health banned workshops and large meetings and it was thought the de-linkage process needed to be completed before any further training The training for the rollout commenced in November 1998 and nation-wide implementation took place during the first quarter of Because of the delay, most trainers had taken up other responsibilities and finances and logistical support were also lost as funds could not be moved to another financial year. The quality of training and follow up was therefore not as good as it was during the pre-test phase. Computerisation Automation was proposed to ease information flow and provide greater flexibility and depth in analysis at district, provincial and headquarters levels as well as at hospitals. To ensure that the calculations and analysis of the HMIS were well understood, automation was scheduled only after the manual system had operated successfully for a quarter. Implementation was to commence with the development and pretest sites by mid-1997 and have a roll out phase within the second quarter of Criteria for automation included the availability of computers, computer literacy on the part of staff, standardization in hardware and software, preparation of a procedures manual, and the development of support systems for maintenance and trouble-shooting. All provincial and 85% of district offices received computers. There were plans to install to transmit data but this was not possible due to a number of problems. Installation of services was centrally funded in districts in Western Province, but even then, there were very unsatisfactory results because of the limited skills in districts, bad telephone connections and few districts with computers. Post-implementation Review The system should have been reviewed against its original objectives by 2001 but this had not taken place. By 2003, The HMIS had evolved into a number of shapes. Initially, during implementation, the system was centrally-directed. After that the centre had very little control over what shape the system took at the lowest levels. By the end of 2003, the HMIS at health centre level had been in operation for five years in 57 districts and 6.5 years in 15 districts. Emerging concerns have come up and it was asked Was the HMIS designed to adjust itself in the wake of these and other changes or does it need human intervention?" Page 102 of 118

103 "If we have to let the system respond to new challenges, what do we want to see and when?" These sorts of questions can go on and on, but the bottom line is "What should we do to keep the system afloat? Different groups of people made different demands on the HMIS. users want to get as much information in their technical areas as possible, data managers on the other hand ay not want to keep too much data which may not be useful. The goal was to find a common ground where the system is not oversimplified to the point of being useless, at the same time not to make it too big to manage. Currently, the HMIS is exactly in this position. Somehow all these interests have to be taken care of, but this has not happened The Hospital Information System Health Centres and 1st Level Hospitals Initially, one system was designed for both health centres and district hospitals. This was quickly changed as there were perceived differences in terms of functions, patient type and patient flow. The existing system could not be used at the health centres, hospitals and health offices as each of these has its unique problems and successes which led to the concept of a separate Hospitals Management Information Systems (Hospital MIS). It was seen as necessary to separate primary health care functions of district hospitals from the first level referral functions. This brought about the idea of Hospital Affiliated Health Centre (HAHC) which meant that Out Patient Departments (OPDs) in district hospitals are separated from the hospitals and placed under the authority of the DHMT. Hospitals should attend to referred and serious cases only. Implementation of the Hospital System The implementation of the hospital MIS did not take place as planned. By the 1st quarter of 1999 all hospitals in the country should have been covered. Unfortunately, only six hospitals from Southern Province were piloted and it was not until mid 2001 that training for both 1st and 2nd level hospitals started. The remaining hospitals were trained by January Problems Implementation of the Hospital MIS was not smooth because adequate attention was not given to the hospitals. Management of first level hospitals has not been strong, nor has improvement of these skills been a priority because of Inadequate Funding and inappropriate training material. The Central Board of Health, because of financial constraints, used materials for the first level hospitals to train second and third level hospitals. Page 103 of 118

104 Review of implementation of the HMIS Plan 10 Background As described previously, the focus of the 1996 HMIS plan was to develop a culture of constant use of data. The HMIS was designed, and health workers trained, in accordance with internationally documented practices, so that data are routinely incorporated into the decision making process. What people often call the HMIS is really just an information pipeline. But in fact the system, in the broader sense, is the interaction of information consumers with the pipeline. In Zambia, the pipeline has been built, and the emphasis needs to be on the people using the information. Therefore, the current objective is to strengthen managers skills and improve data use through strong feedback loops.. The evolution of health reform policy has strongly influenced the HMIS. The MoH has rethought its reform strategy several times since 1997: the national CBoH has restructured twice, the province has been redefined and relocated. During the several administrative restructures, the HMIS institutional home has also changed. In other words, while districts and facilities have remained fairly stable, the upper support lines, at provincial and national levels, have changed several times. Therefore, there is a need to strengthen the information skills of new national and provincial officers, while identifying and disseminating best practices in district and facility information use. Staffing patterns throughout the CBoH are lean, including the HMIS unit. No one person is responsible for supervising/ stimulating management use of the information. The two documentation specialists have several responsibilities, including dissemination of HMIS information; the single information technology specialist s numerous responsibilities include support of the HMIS. This staffing pattern is not adequate for the tasks, so several partners provide assistance in the form of local consultants (two trainers and two data entry clerks) and long term advisors with HMIS responsibilities in their portfolios. In general, staffing at provincial, district, and facility levels has gone down by about 10-15% in recent years. It has not yet been possible to incorporate HMIS training into pre-service curricula or to establish an ongoing in-service HMIS training system. The high staff turnover, combined with the absence of ongoing training mechanisms, means that new staff do not have the basic skills needed to use the HMIS. International support for the HMIS has been considerable. Both Danida and USAID have provided continuous support for the HMIS since the mid 90s. Danida focuses on national and provincial support, including automation. The Zambia Integrated Health Project (ZIHP), a USAID-funded project to support initiatives in demonstration districts, provides technical support to the HMIS by using district field visits to inform the national level, by working with provincial teams, and by developing district strengths in the demonstration districts. The Government of the Netherlands supports physicians acting as advisors to selected districts and provinces; these Dutch doctors have also provided considerable technical assistance (and valuable unrestrained criticism) in HMIS implementation and ongoing use. 10 Based on the 2000 RHINO presentation : ESTABLISHMENT OF AN INFORMATION CULTURE: ACHIEVEMENTS AND CHALLENGES IN ZAMBIA by Alex Simwanza and Mary Church, November 2003:The HMIS in Zambia A trace on the implementation steps by Paul Chishimba et al Page 104 of 118

105 Achievements and constraints In spite of numerous constraints of an organisational and development nature, human resource weaknesses and problems of system design, there have been considerable achievements in implementing the HMIS. While the information pipeline has been established, the HMIS does not consistently produce data of the quality required to support good decisions, and therefore is still not broadly used for evidence based decision making. The root problems are systematic and require changing the way in which people interact with information, rather than quick technical fixes such as indicators and computer systems. Achievements There is a good coverage of the HMIS in public health institutions throughout the country, with all 72 districts trained and the majority of district hospitals participating actively. Indicators and case definitions A small number of indicators (74) have been agreed for routine monitoring; these indicators reflect health status and service delivery, as well as drugs and supplies, finances, etc. Standard case definitions were introduced for the first time nationwide and these are contained in planning handbooks. These have been updated and sets of indicators developed for monitoring the National strategic health plan and the MoH and CBoH. Data collection and flow The data collection instruments have been reduced from some 36 to 2 forms. The information flow has been rationalized into a single channel from multiple channels, thereby improving completeness of submission (100% for 2002 data11) Standard guidelines have been produced whereby each facility, district, and province, as well as the national office, is directed to assess its information and performance quarterly. Training and documentation There was a heavy investment in training, both in data collection and information use, and in classroom and on the job training. Documentation, both for training and procedures manuals, was also a substantial investment. A user reference guide should have been produced, but has not been to date. Plans have been made to incorporate HMIS into the pre-service curriculum, but these have not been implemented to date. While it is difficult to quantify the impact of these investments, it seems on the job training should be emphasized as much as possible. Ongoing pre-service and in-service training is also essential. Data timeliness, use and dissemination 11 The HMIS in Zambia p24 Page 105 of 118

106 Timeliness has improved and data are transferred fairly quickly and are available at each level, with many fewer human resources, than in the past. Data consistency has been assessed as 70% 12 and there was 100% report completeness for 2002 data. Data use has been significantly increased, with graphs in many facilities and a good understanding of the monitoring of trends. Planning is supposed to be based on HMIS data, though this is weak because of poor data quality. Dissemination of data has been improved, with quarterly reports regularly produced for all levels in the form of key indicators with comments. Four annual bulletins have also been produced, and an annual statistical report for 2002 covering all key indicators. Computerisation A computer program has been developed and this is in use in 85% of districts for the quarterly capture of data and creating of reports that show trends and cumulative results. All districts have accounts for sending data to the province for integration into the provincial database, which is in turn sent to the national level. The data collection program has had data error trapping procedures incorporated and feedback is more easily produced. Lead provinces The Western and Southern Provinces were drawn into the implementation cycle early and by the year 2000 had advanced well. The Western Province, in particular, has been a focus for the Netherlands for several decades, and some of the decentralization processes were tested there before national implementation. It was hoped that these areas would readily adopt the practice of using information for management. And indeed these provinces and districts provide some of the best examples of how local expertise can develop. Even these areas report that improvement at facility level is very slow. It has not been possible to devise a mechanism to transfer the skills developed in these provinces to other areas of the country. Challenges The HMIS in Zambia was designed based on the assumptions of the Health Reforms; changes in the implementation and content of the reform impacted on the success of the HMIS implementation. Most challenges presented are seen as a result of design oversights, general system-wide problems and human errors. Policy framework Policy weakness is a serious problem with key policies outdated or absent (For example the statistics act dates back to 1964) implying that there is no clear framework for the HMIS. Similarly, the lack of an overall Monitoring and evaluation framework means that there is no clear overview of the interaction between the various components of the health M&E process. There is confusion of the scope of the various systems. Unfinished design The overall monitoring and evaluation framework has not been clarified, leading to confusion over the roles and responsibilities of the HMIS, the HIS and overall M&E. A number of subsystems human resources, drugs and assets - have not been completed and the entire Community based information system remains uncompleted. There is no finalization on a number of issues such as the issue of home-based or facility- held cards; The computerised linkages between the different subsystems and the links to other vertical programs have also not been finalized, making it difficult to get either detailed information or an overview of the results of the system. 12 Mid term review p 27 Page 106 of 118

107 Inflexibility The system has proved not to be responsive to newly emerging issues and new programs such as MDGs, the Global alliance, HIV/AIDS and others. Without strong leadership, the programmed reviews of the system have not occurred and the result is a system that largely does not answer to current problems. Fragmentation This remains a problem with vertical programs setting up their own parallel systems to get data more frequently and accurately. These systems do not relate to the HMIS and it appears that the HMIS does not have the capacity to incorporate even the relevant data. Data from these key programs are therefore not easily available to the health system. Planning and leadership The national working group proposed to consolidate initial national support for the HMIS never developed into an institutional support framework. Besides the ongoing advocacy that such an advisory group should have provided, there was no mechanism for continuing toplevel policy support, refinement of the information system and ongoing planning for the postimplementation phase. The frequent changes of the overall health management structures in the health reform process mean that leadership has not been consistent or strong from provincial or national level. There has been a decline in respect for the administrative structures and an unclear public information agenda. In addition, the proposed ongoing planning processes for the post-implementation phase have not been carried out. Even where plans were made, they could often not be implemented because of budget constraints. This weak leadership is equally evident at lower levels, where it is observed that in districts or facilities where the manager is interested in the HMIS, data quality improves. Information technology The computer program developed was seen to have numerous weaknesses and while efforts have been made to introduce error-trapping mechanisms and to improve feedback, these have not improved use of the program. Data is often not collected electronically at district or hospital level because of limitations in skills and infrastructure. Efforts to set up and internet at district level have not worked well. There are no links with FAMS or Human resource data to ensure assessment of efficiency. Private sector involvement In spite of recognition that the private and non-profit sector had considerable contribution to make in terms of innovation in HMIS development, they were hardly involved in the HMIS development or implementation process, and their contribution to the overall health service delivery effort is not included in routine statistics. Documentation In spite of good intentions, documentation of the system has been poor, with few guidelines available to introduce workers to how to use the system. A user reference guide was proposed, but this has to date not been produced. Training manuals for hospitals are still lacking. Page 107 of 118

108 Human Resources There is a shortage of trained HMIS staff at all levels, both in numbers and in skills, and the turnover of staff is high, leading to a poor institutional memory. Many HMIS posts are not filled, (only 5/9 provinces have data management specialists) and staff that are in post often do not have the skills to perform their roles, and have many other responsibilities. There appears to be poor team work, with little participation of program managers in data use. Senior managers do not go for HMIS training, with a resulting weakness in leadership of the system. This is exacerbated by the frequent changes in management structures at provincial/regional and at national level. Most trainers have moved on to other activities and their activities of in-service training are not carried out. Similarly, support and supervision of the HMIS are weak. Vertical programs appoint focal persons to collect data to make up for perceived shortcomings in the HMIS Data quality and use This remains generally poor, and an assessment showed that this was 90% human failure. The consistency of data, at 70% was considered preoccupying but not alarming. The quality of data and assessment vary considerably from place to place and there are reports of adulteration and falsification of data, particularly at higher levels 13. This results in lack of faith in the entire system, and data is little used for decision making at any level, little use of sector performance indicators to hold the ministry accountable and weak data on gender equity. The non-routine data collection processes proposed in the plan have not been implemented, meaning that there is no cross-checking of the validity of routine data through surveys or record reviews. The Hospital Management information system The Hospital MIS for second and third levels appears not to have got off the ground, and remains weak at primary level. Support for this area will be clearly addressed in the Plan of Action. 13 ibid p 19 Page 108 of 118

109 Annex 8 HMIS Forms Disease Aggregation Form for Health Institutions and Districts HIA.1 District: Health Institution Month/Quarter: Year: Part 1: Completed by all health institutions and District Health Offices OPD First Attendance IPD Discharge Deaths DIAGNOSES 1.0 IMMUNISABLE DISEASES *** Acute Flaccid Paralysis (suspected polio) 1.10 Diphtheria 1,20 *** Measles 1.30 *** Neonatal Tetanus 1.40 Tetanus (non neonatal) 1.50 Whooping Cough (pertussis) COMMUNICABLE DISEASES AIDS (suspected or confirmed) 2,05 Bilharzia (schistosomiasis) 2,10 *** Diarrhoea: bloody (suspected dysentery) Diarrhoea: non-bloody Diarrhoea with severe dehydration *** (suspected cholera-see case definitions) Ear/nose/throat infections 2,20 Eye infections 2,25 Intestinal worms 2,30 Leprosy (suspected or confirmed) 2.35 Malaria 2,40 *** Meningitis 2,45 Pelvic inflammatory disease 2,50 *** Plague 2,55 *** Rabies 2.60 Respiratory Infection: pneumonia Respiratory Infection: non-pneumonia Sexually Transmitted Diseases 2,70 Skin infections 2,75 Tuberculosis (suspected and confirmed) 2,80 *** Typhoid Fever 2,85 Urinary Tract Infections (non-std) 2,90 *** Yellow Fever 2,95 Communicable Disease (specify) Communicable Disease (specify) Communicable Disease (specify) Communicable Disease (specify) All other communicable diseases (total) 2, MATERNAL AND PERINATAL HEALTH Complications of pregnancy 3,05 Complications of delivery 3,10 Any perinatal conditions 3, NUTRITION Protein Energy Malnutrition (PEM) 4,05 Other nutritional disorders 4,99 *** Indicates Notifiable Disease under 5 years (a) 5 years and above (b) total (c) under 5 years (d) 5 years and above (e) total (f) under 5 years (g) 5 years and above (h) total (i) Page 109 of 118

110 Disease Aggregation Form for Health Institutions and Districts HIA OTHER DISEASES Anaemia 5,05 Cardio-vascular diseases 5,10 Dental diseases 5,15 Digestive system: (not infectious) 5,20 Ear/nose/throat diseases (not infectious) 5,25 Eye diseases (not infectious) 5,30 Genital-Urinary diseases (other than above) 5,35 Mental disorders 5,40 Muscular skeletal and connective tissue (not trauma) 5,45 Neoplasm (All types) 5,50 Nervous System Disorders: Epilepsy 5,55 Nervous System Disorders: Other 5,60 Poisoning 5,65 Pulmonary diseases (not infectious) 5,70 Pyrexia of Unknown Origin (PUO) 5,75 Skin Diseases (not infectious) 5,80 Substance Abuse 5,85 Trauma: accidents, injuries, wounds, burns 5,90 OTHER DIAGNOSES 5, TOTAL DIAGNOSES Total (sum of all preceding rows)-required only at district 10.0 *note: OPD and IP Attendance aggregation required at district level only; at health facilities aggregation is optional and depends on instructions from the district. All institutions and districts should total deaths DISTRICT-SELECTED DISEASES/CONDITIONS Specify Specify Specify Specify Part 2: Completed by district health offices and health institutions with diagnostic capabilities in HIV, Leprosy or TB Age Group DIAGNOSES under 12 years (a) 12 years and older (b) HIV: Laboratory Confirmed Diagnosis HIV.1 HIV Tests Performed HIV.2 New Leprosy MB LEP.1 New Leprosy PB LEP.2 Under treatment Leprosy MB LEP.3 Under treatment Leprosy PB LEP.4 New Tuberculosis Extrapulmonary TB.1 New Tuberculosis Pulmonary (sputum positive) TB.2 New Tuberculosis Pulmonary (sputum negative) TB.3 Tuberculosis Relapse (old) (smear positive) TB.4 Tuberculosis Relapse (old) (smear negative) TB.5 total (c) Part 3: Completed by District TB Control Officer Cohort Analysis Smear Positive TB cases registered in quarter ending months earlier: quarter of year Results at the end of the full treatment period (the quarter the report completed) Started in quarter months earlier Smear negative (cured) Smear not done (treatment completed) Smear positive Died defaulted transferred Total (a) (b) (c) (d) (e) (f) (g) (h) CA.1 New smear positive CA.2 Smear positive retreatment Page 110 of 118

111 Health Centre Service Delivery Aggregation Form HIA.2 District: Health Centre: Year: Catchment Population: 1. Maternal Health Month Month Month QUARTER 1,1 Ante and Postnatal Care Antenatal Visits First attendances Reattendances Total visits ( plus ) TT Immunisation Pregnancies protected by TT Postnatal Visits First Postnatal Visits ,2 Deliveries Normal Complicated Total deliveries (1.2.1 plus 1.2.2) 1,2 Live births Still Births ,3 Perinatal Conditions Low birth weight 1,3 1,4 Maternal Mortality Maternal Deaths 1,4 2. Family Planning Month Month Month QUARTER 2,1 Family Planning (all methods) New Acceptors Revisits Total Contacts (2.1.1 plus 2.1.2) 2,1 3. Child Health Month Month Month QUARTER 3,1 Immunization BCG under 1 year DPT III under 1 year OPV III under 1 year Measles under 1 year Fully Immunized Child under 1 year ,2 Nutrition Weight below lower line under 5 years Under 5s weighed Preventive Care Contacts QUARTER 4,1 Child Health Under 5 Clinic Attendance (under 5s weighed from 3.2.2) 4,1 4,2 Maternal Health Antenatal Clinic (from 1.1.1) Postnatal Clinic (from 1.1.3) Family Planning Clinic (from 2.1) Total Maternal Health Attendances (4.2.1 plus plus 4.2.3) 4,2 4,9 Total Preventive Contacts Total Contacts (4.1 plus 4.2) 4,9 5. Human Resources QUARTER Qualified staff in Health Centre 5,1 (per Human Resources Register) Qualified staff serving in institution if HC; Qualified staff serving in OPD if hospital; (further instructions in procedures manual) ,2 trained Traditional Birth Attendants (ttbas) Reporting at least once in quarter Deliveries assisted by ttbas ,3 Community Health Workers (CHWs) Reporting at least once in quarter Clients seen by CHWs Page 111 of 118

112 Health Centre Service Delivery Aggregation Form HIA.2 6. Curative Care Contacts Month Month Month QUARTER 6,1 Outpatient Visits First attendances under First attendances 5 and over Total First Attendances (6.1.1 plus 6.1.2) Reattendances under Reattendances 5 and over Total Reattendances (6.1.4 plus 6.1.5) Total under 5 visits (6.1.1 plus 6.1.4) Total 5 and over visits (6.1.2 plus 6.1.5) Total Visits (6.1.3 plus 6.1.6) ,2 Inpatient Admissions Under and over Total Admissions (6.2.1 plus 6.2.2) 6,2 6,3 Deaths Under and over Total Deaths (6.3.1 plus 6.3.2) 6,3 6,4 Bed Utilisation Total patient bed days Number of beds in institution Drugs and Supplies Month Month Month QUARTER 7,1 Drug Kits Number of drug kits opened 7,1 7,2 Health Centre Basic Drugs: In Stock 1 if drug was always in stock (never out of stock) during the month 0 if drug was out of stock at any time during month (further instructions in procedures manual) Chloroquine Paracetamol Cotrimoxazole Oral Contraceptives BCG vaccine DPT vaccine OPV vaccine Measles vaccine TT vaccine ,3 Rational Drug Prescription in OPD Number of patients for whom antibiotics were prescribed out of 40 OPD patient sample (further instructions in procedures manual) 7,3 8. Environmental Health Month Month Month QUARTER 8,1 Inspections Number of premises inspected Number of inspected premises in compliance Number of food inspections performed Number of food inspections resulting in seizure and disposal of food stuffs ,2 Sampling Water samples taken Number of times salt tested for iodine ,3 Rodent and Vector Control Complaints received Complaints attended to Note: Extra lines are provided so that each district can collect data for local indicators associated with environmental health activities (further instructions in procedures manual). 9. Supportive Supervision QUARTER 9,1 Supportive Supervision by DHMT 1 if DHMT provided supportive supervision with written feedback at least once in quarter 0 if supportive supervision with written feedback was not provided (further instructions in procedures manual) Page 112 of 118

113 Hospital Service Service Delivery Aggregation Form HIA.3 District: Hospital: Year: Catchment Population: 1. Maternal Health Month Month Month QUARTER 1,1 Ante and Postnatal Care (not applicable) 1,2 Deliveries Normal Complicated Total deliveries (1.2.1 plus 1.2.2) 1,2 Live births Still Births ,3 Perinatal Conditions Low birth weight 1,3 1,4 Maternal Mortality Maternal Deaths 1,4 2. Family Planning (not applicable) 3. Child Health (not applicable) 4. Preventive Care Contacts (not applicable) 5. Human Resources QUARTER Qualified staff in Hospital 5,1 (per Human Resources Register) Qualified staff serving in OPD (further instructions in procedures manual) Qualified staff serving in wards (further instructions in procedures manual) Qualified support staff (further instructions in procedures manual) Total qualified staff (5.1.1 plus plus 5.1.3) Curative Care Contacts Month Month Month QUARTER 6,1 Outpatient Visits Referred Outpatients First attendances under First attendances 5 and over Total First Attendances ( plus ) Reattendances under Reattendances 5 and over Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Bypass Fee Outpatients First attendances under First attendances 5 and over Total First Attendances ( plus ) Reattendances under Reattendances 5 and over Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Total Hospital Outpatients First attendances under 5 ( plus ) First attendances 5 and over ( plus ) Total First Attendances ( plus ) Reattendances under 5 ( plus ) Reattendances 5 and over ( plus ) Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Page 113 of 118

114 Hospital Service Service Delivery Aggregation Form HIA.3 6,2 Inpatient Admissions Under and over Total Admissions (6.2.1 plus 6.2.2) 6,2 6,3 Deaths Under and over Total Deaths (6.3.1 plus 6.3.2) 6,3 6,4 Bed Utilisation Total patient bed days Number of beds in institution Drugs and Supplies Month Month Month QUARTER 7,1 Drug Kits (not applicable) 7,2 Health Centre Basic Drugs: In Stock (not applicable) 7,3 Rational Drug Prescription in OPD Number of patients for whom antibiotics were prescribed out of 40 OPD patient sample (further instructions in procedures manual) 7,3 7,4 Hospital Basic Medical Supplies: In Stock 1 if drug was always in stock (never out of stock) during the month 0 if drug was out of stock at any time during month (further instructions in procedures manual) Fansidar Amoxycillin Benzyl Penicillin Rifampicin / Isoniazid Ketamine Lancets RPR Screening Kits HIV Screening Kits ,5 Drug Expiration Units of drugs or supplies passing expiration date Any drug Uristix RPR Screening Kits HIV Screening Kits Units of drugs or supplies passing expiration date (7.5.1 plus plus plus 7.5.4) 7,5 8. Environmental Health (not applicable) 9. Supportive Supervision (not applicable) Page 114 of 118

115 District Service Delivery Aggregation Form District: HIA.4 Year: Catchment Population: 1. Maternal Health Month Month Month QUARTER 1,1 Ante and Postnatal Care Antenatal Visits First attendances Reattendances Total visits ( plus ) TT Immunisation Pregnancies protected by TT Postnatal Visits First Postnatal Visits ,2 Deliveries Normal Complicated Total deliveries (1.2.1 plus 1.2.2) 1,2 Live births Still Births ,3 Perinatal Conditions Low birth weight 1,3 1,4 Maternal Mortality Maternal Deaths 1,4 2. Family Planning Month Month Month QUARTER 2,1 Family Planning (all methods) New Acceptors Revisits Total Contacts (2.1.1 plus 2.1.2) 2,1 3. Child Health Month Month Month QUARTER 3,1 Immunization BCG under 1 year DPT III under 1 year OPV III under 1 year Measles under 1 year Fully Immunized Child under 1 year ,2 Nutrition Weight below lower line under 5 years Under 5s weighed Preventive Care Contacts QUARTER 4,1 Child Health Under 5 Clinic Attendance 4,1 ( d 5 i h d f 3 2 2) 4,2 Maternal Health Antenatal Clinic (from 1.1.1) Postnatal Clinic (from 1.1.3) Family Planning Clinic (from 2.1) Total Maternal Health Attendances (4.2.1 plus plus 4.2.3) 4,2 4,9 Total Preventive Contacts Total Contacts (4.1 plus 4.2) 4,9 5. Human Resources QUARTER 5,1 Qualified staff in Instiution (per Human Resources Register) Qualified staff in Health Centre Qualified staff serving in Health Centre (further instructions in procedures manual) Qualified staff in Hospital Qualified staff serving in hospital OPD (further instructions in procedures manual) Qualified staff serving in wards of hospital (further instructions in procedures manual) Qualified support staff serving in hospital (further instructions in procedures manual) Total qualified staff ( plus plus ) ,2 trained Traditional Birth Attendants (ttbas) Reporting at least once in quarter Deliveries assisted by ttbas ,3 Community Health Workers (CHWs) Reporting at least once in quarter Clients seen by CHWs Page 115 of 118

116 District Service Delivery Aggregation Form HIA.4 6. Curative Care Contacts Month Month Month QUARTER 6,1 Outpatient Visits Referred Outpatients at Hospital First attendances under First attendances 5 and over Total First Attendances ( plus ) Reattendances under Reattendances 5 and over Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Bypass Fee Outpatients at Hospital First attendances under First attendances 5 and over Total First Attendances ( plus ) Reattendances under Reattendances 5 and over Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Total Outpatients at Hospital First attendances under 5 ( plus ) First attendances 5 and over ( plus ) Total First Attendances ( plus ) Reattendances under 5 ( plus ) Reattendances 5 and over ( plus ) Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Total Health Centre Outpatients First attendances under First attendances 5 and over Total First Attendances ( plus ) Reattendances under Reattendances 5 and over Total Reattendances ( plus ) Total under 5 visits ( plus ) Total 5 and over visits ( plus ) Total Visits ( plus ) Total Outpatients First attendances under 5 ( plus ) First attendances 5 and over ( plus ) Total First Attendances (6.1.1 plus 6.1.2) Reattendances under 5 ( plus ) Reattendances 5 and over ( plus ) Total Reattendances (6.1.4 plus 6.1.5) Total under 5 visits (6.1.1 plus 6.1.4) Total 5 and over visits (6.1.2 plus 6.1.5) Total Visits (6.1.3 plus 6.1.6) ,2 Inpatient Admissions Inpatient Admissions at Hospitals Under and over Total Admissions ( plus ) Inpatient Admissions at Health Centres Under and over Total Admissions ( plus ) Total Inpatient Admissions Under 5 ( plus ) and over ( plus ) Total Admissions (6.2.1 plus 6.2.2) 6,2 6,3 Deaths Under and over Total Deaths (6.3.1 plus 6.3.2) 6,3 6,4 Bed Utilisation at Hospitals Bed Utilisation at Hospitals Total patient bed days Number of beds in hospitals Bed Utilisation at Health Centres Total patient bed days Number of beds in Health Centres Total Bed Utilisation Total patient bed days ( plus ) Number of beds ( plus ) Page 116 of 118

117 District Service Delivery Aggregation Form HIA.4 7. Drugs and Supplies Month Month Month QUARTER 7,1 Drug Kits Number of drug kits opened 7,1 7,2 Health Centre Basic Drugs: In Stock enter number of Health Centres in which drug or supply was always in stock (never out of stock) during the month (further instructions in procedures manual) Chloroquine Paracetamol Cotrimoxazole Oral Contraceptives BCG vaccine DPT vaccine OPV vaccine Measles vaccine TT vaccine ,3 Rational Drug Prescription in OPD Number of patients for whom antibiotics were prescribed out of 40 OPD patient sample (further instructions in procedures manual) 7,3 7,4 Hospital Basic Medical Supplies: In Stock enter number of hospitals in which drug or supply was always in stock (never out of stock) during the month (further instructions in procedures manual) Fansidar Amoxycillin Benzyl Penicillin Rifampicin / Isoniazid Ketamine Lancets RPR Screening Kits HIV Screening Kits number of hospitals in District ,5 Drug Expiration Units of drugs or supplies passing expiration date Any drug Uristix RPR Screening Kits HIV Screening Kits Units of drugs or supplies passing expiration date (7.5.1 plus plus plus 7.5.4) 7,5 8. Environmental Health Month Month Month QUARTER 8,1 Inspections Number of premises inspected Number of inspected premises in compliance Number of food inspections performed Number of food inspections resulting in seizure and disposal of food stuffs ,2 Sampling Water samples taken Number of times salt tested for iodine ,3 Rodent and Vector Control Complaints received Complaints attended to Note: Extra lines are provided so that each district can collect data for local indicators associated with environmental health activities (further instructions in procedures manual). 9. Supportive Supervision QUARTER 9,1 Supportive Supervision by DHMT number of Health Centres to which DHMT provided supportive supervision with written feedback at least once in quarter (further instructions in procedures manual) number of Health Centres in District District Finances and Reports QUARTER 10,1 District Expenses Total expended by District (in quarter) Expenditures on District Office (in quarter) Expenditures on Hospital (in quarter) Expenditures on Health Centres and Community (in quarter) Expenditures on allowances (in quarter) Expenditures on emergency drugs (in quarter) Expenditures on fuel (in quarter) Expenditures on capital investment (in quarter) Total District budget for drugs and medical supplies (in quarter) District budget for drugs (in quarter) District expenditures on drugs (in quarter) District budget for laboratory supplies (in quarter) District expenditures on laboratory supplies (in quarter) ,2 Reports Total number of institutions that should report Total number of institutions that submitted HIA.1 and HIA.2 or HIA Total number of institutions that submitted HIA.1 and HIA.2 or HIA.3 within 3 weeks after the end of the quarter Page 117 of 118

118 Annex 9 Presentations of results 1. Presentation of the Inception Report, Monitoring and Evaluation Subcommittee, Ministry of Health, 13 July Slides from PowerPoint presentation are found on following pages. 2. Presentation of the Draft Assessment Report, Implementation Review Subcommittee, Ministry of Health, 29 July Presentation of the Draft Assessment Report and of the Plan of Action, Monitoring and Evaluation Subcommittee, Ministry of Health, 3 August Slides from PowerPoint presentation are found on following pages. 4. Presentation of the Plan of Action, Implementation Review Subcommittee, Ministry of Health, 5 August Presentation of the Draft Assessment Report and of the Plan of Action, National Consensus Workshop, Mulungushi International Conference Centre, Lusaka, 16 September Slides from PowerPoint presentation are found on following pages. Page 118 of 118