Attachment_J.15_ZamNIS_Scoping_Report.pdf

Zambia Nutrition Information System (ZamNIS) Scoping Exercise

Report on Activities, Documents, and Recommendations November 2016 - August 2017

ZamNIS Scoping Exercise

Report on Activities, Documents, and Recommendations - August 2017

Table of Contents Acronyms 3

1. Introduction 5

2. Background 5

2.1 Situational Analysis 5

3. Activities 6

3.1 Steering Committee Formation 6

3.2 Field-based data collection and observation 6

Ministry of Agriculture 7 Ministry of Community Development and Social Welfare 7 Ministry of Fisheries and Livestock 8 Ministry of General Education 8 Ministry of Health 9 Ministry of Water Development, Sanitation, and Environmental Protection 9 Partners 10 Overarching Bottlenecks 10

3.3 Indicator, reporting flow, and user requirements workshop 11 Indicator List 11 Reporting Flow 13 User Requirements 13

3.4 Data source and technical document review workshop 14

4. Final documents and Recommendations 14

4.1 User Roles and Technical Requirements 14

Recommendations 14

4.2 Indicator List and Data Sourcing 15

Indicator List 15 Data Sourcing 15 Recommendations 16

4.3 Surveillance and Supervision Protocol 17

Recommendations 17

4.4 Disaster Recovery Plan Guidance 17

4.5 Technology Solution 19

Data Warehouse Characteristics 19 Proposed Technology Solution 21

4.6 Implementation Framework 25

References 26 ANNEX 1: ZamNIS Steering Committee Members and Workshop Participants 27 ANNEX 2: ZamNIS Steering Committee - Terms of Reference 29 ANNEX 3: ZamNIS Data Collection Assessment Tool 31

ANNEX 4: Proposed ZamNIS Indicator List (Attached Separately) ANNEX 5: NFNC Partner Ministry Reporting Flow 33 ANNEX 6: User Roles and Technical Requirements for the Proposed ZamNIS 39 ANNEX 7: Surveillance and Supervision Protocol for the Proposed ZamNIS 63 ANNEX 8: Implementation Framework 64

Acronyms AIR American Institute for Research BI Business Intelligence CARE Cooperative for Assistance and Relief Everywhere CRS Catholic Relief Services CSO Central Statistics Office DFID Department for International Development (UK) DHS Demographic and Health Survey EU European Union FISP Farmer Input Support Programme FSP Food Security Pack GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit (German

Corporation for International Development) GRZ Government of the Republic of Zambia ICT Information, Communication, Technology IEC Information, Education, Communication IT Information Technology HISP Health Information Systems Project HMIS Health Management Information System IMAM Integrated Management of Acute Malnutrition MCDP Most Critical Days Program MCDSW Ministry of Community Development and Social Welfare M&E Monitoring and Evaluation MFL Ministry of Fisheries and Livestock MIS Management Information System MNDP Ministry of National Development and Planning MOA Ministry of Agriculture MOGE Ministry of General Education MOH Ministry of Health MWDSEP Ministry of Water Development, Sanitation, and Environmental Protection NCD Non Communicable Disease NFNC National Food and Nutrition Commission NGO Non-governmental Organization NIPN National Information Platform for Nutrition NWASCO National Water Supply and Sanitation Council UiO University of Oslo OLAP Online Analytic Processing SAG Sanitation Action Group SBH Systems for Better Health SC Steering Committee SCT Social Cash Transfer STC Save the Children SUN Scaling Up Nutrition UNZA University of Zambia

USAID United States Agency for International Development WASH Water, Sanitation, and Hygiene WFP World Food Programme WHO World Health Organization ZamNIS Zambia Nutrition Information System

1. Introduction The Zambia Nutrition Information System (ZamNIS) is a proposed web-based data warehouse for the collation, aggregation, and data visualization of nutrition and nutrition-related indicators and information. ZamNIS seeks to provide a singular database for accessing cross-sectoral programmatic information from implementing partners, nutrition research, periodic population-level surveys, and government ministries. With financial support from the World Health Organization (WHO) country office, under the guidance of Zambia’s National Food and Nutrition Commission (NFNC), Akros conducted a scoping exercise to determine the feasibility of implementing ZamNIS, this report outlines the results of the assessment. The ZamNIS scoping exercise was a collaborative effort made possible by continued input and feedback from stakeholders and nutrition professionals in Zambia.

2. Background

2.1 Situational Analysis

While Zambia has made great strides in reducing mortality in infants and children under five, at 40% prevalence, stunting amongst these groups remains unacceptably high (CSO, 2015).

Stunting, when the height of a child is less than expected for his or her age, is indicative of long-term malnutrition and micronutrient deficiency. Studies suggest that the majority of brain development occurs within the first 1,000 days of life, from conception to age two, and that chronic malnutrition during this period can affect cognitive ability, academic achievement, maternal reproductive outcomes, and the availability of economic productive human capital—a crucial requirement for low and middle-income countries seeking sustainable development (Dewey & Begum, 2011). Achieving reductions in stunting among children under five is of prime importance for the Government of the Republic of Zambia (GRZ), cooperating partners, and implementing partners.

NFNC is the designated convening body to coordinate nutrition in Zambia, under the Ministry of Health (MOH). NFNC provides leadership at the national-level aimed at reducing undernutrition and garners political support for an effective multisectoral response. In particular, six key ministries are seen as essential to collaborative efforts to reduce undernutrition: Health, Community Development and Social Welfare, Agriculture, Education, Fisheries and Livestock, and Water Development Sanitation and Environmental Protection.

A significant aspect of the NFNC mandate is the provision of metrics to stakeholders that detail nutrition implementation progress, achievements toward national nutrition objectives, bottlenecks, and the effects and impacts of various national nutrition activities. Currently, there is no information system in Zambia for systematic monitoring of nutrition programmes and projects against set targets—a major impediment to fulfillment of the NFNC mandate.

Further, NFNC has limited access to infrastructure for storing, processing, and analyzing food and nutrition data and producing the information required for decision-making and policy formulation (Kwofie, 2015).

3. Activities From November 2016 to August 2017, NFNC contracted Akros to conduct a scoping exercise to determine the feasibility of implementing ZamNIS. To better understand the state of nutrition-related data collection and dissemination in Zambia, Akros and NFNC conducted qualitative data collection in national-level ministries and in four districts, two of which are recipients of Scaling Up Nutrition (SUN) funding. Through a series of workshops and steering committee meetings, Akros conducted consultative meetings to determine organizational hierarchies, reporting pathways, and stakeholder requirements for a proposed national nutrition database.

3.1 Steering Committee Formation

Under NFNC leadership, M&E staff and nutrition professionals from line ministries, statutory bodies, donors, implementing partners, academia, and other key stakeholders were invited to form the ZamNIS Steering Committee to guide the scoping exercise and to provide consultative input into the process. The first steering committee meeting was held on 18 November 2016, where participants agreed on terms of reference for membership and a workplan for the scoping exercise (Annex 2). From this meeting, eight core members formed the steering committee and participated consistently in scoping exercises (Annex 1).

3.2 Field-based data collection and observation

In December 2016, Akros and NFNC visited the six nutrition-related line ministries at national level and four districts to conduct qualitative interviews and observations regarding nutrition data collection amongst partners and line ministries at district and sub-district levels (Figure 1). Two of the districts visited were recipients of SUN-funding (Lundazi and Chipata), while the other two relied solely on government inputs and partner programs for nutrition-related activity implementation (Kafue and Chisamba).

During this period of qualitative observation, specific information was elicited to better understand the flow of nutrition-related data from sub-district data collection processes to national-level submission, geographic reporting areas, cadres involved in reporting, indicators collected, and reporting tools used. To achieve this, Akros used principles from the data use cycle (Figure 2) to inform a set of questions used during visits to the ministries, districts, and partners (Annex 3).

Figure 2: The data use cycle for information systems

The following observations were made while visiting implementing partners and NFNC partner ministries at national, district, and sub-district level:

Ministry of Agriculture

Nutrition-related Activities

● Farmer Input Supply Programme (FISP)

● Crop diversity demonstrations and promotion

● Food preparation and preservation

Data sources

● Tally Sheets (SUN-provided)

● Activity attendance registers

● FISP Register

Bottlenecks/Challenges

● Turn out for activities is low during planting and harvesting seasons

● Do not use any form of data collection outside of FISP register and attendance sheets

Ministry of Community Development and Social Welfare

Nutrition-related Activities

● Food Security Packs (FSP)

● Cookery demonstrations

● Public Welfare Assistance Scheme (Social Cash Transfer)

● Women’s empowerment and small loans programs

● Alternative Livelihood Programme

Data sources

● FSP Register

● SCT Register

● Activity attendance register

● Tally Sheets (SUN-provided)

Databases

● Integrated Single Registry: Under development, will integrate the existing three data systems (NGO MIS, Zo MIS, SCT MIS). Will allow all social protection programs to be managed under one system and will help to remove duplicate beneficiaries

Bottlenecks/Challenges

● IEC activities implemented in cooperation with other ministries are often double-counted

● Databases don’t seem to be used effectively and are housed at national level so cannot be used for data collection in the districts

● No consistent way of collecting activity data in the districts

● Social Welfare Department and Community Development Department operate independently of one another

● No specialized M&E cadres within this Ministry, no set targets for achievement

Ministry of Fisheries and Livestock

Nutrition-related Activities

● Dietary diversity promotion

● Aquaculture and livestock support

● Fish and livestock production tracking

Data sources

● Activity attendance registers

● Tally Sheets (SUN Provided)

Databases

● Epi Info for animal disease surveillance

Bottlenecks/Challenges

● Institutional structure is weak since split from MoA

● Few existing legal instruments to support MFL activity implementation

● No form of data collection outside of SUN-fund districts

Ministry of General Education

Nutrition-related Activities

● School Health and Nutrition Program (nutrition messaging, school health days)

● Partner-led School Feeding/Home Grown School Feeding

● WASH (including infrastructure)

Data sources

● Annual School Census

● Monthly Return Forms

Databases

● Ed Assist (compilation of annual school census)

● EMIS (Pilot only in Namwala District)

Bottlenecks/Challenges

● Monthly return forms are not consistent across districts

● Districts do not have access to analysis features of Ed Assist

● Annual data does not reflect regular program updates

● For reliable and consistent data collection outside of the annual school census, schools often require additional resources from partners (to implement the programs and train in data collection)

● Funding for nutrition activities is often issued late in the quarter for which reporting should take place

Ministry of Health

Nutrition-related Activities

● Growth monitoring and promotion

● MCH Supplementation, deworming

● Infant and young child feeding counselling (Breastfeeding Promotion)

● Antenatal, delivery, and postnatal care

● Inpatient and outpatient therapeutic management of malnutrition

Data Sources

● HIA2 Form

● Patient Registers

● Tally Sheets (some SUN-provided)

Databases

● Health Management Information System (HMIS)

● Integrated Management of Acute Malnutrition (IMAM) Database

Bottlenecks/Challenges

● Stunting is the overarching impact indicator and it cannot be measured because the majority of health facilities do not have height boards or mats. Height is not measured, only weight and age (underweight, generally not a very indicative indicator)

● Double counting on IEC activities implemented with other ministries

● Most districts are not using HMIS to compile monthly reporting forms for SUN, rather they transfer numbers from HIA2 forms and tally sheets. Many find these numbers then do not match what is in HMIS.

● Disaggregation (age groups) are different in HMIS than in other partner reporting forms

● HIA2 forms are not comprehensive, this is being addressed by the HMIS project -need to strike a balance between data collection and seeing patients (data systems should improve services, not impede them)

Ministry of Water Development, Sanitation, and Environmental Protection

Nutrition-Related Activities

● Rural water and sanitation infrastructure tracking

● WASH Promotion

● Communal borehole drilling and maintenance

Data sources

● Community Led Total Sanitation (SAG data collection forms)

● Area Pump Mender Reports

Databases

● WASH MIS

Bottlenecks/Challenges

● No WASH data collection in peri-urban and urban areas (NWASCO tracks sanitation in urban planned areas, but they are not engaged)

● Current water point questions on SAG forms leaves out those villages with no water points (included on area pump mender reports)

● Discrepancies between WASH MIS and partner reporting forms

● Community volunteers often have to cover large distances, which is really difficult to ensure quality and accurate reporting

● Number of people reached with WASH and nutrition messaging is not tracked

Partners

Implementing and cooperating partners visited during qualitative data collection included the European Union, German Corporation for International Development (GIZ), Catholic Relief Services (CRS), CARE International, and the World Food Programme (WFP). These agencies and organizations fund and/or manage a number of nutrition-related activities in Zambia, including, but not limited to: school feeding, nutrition counselling, market research on food pricing and availability, group care models for promoting infant and young child feeding practices, improving services for farmers through MOA support, promoting dietary diversity, improving nutrition IEC materials for extension officers, sanitation and women’s empowerment interventions, and enhancing data collection practices. Each organization and/or project had its own target groups and indicator definitions dependent upon goals and objectives of projects. Despite these differences, the majority of partners used stunting and dietary diversity scores as key indicators for measuring impact.

Overarching Bottlenecks

Overall, a number of district and sub-district ministerial staff reported similar challenges in nutrition-related data collection, aggregation, and use:

● Limited transport to implement programs, collect data, and submit reports. Results in delays, particularly during rainy season.

● The mobile networks are inconsistent and/or unreliable in some areas which makes data submission online difficult.

● Sub-district reporting structures across NFNC partner ministries are vastly geographically different, this will make it hard to disaggregate and compare data below district level.

● Data use at sub-national level for most ministries is suboptimal, and many responsible for data collection do not understand the importance of the data

● With the exception of the WASH MIS, partner ministries use paper-based reporting up to district level (and higher) this requires budgeting for printing, which is sometimes left to partners.

● There are few incentives for community volunteers to submit reports and statistics

● SUN-funded districts are heavily dependent upon the program to provide tally sheets for activities, which is unsustainable and not consistent with the reality in non-SUN fund districts.

● Determination of target population denominators is varied between ministries.

● Balancing the need for information with the ability to provide services and reduce interruptions to these services is incredibly difficult.

3.3 Indicator, reporting flow, and user requirements workshop

In February 2017, Akros, under the guise of NFNC, held a four-day workshop with steering committee members and nutrition stakeholders to determine indicator feasibility, the flow of reporting through ministries, and nutrition reporting requirements for each ministry.

Indicator List

Under ZamNIS, NFNC does not seek to create additional data collection burdens for partners and ministries, but rather to capitalize on the data collection systems that already exist to obtain nationally-representative nutrition-related information. Additionally, while support for nutrition work in Zambia is largely concentrated in areas identified as having a high burden of chronic malnutrition (stunting in children under five years), NFNC envisions ZamNIS as a functional data warehouse for nutrition information from all districts in the country.

Using information elicited from qualitative data collection and through consultation with NFNC, Akros identified key indicators for measuring national progress toward improved nutrition. Source documents for suggested indicators included: (1) The First 1000 Most Critical Days Programme (MCDP) M&E Plan (plus draft indicators for Phase II of 1000 MCDP), (2) the Compendium of Indicators for Nutrition-Sensitive Agriculture from the Food and Agriculture Organization of the United Nations, (3) suggested World Bank indicators for Zambia’s First 1000 MCDP, (4) Indicators and a Monitoring Framework for the Sustainable Development Goals from the Sustainable Development Solutions Network, and (5) WHO global nutrition indicators.

Using this list, Akros asked stakeholders to review indicators relevant to their ministry and/or implementation work to determine the feasibility and current quality of data collection. From this review, indicators were narrowed down and refined according to known capacity to collect indicator data and usefulness of the indicator. Akros sent a finalized list of indicators to stakeholders for feedback (Annex 4). This feedback is listed below with responses to stakeholder recommendations:

● Routine vs survey data indicators: Collection of stunting information (height for age) at facility-level through HMIS is challenged by a lack of height boards. While stunting is an indicator within HMIS, it is likely collected from few facilities, or only those specialized inpatient and outpatient nutrition treatment facilities participating in the Integrated Management for Acute Malnutrition (IMAM) program. Akros recommends creating two separate indicators for stunting, a routine stunting indicator whose data source is HMIS and a survey stunting indicator whose data source is DHS.

● Inclusion of disaggregations for DHS indicators: Should ZamNIS be built, final indicators will be chosen by NFNC in conjunction with the system build contractor.

This list serves merely as a starting point and is subject to change as data collection standards and methods in Zambia advance. As such, indicators listed here are defined by their source and include disaggregations from their source. NFNC does not seek to duplicate data collection efforts, but rather seeks to pull from what already exists in the country. While some sources, such as DHS, include an exceptional number of indicator disaggregations that may not be useful to ZamNIS, they are listed here merely for completeness of information. At system build, NFNC may dictate what indicator disaggregations they would like to receive from CSO.

● Usefulness of indicators: Many of the indicators listed will be collected secondarily through data set exports from primary sources. As such, it requires minimal effort to include multiple indicators from a pre existing data source or information system.

Some indicators may prove more useful to certain stakeholders than others.

However, a large number of indicators with limited usefulness may prove cumbersome and confusing for end-users. Further, stakeholders identified a multitude of indicators useful for ZamNIS, however, they are not currently collected in a systematic manner in Zambia. Should this change, NFNC may choose to include additional indicators or exclude indicators at their discretion at any time during the build or implementation of ZamNIS.

● Completeness and accuracy of indicators: Due to NFNC’s limited resources and a desire for sustainability, the proposed ZamNIS calls for contributions from line ministries, partners, and research entities. As a result, NFNC may provide guidance on collecting nutrition-related information, but ministries and partners are ultimately responsible for the quality of information collected within their own information systems. The quality of information within ZamNIS will be inherently dependent upon cooperation with entities providing data to ZamNIS.

● Indicator gaps under strategic direction areas: After compiling indicators, stakeholders recommended separating indicators into strategic direction areas associated with the National Food and Nutrition Strategy. Once this was done, it became clear that there are substantial gaps in information related to particular strategic direction areas (SD6: HIV/AIDS and nutrition, SD7: NCD prevention, SD8:

Emergency preparedness, SD10: Monitoring, evaluation, and research, SD11:

Communication and advocacy). While industry standard indicators exist to measure a number of these strategic direction areas, many are not implemented in Zambia. As useful and reliable indicators are made available to NFNC, future iterations of ZamNIS may expand upon the compiled list.

● Stakeholder proposed indicators for future consideration and/or iterations of ZamNIS (should standardized data collection procedures be implemented for each):

○ Number of individuals and caregivers provided nutrition counseling

○ Number of children age 0-59 months provided services during child health week (deworming and vitamin A)

○ Number of farmers practicing diversification of high nutrition value crops

○ Number of farmers reporting consumption of high nutrition value crops

○ Number of households receiving dietary diversification extension services

○ Number of food preparation demonstrations conducted focused on high nutritional value foods (including animal-based protein)

○ Percentage of national budget allocated to nutrition used for nutrition activities

○ Prevalence of anaemia in children under 5 (this indicator was originally included in the draft indicator list, but a source could not be identified, not in MICS or DHS)

○ NCD morbidity

○ Immunisation coverage for children under 5

○ Malaria morbidity and treatment rates

Reporting Flow

Akros staff led NFNC partner ministry representatives and nutrition stakeholders through an exercise to determine reporting mechanisms for each entity and the flow of these reports from site-level data collection to national aggregation (Annex 5). Through this exercise, participants identified reporting forms, stakeholders/cadres at each level, responsible individuals for data collection and aggregation, and ministry-specific geographic catchment areas for reporting.

The resulting reporting flow helped to identify specific individuals eligible for data entry roles in ZamNIS and to clarify comparative geographic disaggregation. Unfortunately, a considerable hindrance to ideal granularity in geographic data analysis is an inability to develop relational data points between disparate catchment areas. Until these areas can be mapped and shapefiles created for GIS analysis, Akros advises that the first iteration of ZamNIS collect and display information for comparison at district level.

User Requirements

Through an interactive activity, Akros staff asked workshop participants to place each identified stakeholder within the levels of his or her ministry or organization into user role categories based on decision-making needs related to nutrition. These user roles included the following:

● Write permissions: Ability to enter data, potential ability to make changes to the system (administration).

● View permissions: Ability to view data, may utilize visualizer tools for basic analysis or to create reports as needed.

● Analysis and validation permissions: Ability to analyze the data in ZamNIS and perform data validation to improve data quality.

From this exercise, Akros was able to piece together a user roles and requirements document to support anticipated ZamNIS user needs (Annex 6), and a supervision and surveillance protocol to outline anticipated reporting pathways (Annex 7).

3.4 Data source and technical document review workshop

Using a final draft of indicators and technical documents, in May 2017, Akros and NFNC reconvened steering committee members and stakeholders to obtain feedback and to confirm data sources for indicators. With respect to user roles, participants advocated for enhanced transparency in nutrition data through ZamNIS, advising greater access to analytic capabilities within the proposed information system. Further, stakeholders reviewed remaining nutrition-related indicators without known data collection methods, and either recommended removal or connected Akros with the greater nutrition stakeholder community to confirm data sources. These indicator sources are noted in section 4.2 and in the final indicator list (Annex 4).

4. Final documents and Recommendations

4.1 User Roles and Technical Requirements

The user roles and technical requirements document details the user types, permissions and authorities granted to each user of the proposed ZamNIS as well as the expected functionalities and outputs of the system (Annex 6). User roles are defined by access to the system and dictate how each user can interact with the system.

Recommendations

● While NFNC indicate that they would like access to research data sets, a lack of standardization in data collection, aggregation, and analysis in research studies creates difficulties in data comparison and analysis and may lead to a ‘bulky’ and complicated information system—which often results in diminished user experience, reduced use of the information system, and lack of confidence in data. Should NFNC choose to work with research entities in Zambia to standardize indicators for entry into ZamNIS, this could be implemented in later iterations of ZamNIS. In the interim, Akros recommends creating a searchable PDF library of nutrition research conducted in Zambia accessible on a public facing website for ZamNIS.

● Data aggregation, analysis, and visualization functionalities of an information system are very sensitive to changes in data elements, indicators, and organizational hierarchy. For this reason, the role of ‘Superuser’ should be limited to few individuals with extensive experience managing information systems. Additionally, data entry roles should only be granted to those individuals who understand the indicators and associated data. Limiting data entry users ensures improved data quality.

● As requested by stakeholders, data analysis user roles and functionality should be made widely available to key stakeholders in Zambia who may find the data useful for policy development and program implementation. By allowing increased access to greater functionality of ZamNIS to a wider cadre of potential users, stakeholders aim to increase transparency and support for nutrition efforts in Zambia.

4.2 Indicator List and Data Sourcing

Indicator List

The attached recommended ZamNIS indicator list includes 104 indicators delineated into strategic direction areas according to the National Food and Nutrition Strategy (Annex 4).

Each of these indicators was chosen in a participatory manner alongside stakeholders according to usefulness in assessing national nutritional status and feasibility of collection based on existence of a primary data source. Each indicator provided includes a description, the desired numerator and denominator (if applicable), available disaggregation from the primary data source, frequency of collection, the source of recommendation for use of the indicator, a confirmed indicator source, and the labor required to collect and submit the indicator into ZamNIS.

Data Sourcing

Each of the recommended indicators is associated with a confirmed indicator source.

Indicator sources are primary data points that are associated with preexisting data collection and reporting methods in Zambia. As iterated upon in the surveillance and supervision protocol (Annex 7), each of these data points will be entered into ZamNIS through a combination of manual web-based data entry, direct electronic data transfer between databases, and data set imports through web-based submissions. The surveillance and supervision protocol also details the location and point of entry specifications for each indicator. Below is a comprehensive list of the data sources identified for the recommended ZamNIS indicators:

Ministry of Agriculture

● Monthly Camp Report

● Farmer Input Support Programme Register

Ministry of Community Development and Social Welfare

● Activity attendance registers

● Social Cash Transfer Beneficiary Register

● Food Security Pack Register

Ministry of Finance

● Yellow Book

Ministry of Fisheries and Livestock

● Aquaculture Reporting Form

● Quarterly Camp Report

● Annual MFL Report

Ministry of General Education

● Annual School Census

● Monthly Return Forms

Ministry of Health

● Health Management Information System (via HIA forms)

● Integrated Management of Acute Malnutrition Monthly Statistics Report

● Health Facility Survey/Census

● Human Resources Reporting: Monthly Facility Staff Returns

Ministry of Water Development, Sanitation, and Environmental Protection

● Water, Sanitation, and Hygiene Management Information System (via WASH

Community Champion Data Collection Forms)

Central Statistical Office

● Zambia Demographic and Health Survey

● Living Conditions Monitoring Survey

National Food and Nutrition Commission

● Micronutrient Survey

● Multisectoral nutrition meeting minutes (via WNCC, DNCC, PNCC, etc)

Disaster Management and Mitigation Unit (under Office of the Vice President)

● Vulnerability Assessment Survey

Health Professions Council of Zambia

● Health Professionals Register

Implementing Partners

● Proposed web-based ZamNIS implementing partner data entry form

Institutional Review Board Entities (University of Zambia and Tropical Diseases Research Centre)

● Register of approved nutrition research protocols

Recommendations

● In order for nutrition research approval statistics and published manuscripts to be included in ZamNIS, NFNC will need to develop an MOU with IRB institutions as well as line ministries involved in research approvals related to nutrition (both human subject or otherwise). This MOU should clearly dictate data submission requests from IRB institutions and line ministries to NFNC, as well as protocols and instructions for researchers to submit published manuscripts to NFNC as part of their approval.

● If ZamNIS is implemented, this recommended indicator list serves merely as a starting point for inclusion. As information systems and data collection procedures mature in Zambia, indicators may change and/or expand over time to include those that are more appropriate for assessing the state of nutrition in Zambia.

● Should these recommended indicators be used in a final version of ZamNIS, standardized indicator definitions will be determined by the primary source of the indicator at the time of system build.

4.3 Surveillance and Supervision Protocol

The ZamNIS surveillance and supervision protocol outlines how data will be entered into the proposed information system and NFNC’s administrative response to data entry, or lack thereof, from partner ministries and stakeholders. Further, this protocol outlines existing nutrition-related data collection processes in Zambia, identifies what data will be entered into ZamNIS, who will be responsible for entering the data, and how frequently the data will be entered. This protocol serves merely as a template and can be changed as needed by NFNC and its partners.

Recommendations

● NFNC will need to establish MOUs with each partner ministry and stakeholders contributing to ZamNIS to ensure data-sharing and cost-sharing for the labor associated with manual data entry or data export into ZamNIS. In order for ZamNIS to be sustainable, with respect to financial and human resources, it is important that stakeholders are responsible for their contributions to the information system.

● In the proposed version of the surveillance and supervision protocol, there are currently no automated reminders for reporting. Should resources be made available to implement this, the system can be designed to send an SMS to individuals responsible for data entry or to remunerate those who report on time. Alternatively, the system can send automated emails to users based on the contact provided in their registration page.

4.4 Disaster Recovery Plan Guidance

An information technology (IT) disaster recovery plan provides a structured approach for responding to unplanned incidents that threaten IT infrastructure. This includes hardware, software, networks, processes and people. Some of the possible disasters that can potentially affect technology infrastructure in Zambia include fire, power outrages, and theft, among others. When an appropriate technology is chosen for ZamNIS, a stepwise disaster plan will need to be created that takes into consideration the information system software, the location of the server, and the security of the online instance.

In the event of a disaster, step-by-step procedures will allow NFNC to recover ZamNIS data and functionality, thereby resuming normal operations in an expedited manner. In order to achieve this, the disaster recovery plan must provide both preventative and response measures to withstand a disaster, and to recover the information system and critical technology platforms such as the network infrastructure, server infrastructure, data storage and database backup systems, organizational software systems and database systems.

Below we highlight the key features of a disaster recovery plan and some considerations specific to ZamNIS and NFNC.

Comprehensive IT disaster recovery plan with a step-by-step standard operating procedure:

For each information system in question (in this case ZamNIS), NFNC’s IT team should gather all detailed information about the system, including, but not limited to an overview of the system, hosting details (either in-house or outsourced), equipment configurations, database configuration and backup systems, server information, and application passwords.

Contact details for individuals and/or organizations supporting components of the information system should also be included. The IT team will establish the scope of disaster recovery activities covering internal elements, external assets, third-party resources, linkages to other ministries/stakeholders/technical working groups. The IT team will then craft step by step instructions applicable to disaster scenarios with potential full-data loss. As plans are developed to maintain the integrity of these information systems in the event of a disaster, they should also be regularly reviewed and updated to meet current needs and situations.

The disaster recovery plan should include an activity checklist to ensure NFNC covers all procedures. This checklist should include information regarding IT inventory, network structures, diagrams, backup data procedures, and proposed employee/stakeholder communications.

Backup strategy: A backup strategy outlines the frequency with which data is backed up (daily, weekly, annual or all) and to where that data is kept. It is very important for organization to have known procedure covering backups. Key business processes and the agreed backup strategy must be very clear. The strategy chosen should provide a for a fully mirrored recovery of the site. A strategy that entails the maintenance of a fully mirrored duplicate site (off site), which will enable instantaneous switching between the live site (headquarters) and the backup site, is recommended. A contact person to take responsibility for this activity should be identified and noted in the disaster recovery plan.

Periodic testing in a simulated environment: NFNC must establish a plan or process to update the disaster recovery plan frequently. A disaster recovery plan procedure needs to tested routinely to determine and validate its appropriateness. IT systems and infrastructure change rapidly. As such, these changes should be accommodated in NFNC’s disaster recovery plan and appropriate amendments should be made to standard operating procedures under the guise of NFNC’s IT director. Routine testing allows NFNC to:

● Assess the extent of the disaster and its impact on ZamNIS and NFNC’s services;

● Respond immediately and effectively to a potential disaster;

● Decide which elements of the disaster recovery plan should be activated;

● Manage the disaster recovery team to maintain vital services provided through

ZamNIS and return to normal operation;

● Ensure employees are notified and allocate responsibilities and activities as required.

Documentation storage: NFNC should ensure copies of the disaster recovery plan (both hard copies and digital copies) are stored in secure locations to be determined by NFNC. A master protected copy should be stored in a specific location with limited access. However, it important to ensure the plan is accessible by several IT staff.

Role and responsibilities and contact information for key personnel: A disaster recovery plan for ZamNIS should identify an emergency response team responsible for critical IT infrastructure disruptions. The plan should further outline the roles and responsibilities of all staff assigned to the IT response team to ensure a coordinated and efficient response.

4.5 Technology Solution

A key barrier to NFNC’s ability to fulfil its mandate is limited access to nutrition-related information. Nutrition-related data in Zambia is located in disparate stand-alone databases and information collection systems—both digital and paper-based—and hosted by numerous stakeholders and government entities. Zambia lacks a single repository for nutrition information, as a result, the data that is available in Zambia is not readily accessible and therefore is not used to inform decision-making for improved nutrition.

A nutrition focused data warehouse, such as that envisioned in ZamNIS, will mitigate lack of information by ensuring appropriate data is available to end users in a timely manner. Unlike operational or programmatic databases, a data warehouse collects consolidated and summarized (aggregated) data that are useful in the decision making process.

Data Warehouse Characteristics

In order for a data warehouse to maintain functionality and long-term sustainability, NFNC should consider the following characteristics when evaluating appropriate technology for ZamNIS:

On-line analytic processing (OLAP): A data warehouse that supports OLAP will enhance timeliness of data processing, allowing decision-makers and analysts access to information in real time. Further, the data warehouse should have the ability to aggregate and disaggregate information from diverse sources so as to enhance data analysis.

Business intelligence (BI) tools: BI tools are integral features of a data warehouse that use data within the system to produce meaningful information which ease interpretation of data.

In particular, BI tools illustrate trends in performance that encourage data use for decision-making. Dashboards and scorecards are two such tools. Dashboards collate key information such as graphs, tables, and maps curated from data points in a “homepage” and allow users to easily establish an overview of the topic displayed, nutrition in this case.

Multiple dashboards can be created to illustrate individual data sets within a specific sub-topic, such as maternal nutrition, or agriculture. Similarly, scorecards function as a tool for monitoring performance by color coding achievements towards targets or cut off points in the form of a table. This allows for easy interpretation of program success and provides organizations with guidance on potential changes in programmatic implementation

Data mining and presentation: Data mining is the process of extracting specific information out of a pool of data. Data mining tools are useful for retrieving relevant knowledge from a large repository of data. Data mining tools should work with OLAP to enable fetching and manipulating data in real-time into disaggregate and aggregate data as required by decision-makers.

Extract, transform, and load (ETL) process: ETL describes the steps required for interoperability between information systems. Data extracted from diverse data sources need to be transformed, mapped, cleaned, and quality checked before importation into a data warehouse. The data warehouse technology chosen for ZamNIS needs to support importation of diverse data sources and should easily facilitate the ETL process.

Scalability: A sustainable data warehouse must have the capability to scale both data sets and increased numbers of users over time. As analytic needs of decision-makers and end-users change, this data repository must be dynamic and adaptable to changes in data needs.

Simplicity: End users of a database such as ZamNIS desire the flexibility and power of robust data models, but may struggle with the resulting complexity of the interface. Simplicity is essential to ease of use, ease of manageability, and effective exploitation and application of the data warehouse. Platform architecture should minimize components in order to simplify the management, administration, and use of integrated data. ZamNIS administrators should be able to effectively balance management of the complexities of the platform’s technology with assisting NFNC and stakeholders to effectively use the data warehouse. To fully utilize the data warehouse in the most cost-effective manner, NFNC needs to choose technologies and architectures that provide extensive functionality, allowing ZamNIS to address a wide variety of data needs while minimizing the effort required to manage and use that technology.

Proof of concept: The database platform that NFNC selects to host ZamNIS should make use of technologies that can be exemplified in other extant systems. Existing systems can demonstrate how current database functionalities might meet the needs and requirements of ZamNIS in a tangible manner.The overall objective of the proof of concept is to find solutions to technical problems, such as how systems can be integrated or throughput can be achieved in a given configuration.

Community support: In countries with limited IT infrastructure and human resources for technology, a global community of experts to draw upon for information system implementation support and troubleshooting is immensely helpful, particularly with respect to long-term sustainability of a database.

Culture of implementing organization: Should ZamNIS be housed and implemented by NFNC, it is important to consider the organizational culture of NFNC with respect to management of and support for information systems. Financial considerations will weigh heavily on the choice of technology to host ZamNIS. Long-term funding prospects for NFNC will need to be sufficient to support system build, management, modification, and upkeep of ZamNIS indefinitely. Further, the technology chosen for ZamNIS should support the functionality desired by NFNC as envisioned in their concept note detailing a proposed national nutrition information system. The system should provide NFNC with desired information so as to allow them to fulfill their mandate as the coordinating body for nutrition in Zambia.

Proposed Technology Solution

Given the above characteristics of a desired database, Akros advises NFNC to adopt a web-based information system platform similar to the District Health Information System 2 (DHIS2) that is capable of importing information across data systems, meets stakeholder reporting requirements, and allows for changes over time as nutrition-related data availability and demand change in Zambia.

DHIS2 is a free and open source program developed and piloted over 20 years ago in South Africa by the University of Oslo (UiO) in conjunction with the Health in Information Systems Program (HISP). DHIS2 functions as a tool for collection, validation, analysis, and presentation of aggregate and patient-based statistical data, tailored (but not limited) to integrated health information management activities. It is a generic tool rather than a pre-configured database application, with an open meta-data model and a flexible user interface that allows the user to design the contents of a specific information system without the need for programming. DHIS2 is a modular web-based software package built with free and open source Java frameworks.

DHIS2 is stable and well-supported by UiO and an international network of experts and software developers. Today, DHIS2 is used in over 60 countries in Africa, Asia, and Latin America as a nation-wide government information system, including Zambia’s HMIS and

WASH MIS.

The key features and purpose of DHIS2 can be summarised as follows:

● Provides a comprehensive data management solution based on data warehousing principles and a modular structure which can easily be customised to the different requirements of a management information system, supporting analysis at different levels of the organisational hierarchy.

● Customisation and local adaptation through the user interface. No programming required to start using DHIS2 in a new setting (country, region, district etc.).

● Provide data entry tools which can either be in the form of standard lists or tables, or can be customised to replicate paper forms.

● Provides different kinds of tools for data validation and improvement of data quality.

● Provides easy to use - one-click reports with charts and tables for selected indicators or summary reports using the design of the data collection tools. Allow for integration with popular external report design tools (e.g. JasperReports) to add more custom or advanced reports.

● Flexible and dynamic (on-the-fly) data analysis in the analytics modules (i.e. GIS, pivot tables, data visualizer, event reports, etc).

● A user-specific dashboard for quick access to the relevant monitoring and evaluation tools including indicator charts and links to favourite reports, maps and other key resources in the system.

● Easy to use user-interfaces for metadata management e.g. for adding/editing datasets or districts.

● Functionality to design and modify calculated indicator formulas.

● User management module for passwords, security, and fine-grained access control (user roles).

● Messages can be sent to system users for feedback and notifications. Messages can also be delivered to email and SMS.

● Users can share and discuss their data in charts and reports using interpretations, enabling an active information-driven user community.

● Functionalities of export-import of data and metadata, supporting synchronisation of offline installations as well as interoperability with other applications.

● Using the DHIS2 Web-API , allow for integration with external software and extension of the core platform through the use of custom apps.

● Further modules can be developed and integrated as per user needs, either as part of the DHIS2 portal user interface or a more loosely-coupled external application interacting through the DHIS2 Web-API.

With reference to the above data warehouse characteristics, DHIS2 meets the desired features in the following ways:

OLAP: DHIS2 is an web-based database which allows multiple users to access the system simultaneously. Users can view and analyze data in real time through the web platform so long as they have an internet connection and compatible web browser.

BI Tools: DHIS2 possesses built in visualization and data analysis applications that allow for creation of maps, tables, and graphs that may be used to customize dashboards (Figure 3) for specific program areas. The host of BI tools available via DHIS2 include: standard reports, data set reports, data completeness reports, static reports, organization unit distribution reports, report tables, charts, pivot tables, and GIS. Pivot tables can further be manipulated to create scorecards for decision makers (Figure 4).

Figure 3: A DHIS2 dashboard in a demo instance

Figure 4: An example scorecard developed from the DHIS2 pivot table application

Data mining and presentation: DHIS2 is optimized for speed of data retrieval and analysis through OLAP. DHIS2 provides one the ability to select data based on characteristics (data element, organization unit/location, and period/time) appropriate for the user and to analyze these points utilizing the BI tools discussed previously (Figure 5).

Data mining is further supported through automated features such as validation and reporting.

ETL Process: DHIS2 is often used as an integrated data warehouse because it has the ability to collate aggregate data from various sources which may exist in other software…