IHRIS Lessons Learned
This page is for collecting lessons learned from the HRIS Strengthening and iHRIS Software Project, to be used in reporting results, building a knowledge base and other Y5 dissemination efforts. Please add to this page to build the collection. You can add your own thoughts under each subhead or add additional items as subheads.
National ownership and capacity building must be priorities from the early stages
Most urgent need is human capacity building in both use of eHealth systems and tech support of those systems (Bellagio)
Capacity-building is often needed in the following areas: developing strong data collection systems, using data for planning and managing human resources and providing technical support for the system and its infrastructure.
Training is critical in ensuring that the HRIS adds value and remains sustainable. Training promotes sustainability through improvements in data analysis and IT skills, as well as through encouraging the use of data for decision-making and problem solving. Finally, training supports the use of practices that improve data accuracy and quality. Such measures are essential, as an HRIS is only useful when data are reliable and valid.
There are too few input staff in the Councils to cope with a combination of staff losses due to sickness, absence and annual leave. This will compromise the quality of information available to senior management on the healthcare sector. Implementation in some districts is held back by the lack of basic computer skills, electricity supply and the shortage of computer equipment. (Uganda)
Training is needed; cannot introduce a new tool and expect people to automatically know how to use it. There is a lack of understanding of the system’s abilities, security, back-up plan and the existence of a Windows version that does not depend on the Internet. (Uganda)
Comprehensive stakeholder identification and involvement from the beginning
The creation of a Stakeholder Leadership Group encourages the exchange of information between users of HRH information and improves the overall utility of the HRIS.
Stakeholders don't know about each other and available data sources. For example, the Chief Nursing Officer in Tanzania expressed dismay at ever knowing how many nurses worked in the public sector or how well qualified they were. One floor down, in the same building, in the TZ Nursing Council, we found extensive work on an Access database with all licensed and registered nurses in the country. Brought them together in the SLG. (HRIS Strengthening presentation)
Data frequently exist in small datasets or paper files that are not necessarily known by all stakeholders. Bringing together HRIS stakeholders, often for the first time in the same room, ensures that information is shared and helps the group quickly reach consensus for making good use of the data that do exist.
Key issues for stakeholders to address are data ownership, policies for data sharing and policy questions that the HRIS will answer
[Need examples.]
Implementing HRH interventions take a long time to change practices and show results =
One major lesson that emerged from this workshop is that people generally know what is wrong with their HR system, they are able to generate a plan and possible solutions – but they tend to get stuck at the implementation stage. Perhaps, this is the stage where we need to focus our attention more and begin to provide direct and sustained assistance with the implementation of specific activities or policy issues.
Place priority on ensuring the accuracy and authority of the government system for public service data.
When an HR system is mandated for a government health sector by a country’s government, the best practice is to strengthen the health sector’s use of the mandated system. In the case of a government-wide system, parallel systems (such as data warehouses) may be used to fill essential gaps.
In Namibia, the Office of the Prime Minister (OPM) mandated an Oracle-based public-sector-wide HR system. The OPM launched this system in the 1990’s and requested that the Namibia Ministries, Departments and Agencies (MDAs) begin entering data. Despite having 20% of the health workforce, the Namibian MOHSW had still not begun entering data or working with the system in 2007. They also had not participated in the specifications development process. The Capacity Project, in response to stakeholder requests, evaluated the options of implementing an independent system vs. supporting the mandated system. Where countries have implemented a government-wide system, the intent is to link payroll and personnel records for the public sector. When this happens, there is a natural incentive for health workers to support its accuracy. In addition, supporting the OPM system to meet the reporting needs of the MOHSW rather than developing a parallel system reduces cost and reporting burden for health workers. Unfortunately, due to the late involvement of the MOHSW in the specifications development process, it is not possible for the first phase of the system to collect and report on data of value and interest to the health sector (such as health competencies). Despite this, the initial system is viewed as a logical first step that can be expanded in the future to meet their needs.
In Botswana, the Department of Public Service Management (DPSM) has mandated a similar system for Botswana MDAs linking payroll and personnel records. This system is in the formative stages, and the Capacity Project, in partnership with BOTUSA, is supporting the MOH to get involved in the planning early to be able to ensure their needs are met by the planned system.
In both Namibia and Botswana, the Oracle HCM system will only capture the public service information, but it will capture this public service information across all ministries. This is especially vital for Botswana, as district health workers (more than half the public sector health workforce) are managed through the President’s Office Regional Administration and Local Governance (PORALG). Health workers also exist and are managed in at least three other MDAs (mining, military, police force and jail system. A comprehensive public sector system will provide a complete view of the public sector health workforce.
As an administrative system for the public sector, Namibia and Botswana HCM systems do not include the private and FBO sectors. Ministries of Health need a complete picture of the health workforce in order to best plan for the country’s needs. This will eventually require a parallel data warehouse to capture private sector data and analyze it along with data from the public sector HRM system. The HRIS Strengthening activities in both countries are encouraged to consider an iHRIS Manage deployment when the time comes to serve as a data warehouse for this information.
Standardization of data, within the country and at the international level, is key to increasing use and effectiveness of data.
In order to integrate certain types of data, inter-ministerial agreement is required on standardization of codes, such as locations. (Uganda)
Using Previously Collected HRH Data to Create a Data Coding Scheme
Capture data at the district level and use it there. Aggregate data at the central level for comprehensive country analysis and support.
From the beginning of Capacity’s work in Uganda, the Project has been interested in developing a personnel system for the Ministry of Health. At the time the project started, the European Union (EU) had a large HR systems project underway to address this need. However, the Nursing Council was very interested in developing a stronger Registry System to track nurses, their training, registration and licensure across the country. Through a successful effort to build a Nursing Council system using Capacity Project’s iHRIS Qualify, the Project was able to extend this work to all four health professional’s councils, where data-entry is now complete and reports are being generated. This work has most recently led to a replacement of the EU system at the Ministry of Health by iHRIS Qualify, with initial data entry well under way. For true value to be realized, however, it is understood that data needs to be entered and used at the district level, with regular aggregation at the central level for a comprehensive view. Pilots are already planned in several districts after the release of the new software.
Rwanda has recently brought its own iHRIS Manage system online with data entry underway. District pilots are planned for early next year.
The iHRIS Manage system will be rolled out to additional districts this year which will improve the timeliness and accuracy of information and will reduce workload at the Ministry. However, this will require the production of regular reports to senior management that analyze workforce trends and interpret their significance to the delivery of healthcare programmes. District data entry will reduce the flow of forms to the Ministry of Health and its data-entry burden. District staff can be slow at inputting data due to lack of basic computer knowledge and poor typing skills. Too few staff have been trained to accommodate sickness and other absences, which affects the continuity of data entry and sustainability of the system. (Uganda)
Establish an interoperability plan with systems to reduce redundancy and associated costs and data conflicts.
Wide agreement among participants that emerging infrastructure and eHealth systems presented a window of opportunity to foster interoperability across geographies, technologies, and programs (Bellagio)
Other Health Information Systems (such as the HMIS) often include or report on HRH data, and frequently offer data of value to HRH planning, training or deployment.
Software industry-standard system architecture frameworks strongly encourage the cost-savings and efficiencies of common use applications and shared data. The following is excerpted from the recommended example architecture principles of The Open Group Architecture Framework (TOGAF) a rapidly growing architecture framework for rationally developing large-scale systems:
“Timely access to accurate data is essential to improving the quality and efficiency of enterprise decision-making. It is less costly to maintain timely, accurate data in a single application, and then share it, than it is to maintain duplicative data in multiple applications. The enterprise holds a wealth of data, but it is stored in hundreds of incompatible stovepipe databases. The speed of data collection, creation, transfer, and assimilation is driven by the ability of the organization to efficiently share these islands of data across the organization.” (Rationale for Principle 10: ‘Data is Shared.’)
Rwanda’s health system is currently experiencing these ‘stovepipe’ databases with seventeen separate data collection forms. The Capacity Project is developing Rwanda’s HRIS to minimize the development of new forms, and actively participating in their strategic efforts to rationalize and harmonize the various eHealth activities.
The iHRIS Manage system is not linked to the MOF, and the HRM Department is still doing payroll manually, which leads to delay in salary payment and inaccuracies. (Uganda)
Free and Open Source software offers best supported and most cost effective model
FOSS offers the best supported, most cost effective model. (Need examples.) (HRIS Strengthening presentation)
Several of the informants felt that because the software is Open Source and therefore requires no licensing fee and is free from viruses, it contributed to sustainability. (Uganda)
Flexibility and adaptation of software is key
Experience suggests that in order to be most useful, HRIS solutions should be designed around country-specific needs, rather than based on a generic, “one-size-fits-all” structure. A step-by-step process that allows for progress at a gradual pace will likely be more effective than a rushed process that calls for rapid changes with little time for adaptation.
HRIS solutions should be designed around country needs as identified by system stakeholders. The HRIS must be flexible and adaptable in meeting emerging needs after it is developed. Use cases have proven to be an effective method of prioritizing stakeholder requirements and communicating those requirements to system developers.
Human resources information and administration requirements are constantly changing. This therefore requires a continued investment in the form of hardware, maintenance, training and programming, even when the current source of funding from the Capacity Project finally ceases. (Uganda)
An effective system is one that is used
In order for a system to be effective, it must be used. A key step in strengthening HRIS is the creation of a culture of routinely using data as a basis for decision making.
Data-driven decision making is a journey not a destination: we are talking about a process for making effective collaborative decisions. This process is characterized by three functional areas:
- collection, integration and dissemination of data to answer policy and management questions;
- analysis and reporting of data
- process and procedures for acting on the data.
In other words, it is not enough to make data available. There must be a process in place for analyzing the information and getting it to the right decision maker at the right time with the power and resources to act on it. Process helps to create frameworks that people can understand, follow and, most importantly, repeat. And it does not have to be complicated. It’s simply showing who is responsible for each step, empowering teams and determining what decisions need to be made. It’s only through such a process (started by this workshop) that a sector-wide culture of inquiry that values the power of data and working together productively to inform sound decision making and improve health service delivery can be established.
Perhaps the most important part of data driven decision making is enabling decision makers to use it: colorful reports, graphs and pie charts will have no effect unless they are combined with active leadership, change management and effective professional development for the key decision makers. Even in Uganda, senior planners and administrators need training and hands-on mentoring and coaching in continuous improvement processes and how to read HR data and apply it to their health sector-wide goals and objectives. It’s a mental shift that can only take place over time – and it cannot be accomplished through a one-off workshop. In fact, most of the positive changes that we are looking for in most parts of the public health systems in Africa are mental and not necessarily governmental.
In Uganda, there is considerable HR data and even “decision-ready information” residing in individual silos – the challenge is to move all these data to relational databases through a secure network to store and access data and present it to decision makers.
Importance of backup plans
As another country sends an emergency message for assistance restoring their data I’m reflecting on the importance of ensuring sustainable back-up plans are in place. Not only must we assist local system administrators with devising an affordable solution we must also automate if possible and check back for status as often as possible. Too often we see back-up plans which are insufficient or exist only on paper, not actually in practice. We have also seen back-up plans which are followed for a period of time and then abandoned due to busy staff or a false sense of security due to a period of time with no system catastrophes. Borrowing from the words of a colleague, back-ups are only useful if they can be used to obtain the required data. And so as we perform regular check-ups to ensure back-ups are conducted in a timely manner we should also inquire as to how often they test the back-up data to ensure it will be useful if needed. --Angela
Collaboration and sharing of ideas and experiences across countries
Keeping the team informed of progress in one country and how to use that experience/information to make lives easier with similar projects in other countries. For example, one HRIS Advisor may plan a DDDM workshop but that knowledge wasn't necessarily transferred to the team such that another Advisor could learn from the experience and mistakes in order to host a DDDM in another country (Angie)
Inter-country sharing prevents mistakes from being repeated, leverages successes and ensures continuity. (HRIS Strengthening presentation)
HRIS leadership
I found it challenging to assert leadership in some places perhaps due to being new. For example, in Namibia Dykki was introduced as the HRIS expert in the assessment interviews. In later implementation work people often asked for Dykki and thought they needed to work with him directly. I'd like to find a better way to keep Dykki involved in the country field support activities without taking any leadership away from the technical lead. (Angie)
Along the same lines, the line of command isn't/wasn't always clear. I still find it somewhat confusing to know when I go to Dykki for approval vs. when I need the approval of the country point person vs. when I need the Chief of Party to approve. In some cases Dykki would approve and then the country point person would find out and give me a different answer. This didn't happen often but I think it was often enough to illustrate that not all of us understand the chain of command and responsibility. (Angie)
Code management
Code management takes a lot of time. Needs a full-time code manager. All code needs to be reviewed before being incorporated into core released software. More time needs to be allotted for documentation. Also need someone (not a developer) to act in a testing/quality assurance role.
Documentation of our HRIS strengthening process
We have some of this now but it was difficult to learn at first until these docs were put together (Angie)
