HIE Library Papers

updated: 09 March 2006

Jeremy B. Clark
Information Systems Advisor to the Namibia Ministry of Health and Social Services

Some observations and thoughts on the phenomenon that it may be difficult to encourage health workers, particularly at remote facilities, to gather, record and report routine data.

1. Most health information systems - the routine ones that collect data from health workers based at facilities throughout a country - have been implemented because managers, bureaucrats and technocrats at the national level think they need to have data that can only be gathered at the level at which services are provided. The national level people may, indeed, legitimately need such data for planning, budgeting, or for responding to requests from inter/multi-national organizations.

2. Historically, in many (dare I say most?) cases, the data to be collected by remote health workers is decided by the bureaucrats and technocrats at the national level, often with advice, even demands from the inter/multi-national partners or organizations, but with no or only token consultation with health workers at the level from which data is being collected. It is unlikely that the staff at the national level has ever worked in a remote health facility; perhaps they have worked at a district level; perhaps they have been on a team that made supervisory visits to remote facilities.

3. If there is any capacity to compile, collate, and make any sense from data reported from facilities, it is available at the national level, so raw data is sent to the national level. The likelihood for feedback to (or promptness/usefulness of such feedback) the source level decreases the further away analysis is done from the source level.

4. In many (most?) cases the remote health worker is disinterested in spending time and effort gathering, recording, and reporting data because much of that data has little or nothing to do with his/her job or the resources allocated to him/her or the facility. What is the motivation? No buy-in, ownership when the decisions were made about what data to collect (or the likelihood that complete and accurate data could, indeed, be collected, given reality at the facility).

In contrast to health services reports, drug or supplies orders are usually accurately and promptly completed and submitted because doing so is (often) closely related to the capacity for the facility to do its job. If money is collected, reports and accounting for it are done promptly and accurately because of periodic visits by the finance people who collect the money and audit the accounts.

The remote health worker has little reason or stake in collecting and reporting service or diagnostic data. His/her basic training or in-service courses, particularly on the EPI or community health services, is likely to have included a unit on keeping statistics and preparing wall charts to monitor immunization coverage or a disease or service of interest to the community.

5. The remote health worker is likely to receive few supervisory visits from district, regional or national level supervisors. Those visits are likely to be brief, admonitory or punitive in their attitude. The "check lists" that are likely to be used during these visits are unlikely to touch on data collection and reporting. The visitors are unlikely to have consulted data that has been dutifully reported before paying a facility a visit. If there is a question about a data matter, the health worker must trot out the copies of the reports that have been retained at the facility. The message here clearly is that the data a facility reports is not consulted or used at higher levels.

6. A well designed routine health information system should include the capacity and some sort of encouragement for facility-based health workers to use at least some of the data they report in their own facility. Using monthly EPI data or ANC data to chart coverage, or monitoring the number of cases of a disease, the number of new latrines, births, or something that is likely to be of interest to the community should be encouraged.

7. Remote heath workers may complain that they receive no feedback on the data they routinely submit. This clearly is the fault of staff at the closest administrative level, usually the district. Useful or meaningful feedback - particularly comparison of performance - patients seen, immunization or antenatal coverage, etc. indicators with neighbouring facilities in the district must come from the district level. If the information system depends on a district only to pass along unprocessed data to the next level, then district staff can hardly be blamed for not providing feedback.

8. A routine health information system must route report forms from facilities through the hierarchy - district --> region --> whatever. Reports from facilities should never be sent directly to the national level. The district and whatever intervening levels should all have some part in and responsibility for handling/processing data report forms, even if such an approach lengthens the time data gets to the national level and introduces the possibility for report forms to go astray. No bureaucracy has interest in forms that it does not process.

A routine information system should provide some measure of capacity and responsibility for data processing and analysis at the most possible remote level. If there is computerized capacity at the district level, then data entry should be done there and standard reports that analyse and provide comparison of data reported from all facilities in the district should be available at that level. Reports that compare similar-type facilities in a health district are fabulous to provide feedback to all facilities in a district.

Even if data entry into a computerized system is not available or possible at the district level, extracting, compiling and feeding back to facilities of two or three indicators as the report forms pass through the district will provide useful information for district administrators and a sense to facility-based workers that a) their data has been received and used for something and b) the ability to compare their facility's activities with those elsewhere in the district.

This same process should be included at every administrative level up to the national level, i.e. regions should provide feedback and comparative data/reports/information to their districts, and the national level to regions, etc.

9. Years ago, in Jamaica (I do not know if they are still doing it) the EPI instituted an annual "competition" for an inscribed trophy cup that was awarded to the district that had most improved or sustained its EPI coverage. The formula used to calculate points for the award included recognition for promptness and completeness of reporting, not only population-weighted coverage improvement or sustainability

The annual award of the Immunization Cup became quite a competitive event, and it contributed to improvement of promptness and completeness of reporting.

10. At a meeting recently to begin to develop a social welfare information system in Namibia, a participant from a very rural district suggested that the system should include the capacity for districts to receive data files for the entire country. She said that her health colleagues (who use the Namibia Health Information System, which depends on people to generate and provide feedback) are often frustrated because they are able only to see and use data from their own district (or region). Technically this presents some problems, but I feel that such feedback should be built into routine systems (if they are computerized).

11. The Internet has the potential to provide prompt and up-to-date feedback to anybody who has a reasonable connection for e-mail or to visit a site from which they may retrieve information they are interested in.

While there is lots of talk about using the Internet for such purposes, I fear that it will be a while for it really to happen. It will be a long time in many countries before reliable, affordable access (or computers) are available at the facility level.

The Namibia HIS has received support from UNICEF to support the registration of a domain (healthnet.org.na) and dial-up Internet/e-mail access because the government's alternative was unreliable, so that data files could reliably be sent FROM districts to regions to the national level. Every level (district, region) that has adequate telephone service has access to and an e-mail address on this system. The accounts allow lower-than-normal-cost dial-up rates from anywhere in the country, and unrestricted Internet access.

While we have made good use of Internet capacity (posting updates to the HIS software rather than circulating diskettes, offering monthly anti-virus definition file updates, etc.), to date the only "feedback" capacity we have used is routine e-mail messages to advise HIS users of problems with the software, work-arounds, announcements of useful or interesting Internet sites, etc. We have not used the potential for downloading national data files that could then be used at the district or regional level to provide comparative analysis.

But even with the availability of Internet access, many districts and regions are not using it, preferring to use the capacity to send data files by diskette. And we find that messages sent by e-mail may not be downloaded for long periods.

Jeremy B. Clark

Additional biographical info: Since 1997 (until July 2002) I have been the Information Systems Advisor to the Namibia Ministry of Health and Social Services. I first came to Namibia (on a UNICEF contract) in 1994 to serve as the (UNICEF) consultant for the development of a national, decentralized health information system. In 1996/97 I spent just over a year in Eritrea helping to develop a similar system. My previous experience has been in Jamaica, where I was a Peace Corps Volunteer serving the Kingston Public Health Department, then worked as a PAHO consultant, then staff member at the office that provided services to Jamaica, Bermuda, and the Cayman Islands. I have also consulted on health information systems projects in Belize and the Turks and Caicos Islands. I have an MPH from UNC Chapel Hill, and got my start in public health administration on a Robert Wood Johnson Foundation project that had me as administrator of a rural community-owned health facility in the mountains of Western North Carolina.