eGovernment for Development

eGovernment for Development
eHealth Case Study No.4

Long-Term History of a Health Management Information System in "Ganesh Medical Institute"

Case Study Authors

Anonymous

Application

"Ganesh Medical Institute" (GMI - not its real name) is one of India's major medical teaching and research facilities with an attached hospital of several hundred bed capacity. In 1989, an application of computers for patient registration was piloted, leading to development from 1993 of a fully-fledged health management information system (HMIS).

Application Description

The HMIS was built with five main functional areas:

• Patient registration - registers the 4,000-5,000 out-patients and 100-150 in-patients who attend the hospital each day, and allocates a unique ID number to each one that is used track their transactions in labs and wards.
• In-patient data - records results for patients from pathology, microbiology, radiology and biochemistry specialisms, and stores blood bank details.
• Operation scheduling - for use by all surgical specialities to plan and record operations, including checking for blood availability and generating operating theatre lists.
• Discharge summary - provides a printed summary report when the patient is discharged from the Institute, particularly for use by supporting medical services.
• Ward routines - provides computerised support for planning and recording of nursing schedules and other ward-based activities.

Role of ICT

The patient registration pilot in 1989 ran on a small LAN with three terminals and a 286-based server. During 1993-96, this pilot was expanded to provide 64 terminals throughout the Institute (wards, major labs, clinical services, surgical services, ancillary services), running the central applications on two 486-based servers. In 2001/2, the central server was upgraded to a PIII machine. A system enabling academic access to the Internet, using 25 PIII clients plus a high-end server was introduced at the same time, but remains unconnected to the HMIS network.

Application Drivers/Purpose

In an overall sense, some drive for the system has come from GMI's wish to be seen as one of India's leading medical institutions, with an expectation that this would be reflected in its adoption of leading-edge management information systems. More specifically, though, the drive came from one individual in GMI who was bitten by the computer bug in the 1980s and wanted to see computers put to good use in the Institute. From the early 1990s, this drive was supplemented by the advocacy of the regional director of India's National Informatics Centre, a central government organisation that undertook most of the software development work for the HMIS.

Stakeholders

The core stakeholders are seen as patients and their families; medical professionals (doctors, nurses); and specialist staff (especially in GMI laboratories and in medical records). As noted above, the National Informatics Centre was involved. So too was the World Health Organisation, providing a grant for development of the pilot registration system.

Health and the Poor

The HMIS cannot really be said to have made much difference to the lives of the poor. Had the system been developed as intended, it would have provided a mechanism to reduce the number of visits by poor patients, e.g. to collect laboratory reports. However, this has not happened to date.

Impact: Costs and Benefits

A WHO grant of c.US$15,000 was used to create the pilot registration system. The first phase of the HMIS cost around US$100,000 for computer hardware plus c.US$50,000 for cabling and networking. Consulting and software development costs were absorbed by the National Informatics Centre. (A second phase plan to expand the system to a total of 192 terminals was not implemented.) The hardware upgrades in 2001/2 cost around US$20,000.

The status of each of the five HMIS functional areas by the start of 2004 was as follows:

• Patient registration - this continues to be operational after more than 10 years; it is popular with medical records staff since it saves the time and effort of an earlier system which required them to create aluminium ID tags for all patients (a considerable task given the thousands of patients who attend each day).
• In-patient data - staff in the pathology lab and in the blood bank continue to use the system: path. lab results for all referred patients can be accessed anywhere on the system, as can information on general availability of blood supplies and specific blood allocations for individual patients. GMI's other specialist sections - microbiology, radiology, biochemistry - have stuck to manual record keeping and report writing.
• Operation scheduling - the system is only used by the plastic and cardiothoracic surgical teams; all others have stuck with manual working methods.
• Discharge summary - only used by around 10% of hospital wards: those which have a functional printer attached. All other wards have no printer or a broken printer and are stuck in a classic "Catch-22" situation: GMI management will only purchase additional hardware if system usage rises from its current low levels, but usage cannot rise unless new printers are purchased!
• Ward routines - now unused following an initial burst of enthusiasm when all 400 nurses were trained. Some nurses were still reluctant to use the system; others did but then left or were promoted. New nurses were not trained and in most wards the UPS power supply points for the HMIS terminals have been moved by nursing staff to support other electrical appliances.

In all, then, the benefits of the system have been very limited.

There has been some development of in-house awareness and skills in GMI, not just the ICT operational skills that staff have required for using the system, but also some higher-level skills: for example, while in-patient registration was developed by NIC, it was GMI staff who developed the out-patient registration module.

Evaluation: Failure or Success?

Two system audits have been conducted - an internal one in 1999/2000, and an external one by a leading Indian software consulting firm in 2003. It is the results of these audits that are described here. They found that most key players still operate in 'manual mode', leading to an overall evaluation that the HMIS has been largely unsuccessful. However, there are now plans to upgrade the system - both hardware and software.

Enablers/Critical Success Factors

1. Enthusiastic hybrid supporters . The core supporters for the project were medical academics who taught themselves IT skills (partly through a "Doctors' Computer Club" that was initiated in the late 1980s). By combining medical and IT competencies in a 'hybrid' manner, these supporters were able to understand key developmental issues from both perspectives.
2. Support from other sources . GMI administration was initially enthusiastic about the project during the early 1990s once they had seen the pilot project. They therefore provided political and material resources to assist the project. The project would also have been untenable without the support from NIC's regional director: he was keen on publicising the value of IT systems in Indian hospitals to ensure that not all infrastructural spending in medicine got channelled towards traditional expenditure such as upgrading radiology equipment.

Constraints/Challenges

1. Absent enforcement during critical junctures . GMI administration provided the support necessary to see the HMIS installed. However, they did not enforce use of the system during a critical period in 1994/5 following implementation. Success of the system at that point depended on pushing hospital staff to abandon their manual working methods, and swap over to the computerised system. But many staff - despite their IT training - felt uncomfortable with the new system and were reluctant to use it. Yet management pulled back from forcing staff to toe the line. As a result, the new system lost impetus and came to run in parallel with the manual system, rather than supplanting it.
2. Staff turnover . In the mid-1990s, the intensive training programme achieved 100% coverage of GMI medical staff. However, around 150 junior doctors leave GMI every year, and up to half that number of nurses also depart. There needed to be an ongoing training programme to cope with this situation but by 1996 there was no training at all: NIC staff had moved on to other projects while computer-literate staff in GMI felt they needed to focus on their main hospital jobs.

Recommendations

1. Plan beyond installation to ongoing operation . Too many e-health projects plan up to the 'grand opening ceremony', but think no further: about how funding is going to be maintained for the system; about how skills are going to be maintained; about the need to upgrade the system on a regular basis to cope with ongoing technical change; about how to truly institutionalise the system. All of this needs to be considered as part of the preparation process.
2. Identify stakeholder incentives . Stakeholders involved with e-health systems always ask themselves "Why should I: what's in it for me?". In this particular case, there was no good answer for too many of GMI's staff - they therefore stuck with, or reverted back to, the manual method of working. As well as thinking about technology, e-health planners must also work out a system of incentives or disincentives for key stakeholders that will ensure that they use the new system. In the initial instance, the carrots and sticks might need to be financial if that is feasible within the public sector.

Further Information

n/a

Case Details

Case Editor : Richard Heeks.
Author Data Sources/Role : Project Management Role.
Outcome : Largely Unsuccessful.
Region : South Asia. Start Date : 1989/1993. Submission Date : January 2004.