Study of the barriers to the implementation of computerised information systems in the National Health Service

Abstract

In order for computerised information systems (CIS) to be utilised to their full potential they must first be successfully implemented. The acquisition and implementation process continues to be an area which is reported by the media to present problems for both public and private sector organisations (Keen, 1994; Collins, 1996). The process was originally considered to be a technical process only, but more recently there has been acknowledgement that there are human implications. The process of acquisition and implementation of CIS in the NHS has, therefore, been studied in order to identify the barriers which might be present. The research commenced by conducting a comprehensive literature search which showed some of the barriers found by previous researchers. A number of theories were examined which it was thought would be helpful in approaching the subject. A change model was then identified (MIT9Os, Scott Morton, 1991) which was used to structure the study and as an aid to analysis. The model would be examined for its utility as a change model in the NHS setting. The research used both quantitative and qualitative methods of data collection. A macro view of the process was initially sought because this approach is seldom taken (Kwon & Zmud, 1987). The major part of the research consisted of 4 case studies and 2 survey questionnaires. One survey questionnaire was sent to IT/IS directors and managers about the process. It was sent to eight NHS Regions (359 questionnaires) and a response rate of 51.5% was attained. An additional questionnaire was sent out to human resource directors in the NHS. This was sent to 400 directors and the response rate was 48%. The study identified a number of barriers to implementation in different elements of the organisation, one of the most important barriers being related to politics/power. It was found that the reasons for acquiring new systems are not always articulated, and if these are not in the interests of powerful stakeholder groups, then implementation may be more difficult. Labour process theory was helpful in examining this aspect (Dent, 1996). Results show that existing models are inappropriate for the majority of CIS implementations. The MIT9Os Model (Scott Morton, 1991) was adapted, adding the element of politics/power, together with money and time as major constraining factors. The element of strategy was given a more prominent position, to indicate a pivotal role and it is argued that the model should not be an equilibrium model because of the constant change necessaiy in the modern business environment. The previous model was expanded to show some of the important issues and questions which need to be addressed by those approaching implementation. However, the research showed that the complexity of the process precluded any simple prescriptive answers to implementation problems being given. Models are little used by practitioners, but if they are used they need to carry a warning note that they are only an aid to preliminary thought, and much other background reading and analysis of the particular situation needs to accompany them.