Degree Name

Doctor of Philosophy


School of Information Technology and Computer Science


This thesis is a study of the design and diffusion of Computerised Health Records (CHRs) among General Practitioners in Australia and Sweden. CHRs are essentially a social (de)construction of meanings and artifacts by actors in social networks through time. These social networks can further be conceptualised as networks of actors belonging to various language or discipline trees, for example, health informatics. Conflict over ascribed meaning is therefore inherent in any political process of social change as not all actors are homogenous. Thus, conflict between actors and actor networks is reflective of the continuing struggle to gain control over the meaning and organisation of society, social structure. Artifacts can be seen as the physical manifestation and imposition of this power struggle. A socio-political approach is therefore, a philosophical term used to describe the afore-mentioned process; technical issues are social, political, economic and legal issues.

CHR design and diffusion is not just a technical design issue confined to a computer programmer's or general practitioner's desktop but encompasses a multitude of broader social design concerns which are inherent reflections upon the political motivations of the actors involved who shape and diffuse the technology in networks. A greater merging of macro and micro sociological views of technology and health is still needed within the health informaticsfield to form a more holistic view of events. Furthermore, CHR diffusion and health informatics need to be viewed and understood as belonging to a more broader landscape of the theories of technology, epistemology and society. The development and diffusion of CHRs (or lack of) are not only a reflection upon the motivations and attitudes of the actors within a social system but upon government policy and funding.

A socio-political approach to CHR design and diffusion among General Practitioners (GPs) is an attempt to try and move away from studying technology just as the design and diffusion of an independent technical or medical artifact but to position the process of design and diffusion within a broader social, political, economic and legal framework in order to help understand the interrelationship between health, technology and society. Therefore, health informatics rather than medical informatics is more reflective of this broader shift in minking.

The comparative study reported was undertaken so as to verify the state of adoption of CHRs among GPs in Australia and Sweden. Responses were gained from a mailout questionnaire to random samples of GPs in both countries (N=600/country). This comparative study is a unique contribution to the health informatics literature, which adds to the existing body of knowledge about CHR design and diffusion, by way of a cross-cultural comparison of GP adoption rates. As a result, some concluding comments and recommendations are offered to assist in the understanding of high and low diffusion rates of CHRs among GPs and the implications for health policy and the formulation of information technology adoption strategies. The mainfindings of the survey conducted indicate that there has been a high rate (72%) of diffusion of computers and CHRs among GPs in-Sweden and a low rate (14%) of diffusion among GPs in Australia. Moreover, use of computers by Australian GPs is still predominantly confined to front desk type applications (e.g. accounts/billing, word processing) as opposed to clinical CHR use (e.g. patient notes, script writing, recall and referral, test ordering). In Sweden, CH R adoption can be seen as more of a result of direct funding availability (a type of authority decision) mandated by governments (local and national) while in Australia the decision to adopt CHRs is more of an individual optional or collective group practice decision.

Results from the survey data indicate what can be called an attitude-behaviour paradox. In the Swedish survey, among the non-users, a positive attitude is associated with an intention to adopt in the near future (within the next 3 years). In the Australian survey, even when the non-users indicate to having a positive attitude they have no intention of adopting CHRs. This finding supports the general belief that new knowledge and attitudes by themselves are not necessarily sufficient to bring about a change in behaviour. Other socio-political reasons need to be considered rather than just behavioural or technical reasons alone. A lack of computer literacy and training was cited as a key barrier to the adoption of CHRs by GPs across both survey samples. The mean year for GP medical education completion for the Australian and Swedish GPs surveyed was 1974 and 1977 respectively. Very few had to undertake any computer related subjects as part of their medical education. A need also exists to clarify social, legal, economic and political debates over a CHR standard, CHR ownership, CHR legal status, stakeholder access rights and responsibilities as well as GP loss of power issues and government funding models.

This research has important implications for public health policy as well as theoretical debates about the process of technology design and diffusion. The health of nations is as equally important as the wealth of nations. Health care provision and research can be seen to underpin the economic well-being of human existence. The design and diffusion of CHRs in primary health care offers an opportunity for changing the organisation and process of health care activities, however, there will always be associated consequences as a result of such change. Epidemiology research in Australia at the general practice level represents a relatively under utilised resource which could be used to help shape the debate over public health policy. The adoption of CHRs can be seen as an opportunity for helping to change this situation at the practice, as well as regional and national levels.

The diffusion framework as advocated by Rogers (1995) needs careful re interpretation and repositioning to reflect changes which have occurred in the social structure of society. Traditional classical diffusion thinking is bound up in its historical roots of social anthropology and rural sociology reflective of the 1930s and 1940s. Classical technology diffusion studies treated social networks more as closed measurable systems in which actor relationships were identifiable. Essentially, social diffusion networks need to be seen as open systems, consisting of both physical and electronic social networks in which actor relationships are not necessarily identifiable and obvious. The focus therefore is not so much upon the spread of the artifact itself, as in past diffusion studies, but more on the spread of information about the artifact and how it is communicated among actors. Therefore, theoretically, CH R diffusion among GPs should be posited in an information and communication network context. Criticisms of classical diffusion thinking need to be taken into account, especially when designing further diffusion research studies.