Dr Anne Young: General practitioner utilisation among women in Australia.

The purpose of this thesis is to explore the determinants of the use of general medical practitioners among women in Australia. General practice is the first point of contact for the provision of health care services in Australia and is the gateway to more intensive and specialised services. Despite the existence of a national health insurance scheme (Medicare) there is growing debate about the equity of access to health services, especially for people living in rural and remote areas of Australia.

This study examined the use of general practitioner (GP) services during 1995 and 1996. The framework for the study was the behavioural model of health service utilisation developed by Andersen, Newman and Aday and others which includes measures of medical need and other individual and societal factors that may predispose, enable or impede use of services. Survey data for 20,000 participants in the Australian Longitudinal Study on Women's Health were linked with data from the Health Insurance Commission (HIC) which administers Medicare. The survey data included a range of questions designed to explore social and environmental aspects of women's lives as well as the psychological and biological determinants of health and health care utilisation. The HIC data provided measures of GP utilisation for each woman and the out of pocket cost associated with each consultation.

Using the linked database, the demographic characteristics, medical history and health service utilisation of very frequent attenders and non-attenders to general practice were described. These data provided insights into the use of GP services that could not be determined from either source alone. Although some findings were consistent across age groups, the profile of frequent attenders differed according to life stages such as pregnancy and menopause. A few case studies highlighted the difficult personal and social circumstances of some women and demonstrated the importance of considering the context in which use of health services took place.

A third source of data for the research was a sub-study to provide additional measures of individual and environmental determinants of health service use and satisfaction with GPs among a sample of almost 5,000 women participating in the longitudinal study. This sub-study provided strong quantitative evidence to support the qualitative reports of the problems faced by women living in country areas. The availability, accessibility and affordability of services were rated lower by women living in rural and remote areas than by women living in urban areas. These substudy results complimented the administrative data from the HIC which revealed inequities in the out of pocket costs of services and in the availability of female GPs.

Many studies have adapted the behavioural model to examine the utilisation of a variety of types of health services, but the model requires more complex analysis than the traditional statistical techniques employed by most researchers. Insights were gained by using structural equation modelling, in addition to multiple linear regression and Poisson regression modelling, and by including out of pocket costs in the model.

The use of GP services was shown to be determined primarily by medical need. Also the out of pocket cost per visit tended to be lower for women with lower socioeconomic status. These findings suggest that the system is equitable in these dimensions. However the out of pocket cost for GP services increased with distance from urban centres, which shows that the charges imposed by GPs are not equitable across place of residence. Higher cost was associated with lower use of services, even among women considered to have high levels of need for care.

The findings from these cross-sectional data suggest a need to regulate the costs of GP services to patients, particularly in rural areas. The study also highlights the need to improve the delivery of GP services especially in rural and remote areas. The long-term impact of the geographical inequity in services on health outcomes for women remains to be determined. Policy implications are discussed and recommendations for future research and monitoring, using cohorts in the longitudinal study, are proposed.