The Link Between Primary Health Care and Health Outcomes for Aboriginal and Torres Strait Islander Australians

Methodology

This review explores the evidence both domestically and internationally as to whether access to high quality primary health care is essential to enhancing Indigenous health status.

Page last updated: June 2008

There is a risk in carrying out literature reviews of this kind that they become reduced to a list of examples, often from the ‘grey’ literature, of those places where, for example, primary health care projects are held to have worked at a local level.

Evidence from this local level – and in particular, from Aboriginal communities and people themselves – is important as a guide to practice. However, policy-makers are often expected to back up such local-level evidence, much of which is not highly substantiated in the more narrow scientific sense, with verifiable information that can stand up in sometimes robust public policy debates.

Although we present a number of case-studies and draw some key lessons from their practice, this report has been prepared as a strategic review for policy makers of what matters in the literature, rather than as a systematic or meta-analysis of the literature. It includes material from the world of professional practice that would not necessarily be included in a systematic review and does not include all of the vast literatures in the field of Indigenous primary health care.

The literature review

We began by carrying out a systematic review of the published literature in peer-reviewed journals about the link between primary health care and health outcomes, concentrating on studies of mortality and / or life expectancy outcomes.

Our reason for limiting our review to life expectancy / mortality measures was twofold. First, with the agreement of the Office for Aboriginal and Torres Strait Islander Health, it was decided that this was necessary to reduce the amount of data to a manageable amount for this review. Second, life expectancy and mortality have the advantage of being the ‘hardest’ of epidemiological measures routinely collected in many countries, allowing greater comparability between different places and over time.

Nevertheless, we note the well-known limitations of using life-expectancy and mortality to measure health. In particular, we need to remember that the aim of health care is not just about preventing mortality but also about preventing and treating sickness and about improving well-being.

With this in mind, the PubMed database was searched using a series of search terms based on MeSH items listed as Major Topics. The search terms for mortality outcomes (Mortality, Mortality/trends, Life Expectancy, and Life Expectancy/trends) were combined with those that might investigate the association between primary health care and mortality (Primary Health Care, Delivery of Health Care/statistics and numerical data, Health personnel/statistics and numerical data, Community Health Services/statistics and numerical data, and Community Health Centres). Top of page

The search revealed a rich body of international evidence in the form of several hundred articles. In order to concentrate on articles of most relevance, the titles and (if available) abstracts were read to identify:
  • studies of mortality trends to assess whether improved levels of primary health care provision over time are associated with increased health outcomes;
  • cross-sectional comparisons of mortality which assess whether different levels of primary health care in different places are associated with differential health outcomes;
  • articles that controlled for known confounders (e.g. income);
  • studies from developed countries (especially Australia, New Zealand, North America and Europe);
  • larger scale studies that compared mortality outcomes at either the national or jurisdictional level (e.g. USA counties or states).
These publications then formed the basis for the data for answering questions about the evidence for the effect of health systems and primary health care on the health of populations.

Once the evidentiary case for the efficacy of primary health care was examined at the broadest level, we turned the focus specifically towards the Indigenous experience, both overseas and in Australia. Two sources of data were brought together and analysed to provide this.

Literature on Indigenous specific mortality / life expectancy

The literature search described above, in our initial work plan and in discussions with the Office of Aboriginal and Torres Strait Islander Health, revealed very few studies directly focusing on Indigenous peoples.

A further examination of the literature was therefore then made specifically on the mortality / life expectancy patterns of Indigenous people in general and Aboriginal and Torres Strait Islander people in particular. These articles formed the basis of answering – on a national or at least jurisdictional level – the key questions about the shifting status of mortality / life expectancy in these populations.

Note that the modified ‘Delphi Method’ described below also yielded a number of research articles relevant to this section.

Capturing expert opinion, community experience and the ‘grey literature’

Simultaneously, we embarked on a process to capture the ‘grey’ literature and seek the views of key stakeholders in the Aboriginal primary health care field, using a modified version of the ‘Delphi Method’.

The Delphi Method is based on a structured process for collecting and distilling knowledge by means of a series of questions to leaders and practitioners in a field. The technique can be summarised in six steps:
1. Select a diverse panel of experts to participate in the exercise.
2. Develop a questionnaire – in our case the questionnaire was organised around the following key queries:
  • Are there published studies and/or write ups in the grey literature that illustrate the effect in Australia of primary health care on the health of Aboriginal and Torres Strait Islander communities?
  • Do you have data / program / descriptive / anecdotal work from your own services that demonstrate an effect?
  • Who else do you suggest we speak to in order to understand the effects of primary health care on the health of Aboriginal and Torres Strait Islander communities?
3. Distribute the questionnaire to the panellists.
4. Follow up the information given to the panellists with an interview.
5. Distribute the questionnaire any further informants nominated by the original panel.
6. Follow up the responses and conclude the panel round.

See Appendices for details of the panellists and the questionnaire.

We also contacted the national and State / Territory representatives of the Aboriginal community controlled health services (National Aboriginal Community Controlled Health Organisation – NACCHO – and affiliates) to seek their views and any data they might have.Top of page