Does breast cancer policy meet the needs of Aboriginal and Torres Strait Islander women in Australia? a review

Abstract

Objective

To evaluate if existing Australian public policy related to screening, diagnosis, treatment and follow up care for breast cancer addresses the needs of and outcomes for Indigenous¹ women?

Methods

This review of policy employed a modified Delphi method via an online panel of experts (n = 13), who were purposively recruited according to experience and expertise. A series of online meetings and online surveys were used for data collection. The aims of the study were to: Identify all existing and current breast cancer policy in Australia;  Analyse the extent to which consideration of Indigenous peoples is included in the development, design and implementation of the policy; and Identify policy gaps and make recommendations as to how they could be addressed. The policies were evaluated using ‘A Guide to Evaluation under the Indigenous Evaluation Strategy, 2020’.

Results

A list of current breast cancer policies (n = 7) was agreed and analysed. Five draft recommendations to improve breast cancer outcomes for Indigenous women were developed and refined by the panel.

Conclusions

Current breast cancer policy in Australia does not address the needs of Indigenous women and requires change to improve outcomes.

Introduction

Breast cancer in Indigenous women in Australia

Breast cancer is the most diagnosed cancer affecting Australian women, and the second largest cause of cancer death in Australian women. While the incidence rate of breast cancer is lower in Indigenous women that non-Indigenous women [1], the mortality rate is higher, with Indigenous women 1.2 times more likely to die from the disease [2]. In New South Wales, Indigenous women are 69% more likely to die from their breast cancer than non-Indigenous women [3]. There are a number of factors that may contribute to this higher mortality rate, including lower participation in screening services, socioeconomic disadvantage, younger age at diagnosis, geographic remoteness, co-morbidities and a more advanced stage of cancer at the time of diagnosis [1, 3, 4]. It is not thought to be due to differences in histological subtype or mammographic density [5, 6].

The role of health policy in Australia

The Australian Policy handbook defines policy as “A statement of government intent, and its implementation through the use of policy instruments.”[7]. In Australia, government policy at the macro level is presented as legislation. Funding typically flows directly from legislation. In order for funding to be allocated, there needs to be a program and associated legislation and policy. Policy is critical to funding, priorities and outcomes, and representation of the issues facing Indigenous women in policy is vital to those issues being addressed in service delivery models and approaches.

Research question

In order to refine our research question, we used PICO format:

• Patient/population: Aboriginal and Torres Strait Islander women.

• Intervention: review Australian breast cancer policy to assess whether it meets the need of the patient/population.

• Comparison: non-Indigenous women in Australia.

• Outcomes: Expert consensus on recommendations for improvement of breast cancer policy in Australia.

Research Question: How does existing Australian policy related to screening, diagnosis, treatment and follow up care for breast cancer address the needs of and outcomes for Indigenous women?

Hypothesis for the study: that existing/current policy for breast cancer in Australia does not adequately address the needs of Indigenous women. And that identifying policy gaps can inform future implementation and improve outcomes for Indigenous women In Australia.

Aims

• Identify all existing and current policy.

• Analyse the extent to which consideration of Indigenous peoples is included in the development and design of the policy.

• Identify policy gaps and make recommendations as to how they could be addressed.

The Delphi Method

The Delphi method- commonly used in reviews- relies on consensus among experts (often referred to as a panel) to answer a research question [8]. Despite variation in design, the Delphi method is based on a series of questions put to the expert panel, whereby the panel members are surveyed for opinions on a particular issue. The questions follow a sequential process, whereby each round/survey is partially based findings of the previous round/survey, allowing for evolution of the study [8]. Results of each round are transparent and deidentified, encouraging reflection on the views of others and possible readjustment of their own perspectives [9]. The deidentification of responses provide protection from bias or perceived negativity [8]. The Delphi method is a framework - where each round augments the previous round - is designed to result in a consensus view.

We have used the Delphi method to look at breast cancer policy; identifying the gaps in policy and synthesizing the evidence regarding the needs of Indigenous women to provide recommendations for policy improvement.

Methodology

Delphi Method

The study utilised the Delphi method via an online panel [10] (see Fig. 1). The panel were purposively recruited by inviting people with knowledge and experience in breast cancer policy and practice, specifically representatives of the breast cancer peak organisations, representatives of the Aboriginal Community Controlled Health sector, consumers and experts in public health policy. The Delphi method was undertaken in five stages over a three month period.

Fig. 1

Delphi process

Fig. 2

Guiding principles for the Indigenous Evaluation Strategy

Search strategy and selection criteria

A systematic hand search of the grey literature (utilising Google search engine) was conducted using the following search terms: “Breast cancer Australia policy”; “Australia breast cancer standard policy”; “Australia breast cancer funding policy”. Table 1 shows inclusion and exclusion criteria were used to select the policies for the study.

Table 1 Inclusion and exclusion criteria

Stage 1

In the first stage or our review, one author (VC) conducted an extensive search (as detailed above) of the grey literature to assemble a list of current policies for breast cancer in Australia. This search was conducted over multiple days during the month of September 2022, employing the stopping rule of ten pages searched (and no mention of policy on pages). The definition of ‘current’ is: an active policy, which has not been superseded. The policies were collated into a spreadsheet with: owner; source; scope; purpose; commencement date; and whether the policy was current or not, extracted.

Invitations to ‘experts’ (see definition above) were sent out via letters/email giving context to the study and a deadline for acceptance. The organisations approached were as follows:

Breast Cancer Network Foundation, Cancer Australia, Cancer Council, Foundation for Breast Cancer Care, Breast SurgANZ, Cancer Institute and an Aboriginal Community Controlled Health Organisation.

Stage 2

The second stage of the review comprised two online meetings with members of the expert panel. These meetings served as an introduction to the project and a review of the collated breast cancer policies, to ascertain if the list was complete. The definition of ‘policy’ (see Results) was refined over several meetings.

Stage 3

The third stage was a process of evaluation to examine how well the policies addressed the needs of Indigenous women. Two authors (VC, KG) of the paper evaluated each policy according to the Australian Government Productivity Commission’s: A Guide to Evaluation under the Indigenous Evaluation Strategy, 2020 [11], specifically Fig. 2 [11].

This evaluation was added to the spreadsheet of policies. Under the following broad categories of: Credible/Useful/Ethical/Transparent, each category was scored either: Yes/No/Partially. Definitions of the categories were as followed:

Credible: trustworthy; grounded on rigorous methodology, and includes Indigenous values, perspectives and knowledges.

Useful: considers the needs of a range of end-users (primarily that of Indigenous people) and of high quality, with findings are available to decision makers, and timed to contribute to policy-making and implementation decisions.

Ethical: improves the quality and consistency of evaluation and ensures that it has a positive impact on Indigenous people.

Transparent: open and honest in intention and information, specifically providing access for Indigenous people.

The evaluation spreadsheet was then disseminated amongst the expert group along with an online survey and each member was invited to provide feedback via the survey and/or an online meeting. The survey and follow up meeting provided further review and discussion about the policies and scoring with a view to reaching consensus.

Stage 4

Stage four consisted of the development of draft recommendations to improve breast cancer policy in Australia. On close examination of the complete list of breast cancer policies and comparing with the evidence of what will improve outcomes for Indigenous women with breast cancer [12], one author (VC) drafted five recommendations in response to the policies and their perceived shortcomings.

The draft recommendations were presented to the panel via an online meeting and/or an online survey for feedback. The feedback was then collated with further discussion and input from the panel to reach consensus where possible. Where consensus was not possible, it was reported.

Stage 5

In the fifth and final stage of the policy review, the first author sent the draft recommendations survey results and comments to the panel, with responses to each comment. Feedback from this process was then incorporated into the draft review of breast cancer policy in Australia and, following discussion of the panel, consensus was reached on newly worded recommendations.

Results

Stage 1

Initially, one author (VC) invited 21 people to be part of the expert group, with a range of expertise and experience (summarised in Table 2. Deidentified experience of panel members), and 14 accepted and participated. One member withdrew from the panel midway through the process due to time constraints, with a total of 13 panel members remaining (not including VC and KG).

Table 2 Deidentified experience of panel members

Stage 2

The agreed definition of policy: “A publicly available statement of intent or expected practice that is not negotiable and requires mandatory compliance (with consequences if not complied with).”

The agreed list of breast cancer policies is shown in Table 3, with evaluation columns added (from Stage Three). There was 100% agreement on the list of policies from the panel.

Table 3 Final list of breast cancer policies (Stage 1), evaluated (Stage 3)

Stage 3

Using the broad categories of the Australian Government Productivity Commission’s: A Guide to Evaluation under the Indigenous Evaluation Strategy, 2020, only one of the policies was deemed Transparent and only one of the other policies was deemed Partially Useful. None of the policies were deemed Ethical or Credible. The panel agreed 100% with the assessment of the policies.

Stage 4

The following five draft recommendations were sent to the panel as an online survey with a brief explanation/justification and there were three options for answers: Agree or Disagree and Comments. In Table 4, we have recorded the number of respondents who agreed or disagreed to each of the recommendations.

Table 4 First round of draft recommendations

Ten (of the 14) members of the panel responded to the survey, either agreeing, disagreeing or neither agreeing nor disagreeing (leaving a comment instead).

Stage 5

The first author collated the panel’s comments regarding the draft recommendations. The first author then responded to these comments and added a Result column to mark whether or not consensus was reached and wording of the recommendation remained the same, or consensus was not reached and extra work needed, as shown in Table 5.

Table 5 Survey results of draft recommendations, with responses

As a result of the first part of Stage Five there was one agreed recommendation and four partially agreed recommendations, with further discussion to reach consensus on the final list of recommendations.

Agreed recommendation

1. That Aboriginal and Torres Strait Islander women are consulted in the formation of policy.

Partially agreed recommendations with outcomes

• 1. Screening for breast cancer is free and accessible at an earlier age.

As mentioned in Table 5, there is scant evidence relating to the early detection of breast cancer for Aboriginal and Torres Strait IslanderI women [13]. There was mention of only one paper found in the grey literature reporting on proportion of recalled cases, leading to the assumption this has not been extensively examined [14]. There is evidence, however, that Indigenous women are more likely to be younger at diagnosis [15].

Through the Delphi process, it became clear that while screening is free from 40 years, active invitation does not begin until 50 years and that there is not enough evidence regarding age of incidence in Indigenous women to pinpoint a younger age than 40 years for free screening. It was also acknowledged that the sensitivity of the mammogram decreases the younger the patient, and that in the < 40 years age range there is not much written about screening benefits due to low incidence, comparatively denser breast tissue, and expensive screening tools [16]. The data on mammographic density in Indigenous women suggests that density is a risk factor for breast cancer just as in non-Indigenous women, but overall density tends to be lower. This means that screening with mammography is still a reasonable strategy to use [17]. Our recommendation for screening being free and accessible at a younger age therefore becomes: Aboriginal and Torres Strait Islander women are actively invited and encouraged to screen from the age of 40 years.

• 2. That women with symptoms are not discouraged from attending breast screening.

As shown in Table 5, there is a level of complexity around the reasons for discouraging women from screening if they have symptoms. The feedback from the panel on this recommendation was that offering solutions within the recommendation makes more sense than discouragement. However, the compounding effect of discouragement for groups that experience higher barriers to screening is still of concern. It is therefore important that policy clearly includes information regarding the alternatives to screening for women with symptoms.

Newly worded recommendation: That policy states the pathway women with symptoms of breast cancer should take (instead of only stating they are not appropriate candidates for screening).

• 3. It is not enough to acknowledge the barriers for Aboriginal and Torres Strait Islander women and not address them or provide appropriate solutions.

There were two policies [18, 19] amongst the complete set of policies which acknowledged and even highlighted the difficulty faced by Indigenous women in screening and treatment of breast cancer. Whilst this acknowledgement is a start, it begs the questions of what can be done about it. These barriers must be addressed in policy. This recommendation was reworded as: That policy must provide appropriate solutions for the barriers that Aboriginal and Torres Strait Islander women face.

• 4. Policy should not be based on assumption of adherence to guidelines by the consumer.

Current policy states: “To be effective on a population basis, a high compliance rate of attendance of women in the appropriate age range for screening mammography is necessary” relies on the “compliance” of attendant women. It is not enough for policy to state what works best, apportioning the burden of responsibility to the consumer. Policy should be written for all Australian women and explicitly address access barriers. This recommendation remains the same.

Discussion

The effectiveness of a policy depends on a number of factors, many of which are government controlled. These include the selection and application of policy instruments in the implementation and operationalisation of policies, the quality of screening equipment and the skill of the readers, decisions about technology, adherence of the women, and developing different approaches for women at very high risk (for example, genetic mutations increasing susceptibility). Our use of the term “high risk” has to carefully distinguish higher incidence (which is not the case with Indigenous women) from higher risk of advanced stage at diagnosis and hence higher mortality.

Policies are how governing bodies set expectations for practice. As previously mentioned, there is a wide body of evidence regarding the poor outcomes for Indigenous women with breast cancer. While there are various guidelines, standards or strategies to improve these outcomes, the evidence is not currently or consistently accounted for in policy. In short, we know what the issues are, we have potential solutions, but we are not implementing these solutions at the policy level.

Throughout the process of undertaking this review, it became clear that Indigenous Australians must be a part of the inception of, consultation for, and writing up of policy, in order for it to better meet the needs of Indigenous women. It is not sufficient to acknowledge poorer outcomes.

Review of the policy

It appears that policy written regarding breast cancer in Australia is predominantly written for low-risk women and focuses solely on screening and diagnosis (and does not include treatment and follow up care). This comes at a cost for higher risk women whose needs are not adequately met by the policy. Whilst we acknowledge that there are high risk pathways to MRI screening in major public hospitals, these are likely to be difficult to reach for Indigenous patients. There is much anecdotal mention of a lack of resources to tailor policy to the needs of higher risk women; however, this is no excuse for what continues to be poorer outcomes for these women. This review looks beyond whether the policy is a good policy and is concerned with whether the policy meets the needs of those who suffer disadvantage in the larger health context. It builds upon the theory that current policy in Australia does not meet the needs of Indigenous women and makes recommendations for how this could be improved.

Final list of recommendations

• 1. That Aboriginal and Torres Strait Islander women are consulted in the formation of policy.

• 2. Aboriginal and Torres Strait Islander women are actively invited and encouraged to screen from the age of 40 years.

• 3. That policy states the pathway women with symptoms of breast cancer should take (instead of only stating they are not appropriate candidates for screening).

• 4. That policy must provide appropriate solutions for the barriers that Aboriginal and Torres Strait Islander women face.

• 5. Policy should not be based on assumption of adherence to guidelines by the consumer.

Our study recognised that there are various guidelines/strategies/action plans produced in an effort to overcome the poorer outcomes for Indigenous people, for example the Optimal Care Pathway for Aboriginal and Torres Strait Islander people with cancer, and that many of these guidelines overlap with the recommendations arising from this review. The issue remains that even with these guidelines, the outcomes are not improving. So while the guidelines and our recommendations align in purpose, they need to be implemented at the policy level.

Conclusion

Since breast cancer is the most common cancer experienced by Indigenous women in Australia, is diagnosed at an earlier age, and is more likely to be experienced with co-morbidities, it is not surprising that the outcomes are poorer. The evidence indicates that the present approach to breast cancer in the Indigenous population is not making a positive difference. The current breast cancer policy in Australia is written for the majority of Australian women and, while the Indigenous population sits at 3.8% [20], the poorer outcomes for Indigenous women are acknowledged as opposed to addressed at the policy level.

The mortality risk of breast cancer in Indigenous women is higher compared with non-Indigenous women, therefore novel policy strategies are required. This study provides recommendations learned in a review of breast cancer policy in Australia. To address the burden of breast cancer among Indigenous women, this study advocates for the update of breast cancer policy in Australia and for policy makers to work alongside Indigenous communities to ensure that outcomes improve.