The risk status, screening history and health concerns of Aboriginal and Torres Strait Islander people attending an Aboriginal Community Controlled Health Service
The work was carried out as part of the corresponding authors' PhD through the University of Newcastle. Data collection was undertaken at Derbarl Yerrigan Aboriginal Health Service in Perth, Western Australia.
Jessica M. Stewart BA/LLB, MPS, Manager, Aboriginal Health Research and Evaluation, Rob W. Sanson-Fisher PhD, Laureate Professor, Sandra Eades PhD, Professor, Michael Fitzgerald MBiostat, Biostatistician. Ms Jessica M. Stewart, Research and Evaluation Unit, Centre for Aboriginal Health, NSW Ministry of Health, 73 Miller Street, Sydney, NSW 2060, Australia. Tel: +61 2 9391 9770; Fax: +61 2 9391 9840; E-mail: firstname.lastname@example.org
Introduction and Aims. Primary health-care services need to maximise prevention activities to improve the health of Aboriginal and Torres Strait Islander people. This study determined Aboriginal and Torres Strait Islander people's risk status and screening history for cardiovascular, diabetes and cancer, and identified opportunities for prevention based on patient's health concerns.
Design and Methods. Consenting adult patients attending an Aboriginal Community Controlled Health Service completed interview surveys. Outcome measures were: patient's self-reported rates of smoking, at-risk alcohol consumption, body mass index and screening rates for cholesterol, diabetes, blood pressure, cervical and breast cancer, the proportion of patients with multiple risk factors and health information patients prioritise to receive from their doctor.
Results. We surveyed 587 patients giving a consent rate of 77%. Patient's self-reported being at risk due to smoking (51%), harmful alcohol consumption in both short term (10%) and long term (10%), and overweight (28%) and obese (49%) body mass index. The proportion of patients who had not been screened within the recommended guidelines was 27% for cholesterol, 24% for diabetes, 40% for blood pressure, 47% for cervical cancer and 54% for breast cancer. The majority (73%) of patients had multiple risk factors. Patients prioritised receiving health information on diabetes (45%), weight (43%) and heart disease (43%).
Discussion and conclusions. Many patients were at high risk, and were concerned about the health risks they experience. Strategies are needed to help clinic staff identify risks, and maximise prevention activities according to best-practice guidelines, particularly to patients who experience multiple risk factors. [Stewart JM, Sanson-Fisher RW, Eades S, Fitzgerald M. The risk status, screening history and health concerns of Aboriginal and Torres Strait Islander people attending an Aboriginal Community Controlled Health Service. Drug Alcohol Rev 2012;31:617–624]
Aboriginal and Torres Strait Islander people, the Indigenous people and traditional owners of Australia, experience higher rates of mortality and morbidity from cardiovascular disease, diabetes and selected cancers compared with non-Indigenous people . These diseases share common preventable risk factors, including smoking, being overweight or obese, risky alcohol consumption, and high blood pressure and cholesterol. In order to prevent these diseases, health service providers need accurate information on whether their patients are at risk, and whether they have been screened within the recommended time period. Yet prevalence rates for risk factors among Aboriginal and Torres Strait Islander people are generally reported through national household surveys rather than from health-care settings . Studies with a smaller number of participants have been undertaken on the risk status of Aboriginal and Torres Strait Islander people in clinical settings other than Aboriginal Community Controlled Health Services (ACCHSs) [3–5], and other studies have used medical records to estimate risk  rather than patient self-report. Few studies have reported risk prevalence data for Aboriginal and Torres Strait Islander people accessing ACCHSs, or on screening rates for preventable disease.
Aboriginal Community Controlled Health Services are independent from government and represent self-determination in the provision of comprehensive primary health care to Aboriginal and Torres Strait Islander people [7–9]. ACCHSs are governed by local community members, and the majority of staff are Aboriginal and Torres Strait Islander people, who provide culturally competent health care and a holistic approach to health and well-being. In 2008–2009, there were 211 ACCHSs funded federally, providing 2.1 million episodes of care to approximately half of the Aboriginal and Torres Strait Islander population in Australia [10,11]. Considering the potential effectiveness of general practitioner-based interventions in the broader Australian community [6–16], ACCHSs are likely to be effective in reducing risk factors and providing appropriate screening to Aboriginal and Torres Strait Islander people in a culturally competent environment. Currently, there is limited reporting on Aboriginal and Torres Strait Islander people who access primary health care, which may contribute to there being modest evidence for the positive effects of primary health-care interventions for Aboriginal and Torres Strait Islander people .
National survey data also indicate that Aboriginal and Torres Strait Islander people often have multiple, complex chronic diseases, which can present challenges for doctors in determining what to prioritise, and in deciding what health advice patients will be most receptive to. This study aims to determine Aboriginal and Torres Strait Islander people's: (1) risk status for smoking, at-risk alcohol consumption, unhealthy body mass index and physical inactivity; (2) rates of under-screening compared with recommended guidelines for diabetes, cholesterol and cervical cancer; and (3) to identify opportunities for prevention based on the health advice patients prioritise to receive from their doctor.
Study design and setting
The cross-sectional study was undertaken in an urban ACCHS, Derbarl Yerrigan Health Service in Perth, Western Australia. The participating ACCHS operates three clinics located more than 20 km apart from each other, servicing the local Aboriginal communities in different urban regions. The study was conducted between November 2007 and July 2009. In May 2009, a total of 13 116 patients were identified as current patients using Communicare records (6043 men and 7078 women). Board of Management approval from the ACCHS was obtained. Ethics approval was obtained from The University of Sydney, The University of Newcastle and the Western Australian Aboriginal Health Information and Ethics Committee. An Aboriginal and Torres Strait Islander community reference group provided important feedback and advice on the cultural and ethical issues regarding the content of the survey and data collection, and the survey was piloted among Derbarl Yerrigan Health Service staff and subsequently refined.
Patients were recruited in the waiting rooms of the three clinics by trained Aboriginal Research Project Officers and informed consent obtained. Patients who were Aboriginal and Torres Strait Islander people aged 18 years and over were randomly selected from the daily printed appointment list to reduce selection bias. Before recruitment commenced on each day, either every patient was approached if there were two interviewers or every second patient on the list if only one interviewer was available. The gender and age of non-consenting patients was recorded. Patient demographic information was compared with national Australian Bureau of Statistics (ABS) census data to determine the representativeness of the patient population of broader Aboriginal and Torres Strait Islander populations in Australia.
Questionnaire and demographic information
Participants completed a self-administered pen and paper questionnaire before the scheduled consultation. Participants with difficulties reading or writing were assisted to complete the survey by a research project officer who read out the questions.
Demographic information was collected on participants' gender, age, highest level of education completed, main source of income and relationship status.
The survey addressed three health domains: (1) prevalence of self-reported behavioural health risk factors, including smoking, unhealthy body mass index, physical inactivity and at-risk alcohol consumption; (2) rates of under-screening compared with recommended guidelines for diabetes, cholesterol and cervical cancer; and (3) health advice patients prioritise to receive from their doctor.
Definitions of risk categories for each risk factor and optimal screening intervals were derived from current national general practice guidelines for clinical preventive services [13–16]. The risk factors and diseases covered in the survey items were selected due to the high burden of disease associated with the health issue, and the potential to effectively reduce the risk factor or undertake preventative screening within a primary health-care setting based on evidence-based guidelines for preventive care. The wording of survey items followed that of the most recent ABS national population survey items [17–19].
Definitions of behavioural risk factors
Smoking status. Participants were defined as daily smokers if they reported smoking tobacco at least once per day, or as an occasional smoker if they self-reported being a smoker, but not daily. Patients were considered an ex-smoker if they had not had a cigarette in the last 12 months, or as never smoked if they had never been a regular smoker .
Alcohol. Alcohol consumption was measured using a 7 day diary recall, a measure which has been shown to provide a more accurate estimate than other self-report measures . The National Health and Medical Research Council (NHMRC) Standard Drinks Chart  was shown to participants to assist recall. Recorded consumption was then converted to standard drinks using the Standard Drinks Chart. A patient was considered at risk if their alcohol consumption levels met NHMRC guidelines  for at-risk alcohol consumption which equated to no more than four standard drinks on any one occasion for risk of alcohol-related injury, and no more than two standard drinks on any day for lifetime risk of harm from alcohol-related harm. These NHMRC guidelines were used by the ABS in conducting the National Aboriginal and Torres Strait Islander Health Survey (NATSIHS) .
Body mass index. Participants were considered overweight if their measured body mass index was 25–30 kg/m2 (overweight) or over 30 kg/m2 (obese) [15,23,24]. Participants' height and weight were measured by project staff after completing the survey using scales in the clinic.
Definitions of screening
Survey questions asked ‘when was the last time your doctor tested you for . . . ?’. Response options reflected the current evidence-based recommendations for screening intervals of the Royal Australasian College of General Practitioners, and were answered in a ‘yes’, ‘no’ or ‘not sure’ format.
Blood pressure. Participants were considered at risk if they had not had their blood pressure checked in the previous 12 months, or 6 months for patients who self-reported as having diabetes .
Cholesterol. Patients were considered at risk if they had not been screened within the previous 2 years as part of an overall cardiovascular risk status assessment .
Diabetes (type 2). Patients were considered ‘at risk’ if they were 35 years or over and had not had fasting blood sugar screening within the previous 12 months. Patients who self-reported having type 2 diabetes were excluded from the analysis of patients ‘at risk’ of type 2 diabetes .
Cervical cancer. Women were considered at risk if they were between 18 and 69 years and had not had a pap smear in the previous 2 years, and had not had a hysterectomy .
Breast cancer. Women were considered at risk if they were between the ages of 50 and 69 years and had not had a mammogram within the last 2 years .
Health information patients prioritised to receive from their doctor
Participants were asked to prioritise three health issues they would most like their doctor to talk to them about with the list, including the following: smoking, alcohol, diabetes, blood pressure, heart disease, weight, cholesterol, cervical cancer, breast cancer or other with a space to specify. Answers were recorded by placing the number 1, 2 or 3 next to the selected items. Responses were weighted with a score of 3 allocated to a patient's first priority, 2 to their second priority and 1 to their third priority. The scores were then aggregated to give an overall prioritisation of health issues by participants.
Univariate comparisons between those who nominated diabetes, heart disease and weight as a top three priority were made using a series of χ2-tests to determine characteristics of patients who prioritised each of these issues. To account for multiple testing, a critical P-value of 0.001 was used. All analyses were performed using sas software version 9.2 .
Of the 758 people invited to join the study, 587 were recruited and 171 people declined to participate, giving a consent rate of 77%. The sample consisted of 385 (66%) women and 202 (34%) men. There were no significant differences in age (χ2 = 8.07, P = 0.089) between consenters and non-consenters; however, the non-consenting group contained a greater proportion of men than the consenting group (χ2 = 6.42, P = 0.011). Demographic information of study participants was also compared with 2006 census data for Aboriginal and Torres Strait Islander people living in the same geographical location, Perth  for gender, Indigenous status, age, education, source of income and relationship status. As shown in Table 1, the sample differed from the census data in all of these factors, suggesting that patients visiting the clinic are not broadly representative of the general Aboriginal and Torres Strait Islander population in the area. In comparison with the national consensus data, the study population had a higher proportion of women, were generally older and less likely to have completed higher schooling levels, but were more likely to have a vocational or other qualification, which may have included the completion of high school equivalent. The higher proportion of women attending health clinics is consistent with results of other general practice studies [27,28].
Table 1. Comparison of demographic characteristics of study participants and participants in the Australian Bureau of Statistics National Census 2006
|Gender (n = 587)|| || || |
| Male||34 (202)||49.8||<0.0001|
| Female||66 (385)||50.2|| |
|Indigenous status (n = 586)|| || || |
| Aboriginal||98 (572)||90||<0.0001|
| Torres Strait Islander||0.9 (5)||6|| |
| Both Aboriginal and Torres Strait Islander||1.5 (9)||4|| |
|Age (years) (n = 587)|| || || |
| 18–19||3.1 (18)||6.7||<0.0001|
| 20–24||8.4 (49)||14.9|| |
| 25–44||44 (256)||47.6|| |
| 45–64||38 (224)||24.8|| |
| 65+||6.8 (40)||6|| |
|Education level (18 years+) (n = 546)|| || || |
| Year 8 or below||18 (99)||23||0.0112|
| Year 9 or equivalent||17 (89)||13|| |
| Year 10 or equivalent||36 (196)||33|| |
| Year 11 equivalent||15 (80)||14|| |
| Year 12 or equivalent||15 (82)||17|| |
|Vocational/other qualification (n = 556)||48 (281)||12||<0.0001|
|Main source of income (n = 568)|| || || |
| Full-time employment||24 (134)||24||<0.0001|
| Part-time employment||8.6 (49)||15|| |
| Centrelink/unemployed/not in labour force||61 (349)||53|| |
| Other||6.3 (36)||8|| |
|Relationship status (n = 577)||18 years+||Includes 15 years+|| |
| Married||19 (107)||24||<0.0001|
| Single/not married||57 (330)||60|| |
| De facto||24 (140)||16|| |
Prevalence of risk factors component
The study found a high prevalence of health risks, with half of all patients being current smokers (51%), 10% of people being at risk due to both short-term or long-term alcohol consumption, and 72% of men and 80% of women being overweight or obese (Table 2).
Table 2. Proportion of study participants (>17 years) who self-reported being at risk status for smoking, alcohol consumption and body mass index
|Current smoker||99 (51%)||196 (51%)||295 (51%)||0.9275|
|Mean (SD) days drinking alcohol||1.2 (1.4)||1.2 (1.5)||1.2 (1.4)||0.8422|
|More than two alcoholic drinks||23 (55%)||41 (56%)||64 (56%)||0.8841|
|More than four alcoholic drinks||21 (50%)||33 (45%)||54 (47%)||0.6199|
|Body mass index|| || || || |
| Underweight (<18.5)||5 (2.8%)||4 (1.2%)||9 (1.8%)||0.0466|
| Overweight (25–30)||55 (31%)||89 (27%)||144 (28%)|| |
| Obese (>30)||74 (41%)||173 (53%)||247 (49%)|| |
A high proportion of patients attending the clinics self-reported not having being screened within recommended national preventive guidelines for Aboriginal and Torres Strait Islander health (Table 3). Of all women surveyed, 47% had not been screened for cervical cancer at recommended intervals, nor had 54% of women for breast cancer. Both men and women were potentially at risk due to under-screening as 40% of patients had not been screened for high blood pressure within recommended guidelines, 27% of patients had not been screened for cholesterol and 24% of patients had not been screened for diabetes within recommended guidelines.
Table 3. Proportion of study participants (>17 years) who self-reported not being screened for diabetes, blood pressure, cholesterol, cervical and breast cancer according to best-practice guidelines
|Diabetes risk||19 (20)||44 (27)||63 (24)||0.1777|
|Blood pressure risk||87 (44)||144 (38)||231 (40)||0.1594|
|Cholesterol risk||55 (28)||103 (27)||158 (27)||0.7107|
|Cervical cancer risk|| ||152 (47)||152 (47)|| |
|Breast cancer risk|| ||51 (54)||51 (54)|| |
Prevalence of multiple risk factors
Analysis of the frequency distribution of patients with multiple risk factors was calculated based on at-risk behaviour and under-screening (Table 4). Less than 2% of all patients were risk-free, with 73% of patients having multiple risk factors. Of these, the greatest proportion of patients had two of the risk factors examined in this study (31%), with 42.7% of patients having three or more risk factors.
Table 4. Proportion of study participants (>17 years) with self-reported multiple health risk factors
|0||5 (2.5)||6 (1.6)||11 (1.9)|
|1||62 (31)||82 (21)||144 (25)|
|2||73 (36)||108 (28)||181 (31)|
|3||46 (23)||106 (28)||152 (26)|
|4||12 (5.9)||48 (12)||60 (10)|
|5||3 (1.5)||29 (7.5)||32 (5.5)|
|6||1 (0.5)||6 (1.6)||7 (1.2)|
Health information patients prioritised to receive from their doctor
Both male and female patients prioritised the same five key health issues (Table 5). Diabetes was the most important health issue patients wanted to receive information about from their doctor, particularly for patients who were already diabetic or had high blood glucose (P < 0.0001). This suggests there is opportunity to reduce the onset of diabetes among patients in this setting. Patients who had not been screened within recommended guidelines for high blood pressure and cholesterol were also keen to receive information on diabetes (P < 0.0001). Patients who were already taking blood pressure medication prioritised receiving information on heart disease (P < 0.0001), as well as diabetes (P < 0.0001) and high blood pressure (P < 0.0001). Female patients who were overweight or obese were keen to receive advice from their doctor on how to lose weight (P < 0.0001) and also prioritised receiving information on high cholesterol (P < 0.0001), smoking cessation (P < 0.0001) and diabetes (P < 0.0003).
Table 5. Health issues study participants prioritised to receive from their doctor
|Diabetes||1 (140)||1 (258)|
|Weight||5 (75)||2 (248)|
|Heart disease||2 (106)||3 (181)|
|Smoking||4 (80)||4 (134)|
|Blood pressure||3 (94)||5 (130)|
|Cervical cancer||—||6 (91)|
|Breast cancer||—||7 (67)|
|Cholesterol||8 (35)||8 (55)|
|Alcohol||7 (42)||9 (54)|
|Prostate cancer||6 (52)||—|
|Other||9 (6)||10 (23)|
This study shows that Aboriginal and Torres Strait Islander people attending ACCHSs are at high risk for developing cardiovascular disease, diabetes and cancer. It also found that most patients had multiple risk factors, which presents challenges for doctors in knowing what health advice and screening to prioritise to ensure they are providing the most effective preventive care. Patients in the study had a high level of awareness about their risk status and health conditions and wanted to receive advice and information from their doctor on health risks and leading chronic diseases.
The study relied on patients to report on their health status and previous screening history. While there are few studies validating self-report of risk factors among Aboriginal and Torres Strait Islander populations, studies suggest that self-reported smoking status is accurate among pregnant Aboriginal women and rural Aboriginal communities [29,30]. Self-reported alcohol consumption is known to be less accurate. One study has validated the use of self-report among urban Aboriginal communities , with increased accuracy using a 7 day retrospective diary [20,32,33], which was also used in this study. This study was also undertaken in one ACCHS in Australia and therefore the findings may not be able to be generalised to other ACCHS client populations. However, a large sample of participants was recruited through three geographically separate clinics. This study also allowed for more time to be spent with patients and therefore could provide more accurate information compared with provider-based survey data [4,27,28], and contributes to the limited published literature on the health of Aboriginal and Torres Strait Islander people living in urban areas .
A strength of this study is that it presents information specifically for Aboriginal and Torres Strait Islander people attending ACCHS clinics, and is one of the first to quantify rates of under-screening for common preventable risk factors and diseases. The prevalence of risk factors that are reported in national household surveys were similar to the findings in this study, including smoking rates of 51% versus 50% in the 2004–2005 NATSIHS, and rates of risky/high-risk drinking of 14% in major cities  compared with 10% in this study; however, the proportion of adult Aboriginal and Torres Strait Islander people living in non-remote areas who were overweight or obese was 60% compared with 77% in this study.
We focused on preventable risk factors for cardiovascular disease, diabetes and cancer as they contribute most to the health disparity between Aboriginal and Torres Strait Islander people and non-Indigenous Australians and therefore present the greatest potential for reducing the health disparity. Risk factors and screening tests were also chosen where strong evidence exists of effective interventions that can be delivered in a primary health-care setting . Evidence exists on the effectiveness of general practitioner-based interventions in non-Indigenous populations that modify patient behaviour to reduce smoking rates , at-risk alcohol consumption  and to encourage weight loss through diet and exercise [38,39]. Evidence also supports the effectiveness of regular screening and intervention in the prevention of cardiovascular disease , diabetes [14,41–43] and some cancers [44,45]. Similar to studies in mainstream primary health care in Australia [27,28,46], opportunities for preventive screening in ACCHS clinics are being missed. In order to be effective change agents for reducing high rates of preventable health risk factors among Aboriginal and Torres Strait Islander people, health providers working in ACCHSs should benefit from having reliable information about the likely risk status of their patients and an understanding of their health concerns.
Although the life expectancy of Aboriginal and Torres Strait Islander people has not improved significantly over the past two decades, changes in disease mortality patterns, such as significant improvements in infectious diseases, are in part due to improvements in the delivery of primary health-care to Aboriginal and Torres Strait Islander people. Similar efforts now need to focus on preventing high rates of chronic disease and associated risk factors. Opportunities appear to be missed, possibly due to competing health issues during consultations, insufficient knowledge of patient risk status and other system barriers common to primary care generally [27,28]. Increased focus on providing behavioural counselling on smoking, diet and exercise, and advice on prevalent health risks and disease, in particular diabetes and heart disease, may result in significant health gains.
Feedback received from doctors at the clinic on how to improve the identification of patient risk status at a local clinic level recommended that electronic patient information systems be enhanced with better automatic prompts for preventive health checks and guided intervention. Greater emphasis at a regional and national level for improving the routine reporting of prevalence data should also be considered to inform health policy and practice.
The authors would like to gratefully thank and acknowledge the Board of Management, staff and patients of Derbarl Yerrigan Health Service, the study research officers, Joan Fraser and Roz Yarran, and the community reference group, all of whom provided important input and support to the project.