Tim Baker, MB BS, Emergency Physician; Scott McCoombe, PhD, Lecturer; Cate Mercer-Grant, BBus, Program manager; Susan Brumby, RN, MHM, Clinical Associate Professor.
Chest pain in rural communities; balancing decisions and distance
Article first published online: 4 APR 2011
© 2011 The Authors. EMA © 2011 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Emergency Medicine Australasia
Volume 23, Issue 3, pages 337–345, June 2011
How to Cite
Baker, T., McCoombe, S., Mercer-Grant, C. and Brumby, S. (2011), Chest pain in rural communities; balancing decisions and distance. Emergency Medicine Australasia, 23: 337–345. doi: 10.1111/j.1742-6723.2011.01412.x
- Issue published online: 14 JUN 2011
- Article first published online: 4 APR 2011
- Accepted 11 January 2011
- agricultural worker;
- chest pain;
- emergency medicine;
- health literacy;
- rural health service
Objective: This pilot study examines the prevalence of cardiac risk factors in a cohort of agricultural workers, assesses their knowledge of local emergency health services and investigates their decision-making abilities with regard to when and how they would seek help when experiencing chest pain.
Methods: Farm men and women were recruited from 20 rural Victorian sites and underwent health assessments for total cholesterol, blood glucose, weight, height and blood pressure. Participants completed a survey to determine their knowledge of chest pain treatment, local emergency services and likely response to chest pain.
Results: Cardiac risk factors within this cohort of 186 adult farming men and women were common, with 61% of men (58/95, 95% confidence interval [CI] 51–70) and 74% of women (68/91, 95% CI 65–83) either overweight or obese. When asked to name their nearest ED, 10% of participants (19/184, 95% CI 7–16) nominated health services or towns where no ED exists. Furthermore, 67% of respondents (123/185, 95% CI 59–73) believed it was safe to travel to hospital by car while potentially having a myocardial infarction.
Conclusions: This cohort of agricultural workers were at considerable risk of experiencing acute coronary events, but many would make decisions about when and how to seek medical help for chest pain that are at odds with published community guidelines. These results highlight the need for education to improve knowledge of local emergency services and address behavioural barriers to accessing care.
Each year in regional Australia approximately 9000 people die of coronary artery disease, with acute myocardial infarction accounting for approximately half of these deaths. Nine hundred of these lives would be saved if mortality rates in regional areas were the same as those in metropolitan areas.1 Heightened regional mortality is partly explained by a higher incidence of cardiac risk factors.2 Delays in initiating treatment for acute cardiac events in rural areas might also contribute.
Treatment in the first 2 h following a myocardial infarction can decrease mortality by half.3 Also, one in four people who experience a myocardial infarction die from cardiac arrest within 1 h of their first chest pain.4 If the person has already reached hospital when they have a cardiac arrest, treatment success rates of between 20% and 40% are described.5 If the arrest occurs outside of hospital care, the survival rate is much lower. In rural Victoria, Australia, where this survey was conducted, survival rates for witnessed cardiac arrests occurring outside of medical care are less than 2%.6
Studies around the world have found many reasons why people with chest pain procrastinate before seeking treatment.7 Two common explanations are the failure to recognize a heart attack when the symptoms are atypical, and delay due to fear and embarrassment, even when symptoms indicate a heart attack is almost certain.8 Living in a rural or remote location has been described as an additional cause for delay.8 Even allowing for travel time, rural men and women with chest pain have been found to wait 30 min longer than their urban counterparts before seeking treatment.9 This delayed decision-making behaviour is not restricted to chest pain, with rural men and women presenting later for a wide range of emergency conditions.10
Farmers and non-town dwellers are often thought to be the slowest rural Australians to seek emergency medical treatment. Reasons given include the poorly defined and stereotypical concept of stoicism and an apparent fatalistic acceptance of supposed outcomes.11 We postulate that a lack of health literacy might also play a role in delaying medical help seeking behaviours in emergencies. Health literacy has been defined as the knowledge and skills to understand and use health resources, including in medical emergencies.12 In rural areas, the understanding of emergency medical service roles and capabilities appears to be adversely affected by a period of change where local district hospitals, emergency or after hours services have been closed with people advised to ring centralized after hours triage or attend alternative larger services.
Over this period, rural and remote Australians might not have received appropriate information explaining how to access and use these new emergency services effectively. It seems conceivable that there is a disconnect between modern emergency services and people more familiar with the old system, which was in place for more than a generation, of a general practice clinic and a small local hospital that dealt with every emergency situation occurring within its district.
The objectives of this pilot study were:
- 1To determine the prevalence of cardiac risk factors in a population of agricultural workers.
- 2To assess their knowledge of local emergency health services, and
- 3To examine decision-making processes among agricultural workers about when and how they would seek help for chest pain.
This cross-sectional study, of exploratory design, was conducted in Victoria, Australia between July 2009 and August 2010, in conjunction with the Sustainable Farm Families (SFF) programme (developed by the National Centre for Farmer Health, Hamilton, Victoria, Australia). SFF is an evidence-based service delivery programme that focuses on the health, well-being and safety of farm men, women and their families Australia wide.13 Participants are primarily recruited to SFF programmes through an industry group, although programmes are open to all farming men and women who meet eligibility criteria. The industry group can be any group within a local community that involves farmers; these groups have included sporting groups, progress associations and livestock groups among many others. Farmers are self-selecting to participate in the programme, although extensive media campaigns are undertaken across rural areas to encourage a broad range of the local agricultural community to participate. The SFF programme is well-documented elsewhere14,15 with participants undertaking health education workshops and having individual physical assessments over a period of 2–3 years.
To be eligible for this chest pain pilot study, participants were required to be aged between 18 and 80 years, speak and understand English. Each person undertaking the present study self-identified as being involved in active farming over the last 5 years as required by SFF selection criteria. As the present study reflected the demographics of farm men and women, both husband and wife were classified as farmers. If other family members such as parents and adult children met the above criteria and were actively employed within the farming enterprise they too were eligible to participate. Farmers were recruited from 20 rural locales across Victoria to undertake the present study. Using Victorian Government Department of Human Services boundaries, participants were recruited from four SFF programmes conducted in the Hume region, six within Loddon Mallee, four within Grampians, four within Barwon South Western and two within Gippsland. A plain language statement approved by South-West multidisciplinary ethics committee and signed consent was obtained from each participant.
Each consenting agricultural worker completed a short chest pain questionnaire on the first morning of the 2 day SFF workshop. The questionnaire asked how long it would take to drive, at legal speed, to local health-care facilities, and the name of the nearest ED. A scenario question required participants to estimate how long they would delay seeking medical advice with chest pain depending on whether they were in town or on farm. Participants responded to a further 12 statements about chest pain behaviour using a 5-point Likert scale. Questions were constructed so they could be later grouped into pairs to determine response consistency. Questions were written to either strongly agree or disagree with published community chest pain guidelines, such as from the National Heart Foundation of Australia's Every Minute Counts campaign.16
In addition to the questionnaire, fasting total cholesterol and blood glucose levels were tested using finger prick capillary samples following a 10 h fast. Weight was measured using calibrated scales and height determined. From this, the body mass index could be calculated. Blood pressure was measured using paired readings from an automated blood pressure machine. People were considered to be at risk of acute coronary events if they had a systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg.17 Other risk factors include having a body mass index ≥25, fasting blood glucose level ≥5.5 mmol/L18 or a total fasting cholesterol measurement ≥5.5 mmol/L.18
Data were collated using Microsoft Excel 2007 and analysed using the statistical software package PASWStatistics 18.0 (SPSS Inc. (Statistical Package for the Social Sciences Inc.) Chicago, Illinois, United States of America). Non-numerical or blank answers to questions were not included in the dataset. Questionnaires with less than 90% of questions answered were to be removed from this analysis, but no participant had less than a 90% response.
As participants in this chest pain pilot study were recruited through their attendance at SFF programmes, there was the possibility of selection bias. Participant self-selection could influence the data presented, as healthy and motivated individuals might be more knowledgeable about, or respond more rapidly to, chest pain. However, this is based on assumptions that people who have poor health do not want to receive free health assessment and advice. SFF programmes are free to attend and the availability of medical bulk billing is lower in rural communities. It is possible that people who were unwell might have been motivated to attend and have their health checked. It is not within the scope of this pilot study to determine whether the selection criteria resulted in a positive or negative skew, but rather to provide an important insight into the cardiac health and emergency knowledge of farmers in Victoria.
With an unknown estimated prevalence (assumed to be 50% for study size calculations), the chosen sample size has sufficient statistical power to measure parameters with a 95% confidence interval (CI) range of 7.5%. CIs were calculated using the Adjusted Wald method. Spearman's rho was used to determine the correlation between time to seek medical advice and distance from the nearest ED.
One hundred and eighty-six farmers (91 female and 95 male) from 20 rural locales in Victoria were enrolled in the study. All completed the survey with questions achieving between 98% and 100% usable response rate. The one exception to this response rate was in the scenario asking how long a participant would wait at home with chest pain. Fifteen per cent of respondents gave a non-numerical answer, such as ‘minutes’, ‘hours’ or ‘days’, or left the question blank.
Cardiac risk factors
The average age was 53.4 (SD 13.0) years for men and 51.8 (SD 11.2) years for women. Cardiac risk factors are listed in Table 1. In this cohort, 61% of men (58/95, 95% CI 51–70) and 74% of women (68/91, 95% CI 65–83) were either overweight or obese. Over half of the participants were hypertensive (52%, 95/186, 95% CI 44–58) and a similar number (53%, 99/186, 95% CI 46–60) had fasting blood glucose levels that required referral for further testing. Twenty-six per cent of participants (49/186, 95% CI 21–33) had raised cholesterol during testing or had previously been identified as having raised cholesterol. Ten per cent (18/186, 95% CI 6–15) returned positive results for hypercholesterolaemia during the SFF programme, with a further 17% of participants (31/186, 95% CI 12–23) using oral hypolipidaemic medication at the time of testing to maintain normal cholesterol levels. Three per cent of participants were current smokers (3/186, 95% CI 0–5) and 26% (48/186, 95% CI 20–33) indicated they had previously smoked.
|Female % (n, 95% CI)||Male % (n, 95% CI)|
|Body mass index ≥25||74.7 (68/91, 64.9–82.6)||61.1 (58/95, 51.0–70.3)|
|Fasting blood glucose ≥5.5 mmol/L||59.3 (54/91, 49.1–68.9)||47.4 (45/95, 37.6–57.3)|
|Total cholesterol ≥5.5 mmol/L||5.5 (5/91, 2.1–12.5)||13.7 (13/95, 8.0–22.2)|
|Already taking lipid lowering agents||12.1 (11/91, 6.7–20.5)||21.1 (20/95, 14.0–30.4)|
|Hypertension (BP ≥140/90 mmHg)||49.5 (45/91, 39.4–59.5)||52.6 (50/95, 42.7–62.4)|
|Current smoker||2.2 (2/91, 0.1–8.1)||3.2 (3/95, 0.7–9.3)|
|Previous smoker||29.7 (27/91, 21.2–39.8)||22.1 (21/95, 14.9–31.5)|
Understanding EDs and local health services
When asked to nominate the nearest ED (Table 2), 38% (71/184, 95% CI 32–46) named hospitals with a doctor in attendance 24 h each day. A further 50% (94/184, 95% CI 44–58) nominated EDs where doctors had to be called in from home, whereas 10% of participants (19/184, 95% CI 7–16) named health services or towns where no ED exists. Some of these towns had a general practice clinic whereas others had health services where the ED had been closed. Furthermore, 14 of these 19 (74%, 95% CI 51–89)) stated that it would be appropriate to travel by car with chest pain to these non-emergency facilities. The average time to drive to the nearest ED named (either staffed or non-staffed) was 24 min (range 2–60 min). In this cohort, 39% of respondents (73/184, 95% CI 33–47) reported that they were closer to their local general practitioner service than the nearest emergency service.
|% (n, 95% CI)|
|Time to drive to nearest ED at legal speed|
|Under 10 min||31.9 (59/185, 25.6–38.9)|
|10–20 min||40.0 (74/185, 33.2–47.2)|
|20–30 min||17.8 (33/185, 12.9–24.0)|
|30–40 min||7.0 (13/185, 4.0–11.7)|
|40 min and over||3.2 (6/185, 1.3–7.0)|
|General practice clinic or ED closer|
|General practice clinic closer||39.2 (73/184, 32.9–46.9)|
|ED closer||7.5 (14/184, 4.5–12.5)|
|Same time to ED and general practice clinic||52.2 (97/184, 45.5–59.8)|
|Type of nearest ED named|
|ED with 24 h doctor present||38.2 (71/184, 31.9–45.8)|
|ED with doctor on call from home||49.5 (94/184, 43.9–58.2)|
|Non-ED||10.2 (19/184, 6.6–15.6)|
Understanding the role of general practice clinics
The majority of participants (70%, 129/185, 95% CI 63–76) thought they should attend the ED directly if suffering chest pain. Upon further questioning, 22% (41/185, 95% CI 17–29) thought it was necessary to ring their local doctor first, and 18% (34/185, 95% CI 13–25) were unsure. In this farming cohort, 55% of participants (102/186, 95% CI 48–62) believed that a local doctor could exclude an acute myocardial infarction with an electrocardiogram (Table 3). Thirty-seven per cent (68/186, 95% CI 30–44) thought that exclusion of acute myocardial infarction required several hours, at least, in hospital.
|Question||Agree % (n, 95% CI)||Disagree % (n, 95% CI)||Unsure % (n, 95% CI)|
|1||Many people having a heart attack mistakenly think that their chest pain is due to a non-serious cause, such as indigestion.||69.2 (128/185, 62.2–75.4 )||6.5 (12/185, 3.6–11.1)||24.3 (45/185, 18.7–31.0)|
|2||It is usually easy for a person to tell if their chest pain is due to a heart attack, rather than indigestion.||12.9 (24/186, 8.8–18.5)||49.5 (92/186, 42.4–56.6)||37.6 (70/186, 31.0–44.8)|
|3||Most people with chest pain at 02.00 hours should seek medical attention immediately.||75.3 (140/186, 68.6–80.9)||5.9 (11/186, 3.2–10.4)||18.8 (35/186, 13.8–25.1)|
|4||Most people with chest pain at 02.00 hours should wait until the morning to seek medical attention.||9.2 (17/184, 5.8–14.4)||82.1 (151/184, 75.9–87.0)||8.7 (16/184, 5.3–13.7)|
|5||It is recommended that a person should seek immediate medical help if they have 10 min of chest pain, even if it is mild.||56.8 (105/185, 49.6–63.7)||14.1 (26/185, 9.7–19.9)||29.2 (54/185, 9.7–19.9)|
|6||It is recommended that people should seek immediate medical help only if they think their chest pain is due to a heart attack.||31.9 (59/185, 25.6–38.9)||56.2 (104/185, 49.0–63.2)||11.9 (22/185, 7.9–17.4)|
|7||It is difficult to be sure that chest pain is not due to a heart attack without at least several hours in hospital.||36.6 (68/186, 30.0–43.7)||(28.0 52/186, 22.0–34.8)||35.5 (66/186, 29.0–42.6|
|8||A doctor in their rooms can safely rule out a nasty cause of chest pain in most people by examining the person and taking an electrocardiograph of the heart.||54.8 (102/186, 47.7–61.8)||14.5 (27/186, 10.1–20.4)||30.7 (57/186, 24.5–37.6)|
|9||If an ambulance can attend quickly, someone worried about chest pain should go to hospital by ambulance.||78.4 (145/185, 71.9–83.7)||8.1 (15/185, 4.9–13.0)||13.4 (25/185, 9.3–19.2)|
|10||It is safe to go to the ED by car with chest pain, as long as the person with chest pain does not drive.||66.5 (123/185, 59.4–72.9)||18.4 (34/185, 13.4–24.6)||15.1 (28/185, 10.6–21.0)|
|11||Most doctors recommend that people with chest pain should go straight to the ED.||69.7 (129/185, 62.8–75.9)||4.9 (9/185, 2.5–9.1)||25.3 (47/185, 19.7–32.2)|
|12||Most doctors recommend that people with chest pain should ring their general practitioner first, before going to the ED.||22.2 (41/185, 16.8–28.7)||59.5 (110/185, 52.3–66.3)||18.4 (34/185, 13.4–24.6)|
|13||People who live a long way from hospital should decide to go to hospital when they have chest pain earlier than patients who live near the hospital.||64.3 (119/185, 57.2–70.9)||28.5 (53/185, 22.6–35.6)||7.0 (13/185, 4.1–11.8)|
|14||It is sensible that people who have to go a long way to hospital wait at home, for an hour or two, to see if the chest pain will go away.||5.4 (10/186, 2.8–9.7)||84.4 (157/186, 78.5–89.0)||10.2 (19/186, 6.6–15.5)|
Impact of distance and delay
When questioned about appropriate emergency decision making, 57% of respondents (105/185, 95% CI 50–64) agreed with the National Heart Foundation of Australia's major community message, that people with 10 min of chest pain should seek medical advice. Most participants recognized that time of day should not influence medical seeking behaviours. In fact, 82% (151/184, 95% CI 76–87) disagreed with the statement that people with chest pain should wait until morning to seek medical help. The cohort appeared less certain with the delays caused by distance with 84% (157/186, 95% CI 79–89) indicating they thought it unwise for people who have a long way to go to hospital to wait for hours to see if pain would go away by itself, yet only 64% (119/185, 95% CI 57–71) responding that more remote patients should move earlier (Table 3). When the time participants would wait with chest pain at home before seeking emergency assistance was plotted against how far the person with chest pain lived from hospital, the opposite was found. The further away a person lived, the longer they would wait on the farm before seeking emergency medical assistance. The median waiting times have been graphed and show a striking trend towards more remote individuals waiting longer with chest pain before seeking help (Fig. 1). However, it should be noted that there was a wide range of responses to this question (from 1 min to 1 day), resulting in a large degree of overlap between groups, and a small, but not statistically significant association (Spearman's rho = 0.112, P= 0.157).
The majority (78%, 145/185, 95% CI 72–84) of participants thought travelling by ambulance was the appropriate means to seek emergency treatment. However, 67% (123/185, 95% CI 59–73) also indicated that travelling by car while potentially having a myocardial infarction was safe. Many thought that it was safe to go by either ambulance or car.
This pilot study provides an interesting and disturbing insight into emergency medical care seeking behaviours of farm men and women in rural Victoria. Of concern is a pervading uncertainty of what to do if experiencing chest pain. This farming cohort were unsure of how long they should wait if experiencing chest pain and hesitant on whether they should first contact their general practitioner or go directly to the nearest ED. Farmers were sometimes unaware that their local hospital no longer (or never) had an ED, and were sceptical on whether they should call an ambulance in emergency scenarios.
The potential for poor emergency outcomes is further compounded by this farming cohort being at heightened risk of experiencing acute coronary events according to clinical cardiac risk factors collected during the present study. For example, 68% of our representative farming sample was either overweight or obese (on body mass index criteria) compared with the national average of 60%.17 We also revealed in this cohort high rates of elevated blood glucose 54% and hypertension (52% compared with 29% nationally17,19) although these comparisons are not age, race or sex standardized. The low rates of current smokers within our cohort (3%) was a health positive and reflects previous studies showing lower than average smoking rates in farmers.20 This very low rate compares favourably with the 20% of Victorians who currently smoke.21
Although our cohort size and non-randomized selection criteria prohibit conclusions to be drawn about cardiac risk factors in the wider population of farmers, they do correlate well with other studies showing increased rates of cardiac risk factors in rural and remote communities.2 More importantly, these results depict a population of farmers that are at risk of cardiac events, and could be adversely affected if they make the wrong decisions. The findings of this pilot study also highlight the need for further investigation into cardiac risk factors in agricultural communities.
Although participants had a theoretical understanding that waiting at home to see if the chest pain would recede was not a good idea, there was little evidence to suggest that they understood that people living further away from treatment should start to move earlier to overcome the unavoidable delay of longer travel time. In fact, there is a suggestion that they might even wait longer when more distant. This finding is in agreement with the often described ‘distance-decay effect’,22 which is summarized by the longer the trip, the greater the disincentive to travel. As commented by one participant ‘Many times, we tend to avoid it (seeking assistance) if possible because of the distance. If we lived closer to facilities, we would be asking advice from experts a lot more often.’
In the present study, many participants felt they should contact their doctor first, or were unsure on what course of action to take, when experiencing chest pain. As outlined by Farmer et al., rural people base their attitudes of appropriate health-care behaviour on their experience with local general practitioners.23 Rural men and women are also reluctant to seek help unless it is a ‘genuine’ emergency. Factors influencing this behaviour include fear of medical treatment, an ongoing relationship with the doctor and reluctance to be thought of as a ‘time-waster’.23 Among the cohort of farmers in the present study, about one-third thought that they should not seek help unless they thought the pain was caused by a heart attack. Furthermore, if chest pain was experienced during the night, approximately one in ten would wait until the morning before seeking treatment.
A close relationship with a general practitioner remains the cornerstone of good rural health care, but contacting or visiting a local doctor at their clinic is not recommended when a patient has chest pain.8 Despite this cohort's beliefs to the contrary, acute coronary syndrome can seldom be excluded in a clinic setting, and it has been shown to delay hospital presentation by at least 1 h.24,25
In the present study, it appears that many farmers saw being driven to hospital as an acceptable alternative to calling the ambulance. Many reasons have been suggested for the reluctance of rural people to use ambulance services.26 Most rural people can recall an anecdote, from word of mouth or the media, where the use of an ambulance service was said to result in a poor health outcome.26 There is a widespread belief that travelling by car is quicker and thus safer. One participant wrote on their questionnaire that it was ‘sometimes quicker to drive to hospital rather than get an ambulance’. Although the wait at home might be shorter in this scenario, the time to care, which is the most important factor, has been shown to be longer on average.24,27,28 Calling the ambulance also provides immediate telephone advice and activates the emergency medical system. Paramedics also have the ability to defibrillate once on the scene. From our cohort, it is not difficult to foresee a tragedy when someone travels by car for 30 min to an ED where the doctor is unavailable or, even worse, to a town without an ED.
Because of the limited size of our cohort and relatively small number of questions, definitive statements on emergency medical care seeking behaviours in rural communities cannot yet be made. However, the need for further investigation using a randomized sample is apparent to determine whether these concerning trends hold true across the remainder of Australia's farming and remote communities.
Improving health literacy among farm men and women is of the utmost importance as they have higher rates of clinical risk factors and appear to be lagging in emergency knowledge and services when compared with their urban and regional counterparts. Furthermore, farmers are generally the most remotely located within a population and would achieve greater benefit from acting rapidly in response to acute myocardial events and other medical emergencies. Programmes addressing behavioural barriers to accessing care8 and improving emergency decision making within the farming cohort might be readily translatable into rural lives saved.
The authors are appreciative of the statistical help provided by Benjamin Philpot from Flinders University and also to acknowledge the continued support of farming communities Australia-wide.
TB and SM developed the questionnaire, analysed results and co-wrote the paper. CM-G recruited programme participants, ensured completion of questionnaires and provided feedback on the paper. SB secured funding for the study, developed the SFF model, oversaw data collection, questionnaire development, data analysis and compilation of the manuscript.
- 1Australian Institute of Health and Welfare (AIHW). Rural, Regional and Remote Health: A Study on Mortality. AIHW Cat. No. PHE 45. Canberra: AIHW (Rural Health Series No. 2), 2003.
- 2Australian Institute of Health and Welfare (AIHW). Epidemic of Coronary Heart Disease and Its Treatment in Australia. AIHW Cat. No. CVD 21. Canberra: AIHW (Cardiovascular Disease Series No. 20), 2002.
- 4Australian Institute of Health and Welfare (AIHW). Heart, Stroke and Vascular Diseases – Australian Facts 2001. Canberra: AIHW (Cardiovascular Disease Series No. 14), 2001.
- 12Australian Bureau of Statistics. Health Literacy, Australia . 2008 Catalogue No. 4233.0. [Cited 22 May 2008.] Available from URL: http://www.abs.gov.au
- 13The Sustainable Farm Families Project: changing attitudes to health. Rural Remote Health 2009; 9: 1012 (online). [Cited 1 November 2010.] Available from URL: http://www.rrh.org.au/publishedarticles/article_print_1012.pdf, , .
- 14Living Longer on the Land – Sustainable Farm Families in Broadacre Agriculture. RIRDC Publication No 08/048. Canberra: Rural Industries Research and Development Corporation, 2008., , .
- 16National Heart Foundation of Australia. Will you recognise your heart attack? 2009. [Cited 4 Aug 2010.] Available from URL: http://www.heartfoundation.org.au
- 19Diabesity and Associated Disorders in Australia – 2000: The Accelerating Epidemic. The Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Melbourne: International Diabetes Institute, 2001., , et al.
- 21Department of Human Services. Victorian Population Health Survey 2007. Victoria: 2008.