INDIGENOUS HEALTH The recording of Aboriginal and Torres Strait Islander status in general practice clinical records: a cross-sectional study Allison Thomson,1 Simon Morgan,2 Peter O’Mara,3,4 Amanda Tapley,2 Kim Henderson,2 Mieke van Driel,5 Christopher Oldmeadow,6 Jean Ball,6 John Scott,2 Neil Spike,7,8 Lawrie McArthur,9 Parker Magin1,2 R ecording of Aboriginal and Torres Strait Islander status in health care settings has been identified as a key issue in addressing the Council of Australian Government’s commitments to ‘Closing the Gap’(CTG).1 Abstract: Objectives: To document the frequency of recording of Aboriginal and Torres Strait Islander status in general practice (GP) clinical records and to establish associations of this recording. Methods: Cross-sectional analysis of recording of patients’ Aboriginal and Torres Strait Islander status in GP clinical records from GP training practices in four Australian states. Accurate recording of Aboriginal and Torres Strait Islander status is important to both the individual patient and to the greater public health for a number of reasons. These include the planning, monitoring and evaluation of health services,1,2 such as estimates of the health disparity gap between Aboriginal and Torres Strait Islanders and nonIndigenous Australians,3 and assessing the effectiveness of public health interventions such as immunisation programs.4 But, most importantly, identification of Aboriginal and Torres Strait Islander status is a vital component of high-quality patient care. Aboriginal and Torres Strait Islander status is considered an important element of clinical decision making due to the differing health needs of this population.5 This is reflected in specific clinical guidelines for Aboriginal and Torres Strait Islander patients, for example, when assessing cardiovascular risk.6 Delivering culturally safe healthcare also requires the clinician to be aware of a patient’s Aboriginal and Torres Strait Islander status.7 In addition, knowledge of a patient’s Aboriginal and Torres Strait Islander status Results: Of the 9,704 clinical records examined, the patients’ Aboriginal and Torres Strait Islander status had been documented in 5,165 (53.2%). Higher rates of recording were associated with older patient age, practices outside a major city, patients who were not new to the practice and the patient being Aboriginal and Torres Strait Islander. In encounters with Aboriginal and Torres Strait Islander patients, the patient’s status had been documented in 82% of records. Those attending larger practices were less likely to have had their status recorded. Conclusions: This is the first report of Aboriginal and Torres Strait Islander status recording in GP clinical records. Almost 20% of Aboriginal and Torres Strait Islander patients did not have their status recorded in the clinical record, with indications that recording may be unsystematic. Implications: Our findings reinforce the need for a systematic approach to identification of Aboriginal and Torres Strait Islander status in general practice and will inform policy and practice in this important area. Key words: Indigenous Australians, health records, personal, family practice, general practice enables patients to be offered specific GPmediated health initiatives such as Aboriginal and Torres Strait Islander health checks and the CTG medication scheme.8 Despite the importance of accurate recording of status, Aboriginal and Torres Strait Islander people are under-identified in many health data sets.2 Data about frequency of recording of Aboriginal and Torres Strait Islander status comes largely from tertiary institutions and has not previously been documented in primary care.1,9 In the hospital setting, it is estimated that 12% of Aboriginal and Torres Strait Islander patients are not correctly identified.10 Aboriginal and Torres Strait Islander status has also been shown to be under-identified on up to 27% of death records.11 General practice is the cornerstone of Australian primary care12 and, as such, has a vital role in the provision of health care for Aboriginal and Torres Strait Islander people.1 While Aboriginal Community Controlled 1. Discipline of General Practice, University of Newcastle, New South Wales 2. General Practice Training Valley to Coast, New South Wales 3. School of Medicine and Public Health, Faculty of Health and Medicine; University of Newcastle, New South Wales 4. Wollotuka Institute, University of Newcastle, New South Wales 5. Discipline of General Practice, University of Queensland 6. Clinical Research Design IT and Statistical Support Unit, Hunter Medical Research Institute, New South Wales 7. Department of General Practice, University of Melbourne, Victoria 8. Victorian Metropolitan Alliance 9. Adelaide to Outback GP Training, South Australia Correspondence to: Professor Parker Magin, General Practice Training Valley to Coast, Newbolds Building – Gavey St, Mayfield, Newcastle, NSW 2304; e-mail: Parker.magin@newcastle.edu.au Submitted: September 2014; Revision requested: November 2014; Accepted: February 2015 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. The authors have stated they have no conflict of interest. Aust NZ J Public Health. 2016; 40 (Suppl. 1) S70-S74; doi: 10.1111/1753-6405.12400 S70 Australian and New Zealand Journal of Public Health © 2015 The Authors Vol. 40 (Suppl. 1) 2016 General practice recording of Aboriginal and Torres Strait Islander status Health Services (ACCHS) have been established to provide care for Aboriginal and Torres Strait Islander people, recent figures suggest that 50–60% of Aboriginal and Torres Strait Islander patients access health care outside of these organisations.13,14 In Australia, Aboriginal and Torres Strait Islander patients account for about 1.6% of general practice consultations.15 In Australian general practice, sub-optimal collection and recording of Aboriginal and Torres Strait Islander status is proposed to be a long-standing problem.1 However, evidence regarding prevalence of Aboriginal and Torres Strait Islander status recording is lacking. Recently, steps have been taken to improve recording with changes to practice accreditation standards specifying that practices must routinely record Aboriginal and Torres Strait Islander status in patient health records.5 It is recommended that questions about Aboriginal and Torres Strait Islander status be part of new patient registration information.9 Previous studies have examined general practice procedures, rather than individual clinical files, and demonstrated that only a minority of mainstream practices had routine identification procedures in place.9,16 major cities through to very remote practices in four Australian states (New South Wales, Victoria, Tasmania and South Australia). Data collection for the ReCEnT project takes place within mainstream general practices. No Aboriginal Community Controlled Health Services are included in the analysis. routinely bulk bills (i.e. there is no out-ofpocket expense to the patient). Practice postcode was used to define the Australian Standard Geographical Classification – Remoteness Area (ASGC-RA)20 (the degree of rurality) and the SEIFA code of the practice location. Characteristics of the participating practices are recorded by each registrar each term, on a paper based questionnaire. Statistical analysis Participating GP registrars undergo faceto-face orientation in how to complete the patient encounter forms and record encounter data on the paper data collection instrument. The collected data includes patient demographics, including Aboriginal and Torres Strait Islander status. Registrars are requested to directly ask each patient their Aboriginal and Torres Strait Islander status according to best practice.18 GP registrars record the details of 60 consecutive patient encounters each training term. For one round of data collection in the second half of 2012, an additional question was added to the data collection instrument. This question was: “Was the patient’s Aboriginal and/or Torres Strait Islander status documented in the clinical record prior to this consultation?” Response options were ‘yes’ or ‘no’. Previous studies have examined barriers to identification of Aboriginal and Torres Strait Islander patients in general practice. Such barriers have included a lack of routine processes for identification,9 lack of awareness of the importance of identification, apprehension regarding reactions of patients to the question, and assumptions that Aboriginal and Torres Strait Islander patients did not attend particular practices.16 This study aimed to document the prevalence of recording of Aboriginal and Torres Strait Islander status in the individual patient’s clinical record and to establish patient and practice associations of this recording in general practice. Outcome factors The primary outcome factor was prior recording of Aboriginal and Torres Strait Islander status in the individual patient’s clinical record. Records from encounters where there was prior recording of status were compared to all other clinical records of patients included in the study. A secondary analysis was performed that included only patients of Aboriginal and Torres Strait Islander background. Records from encounters where prior recording of status had occurred were compared to all other clinical records of Aboriginal and Torres Strait Islander patients. Independent variables Methods This was a cross-sectional analysis of practice data collected during general practice consultations as part of the Registrar Clinical Encounters in Training (ReCEnT) cohort study. The ReCEnT study methodology has been described in detail elsewhere.17 Briefly, ReCEnT is an ongoing cohort study of the clinical and educational content of GP registrar encounters with patients. It is undertaken in four general practice regional training providers (RTPs), encompassing Vol. 40 (Suppl. 1) 2016 Independent variables were categorised as patient or practice factors. Patient factors were age, gender, Aboriginal and Torres Strait Islander status, non-English speaking background (NESB), new patient to registrar, and new patient to practice (that is, not having previously attended the practice). Practice factors included rurality, decile of the Socioeconomic Index for Area (SEIFA) Relative Index of Disadvantage,19 number of GPs working at the practice, use of computerised medical records and whether the practice Analysis was performed on one round of data collection from 2012. For the primary outcome of whether Aboriginal and Torres Strait Islander status had been recorded in the clinical record, univariable associations with independent variables were tested with logistic regression within a generalised estimating equations (GEE) framework to account for clustering of consultations within registrars. Multivariable analysis was then performed using logistic regression within a GEE framework to account for clustering of consultations within registrars. For the second analysis confined to patients of Aboriginal and Torres Strait Islander background, univariable associations with independent variables were tested with logistic regression within a GEE framework. Multivariable analysis was then performed using logistic regression within a GEE framework and exact methods were used when there were small numbers of observations. For both outcomes, variables with a p-value <0.20 and a relevant effect size in the univariable analysis were included in the multiple regression model. To analyse the effect of missing data on results, further analyses employing multiple imputation were performed using chained regressions to generate 10 imputed data sets. Missing values were predicted (under the missing at random assumption) using an iterative series of appropriate regression models conditional on the observed value of the outcome variable. Coefficients and standard errors for the variability between imputations were combined using the method of Rubin.21 All regression analyses were performed using SAS v9.4 (SAS Institute Inc., Cary, NC, USA) and multiple imputation using Stata v13 (Statacorp, College Station, TX, USA). Statistical significance was set at the 0.05 level. Ethics approval for the study was obtained from the Human Research Ethics Committee of the University of Newcastle, NSW. H-20120110. Australian and New Zealand Journal of Public Health © 2015 The Authors S71 Thomson et al. Results The response rate of registrars for the ReCEnT project for this round of data collection was 94.7%. There were 13,078 patient encounters, collected from 152 practices by 217 registrars. The question regarding prior recording of Aboriginal and Torres Strait Islander status was answered in 9,704 encounters (74.2%). Table 1 shows the characteristics of the patients and practices in this study and their univariable associations with status recording. Of the 9,704 valid encounters, 3,675 (37.9% [95%CI: 36.9-38.8]) were with male patients. The mean age of patients was 39.9 years. Of these encounters, 117 (1.2% [95%CI: 0.9-1.4]) were with patients who identified as Aboriginal and Torres Strait Islander. Practices were from locations across ASGC-RA20 codes 1 to 5, reflecting major cities through to very remote practice locations. Univariable associations of status recording in Aboriginal and Torres Strait Islander patients are presented in Table 2. Table 1: Characteristics associated with recording the patient’s Aboriginal and Torres Strait Islander status (n=9,704a). Variable Patient age group Class <15 15 to <31 929 (45%) 1,139 (55%) 31 to <55 1,451 (48%) 1,591 (52%) Patient gender 55+ Male 1,234 (44%) 1,776 (48%) 1,569 (56%) 1,899 (52%) 0.37 Aboriginal or Torres Strait Islander Female No 2,681 (46%) 4,518 (47%) 3,170 (54%) 5,069 (53%) <0.001 Non-English Speaking Background Yes No 21 (18%) 4,096 (46%) 96 (82%) 4,778 (54%) 0.55 New patient to the registrar Yes No 270 (52%) 1,899 (43%) 249 (48%) 2,502 (57%) <0.001 New patient to surgery Yes No 2,546 (49%) 4,051 (45%) 2,612 (51%) 4,898 (55%) <0.001 Practice sizec Yes Small 445 (67%) 1,325 (47%) 223 (33%) 1,504 (53%) 0.89 Does the practice routinely bulk billd Large No 3,175 (47%) 3,860 (49%) 3,563 (53%) 3,939 (51%) 0.025 Rurality Yes Major city 628 (34%) 2,979 (53%) 1,199 (66%) 2,682 (47%) 0.021 Inner regional 1,101 (37%) 1,837 (63%) 459 (42%) 646 (58%) 6.6 (2.2) 6.4 (2.2) Of practices included in our study, 98.2% used computerised medical records. Outer regional/ Remote/Very remote mean (SD) SEIFA Index (decile) All patients including non-Aboriginal and Torres Strait Islander patients 0.60 a: Numbers may not add to 9704 due to missing data b: Logistic regression adjusted for clustering c: Practices defined as small if less than 6 GPs were working in the practice d: Practices were defined as routinely bulk-billing if all patients were routinely bulk-billed. In 53.2% (5,165 of 9,740 [95%CI: 52.2-54.2]) of all patient encounters, the patient’s Aboriginal or Torres Strait Islander status had been documented in the clinical record prior to the consultation. Aboriginal and Torres Strait Islander patients The regression model (including records of both Aboriginal and Torres Strait Islander and non-Aboriginal and Torres Strait Islander patients) with outcome factor ‘Aboriginal and Torres Strait Islander status recorded’ is presented in Table 3. Prior recording of Aboriginal and Torres Strait Islander status was significantly associated with older patient age and the patient being Aboriginal and Torres Strait Islander. Patients who were new to the practice or new to the registrar were significantly less likely to have had their status recorded previously. For patients of Aboriginal and Torres Strait Islander background, the univariable findings of association of status recording are presented in Table 2 and the multivariable model is presented in Table 3. Of Aboriginal and Torres Strait Islander patients, 82.0% (96 of 117 [95%CI: 73.9-88.5]) had their status documented prior to the encounter. Aboriginal and Torres Strait Islander patients attending a doctor at a larger practice size (6 or more GPs) were significantly less likely (OR of 0.15 [95%CI: 0.04-0.65]) to have their status recorded prior to that visit. Compared to patients attending practices in major cities, patients attending practices in inner regional areas were more likely to have their status recorded prior to the consultation. However, the increased odds of patients in outer regional and remote areas (compared to major cities) having their status recorded did not reach statistical significance. S72 Aboriginal and Torres Strait Islander Status recorded No (n=4,539) Yes (n=5,165) pb 0.004 816 (48%) 881 (52%) Imputation analyses There were missing data of 25.8% for the outcome variable. In the imputation analyses (see Supplementary Table 1, available online) for all patients (Aboriginal and Torres Strait Islander and non-Aboriginal and Torres Strait Islander patients), the same variables remained in the model. All significant variables remained significant. The non-significant finding of bulk-billing practices being more likely to have status recorded became statistically significant, as did recording in outer regional/remote/very remote locations. In records of Aboriginal and Torres Strait Islander patients only (examining our secondary outcome), NESB dropped out of the imputed multivariable model. Discussion Summary of main findings Patients were found to have their Aboriginal and Torres Strait Islander status documented prior to the consultation in 53.2% of encounters. Associations of recording of status were older patients, Aboriginal and Torres Strait Islander patients and the patient not being new to the practice. Of the Aboriginal and Torres Strait Islander patients in this study, four in five were found to have their status documented prior to the encounter. In this smaller number of consultations, the only significant association of recording of status was size of practice, with those attending larger practices less likely to have their status recorded. Australian and New Zealand Journal of Public Health © 2015 The Authors Vol. 40 (Suppl. 1) 2016 General practice recording of Aboriginal and Torres Strait Islander status Strengths and limitations A major strength of our study is the high response rate, with 94.7% of registrars participating, and the conduct of the study across four Australian states and all ASGC-RS classifications. This response rate is exceptional for studies of general practitioners.22 While the response rate is high, there was 25.8% missing data for the outcome variable. For the primary analysis; however, use of multiple imputation techniques strongly supported the findings of our analysis. Another strength of the study is the faceto-face orientation of the ReCEnT project, including orientation to best practice in asking Aboriginal and Torres Strait Islander status.18 We are thus confident Table 2: Characteristics associated with recording the patient’s Aboriginal and Torres Strait Islander status for Aboriginal and Torres Strait Islander patients (n=117a). Variable Patient age group Aboriginal and Torres Strait Islander Status recorded No (n=21) Yes (n=96) pb 0.46 26 (76%) 8 (24%) Class <15 15 to <31 5 (14%) 30 (86%) 7 (21%) 26 (79%) Patient gender 1 (6.7%) 11 (28%) 14 (93%) 29 (73%) 0.057 Non-English Speaking Background Female No 10 (13%) 19 (17%) 66 (87%) 91 (83%) 0.15 (Exact) New patient to the registrar Yes No 2 (50%) 9 (15%) 2 (50%) 51 (85%) 0.34 New patient to surgery Yes No 12 (22%) 18 (18%) 43 (78%) 84 (82%) 0.72 (Exact) Practice sizec Yes Small 3 (21%) 3 (6.1%) 11 (79%) 46 (94%) 0.004 Does the practice routinely bulk billd Large No 18 (27%) 19 (22%) 49 (73%) 67 (78%) 0.051 Rurality Yes Major city 2 (6.5%) 9 (17%) 29 (94%) 44 (83%) 0.92 Inner regional 8 (20%) 32 (80%) Outer regional/Remote/Very remote mean (SD) 4 (17%) 5.6 (2.1) 20 (83%) 5.3 (2.6) SEIFA Index (decile) 0.64 a: Numbers may not add to 117 due to missing data b: Logistic regression adjusted for clustering c: Practices defined as small if less than 6 GPs were working in the practice d: Practices were defined as routinely bulk-billing if all patients were routinely bulk-billed. Table 3: Results of Multivariable analyses. Univariable Variable Adjusted for other variables in the model OR (95% CI) P Class OR (95% CI) P Predictors of recording the patient’s Aboriginal and Torres Strait Islander status (all patients) 1.11 (1.01–1.22) 0.032 1.11 (1.01–1.21) Patient age group 15 to <31 0.037 1.11 (1.01–1.22) 0.037 Referent age <15 31 to <55 1.13 (1.03–1.24) 0.008 0.004 1.18 (1.06–1.32) <.001 1.22 (1.10–1.36) 55+ Aboriginal or Torres Strait Islander Yes 2.64 (1.74–4.01) <.001 3.09 (1.91–5.00) <0.001 New patient to the registrar Yes 0.82 (0.75–0.89) <.001 0.89 (0.82–0.97) 0.007 New patient to surgery Yes 0.51 (0.42–0.62) <.001 0.53 (0.43–0.64) <0.001 Does the practice routinely bulk bill Yes 1.91 (1.08–3.35) 0.025 1.79 (0.96–3.34) 0.068 Rurality Inner regional 1.96 (1.21–3.17) 0.006 2.06 (1.22–3.48) 0.007 Referent Major City Outer regional/ 1.51 (0.77–2.96) 0.23 1.68 (0.79–3.58) 0.18 Remote/Very remote Predictors of recording the patient’s Aboriginal and Torres Strait Islander status for Aboriginal and Torres Strait Islander patients Patient gender Female 2.50 (0.96–6.55) 0.061 2.19 (0.77–6.27) 0.14 Non-English Speaking Background No 4.79 (0.63–36.15) 0.15 (Exact) 9.59 (0.97–95.17) 0.050 (Exact) Practice size Large 0.18 (0.05–0.64) 0.009 0.15 (0.04–0.65) 0.011 Does the practice routinely bulk bill Yes 4.11 (0.9–18.82) 0.068 4.81 (0.75–31.06) 0.099 Vol. 40 (Suppl. 1) 2016 A potential limitation is that while there is excellent generalisability to Australian GP practices that train registrars, these practices may differ from non-training practices. A further limitation is that while the findings of the primary analysis regarding the associations of recording Aboriginal and Torres Strait Islander status in the full practice population are robust, interpretation of the associations of recording in Indigenous patients is limited by the smaller sample size. Comparison with existing literature and interpretation of findings 55+ Male 31 to <55 of the reliability of data for this important independent variable (the Aboriginal and Torres Strait Islander status of each patient). The percentage of patients in mainstream general practice in our study who identified as Aboriginal or Torres Strait Islander (1.2%) is broadly consistent with previously published data.15,23 To our knowledge, this is the first time prevalence and associations of documentation of Aboriginal and Torres Strait Islander status in the clinic record has been described in the general practice setting (the setting of a substantial proportion of Aboriginal and Torres Strait Islander health care delivery). As a result, there is no literature with which to compare our prevalence of 53.2% documentation of status. We found a strong positive association between recording of Aboriginal and Torres Strait Islander status and the patient being Aboriginal and Torres Strait Islander. This has not been described before and may reflect patients self-identifying to their GP. Our finding of 82% of Aboriginal and Torres Strait Islander patients having had their status recorded in the general practice clinical file is comparable with hospital findings from the Australian Institute of Health and Welfare (AIHW). In hospitals, 88% of Aboriginal and Torres Strait Islander patients were found to be correctly identified on admission records.10 The same report found that the percentage of Aboriginal and Torres Strait Islander patients correctly identified increased with remoteness. Although we found that there was an increased recording of status for patients in inner regional areas compared to major cities, this association was not statistically significant for outer regional, remote and very remote areas. When we examined only Aboriginal and Torres Strait Islander patients, there was no difference in recording of status by degree of rurality. Australian and New Zealand Journal of Public Health © 2015 The Authors S73 Thomson et al. The proportion of patients new to the practice with status recorded in our study was 33.4% (see Table 1). Previous literature has suggested that only one-third of practices had methods in place to routinely collect Aboriginal or Torres Strait Islander status for new patients.9 General practice staff have previously described that identification of Aboriginal and Torres Strait Islander status in new patients would be easier than in pre-existing patients.9 Our findings suggest that although it may be perceived as easier, actual recording of new patients’ Aboriginal and Torres Strait Islander status is relatively infrequent. Implications for practice and policy We found that younger patient age was associated with less recording of Aboriginal and Torres Strait Islander status. This may suggest that Aboriginal and Torres Strait Islander status is not being routinely and systematically recorded in the clinical record by practices. An interpretation is that status tends to be recorded not systematically but, rather, only when a condition associated with Aboriginal and Torres Strait Islander status (especially a chronic condition, for example diabetes) develops. However, for the clinician, knowing the Aboriginal and Torres Strait Islander status of a child is just as important as knowing the status of an adult, e.g. in the management of otitis media.24 Our findings should encourage clinicians to especially consider asking and recording the Aboriginal and Torres Strait Islander status of younger patients in their practice. Patients attending practices in inner regional locations had almost twice the odds of having their status recorded than those in major cities. Given that 32% of Aboriginal and Torres Strait Islander people live in urban areas,25 our findings suggest there is an imperative to improve recording in major cities. This study took place after changes were made to general practice accreditation, making collection and recording of Aboriginal and Torres Strait Islander status on all patients (regardless of ethnic background) a standard for practice accreditation.5 Our findings – that nearly half the clinical records did not have Aboriginal and Torres Strait Islander health status documented, and that nearly 20% of Aboriginal and Torres Strait Islander patients remained unidentified – suggest that further work remains to be done. Best practice guidelines regarding identification of Aboriginal and Torres Strait Islander status in general practice recommend a systems- S74 based approach to identification.9 This includes questions about Aboriginal and Torres Strait Islander status as part of new patient registration information.9 Our findings regarding new patients to the surgery being less likely to be recorded suggest that this is not occurring at a practice level, or at least status is not being entered into the file prior to the clinician seeing the patient. This is consistent with the contention that ensuring all patients are asked about Aboriginal and Torres Strait Islander status “cannot easily be centrally mandated or enforced.”1 It is perhaps encouraging that the majority of Aboriginal and Torres Strait Islander patients were recorded in the clinical record. However, given the importance of this to the individual patient, and that just under 20% of patients who identify as Aboriginal or Torres Strait Islander did not have their status recorded, there is still considerable room for improvement on this important public health issue. Promotion and education around the importance of the issue and implementation of best practice guidelines2 is imperative for policy makers and practitioners alike. Our findings suggest increased attention be given to recording the status of young patients and new patients, and improving recording in practices in urban locations. References 1. 2. 3. 4. 5. 6. 7. 8. 9. Australian Institute of Health and Welfare. Taking the Next Steps: Identification of Aboriginal and Torres Strait Islander Status in General Practice. Catalogue No.: IHW 100. Canberra (AUST): AIHW; 2013. Australian Institute of Health and Welfare. National Best Practice Guidelines for Collecting Indigenous Status in Health Data Sets. Catalogue No.: IHW 29. Canberra (AUST): AIHW; 2010. Randall DA, Lujic S, Leyland AH, Jorm LR. Statistical methods to enhance reporting of Aboriginal Australians in routine hospital records using data linkage affect estimates of health disparities. Aust N Z J Public Health. 2013;37:442-9. Naidu L, Chiu C, Habig A, Lowbridge C, Jayasinghe S, Wang H, et al. Vaccine preventable diseases and vaccination coverage in Aboriginal and Torres Strait Islander people, Australia 2006-2010. Commun Dis Intell Q Rep. 2013;37 Suppl:1-95. Royal Australian College of General Practitioners. Standards for General Practices. 4th ed. South Melbourne (AUST): RACGP; 2010. National Vascular Disease Prevention Alliance. Guidelines for the Assessment of Absolute Cardiovascular Disease Risk. Melbourne (AUST): National Stroke Foundation; 2012. Liaw ST, Lau P, Pyett P, Furler J, Birchill M, Rowley K, et al. Successful chronic disease care for Aboriginal Australians requires cultural competence. Aust N Z J Public Health. 2011;35:238-48. Department of Human Services. Closing the Gap- PBS Co-payment Measure [Internet].Canberra (AUST): Commonwealth of Australia; 2010 [cited 2013 Aug 21]. Available from: http://www.medicareaustralia.gov.au/ provider/pbs/prescriber/closing-the-gap.jsp - N10009 Kelaher M, Parry A, Day S, Paradies Y, Lawlor J, Solomon L. Improving the Identification of Aboriginal and Torres Strait Islander People in Mainstream General Practice. Melbourne (AUST): The Lowitja Institute; 2010. 10. Australian Institute of Health and Welfare. Indigenous Identification in Hospital Separations Data – Quality Report. Catalogue No.: IHW 90. Canberra (AUST): AIHW; 2013. 11. Briffa TG, Sanfilippo FM, Hobbs MS, Ridout SC, Katzenellenbogen JM, Thompson PL, et al. Underascertainment of Aboriginality in records of cardiovascular disease in hospital morbidity and mortality data in Western Australia: A record linkage study. BMC Med Res Methodol. 2010;10:111. 12. Britt H, Miller GC, Charles J, Valenti L, Fahridin S, Pan Y, et al. General Practice in Australia, Health Priorities and Policies 1998 to 2008 [Internet]. Catalogue No.: GEP 24. Canberra (AUST): AIHW; 2009 [cited 2013 Apr 4]. Available from: http://www.aihw.gov.au/publicationdetail/?id=6442468257%3E 13. Australian Health Ministers’Advisory Council. Aboriginal and Torres Strait Islander Health Performance Framework: 2010 Report. Canberra (AUST): AHMAC; 2011. 14. Couzos S, Thiele DD. Closing the gap depends on ACCHSs. Med J Aust. 2009;190:541. 15. Britt H, Miller GC, Henderson J, Charles J, Valenti L, Harrison C, et al. General Practice Activity in Australia 2011–12. General Practice Series No.: 31. Sydney (AUST): Sydney University Press; 2012. 16. Kehoe H, Lovett RW. Aboriginal and Torres Strait Islander health assessments - barriers to improving uptake. Aust Fam Physician. 2008;37:1033-8. 17. Morgan S, Magin PJ, Henderson KM, Goode SM, Scott J, Bowe SJ, et al. Study protocol: The registrar clinical encounters in training (ReCEnT) study. BMC Fam Pract. 2012;13:50. 18. National Faculty of Aboriginal and Torres Strait Islander Health. Identification of Aboriginal and Torres Strait Islander People in Australian General Practice [Internet]. East Melbourne (AUST): RACGP; 2011 [cited 2013 Aug 21]. Available from: http://www.racgp.org.au/ yourracgp/faculties/aboriginal/guides/identification/ 19. Australian Bureau of Statistics. 2039.0 - Information Paper: An Introduction to Socio-economic Indexes of Areas (SEIFA) [Internet]. Canberra (AUST): ABS; 2006 [cited 2013 Jul 17]. Available from: http://www.abs. gov.au/ausstats/abs@.nsf/mf/2039.0/ 20. Australian Bureau of Statistics. ASGC Remoteness Classification: Purpose and Use. Census Paper No.: 03/01. Canberra (AUST): ABS; 2003. 21. Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York (NY): Wiley; 1987. 22. Bonevski B, Magin P, Horton G, Foster M, Girgis A. Response rates in GP surveys - trialling two recruitment strategies. Aust Fam Physician. 2011;40:427-30. 23. Zhang C, Valenti L, Britt H. General practice encounters with Aboriginal and Torres Strait Islander people. Aust Fam Physician. 2014;43:15. 24. Darwin Otitis Guidelines Group. Recommendations for Clinical Care Guidelines on the Management of Otitis Media in Aboriginal and Torres Strait Islander Populations. Canberra (AUST): Commonwealth Department of Health and Ageing; 2010. 25. Australian Bureau of Statistics. 4704.0. - The Health and Welfare of Australia’s Aboriginal and Torres Strait Islander Peoples. Canberra (AUST): ABS; 2012. Supporting Information Additional supporting information may be found in the online version of this article: Supplementary Table 1: Results of analyses including imputed data. Australian and New Zealand Journal of Public Health © 2015 The Authors Vol. 40 (Suppl. 1) 2016